Effects of Deep Brain Stimulation on Speech Motor Planning/Programming in Patients with Parkinson's Disease

jones_h.pdf

jones_h_Page_043.txt

jones_h_Page_077.txt

jones_h_Page_097.txt

jones_h_Page_044.txt

jones_h_Page_010.txt

jones_h_Page_041.txt

jones_h_Page_063.txt

jones_h_Page_101.txt

jones_h_Page_102.txt

jones_h_Page_029.txt

jones_h_Page_025.txt

jones_h_Page_036.txt

jones_h_Page_070.txt

jones_h_Page_075.txt

jones_h_Page_092.txt

jones_h_Page_012.txt

jones_h_Page_023.txt

jones_h_Page_005.txt

jones_h_Page_039.txt

jones_h_Page_033.txt

jones_h_Page_060.txt

jones_h_Page_020.txt

jones_h_Page_090.txt

jones_h_Page_079.txt

jones_h_Page_059.txt

jones_h_Page_098.txt

jones_h_pdf.txt

jones_h_Page_038.txt

jones_h_Page_078.txt

jones_h_Page_031.txt

jones_h_Page_030.txt

jones_h_Page_014.txt

jones_h_Page_019.txt

jones_h_Page_057.txt

jones_h_Page_084.txt

jones_h_Page_085.txt

jones_h_Page_046.txt

jones_h_Page_054.txt

jones_h_Page_094.txt

jones_h_Page_022.txt

jones_h_Page_087.txt

jones_h_Page_028.txt

jones_h_Page_093.txt

jones_h_Page_083.txt

jones_h_Page_071.txt

jones_h_Page_067.txt

jones_h_Page_052.txt

jones_h_Page_050.txt

jones_h_Page_073.txt

jones_h_Page_058.txt

jones_h_Page_015.txt

jones_h_Page_011.txt

jones_h_Page_004.txt

jones_h_Page_007.txt

jones_h_Page_082.txt

jones_h_Page_009.txt

jones_h_Page_072.txt

jones_h_Page_034.txt

jones_h_Page_027.txt

jones_h_Page_091.txt

jones_h_Page_002.txt

jones_h_Page_017.txt

jones_h_Page_065.txt

jones_h_Page_021.txt

jones_h_Page_068.txt

jones_h_Page_066.txt

jones_h_Page_053.txt

jones_h_Page_047.txt

jones_h_Page_080.txt

jones_h_Page_095.txt

jones_h_Page_026.txt

jones_h_Page_006.txt

jones_h_Page_055.txt

jones_h_Page_032.txt

jones_h_Page_076.txt

jones_h_Page_051.txt

jones_h_Page_056.txt

jones_h_Page_045.txt

jones_h_Page_049.txt

jones_h_Page_035.txt

jones_h_Page_037.txt

jones_h_Page_018.txt

jones_h_Page_088.txt

jones_h_Page_003.txt

jones_h_Page_048.txt

jones_h_Page_013.txt

jones_h_Page_024.txt

jones_h_Page_100.txt

jones_h_Page_099.txt

jones_h_Page_086.txt

jones_h_Page_081.txt

jones_h_Page_061.txt

jones_h_Page_089.txt

jones_h_Page_074.txt

jones_h_Page_042.txt

jones_h_Page_008.txt

jones_h_Page_062.txt

jones_h_Page_016.txt

jones_h_Page_069.txt

jones_h_Page_040.txt

jones_h_Page_064.txt

jones_h_Page_001.txt

jones_h_Page_096.txt

LIST OF TABLES


Table page

2-1 Auditory-perceptual dysarthria severity scale. ...................................... ............... 26

2-2 Power analysis to detect differences between 'switch' and 'no switch' speech
responses..................... .......................................27

2-3 Powers corresponding to r, ratio of mean SRT between the 'on' and 'off stimulation
conditions, based on sensitivity analysis. ........................................ ....... ............... 28

4-1 Individual and group descriptive data............. ............. ....... .................... 62

4-2 Summary statistics for speech reaction time (SRT) and response accuracy by priming
condition and stim ulation state. .............................................. .............................. 63

4-3 Mean difference and p-values for speech reaction time (SRT) and response accuracy
by prim ing condition and stimulation state ................................................ ............... 64










previous target (6), initial sound or syllable repetition (5), no speech response (3), and production

of the prime in a 'switch' task (2). A combination of two or more of these error types was also

encountered in five trials. Please see Chapter 2 for methods and operational definitions used in

determining error type.

These errors in response accuracy are likely best explained by a variety of hypothesized

mechanisms. Premature responses are generally the most common type of errors described in RT

studies and they are generally accepted to reflect decreased inhibition of movement execution

and these types of errors comprised approximately one-third (33%) of the total errors in our

study. Spencer & Rogers (2005) reported that premature responses (or "early responders")

occurred in 25% of the total errors produced by subjects with hypokinetic dysarthria.

The next two most frequently occurring error types, phonological (33%) and

lexical/semantic errors (7%), comprised approximately 40% of the total errors. As defined in

Chapter 2, phonological errors include single phoneme omissions, deletions, substitutions, and

transpositions, while lexical/semantic errors consist of whole real word substitutions. These two

errors types are best considered to be linguistically based errors, rather than errors occurring at

the motor level. Using the model of Van der Merwe (1997), the hypothesized level of disruption

would be at the level of linguistic symbolic planning, rather than the level which is the focus of

this study, motor planning/programming. Although the relatively high number of language based

errors was unexpected, disorders of language function in PD such as subtle declines in

expressive language function are being recognized with increasing regularity (Ellis et al., 2006;

Ellis & Rosenbek, 2007). Additionally, in our group of participants, we conducted discourse

analysis of repetition of multisyllabic words, repetition of sentences, and connected speech in

order to determine linguistic competency. Some evidence of language disturbance was found in












32 32


30-





I 20





10-
-5
10-

6 6 6




PR PHONO LS PPP PPT ME ISR NR P
Type of errors


Figure 5-1. Distribution of errors differentiated among error type. PR = premature
response, PHONO = phonological error, LS = lexical/semantic error, PPP = production of
previous prime, PPT = production of previous target, ME = multiple errors (i.e., two or
more of other error types), ISR = initial sound repetition, NR = no response, P =
production of the prime in 'switch' trial.










improvements with DBS. Further study in patients with more severe dysarthria would be

beneficial to more completely understand the speech effects of DBS in individuals with PD.










& Rogers, 2005). Stimuli were linguistically controlled so that all effects were due to the

experimental manipulation.



Sixteen prime-target pairs were used in the experiment. Each of these sixteen prime-

target pairs was presented four times. Three of the presentations for each of the prime-target

pairs were the no-switch condition (i.e., shopper-shopper), while one of the four presentations

was the switch condition (i.e., shopper-chopper). This ensured that subjects expected the prime

to accurately provide information regarding the target.

Equipment

The equipment configuration is shown in cartoon format in Figure 2-3. A laptop

computer (Compaq Presario V2000) was used to run the E-Prime computer software

(Psychology Software Tools) used in the response priming procedure. Stimuli were presented on

a separate 19" computer monitor (Planar PL1910M) placed in front of the subject. Presentation

of stimuli and calculation of response latencies were managed by the E-Prime program and a

Serial Response Box (SRB) (Psychology Software Tools). Registration of speech onset, as

measured by the onset of voicing, was measured using an accelerometer (PCB Piezotronics

352C22) placed inferior to the thyroid cartilage with adhesive tape. The accelerometer was

powered by a portable power source (PCB Piezotronics Model 480C02 ICP Signal Conditioner)

and the signal captured by this transducer activated the Voice Key of the Serial Response Box to

measure response latency.

High-quality acoustic recordings were made during administration of the response

priming procedure, FAS, and Stroop using a high-quality digital recorder (Marantz PMD671)

and a head-mounted microphone (Shure SM10A) positioned two centimeters from the left corner










suggest that disordered maintenance of speech motor programs may result in "abnormally placed

pauses, difficulty with progression through an utterance and difficulty initiating articulation" (p.

348). Similarly, disordered switching of speech motor programs may be associated with

"difficulty stopping an ongoing response, marked hesitations between movement segments, and

occasional inability to switch from one to another movement" (Spencer & Rogers, 2005, p. 348).

To test the notion that speech motor planning/programming is disrupted in individuals with

PD and hypokinetic dysarthria, Spencer and Rogers (2005) employed a RT paradigm. Ten

participants with PD and hypokinetic dysarthria and 15 normal controls were tested using a

response priming procedure in which participants were provided with a prime to supply

information regarding target (see Chapter 2 "Experimental Procedures" for more details). The

primary dependent variable was SRT. Results provided preliminary evidence for the notion that

maintenance and switching of speech motor programs is disordered in participants with PD and

hypokinetic dysarthria.

Reaction Time/Speech Reaction Time

The use of RT experiments being used as an index of various underlying neural processes

has a history that dates back to at least the mid-19th century (Smith, 2004). The basic RT

paradigm (to which an infinite array of variables can be added) involves presentation of a

stimulus to subjects who are then required to start a movement as quickly as possible (Hallett,

1990). RT is defined as the temporal duration between the presentation of the stimulus and the

initiation in movement. Adaptations to the RT paradigm usually increase the complexity of the

task in any number of ways, by methods such as demanding more complicated movements or

adding a pause between presentation of the stimulus and the command for movement. Increased

complexity invariably results in increases in RT. RT in individuals with PD has been studied










expected ('no-switch' condition) or with a low probability by an unexpected target word

('switch' condition). The primary dependent measure was SRT 'on' and 'off DBS.

Twelve participants completed the study. Statistically significant differences (p < 0.05)

were found in SRT between the 'no switch' and 'switch' conditions, regardless of DBS state.

Significant differences were also found in SRT in the 'no switch' condition (i.e., subjects

produced a word more quickly 'on' versus 'off stimulation). No differences across stimulation

conditions in the 'switch' condition were observed. These findings suggest that the greater

complexity of the 'switch' condition requires increased speech motor planning/programming

processes which can be measured temporally. DBS was also found to improve SRT in the 'no

switch' condition, suggesting that maintenance of speech motor programs may be improved by

DBS. No differences between DBS states were found in the 'switch' condition, suggesting that

DBS has little influence on the multiple processes involved in a motor program switching task.










fashion. Using the Medtronic Access Review device, a trained research assistant turned subjects

'on' or 'off allowing subjects and the principal investigator to remain blinded to the test

condition. Thirty minutes later, the first test session was initiated and completed in

approximately 30 minutes. The second two-hour DBS washout was then started and this was

followed by implementation of the second stimulation condition by the research assistant.

Subjects that were tested 'on' DBS during then first test session were tested 'off DBS during the

second test session, and vice versa. The second test session was also started 30 minutes after the

condition was initiated and completed in approximately 30 minutes. Following completion of

both test sessions, all subjects were turned 'on' DBS and took their anti-PD medications.

Experimental Procedures

Experimental sessions were conducted primarily at the homes of the participants, as well

as the UF Speech and Hearing Center in select cases. A response priming procedure based on the

work of Spencer (Spencer & Rogers, 2006; Spencer, 2006) was utilized. As shown in Figure 2-2,

in this paradigm, participants are provided with a prime to supply information regarding a target.

Over presentation of a series of target words for speech production, the prime word was followed

with a high probability (75% of trials) by the primed target as expected in the 'no switch'

condition, such as with the prime-target pair "shopper-shopper". In 25% of trials, however, the

subject was presented with an incorrect prime, discovered upon presentation of the command for

movement (i.e., the target word). In other words, the prime-target pair did not match in the

'switch' condition (i.e., "shopper-chopper"). Subjects were trained to read the target word aloud

as quickly and accurately as possible and were further instructed to be prepared to say the prime

word due to the high likelihood it would match the upcoming target. Each trial began with a

signal (i.e., +) on a computer screen followed by visual presentation of the prime word for










one-third (4/12) of our participants. As shown in Table 4-1, of these four participants, subject 2

produced one phonological error in repetition, subject 3 made one semantic error in connected

speech, subject 4 produced a total of three errors in connected speech (two phonological and one

semantic error), and subject 6 was found to produce one phonological error in repetition and two

errors in connected speech (one phonological and one semantic). The influence of DBS on

language function in individuals with PD and DBS has received little attention other than with

verbal fluency tasks which "represent an almost exclusively applied index of linguistic

proficiency...in this population (Whelan, Murdoch, Theodoros, Silburn, & Hall, 2005, p. 93).

Whelan et al. (2005) provide some of the only available data on the language specific effects of

STN DBS in two patients. Although language changes varied with time in both the positive and

negative directions, they primarily report changes in "divergent language production

proficiency" and "lexical-semantic manipulation skills" (Whelan, Murdoch, Theodoros, Silburn,

& Hall, 2005, p. 99). Further studies of the language effects of DBS and PD await completion.

The other 27% of errors produced included a variety of error types, including production

of a previous prime (6%), production of a previous target (6%), initial sound or syllable

repetition (5%), multiple errors (5%), no speech response (3%), and production of the prime in a

'switch' task (2%). Many of these error types can be hypothesized to occur at the level of motor

planning/programming. For example, errors such as production of a previous prime or target

both seem to reflect an inability to inhibit previous motor programs. Production of the prime in

the 'switch' condition occurred infrequently, but provides a direct example of disordered

switching of speech motor programs. Van der Merwe (1997) suggests that errors such as initial

sound repetitions (i.e., neurogenic dysfluency) may also be due to deficits at the level of motor

planning/programming.










colleagues (1999, 2000; 2001; see also Pinto, Gentil, Fraix, Benabid & Pollak, 2003) merit

further scrutiny for those interested in the possible influence of DBS on speech motor

planning/programming in PD.

Gentil and colleagues (1999) investigated the effect of bilateral STN DBS on speech and

nonspeech oromotor function in 10 patients with PD and 14 healthy controls. Patients were

tested off medication in both 'on' and 'off stimulation conditions. The 'off stimulation

condition assessment occurred 1 hour after discontinuing stimulation. Perceptual and kinematic

measurement approaches were utilized. Perceptual measurement was limited to using the score

from item 18 of the Unified Parkinson's Disease Rating Scale (UPDRS). Kinematic assessment

procedures utilized force transducers to determine ramp-and-hold force contractions and

maximal strength of the lips and tongue. Kinematics revealed a number of improvements in lip

and tongue function 'on' stimulation, including increased maximal strength, increased accuracy

in reaching a target, increased precision during the hold phase, and decreased RT. Results of

perceptual assessment also revealed improved speech function in the 'on' versus 'off

stimulation condition. It is the kinematic data which is most scientifically rigorous and which is

of primary interest for the present discussion on speech motor planning/programming. These

data which improvements in RT 'on' as compared to 'off STN DBS for a nonspeech oromotor

movement suggest that processes involved in speech motor planning/programming may be

positively influenced by DBS. Indeed, the measure ofRT (as discussed in detail later in this

chapter) is an accepted measurement technique to quantify preparatory aspects of movement

(e.g., motor planning/programming) and this study may reflect a pioneering effort to measure

non-execution level speech changes with DBS in individuals with PD.








Digital
Audio
Recorder


U


Head-Set Mic


Signal
Conditioner


Serial
Laptop Response
Computer Box /


Figure 2-3. Equipment configuration


Monitor


K


I


r










Research Question 2

Are there significant differences in SRT in the 'on' versus 'off' DBS state when

participants with PD produce a word in the 'no switch' condition?

It was predicted that participants will have improved SRT in single words in the 'no

switch' condition when 'on' versus 'off DBS. This was expected due improved maintenance of

the speech motor program in the 'on' stimulation condition.

Research Question 3

Are there significant differences in SRT in the 'on' versus 'off' DBS state when

participants with PD produce a word in the 'switch' condition?

It was predicted that SRT will be improved in single words in the 'switch' condition in

the 'on' versus 'off' stimulation condition. This was expected due to improved ability to switch

speech motor programs in the 'on' stimulation condition.

Secondary and Exploratory Aims

The secondary aims of this study were to determine the effects of the experimental

manipulations on speech response accuracy. It was expected that response accuracy would differ

in response to priming condition. That is, more errors were anticipated in the 'switch' versus the

'no switch' condition. No difference in response accuracy was expected in response to DBS

state. For exploratory purposes, measures of neuropsychological performance (i.e., verbal

fluency and response inhibition) were conducted both 'on' and 'off DBS.










CHAPTER 2
MATERIALS AND METHODS

Experiment Overview

A reaction time (RT) paradigm was employed to measure the effects of DBS on

maintaining and switching of speech motor programs in individuals with PD and hypokinetic

dysarthria. Double blind testing was completed in the 'on' and 'off DBS states using a response

priming procedure in which participants were provided with a prime to supply information

regarding target. Over a series of targets, the prime word was followed with a high probability by

the primed target as expected ('no-switch' condition) or with a low probability by an unexpected

target word ('switch' condition). The task of the subjects was to speak the target word aloud as

quickly and accurately as possible. The primary dependent measure was SRT.

Participants

Subjects

Twelve participants with PD and DBS completed the study. Subjects were recruited

through the Movement Disorders Clinic at the University of Florida (UF) and the Speech and

Hearing Center at UF based on the following criteria:

Inclusion Criteria

Inclusion criteria included patients age 25 85 years old with a diagnosis of "probable"

idiopathic PD as determined by a neurologist with expertise in the evaluation of movement

disorders, six months to two years status-post unilateral left or bilateral GPi or STN DBS,

medically optimized and stable on anti-PD and psychotropic medications for at least 30 days at

the time of the screening visit, the ability to read words and sentences aloud, and completion of

the informed consent to participate in the study.










Assessment of linguistic competency during repetition and connected speech suggested

intact linguistic systems in eight of 12 (75%) participants. Four subjects produced a total of eight

errors during repetition (n = 2) or connected speech (n = 6). Five errors were determined to be

phonological and 3 were semantic.

Experimental Results

Table 4-2 presents the summary statistics for both the primary and secondary research

questions. The mean and standard deviation for SRT and response accuracy are shown across

priming (i.e., 'no switch' or 'switch') and stimulation (i.e., DBS 'on' or 'off) conditions. Mean

SRT in the 'no switch' condition was 615.24 ms (SD = 96.77) 'on' DBS and 671.38 ms (SD

=113.05) 'off DBS. Mean SRT in the 'switch' condition was 717.09 ms (SD= 89.11) 'on' DBS

and 728.67 ms (SD = 98.35) 'off DBS. Mean number of errors (per 16 responses) in the 'no

switch' condition was 0.69 (SD = 0.85) 'on' DBS and 1.06 (SD = 0.74). Mean number of errors

in the 'switch' condition was 1.50 (SD = 1.17)'on' DBS and 1.50 (SD = 1.98)'off DBS.

Primary Aims

Research question 1 ('no-switch' vs. 'switch')

Statistical significance was set at the 0.05 level for all analyses performed. Table 4-3

shows mean difference and p-values for all comparisons. Separate Wilcoxon signed-rank tests

were conducted which revealed statistically significant differences in SRT in the predicted

direction between 'switch' and 'no-switch' conditions in both the 'on' DBS (signed rank = 1, p =

0.0010) and 'off DBS (signed rank = 12, p = 0.0342) states. Furthermore, Fisher's combination

method revealed significant differences in SRT overall across DBS conditions (Fisher's

combination test statistic = 20.57, p = 0.0040). In other words, subjects produced a speech

response more quickly in the 'no-switch' versus 'switch' condition, regardless of whether DBS










Desbonnet, L., Temel, Y., Visser-Vandewalle, V., Blokland, A., Hornikx, V., &
Steinbusch, H. W. (2004). Premature responding following bilateral stimulation of the rat
subthalamic nucleus is amplitude and frequency dependent. Brian Research, 1008, 198-204.

Dirnberger, G., Reumann, M., Endl, W., Lindinger, G., Lang, W., & Rothwell. J. C.
(2000). Dissociation of motor preparation from memory and attentional processes using
movement-related cortical potentials. Experimental Brain Research, 135, 231-240.

Donovan, N. J., Velozo, C. A., & Rosenbek, J. C. (in press) The Communicative
Effectiveness Survey: Investigating its item-level psychometrics. Journal of Medical Speech
Pathology.

Draper, I. T., & Johns, R. J. (1964). The disordered movement in parkinsonism and the
effect of drug treatment. Bulletin of the Johns Hopkins Hospital, 115, 465-480.

Dromey, C., Kumar, R., Lang, A. E., & Lozano, A. M. (2000). An investigation of the
effects of subthalamic nucleus stimulation on acoustic measures of voice. Movement Disorders,
15, 1132-1138.

Duffy, J. R. (2005). Motor speech disorders: Substrates, differential diagnosis, and
management (2nd edition). St. Louis: MO: Elsevier Mosby.

Ellis, C., Okun, M. S., Gonzalez-Rothi, L. J., Crosson, B., Rogalski, Y.,
& Rosenbek, J. C. (2006). Expressive language use after PD: Deficits in use but
not form. Movement Disorders, 21, 97-98.

Ellis, C., & Rosenbek, J. C. (2007). The basal ganglia and expressive language: A review
and directions for future research. Communicative Disorders Review, 1, 1-15.

Esselink, R. A. J., de Bie, R. M. A., de Haan, R. J., Lenders, M. W. P. M., Nijssen, P.C.
G., Staal, M. J., et al. (2004). Unilateral palidotomy vs bilateral subthalamic nucleus stimulation
in PD: A randomized trial. Neurology, 62, 201-207.

Evarts, E. V., Teravainen, H., & Calne, D. B. (1981). Reaction time in Parkinson's
disease. Brain, 104, 167-186.

Gale, J. T., Amirnovin, R., Williams, Z. M., Flaherty, A. W., & Eskander, E. N. (in
press). From symphony to cacophony: Pathophysiology of the human basil ganglia in Parkinson
disease. Neuroscience and Biobehavioral Reviews.

Garavan, H., Ross, T. J., & Stein, E. A. (1999). Right hemispheric dominance of
inhibitory control: An event-related functional MRI study. Proceedings of the NationalAcademy
of Sciences of the United States of America, 96, 8301-8306.










LIST OF ABBREVIATIONS

CES Communicativeness Effectiveness Survey

CMA Cingulate motor area

DBS Deep brain stimulation

GPe Globus pallidus pars externa

GPi Globus pallidus internus

IC Internal capsule

MC Motor cortex

MMSE Mini-Mental State Examination

PD Parkinson's disease

PET Positron emission tomography

PMC Premotor cortex

RT Reaction time

SIT Sentence Intelligibility Test

SLPs Speech-language pathologists

SMA Supplementary motor area

SNpc Substantia nigra pars compact

SNpr Substantia nigra pars reticulata

SRB Serial Response Box

SRT Speech reaction time

STN Subthalamic nucleus

UF University of Florida

UPDRS Unified Parkinson's Disease Rating Scale

WMS-III Wechsler Memory Scale 3rd Edition










These RT data have been replicated by Gentil and colleagues. In 2000, they further

investigated the influence of bilateral stimulation in 10 patients with STN DBS and used similar

experimental procedures as described above. Improved RT in the 'on' STN DBS condition for a

nonspeech oromotor movement was again reported. This group of researchers (Pinto, Gentil,

Fraix, Benabid, & Pollak, 2003) continued to use their experimental protocol in their largest

group to date 26 patients with PD and bilateral STN DBS and improved RT was again found

in the 'on' stimulation condition.

A Model of Speech Motor Control

Numerous models of motor control which share many features including concepts

consistent with the concept of motor programs have been proposed, including those from

Schmidt (1975) and Brooks (1986). Speech specific models are also available, such as Van der

Merwe's (1997) proposed four-phase framework.

Four Phase Model

Van der Merwe (1997) has proposed a model for speech motor control based on four

phases: linguistic-symbolic planning, motor planning, motor programming, and execution. Let us

consider each phase of the model in greater detail, with particular attention to motor planning

and motor programming.

Linguistic Symbolic Planning

Linguistic symbolic planning is considered to be a non-motor stage during which the

intent to communicate originates due to individual behavioral needs and environmental demands.

A message in compiled during this level of processing, requiring semantic, syntactic,

morphological, and phonological planning. For example, phonological planning involves

selection and combination of phonemes "in accordance with the phontoactic rules of the










Table 2-3. Powers corresponding to r, ratio of mean SRT between the 'on' and 'off stimulation
conditions, based on sensitivity analysis.



R 1.05 1.06 1.07 1.08 1.09 1.1

Power











Pillon, B., Ardouin, M. A., Damier, P., Krack, P., Houete, J. L., Klinger, H., et al. (2000).
Neuropsychological changes between "off" and "on" STN of GPi stimulation in Parkinson's
disease. Neurology, 55, 411-418.

Pinto, S., Gentil, M., Fraix, V., Benabid, A.-L., & Pollak, P. (2003). Bilateral subthalamic
stimulation effects on oral force control in Parkinson's disease. Journal of Neurology, 250,
179-187.

Pinto, S., Thobois, S., Costes, N., Le Bars, D., Benabid, A.-L., Broussole, E., et al.
(2004). Subthalamic nucleus stimulation in dysarthria in Parkinson's disease: A PET study.
Brain, 127, 602-615.

Robertson, C., & Flowers, K. A. (1990). Motor set in Parkinson's disease. Journal of
Neurology, Neurosurgery, and Psychiatry, 53, 583-592.

Rodriguez-Oroz, M. C., Obeso, J. A., Lang, A. E., Houeto, J.-L., Pollak, P., Rehncrona,
S., et al. (2005). Bilateral deep brain stimulation in Parkinson's disease: A multicentre study with
4 years follow-up. Brain, 128, 2240-2249.

Romero, D. H, Van Gemmert, A. W., Adler, C. H., Bekkering, H., & Stelmach, G. E.
(2003). Altered aiming movements in Parkinson's disease patients and elderly adults as a
function of delays in movement onset. Experimental Brain Research, 151, 249-261.

Romito, L. M., Scerrati, M., Contarino, M. F., lacoangeli, M., Bentiboglio, A. R., &
Albanese, A. (2003). Bilateral high frequency subthalamic stimulation in Parkinson's disease:
Long-term neurological follow-up. Journal of Neurological Sciences, 43, 119-128.

Rothlind, J. C., Cockshott, R. W., Starr, P. A., & Marks, Jr., W. J. (2007).
Neuropsychological performance following staged bilateral pallidal or subthalamic nucleus deep
brain stimulation for Parkinson's disease. Jouraml of the International Neuropsychological
Society, 13, 68-79.

Rousseaux, M., Krystkowiak, P., Kozlowski, O., Ozsancak, C., Blond, S., & Destee, A.
(2004). Effects of subthalamic nucleus stimulation on parkinsonian dysarthria and speech
intelligibility. Journal ofNeurology, 251, 327-334.

Roy, E. A., Saint-Cyr, J., Taylor, A., & Lang, A. (1993). Movement sequencing disorders
in Parkinson's disease. International Journal ofNeuroscience, 73, 183-194.

Rubchinsky, L. L., Kopell, N., & Sigvardt, K. A. (2003) Modeling facilitation and
inhibition of competing motor programs in basal ganglia subthalamic nucleus-pallidal circuits.
Proceedings of the National Academy of Sciences of the United States of America, 100,
14427-14432.










in patients with PD. The use of thorough washouts of DBS is also an important aspect of this

work. Many previous data are collected from studies with inadequate washouts and we utilized

two two-hour washouts before the DBS condition was implemented. The strict

inclusion/exclusion criteria allowed us to obtain a homogenous and representative sample of

individuals with PD. Finally, our use of a SRT paradigm allowed us to measure the primary

variable of interest using an instrumental, objective measurement tool.

Weaknesses of the study are also present. The small sample size (n = 12) is a clear

limitation which highlights the preliminary nature of our findings. However, the sample size

appeared to be sufficient to answer the primary research questions. The strength of the study is

also compromised by the fact that the exact implantation site is unknown in 75% (9/12) of

participants. Although all participants underwent unilateral or staged bilateral procedures, three-

quarters of participants were recruited from a larger surgical trial which seeks to compare the

effects of GPi and STN DBS in a blinded fashion. Although the exact surgical sites will be

known upon completion of the larger trial, at the present, determining differential effects of GPi

or STN DBS on SRT is impossible. Our incomplete knowledge regarding the exact processes

involved in speech motor planning/programming and their neurophysiological correlates is

another limitation. For example, the anatomical locations of the speech sensorimotor store and

buffer have not been established experimentally. Another limitation of the study is the quality of

the digital recordings obtained during the appropriate portions of the screening and test sessions.

These recordings were sufficient for the purposes of the present study, which include aiding in

the description of study participants by determining linguistic competency, intelligibility scores,

and speech diagnosis and severity, as well as for determining response accuracy. However, due

to the fact that the recordings were made during screening and testing sessions conducted at










EFFECTS OF DEEP BRAIN STIMULATION ON SPEECH MOTOR PLANNING/
PROGRAMMING IN PATIENTS WITH PARKINSON' S DISEASE


















By

HARRISON N. JONES


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2007










sensorimotor store and loaded the motor program into the buffer. This task requires several

processes to occur prior to movement execution. The incorrect motor program in the buffer must

be inhibited, the correct motor subprograms must be retrieved from the store, and the motor

program must be loaded into the buffer. Due to these additional processes and the increased

complexity of the 'switch' versus 'no switch' condition, RT will be increased for these tasks.










responses, with a standard deviation of 0.0805. If it assumed that the differences in log-

transformed response latency have the same distribution in our study for each stimulation

condition, then the Wilcoxon signed-rank sum test at the 0.05 Type-I error level has sufficient

power to detect differences between 'switch' and 'no switch' speech responses as shown in

Table 2-2.

On the other hand, for differences between 'on' and 'off stimulation conditions, we did

not have an estimate of effect size. Thus, a sensitivity analysis with regard to difference

magnitude was completed with the assumption that the sample size would be 15 to test the

hypothesis using the Wilcoxon signed-rank sum test at the 0.05 Type-I error level. Table 2-3

provides powers corresponding to r, ratio of mean SRT between the 'on' and 'off stimulation

conditions. This table suggests that even if the difference between the 'on' and 'off stimulation

conditions is only half of the difference between the 'switch' and 'no switch' responses observed

in the Spencer and Rogers study from 2005, we will still have reasonable power.











APPENDIX B
EXPERIMENTAL STIMULI


Jump-JUMP


Zipper-ZIPPER


2 Shopper-SHOPPER I I
3 Shining-SHINING
4 Jealous-JEALOUS Thug-THUG
5 Shake-SHAKE Shake-SHAKE
6 0 Jealous-JEALOUS
7 Zipper-ZIPPER Jump-JUMP
8 Ramp-RAMP Veal-VEAL
9 Jingle-JINGLE Rowing-ROWING
10 Jade-JADE
11 Cheer-CHEER
12 Thug-THUG
13 Rowing-ROWING Cheer-CHEER
14 Veal-VEAL Jingle-JINGLE
15 Chatter-CHATTER spo"
16 Shining-SHINING Chatter-CHATTER
17 Jump-JUMP Sheep-SHEEP
18 Jealous-JEALOUS Zipper-ZIPPER
19 Shake-SHAKE Ramp-RAMP
20 Jade-JADE Liver-LIVER


21 Sheep-SHEEP
22
23 Liver-LIVER
24 Veal-VEAL
25_
26
27 Ramp-RAMP
28 Jade-JADE
29 Jingle-JINGLE
30
31 Rowing-ROWING
32 Cheer-CHEER


Jumo-JUMP


I Rowing-ROWING


33 Shining-SHINING Thug-THUG
34 Thug-THUG
35 Shake-SHAKE
36 Shopper-SHOPPER Veal-VEAL
37 -Cheer-CHEER
38 Jump-JUMP Liver-LIVER

40 Liver-LIVER Shopper-SHOPPER
41 Sheep-SHEEP Jingle-JINGLE
42 Chatter-CHATTER Shining-SHINING
43 *ealowing-ROWING
44 Sheep-SHEEP
45 Jade-JADE
46 Ramp-RAMP
47 Jade-JADE
48 Zipper-ZIPPER
49 Shake-SHAKE Jealous-JEALOUS
50 Chatter-CHATTER Chatter-CHATTER
51 Rowing-ROWING Veal-VEAL
52 Thug-THUG
53 Veal-VEAL Shopper-SHOPPER
54 Liver-LIVER
55 Shopper-SHOPPER Cheer-CHEER
56 Liver-LIVER Ramp-RAMP
57 Jingle-JINGLE Sheep-SHEEP
58 Shining-SHINING
59 Thug-THUG Jump-JUMP
60 Zipper-ZIPPER Shake-SHAKE
61 Sheep-SHEEP
62 Jealous-JEALOUS Chatter-CHATTER
63 Ramp-RAMP Jade-JADE
64 Cheer-CHEER Jingle-JINGLE


= Switch = No switch















Tur DBS ON
(11:00 a.m.)
N=6


Tum DBS OFF
DBS Washout 1
(9:00 a.m.)



Randomization


SKeep DBS OFF
(11:00 a.m.)
N=6


Test session 1: Response Priming Test, FAS, Stroop
(11:30 a.m.)


DBS OFF for all subjects
(12:00 a.m.)


Test session 2: Response Priming Test, FAS, Stroop
(2:30 p.m.)





DBS ON for all subjects
(3:00 p.m.)


Figure 2-1. Typical experimental timeline.


2













memory" associated with STN DBS (along with "moderate declines...in semantic and phonemic

verbal fluency" (Parsons et al., 2006, p. 578). Studies which attempt to determine the effects of

the 'on' and 'off DBS state are emerging. Jahanshahi et al. (2000) reported that bilateral GPi

and STN DBS improved Stroop control trial performance 'on' versus 'off DBS, while the 'on'

STN DBS state worsened performance on the interference portion of the Stroop. Pillion and

colleagues (2000) found that bilateral STN DBS improved performance in the word and the color

portions of the Stroop, though more errors were noted 'on' DBS in the interference condition of

the Stroop color test. Comparing the exact findings between these two studies and our own is

made challenging by the fact that each study used a different version of the Stroop. Insufficient

power does not appear to explain our nonsignificant findings as Pillon et al. found significant

group differences in a similar group of 13 subjects (six with GPi DBS & with seven with STN

DBS). Regardless, many more data are needed understand he effects of DBS state on measures

of cognitive function, including response inhibition.

Strengths/Weaknesses

This study has several strengths which allow it to make a contribution to the

understanding of the speech effects of DBS in PD, particularly at the level of motor

planning/programming. The experimental design was rigorous and controlled for many threats to

internal and external validity. Particular strengths of the design include the use of double blind

testing, a thorough washout for both stimulation and medication effects, rigid inclusion/exclusion

criteria, and the use of an objective measurement approach for determining speech effects (i.e.,

SRT).

These strengths are not insignificant. To our knowledge, double blind testing has not

previously been conducted in the literature which has been focused on the speech effects of DBS










Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

EFFECTS OF DEEP BRAIN STIMULATION ON SPEECH MOTOR PLANNING/
PROGRAMMING IN PATIENTS WITH PARKINSON' S DISEASE

By

Harrison N. Jones

December 2007

Chair: John C. Rosenbek
Major: Rehabilitation Science

The primary purpose of this study was to measure the effects of deep brain stimulation

(DBS) on maintaining and switching speech motor programs in individuals with Parkinson's

disease (PD) and hypokinetic dysarthria. Recent literature suggests that at least a portion of the

underlying mechanism of hypokinetic dysarthria in individuals with PD may be related to

deficits in speech motor planning/programming, including maintaining and switching motor

programs (Spencer & Rogers, 2005; Van der Merwe, 1997). Although the effects of DBS on

speech motor planning/programming have not been previously explored, DBS has been shown to

have a positive influence on these processes in the limbs and we posited that DBS would

similarly benefit speech maintenance and switching.

A reaction time paradigm was employed to measure the effects of DBS on maintaining

and switching of speech motor programs in individuals with PD. Double blind testing was

completed in the 'on' and 'off DBS conditions using a response priming procedure in which

participants were provided with a prime word to supply information regarding target word. Over

a series of targets, the prime was followed with a high probability by the primed target as










and/or debate were required, although these final consensus judgments were used to determine

accuracy of each response. Types of errors included the following:

Production of the prime Responses in which participants do not "switch" their speech response.

For example, "shopper-chopper" is the prime-target pair and the production is "shopper" rather

than "chopper". Subjects may self-correct.

Partial production of prime Responses in which participants partially produce the prime word

such as the first sound or syllable. For example, the prime-target pair is "shopper-chopper" and

the subject responds "sh... chopper". Subjects may self-correct.

Initial sound repetition Responses in which participants repeat the initial sound or syllable of

the target word. For example, in the case of the word pair "shopper-shopper", the subject

responds "sh... shopper".

Production of previous target Responses in which the subject produces a previously presented

target.

No response Subject does not produce a speech response upon command.

Phonological error Single phoneme omission, deletion, substitution, transposition, etc.

Lexical/semantic error Responses in which participants substituted a whole real word for the

target word.

Multiple errors Two or more errors from the above list were combined in a single

response.

Reliability

Intra-rater Reliability

Judge one completed perceptual assessment to determine the accuracy of each response on two

separate occasions. Point-by-point analysis was conducted for 100% of each subject's responses










Another alternative explanation for these findings is that the differences we found are due

to execution level speech deficits, rather than the deficits in speech motor

planning/programming. This explanation does not appear plausible, as the primary dependent

measure of SRT was calculated before speech was executed. The response accuracy data also

support deficits primarily at the level of motor planning/programming, although deficits at the

level of execution also appeared to be present. This notion is based on the perceptual assessment

that sound distortions occurred during production of words in the response priming test on 4% of

all trials. Sixty-five sound distortions were perceived in 1,526 trials, 38 while subjects were 'on'

DBS and 27 'off DBS. Although Van der Merwe (1997) suggests that distortions may occur due

to programming level deficits, it is traditional to attribute this type of error to execution level

deficits. Therefore, it appears plausible that the hypokinetic dysarthria of our subjects was

influenced by deficits in both planning/programming and execution.

Discussion Conclusion

We conducted an experiment in subjects with DBS and PD in which participants

completed a response priming protocol in two priming conditions (i.e., 'switch' and 'no switch')

both 'on' and 'off DBS. SRT was found to be significantly different across the priming

conditions in that subjects produced a word more quickly in the 'no switch' versus the 'switch'

condition. Our participants were also found to produce a word more quickly in the 'no switch'

condition when 'on' versus 'off DBS. The proposed mechanism of this improvement is an

increased ability to maintain the motor program in the buffer prior to the command for execution.

SRT was not significantly different in the 'switch' condition across DBS states, suggesting that

DBS has little influence on mechanisms involved in switching of speech motor programs (i.e.,










include "slow, struggling speech with distortion and even apparent substitution" such as

associated with apraxia of speech (p. 17).

Motor Programming

It should be acknowledged that differentiation between speech motor planning and motor

programming is extremely challenging and has not, to our knowledge, been demonstrated

experimentally. However, during this hypothesized level of speech production, motor programs

are finally selected and sequenced for movements of the necessary muscles for speech

production. Information regarding "spatio-temporal and force dimensions such as muscle tone,

rate, direction, and range of movements" is controlled by muscle specific programs influenced

by external feedback (Van der Merwe, 1997, p. 7). Motor programming involves multiple neural

areas including the basal ganglia, SMA, lateral cerebellum, and MC. Speech change associated

with deviant speech motor programming is hypothesized to include "sound distortion, defects in

speech rate, and/or problems in the initiation of movement" (Van der Merwe, 1997, p. 18).

Execution

At the level of execution, the plans and programs of the previous phases in speech

production result in muscle movements and speech production. Feedback to higher levels in the

motor system is an integral part of this phase, particularly during development. Feedback may

take the form of tactile-kinematic and acoustic information, for example. Neural areas involved

in execution include the "motor cortex, the lower motor neurones, peripheral nerves, and motor

units in the muscles" (Van der Merwe, 1997, p. 16). Areas involved with preparatory aspects of

speech movement (i.e., motor planning and motor programming) are also active during

execution, including cortical (SMA) and subcortical regions (the basal ganglia and thalamus).

Although not directly addressed by this model, execution level errors, for example, are often










model proposed by Sternberg and colleagues (1978, 1980, 1988 & 1990), the hypothesized levels

of processing that must occur in the 'switch' condition before movement can be executed include

inhibition of the unwanted motor program held in the buffer, retrieval of the appropriate

subprograms from the sensorimotor store, and the loading of a new motor program into the

buffer. When results from research questions 2 and 3 are both considered, the data suggest that

DBS improves performance at the level of the buffer, but not retrieval of subprograms from the

store and/or inhibition of the unwanted motor program.

Secondary Aims

The secondary aims of this study were to determine the effects of the experimental

manipulations on response accuracy. Overall, errors were infrequently encountered. We

anticipated that response accuracy would be significantly decreased in the 'switch' versus 'no

switch' condition. Contrary to this expectation, response accuracy was not significantly

influenced by the priming condition (p = 0.15). When collapsed across DBS conditions, subjects

produced a mean of 1.50 errors per 16 responses in the 'switch' condition compared to 0.88

errors in the 'no switch' condition. Consistent with our expectations, stimulation condition also

did not significantly influence response accuracy in either priming condition. In the 'no switch'

condition, subjects produced a mean of 0.69 errors per 16 responses 'on' stimulation and 1.06

errors 'off DBS (p = 0.18). In the 'switch' condition, subjects produced a mean of 1.50 errors

per 16 responses both 'on' and 'off DBS (p = 0.72).

These non-significant differences in response accuracy are likely due to a lack of power

due to the low number of errors observed. Quite simply, errors did not occur frequently enough

or there were an insufficient number of stimuli to elicit a sufficient number of errors to find

significant group differences. Of note, power analysis was conducted based on SRT data rather










Pinto et al. (2004) further suggest that the patterns of abnormal cortical activation in patients

with PD and hypokinetic dysarthria appear to normalize with improvements in speech when 'on'

STN stimulation. Other research also indicates that STN DBS has some benefit for speech

(Dromey, Kumar, Lang, & Lozano, 2000; Hoffman-Ruddy, Schulz, Vitek & Evatt, 2001;

Rousseaux, Krystkowiak, Kozlowski, Ozsancak, Blond, & Destee, 2004), though speech changes

may be dependent on parameters of stimulation being applied (Tornqvist, Schalen, & Rehncrona,

2005) or the hemisphere be stimulated (Santens, De Letter, Van Borsel, De Reuck & Caemaert,

2003; Wang, Metman, Bakay, Arzbaecher,,& Bernard, 2003). However, dysarthria is also

commonly reported as an adverse event in many surgical trials (The deep brain stimulation for

Parkinson's disease study group, 2001; Esselink et al., 2004; Krack et al., 2003; Kumar et al.,

1998a, b; Ostergaard, Sunde, & Dupont, 2002; Rodriguez-Oroz et al., 2005; Romito et al., 2003;

Schupback et al., 2005; Thobois et al., 2002) and substantial improvement in speech would not

be a surgical goal in most cases.

The aforementioned studies designed to evaluate the speech effects of surgery have

primarily targeted measurement at the execution phase of movement, rather than preparatory

motor processes such as planning/programming (see discussion of Models of Motor Control

below). This is not surprising considering that the hypokinetic dysarthria encountered in PD has

traditionally been conceptualized as a disorder of execution level processes (Darley, Aronson, &

Brown, 1969a, b, 1975; Duffy, 2005; Yorkston, Beukelman, Strand, Bell, 1999). However,

possible contribution of speech motor planning/programming deficits in PD are being

increasingly recognized (Spencer, 2006; Spencer & Rogers, 2005; Van der Merwe, 1997),

though little work has targeted processes involved in speech motor planning/programming to

determine the possible influence of DBS. However, upon close inspection, data from Gentil and










cortex, and the anterior cingulate/medial orbitofrontal cortices. This arrangement allows the

basal ganglia to influence multiple diverse processes of the frontal lobes including movement,

behavior, cognition, language, and emotion (Alexander, DeLong & Strick, 1986). Indeed, as

stated by Murdoch (2004), thishs anatomical arrangement (allows) the output from the basal

ganglia (to gain) access to multiple areas of the frontal lobe.., and provides a basic

neuroanatomical mechanism whereby these subcortical structures can influence aspects of

behaviour, cognition and language as well as motor function" (p. 235).

The focus of the present paper is on preparatory aspects of movement, specifically speech

motor planning/programming. Van der Merwe's four phase model of speech production (1997;

see full discussion below) suggests that many of the aforementioned neural regions are involved

in these activities. Motor planning, for example, is suggested to be accomplished largely in PMC

and SMA (i.e., the motor association area), while motor programming is thought to comprise

areas such as the basal ganglia, cerebellum, SMA, and MC.

Circuitry in PD

Although the exact role of dopaminergic input to the striatum is not completely understood

(Mink, 1996), the degeneration of dopaminergic cells in SNpc is a hallmark feature of PD

(Bergman & Deuschl, 2002). Gale and colleagues (in press) state the at the neurophysiological

level, loss of dopaminergic neurons in PD leads to the clinically observed manifestations of the

disease due to derangements of firing rates, neuronal selectivity, and the firing patterns of (basal

ganglia) neurons" (p. 2). Figure 3-3 illustrates the effect of PD on the circuitry of the basal

ganglia as shown in the 'standard model'. In the indirect pathway, loss of striatal dopamine leads

to excessive inhibition of GPe, leading to decreased inhibition of STN and the delivery of

excessive excitatory drive to the basal ganglia output nuclei (i.e., GPi and SNpr). This is










LIST OF FIGURES


Figure page

2-1 T typical experim mental tim line ........................................ .............................................29

2-2 R response prim ing protocol ........................................................................ .................. 30

2-3 E quipm ent configuration ......................................................................... ....................3 1

3-1 Basal ganglia structures and surrounding areas ..................................... ............... 53

3-2 The intrinsic and extrinsic circuitry of the basal ganglia under normal conditions ..........54

3-3 The intrinsic and extrinsic circuitry of the basal ganglia in Parkinson's diseease.............55

3-4 U nilateral deep brain stim ulation (DB S) ........................................ ....................... 56

4-1 Mean speech reaction time (SRT) in milliseconds (ms) in the 'no switch' and switch'
conditions ............... ...... .... ............. ..........................................65

4-2 Mean speech reaction time (SRT) in milliseconds (ms) in the 'no switch' and
'sw itch' conditions 'on' and 'off D B S ........................................ ......................... 66

5-1 Distribution of errors differentiated among error type ............................................... 85










than response accuracy and a low number of errors are typically encountered in RT experiments,

so perhaps this finding is not surprising.

Once again, the limb literature in on RT in PD may be informative, particularly with

regard to the lack of effects of DBS on response accuracy. For example, Schubert et al. (2002)

report no statistical difference in response accuracy across stimulation conditions in a variety of

RT paradigms, including a visual simple RT task, a visual choice RT task, and an auditory

choice RT task. Other researchers have reported similar findings in simple (Temel et al., 2006)

and choice RT (van den Wildenberg et al. 2006) experiments in individuals with PD and DBS.

These previous studies, along with our present findings, suggest that DBS may have little

influence on response accuracy in simple and choice RT experiments. Again, this is likely

explained by the relatively low frequency of errors encountered in RT experiments. For example,

we found an error rate of 6.5% on all trials completed by our study participants. Temel and

colleagues (2006) report errors occurred in "about 5%" of trials, while Van den Wildenberg et al.

(2006) found a 2% error rate. Due to the low occurrence of errors, the current study was likely

underpowered to detect differences in response accuracy.

Our participants made ninety-nine errors in a total of 1,526 trials. Our error rate of 6.5%

is similar to the error rate of 6.4% reported by Spencer and Rogers (2005) in their participants

with dysarthria (both hypokinetic and ataxic types). Of course, as previously discussed in

Chapter 3, the response priming procedure we utilized was based on this and subsequent work

(Spencer, 2006), so the consistency of error rates in subjects with dysarthria is reassuring. Figure

5-1 shows the distribution of errors in the present study. Out of the 99 total errors, 64 (65%) were

either a premature response (32) or a phonologic error (32). The other 35% included the

following error types: lexical/semantic (7), production of a previous prime (6), production of a










1000ms. The prime was followed by a blank screen for 250ms and then the target word was

presented for 1000ms. Response latency was measured temporally from the presentation of the

target. Response accuracy was determined using broad phonetic transcription scored online with

later verification and analysis of errors. Further details can be found under the "Scoring" and

"Equipment" sections of this chapter.

Subjects were also administered two neuropsychological tests during each of the two test

sessions: the FAS test of verbal fluency and the Stroop. Color and Word Test. During the FAS,

participants were provided with one-minute to produce as many words as possible starting with

each of the three letters F, A, and S (Benton & Hamsher, 1976). The Stroop has three sets of

stimuli: color words printed in black ink, symbols (i.e., X) printed in color ink, and color words

from the first set of stimuli printed with incongruous colors from the second set of stimuli.

Subjects were asked to move through each set of stimuli reading words or naming colors as

quickly as possible (Golden & Freshwater, 2002).

Training session

Prior to the first test session, participants were trained in the experimental task for

approximately 10 minutes (see Appendix A for the training session stimuli). Following this

training period, if a subject was unable to perform the experimental task, it was planned to

discontinue testing, though this did not occur.

Stimuli

The prime and target words consisted of one- and two-syllable words (see Appendix B)

from Spencer (2006) (adapted from Spencer & Rogers, 2005). Word rimes were maintained from

prime to target. It was required that the onset of the prime needed to share two features with the

target and be a highly marked phoneme. A balance for word frequency was maintained (Spencer













Table 4-1. Individual and group descriptive data.


a es a a't a2 Sfea a eS

a!S
'tsE EsE a aasaxffi ^ I


a2
* 0> 1 ffl- P BO C <



I I I I 1 1 it




11111~ h11iJI
s Q e11111 o 0 EOf &


g


j 3!



( N











rM v



t an r- co o t LA aO




^ 0 0 (0 (0 Q co Q a o Qa
2 0Q .8 .l .3 .l CO -J ..J .1 ...J

__ I
-1 (h0 tC U' (0 tllh ~VC~ )










Many of the structures of the basal ganglia, as well as other surrounding structures such as the

thalamus and internal capsule (IC), are shown in Figure 3-1.

Normal Basal Ganglia Intrinsic/Extrinsic Circuitry

The extrinsic and intrinsic circuitry of the basal ganglia is of course complex and will only

be briefly reviewed, with an emphasis on what some have begun to call the 'standard model'

(Gale, Amimovin, Williams, Flaherty, & Eskandar, in press) This model suggests a number of

separate, functionally, and anatomically segregated corticobasal ganglion circuits, each of which

has two pathways through the basal ganglia: the inhibitory direct pathway and the excitatory

indirect pathway (DeLong & Wichmann, 2007). As shown in Figure 3-2, under normal

conditions, the putamen receives excitatory input from multiple cortical areas, including the

portions of the motor cortex (MC), premotor cortex (PMC), supplementary motor area (SMA),

and cingulate motor area (CMA) (Alexander, DeLong & Strick, 1986; DeLong, 1990; DeLong &

Wichmann, 2007; Mink, 1996). Output of the putamen is thought to be neuromodulated through

the direct and indirect pathways. The direct pathway projects inhibitory signals to the output

nuclei, mainly GPi and SNpr. The indirect pathway sends inhibitory signals from the striatum

(primarily the putamen) to GPe and GPe in turn sends inhibitory projections to the STN. In

addition receiving inhibitory projections from GPe, the STN also receives excitatory projections

directly from the cortex. Finally, the STN projects excitatory signals to the main output nuclei

(GPi and SNpr) where both the direct and indirect pathways to converge to deliver a balance of

excitatory and inhibitory signals, resulting in tonic, rapid inhibitory GABAergic projections to

the thalamus (Hikosaka, 2007). This finally results in excitatory projections back to cortical areas

including MC, PMC, SMA, and CMA (Alexander, DeLong & Strick, 1986; DeLong, 1990;

DeLong & Wichmann, 2007), as well as the dorsolateral prefrontal area, lateral orbitofrontal










during each of their two test sessions to determine whether agreement regarding accuracy was

present. Intra-class correlation coefficients were determined.

Inter-rater Reliability

A point-by-point analysis was completed to compare the level of agreement between the

two judges as to whether each individual response was correct or incorrect. Intra-class

correlation coefficients were determined.

Data Analysis

Average response latency and number of errors were calculated for each participant by

the stimulation condition ('on' and 'off DBS) and by the prime-target relationship ('switch' and

'no switch' responses). Summary statistics are provided by the stimulation condition and by the

prime-target relationship.

Furthermore, formal statistical inferences were conducted using nonparametric

procedures. Separate Wilcoxon signed-rank tests were used to determine whether significant

differences were present between the 'on' and 'off stimulation conditions for 'switch' and 'no

switch' speech responses in terms of SRT and response accuracy. Fisher's combination method

was used to further analyze the data and combine results from the 'on' and 'off stimulation

conditions and the 'switch' and 'no switch' test conditions to obtain overall p-values.

Sample Size and Power Consideration

Power analysis was conducted based on data for participants with hypokinetic dysarthria

published by Spencer and Rogers (2005). As the study population and experimental stimuli vary

between the two experiments, a conservative approach was utilized whenever possible. The

mean difference in log-transformed response latency (called speech reaction times in the

published paper) is 0.0966 unit larger (i.e., 10% longer) for 'switch' versus 'no switch' speech










unable to reliably make these measurements due to the presence of extraneous noise in the

acoustic signal. This was presumably due to the environments the digital recordings were made

(i.e., subject's homes and clinical environments). Although this was convenient for participants

and aided in recruitment, it would be a significant improvement to make acoustic recordings in a

sound treated room to more purely capture the speech signal. This would also provide the high-

quality digital recordings necessary for acoustic analysis of the speech signal which would allow

for additional insights at the level of execution. Although acoustic analysis of the speech effects

of DBS in subjects with PD has been completed previously (Dromey et al., 2000; Hoffman-

Ruddy et al. 2001, Wang et al., 2003), these studies suffer methodological limitations such as

unspecified stimulation washouts and small sample sizes.

Another interesting area for future study would be modification of the paradigm in order

to compare limb planning/programming with speech planning/programming. This is important

because of the differential responses across the corticospinal and corticobulbar systems to

treatments for PD (e.g., DBS, levodopa) that are commonly reported in the literature. However,

data from Adams and colleagues (2004) suggest that the reported differential response of these

systems to levodopa, for example, may be due to differences in the measurement approaches

used rather than true differences. If appropriately modified, the employed experimental paradigm

would allow for comparisons across these two systems using the same measurement approach

(i.e., RT). Such an approach could facilitate further understanding of how treatments for PD

influence different movements.

Finally, overall, the participants in our study were judged to have only mild-moderate

dysarthria on average. Only two of the 12 were on the more severe end of the severity spectrum

with moderate-severe dysarthria. This may have caused a ceiling effect in terms of speech











Thalamus


IC

Putamen

GPe

GPi

STN

4SN
















Figure 3-1. Basal ganglia structures and surrounding areas. IC = internal capsule, GPe = globus
pallidus pars externa, GPi = globus pallidus internus, STN = subthalamic nucleus, SN =
substantial nigra.










Marks, Jr., W. J. (2005). Programming thalamic and palatal deep brain stimulators for the
treatment of movement disorders (CD-ROM). 2005 American Academy ofNeurology.

Marsden, C. D., & Obeso, J. A. (1994). The functions of the basil ganglia and the
paradox of stereotaxic surgery in Parkinson's disease. Brain, 117, 877-897.

Mink, J. W. (1996). The basal ganglia: Focused selection and inhibition of competing
motor programs. Progress in Neurobiology, 50, 381-425.

Mink, J. W. (2003). The basal ganglia and involuntary movements. Neurological Review,
60, 1365-1368.

Mink, J. W., & Thach, W. T. (1993). Basal ganglia circuits and their role in behavior.
Current Biology, 3, 950-957.

Montgomery, Jr., E. B., Baker, K. B., Lyons, K. & Koller, W. C. (2000). Motor initiation
and execution in essential tremor and Parkinson's disease. Movement Disorders, 15, 511-515.

Morrison, C. E., Borod, J. C., Perrine, K., Beric, A., Brin, M. F., Rezai, A., et al. (2004).
Neuropsychological functioning following bilateral subthalamic nucleus stimulation in
Parkinson's disease. Archives of Clinical Neuropsychology, 19, 165-181.

Muller, T., Eising, E., Khun, W., Buttner, T., Coenen, H. -H., & Przuntek, H. (1999).
Delayed motor response correlates with striatal degeneration in Parkinson's disease. Acta
Neurologica Scandinavica, 100, 227-230.

Obwegeser, A. A., Uitti, R. J., Witte, R. J., Lucas, J. A., Turk, M. F., & Wharen, Jr., R. E.
(2001). Quantitative and qualitative outcome measures after deep brain stimulation to treat
disabling tremors. Neurosurgery, 48, 274-284.

Okun, M. S., & Foote, K. D. (2005). Subthalamic nucleus versus globus pallidus intema
deep brain stimulation, the rematch: Will pallidal deep brain stimulation make a triumphant
return?. Archives ofNeurology, 62, 533-536.

Ostergaard, K., Sunde, N., & Dupont, E. (2002). Effects of bilateral stimulation of the
subthalamic nucleus in patients with severe Parkinson's disease and motor fluctuations.
Movement Disorders, 17, 693-700.

Parsons, T. D., Rogers, S. A., Braaten, A. J., Woods, S. P., & Troster, A. J. (2006).
Cognitive sequelae of subthalamic nucleus deep brain stimulation in Parkinson's disease: A
meta-analyses. Lancet Neurology, 5, 578-588.

Pascual-Leone, A., Valls-Sole, J., Brasil-Neto, J. P., Cohen, L. G., & Hallett, M. (1994).
Akinesia in Parkinson's disease. I. Shortening of simple reaction time with focal, single-pulse
transcranial magnetic stimulation. Neurology, 44, 884-891.










Dirnberger, Reuman, Endl, Lindinger, Lang & Rothwell, 2000; Harrington & Haaland, 1991;

Inzelberg, Plotnik, Flash, Schechtman, Shahar, & Korczyn, 2001; Kropotov & Etlinger, 1999;

Robertson & Flowers, 1990; Roy, Saint-Cyr, Taylor, & Lang, 1993; Rubchinsky, Kopell, &

Sigvardt, 2003; Stelmach, Garcia-Colera, & Martin, 1989; Weiss, Stelmach, & Hefter, 1997).

Disordered maintenance and switching of motor programs has been proposed to explain some of

the primary motor symptomatology of PD, such as akinesia.

Speech Motor Programming Maintenance and Switching

Deficits in speech motor planning/programming in patients with PD are being increasingly

recognized (Van der Merwe, 1997). As noted by Van der Merwe (1997), "this complicates our

traditional view of dysarthria as a motor execution problem" (p. 18). The fact that the basal

ganglia appear to be involved "in both motor programming and execution suggests the

possibility of dual symptomatology in certain types of dysarthria" (Van der Merwe, 1997, p. 18).

The author then specifically suggests that the hypokinetic dysarthria of PD is one of the

dysarthria types in which "Coexisting problems in both motor programming and motor execution

would seem to be present..." (Van der Merwe, 1997, p. 18)

Spencer and Rogers (2005; see also Spencer 2006) have pioneered the study of speech

motor planning/programming deficits in dysarthria types traditionally associated with execution

level dysfunction. These authors suggest that the role of the basal ganglia in motor

planning/programming has been "illuminated by converging evidence from limb RT studies of

adults with Parkinson's disease" which show impairments in two primary areas: maintenance

and switching (Spencer & Rogers, 2005, p. 348). Furthermore, the presence of motor

planning/programming deficits in individuals with hypokinetic dysarthria is supported by

specific speech symptoms commonly encountered in this population. Spencer and Rogers (2005)










Table 4-2. Summary statistics for speech reaction time (SRT) and response accuracy by priming
condition and stimulation state. SRT data is in milliseconds (ms) and the number of
errors is per 16 responses.





No switch 615.24 96.77 671.38 113.05 643.31 106.83
Switch 717.09 89.11 728.67 110.50 722.88 98.35
No switch 0.69 0.85 1.06 0.74 0.88 0.8
Switch 1.5 1.17 1.5 1.98 1.5 1.59










inhibitory neurotransmitters. Neurons located downstream from the stimulation site are also

likely to be influenced by DBS. This notion is supported by functional imaging studies revealing

changes in cortical activity associated with improvements in motor function (Lozano et al.,

2002). For example, Davis et al. (1997) used positron emission tomography (PET) to reveal an

increase in regional cerebral blood flow in the PMC when GPi DBS improved motor

symptomatology. Finally, when considering the mechanism of DBS, it is important to consider

that DBS likely influences different neural targets in different ways (Lozano et al., 2002).

DBS and speech

Although speech disturbance in the form of hypokinetic dysarthria is common in

individuals with PD, speech improvement is not specifically targeted by DBS surgery. The

effects of neurosurgical treatments for PD on speech function have only recently begun to

receive systematic attention, particularly STN DBS (for more detailed reviews see Jones,

Kendall, Sudhyadhom, & Rosenbek, in press; Schulz, 2002; Schulz & Grant, 2000). To grossly

simplify what is developing to be a fairly substantial body of work, the influence of STN DBS on

speech has been studied using a variety of sophisticated measurement approaches, including

instrumental approaches such as acoustic analysis and kinematic measurement. Gentil and

colleagues (1999, 2000; 2001; see also Pinto, Gentil, Fraix, Benabid & Pollak, 2003) have been

pioneers in the study of the speech effects of STN DBS and have conducted a program of

research using perceptual, kinematic, and acoustic analyses. Considered overall, these data

suggest a number changes 'on' versus 'off STN DBS, including perceptual improvements in

speech, improved lingual and labial strength and control, and improvements in the acoustic

speech signal (e.g., increased fundamental frequency variability in sentences and decreased

fundamental frequency variability during sustained vowel production). PET scan data from











BIOGRAPHICAL SKETCH

Harrison N. Jones completed his B.A. in communication disorders at North Carolina State

University in 1996, followed by his M.A. in communication disorders at Appalachian State

University in 1998. Upon completion of his master's degree, Mr. Jones started his clinical

fellowship as a speech-language pathologist, which he completed in 1999. Since beginning his

fellowship, he has continued to practice as a speech-language pathologist with expertise in the

evaluation and treatment neurogenic communication and swallowing disorders in adults. He has

practiced at a variety of institutions including Duke University Medical Center and Shands

Hospital at the University of Florida. Mr. Jones enrolled at the University of Florida in pursuit of

his Ph.D. in rehabilitation science in 2004. His broad area of research interest is in motor speech

disorders in patients with neurological disease. He is particularly interested in preparatory

aspects of speech production (e.g., motor planning/programming) and speech disorders in

individuals with Parkinson's disease. Following completion of his Ph.D., Mr. Jones will return to

Duke University to join the academic faculty as an assistant professor. His primary

responsibilities will be to conduct a programmatic line of research in his areas of scientific

interest and provide evaluation and treatment services to adults with neurogenic speech and

swallowing disorders.











Trepanier, L. L., Kumar, R., Lozano, A. M., Lang, A. E., & Saint-Cyr, J. A. (2000).
Neuropsychological outcome of GPi pallidotomy and GPi or STN deep brain stimulation in
Parkinson's disease. Brain Cognition, 42, 324-347.

Van der Merwe, A. (1997). A theoretical framework for the characterization of
pathological speech sensorimotor control. In M. R. McNeil (Ed.) Clinical management of
sensorimotor speech disorders (pp. 1-25). New York: Thieme.

van den Wildenberg, W. P., van Boxtel, G. J., van der Molen, M. W., Bosch, D. A.,
Speelman, J. D., & Brunia, C. H. (2006). Stimulation of the subthalamic region facilitates the
selection and inhibition of motor responses in Parkinson's disease. Journal of Cognitive
Neuroscience, 18, 626-636.

Vink, M., Kahn, R. S., Raemaekers, M., van den Heuvel, M., Boersma, M., & Ramsey,
N. F. (2005). Function of striatum beyond inhibition and execution of motor responses. Human
Brain Mapping, 25, 336-344.

Vitek, J. L., & Walter, B. L. (2005). Surgical treatments for Parkinson's disease (CD-
ROM). 2005 American Academy ofNeurology.

Volkmann, J. (2004). Deep brain stimulation for the treatment of Parkinson's disease.
Journal of Clinical Neurophysiology, 21, 617.

Wang, E., Metman, L. V., Bakay, R., Arzbaecher, J., & Bernard, B. (2003). The effect of
unilateral electrostimulation of the subthalamic nucleus on respiratory/phonatory subsystems of
speech production in Parkinson's disease A preliminary report. Clinical Linguistics &
Phonetics, 17, 283-289.

Weiss, P., Stelmach, G. E., & Hefter, H. (1997). Programming of a movement sequence
in Parkinson's disease. Brain, 120, 91-102.

Whelan, B. -M., Murdoch, B. E., Theodoros, D. G., Silburn, P., & Hall, B. (2005).
Beyond verbal fluency: Investigating the long-term effects of bilateral subthalamic (STN) deep
brain stimulation (DBS) on language function in two cases. Neurocase, 11, 93-102.

Weismer, G. (2007). Neural perspectives on motor speech disorders. In G. Weismer (Ed.)
Motor speech disorders (pp. 57-91). San Diego, CA: Plural Publishing.

Witt, K., Pulkowski, U., Herzog, J., Lorenz, D., Hamel, W., Deuschl, G., et al. (2004).
Deep brain stimulation of the subthalamic nucleus improves cognitive flexibility but impairs
response inhibition in Parkinson's disease. Archives ofNeurology, 61, 697-700.

Yorkston, K. M., Beukelman, D. R., Strand, E. A., & Bell, K. R. (1999). Management of
Motor Speech Disorders in Children andAdults (2nd Edition). Austin, TX: Pro-Ed Inc.






























2007 Harrison N. Jones










associated with peripheral muscle changes such as impaired strength or tone. Current

conceptualization of all dysarthria types, including the dysarthria of PD, suggests that these

speech disorders and their resultant errors are due primarily (if not exclusively) from deviant

processes of execution.

Motor Planning/Programming

Van der Merwe's four phase model of speech control provides useful theoretical

constructs for research conducted in this area. In turn, notions about speech motor control will

continue to be refined with experimental data. Data supporting differentiation between the

processes of speech motor planning and speech motor programming would be particularly

valuable, and, as noted by Spencer & Rogers (2005), has not yet occurred. Therefore, for the

present discussion, we will combine these two phases of speech motor control into one motor

planning/programming.

The notion that preparatory activities consistent with processes involved in motor

planning/programming which occur prior to movement execution are commonly encountered in

the literature. Indeed, as noted by Schmidt in 1975, the notion of motor programs or "a set of

stored muscle commands ready for action at any given time has probably been with us for a very

long time" (p. 231). Early contributions from Lashley (1917) and Henry and Rogers (1960) have

continued to be refined by a number of researchers who have proposed models consistent with

concepts of speech motor planning/programming (Klapp, 2003; Levelt & Wheeldon, 1994;

Schmidt, 1975; Sternberg, Knoll, Monsell, & Wright, 1988; Sternberg, Knoll, & Turock, 1990;

Sternberg, Monsell, Knoll, & Wright, 1978; Sternberg, Wright, Knoll, & Monsell, 1980; Van der

Merwe, 1997).










CHAPTER 4
RESULTS

Participants

A total of 12 participated in the experiment. An additional eight individuals entered the

screening process but did not meet inclusion criteria or were withdrawn (3 had surgery

completed an outside facility, 2 failed screening due to a Spatial Span Subtest score < 7, 1

subject had severe tremor 'off DBS, 1 subject was unstable on anti-Parkinson's disease

medications, and 1 subject had a local skin reaction to the accelerometer).

Table 4-1 shows individual and group descriptive data for the 12 participants. Mean age

was 61 years (sd = 8.28). Nine of the participants were male (75%) and three were female (25%).

Three patients had undergone STN surgery (25%) and in 9 subjects the exact surgical site (GPi

or STN) was unknown due to participation in a larger double-blinded study. Eight of 12 (67%)

had undergone a unilateral DBS surgery and 4 (33%) had undergone bilateral DBS surgery.

Mean duration status-post surgery in months following surgery at the time of screening was 13.5

months (sd = 5.45). Half of the participants (6/12) were first randomized to the 'on' stimulation

test condition and the other half were first tested 'off stimulation.

Mean years of education was 14.16 (sd = 3.69) with the mode being 12 years (i.e., a high

school diploma). Mean MMSE was 28.67 out of 30 (sd = 0.98) Mean standard scores for the

Spatial Span Subtests were a forward score of 10.17 (sd = 2.04) and backward score of 10.83 (sd

= 1.99).

Perceptual judgment of dysarthria type was hypokinetic in all participants. The mean of

dysarthria severity ratings was 3.08 (sd = 1.31) and the mode was 2. All participants reported an

unremarkable speech and language developmental history. Mean intelligibility score across the

two listeners was 93.25% (sd = .06). Mean CES score was 33.03 out of 56 (sd = 8.66).










Smiley-Oyen, Lowry, & Kerr, 2007; Smiley-Oyen & Worringham, 2001; Sternberg et al., 1978).

Indeed, as previously discussed, choice reaction time is always longer than simple reaction

(Hallett, 1990). In the present experiment, results which did not support longer SRT for the

'switch' condition would cast great suspicion on the employed experimental paradigm. As the

'no switch' condition can be conceptualized as a simple RT experiment and the 'switch'

condition can be regarded as a choice RT experiment, results which did not show increased SRT

when switching speech motor programs would suggest a fatal flaw in the study as choice RT is

invariably longer than simple RT.

To more explicitly describe the proposed underlying mechanisms involved in speech

motor planning/programming, it is necessary to turn to the model proposed by Sternberg and

colleagues (1978, 1980, 1988 & 1990). Based on this model, in the 'no switch' condition,

subjects are hypothesized to retrieve subprograms for movement from the sensorimotor store.

These subprograms are loaded into the motor buffer and comprise the motor program.

Participants are able to prepare their speech responses prior to the command for speech execution

and maintain them in the buffer. In contrast, in the more complex 'switch' condition, subjects are

unable to prepare the desired motor program in advance, which requires that several additional

processes must occur before speech can be produced when the command for execution is

provided. These processes include inhibition of the previously prepared and now undesired

motor program, retrieval of new subprograms from the sensorimotor store, and the loading of

these subprograms into the buffer to construct a new motor program. Thus, as found in this

experiment, SRT should be longer in the 'no switch' condition due to the additional processes

required and the increased complexity of this task. It could further be argued that the temporal

difference between the 'no switch' and 'switch' conditions is a measure of these additional










In 1988, Sternberg and colleagues continued to clarify and refine their model of they now

call a "subprogram-selection model" in a paper emphasizing the concept that the motor program

"is operated upon by a series of selection and command processes" (p. 184). Before a movement

subprogram can be executed, it must first be accessed from the buffer. After the subprogram is

accessed from the buffer, "The command process... causes it to be 'executed'". Speech

production "is thus controlled by an alternating sequence of selection and command processes"

(p. 184).

The model presented by Sternberg and colleagues has its limitations but it provides a

sensible theoretical framework for research studying processes involved in motor

planning/programming in general and speech more specifically. This model can be applied to

concepts in the current motor control literature, such as motor programming maintenance and

switching.

Motor Program Maintenance

Sternberg et al.'s notion that motor programs are held in the buffer until a command for

movement is provided appears analogous to the ability to hold or maintain a motor program prior

to movement execution. Maintenance of a motor program has been studied extensively in both

normal and disordered populations. Hallett (1990) describes the common method for studying

maintenance of a motor program using the delayed response paradigm. In this situation,

"information about the movement is completely specified, but then the information is withdrawn

for a period of time before the stimulus to move is delivered" (Hallett, 1990, p. 588). RT serves

as an index of maintenance of the motor program in the buffer. In some disordered populations

(including individuals with PD), the contents of the buffer have been found to decay over time,

disrupting the maintenance of motor programs as measured by RT.











Table 4-3. Mean difference and p-values for speech reaction time (SRT) and response accuracy
by priming condition and stimulation state.


'stastically significant p < 0.05










FAS Test

Separate Wilcoxon signed-rank tests were conducted for each stimuli item of the FAS

(i.e., the letters F, A, and S) in the 'on' versus 'off DBS condition. No statistical differences in

phonemic verbal fluency between DBS conditions were found for any of the three stimuli letters

(F, p = 0.36, A, p = 0.25; S, p = 0.99), or across all stimuli (p = 0.99). For the stimulus letter F,

the mean number of correct response across subjects was 10.92 (SD = 4.34, range = 15) in the

'on' DBS condition and 11.92 (SD = 4.54, range = 16) 'off DBS. Mean number of correct

productions for the stimulus letter A was 9.75 (SD = 4.85, range 18) 'on' DBS and 8.58 (SD =

3.63, range = 12) 'off DBS. A mean of 10.92 (SD = 5.16, range = 19) correct productions were

observed 'on' DBS compared to 11.42 (SD = 4.38, range = 15) 'off DBS for stimulus letter S.

When these values were combined across all three stimulus letters, the mean number of correct

productions 'on' DBS was 31.58 (SD = 13.12, range 51) and 31.92 (SD = 11.40, range = 41)

'off DBS.

A pre-post decline in phonemic and semantic verbal fluency is perhaps the most

common neuropsychological finding after GPi and STN DBS surgery (Daniele et al., 2003; De

Gaspari et al., 2006; Morrison et al., 2004; Parsons, Rogers, Braaten, Woods, & Troster, 2006;

Rothlind, Cockshott, Starr, & Marks, Jr., 2007; Saint-Cyr, Trepanier, Kumar, Lozano, & Lang,

2000; Trepanier, Kumar, Lozano, Lang, & Saint-Cyr, 2000). In contrast to this well-established,

persistent decline in verbal fluency pre-post DBS surgery, the influence of post-operative DBS

state (i.e., 'on' and 'off stimulation) on this measure has received very little attention. However,

Schroeder and colleagues (2003) provide some guidance in their study of phonemic verbal

fluency in seven subjects 'on' and 'off stimulation. Phonemic verbal fluency was found to

significantly decline in the 'on' versus the 'off DBS condition. PET results revealed this decline










patient's homes or in a clinic environment, they are of insufficient quality to use for additional,

more precise analyses of the speech signal, such as temporal measurement of word length or

acoustic analyses.

Alternative Explanations

Alternative explanations for our findings must be considered. Chief among these

alternative considerations must be the notion that the findings are due to a different process than

speech motor planning/programming. A variety of cognitive influences could be used to explain

our findings including global cognitive function, attention and concentration, and working

memory. However, the MMSE was used to screen for global cognitive dysfunction and

participants with MMSE < 26 were excluded. Mean MMSE was 28.67 (SD = 0.98). Changes in

overall cognitive function thus appear unlikely to explain our findings for SRT as global

cognitive function was intact for our participants. The forward and backward portions of the

WMS-III Spatial Span were used to screen for disorders of attention and working memory.

Individuals with standard scores < 7 on either subtest were excluded. Mean Spatial Span forward

standard score was 10.17 (SD = 2.04) and Spatial Span backward standard score was 10.83 (SD

= 1.99). These means are both within the normal range and thusly, changes in attention or

working memory appear unlikely to explain our findings.

The response priming procedure we utilized also does not seem to support cognitive

mechanisms such as working memory to explain our findings. This testing procedure was

designed to make little demand on cognitive function in general. Additionally, we argue that this

paradigm is not a test of working memory, as subjects were provided with the target immediately

upon command for execution.










reinforced by reduced inhibition to GPi/SNpr delivered through the direct pathway.

Cumulatively, this "imbalance" between the indirect and direct pathways results in excessive

thalamic inhibition, finally resulting in reduced excitatory projections back to the cortex and

inhibiting intended movement (DeLong, 1990). Thus, the effect of striatal dopaminergic loss is

"inhibition of cortically initiated movement, to cause akinesia (loss of movement), hypokinesia

(reduction of movement) and bradykinesia (slowness of movement)" (Marsden & Obeso, 1994,

p. 878).

Other work has continued to refine the role of the BG in movement. For example, Mink

(1996; 2003; Mink & Thach, 1991) has proposed a model based on a series of experimental

observations which suggests the role of the basal ganglia in normal movement is to facilitate

desired motor programs while inhibiting other unwanted motor programs. According to this

model of focused selection and inhibition of competing motor programs, voluntary movement is

initiated by cortical mechanisms. The basal ganglia facilitate movement by decreasing inhibition

of desired motor programs while simultaneously increasing inhibition of competing motor

programs (Mink, 1996; Mink, 2003; Rubchinsky, Kopell, & Sigvardt, 2003) through GABAergic

output influencing cortical (and brainstem) motor mechanisms (Hikosaka, 2007; Mink, 2003).

The standard model of basal ganglia circuitry continues to be influential but it has been

suggested that this model requires substantial revision and refinement to account for advances in

understanding (Gale et al., in press; Greybiel, 2005). Nevertheless, this model continues to have

heuristic value and in particular, provides a logical rationale for surgical interventions.

Deep Brain Stimulation

A number of neurosurgical approaches to the treatment of PD have been developed over

the last century, including ablative procedures (e.g., pallidotomy), and, more recently, DBS. DBS











Research Question 2 .................. ................................ .. ....... .. ...........69
R research Q question 3 ................................................................70
S e c o n d a ry A im s ......................................................................................................................7 3
E x p lo rato ry A im s....................................................................................................................7 7
F A S T est .................................................................................................78
Stroop C olor and W ord T est.................................... .............................. ................. .. 79
Strengths/Weaknesses ................................ ....................... ..................80
A alternative E explanations ................................................................82
D discussion Conclusion ............................................. 83

6 F U T U R E W O R K .............................................................................................................. 8 6

APPENDIX

A TRAINING SESSION STIMULI ....................................... 89

B EXPERIM EN TAL STIM ULI .................................................... ...............90

L IST O F R E F E R E N C E S ............. ................. ...............................................................9 1

B IO G R A PH ICA L SK ETCH ........ ............ ......................................................................102










the value 0.68, which indicates strong agreement between the raters, and the confidence interval

of (0.59, 0.77) confirms that one can reject the null hypothesis of no agreement. In addition, the

percentage of task items agreed by the two raters range from 92% to 100% for the twelve

subjects. Intra-class correlation coefficient was determined to be 0.82 with a 95% confidence

interval [0.71, 0.90].










was 'on' or 'off. As shown in Figure 4-1, when collapsed across stimulation conditions, mean

SRT in the 'no switch' condition was 643.38 ms (SD = 106.83) and 722.88 ms (SD = 98.35) in

the 'switch' condition.

Research question 2 ('no-switch' condition 'on' vs. 'off' DBS)

Separate Wilcoxon signed-rank tests were conducted which revealed statistically

significant differences in SRT in the 'no switch' condition between 'on' and 'off DBS states

(signed rank = 10, p = 0.0210) in the predicted direction. That is, subjects produced a speech

response more quickly in the 'no switch' condition when 'on' versus 'off DBS. As shown in

Figure 4-2, mean SRT in the 'no switch' condition was 615.24 ms (SD = 96.77) 'on' DBS and

671.38 ms (SD = 113.05) 'off DBS.

Research question 3 ('switch' condition 'on' vs. 'off' DBS)

Separate Wilcoxon signed-rank tests were conducted which revealed no significant

differences in SRT in the 'switch' condition (signed rank = 30, p = 0.5186) between 'on' and

'off DBS states (i.e., no difference in SRT was observed in the 'switch' condition regardless of

whether DBS was 'on' or 'off). As shown in Figure 4-2, mean SRT in the 'switch' condition

was 717.09 ms (SD = 89.11) 'on' DBS and 728.67 ms (SD = 110.50) 'off DBS.

Secondary Aims

Separate Wilcoxon signed-rank tests were conducted for both the DBS 'on' and 'off

conditions. In the 'off DBS condition, no statistical difference (signed rank = 17, p = 0.5469)

was found in response accuracy between 'switch' and 'no-switch' conditions. In the 'on' DBS

condition, response accuracy was also not statistically significant (signed rank = 9, p = 0.0605),

thought there was a trend toward significant response accuracy results in the predicted direction

(i.e., subjects produced more errors on average in the 'switch' condition versus the 'no switch'










second pause between presentations. They were asked to orthographically transcribe the

sentences and enter them into the SIT program via a computer keypad. Scorers were encouraged

to pause sentence playback as needed in order to accurately transcribe the sentences, although

each sentence was heard only twice and no adjustments in volume were permitted.

A discourse analysis was completed by two SLPs experienced in discourse analysis (DK

& HJ) based on acoustic recordings of repetition of multisyllabic words, repetition of sentences,

and connected speech in order to determine linguistic competency. These productions were

assessed for the presence of linguistic errors, specifically phonologic errors (i.e., substitutions,

omissions, transpositions, etc.) and/or semantic or verbal errors. Any differences in analysis led

to re-listening and debate in order to make a final determination regarding the presence and type

of linguistic errors.

Experimental Sessions

Subjects who met all study criteria following completion of the screening session were

scheduled for the experimental session one to 30 days later. Please see Figure 2-1 for an example

of the typical experimental timeline. Subjects were tested 'off their anti-PD medication,

including levodopa. The 'off medication condition was defined as at least 12 hours off anti-PD

medications. Testing was conducted with left-brain DBS in the 'on' and 'off states. If subjects

had bilateral DBS, the right-brain DBS was turned 'off for the duration of the experimental

session. Two two-hour washouts of DBS were completed during each experimental session

during which all DBS therapy was discontinued. Two test sessions in which a battery of tests

was administered (see "Experimental Procedures" below) while subjects were 'on' and 'off

DBS were completed for each participant. Following the first DBS washout, subjects were quasi

randomly assigned to the first stimulation condition (i.e., 'on' or 'off DBS) in a counterbalanced










SRT in the 'No Switch' Condition is a Measure of Speech Motor Program Maintenance

In the present experiment, SRT in the 'no switch' condition serves as a measure of the

maintenance of speech motor programs 'on' and 'off DBS. In this condition, subjects are

visually presented with a prime word and instructed to speak this word as quickly and clearly as

possible upon presentation of the command for movement (i.e., the target word). Although the

experiment was not originally conceived in this manner, this paradigm satisfies the criteria for a

simple RT experiment in that subjects are provided with complete information regarding the

expected movement in advance. The subprogram retrieval model of Sternberg and colleagues

suggests that this allows participants to retrieve subprograms from the sensorimotor store and

load the motor program into the buffer prior to movement execution. Much like many other

simple RT experiments, the complexity of this task in the present experiment was increased by

adding a 250 ms delay between stimulus presentation (i.e., the prime word) and presentation of

the command for movement (i.e., the target word).

SRT in the 'Switch' Condition is a Measure of Speech Motor Program Switching

In the present experiment, SRT in the 'switch' condition serves as a measure of the

switching of speech motor programs 'on' and 'off DBS. In this condition, subjects are visually

presented with a stimulus (i.e., the prime word) which does not accurately inform the requested

movement upon receipt of the command for movement (i.e., the target word). In other words, the

prime unexpectedly does not match the target. This paradigm appears to generally satisfy the

criteria for a choice RT experiment in that subjects are provided with incomplete information

regarding the expected movement until the command for movement is presented. However, the

complexity is again increased by the presentation of an incorrect prime. Presumably, according

the model of Sternberg participants have already retrieved incorrect motor subprograms from the














OFF DBS
I 671.38 ms*


800-



700-



600 -



500 -


ON DBS
I 717.09 ms


NO SWITCH


OFF DBS
I 728.67 ms


SWITCH


DBS Condition
statistically significant, p < 0.05


Figure 4-2. Mean speech reaction time (SRT) in milliseconds (ms) in the 'no switch' and
'switch' conditions 'on' and 'off DBS.


ON DBS
T 615.24 ms























Connecting wire






IPG / neurostimulator


-~


Figure 3-4. Unilateral deep brain stimulation (DBS). IPG = internal pulse generator.





























urement of
begins here


Prime Target
1000 ms 1000 ms



Figure 2-2. Response priming protocol.










language, and it is portrayed as a linguistic-symbolic function within the proposed framework"

(Van der Merwe, 1997, p. 9). These processes are thought to be accomplished primarily by

temporal-parietal areas, including Broca's and Wernicke's areas. Errors attributed to the level of

linguistic-symbolic planning are due to semantic, lexical, syntactic, morphological, and

phonological errors associated with language based disorders (i.e., aphasia). For example, Van

der Merwe suggests deficits in phonological planning will result in phoneme substitutions and

transpositions.

Motor Planning

Motor planning, along with the next level of the model, motor programming, is the area

of interest in the present experiment. Van der Merwe conceptualizes motor planning phase as

being "mediated by the by the 'highest' level of the motor hierarchy" (p. 9). Motor planning

involves "gradual transformation of symbolic units (phonemes) to a code that can be handled by

a motor system" (p. 9). Van der Merwe (1997) suggests that speakers learn "core motor plans"

during development and adaptation. These motor plans include goals in the form of spatial and

temporal specifications for movement which are stored in sensorimotor memory. During

acquisition of a core motor program, such as during development, this model suggests increased

usage of external feedback, such as that from proprioception and audition. Following retrieval of

the core motor plan, planning continues in order so that the "consecutive movements necessary

to fulfill the spatial and temporal goals" can be met (p. 11). Motor plan subroutines such as velar

lift and tongue placement are then specified for the planned production. Motor planning is

thought be accomplished in multiple neural regions, particularly the motor association areas of

PMC and SMA, as well as prefrontal and parietal association areas. Van der Merwe (1997)

suggests the changes in speech that may be encountered due to deficits in motor planning to

















LIST OF REFERENCES


Adams, S. G., Jog, M., Eadie, T., Dykstra, A., Gauthier, G., & Vercher, J. -L. (2004).
Jaw and finger movements during visual and auditory motor tracking in Parkinson disease.
Journal of Medical Speech-Language Pathology, 12, 125-130.

Alexander, G. E., DeLong, M. R., & Strick, P. L. (1986). Parallel organization of
functionally segregated circuits linking basal ganglia and cortex. Annual Review of
Neuroscience, 9, 357-381.

Aron, A. R., Fletcher, P. C., Bullmore, E. T., Sahakian, B. J., & Robbins, T. W. (2003).
Stop-signal inhibition disrupted by damage to right inferior frontal gyrus in humans. Nature
Neuroscience, 6, 115-116.

Aron, A. R., Robbins, T. W., & Poldrack, R. A. (2004). Inhibition and the right inferior
frontal cortex. Trends in Cognitive Sciences, 8, 170-177.

Benabid, A. L. (2003). Deep brain stimulation for Parkinson's disease. Current Opinion
in Neurobiology, 13, 693-706.

Benabid, A. L., Pollak, P., Gao, B., Hoffmann, D., Limousin, P., Gay, E., et al. (1996).
Chronic electrical stimulation of the ventralis intermedius nucleus of the thalamus as a treatment
of movement disorders. Journal ofNeurosurgery, 84, 203-214.

Benabid, A. L., Pollak, P., Gross, C., Hoffmann, D., Benazzouz, A., Gao, B., et al.
(1994). Acute and long-term effects of subthalamic nucleus stimulation in Parkinson's disease.
Stereotactic and Functional Neurosurgery, 62, 76-84.

Benecke, R., Rothwell, J. C., Dick, P. R., Day, B. L., & Marsden, C. D. (1987).
Disturbance of sequential movements in patients with Parkinson's disease. Brain, 110, 361-379.

Benton, A. L., & Hamsher, K. (1976). Multilingual aphasia examination: Manual of
instruction. Iowa City: University of Iowa.

Bergman, H., & Deuschl, G. (2002). Pathophysiology of Parkinson's disease: From
clinical neurology to neuroscience and back. Movement Disorders, 17, S28-S40.

Bloxham, C. A., Mindel, T. A., & Frith, C. D. (1984). Initiation and execution of
predictable and unpredictable movements in Parkinson's disease. Brain, 107, 371-384.










Exploratory Aims

For exploratory purposes, two measures of neuropsychological performance, the FAS test

and the Stroop Color and Word Test, were administered both 'on' and 'off DBS. The FAS is a

test of phonemic verbal fluency in which an individual's ability to generate words beginning

with the letters F, A, and S is counted. The FAS and other tests of phonemic verbal fluency are

commonly encountered in the neuropsychological literature, as are similar tests such as tests of

semantic verbal fluency (e.g., animals). Normative data is available from Tombaugh, Kozak, and

Rees (1999) from a large group stratified by age and education level are available for these

particular stimuli.

The Stroop Color and Word Test has three sections the word section, the color section,

and the color-word section. The word section has randomly ordered stimulus items comprising

three different color words (i.e., red, green, and blue) printed in black ink. Subjects are instructed

to correctly read aloud as many of the printed words as possible in 45 seconds. Similarly, the

color section comprises symbols (i.e., XXXX) printed in red, green, and blue ink and subjects are

instructed to correctly read aloud as many of the printed colors as possible in 45 seconds. Finally,

the color-word section comprises the color words from the word page printed in incongruent

colors from the color page. For example, the word "red" is shown printed in blue ink. Subjects

are again provided 45 seconds and instructed to name as many of the colors the words are printed

in as possible (rather than the word that is printed). The Stroop Color and Word Test is thought

to be a test of an individual's ability to volitionally inhibit automatic word reading to produce the

required response. Our findings on these tests and their interpretation follow.










ACKNOWLEDGMENTS

I thank all of the countless individuals and institutions who have supported me during

completion of my Ph.D. For their steadfast love and support, I thank my wife Carlee, my son

Nate, my mother Linda, and the memory of my father Mike. For their professional and personal

mentoring, I thank the all the members of my dissertation committee Drs. John C. Rosenbek,

Diane L. Kendall, Michael S. Okun, and Craig Velozo. I would further like to extend my deepest

gratitude to Dr. Rosenbek for all he has taught me about science and life. I will forever be

indebted to him for his kind but intellectually rigorous mentorship throughout my years at the

University of Florida (UF). Perhaps someday in the future, I can repay the debt in kind to

another. I thank Sam Wu for his guidance and assistance with my statistical analyses and Hubert

Fernandez for his expertise in experimental design. I thank Nan Musson, Jim Korner & Leslie

Gonzalez-Rothi for their support during completion of my degree. I thank Chris Edghill for his

expertise in the graphic design of the figures contained within this work. I thank all of my

research assistants Kaitlin Kobaitri, Bashar Mourad, Kristin Shaffer, Tina Tso, & Lauren

Laube for their volunteer work which was so critical for the success of this project. Finally, for

their support of my education and research, I thank the Office of Research and Development,

Rehabilitation, R&D Service, Department of Veterans Affairs, the Brain Rehabilitation Research

Center at the Malcom Randall VAMC, the UF Movement Disorders Center, the UF Department

of Communicative Disorders, and the faculty, staff, and my fellow students in the UF

Rehabilitation Science Doctoral Program.










planning/programming phase. Contrary to our prediction, no significant differences in SRT were

observed in the 'switch' condition regardless of whether DBS was 'on' or 'off. Mean SRT in the

'switch' condition was 717.09 ms 'on' DBS and 728.67 ms 'off DBS.

This negative finding was unexpected considering previous data in the limb literature

showing improvement in choice RT with DBS of GPi (Schubert et al., 2002) and STN in humans

(Kumru et al., 2003; Temel et al., 2005; van den Wildenberg et al., 2006) and rodents (Temel et

al., 2005). However, close inspection of these previous studies reveal the presence of at least two

important differences in comparison to the present work. One, the choice RT paradigms utilized

in previous work did not utilize use of a prime to prepare subjects for the upcoming movement

command. This is in contrast to our study which utilized a presentation of a prime word in all

trials. Additionally, in the 'switch' condition of our study, the prime word unexpectedly did not

match the target word. This required subjects to activate the desired motor program and inhibit

the unwanted motor program. This additional process of inhibition was not required by the

simple RT paradigms in the previous work. The second important difference between this study

and prior limb research concerns population differences. Subjects in the previous work almost

invariably underwent bilateral procedures. In contrast, only one-third (4/12) of our participants

had bilateral DBS and two-thirds (8/12) had underwent unilateral left only surgery. In

participants who had undergone bilateral DBS and consequently had right DBS, the lead in this

hemisphere was turned 'off for the entire study and not manipulated experimentally. These two

differences are particularly important because movement inhibition has been suggested to be a

bilateral (Liddle, Kiehl, & Smith, 2001; Leung & Cai, 2007) but primarily right dominant

process mediated by prefrontal cerebral circuits (Aron, Robbins, & Poldrack, 2004; Aron et al.,

2003; Chambers et al., 2006; Garavan et al., 1999; Konishi et al., 1999; Leung & Cai, 2007;










is a surgical procedure in which electrical stimulation is delivered to neural targets through

chronically implanted leads. Quadrapolar electrodes are connected to an internalized

programmable neurostimulator usually placed below the clavicle as shown in Figure 3-4

(Benabid, 2003). Thalamic, pallidal, and STN DBS are all recognized to have a beneficial effect

on the motor symptoms of PD (Benabid, 2003; Gross, 2004; Rodriguez-Oroz et al., 2005;

Volkmann, 2004). After a period of intense use and then declining interest, neurosurgical

treatments for PD have once again become popular, primarily due to limitations in the medical

management of PD, advances in the understanding of the intrinsic and extrinsic circuitry of the

basal ganglia, surgical technique, neuroimaging, and microelectrode recording techniques

(Koller, Pahwa, Lyons & Albanese, 1999). In comparison to ablative procedures, DBS may have

many advantages, including a decreased occurrence of adverse events, minimal permanent

lesions, and an increased ability to perform bilateral procedures without adverse events such as

speech and swallowing problems (Benabid et al., 1996; Ghika et al., 1998; Koller et al., 1999;

Obwegeser et al., 2001; Pinto et al., 2004; Rodriguez-Oroz et al., 2005). Adjustments in the

parameters of stimulation can also facilitate individualization in the treatment and minimize side

effects.

DBS in PD

Although the optimal surgical target for DBS in PD remains unknown (Okun & Foote,

2005), the most common surgical targets are the thalamus, GPi, and STN. Influential early

studies by Benabid and colleagues (1994) and Limousin et al. (1998) suggested significant

benefit to STN DBS and this quickly became the surgical treatment of choice for most centers.

However, it has been suggested that further trials comparing the benefit of GPi versus STN DBS

need completion (and are in fact in progress) (Okun & Foote, 2005).










extensively and is generally found to be impaired in comparison to age-matched normal controls

(Bloxham, Mindel, & Frith, 1984; Draper & Johns, 1964; Evarts, Teravainen, & Calne, 1981;

Gueye, Viallet, Legallet, & Trouche, 1998; Labutta, Miles, Sanes, & Hallett, 1994; Montgomery,

Baker, Lyons, & Koller, 2000; Muller, Eising, Khun, Buttner, Coenen, & Przuntek, 1999; Temel

et al., 2006). There are two general types of RT conditions: simple RT and choice RT.

Simple and Choice Reaction Time

In the simple RT test paradigm, "the expected movement is described completely,

without ambiguity" (Hallett, 1990, p. 587). This allows subjects to fully prepare (or plan and

program) the required movements in advance of the command to execute movement. Simple RT

experiments can still increase the complexity of the task, most often by adding a delay between

stimulus and command. Using the model of Sternberg et al., this would require participants to

maintain the motor program in the buffer until the command to execute movement is provided.

In the choice RT paradigm, subjects are not provided "a complete description of the

required movement" until "the stimulus that calls for the movement initiation" (p. 587) is

delivered. Since subjects are not able to plan/program movements in advance, choice RT is

always longer than simple RT. Like simple RT, choice RT is also influenced by complexity

factors. For example, providing incorrect information about the required movement in advance

of the command for movement increases complexity (and thus RT). According to the model of

Sternberg et al., the increase RT in the choice RT paradigm is explained by the additional

required planning/programming processes required with this task. These processes include

retrieving the appropriate motor subprograms from the sensorimotor store and loading them into

the buffer.









800-



700-



600 -



500 -



400-


T 722.88 ms*


T 643.31 ms


NO SWITCH


SWITCH


Priming Condition
statistically significant, p < 0.05
Data collapsed across DBS stimulation conditions





Figure 4-1. Mean speech reaction time (SRT) in milliseconds (ms) in the 'no switch' and
'switch' conditions.










Hoffman-Ruddy, B., Schulz, G., Vitek, J., & Evatt, M. (2001). A preliminary study of the
effects of subthalamic nucleus deep brain stimulation on voice and speech characteristic in
Parkinson's disease. Clinical Linguistics & Phonetics, 15, 97-101.

Inzelberg, R., Plotnik, M., Flash, T., Schechtman, E., Shahar, I., & Korczyn, A. D.
(2001). Mental and motor switching in Parkinson's disease. Journal of Motor Behavior, 33,
377-385.

Jahanshahi, M., Ardouin, C. M., Brown, R. G., Rothwell, J. C., Obeso, J., Albanese, A.,
et al. (2000). The impact of deep brain stimulation on executive function in Parkinson's disease.
Brain, 123, 1142-1154.

Jones, H. N., Kendall, D. L., Sudhyadhom, A., & Rosenbek, J. C. (in press). The effects
of lesion therapy and deep brain stimulation on speech function in patients with Parkinson's
disease. Communicative Disorders Review.

Jones, D. L., Phillips, J. G., Bradshaw, J. L., lansek, R., & Bradshaw, J. A. (1992).
Programming of single movements in Parkinson's disease: Comparison with Huntington's
disease. Journal of Clinical and Experimental Neuropsychology, 14, 762-772.

Keller, E. (1987). The cortical representation of motor processes of speech. In E. Keller
& M. Gopnik (Eds.). Motor and sensory processes of language (pp. 125-162). Hillsdale, NJ:
Lawrence Erlbaum Associates.

Kent, R. D., Kent, J. F., & Rosenbek, J. C. (1987). Maximum performance tests of speech
production. Journal of Speech and Hearing Disorders, 52, 367-387.

Kent, R. D., Kent, J. F., Rosenbek, J. C., Vorperian, H. K., & Weismer, G. (1997). A
speaking task analysis of the dysarthrias in cerebellar disease. Folia Phoniatrica et Logopaedica,
49, 63-82.

Kent, R. D., & Rosenbek, J. C. (1982). Prosodic disturbance and neurologic lesion. Brain
and Language, 15, 259-291.

Klapp, S. T. (2003). Reaction time analysis of two types of motor preparation for speech
articulation: Action as a sequence of chunks. Journal of Motor Behavior, 35, 135-150.

Koller, W. C., Pahwa, R., Lyons, K. E., & Albanese, A. (1999). Surgical treatment of
Parkinson's disease. Journal of the Neurological Sciences, 167, 1-10.

Konishi, S., Nakajima, K., Uchida, I., Kikyo, H., Kameyama, M., & Miyashita, Y.
(1999). Common inhibitory mechanism in human inferior prefrontal cortex revealed by event-
related functional MRI. Brain, 122, 981-991.










Saint-Cyr, J. A., Trepanier, L. L., Kumar, R., Lozano, A. M., & Lang A. E. (2000).
Neuropsychological consequences of chronic bilateral stimulation of the subthalamic nucleus in
Parkinson's disease. Brain, 123, 2091-2108.

Santens, P., De Letter, M., Van Borsel, J., De Reuck, J., & Caemaert, J. (2003).
Lateralized effects of subthalamic nucleus stimulation on different aspects of speech in
Parkinson's disease. Brain and Language, 87, 253-258.

Schmidt, R. A. (1975). A schema theory of discrete motor skill learning. Psychological
Review, 82, 225-260.

Schroeder, U., Kuehler, A., Lange, K. W., Haslinger, B., Tronnier, V. M., Krause, M., et
al. (2003). Subthalamic nucleus stimulation affects a frontotemporal network: A PET study.
Annals ofNeurology, 54, 445-450.

Schubert, T., Volkmann, J., Muller, U., Sturm, V., Voges, J., Freund, H. J., et al. (2002).
Effects of pallidal deep brain stimulation and levodopa treatment on reaction-time performance
in Parkinson's disease. Experimental Brain Research, 144, 8-16.

Schulz, G. M. (2002). The effects of speech therapy and pharmacological treatments on
voice and speech in Parkinson's disease: A review of the literature. Current Medicinal
Chemistry, 9, 1359-1366.

Schulz, G. M., & Grant, M. K. (2000). Effects of speech therapy and pharmacologic and
surgical treatments on voice and speech in Parkinson's disease: A review of the literature.
Journal of Communication Disorders, 33, 59-88.

Schupbach, W. M. M., Chastan, N., Welter, M. L., Houeto, J. L., Mesnage, V., Bonnet,
A. M., et al. (2005). Stimulation of the subthalamic nucleus in Parkinson's disease: A 5 year
followup. Journal ofNeurology, Neurosurgery, & Psychiatry, 76, 1640-1644.

Sheridan, M. R., Flowers, K. A., & Hurrell, J. (1987). Programming and execution of
movement in Parkinson's disease. Brain, 110, 1247-1271.

Smiley-Oyen, A. L., Lowry, K. A., & Kerr, J. P. (2007). Planning and control of
sequential rapid aiming in adults with Parkinson's disease. Journal of Motor Behavior, 39,
103-114.

Smiley-Oyen, A. L., & Worringham, C. J. (2001). Peripheral constraint versus on-line
programming in rapid aimed sequential movements. Acta Psychologica, 108, 219-245.

Smith, D. J. (2004). Motor programming. Retrieved October 8, 2008 from
http://www.smithsrisca.demon.co.uk/motor-programming.html










was accompanied by decreased regional cerebral blood flow in several areas including "the right

orbitofrontal cortex, the left inferior temporal gyms, and the left inferior frontal/insular cortex"

(Schroeder et al., 2003, p. 447). The neurophysiological mechanism proposed by these authors

for this decline with STN DBS was "decreased activation of a left-sided network... incorporating

the inferior frontal cortex, the insular cortex, and the temporal cortex" (Schroeder et al., 2003, p.

447). Our data are not consistent with these findings but other recent work supports our findings.

Witt et al. (2004) studied the effects of STN DBS on verbal fluency (including phonemic

fluency) in 23 subjects with PD and found no change in verbal fluency between the 'on' and

'off DBS states. The differences between these two previous studies might be explained by

differences in the cognitive status of the participants. That is, Schroeder et al. (2003) did not

appear to screen or assess cognitive status, while Witt and colleagues (2004) excluded

participants with evidence of cognitive dysfunction, as did our current study. Regardless, the

effects of DBS state on verbal fluency in individuals with PD demands further systematic

attention.

Stroop Color and Word Test

Separate Wilcoxon signed-rank tests were conducted for each of the subtests of the

Stroop in the 'on' versus 'off DBS conditions. No statistical differences in t-scores were found

for the word section (p = 0.49), the color section (p = 0.30), or the color-word section (p = 0.28).

Further, no significant differences in interference score were found (p = 0.56). Overall, GPi and

STN DBS have generally been found to be well tolerated procedures in terms of associated

cognitive decline besides the common and persistent decline in verbal fluency described above

(Daniele et al., 2003; Limousin et al., 1998; Pillon et al., 2000). However, meta-analysis has

revealed "significant, albeit small, declines in executive functions and verbal learning and











Table 2-2. Power analysis to detect differences between 'switch' and 'no switch' speech
responses.


N 10 11 12 13 14 15

Power










Gentil, M., Chaubin, P., Pinto, S., Pollak, P., & Benabid, A. L. (2001). Effect of bilateral
stimulation of the subthalamic nucleus on parkinsonian voice. Brain and Language, 78, 233-240.

Gentil, M., Garcia-Ruiz, P., Pollak, P., & Benabid, A.-L. (1999). Effect of stimulation of
the subthalamic nucleus on oral control of patients with parkinsonism. Journal ofNeurology,
Neurosurgery, & Psychiatry, 67, 329-333.

Gentil, M., Garcia-Ruiz, P., Pollak, P., & Benabid, A.-L. (2000). Effect on bilateral deep-
brain stimulation on oral control of patients with parkinsonism. European Neurology, 44,
147-152.

Gentilucci, M., & Negrotti, A. (1999a). The control of an action in Parkinson's disease.
Experimental Brain Research, 129, 269-277.

Gentilucci, M., & Negrotti, A. (1999b). Planning and executing an action in Parkinson's
disease. Movement Disorders, 14, 69-79.

Ghika, J., Villemure, J.-G., Fankhauser, H., Favre, J., Assal, G., & Ghika-Schmid, F.
(1998). Efficiency and safety of bilateral contemporaneous pallidal stimulation (deep brain
stimulation) in levodopa-responsive patients with Parkinson's disease with severe motor
fluctuations: A 2 year follow-up review. Journal ofNeurosurgery, 89, 713-718.

Golden, C. J., & Freshwater, S. M. (2002). The Stroop color and word test: A manual for
clinical and experimental uses. Wood Dale, IL: Stoelting Co.

Grill, W. M., Snyder, A. N., & Miocinovic, S. (2004). Deep brain stimulation creates an
informational lesion of the stimulated nucleus. Neuroreport, 15, 1137-1140.

Gross, R. E. (2004). Deep brain stimulation in the treatment of neurological and
psychiatric disease. Expert Review ofNeurotherapeutics, 4, 465-478.

Gueye, L., Viallet, F., Legallet, E., & Trouche, E. (1998). The use of advance information
for motor preparation in Parkinson's disease: Effects of cueing and compatibility between
warning and imperative stimuli. Brain and Cognition, 38, 66-86.

Harrington, D. L., & Haaland, K. Y. (1991). Sequencing in Parkinson's disease:
Abnormalities in programming and controlling movement. Brain, 114, 99-115.

Henry, F. M., & Rogers, D. E. (1960). Increased response latency for complicated
movements and a "memory drum" theory of neuromotor reaction. The Research Quarterly, 31,
448-458.

Hikosaka, 0. (2007). GABAergic out put of the basal ganglia. Progress in Brain
Research, 160, 209-226.










inhibition of unwanted motor programs or retrieval of new subprograms from the sensorimotor

store).

Traditionally, the speech deficits in individuals with hypokinetic dysarthria (and indeed

all dysarthria types) have been considered to be execution level deficits (Darley, Aronson, &

Brown, 1969a, b, 1975; Duffy, 2005; Yorkston et al. 1999). However, this conceptualization of

dysarthria as strictly an execution level disorder has been questioned by Kent and associates

(Kent & Rosenbek, 1982; Kent et al. 1997), as well as more recent experimental findings from

Spencer & Rogers (2005; see also Spencer 2006). Our present findings also support the notion

that individuals with PD and hypokinetic dysarthria have speech deficits at the level of motor

planning/programming. Furthermore, our findings suggest that these planning/programming

deficits can be measured using a SRT paradigm. Finally, DBS of the GPi and/or STN appears to

differentially influence the motor planning/programming processes required in the different

priming conditions of our experiment. In other words, our findings suggest that DBS is

associated with an improvement in the maintenance of the speech motor program in the buffer

but not the multiple processes involved in switching of speech motor programs.










Krack, P., Batir, A., Van Blercom, N., Chabardes, S., Fraix, V., Ardoun, C., et al. (2003).
Five-year follow-up of bilateral stimulation of the subthalamic nucleus in advanced Parkinson's
disease. New England Journal of Medicine, 349, 1925-1934.

Kropotov, J. D., & Etlinger, S. C. (1999). Selection of actions in the basal ganglia-
thalamocortical circuits: Review and model. International Journal ofPsychophysiology, 31,
197-217.

Kumar, R., Lozano, A. M., Kim, Y. J., Hutchison, W. D., Sime, E., Halket, E. et al.
(1998a). Evaluation of subthalamic nucleus deep brain stimulation in advanced Parkinson's
disease. Neurology, 51, 850-855.

Kumar, R., Lozano, A. M., Montgomery, E., & Lang, A. E. (1998b). Pallidotomy and
deep brain stimulation of the pallidum and subthalamic nucleus in advanced Parkinson's disease.
Movement Disorders, 13, 73-82.

Kumru, H., Summerfield, C., Valldeoriola, F., & Valls-Sole, J. (2003). Effects of
subthalamic nucleus stimulation on characteristics of EMG activity underlying reaction time in
Parkinson's disease. Movement Disorders, 19, 94-100.

Labutta, R. J., Miles, R. B., Sanes, J. N., & Hallett, M. (1994). Motor program memory
storage in Parkinson's disease patients tested with a delayed response task. Movement Disorders,
9, 218-222.

Lashley, K. S. (1917). The accuracy of movement in the absence of excitation from a
moving organ. American Journal ofPhysiology, 43, 169-194.

Leung, H. -C., & Cai, W. (2007). Common and differential ventrolateral prefrontal
activity during inhibition of hand and eye movements. Journal ofNeuroscience, 27, 9893-9900.

Levelt, W. J. & Wheeldon, L. (1994). Do speakers have access to a mental syllabary?.
Cognition, 50, 239-269.

Liddle, P. F., Kiehl, K. A., & Smith, A. M. (2001). Event-related fMRI study of response
inhibition. Human Brain Mapping, 12, 100-109.

Limousin, P., Krack, P., Pollak, P., Benazzouz, A., Ardoun, C., Hoffmann, D. et al.
(1998). Electrical stimulation of the subthalamic nucleus in advanced Parkinson's disease. The
New England Journal of Medicine, 339, 1105-1111.

Lozano, A. M., Dostrovsky, J., Chen, R., & Ashby, P. (2004). Deep brain stimulation for
Parkinson's disease: Disrupting the disruption. Lancet Neurology, 1, 225-231.










Vink et al., 2005). Especially pertinent to the present discussion, Aron and colleagues (2004)

assert that "Converging evidence the right frontal cortex might subserve inhibitory processes

underlying switching" (p. 171). Since left DBS primarily influences the ipsilateral hemisphere,

the lack of significance in SRT in the 'switch' condition found in our study may be due to a

laterality effect precipitated by the inhibition demands of this task. In other words, if a large

portion of the 'switch' task involves right hemisphere mediated inhibition mechanisms, it

appears that left DBS would have little opportunity to influence these contralateral neural

mechanisms.

The negative findings for this research question lead us to consider alternative

explanations such as a lack of power. However, the study does not appear to be underpowered to

answer this research question. Power was sufficient to answer research questions 1 and 2. The

positive findings in research question 1 provide support for the experimental paradigm. All

variables between research questions 2 and 3 remained constant except the priming condition

and power was sufficient to answer research question 2. Furthermore, at a glance, mean SRT

differed little between stimulation conditions (i.e., 717.09 ms 'off DBS and 728.67 ms 'on'

DBS) and the p-value of 0.52 do not suggest a trend in the data. Of course, the participation of

additional subjects would increase the likelihood of detecting statistically significant group

differences, but in this case, if such differences were found, they would be unlikely to be

meaningful.

Perhaps the most parsimonious explanation for these findings is that stimulation only had

a positive influence on maintenance of motor programs in the buffer rather than on the additional

processes necessary for switching. The 'switch' condition was designed to be a more complex

test of motor planning/programming due to these additional processes. To again turn to the










Subprogram Retrieval Model

The work of Sternberg and colleagues (1978, 1980, 1988, & 1990) has been particularly

useful in attempting to define the theoretical underpinnings of the processes involved in speech

motor planning/programming. In 1978, Sternberg et al. proposed a "subprogram retrieval" model

(later renamed as the "subprogram-selection" model" by Sternberg et al., 1988) to explain their

data on preparation of rapid movement sequences during speech and typewriting. In this model, a

motor program or "representation of the entire response... is constructed before the response

starts" (Sternberg et al., 1978, p. 133). The motor program comprises "a set of linked

subprograms, one for each unit of the response" that are "retained in a special motor-program

buffer...distinct from ordinary short-term memory" until the command for movement is received

(Sternberg et al., 1978, p. 133). Subprograms are retrieved from the sensorimotor store and

loaded into the buffer prior to movement. Keller (1987) suggests that the subprograms or

"aggregates of muscle commands" are learned and are then to be found in the store ready for

retrieval as needed, rather than freshly generated each time speech is to be produced (p. 135).

Upon command for movement, the first subprogram is located in the buffer in order to initiate

movement execution. Sternberg and colleagues (1978) further describe their proposed

mechanism for retrieval of subprograms from the buffer: "The particular retrieval mechanism

suggested by our results is self-terminating sequential search through a nonshrinking buffer,

rather than, for example, a process of direct access... The search is presumably necessary because

the necessary subprograms are not arranged in the buffer in the order in which they must be

executed" (p. 147). Additionally, Sternberg and colleagues suggest the size or unit of the speech

subprograms in their model is the "stress group or 'metric foot' (a segment of speech associated

with a primary stress)" (1978, p. 136)










CHAPTER 6
FUTURE WORK

Our experience with this research project suggests many avenues for future work in this

area. Experiments to further determine how robust the positive influences of DBS on maintaining

speech motor programs in the buffer are warranted. This can be accomplished in a number of

ways, such as by varying the length of delay or with the use of articulatory suppression between

presentation of the prime and target.

Further investigations into the laterality effects of DBS in PD on speech motor program

maintenance and switching are also warranted. As previously discussed in Chapter 5, our lack of

significant differences in the 'switch' condition between the 'on' and 'off DBS states may have

been due to the fact that subjects with unilateral left DBS were targeted for recruitment due to

the critical nature of the left hemisphere in speech and language. Although subjects with bilateral

DBS also participated, right DBS was turned 'off for the entirety of the experiment. Since the

inhibition process involved in the 'switch' condition may rely heavily on right hemisphere

cerebral circuitry, a comparison between PD subjects with right and left DBS may assist in

further determining the effects of DBS on the switching of speech motor programs. It might be

expected that right hemisphere DBS would improve performance in the 'switch' condition due to

an improved ability to inhibit unwanted motor programs. Such a paradigm would also allow a

comparison on the effects of left and right DBS on maintenance of speech motor programs.

Subsequent work in this area may be improved by collecting execution level speech data

in addition to the motor planning/programming variables studied in the present experiment. For

example, data such as the duration of movement during target speech productions would target

the level of execution and complement the SRT planning/programming data we collected. We

attempted to complete post-hoc analysis of movement time in the present experiment, but were











Table 2-1. Auditory-perceptual dysarthria severity scale.


0 Normal speech

1 Slight dysarthria

2 Mild dysarthria

3 Mild-moderate dysarthrla


































To my extraordinary wife Carlee your steadfast love, support, and encouragement have
allowed me to achieve my dreams










processes. In other words, the average difference across priming conditions of 79.57 ms may

reflect the additional time required to inhibit the undesired motor program, retrieve new

subprograms from the sensorimotor store, and load them into the buffer to construct a new motor

program.

Research Question 2

Research question 2 was designed to determine whether a significant difference exists in

SRT in the 'on' DBS versus the 'off DBS state when participants with PD produce a word in the

'no switch' condition. The hypothesized theoretical level of processing of this question was

maintenance of the motor program in the buffer at the level of motor planning/programming. We

were interested in determining if DBS influences maintenance of the motor program in the buffer

after retrieval from the sensorimotor store. As predicted, statistically significant differences in

SRT were found which revealed faster performance 'on' DBS when compared to 'off DBS.

Mean SRT in the 'no switch' condition was 615.24 ms 'on' DBS and 671.38 ms 'off DBS. This

leads us to conclude that DBS directly influences speech motor program maintenance.

There are data available to suggest that motor program maintenance is impaired in PD

(Gentilucci and Negrotti, 1999a, b; Gueye et al., 1998; Jones et al., 1992; Labutta et al., 1994;

Pascual-Leone, Valls-Sole, Brasil-Neto, Cohen & Hallett, 1994; Romero et al., 2003; Sheridan,

Flowers, & Hurrell, 1987; Stelmach, Garcia-Colera & Martin, 1989). Although sufficient detail

regarding the theoretical underpinnings of many of these studies is not provided, the model of

Sternberg et al. (1978, 1980, 1988 & 1990) suggests this deficit may occur while motor programs

are held in the buffer prior to command for movement execution. Our findings suggest that

maintenance of speech motor programs in the buffer may be enhanced by DBS in individuals

with PD, at least in simple RT experiments. On average, SRT was improved by 56.14 ms in the










CHAPTER 3
LITERATURE REVIEW

This chapter will review the necessary literature to support the rationale of the study. This

will include discussion of: (1) The basal ganglia and its internal and external circuitry; (2) DBS

as treatment for PD and possible mechanisms of benefit; (3) Motor planning/programming with

an emphasis on maintaining and switching motor programs, (4) The possibility of motor

planning/programming deficits (including speech) in PD; and (5) The use of reaction time

experiments to test aspects of motor planning/programming.

The Basal Ganglia

The basal ganglia comprise several subcortical nuclei critically involved in movement

and posture. Although this study emphasizes the role of the basal ganglia in movement

(specifically its planning/programming), it should be recognized that modern conceptualization

of basal ganglia function emphasizes a number of functionally segregated circuits involved in a

number of diverse motor, cognitive, and limbic functions (Alexander, DeLong & Strick, 1986).

Due to the intrinsic/extrinsic basal ganglia circuitry, these structures have the opportunity to

influence diverse cortical areas including those thought to be involved in speech motor

planning/programming.

Anatomy

First, the normal basal ganglia anatomy will be described. The basal ganglia are often

conceptualized as a group of input structures, output structures, and intrinsic nuclei. The two

primary input structures are the striatum (comprised of the caudate and putamen) and the

subthalamic nucleus (STN). The two primary output structures are globus pallidus internus (GPi)

and substantial nigra pars reticulata (SNpr). The intrinsic nuclei of the basal ganglia include

globus pallidus pars externa (GPe) and substantial nigra pars compact (SNpc) (Mink, 1996).










Surgical treatments are usually performed in those patients with advanced PD who have

disabling motor symptoms that are insufficiently controlled with medical management. GPi and

STN DBS are recognized to have a beneficial effect on symptoms such as tremor, rigidity,

bradykinesia, dyskinesia, and postural/gait abnormalities in patients with PD who are

insufficiently managed with pharmacological therapy (Benabid, 2003; Gross, 2004; Rodriguez-

Oroz et al., 2005; Volkmann, 2004). Also important is the absence of significant cognitive

impairments, an understanding of the surgical risks, and realistic post-operative expectations

(Marks Jr., 2005; Vitek & Walter, 2005). For GPi or STN DBS, the best predictor of outcome

seems to be a patient's response to levodopa (Benabid, 2003).

Mechanisms of DBS

The exact neurophysiological mechanisms for the improvement in motor

symptomatology with DBS in PD are unknown. Due to the similarities in improvements in motor

functioning following ablative lesioning procedures, it has been suggested that "DBS acts as a

transient electrical inactivation or reversible lesion to block the output of dysfunctional targets"

(Lozano, Dostrovsky, Chen, & Ashby, 2002, p. 226). However, current conceptualization of the

mechanisms of DBS suggests that this is a vast oversimplification (Lozano et al., 2002;

Desbonnet et al., 2004; Grill, Snyder, & Miocinovic, 2004). For example, Temel and colleagues

(2005) state that "an increasing amount of data suggests that categorizing DBS as being

'inhibitory' and thus equating its effects to those of a lesion.. is...an oversimplification of what

is a highly complex and multi-faceted technique" (p. 397). Lozano and colleagues (2002) review

multiple possible mechanisms of DBS including facilitative, inhibitory, and downstream effects.

Facilitative effects likely include activation of large axons. Inhibitory effects may include partial

or complete blocking of neuronal firing, possibly due to depolarization and/or the release of










CHAPTER 1
INTRODUCTION

Primary Aims

The primary purpose of this study was to measure the effects of deep brain stimulation

(DBS) on maintaining and switching speech motor programs in individuals with Parkinson's

disease (PD) and hypokinetic dysarthria using a speech reaction time (SRT) paradigm. Recent

literature suggests that at least a portion of the underlying mechanism of hypokinetic dysarthria

in individuals with PD may be related to deficits in speech motor planning/programming,

including maintaining and switching motor programs (Spencer & Rogers, 2005; Van der Merwe,

1997). Although the effects of DBS on speech motor planning/programming have not, to our

knowledge been previously explored, DBS has been shown to positively influence maintaining

and switching motor programs in the limbs and we posited that DBS would similarly improve

motor speech program maintenance and switching. In order to test the influence of DBS on these

processes, two priming conditions were tested (i.e., 'switch' or 'no switch') in both DBS states

(i.e., 'on' and 'off stimulation). The primary dependent variable was SRT.

The following null hypothesis was addressed:

There is no significant difference in SRT across DBS states (i.e., 'on' and 'off" stimulation) or

priming condition (i.e., 'switch' and 'no switch').

Research Question 1

Are there significant differences in SRT when subjects with PD and DBS produce a

word in the 'switch' and 'no switch' conditions?

It was predicted that SRT will be faster when producing a word in the 'no switch' versus

'switch' condition, regardless of DBS state. This was expected due to the increased complexity

of the 'switch' condition on processes involved in speech motor programming/planning.










Participant Description

A speech evaluation was completed including maximum performance testing of the

speech mechanism (Kent, Kent, & Rosenbek, 1987), repetition of multisyllabic words and

sentences, elicitation of a connected speech sample, and determination of intelligibility using the

short form of the SIT. All components of the speech evaluation were recorded using a high-

quality digital audio recorder (Marantz PMD671) and a head-mounted microphone (Shure

SM10A) positioned two centimeters from the left corner of the subject's mouth. The

Communicativeness Effectiveness Survey (CES), an eight-item questionnaire using a seven-

point scale in which individuals make judgments about their ability to communicate effectively

during everyday activities, was also administered (Donovan, Velozo, & Rosenbek, in press).

A speech diagnosis regarding the presence, type, and severity of dysarthria was later

determined for each of the participants based on acoustic recordings of maximum performance

testing, word and sentence repetition, and connected speech samples. Two speech-language

pathologists (SLPs) (JR & HJ) experienced in the evaluation of neurogenic speech disorders used

perceptual assessment to independently determine whether dysarthria was present and, if so, the

type and severity based on a seven-point scale (see Table 2-1) using the Mayo Clinic

classification terminology (Darley, Aronson, & Brown, 1969a, b, 1975; Duffy, 2005). Any

differences in speech diagnosis or severity of dysarthria led to re-listening and debate to make a

final consensus decision.

To determine intelligibility, acoustic recordings of sentences from the SIT were presented

via headphone to two undergraduate students with normal hearing who served as intelligibility

scorers. These individuals were inexperienced in communicating with individuals with

dysarthria. Each sentence was presented at a fixed volume to each scorer two times with a three










Brooks, V. B. (1986). The neural basis of motor control. New York: Oxford University
Press.
Brown, R. G., Dowsey, P. L., Brown, P., Jahanshahi, M., Pollak, P., Benabid, A. L., et al.
(1999). Impact of deep brain stimulation on upper limb akinesia in Parkinson's disease. Annals of
Neurology, 45, 473-288.

Chambers, C. D., Bellgrove, M. A., Stokes, M. G., Henderson, T. R., Garavan, H.,
Robertson, I. H., et al. (2006). Executive "brake failure" following deactivation of human frontal
lobe. Journal of Cognitive Neuroscience, 18, 444-455.

Contreras-Vidal, J. L., & Stelmach, G. E. (1996). Effects of Parkinsonism on motor
control. Life Sciences, 58, 165-176.

Daniele, A., Albanese, A., Contarino, M. F., Zinzi, P., Barbier, A., Gasparini, F., et al.
(2003). Cognitive and behavioral effects of chroinc stimulation on cognitive function in
Parkinson's disease. Journal ofNeurology, Neurosurgery, & Psychiatry, 74, 175-182.

Darley, F. L., Aronson, A. E., & Brown, J. R. (1969a). Clusters of deviant speech
dimensions in the dysarthrias. Journal of Speech and Hearing Research 12, 462-496.

Darley, F. L., Aronson, A. E., & Brown, J. R. (1969b). Differential diagnostic patterns of
dysarthria. Journal of Speech and Hearing Research, 12, 246-269.

Darley, F. L., Aronson, A. E., & Brown, J. R. (1975). Motor Speech Disorders.
Philadelphia: W. B. Saunders Company.

Davis, K. D., Taub, E., Houle, S., Lang, A. E., Dosdrovsky, J. O., Tasker, R. R., et al.
(1997). Globus pallidus stimulation activates the cortical motor system during alleviation of
parkinsonian symptoms. Nature Medicine, 3, 671-674.

De Gaspari, D., Siri, C., Di Gioia, M., Antonini, A., Isella, V., Pizzolato, A., et al. (2006).
Clinical correlates and cognitive underpinnings of verbal fluency impairment after chronic
subthalamic stimulation in Parkinson's disease. Parkinsonism andRelatedDisorders, 12,
289-295.

The deep-brain stimulation for Parkinson's disease study group. (2001). Deep-brain
stimulation of the subthalamic nucleus or the pars internal of the globus pallidus in Parkinson's
disease. New England Journal of Medicine, 345, 956-963.

DeLong, M. R. (1990). Primate models of movement disorders of basal ganglia origin.
Trends in Neuroscience, 13, 281-285.

DeLong, M. R., & Wichmann, T. (2007). Circuits and circuit disorders of the basal
ganglia. Neurological Review, 64, 20-24.











TABLE OF CONTENTS



A C K N O W L E D G M E N T S ..............................................................................................................4

LIST OF TABLES .................. .............. ............................

LIST O F FIG U RE S ................................................................. 9

A B S T R A C T ........................................... ................................................................ 1 1

CHAPTER

1 INTRODUCTION ............... ............................ .............................. 13

P rim ary A im s ........................................................................... 13
R research Q question 1 ............................................ .. .. .......... ..... ...... 13
R research Q question 2 .................. ................................ ........ .. ............ 14
Research Question 3 ......................................... .......... ..............14
Secondary and Exploratory A im s............................................................................ ...... 14

2 M A TER IA L S A N D M ETH O D S ........................................ .............................................15

E xperim ent O overview .....................................................................15
P artic ip an ts .........................................................................1 5
S u b je cts ................... ...................1...................5..........
Inclusion Criteria ..................................................................... ......... 15
Exclusion C riteria................................................... 16
Screening Session ................................................................ ..... ..... ......... 16
E lig ib ility D term in action ........................................................................................... 16
Participant D description .............................................................................. 17
Experimental Sessions ....................................................... ........... ............... 18
Experim mental Procedures ................................... ................ .......................19
T rain in g session ................................................................................ 2 0
S tim u li ................... ...................2...................0.........
E q u ip m en t ................... ...................2...................1..........
S c o rin g ................................ .............................................................................................. 2 2
Speech Reaction Tim e ............................................................................. 22
R e sp o n se A ccu racy ................................................................................................... 2 2
R liability ........................................................... 23
In tra-rater R eliab ility ................................................................ ...............................2 3
In ter-rater R eliab ility ................................................................ ...............................2 4
D ata A nalysis................... .................................. ...... .. .. ............. 24
Sam ple Size and Pow er Consideration ........................................................................ ... 24










Exclusion Criteria

Exclusion criteria included history positive for previous neurosurgery for PD (e.g.,

pallidotomy or thalamotomy), DBS surgery completed at an outside institution, thalamic DBS,

recent (less than three months previous) or significant stroke, Mini-Mental State Examination

score (MMSE) < 26, and Wechsler Memory Scale 3rd Edition (WMS-III) Spatial Span Forward

or Backward Subtests standard score < 7, history or presence of aphasia, inability to discontinue

anti-PD medication overnight and during the test session, inability to discontinue DBS for at

least six hours, inability to perform the study tasks for reasons such as an incapacity to read

words and sentences aloud or produce intelligible speech, severe motor symptoms causing

extreme difficulty/inconvenience when medications are withheld and/or DBS device is turned

off, or other significant medical illness that prevents meaningful participation in the study.

Screening Session

In order to determine eligibility and to further describe participants, a screening visit was

first conducted. Screenings sessions were primarily conducted at the subject's homes and the

Speech and Hearing Center at the University of Florida.

Eligibility Determination

During the screening visit, informed consent was first obtained. Next, participants were

asked questions about their demographic information and current and past medical health status

for inclusion/exclusion purposes. This was followed by administration of the MMSE and the

WMS-III Spatial Span Forward and Backward Subtests to screen general cognitive function,

attention and concentration, and working memory. Ability to read sentences aloud was

determined during administration of the Sentence Intelligibility Test (SIT) (see "Participant

description" below for more details).









APPENDIX A
TRAINING SESSION STIMULI


Hill-HillB

Beat-Beat


4 Hum-Hum
5 Lady-Lady

7 Mall-Mall
89 Hea
9 Heat-Heat


10 Rowing-Rowing


Nut-Nut
Howing-Howing
Heat-Heat

Dill-Dill

Ship-Ship
Men-Men

Sheet-Sheet


1
2
3


E = Switch = No switch


I Sim St St


I









































Figure 3-2. The intrinsic and extrinsic circuitry of the basal ganglia under normal conditions.
SMA= supplementary motor area, PMC = premotor cortex, MC = motor cortex, SNpc =
substantial nigra pars compact, Dl = striatal output receptor type D1, D2 = striatal output
receptor type D2, GPe = globus pallidus pars externa, STN = subthalamic nucleus, GPi =
globus pallidus internus, SNpr = substantial nigra pars reticulata, VA = ventral anterior nucleus
of the thalamus, VL = ventral lateral nucleus of the thalamus, CM = centrum medianum.










CHAPTER 5
DISCUSSION

The following discussion of our research findings will be outlined as related to the

primary, secondary, and exploratory research questions detailed in Chapter 1. This will be

followed by a more general discussion including the strengths and weaknesses of the study and

alternative hypotheses. Directions for future research will be covered in Chapter 6.

Primary Aims

Research Question 1

Research question 1 was designed to determine whether a significant difference exists in

SRT between the 'switch' and 'no switch' conditions in patients with PD and DBS. While both

conditions were hypothesized to target the level of speech motor planning/programming, the

'switch' condition was expected to require additional, more complex processes for successful

completion (i.e., inhibition of the unwanted motor program, retrieval of new subprograms from

the sensorimotor store, and the loading of these subprograms into the buffer in order to construct

a new motor program). Consistent with our predictions, we found subjects responded

significantly faster in terms of SRT in the 'no-switch' condition regardless of whether DBS was

'on' or 'off. Collapsed across stimulation condition, mean SRT for the 'no switch' task was

643.31 ms and 722.88 ms. for the 'switch' task.

These data validate the response priming paradigm and support the critical notion that the

'switch' condition is more complex than the 'no switch' condition due to the increased demands

of this task on processes involved in speech motor planning/programming. These data support

use of this paradigm to measure processes involved in speech motor planning/programming.

Additionally, these findings are consistent with the modern RT literature in which a variety of

variables, including complexity, are known to influence reaction time (Henry & Rogers, 1960;










of the patient's mouth. Digital video recordings (Sanyo VPC-C6) were also made during all

experimental tasks.

Scoring

Speech Reaction Time

SRT was measured by E-Prime computer software and the voice key of the SRB. Only

correct productions were used to calculate SRT. Subjects were trained to not produce a speech

response until the target word was presented. SRT was measured starting with the command for

movement (i.e., presentation of the target) until the onset of voicing registered for each response.

Responses during the pause or less than 250ms after presentation of the target were scored as a

premature response. Premature responses were considered to be incorrect and were combined

with other errors from the response accuracy assessment to determine the total number of

incorrect responses.

Response Accuracy

Acoustic recordings from the response priming test were used to determine the accuracy

of responses over two sessions. During the first session, judge one (HJ) independently listened to

all responses using high quality headphones (Sony MDR-V6). Each response was listened to as

many times as necessary to be able to use broad phonetic transcription the record the responses

and determine if a production was correct or incorrect. During the second session, a second judge

(DK) was added and this process was completed with both judges simultaneously listening to the

acoustic recordings with headphones. When the two judges did not agree on whether a response

was correct or incorrect, a consensus decision was reached with further listening and debate.

Agreement between the judges was not considered to occur in cases where further listening










Motor Program Switching

The ability to rapidly switch motor programs has also received significant experimental

attention. In such experiments, subjects are (often unexpectedly) presented with a prime stimulus

that does not accurately or completely inform the target stimulus. Viewed through the model of

Sternberg and colleagues, this task requires subjects to clear the prepared (and inaccurate) motor

program in the buffer, search the sensorimotor store for the newly required motor subprograms,

and load the motor program into the buffer before execution. Switching motor programs requires

increased RT due to the increased complexity of this task. Difficulty switching has been

hypothesized to be related to impairments inhibiting or modifying a motor program (Inzelberg et

al., 2001; Kropotov & Etlinger, 1999; Mink, 1996) or in the activation of a new motor program

(Haaland & Harrington, 1991).

Deviant Motor Programming in PD

In individuals with PD, deficits in aspects of motor programming are being recognized

with increasing regularity, both in the limbs and, less commonly, the speech mechanism.

Limb Motor Programming Maintenance and Switching

A substantial literature supporting the concept of impaired maintenance of motor

programs for limb movements in PD is available (Gentilucci and Negrotti, 1999a, b; Gueye,

Viallet, Legallet & Trouche, 1998; Jones, Phillips, Bradshaw, lansek, & Bradshaw, 1992;

Pascual-Leone, Valiis-Sole, Brasil-Neto, Cohen & Hallett, 1994; Romero, Van Gemmert, Adler,

Bekkering, & Stelmach, 2003; Sheridan, Flowers, & Hurrell, 1987; Stelmach, Garcia-Colera &

Martin, 1989), though this finding has not been replicated by all studies (Labutta, Miles, Sanes,

& Hallett, 1994). PD patients have also been found to have disordered ability to switch motor

programs (Benecke, Rothwell, Dick, Day & Marsden, 1998; Contreras-Vidal & Stelmach, 1996;










condition under the 'on' DBS state but this difference was not significant). Fisher's combination

method was used to determine overall differences in response accuracy across DBS conditions

and results were not significant (Fisher's combination test statistic = 6.82, p = 0.1526).

Separate Wilcoxon signed rank tests revealed no significant differences in response

accuracy in the 'no switch' (signed rank = 14, p = 0.1816) or 'switch' (signed rank = 11.5, p=

0.7188) conditions when 'on' versus 'off DBS states were compared. In other words, no

difference in response accuracy was observed in the 'no switch' or 'switch' conditions regardless

of whether DBS was 'on' or 'off'. Fisher's combination was used to determine overall

differences in response accuracy across DBS conditions and results were not significant (Fisher's

combination test statistic = 4.07, p = 0.3881).

Reliability

Intra-rater Reliability

Judge one completed perceptual assessment to determine the accuracy of each response

on two separate occasions. Point-by-point analysis was conducted for each subject's responses

during each of their two test sessions to determine whether agreement regarding accuracy was

present. The kappa coefficient has the value 0.79, which indicates strong agreement between the

separate rating sessions and the confidence interval of (0.70, 0.87) confirms that one can reject

the null hypothesis of no agreement. Additionally, the percentage of task items agreed in the two

occasions range from 86% to 100% for the twelve subjects. Intra-class correlation coefficient

was determined to be 0.82 with a 95% confidence interval [0.71, 0.90].

Inter-rater Reliability

A point-by-point analysis was completed to compare the level of agreement between two

judges as to whether each individual response was correct or incorrect. The kappa coefficient has










Spencer, K. A. (2006, March). Effect of medication withdrawal on response preparation
in Parkinson's disease: Preliminary evidence. Poster presentation at the Speech Motor Control
conference, Austin, TX.

Spencer, K. A., & Rogers, M. A. (2005). Speech motor programming in hypokinetic and
ataxic dysarthria. Brain and Language, 94, 347-366.

Stelmach, G. E., Garcia-Colera, A., & Martin, Z. E. (1989). Force transition control
within a movement sequence in Parkinson's disease. Journal ofNeurology, 236, 406-410.

Sternberg, S., Knoll, R. L., Monsell, S., & Wright, C. E. (1988). Motor programs and
hierarchical organization in the control of rapid speech. Phonetica, 45, 175-197.

Sternberg, S., Knoll, R. L., & Turock, D. L. (1990). Hierarchical control in the execution
od action sequences: Tests of two invariance properties. In M. Jeannerod (Ed.) Attention and
performance XIII: Motor representation and control (pp. 3-55). Hillsdale, NJ: Lawrence
Erlbaum Associates.

Sternberg, S., Monsell, S., Knoll, R. L., & Wright, C. E. (1978). The latency and duration
of rapid movement sequences: Comparisons of speech and typewriting. In G. E. Stelmach (Ed.)
Information processing in motor control and learning (pp. 117-152). New York: Academic
Press.

Sternberg, S., Wright, C. E., Knoll, R. L., & Monsell, S. (1980). Motor programs in rapid
speech: Additonal evidence. In R. A. Cole (Ed.) Perception and production affluent speech
(pp.507-534). Hillsdale, NJ: Lawrence Erlbaum Associates.

Temel, Y., Blokland, A., Ackermans, L., Boon, P., van Kranen-Mastenbroek, V. H. J. N.,
Beuls, E. A. M., et al. (2005). Differential effects of subthalamic nucleus stimulation in advanced
Parkinson disease on reaction time performance. Experimental Brain Research, 169, 389-399.

Temel, Y., Visser-Vandewalle, V., Aendekerk, B., Rutten, B., Tan, S., Scholtissen, B., et
al. (2005). Acute and separate modulation of motor and cognitive performance in parkinsonian
rats by bilateral stimulation of the subthalamic nucleus. Experimental Neurology, 193, 43-52.

Thobois, S., Mertens, P., Guenot, M., Hermier, M., Mollion, H., Bouvard, M., et al.
(2002). Subthalamic nucleus stimulation in Parkinson's disease: Clinical evaluation of 18
patients. Journal ofNeurology, 249, 529-534.
Tombaugh, T. N., Kozak, J., & Rees, L. (1999). Normative data stratified by age and
education for two measures of verbal fluency: FAS and animal naming. Archives of Clinical
Neuropsychology, 14, 167-177.

Tomqvist, A. L., Schalen, L., & Rehncrona, S. (2005). Effects of difference electrical
parameter settings on the intelligibility of speech in patients with Parkinson's disease treated
with subthalamic deep brain stimulation. Movement Disorders, 20, 416-423.

PAGE 1

1 EFFECTS OF DEEP BRAIN STIMULATION ON SPEE CH MOTOR PLANNING/ PROGRAMMING IN PATIENTS WITH PARKINSONS DISEASE By HARRISON N. JONES A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2007

PAGE 2

2 2007 Harrison N. Jones

PAGE 3

3 To my extraordinary wife Carlee your stead fast love, support, and encouragement have allowed me to achieve my dreams

PAGE 4

4 ACKNOWLEDGMENTS I thank all of the countless individuals a nd institutions who have supported me during completion of my Ph.D. For thei r steadfast love and support, I thank my wife Carlee, my son Nate, my mother Linda, and the memory of my father Mike. Fo r their professional and personal mentoring, I thank the all the members of my dissertation committee Drs. John C. Rosenbek, Diane L. Kendall, Michael S. Okun, and Craig Velo zo. I would further like to extend my deepest gratitude to Dr. Rosenbek for all he has taught me about science and life. I will forever be indebted to him for his kind but intellectually rigorous ment orship throughout my years at the University of Florida (UF). Perhaps someday in the future, I can repay the debt in kind to another. I thank Sam Wu for his guidance and assi stance with my statistical analyses and Hubert Fernandez for his expertise in experimental de sign. I thank Nan Musson, Jim Korner & Leslie Gonzalez-Rothi for their support during completion of my degree. I thank Chris Edghill for his expertise in the graphic design of the figures contained within this work. I thank all of my research assistants Kaitlin Kobaitri, Bashar Mourad, Kristin Shaffer, Tina Tso, & Lauren Laube for their volunteer work which was so criti cal for the success of this project. Finally, for their support of my education a nd research, I thank the Office of Research and Development, Rehabilitation, R&D Service, Department of Vetera ns Affairs, the Brain Rehabilitation Research Center at the Malcom Randall VAMC, the UF Movement Disorders Center, the UF Department of Communicative Disorders, and the faculty, staff, and my fellow students in the UF Rehabilitation Science Doctoral Program.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........8 LIST OF FIGURES.........................................................................................................................9 ABSTRACT...................................................................................................................................11 CHAPTER 1 INTRODUCTION................................................................................................................. .13 Primary Aims................................................................................................................... .......13 Research Question 1........................................................................................................13 Research Question 2........................................................................................................14 Research Question 3........................................................................................................14 Secondary and Exploratory Aims...........................................................................................14 2 MATERIALS AND METHODS...........................................................................................15 Experiment Overview............................................................................................................ .15 Participants.............................................................................................................................15 Subjects............................................................................................................................15 Inclusion Criteria.............................................................................................................15 Exclusion Criteria............................................................................................................16 Screening Session.............................................................................................................. .....16 Eligibility Determination.................................................................................................16 Participant Description....................................................................................................17 Experimental Sessions............................................................................................................18 Experimental Procedures........................................................................................................ 19 Training session............................................................................................................... .......20 Stimuli........................................................................................................................ .............20 Equipment...............................................................................................................................21 Scoring....................................................................................................................................22 Speech Reaction Time.....................................................................................................22 Response Accuracy.........................................................................................................22 Reliability...............................................................................................................................23 Intra-rater Reliability.......................................................................................................2 3 Inter-rater Reliability.......................................................................................................2 4 Data Analysis..........................................................................................................................24 Sample Size and Power Consideration...................................................................................24

PAGE 6

6 3 LITERATURE REVIEW.......................................................................................................32 The Basal Ganglia.............................................................................................................. .....32 Anatomy..........................................................................................................................32 Normal Basal Ganglia Intrinsic/Extrinsic Circuitry........................................................33 Circuitry in PD................................................................................................................ 34 Deep Brain Stimulation..........................................................................................................35 DBS in PD...................................................................................................................... ........36 Mechanisms of DBS...............................................................................................................37 DBS and speech......................................................................................................................38 A Model of Speech Motor Control.........................................................................................41 Four Phase Model............................................................................................................41 Linguistic Symbolic Planning.........................................................................................41 Motor Planning................................................................................................................4 2 Motor Programming........................................................................................................43 Execution.........................................................................................................................43 Motor Planning/Programming................................................................................................44 Subprogram Retrieval Model.................................................................................................45 Motor Program Maintenance...........................................................................................46 Motor Program Switching...............................................................................................47 Deviant Motor Programming in PD.......................................................................................47 Limb Motor Programming Maintenance and Switching.................................................47 Speech Motor Programming Maintenance and Switching..............................................48 Reaction Time/Speech Reaction Time...................................................................................49 Simple and Choice Reaction Time..................................................................................50 SRT in the No Switch Condition is a Measure of Speech Motor Program Maintenance.................................................................................................................51 SRT in the Switch Condition is a Meas ure of Speech Motor Program Switching.......51 4 RESULTS...................................................................................................................... .........57 Participants.............................................................................................................................57 Experimental Results..............................................................................................................58 Primary Aims................................................................................................................... 58 Research question 1 (noswitch vs. switch)........................................................58 Research question 2 (no-switch condition on vs. off DBS).............................59 Research question 3 (switch c ondition on vs. off DBS)..................................59 Secondary Aims...............................................................................................................59 Reliability...............................................................................................................................60 Intra-rater Reliability.......................................................................................................6 0 Inter-rater Reliability.......................................................................................................6 0 5 DISCUSSION................................................................................................................... ......67 Primary Aims................................................................................................................... .......67 Research Question 1........................................................................................................67

PAGE 7

7 Research Question 2........................................................................................................69 Research Question 3........................................................................................................70 Secondary Aims................................................................................................................. .....73 Exploratory Aims............................................................................................................... .....77 FAS Test..........................................................................................................................78 Stroop Color and Word Test............................................................................................79 Strengths/Weaknesses........................................................................................................... ..80 Alternative Explanations....................................................................................................... .82 Discussion Conclusion.......................................................................................................... ..83 6 FUTURE WORK.................................................................................................................. ..86 APPENDIX A TRAINING SESSION STIMULI...........................................................................................89 B EXPERIMENTAL STIMULI................................................................................................90 LIST OF REFERENCES...............................................................................................................91 BIOGRAPHICAL SKETCH.......................................................................................................102

PAGE 8

8 LIST OF TABLES Table page 2-1 Auditory-perceptual dysarthria severity scale...................................................................26 2-2 Power analysis to detect differences between switch and no switch speech responses...................................................................................................................... ......27 2-3 Powers corresponding to r, ratio of mean SRT between the on and off stimulation conditions, based on sensitivity analysis...........................................................................28 4-1 Individual and group descriptive data................................................................................62 4-2 Summary statistics for speech reaction time (SRT) and response accuracy by priming condition and stimulation state. ........................................................................................63 4-3 Mean difference and p-values for spe ech reaction time (SRT) and response accuracy by priming condition and stimulation state........................................................................64

PAGE 9

9 LIST OF FIGURES Figure page 2-1 Typical experimental timeline...........................................................................................29 2-2 Response priming protocol................................................................................................30 2-3 Equipment configuration...................................................................................................31 3-1 Basal ganglia structur es and surrounding areas ................................................................53 3-2 The intrinsic and extrinsic circuitry of the basal ganglia under normal conditions...........54 3-3 The intrinsic and extrinsic circuitry of the basal ganglia in Parkinsons diseease.............55 3-4 Unilateral deep brain stimulation (DBS)...........................................................................56 4-1 Mean speech reaction time (SRT) in milliseconds (ms) in the no switch and switch conditions...........................................................................................................................65 4-2 Mean speech reaction time (SRT) in milliseconds (ms) in the no switch and switch conditions on and off DBS.............................................................................66 5-1 Distribution of errors di fferentiated among error type......................................................85

PAGE 10

10 LIST OF ABBREVIATIONS CES Communicativeness Effectiveness Survey CMA Cingulate motor area DBS Deep brain stimulation GPe Globus pallidus pars externa GPi Globus pallidus internus IC Internal capsule MC Motor cortex MMSE Mini-Mental State Examination PD Parkinsons disease PET Positron emission tomography PMC Premotor cortex RT Reaction time SIT Sentence Intelligibility Test SLPs Speech-language pathologists SMA Supplementary motor area SNpc Substantia nigra pars compacta SNpr Substantia nigra pars reticulata SRB Serial Response Box SRT Speech reaction time STN Subthalamic nucleus UF University of Florida UPDRS Unified Parkinsons Disease Rating Scale WMS-III Wechsler Memory Scale 3rd Edition

PAGE 11

11 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy EFFECTS OF DEEP BRAIN STIMULATION ON SPEE CH MOTOR PLANNING/ PROGRAMMING IN PATIENTS WITH PARKINSONS DISEASE By Harrison N. Jones December 2007 Chair: John C. Rosenbek Major: Rehabilitation Science The primary purpose of this study was to meas ure the effects of d eep brain stimulation (DBS) on maintaining and switching speech motor programs in individuals with Parkinsons disease (PD) and hypokinetic dysart hria. Recent literature suggests that at least a portion of the underlying mechanism of hypokinetic dysarthria in individuals with PD may be related to deficits in speech motor planning/program ming, including maintaining and switching motor programs (Spencer & Rogers, 2005; Van der Merwe, 1997). Although the effects of DBS on speech motor planning/programming have not been previously explored, DBS has been shown to have a positive influence on these processes in the limbs and we posited that DBS would similarly benefit speech maintenance and switching. A reaction time paradigm was employed to measure the effects of DBS on maintaining and switching of speech motor programs in i ndividuals with PD. Double blind testing was completed in the on and off DBS conditions using a response priming procedure in which participants were provided with a prime word to supply information regard ing target word. Over a series of targets, the prime was followed w ith a high probability by the primed target as

PAGE 12

12 expected (no-switch condition) or with a lo w probability by an unexpected target word (switch condition). The primary dependent m easure was SRT on and off DBS. Twelve participants completed the study. Stat istically significant differences (p < 0.05) were found in SRT between the no switch and switch conditions, regardless of DBS state. Significant differences were al so found in SRT in the no sw itch condition (i.e., subjects produced a word more quickly on versus off stimulation). No differences across stimulation conditions in the switch condition were obser ved. These findings sugg est that the greater complexity of the switch condition require s increased speech mo tor planning/programming processes which can be measured temporally. DBS was also found to improve SRT in the no switch condition, suggesting that maintenance of speech motor programs may be improved by DBS. No differences between DBS states were found in the switch c ondition, suggesting that DBS has little influence on the multiple processe s involved in a motor program switching task.

PAGE 13

13 CHAPTER 1 INTRODUCTION Primary Aims The primary purpose of this study was to measure the effects of d eep brain stimulation (DBS) on maintaining and switching speech motor programs in individuals with Parkinsons disease (PD) and hypokinetic dysarthria using a speech reaction time (S RT) paradigm. Recent literature suggests that at least a portion of the underlying me chanism of hypokinetic dysarthria in individuals with PD may be related to deficits in speech motor planning/programming, including maintaining and switching motor programs (Spencer & Rogers, 2005; Van der Merwe, 1997). Although the effects of DBS on speech motor planning/programming have not, to our knowledge been previously explor ed, DBS has been shown to positively influence maintaining and switching motor programs in the limbs a nd we posited that DBS wo uld similarly improve motor speech program maintenance and switching. In order to test the influence of DBS on these processes, two priming conditions were tested (i.e., switch or no switch) in both DBS states (i.e., on and off stimulation). The primary dependent variable was SRT. The following null hypothesis was addressed: There is no significant difference in SRT across DBS states (i.e., on and off stimulation) or priming condition (i.e., switch and no switch). Research Question 1 Are there significant differences in SRT when subjects with PD and DBS produce a word in the switch and no switch conditions? It was predicted that SRT will be faster when producing a word in the no switch versus switch condition, regardless of DB S state. This was expected due to the increased complexity of the switch condition on pr ocesses involved in speech motor programming/planning.

PAGE 14

14 Research Question 2 Are there significant differences in SRT in the on versus off DBS state when participants with PD produce a wo rd in the no switch condition? It was predicted that partic ipants will have improved SRT in single words in the no switch condition when on versus off DBS. This was expected due improved maintenance of the speech motor program in th e on stimulation condition. Research Question 3 Are there significant differences in SRT in the on versus off DBS state when participants with PD produce a word in the switch condition? It was predicted that SRT will be improved in single words in the switch condition in the on versus off stimulati on condition. This was expected due to improved ability to switch speech motor programs in the on stimulation condition. Secondary and Exploratory Aims The secondary aims of this study were to determine the effects of the experimental manipulations on speech response accuracy. It was expected that response accuracy would differ in response to priming condition. That is, more errors were anticipated in the switch versus the no switch condition. No difference in respons e accuracy was expected in response to DBS state. For exploratory purposes, measures of neuropsychological performance (i.e., verbal fluency and response inhi bition) were conducted both on and off DBS.

PAGE 15

15 CHAPTER 2 MATERIALS AND METHODS Experiment Overview A reaction time (RT) paradigm was empl oyed to measure the effects of DBS on maintaining and switching of speech motor pr ograms in individuals with PD and hypokinetic dysarthria. Double blind testing wa s completed in the on and o ff DBS states using a response priming procedure in which participants were provided with a prime to supply information regarding target. Over a series of targets, th e prime word was followed with a high probability by the primed target as expected (no-switch condi tion) or with a low probability by an unexpected target word (switch condition). The task of the subjects was to speak the target word aloud as quickly and accurately as possible. The prim ary dependent measure was SRT. Participants Subjects Twelve participants with PD and DBS co mpleted the study. Subjects were recruited through the Movement Disorders Clinic at the Un iversity of Florida (UF) and the Speech and Hearing Center at UF based on the following criteria: Inclusion Criteria Inclusion criteria included patients age 25 85 years old with a dia gnosis of probable idiopathic PD as determined by a neurologist with expertise in the evaluation of movement disorders, six months to two years status-post unilateral left or bilateral GPi or STN DBS, medically optimized and stable on anti-PD and psyc hotropic medications for at least 30 days at the time of the screening visit, the ability to read words and sentences aloud, and completion of the informed consent to participate in the study.

PAGE 16

16 Exclusion Criteria Exclusion criteria included history positiv e for previous neurosurgery for PD (e.g., pallidotomy or thalamotomy), DBS surgery comple ted at an outside inst itution, thalamic DBS, recent (less than three months previous) or si gnificant stroke, Mini-Mental State Examination score (MMSE) < 26, and Wechsler Memory Scale 3rd Edition (WMS-III) Spatial Span Forward or Backward Subtests standard score < 7, history or pres ence of aphasia, inability to discontinue anti-PD medication overnight and during the test sessi on, inability to discontinue DBS for at least six hours, inability to perf orm the study tasks for reasons su ch as an incapacity to read words and sentences aloud or produce intellig ible speech, severe motor symptoms causing extreme difficulty/inconvenience when medications are withheld and/or DBS device is turned off, or other significant medical illness that pr events meaningful participation in the study. Screening Session In order to determine eligibility and to furt her describe participants a screening visit was first conducted. Screenings sessi ons were primarily conducted at the subjects homes and the Speech and Hearing Center at the University of Florida. Eligibility Determination During the screening visit, informed consent was first obtained. Next, participants were asked questions about their demographic informati on and current and past medical health status for inclusion/exclusion purposes. This was followed by administration of the MMSE and the WMS-III Spatial Span Forward and Backward Subt ests to screen general cognitive function, attention and concentration, and working me mory. Ability to read sentences aloud was determined during administration of the Sentence Intelligibility Test (SIT) (see Participant description below for more details).

PAGE 17

17 Participant Description A speech evaluation was completed including maximum performance testing of the speech mechanism (Kent, Kent, & Rosenbek, 1987), repetition of multisyllabic words and sentences, elicitation of a connected speech samp le, and determination of intelligibility using the short form of the SIT. All components of the speech evaluation were recorded using a highquality digital audio recorder (Marantz PMD671) and a head-mounted microphone (Shure SM10A) positioned two centimeters from the le ft corner of the subjects mouth. The Communicativeness Effectiveness Survey (CES), an eight-item questi onnaire using a sevenpoint scale in which individuals make judgments about their ability to communicate effectively during everyday activities, was also administered (Donovan, Velozo, & Rosenbek, in press). A speech diagnosis regarding the presence, ty pe, and severity of dysarthria was later determined for each of the participants based on acoustic recordings of maximum performance testing, word and sentence repetition, and connected speech samples. Two speech-language pathologists (SLPs) (JR & HJ) experienced in th e evaluation of neurogenic speech disorders used perceptual assessment to independe ntly determine whether dysarthr ia was present and, if so, the type and severity based on a seven-point sc ale (see Table 2-1) using the Mayo Clinic classification terminology (Darley, Arons on, & Brown, 1969a, b, 1975; Duffy, 2005). Any differences in speech diagnosis or severity of dysarthria led to re-listening and debate to make a final consensus decision. To determine intelligibility, aco ustic recordings of sentences from the SIT were presented via headphone to two undergradu ate students with normal hearing who served as intelligibility scorers. These individuals were inexperien ced in communicating with individuals with dysarthria. Each sentence was presented at a fixed volume to each scorer two times with a three

PAGE 18

18 second pause between presentations. They were asked to orthographically transcribe the sentences and enter them into the SIT program via a computer keypad. Scorers were encouraged to pause sentence playback as needed in order to accurately transcribe the sentences, although each sentence was heard only twice and no adjustments in volume were permitted. A discourse analysis was completed by two SL Ps experienced in discourse analysis (DK & HJ) based on acoustic recordings of repetition of multisyllabic words, repetition of sentences, and connected speech in order to determine linguistic competency. These productions were assessed for the presence of linguistic errors, specifically phonologic errors (i.e., substitutions, omissions, transpositions, etc.) and/ or semantic or verbal errors. Any differences in analysis led to re-listening and debate in or der to make a final determination regarding the presence and type of linguistic errors. Experimental Sessions Subjects who met all study criteria followi ng completion of the sc reening session were scheduled for the experimental session one to 30 days later. Please see Figure 2-1 for an example of the typical experimental timeline. Subject s were tested off their anti-PD medication, including levodopa. The off medi cation condition was defined as at least 12 hours off anti-PD medications. Testing was conducted w ith left-brain DBS in the on and off states. If subjects had bilateral DBS, the right-brain DBS was turn ed off for the duration of the experimental session. Two two-hour washouts of DBS were co mpleted during each experimental session during which all DBS therapy was discontinued. Two test sessions in which a battery of tests was administered (see Experimental Procedures below) while subjects were on and off DBS were completed for each participant. Follow ing the first DBS washout, subjects were quasi randomly assigned to the first stimulation condition (i.e., on or off DBS) in a counterbalanced

PAGE 19

19 fashion. Using the Medtronic Access Review device, a trained research assi stant turned subjects on or off allowing subjects a nd the principal investigator to remain blinded to the test condition. Thirty minutes later, the first te st session was initiated and completed in approximately 30 minutes. The second two-hour DBS washout was then started and this was followed by implementation of the second stimul ation condition by the research assistant. Subjects that were tested on DBS during then first test session were tested off DBS during the second test session, and vice versa. The second test session was al so started 30 minutes after the condition was initiated and completed in approximately 30 minutes. Fo llowing completion of both test sessions, all subjects were turned on DBS and took their anti-PD medications. Experimental Procedures Experimental sessions were conducted primarily at the homes of the participants, as well as the UF Speech and Hearing Center in select cases. A response priming procedure based on the work of Spencer (Spencer & Rogers, 2006; Spen cer, 2006) was utilized. As shown in Figure 2-2, in this paradigm, participants are provided with a prime to supply inform ation regarding a target. Over presentation of a series of target words for speech production, the prime word was followed with a high probability (75% of trials) by the pr imed target as expected in the no switch condition, such as with the prime-target pair shopper-shopper. In 25% of trials, however, the subject was presented with an incorrect prime, discovered upon presentation of the command for movement (i.e., the target word). In other word s, the prime-target pair did not match in the switch condition (i.e., shopper-chopper). Subjects were trained to read the target word aloud as quickly and accurately as possible and were furt her instructed to be prepared to say the prime word due to the high likelihood it would match the upcoming target. Each trial began with a signal (i.e., +) on a computer sc reen followed by visual presentation of the prime word for

PAGE 20

20 1000ms. The prime was followed by a blank screen for 250ms and then the target word was presented for 1000ms. Response latency was measur ed temporally from the presentation of the target. Response accuracy was determined using br oad phonetic transcription scored online with later verification and analysis of errors. Furt her details can be found under the Scoring and Equipment sections of this chapter. Subjects were also administer ed two neuropsychological tests during each of the two test sessions: the FAS test of verbal fluency and the Stroop. Color and Word Test. During the FAS, participants were provided with one-minute to produce as many words as possible starting with each of the three letters F, A, and S (Benton & Hamsher, 1976). The Stroop has three sets of stimuli: color words printed in black ink, symbols (i.e., X) printed in color ink, and color words from the first set of stimuli pr inted with incongruous colors from the second set of stimuli. Subjects were asked to move th rough each set of stimuli reading words or naming colors as quickly as possible (Golden & Freshwater, 2002). Training session Prior to the first test session, participants were trained in the experimental task for approximately 10 minutes (see Appendix A for the training session stimuli). Following this training period, if a subject was unable to perform the experimental task, it was planned to discontinue testing, though this did not occur. Stimuli The prime and target words consisted of one and two-syllable words (see Appendix B) from Spencer (2006) (adapted from Spencer & R ogers, 2005). Word rimes were maintained from prime to target. It was required th at the onset of the prime needed to share two features with the target and be a highly marked phoneme. A balance for word frequency was maintained (Spencer

PAGE 21

21 & Rogers, 2005). Stimuli were linguistically controlled so that all effects were due to the experimental manipulation. Sixteen prime-target pairs were used in th e experiment. Each of these sixteen primetarget pairs was presented four times. Three of the presentations for each of the prime-target pairs were the no-switch condition (i.e., shopper-s hopper), while one of the four presentations was the switch condition (i.e., s hopper-chopper). This ensured that subjects expected the prime to accurately provide information regarding the target. Equipment The equipment configuration is shown in cartoon format in Figure 2-3. A laptop computer (Compaq Presario V2000) was used to run the E-Prime computer software (Psychology Software Tools) used in the respon se priming procedure. Stimuli were presented on a separate 19 computer monitor (Planar PL1910M) placed in front of the subject. Presentation of stimuli and calculation of response latenc ies were managed by the E-Prime program and a Serial Response Box (SRB) (Psychology Softwa re Tools). Registration of speech onset, as measured by the onset of voicing, was measured using an accelerometer (PCB Piezotronics 352C22) placed inferior to the thyroid cartilage with adhesive tape. The accelerometer was powered by a portable power source (PCB Piezotronics Model 480C02 ICP Signal Conditioner) and the signal captured by this tr ansducer activated the Voice Key of the Serial Response Box to measure response latency. High-quality acoustic recordings were ma de during administration of the response priming procedure, FAS, and Stroop using a high -quality digital record er (Marantz PMD671) and a head-mounted microphone (Shure SM10A) positioned two centimeters from the left corner

PAGE 22

22 of the patients mouth. Digital video recordi ngs (Sanyo VPC-C6) were also made during all experimental tasks. Scoring Speech Reaction Time SRT was measured by E-Prime computer soft ware and the voice key of the SRB. Only correct productions were used to calculate SR T. Subjects were trained to not produce a speech response until the target word was presented. SRT was measur ed starting with the command for movement (i.e., presentation of th e target) until the onset of voicing registered for each response. Responses during the pause or less than 250ms after presentation of the target were scored as a premature response. Premature responses were c onsidered to be incorrect and were combined with other errors from the re sponse accuracy assessment to determine the total number of incorrect responses. Response Accuracy Acoustic recordings from the response primi ng test were used to determine the accuracy of responses over two sessions. Du ring the first session, judge one (H J) independently listened to all responses using high quality headphones (Sony MDR-V6). Each response was listened to as many times as necessary to be able to use broa d phonetic transcription the record the responses and determine if a production wa s correct or incorrect. During the second session, a second judge (DK) was added and this process was completed with both judges simultane ously listening to the acoustic recordings with headphones. When the two judges did not agree on whether a response was correct or incorrect, a consensus decision was reached with further listening and debate. Agreement between the judges wa s not considered to occur in cases where further listening

PAGE 23

23 and/or debate were required, although these fina l consensus judgments were used to determine accuracy of each response. Types of errors included the following: Production of the prime Responses in which participants do not switch their speech response. For example, shopper-chopper is the prime-targ et pair and the production is shopper rather than chopper. Subjec ts may self-correct. Partial production of prime Responses in which participants partially produce the prime word such as the first sound or syllable. For example, the prime-target pair is shopper-chopper and the subject responds shchopper Subjects may self-correct. Initial sound repetition Responses in which participants repe at the initial sound or syllable of the target word. For example, in the case of the word pair shopper-shopper, the subject responds shshopper. Production of previous target Responses in which the subject produces a previously presented target. No response Subject does not produce a speech response upon command. Phonological error Single phoneme omission, deletion, substitution, transposition, etc. Lexical/semantic error Responses in which participants substituted a whole real word for the target word. Multiple errors Two or more errors from the above list were combined in a single response Reliability Intra-rater Reliability Judge one completed perceptual assessment to determine the accuracy of each response on two separate occasions. Point-by-point analysis wa s conducted for 100% of each subjects responses

PAGE 24

24 during each of their two test sessions to determine whether agreement regarding accuracy was present. Intra-class correlation co efficients were determined. Inter-rater Reliability A point-by-point analysis was completed to compare the level of agreement between the two judges as to whether each individual respon se was correct or incorrect. Intra-class correlation coefficients were determined. Data Analysis Average response latency and number of errors were calculated for each participant by the stimulation condition (on a nd off DBS) and by the prime-ta rget relationship (switch and no switch responses). Summary statistics are provided by the stimulation condition and by the prime-target relationship. Furthermore, formal statistical infere nces were conducted using nonparametric procedures. Separate Wilcoxon signed-rank tests were used to determine whether significant differences were present between the on and o ff stimulation conditions for switch and no switch speech responses in te rms of SRT and response accuracy. Fisher's combination method was used to further analyze the data and combin e results from the on and off stimulation conditions and the switch and no switch test conditions to obt ain overall p-values. Sample Size and Power Consideration Power analysis was conducted based on data for participants with hypokinetic dysarthria published by Spencer and Rogers (2005). As the study population and experimental stimuli vary between the two experiments, a conservative a pproach was utilized whenever possible. The mean difference in log-transformed response latency (called speech reaction times in the published paper) is 0.0966 unit larger (i.e., 10% l onger) for switch versus no switch speech

PAGE 25

25 responses, with a standard deviation of 0.0805. If it assumed that the differences in logtransformed response latency have the same di stribution in our study for each stimulation condition, then the Wilcoxon signed-rank sum test at the 0.05 Type-I error level has sufficient power to detect differences between switch and no switch speech responses as shown in Table 2-2. On the other hand, for differences between on and off stimulation conditions, we did not have an estimate of effect size. Thus, a se nsitivity analysis with regard to difference magnitude was completed with the assumption th at the sample size would be 15 to test the hypothesis using the Wilcoxon signed-rank sum test at the 0.05 Type-I error level. Table 2-3 provides powers corresponding to r, ratio of mean SRT between the on and off stimulation conditions. This table suggests that even if the difference between the on and off stimulation conditions is only half of the difference between the switch and no sw itch responses observed in the Spencer and Rogers study from 2005, we will still have reasonable power.

PAGE 26

26 Table 2-1. Auditory-perceptual dysarthria severity scale.

PAGE 27

27 Table 2-2. Power analysis to detect differences between sw itch and no switch speech responses.

PAGE 28

28 Table 2-3. Powers corresponding to r, ratio of m ean SRT between the on and off stimulation conditions, based on sensitivity analysis.

PAGE 29

29 Figure 2-1. Typical ex perimental timeline.

PAGE 30

30 Figure 2-2. Response priming protocol.

PAGE 31

31 Figure 2-3. Equipment configuration

PAGE 32

32 CHAPTER 3 LITERATURE REVIEW This chapter will review the necessary literature to support the rationale of the study. This will include discussion of: (1) The basal ganglia a nd its internal and external circuitry; (2) DBS as treatment for PD and possible mechanisms of benefit; (3) Motor planning/programming with an emphasis on maintaining and switching moto r programs, (4) The possibility of motor planning/programming deficits (including speech ) in PD; and (5) The use of reaction time experiments to test aspects of motor planning/programming. The Basal Ganglia The basal ganglia comprise several subcortical nuclei critically involved in movement and posture. Although this study emphasizes the role of the basal ganglia in movement (specifically its planning/program ming), it should be recognized that modern conceptualization of basal ganglia function emphasizes a number of functionally segregated circuits involved in a number of diverse motor, cognitive, and limbi c functions (Alexander, DeLong & Strick, 1986). Due to the intrinsic/extrinsic basal ganglia circ uitry, these structures have the opportunity to influence diverse cortical areas including those thought to be involved in speech motor planning/programming. Anatomy First, the normal basal ganglia anatomy will be described. The basal ganglia are often conceptualized as a group of i nput structures, output structures and intrinsic nuclei. The two primary input structures are the striatum (com prised of the caudate and putamen) and the subthalamic nucleus (STN). The two primary output structures are globus pa llidus internus (GPi) and substantia nigra pars reticulata (SNpr). Th e intrinsic nuclei of the basal ganglia include globus pallidus pars externa (GPe) and substan tia nigra pars compacta (SNpc) (Mink, 1996).

PAGE 33

33 Many of the structures of the basal ganglia, as well as other surrounding structures such as the thalamus and internal capsule (IC), are shown in Figure 3-1. Normal Basal Ganglia Intrinsic/Extrinsic Circuitry The extrinsic and intrinsic circuitry of the basa l ganglia is of course complex and will only be briefly reviewed, with an emphasis on what some have begun to call the standard model (Gale, Amirnovin, Williams, Flaherty, & Eskandar, in press) This model suggests a number of separate, functionally, and anatomically segregat ed corticobasal ganglion circuits, each of which has two pathways through the ba sal ganglia: the inhibitory direct pathway and the excitatory indirect pathway (DeLong & Wichmann, 2007) As shown in Figure 3-2, under normal conditions, the putamen receives excitatory inpu t from multiple cortical areas, including the portions of the motor cortex (MC), premotor co rtex (PMC), supplementary motor area (SMA), and cingulate motor area (CMA) (Alexander, DeLong & Strick, 1986; DeLong, 1990; DeLong & Wichmann, 2007; Mink, 1996). Output of the putam en is thought to be neuromodulated through the direct and indirect pathways The direct pathway projects i nhibitory signals to the output nuclei, mainly GPi and SNpr. The indirect pathwa y sends inhibitory signals from the striatum (primarily the putamen) to GPe and GPe in turn sends inhibitory projections to the STN. In addition receiving inhibitory proj ections from GPe, the STN also receives excitatory projections directly from the cortex. Finall y, the STN projects excitatory si gnals to the main output nuclei (GPi and SNpr) where both the direct and indirect pathways to converge to deliver a balance of excitatory and inhibitory signals resulting in tonic, rapid inhibitory GABAergic projections to the thalamus (Hikosaka, 2007). This finally results in excitatory projections back to cortical areas including MC, PMC, SMA, and CMA (Alexander, DeLong & Strick, 1986; DeLong, 1990; DeLong & Wichmann, 2007), as well as the dorsolateral prefrontal area, lateral orbitofrontal

PAGE 34

34 cortex, and the anterior cingulat e/medial orbitofrontal cortices. This arrangement allows the basal ganglia to influence multiple diverse processes of the frontal lobes including movement, behavior, cognition, language, and emotion (Alexander, DeLong & Strick, 1986). Indeed, as stated by Murdoch (2004), (t)hi s anatomical arrangement (allows) the output from the basal ganglia (to gain) access to multiple areas of the frontal lobeand provides a basic neuroanatomical mechanism whereby these subcor tical structures can influence aspects of behaviour, cognition and language as well as motor function (p. 235). The focus of the present paper is on preparat ory aspects of movement, specifically speech motor planning/programming. Van der Merwes four phase model of speech production (1997; see full discussion below) suggests that many of the aforementioned neural regions are involved in these activities. Motor planning, for example, is suggested to be accomplished largely in PMC and SMA (i.e., the motor association area), while motor programming is thought to comprise areas such as the basal ganglia, cerebellum, SMA, and MC. Circuitry in PD Although the exact role of dopami nergic input to the striatum is not completely understood (Mink, 1996), the degeneration of dopaminergic cel ls in SNpc is a hallmark feature of PD (Bergman & Deuschl, 2002). Gale and colleagues (in press) state the at the neurophysiological level, loss of dopaminergic neurons in PD leads to the clinically observed manifestations of the disease due to derangements of firing rates, neur onal selectivity, and the firing patterns of (basal ganglia) neurons (p. 2). Figure 33 illustrates the effect of PD on the circuitry of the basal ganglia as shown in the standard model. In the indir ect pathway, loss of striatal dopamine leads to excessive inhibition of GPe, leading to decreased inhibition of STN and the delivery of excessive excitatory drive to the basal ganglia output nuclei (i.e., GPi and SNpr). This is

PAGE 35

35 reinforced by reduced inhibition to GPi/S Npr delivered through the direct pathway. Cumulatively, this imbalance between the indire ct and direct pathways results in excessive thalamic inhibition, finally resulting in reduced excitatory projections b ack to the cortex and inhibiting intended movement (DeLong, 1990). Thus, th e effect of striatal dopaminergic loss is inhibition of cortically initiat ed movement, to cause akinesia (loss of movement), hypokinesia (reduction of movement) and bra dykinesia (slowness of move ment) (Marsden & Obeso, 1994, p. 878). Other work has continued to refine the role of the BG in movement. For example, Mink (1996; 2003; Mink & Thach, 1991) has proposed a m odel based on a series of experimental observations which suggests the role of the basal ganglia in normal movement is to facilitate desired motor programs while inhibiting other unwanted motor programs. According to this model of focused selection and inhibition of competing motor programs, voluntary movement is initiated by cortical mechanisms. The basal ganglia facilitate movement by decreasing inhibition of desired motor programs while simultaneous ly increasing inhibition of competing motor programs (Mink, 1996; Mink, 2003; Rubchinsky, Kope ll, & Sigvardt, 2003) through GABAergic output influencing cortical (and brainstem) motor mechanisms (Hikosaka, 2007; Mink, 2003). The standard model of basal ganglia circuitry continues to be influential but it has been suggested that this model requires substantial revision and refinement to account for advances in understanding (Gale et al., in pres s; Greybiel, 2005). Ne vertheless, this model continues to have heuristic value and in particular, provides a logical rationale for su rgical interventions. Deep Brain Stimulation A number of neurosurgical approaches to the treatment of PD have been developed over the last century, including ablative procedures (e.g., pallidotomy), and, more recently, DBS. DBS

PAGE 36

36 is a surgical procedure in which electrical stimulation is deli vered to neural targets through chronically implanted leads. Quadrapolar elec trodes are connected to an internalized programmable neurostimulator usually placed below the clavicle as shown in Figure 3-4 (Benabid, 2003). Thalamic, pallidal, and STN DBS ar e all recognized to have a beneficial effect on the motor symptoms of PD (Benabid, 2003; Gross, 2004; Rodriguez-Oroz et al., 2005; Volkmann, 2004). After a period of intense use a nd then declining interest, neurosurgical treatments for PD have once again become popular, primarily due to limitations in the medical management of PD, advances in the understanding of the intrinsic and extrinsic circuitry of the basal ganglia, surgical tec hnique, neuroimaging, and microele ctrode recording techniques (Koller, Pahwa, Lyons & Albanese, 1999). In comp arison to ablative procedures, DBS may have many advantages, including a decreased occurr ence of adverse events, minimal permanent lesions, and an increased ability to perform bilateral procedures without adverse events such as speech and swallowing problems (Benabid et al ., 1996; Ghika et al., 1998; Koller et al., 1999; Obwegeser et al., 2001; Pinto et al., 2004; Rodriguez-Oroz et al., 2005). Adjustments in the parameters of stimulation can also facilitate in dividualization in the trea tment and minimize side effects. DBS in PD Although the optimal surgical target for DBS in PD remains unknown (Okun & Foote, 2005), the most common surgical targets are th e thalamus, GPi, and STN. Influential early studies by Benabid and colleagues (1994) and Limousin et al. (1998) suggested significant benefit to STN DBS and this quickly became the surgical treatment of choice for most centers. However, it has been suggested that further tria ls comparing the benefit of GPi versus STN DBS need completion (and are in fact in progress) (Okun & Foote, 2005).

PAGE 37

37 Surgical treatments are usually performed in those patients with advanced PD who have disabling motor symptoms that are insufficiently controlled with medical management. GPi and STN DBS are recognized to have a beneficial effect on symptoms such as tremor, rigidity, bradykinesia, dyskinesia, and postural/gait abnormalities in patients with PD who are insufficiently managed with pharmacological th erapy (Benabid, 2003; Gross, 2004; RodriguezOroz et al., 2005; Volkmann, 2004). Also important is the absence of significant cognitive impairments, an understanding of the surgical ri sks, and realistic postoperative expectations (Marks Jr., 2005; Vitek & Walter, 2005). For GPi or STN DBS, the best predictor of outcome seems to be a patients response to levodopa (Benabid, 2003). Mechanisms of DBS The exact neurophysiological mechanisms for the improvement in motor symptomatology with DBS in PD are unknown. Due to the similarities in improvements in motor functioning following ablative lesi oning procedures, it has been suggested that DBS acts as a transient electrical inactivation or reversible le sion to block the output of dysfunctional targets (Lozano, Dostrovsky, Chen, & Ashby, 2002, p. 226). Howe ver, current concep tualization of the mechanisms of DBS suggests that this is a vast oversimplificati on (Lozano et al., 2002; Desbonnet et al., 2004; Grill, Snyder, & Miocinovi c, 2004). For example, Temel and colleagues (2005) state that an increasing amount of da ta suggests that categorizing DBS as being inhibitory and thus equating its effects to those of a lesionisan oversimplification of what is a highly complex and multi-faceted technique (p. 397). Lozano and colleagues (2002) review multiple possible mechanisms of DBS including facilitative, inhibitory, and downstream effects. Facilitative effects likely include activation of large axons. Inhibito ry effects may include partial or complete blocking of neuronal firing, possibly due to depolarization an d/or the release of

PAGE 38

38 inhibitory neurotransmitters. Neurons located downstream from the stimulation site are also likely to be influenced by DBS. This notion is supported by functional im aging studies revealing changes in cortical activity associated with improvements in motor function (Lozano et al., 2002). For example, Davis et al. (1997) used positron emission tomography (PET) to reveal an increase in regional cerebral blood flow in the PMC when GPi DBS improved motor symptomatology. Finally, when considering the mechanism of DBS, it is important to consider that DBS likely influences diffe rent neural targets in differe nt ways (Lozano et al., 2002). DBS and speech Although speech disturbance in the form of hypokinetic dysarthria is common in individuals with PD, speech improvement is not specifically targeted by DBS surgery. The effects of neurosurgical treatments for PD on speech function have only recently begun to receive systematic attention, pa rticularly STN DBS (for more detailed review s see Jones, Kendall, Sudhyadhom, & Rosenbek, in press; Sc hulz, 2002; Schulz & Grant, 2000). To grossly simplify what is developing to be a fairly subs tantial body of work, the influence of STN DBS on speech has been studied using a variety of sophisticated measurement approaches, including instrumental approaches such as acoustic anal ysis and kinematic measurement. Gentil and colleagues (1999, 2000; 2001; see also Pinto, Gentil, Fraix, Benabid & Pollak, 2003) have been pioneers in the study of the speech effects of STN DBS and have conducted a program of research using perceptual, kinematic, and acoustic analyses. C onsidered overall, these data suggest a number changes on versus off STN DBS, including perceptual improvements in speech, improved lingual and labial strength and control, and improvements in the acoustic speech signal (e.g., increased fundamental freque ncy variability in sentences and decreased fundamental frequency variability during sustai ned vowel production). PET scan data from

PAGE 39

39 Pinto et al. (2004) further suggest that the patterns of abnormal cortical activation in patients with PD and hypokinetic dysarthria appear to normalize with improvements in speech when on STN stimulation. Other research also indicates that STN DBS has some benefit for speech (Dromey, Kumar, Lang, & Lozano, 2000; Hoff man-Ruddy, Schulz, Vitek & Evatt, 2001; Rousseaux, Krystkowiak, Kozlowski, Ozsancak Blond, & Destee, 2004), though speech changes may be dependent on parameters of stimulati on being applied (Tornqvist, Schalen, & Rehncrona, 2005) or the hemisphere be stimulated (Santens De Letter, Van Borsel, De Reuck & Caemaert, 2003; Wang, Metman, Bakay, Arzbaecher,,& Bernard, 2003). However, dysarthria is also commonly reported as an adverse event in many surg ical trials (The deep brain stimulation for Parkinsons disease study group, 2001; Esselink et al., 2004; Krack et al ., 2003; Kumar et al., 1998a, b; Ostergaard, Sunde, & Dupont, 2002; Rodri guez-Oroz et al., 2005; Romito et al., 2003; Schupback et al., 2005; Thobois et al., 2002) and s ubstantial improvement in speech would not be a surgical goal in most cases. The aforementioned studies designed to evalua te the speech effects of surgery have primarily targeted measurement at the execution phase of movement, rather than preparatory motor processes such as planni ng/programming (see discussion of Models of Motor Control below). This is not surprising considering that the hypokinetic dysarthria encountered in PD has traditionally been conceptualized as a disorder of execution level processes (Darley, Aronson, & Brown, 1969a, b, 1975; Duffy, 2005; Yorkston, Be ukelman, Strand, Bell, 1999). However, possible contribution of speech motor planni ng/programming deficits in PD are being increasingly recognized (Spen cer, 2006; Spencer & Rogers, 2005; Van der Merwe, 1997), though little work has targeted processes involved in speech motor planning/programming to determine the possible influence of DBS. Howeve r, upon close inspection, data from Gentil and

PAGE 40

40 colleagues (1999, 2000; 2001; see also Pinto, Ge ntil, Fraix, Benabid & Pollak, 2003) merit further scrutiny for those interested in th e possible influence of DBS on speech motor planning/programming in PD. Gentil and colleagues (1999) investigated th e effect of bilateral STN DBS on speech and nonspeech oromotor function in 10 patients with PD and 14 healthy controls. Patients were tested off medication in both on and off stimulation conditions. The off stimulation condition assessment occurred 1 hour after discontinuing stimulation. Perceptual and kinematic measurement approaches were utilized. Perceptu al measurement was limited to using the score from item 18 of the Unified Parkinsons Dis ease Rating Scale (UPDRS). Kinematic assessment procedures utilized force tr ansducers to determine ramp-a nd-hold force contractions and maximal strength of the lips and tongue. Kinema tics revealed a number of improvements in lip and tongue function on stimulation, includi ng increased maximal strength, increased accuracy in reaching a target, increased precision during the hold phase, and decreased RT. Results of perceptual assessment also revealed improve d speech function in the on versus off stimulation condition. It is the kinematic data whic h is most scientifically rigorous and which is of primary interest for the present discu ssion on speech motor planning/programming. These data which improvements in RT on as compar ed to off STN DBS for a nonspeech oromotor movement suggest that processes involved in speech motor planning/programming may be positively influenced by DBS. Indeed, the measure of RT (as discussed in detail later in this chapter) is an accepted measurement technique to quantify preparatory aspects of movement (e.g., motor planning/programming) and this stud y may reflect a pioneerin g effort to measure non-execution level speech changes with DBS in i ndividuals with PD.

PAGE 41

41 These RT data have been replicated by Gentil and colleagues. In 2000, they further investigated the influence of bilateral stimulati on in 10 patients with STN DBS and used similar experimental procedures as described above. Im proved RT in the on STN DBS condition for a nonspeech oromotor movement was again reported. This group of researchers (Pinto, Gentil, Fraix, Benabid, & Pollak, 2003) continued to use th eir experimental protocol in their largest group to date 26 patients with PD and bilateral STN DBS and improved RT was again found in the on stimulation condition. A Model of Speech Motor Control Numerous models of motor control whic h share many features including concepts consistent with the concept of motor program s have been proposed, including those from Schmidt (1975) and Brooks (1986). Speech specific models are also available, such as Van der Merwes (1997) proposed f our-phase framework. Four Phase Model Van der Merwe (1997) has proposed a model for speech motor control based on four phases: linguistic-symbolic planning, motor planning, motor programming, and execution. Let us consider each phase of the model in greater detail, with particul ar attention to motor planning and motor programming. Linguistic Symbolic Planning Linguistic symbolic planning is considered to be a non-motor stage during which the intent to communicate originates due to individu al behavioral needs and environmental demands. A message in compiled during this level of processing, requiring semantic, syntactic, morphological, and phonological planning. For example, phonological planning involves selection and combination of phonemes in accordance with the phontoactic rules of the

PAGE 42

42 language, and it is portrayed as a linguistic-symbolic function within the proposed framework (Van der Merwe, 1997, p. 9). These processes are thought to be accomplished primarily by temporal-parietal areas, including Brocas and Wernic kes areas. Errors attri buted to the level of linguistic-symbolic planning are due to seman tic, lexical, syntactic, morphological, and phonological errors associat ed with language based disorders (i.e., aphasia). For example, Van der Merwe suggests deficits in phonological pl anning will result in phoneme substitutions and transpositions. Motor Planning Motor planning, along with the next level of the model, motor programming, is the area of interest in the present experiment. Van de r Merwe conceptualizes motor planning phase as being mediated by the by the hig hest level of the motor hierarchy (p. 9). Motor planning involves gradual transformation of symbolic units (phonemes) to a code that can be handled by a motor system (p. 9). Van der Merwe (1997) s uggests that speakers learn core motor plans during development and adaptation. These motor plan s include goals in the form of spatial and temporal specifications for movement which are stored in sensor imotor memory. During acquisition of a core motor progra m, such as during development, this model suggests increased usage of external feedback, such as that from proprioception and audition Following retrieval of the core motor plan, planning continues in orde r so that the consecutive movements necessary to fulfill the spatial and temporal goals can be me t (p. 11). Motor plan subroutines such as velar lift and tongue placement are then specified for the planned production. Motor planning is thought be accomplished in multiple neural regions, particularly the motor association areas of PMC and SMA, as well as prefrontal and pa rietal association areas Van der Merwe (1997) suggests the changes in speech that may be enco untered due to deficits in motor planning to

PAGE 43

43 include slow, struggling speech with distorti on and even apparent substitution such as associated with apraxia of speech (p. 17). Motor Programming It should be acknowledged that differentia tion between speech moto r planning and motor programming is extremely challenging and has not, to our knowledge, been demonstrated experimentally. However, during this hypothesi zed level of speech production, motor programs are finally selected and sequenced for move ments of the necessary muscles for speech production. Information regarding s patio-temporal and force dimensions such as muscle tone, rate, direction, and range of movements is controlled by muscle specific programs influenced by external feedback (Van der Merwe, 1997, p. 7). Motor programming involves multiple neural areas including the basal ganglia, SMA, lateral cerebellum, and MC. Speech change associated with deviant speech motor programming is hypothesi zed to include sound distortion, defects in speech rate, and/or problems in the initia tion of movement (Van der Merwe, 1997, p.18). Execution At the level of execution, the plans and programs of the previous phases in speech production result in muscle movements and speech production. Feedback to higher levels in the motor system is an integral part of this phase particularly during development. Feedback may take the form of tactile-kinematic and acoustic information, for example. Neural areas involved in execution include the motor cortex, the lowe r motor neurones, periph eral nerves, and motor units in the muscles (Van der Merwe, 1997, p. 16) Areas involved with pr eparatory aspects of speech movement (i.e., motor planning and motor programming) are also active during execution, including cortical (SMA) and subcortical regions (the basal ga nglia and thalamus). Although not directly addressed by this model, execution level errors, for example, are often

PAGE 44

44 associated with peripheral muscle changes su ch as impaired strength or tone. Current conceptualization of all dysarthria types, incl uding the dysarthria of PD, suggests that these speech disorders and their resultant errors are due primarily (if not exclusively) from deviant processes of execution. Motor Planning/Programming Van der Merwes four phase model of speech control provides useful theoretical constructs for research conducted in this area. In turn, notions about speech motor control will continue to be refined with experimental data. Data supporting diffe rentiation between the processes of speech motor planning and speech motor programming woul d be particularly valuable, and, as noted by Spencer & Rogers (2 005), has not yet occurre d. Therefore, for the present discussion, we will combin e these two phases of speech mo tor control into one motor planning/programming. The notion that preparatory activities cons istent with processes involved in motor planning/programming which occu r prior to movement execution are commonly encountered in the literature. Indeed, as noted by Schmidt in 1975, the notio n of motor programs or a set of stored muscle commands ready for action at any gi ven time has probably been with us for a very long time (p. 231). Early contributions from Lash ley (1917) and Henry and Rogers (1960) have continued to be refined by a number of research ers who have proposed m odels consistent with concepts of speech motor planning/program ming (Klapp, 2003; Levelt & Wheeldon, 1994; Schmidt, 1975; Sternberg, Knoll, Monsell, & Wright, 1988; Ster nberg, Knoll, & Turock, 1990; Sternberg, Monsell, Knoll, & Wright, 1978; Ster nberg, Wright, Knoll, & Monsell, 1980; Van der Merwe, 1997).

PAGE 45

45 Subprogram Retrieval Model The work of Sternberg and colleagues ( 1978, 1980, 1988, & 1990) has been particularly useful in attempting to define the theoretical und erpinnings of the processes involved in speech motor planning/programming. In 19 78, Sternberg et al. proposed a subprogram retrieval model (later renamed as the subprogr am-selection model by Sternberg et al., 1988) to explain their data on preparation of rapid move ment sequences during speech a nd typewriting. In this model, a motor program or representation of the entire responseis constructed before the response starts (Sternberg et al., 1978, p. 133). The motor program comprises a set of linked subprograms one for each unit of the response that are retained in a special motor-program buffer distinct from ordinary short-term memory until the command for movement is received (Sternberg et al., 1978, p. 133). Subprograms are retrieved from the sensorimotor store and loaded into the buffer prior to movement. Ke ller (1987) suggests that the subprograms or aggregates of muscle commands are learned and are then to be found in the store ready for retrieval as needed, rather than freshly genera ted each time speech is to be produced (p. 135). Upon command for movement, the first subprogram is located in the buffer in order to initiate movement execution. Sternberg and colleague s (1978) further desc ribe their proposed mechanism for retrieval of subpr ograms from the buffer: The particular retrieval mechanism suggested by our results is self-terminating sequential search through a nonshrinking buffer, rather than, for example, a process of direct acc essThe search is presumably necessary because the necessary subprograms are not arranged in the buffer in the order in which they must be executed (p. 147). Additionally, Sternberg and colleagues suggest the size or unit of the speech subprograms in their model is the stress group or metric foot (a segment of speech associated with a primary stress) (1978, p. 136)

PAGE 46

46 In 1988, Sternberg and colleagues continued to clarify and refine their model of they now call a subprogram-selection model in a paper em phasizing the concept that the motor program is operated upon by a series of selection and comm and processes (p. 184). Before a movement subprogram can be executed, it must first be a ccessed from the buffer. After the subprogram is accessed from the buffer, The command processcauses it to be executed. Speech production is thus controlled by an alternating sequence of se lection and command processes (p. 184). The model presented by Sternberg and collea gues has its limitations but it provides a sensible theoretical framework for resear ch studying processes involved in motor planning/programming in general and speech more specifically. This mode l can be applied to concepts in the current motor control literatur e, such as motor progr amming maintenance and switching. Motor Program Maintenance Sternberg et al.s notion that motor program s are held in the buffer until a command for movement is provided appears anal ogous to the ability to hold or maintain a motor program prior to movement execution. Maintenanc e of a motor program has been studied extensively in both normal and disordered populations. Hallett ( 1990) describes the common method for studying maintenance of a motor program using the de layed response paradigm. In this situation, information about the movement is completely sp ecified, but then the information is withdrawn for a period of time before the stimulus to move is delivered (Hallett, 1990, p. 588). RT serves as an index of maintenance of the motor program in the buffer. In some disordered populations (including individuals with PD), the contents of the buffer have been found to decay over time, disrupting the maintenance of mo tor programs as measured by RT.

PAGE 47

47 Motor Program Switching The ability to rapidly switch motor programs has also received significant experimental attention. In such experiments, subjects are (oft en unexpectedly) presented with a prime stimulus that does not accurately or completely inform th e target stimulus. Viewed through the model of Sternberg and colleagues, this ta sk requires subjects to clear th e prepared (and inaccurate) motor program in the buffer, search the sensorimotor store for the newly required motor subprograms, and load the motor program into the buffer before execution. Switching motor programs requires increased RT due to the increased complexity of this task. Difficulty switching has been hypothesized to be related to impairments inhibiti ng or modifying a motor program (Inzelberg et al., 2001; Kropotov & Etlinger, 1999; Mink, 1996) or in the activation of a new motor program (Haaland & Harrington, 1991). Deviant Motor Programming in PD In individuals with PD, deficits in asp ects of motor programming are being recognized with increasing regularity, bot h in the limbs and, less commonly, the speech mechanism. Limb Motor Programming Maintenance and Switching A substantial literature s upporting the concept of impaired maintenance of motor programs for limb movements in PD is availabl e (Gentilucci and Negrotti, 1999a, b; Gueye, Viallet, Legallet & Trouche, 1998; Jones, Ph illips, Bradshaw, Iansek, & Bradshaw, 1992; Pascual-Leone, Valiis-Sole, Brasil-Neto, Cohen & Hallett, 1994; Romero, Van Gemmert, Adler, Bekkering, & Stelmach, 2003; Sheridan, Flowers, & Hurrell, 1987; Stelmach, Garcia-Colera & Martin, 1989), though this finding ha s not been replicated by all st udies (Labutta, Miles, Sanes, & Hallett, 1994). PD patients have also been found to have disordered ability to switch motor programs (Benecke, Rothwell, Dick, Day & Marsden, 1998; Contreras-Vi dal & Stelmach, 1996;

PAGE 48

48 Dirnberger, Reuman, Endl, Lindinger, Lang & Rothwell, 2000; Harrington & Haaland, 1991; Inzelberg, Plotnik, Flash, Schechtman, Shah ar, & Korczyn, 2001; Kropotov & Etlinger, 1999; Robertson & Flowers, 1990; Roy, Saint-Cyr, Taylor, & Lang, 1993; Rubchinsky, Kopell, & Sigvardt, 2003; Stelmach, Garcia-Colera, & Martin, 1989; Weiss, Stelmach, & Hefter, 1997). Disordered maintenance and switching of motor programs has been proposed to explain some of the primary motor symptomatology of PD, such as akinesia. Speech Motor Programming Maintenance and Switching Deficits in speech motor planning/programming in patients with PD are being increasingly recognized (Van der Merwe, 1997). As noted by Van der Merwe (1997), this complicates our traditional view of dysarthria as a motor execut ion problem (p. 18). The fact that the basal ganglia appear to be involv ed in both motor programming and execution suggests the possibility of dual symptomatology in certain ty pes of dysarthria (Van der Merwe, 1997, p. 18). The author then specifically suggests that th e hypokinetic dysarthria of PD is one of the dysarthria types in which Coexisting problems in both motor programming and motor execution would seem to be present (Van der Merwe, 1997, p. 18) Spencer and Rogers (2005; see also Spencer 2006) have pioneered the study of speech motor planning/programming deficits in dysarthria types traditionally associated with execution level dysfunction. These authors suggest that the role of the basa l ganglia in motor planning/programming has been illuminated by converging evidence from limb RT studies of adults with Parkinsons disease which show impairments in two primary areas: maintenance and switching (Spencer & Rogers, 2005, p. 348). Furthermore, the presence of motor planning/programming deficits in individua ls with hypokinetic dysarthria is supported by specific speech symptoms commonly encountered in this population. Spencer and Rogers (2005)

PAGE 49

49 suggest that disordered maintenance of speech motor programs may result in abnormally placed pauses, difficulty with progression through an utte rance and difficulty ini tiating articulation (p. 348). Similarly, disordered switching of speech motor programs may be associated with difficulty stopping an ongoing response, marked hesitations between movement segments, and occasional inability to switch from one to anothe r movement (Spencer & Rogers, 2005, p. 348). To test the notion that speech motor planning/p rogramming is disrupted in individuals with PD and hypokinetic dysarthria, Spencer and Rogers (2005) employed a RT paradigm. Ten participants with PD and hypoki netic dysarthria and 15 normal c ontrols were tested using a response priming procedure in which participants were provided with a prime to supply information regarding target (see Chapter 2 E xperimental Procedures for more details). The primary dependent variable was SRT. Results prov ided preliminary evidence for the notion that maintenance and switching of speech motor programs is disordered in participants with PD and hypokinetic dysarthria. Reaction Time/Speech Reaction Time The use of RT experiments being used as an index of various underlyi ng neural processes has a history that dates back to at least the mid-19th century (Smith, 2004). The basic RT paradigm (to which an infinite array of variab les can be added) invol ves presentation of a stimulus to subjects who are then required to start a movement as quickly as possible (Hallett, 1990). RT is defined as the temporal duration between the presentation of the stimulus and the initiation in movement. Adaptations to the RT pa radigm usually increase the complexity of the task in any number of ways, by methods such as demanding more complicated movements or adding a pause between presentation of the stim ulus and the command fo r movement. Increased complexity invariably results in increases in RT RT in individuals with PD has been studied

PAGE 50

50 extensively and is generally found to be impaired in comparison to age-matched normal controls (Bloxham, Mindel, & Frith, 1984; Draper & Jo hns, 1964; Evarts, Teravainen, & Calne, 1981; Gueye, Viallet, Legallet, & Tr ouche, 1998; Labutta, Miles, Sanes, & Hallett, 1994; Montgomery, Baker, Lyons, & Koller, 2000; Muller, Eising, Kh un, Buttner, Coenen, & Przuntek, 1999; Temel et al., 2006). There are two ge neral types of RT conditions : simple RT and choice RT. Simple and Choice Reaction Time In the simple RT test paradigm, the expected movement is described completely, without ambiguity (Hallett, 1990, p. 587). This allo ws subjects to fully prepare (or plan and program) the required movements in advance of the command to execute movement. Simple RT experiments can still increase the complexity of the task, most often by adding a delay between stimulus and command. Using the model of Sternberg et al., this would re quire participants to maintain the motor program in the buffer until the command to execute movement is provided. In the choice RT paradigm, subjects are not provided a complete description of the required movement until the stimulus that calls for the movement initiation (p. 587) is delivered. Since subjects are not able to plan/program movement s in advance, choice RT is always longer than simple RT. Like simple RT, choice RT is also influenced by complexity factors. For example, providing incorrect information about the required movement in advance of the command for movement increases complex ity (and thus RT). According to the model of Sternberg et al., the increase RT in the choi ce RT paradigm is explained by the additional required planning/programming pr ocesses required with this ta sk. These processes include retrieving the appropriate motor subprograms from the sensorimotor store and loading them into the buffer.

PAGE 51

51 SRT in the No Switch Condition is a Measure of Speech Motor Program Maintenance In the present experiment, SRT in the no sw itch condition serves as a measure of the maintenance of speech motor programs on and off DBS. In this condition, subjects are visually presented with a prime wo rd and instructed to speak this word as quickly and clearly as possible upon presentation of the command for m ovement (i.e., the target word). Although the experiment was not originally conceived in this manner, this paradigm satisfies the criteria for a simple RT experiment in that subjects are provided with complete in formation regarding the expected movement in advance. The subprogram retrieval model of Sternberg and colleagues suggests that this allows partic ipants to retrieve subprograms fr om the sensorimotor store and load the motor program into the buffer prio r to movement execution. Much like many other simple RT experiments, the complexity of this task in the present experiment was increased by adding a 250 ms delay between st imulus presentation (i.e., the prime word) and presentation of the command for movement (i.e., the target word). SRT in the Switch Condition is a Measure of Speech Motor Program Switching In the present experiment, SRT in the sw itch condition serves as a measure of the switching of speech motor programs on and off DB S. In this condition, subjects are visually presented with a stimulus (i.e., the prime word ) which does not accurately inform the requested movement upon receipt of the command for movement (i.e., the target word). In other words, the prime unexpectedly does not match the target. This paradigm appears to generally satisfy the criteria for a choice RT experiment in that subj ects are provided with incomplete information regarding the expected movement until the comma nd for movement is presented. However, the complexity is again increased by the presentation of an incorrect prime. Presumably, according the model of Sternberg participants have alrea dy retrieved incorrect motor subprograms from the

PAGE 52

52 sensorimotor store and loaded the motor progra m into the buffer. This task requires several processes to occur prior to movement execution. The incorrect motor program in the buffer must be inhibited, the correct moto r subprograms must be retrieved from the store, and the motor program must be loaded into the buffer. Due to these additio nal processes and the increased complexity of the switch vers us no switch condition, RT will be increased for these tasks.

PAGE 53

53 Figure 3-1. Basal ganglia structures and surrounding areas. IC = internal capsule, GPe = globus pallidus pars externa, GPi = globus pallidus internus, STN = subthalamic nucleus, SN = substantia nigra.

PAGE 54

54 Figure 3-2. The intrinsic and extrinsic circuitr y of the basal ganglia under normal conditions. SMA= supplementary motor area, PMC = premotor cortex, MC = motor cortex, SNpc = substantia nigra pars compacta, D1 = striatal output receptor type D1, D2 = striatal output receptor type D2, GPe = globus pallidus pars externa, STN = subthalamic nucleus, GPi = globus pallidus internus, SNpr = substantia nigr a pars reticulata, VA = ventral anterior nucleus of the thalamus, VL = ventral lateral nucleus of the thalamus, CM = centrum medianum.

PAGE 55

55 Figure 3-3. The intrinsic and extr insic circuitry of the basal gang lia in Parkinsons disease. SMA = supplementary motor area, PMC = premotor co rtex, MC = motor corte x, SNpc = substantia nigra pars compacta, D1 = striatal output recepto r type D1, D2 = striatal output receptor type D2, GPe = globus pallidus pars externa, STN = subthalamic nucleus, GPi = globus pallidus internus, SNpr = substantia nigra pars reticulata VA = ventral anterior nu cleus of the thalamus, VL = ventral lateral nuc leus of the thalamus, CM = centrum medianum.

PAGE 56

56 Figure 3-4. Unilateral deep brain stimulation (DBS). IPG = internal pulse generator.

PAGE 57

57 CHAPTER 4 RESULTS Participants A total of 12 participated in the experiment. An additional eight individuals entered the screening process but did not meet inclusion criteria or were withdrawn (3 had surgery completed an outside facility, 2 failed screeni ng due to a Spatial Span Subtest score < 7, 1 subject had severe tremor off DBS, 1 s ubject was unstable on anti-Parkinsons disease medications, and 1 subject had a local sk in reaction to the accelerometer). Table 4-1 shows individual and group descriptive data for the 12 participants. Mean age was 61 years (sd = 8.28). Nine of th e participants were male (75%) and three were female (25%). Three patients had undergone STN surgery (25%) a nd in 9 subjects the exact surgical site (GPi or STN) was unknown due to participation in a larger double-blinded study. Eight of 12 (67%) had undergone a unilateral DBS surgery and 4 (33%) had undergone bilateral DBS surgery. Mean duration status-post surgery in months fo llowing surgery at the time of screening was 13.5 months (sd = 5.45). Half of the participants (6/12) were first randomized to the on stimulation test condition and the other half we re first tested off stimulation. Mean years of education was 14.16 (sd = 3.69) with the mode being 12 years (i.e., a high school diploma). Mean MMSE was 28.67 out of 30 (s d = 0.98) Mean standard scores for the Spatial Span Subtests were a forward score of 10.17 (sd = 2.04) and backward score of 10.83 (sd = 1.99). Perceptual judgment of dysar thria type was hypokinetic in all participants. The mean of dysarthria severity ratings was 3.08 (sd = 1.31) and the mode was 2. All participants reported an unremarkable speech and language developmental hi story. Mean intelligib ility score across the two listeners was 93.25% (sd = .06). Mean CES score was 33.03 out of 56 (sd = 8.66).

PAGE 58

58 Assessment of linguistic competency during repetition and connected speech suggested intact linguistic systems in eight of 12 (75%) participants Four subjects produced a total of eight errors during repetition (n = 2) or connected speech (n = 6). Five errors were determined to be phonological and 3 were semantic. Experimental Results Table 4-2 presents the summary statistics fo r both the primary and secondary research questions. The mean and standard deviation for SRT and response accuracy are shown across priming (i.e., no switch or switch) and stimulation (i.e., DBS on or off) conditions. Mean SRT in the no switch condition was 615.24 ms (SD = 96.77) on DBS and 671.38 ms (SD =113.05) off DBS. Mean SRT in the switc h condition was 717.09 ms (SD= 89.11) on DBS and 728.67 ms (SD = 98.35) off DBS. Mean number of errors (per 16 responses) in the no switch condition was 0.69 (SD = 0.85) on DBS and 1.06 (SD = 0.74). Mean number of errors in the switch condition was 1.50 (SD = 1.17) on DBS and 1.50 (SD = 1.98) off DBS. Primary Aims Research question 1 (no-switch vs. switch) Statistical significance was se t at the 0.05 level for all an alyses performed. Table 4-3 shows mean difference and p-values for all comparisons. Separate Wilcoxon signed-rank tests were conducted which revealed statistically si gnificant differences in SRT in the predicted direction between switch and no-switch conditions in both th e on DBS (signed rank = 1, p = 0.0010) and off DBS (signed rank = 12, p = 0.0342) st ates. Furthermore, Fishers combination method revealed significant differences in SRT overall across DBS conditions (Fishers combination test statistic = 20.57, p = 0.0040). In other words, subjects produced a speech response more quickly in the no-switch versus switch condition, regardless of whether DBS

PAGE 59

59 was on or off. As shown in Figure 4-1, when collapsed across stimulation conditions, mean SRT in the no switch condition was 643.38 ms (SD = 106.83) and 722.88 ms (SD = 98.35) in the switch condition. Research question 2 (no-switch condition on vs. off DBS) Separate Wilcoxon signed-rank tests were conducted which revealed statistically significant differences in SRT in the no switch condition between on and off DBS states (signed rank = 10, p = 0.0210) in the predicted di rection. That is, subjects produced a speech response more quickly in the no switch conditio n when on versus o ff DBS. As shown in Figure 4-2, mean SRT in the no switch c ondition was 615.24 ms (SD = 96.77) on DBS and 671.38 ms (SD = 113.05) off DBS. Research question 3 (switch condition on vs. off DBS) Separate Wilcoxon signed-rank tests were conducted which revealed no significant differences in SRT in the switch condition (signed rank = 30, p = 0.5186) between on and off DBS states (i.e., no difference in SRT was ob served in the switch condition regardless of whether DBS was on or off). As shown in Figure 4-2, mean SRT in the switch condition was 717.09 ms (SD = 89.11) on DBS a nd 728.67 ms (SD = 110.50) off DBS. Secondary Aims Separate Wilcoxon signed-rank tests were c onducted for both the DBS on and off conditions. In the off DBS condition, no stat istical difference (signed rank = 17, p = 0.5469) was found in response accuracy between switch and no-switch conditi ons. In the on DBS condition, response accuracy was also not statis tically significant (si gned rank = 9, p = 0.0605), thought there was a trend toward significant resp onse accuracy results in the predicted direction (i.e., subjects produced more erro rs on average in the switch condition versus the no switch

PAGE 60

60 condition under the on DBS stat e but this difference was not si gnificant). Fishers combination method was used to determine overall differences in response accuracy across DBS conditions and results were not significant (Fishers combination test sta tistic = 6.82, p = 0.1526). Separate Wilcoxon signed rank tests reveal ed no significant differences in response accuracy in the no switch (signed rank = 14, p = 0.1816) or switch (signed rank = 11.5, p = 0.7188) conditions when on versus off DBS states were compared. In other words, no difference in response accuracy was observed in th e no switch or switch conditions regardless of whether DBS was on or off. Fishers combination was used to determine overall differences in response accuracy across DBS conditi ons and results were not significant (Fishers combination test statistic = 4.07, p = 0.3881). Reliability Intra-rater Reliability Judge one completed perceptual assessment to determine the accuracy of each response on two separate occasions. Point-by-point analys is was conducted for each subject's responses during each of their two test sessions to determine whether agreement regarding accuracy was present. The kappa coefficient has the value 0.79, which indicates strong agreement between the separate rating sessions and the confidence interv al of (0.70, 0.87) confirms that one can reject the null hypothesis of no agreemen t. Additionally, the percentage of task items agreed in the two occasions range from 86% to 100% for the twelve subjects. Intra-class correlation coefficient was determined to be 0.82 with a 95% confidence interval [0.71, 0.90]. Inter-rater Reliability A point-by-point analysis was completed to compare the level of agreement between two judges as to whether each indivi dual response was correct or inco rrect. The kappa coefficient has

PAGE 61

61 the value 0.68, which indicates st rong agreement between the rate rs, and the confidence interval of (0.59, 0.77) confirms that one can reject the null hypothesis of no agreement. In addition, the percentage of task items agr eed by the two raters range from 92% to 100% for the twelve subjects. Intra-class correlati on coefficient was determined to be 0.82 with a 95% confidence interval [0.71, 0.90].

PAGE 62

62 Table 4-1. Individual and group descriptive data.

PAGE 63

63 Table 4-2. Summary statistics for speech reaction time (SRT) and response accuracy by priming condition and stimulation state. SRT data is in milliseconds (ms) and the number of errors is per 16 responses.

PAGE 64

64 Table 4-3. Mean difference and p-values for speech reaction time (SRT) and response accuracy by priming condition and stimulation state.

PAGE 65

65 Figure 4-1. Mean speech reaction time (SRT) in milliseconds (ms) in the no switch and switch conditions.

PAGE 66

66 Figure 4-2. Mean speech reaction time (SRT) in millisecond s (ms) in the no switch and switch conditions on and off DBS.

PAGE 67

67 CHAPTER 5 DISCUSSION The following discussion of our research findings will be outlined as related to the primary, secondary, and explorator y research questions detailed in Chapter 1. This will be followed by a more general discussion includi ng the strengths and wea knesses of the study and alternative hypotheses. Direct ions for future research wi ll be covered in Chapter 6. Primary Aims Research Question 1 Research question 1 was designed to determine whether a significant difference exists in SRT between the switch and no switch condi tions in patients with PD and DBS. While both conditions were hypothesized to target the le vel of speech motor planning/programming, the switch condition was expected to require additional, more co mplex processes for successful completion (i.e., inhibition of the unwanted mo tor program, retrieval of new subprograms from the sensorimotor store, and the loading of these subprograms into the buffer in order to construct a new motor program). Consistent with our predictions, we found subjects responded significantly faster in terms of SRT in the no-switch condition regardless of whether DBS was on or off. Collapsed across stimulation cond ition, mean SRT for the no switch task was 643.31 ms and 722.88 ms. for the switch task. These data validate the response priming para digm and support the critical notion that the switch condition is more complex than the no switch condition due to the increased demands of this task on processes involved in speech motor planning/programming. These data support use of this paradigm to measure processes involved in speech mo tor planning/programming. Additionally, these findings are consistent with th e modern RT literature in which a variety of variables, including complexit y, are known to influence reacti on time (Henry & Rogers, 1960;

PAGE 68

68 Smiley-Oyen, Lowry, & Kerr, 2007; Smiley-Oyen & Worringham, 2001; Sternberg et al., 1978). Indeed, as previously discussed, choice reacti on time is always longer than simple reaction (Hallett, 1990). In the present experiment, re sults which did not support longer SRT for the switch condition would cast great suspicion on the employed experimental paradigm. As the no switch condition can be conceptualized as a simple RT experiment and the switch condition can be regarded as a choice RT experi ment, results which did not show increased SRT when switching speech motor programs would suggest a fatal flaw in the study as choice RT is invariably longer than simple RT. To more explicitly describe the proposed underlying mechanisms involved in speech motor planning/programming, it is necessary to turn to the model proposed by Sternberg and colleagues (1978, 1980, 1988 & 1990). Based on this model, in the no switch condition, subjects are hypothesized to retrieve subprograms for movement from the sensorimotor store. These subprograms are loaded into the moto r buffer and comprise the motor program. Participants are able to prepare their speech responses prior to the command for speech execution and maintain them in the buffer. In contrast, in the more complex switch condition, subjects are unable to prepare the desired motor program in advance, which requires that several additional processes must occur before speech can be produced when the command for execution is provided. These processes include inhibition of the previously prepared and now undesired motor program, retrieval of new subprograms from the sensorimotor store, and the loading of these subprograms into the buffer to construc t a new motor program. Thus, as found in this experiment, SRT should be longer in the no switch condition due to the additional processes required and the increased complexity of this tas k. It could further be ar gued that the temporal difference between the no switch and switch conditions is a measure of these additional

PAGE 69

69 processes. In other words, the average diffe rence across priming conditions of 79.57 ms may reflect the additional time required to inhib it the undesired motor program, retrieve new subprograms from the sensorimotor store, and load them into the buffer to construct a new motor program. Research Question 2 Research question 2 was designed to determine whether a significant difference exists in SRT in the on DBS versus the off DBS state wh en participants with PD produce a word in the no switch condition. The hypothesized theoretical level of processing of this question was maintenance of the motor program in the buffer at the level of motor planning/programming. We were interested in determining if DBS influences maintenance of the motor program in the buffer after retrieval from the sensorimotor store. As predicted, statistically significant differences in SRT were found which revealed faster performa nce on DBS when compared to off DBS. Mean SRT in the no switch condition was 615.24 ms on DBS and 671.38 ms off DBS. This leads us to conclude that DBS directly in fluences speech motor program maintenance. There are data available to suggest that mo tor program maintenance is impaired in PD (Gentilucci and Negrotti, 1999a, b; Gueye et al., 1998; Jones et al., 1992; Labutta et al., 1994; Pascual-Leone, Valls-Sole, Brasil-Neto, Cohen & Hallett, 1994; Romero et al., 2003; Sheridan, Flowers, & Hurrell, 1987; Stelmach, Garcia-C olera & Martin, 1989). Although sufficient detail regarding the theoretical underpin nings of many of these studies is not provided, the model of Sternberg et al. (1978, 1980, 1988 & 1990) suggests this deficit may occur while motor programs are held in the buffer prior to command for movement execution. Our findings suggest that maintenance of speech motor programs in the bu ffer may be enhanced by DBS in individuals with PD, at least in simple RT experiments. On average, SRT was improved by 56.14 ms in the

PAGE 70

70 no switch condition when subj ects were on DBS. These data are supported by the limb movement literature in which GPi or STN DBS improves simple RT performance (Brown et al., 1999; Kumru, Summerfield, Valldeoriola, & Valle-Sole, 2003; Schubert et al., 2002; Temel et al., 2006; van den Wild enberg et al., 2006). Of course, as reminded by Keller (1987) and Weismer (2007) among others, caution must be used when making inferences about speech movements based on other kinds of movements (e.g., limb movement). Thankfully, there is a burgeoning literature which may provide more direct support for the positive effects of DBS on maintenance of motor programs in the buffer in patients with PD. Data from the programmatic study of the effects of STN DBS on oromotor movements by Gentil and colleagues (1999, 2000; 2001; see also Pinto, Gentil, Fraix, Benabid & Pollak, 2003) have consistently shown improved la bial and lingual RT when subjects are on versus off DBS. In the simple RT paradigm us ed by these researchers, subjects produce a target force level following presentation of a stimulus fo r response. Simple RT (along with a variety of other measurements) is measured using transduc ers affixed to the lower lip and tongue and has consistently been found to be improved by STN DBS. Of note, although pa rticipants in these experiments produced several different target force levels, this work appe ars consistent with a simple RT rather than a choice RT paradigm as subjects produced repe titions of each of the requested target force levels in block. Research Question 3 Research Question 3 was designed to determine whether a significant difference exists in SRT in the on DBS versus the off DBS state when participants with PD produce a word under the switch condition. The hypothesized theoretical level of processing of this question was switching of maintenance motor programs at the le vel of the sensorimotor store during the motor

PAGE 71

71 planning/programming phase. Contrary to our pr ediction, no significant diffe rences in SRT were observed in the switch condition regardless of whether DBS was on or off. Mean SRT in the switch condition was 717.09 ms on DBS and 728.67 ms off DBS. This negative finding was unexpected consider ing previous data in the limb literature showing improvement in choice RT with DBS of GPi (Schubert et al., 2002) and STN in humans (Kumru et al., 2003; Temel et al., 2005; van den Wildenberg et al., 2006) and rodents (Temel et al., 2005). However, close inspection of these previous studies reveal the presence of at least two important differences in comparison to the pres ent work. One, the choice RT paradigms utilized in previous work did not utili ze use of a prime to prepare s ubjects for the upcoming movement command. This is in contrast to our study which u tilized a presentation of a prime word in all trials. Additionally, in the sw itch condition of our study, the pr ime word unexpectedly did not match the target word. This required subjects to activate the desired motor program and inhibit the unwanted motor program. This additional process of inhibition was not required by the simple RT paradigms in the previous work. Th e second important difference between this study and prior limb research concerns population differences. Subjects in the previous work almost invariably underwent bilateral procedures. In cont rast, only one-third (4/12) of our participants had bilateral DBS and two-thirds (8/12) had underwent unilateral le ft only surgery. In participants who had undergone bila teral DBS and consequently had right DBS, the lead in this hemisphere was turned off for the entire study and not manipulated expe rimentally. These two differences are particularly important because m ovement inhibition has been suggested to be a bilateral (Liddle, Kiehl, & Smith, 2001; Le ung & Cai, 2007) but primarily right dominant process mediated by prefrontal cerebral circui ts (Aron, Robbins, & Poldrack, 2004; Aron et al., 2003; Chambers et al., 2006; Garavan et al., 1999; Konishi et al., 1999; Leung & Cai, 2007;

PAGE 72

72 Vink et al., 2005). Especially pe rtinent to the present discus sion, Aron and colleagues (2004) assert that Converging evidence the right frontal cortex might subserve inhibitory processes underlying switching (p. 171) Since left DBS primarily influe nces the ipsilateral hemisphere, the lack of significance in SRT in the switch condition found in our study may be due to a laterality effect precipitated by the inhibition demands of this ta sk. In other words, if a large portion of the switch task involves right he misphere mediated inhibition mechanisms, it appears that left DBS would ha ve little opportunity to influe nce these contralateral neural mechanisms. The negative findings for this research que stion lead us to consider alternative explanations such as a lack of power. However, the study does not appear to be underpowered to answer this research question. Power was suffici ent to answer research questions 1 and 2. The positive findings in research question 1 provide support for the experimental paradigm. All variables between research quest ions 2 and 3 remained constant except the priming condition and power was sufficient to answer research qu estion 2. Furthermore, at a glance, mean SRT differed little between stimulation conditions (i.e., 717.09 ms off DBS and 728.67 ms on DBS) and the p-value of 0.52 do not suggest a trend in the data. Of course the participation of additional subjects would increase the likelihood of detecting statistically significant group differences, but in this case, if such differences were found, they would be unlikely to be meaningful. Perhaps the most parsimonious explanation for these findings is that stimulation only had a positive influence on maintenan ce of motor programs in the buffe r rather than on the additional processes necessary for switching. The switch condition was designed to be a more complex test of motor planning/programming due to these additional processes. To again turn to the

PAGE 73

73 model proposed by Sternberg and colleagues (1978, 1980, 1988 & 1990), the hypothesized levels of processing that must occur in the switch condition before movement can be executed include inhibition of the unwanted motor program held in the buffer, retrieval of the appropriate subprograms from the sensorimotor store, and the loading of a new motor program into the buffer. When results from research questions 2 and 3 are both considered, the data suggest that DBS improves performance at the level of the bu ffer, but not retrieval of subprograms from the store and/or inhibition of th e unwanted motor program. Secondary Aims The secondary aims of this study were to determine the effects of the experimental manipulations on response accuracy. Overall, errors were infrequently encountered. We anticipated that response accuracy would be sign ificantly decreased in the switch versus no switch condition. Contrary to this expecta tion, response accuracy was not significantly influenced by the priming condition (p = 0.15). When collapsed across DBS conditions, subjects produced a mean of 1.50 errors per 16 responses in the switch cond ition compared to 0.88 errors in the no switch conditi on. Consistent with our expecta tions, stimulation condition also did not significantly influence response accuracy in either priming condition. In the no switch condition, subjects produced a mean of 0.69 erro rs per 16 responses on stimulation and 1.06 errors off DBS (p = 0.18). In the switch c ondition, subjects produced a mean of 1.50 errors per 16 responses both on and off DBS (p = 0.72). These non-significant differences in response accuracy are like ly due to a lack of power due to the low number of errors observed. Quite simply, errors did not occur frequently enough or there were an insufficient number of stimuli to elicit a sufficient num ber of errors to find significant group differences. Of note, power anal ysis was conducted based on SRT data rather

PAGE 74

74 than response accuracy and a low number of errors are typically encountered in RT experiments, so perhaps this finding is not surprising. Once again, the limb literature in on RT in PD may be informative, particularly with regard to the lack of effects of DBS on res ponse accuracy. For example, Schubert et al. (2002) report no statistical difference in response accuracy across stimulation conditions in a variety of RT paradigms, including a visual simple RT ta sk, a visual choice RT task, and an auditory choice RT task. Other researchers have reported si milar findings in simple (Temel et al., 2006) and choice RT (van den Wildenberg et al. 2006) experiments in individuals with PD and DBS. These previous studies, along with our presen t findings, suggest that DBS may have little influence on response accuracy in simple and ch oice RT experiments. Again, this is likely explained by the relatively low fr equency of errors encountered in RT experiments. For example, we found an error rate of 6.5% on all trials completed by our study participants. Temel and colleagues (2006) report errors occurred in about 5% of trials, while Van den Wildenberg et al. (2006) found a 2% error rate. Due to the low oc currence of errors, the current study was likely underpowered to detect differences in response accuracy. Our participants made ninety-nine errors in a total of 1,526 trials. Ou r error rate of 6.5% is similar to the error rate of 6.4% reported by Spencer and Rogers (2005) in their participants with dysarthria (both hypokinetic and ataxic types). Of course, as previously discussed in Chapter 3, the response priming procedure we ut ilized was based on this and subsequent work (Spencer, 2006), so the consistency of error rates in subjects with dysarth ria is reassuring. Figure 5-1 shows the distribution of errors in the present study. Out of the 99 total errors, 64 (65%) were either a premature response (32) or a phonol ogic error (32). The other 35% included the following error types: lexical/seman tic (7), production of a previous prime (6), production of a

PAGE 75

75 previous target (6), initial sound or syllable re petition (5), no speech response (3), and production of the prime in a switch task (2). A combinati on of two or more of th ese error types was also encountered in five trials. Pleas e see Chapter 2 for methods and operational definitions used in determining error type. These errors in response accura cy are likely best explaine d by a variety of hypothesized mechanisms. Premature responses are generally the most common type of errors described in RT studies and they are generally accepted to reflect decreased inhibition of movement execution and these types of errors comprised approximately one-third (33%) of th e total errors in our study. Spencer & Rogers (2005) reported that premature responses (or early responders) occurred in 25% of the tota l errors produced by subjects w ith hypokinetic dysarthria. The next two most frequently occurr ing error types, phonol ogical (33%) and lexical/semantic errors (7%), comprised approxima tely 40% of the total errors. As defined in Chapter 2, phonological e rrors include single phoneme omissions, deletions, substitutions, and transpositions, while lexical/semantic errors cons ist of whole real word substitutions. These two errors types are best considered to be linguistically based errors, rather than errors occurring at the motor level. Using the model of Van der Merwe (1997), the hypothesized level of disruption would be at the level of linguistic symbolic planning, rath er than the level wh ich is the focus of this study, motor planning/programming. Although th e relatively high number of language based errors was unexpected, disorders of language f unction in PD such as subtle declines in expressive language function are being recognized with increasing regularity (Ellis et al., 2006; Ellis & Rosenbek, 2007). Additionall y, in our group of participants, we conducted discourse analysis of repetition of multisyl labic words, repetition of sent ences, and connected speech in order to determine linguistic competency. Some evidence of language disturbance was found in

PAGE 76

76 one-third (4/12) of our participan ts. As shown in Table 4-1, of th ese four participants, subject 2 produced one phonological error in repetition, subject 3 made one semantic error in connected speech, subject 4 produced a total of three erro rs in connected speech (two phonological and one semantic error), and subject 6 was found to pr oduce one phonological error in repetition and two errors in connected speech (one phonological and one semantic). The influence of DBS on language function in individuals with PD and DBS has received little attention other than with verbal fluency tasks which represent an al most exclusively applie d index of linguistic proficiencyin this population (Whelan, Murd och, Theodoros, Silburn, & Hall, 2005, p. 93). Whelan et al. (2005) provide some of the only av ailable data on the language specific effects of STN DBS in two patients. Although language cha nges varied with time in both the positive and negative directions, they pr imarily report changes in d ivergent language production proficiency and lexical-sem antic manipulation skills (Whelan, Murdoch, Theodoros, Silburn, & Hall, 2005, p. 99). Further studies of the langu age effects of DBS and PD await completion. The other 27% of errors produ ced included a variety of erro r types, incl uding production of a previous prime (6%), production of a prev ious target (6%), init ial sound or syllable repetition (5%), multiple errors (5%), no speech response (3%), and production of the prime in a switch task (2%). Many of these error types ca n be hypothesized to occur at the level of motor planning/programming. For example, errors such as production of a previous prime or target both seem to reflect an inability to inhibit pr evious motor programs. Production of the prime in the switch condition occurred infrequently, but provides a direct example of disordered switching of speech motor programs. Van der Merwe (1997) suggests that errors such as initial sound repetitions (i.e., neurogenic dysfluency) may al so be due to deficits at the level of motor planning/programming.

PAGE 77

77 Exploratory Aims For exploratory purposes, two measures of neuropsychological performance, the FAS test and the Stroop Color and Word Test, were admini stered both on and off DBS. The FAS is a test of phonemic verbal fluency in which an individuals ability to generate words beginning with the letters F, A, and S is counted. The FA S and other tests of phonemic verbal fluency are commonly encountered in the neurop sychological literature, as are si milar tests such as tests of semantic verbal fluency (e.g., animals). Normativ e data is available from Tombaugh, Kozak, and Rees (1999) from a large group stratified by age and education level are available for these particular stimuli. The Stroop Color and Word Test has three sect ions the word section, the color section, and the color-word section. The word section ha s randomly ordered stimulus items comprising three different color words (i.e., re d, green, and blue) printed in black ink. Subjects are instructed to correctly read aloud as many of the printed words as possible in 45 seconds. Similarly, the color section comprises symbols (i.e., XXXX) prin ted in red, green, and bl ue ink and subjects are instructed to correctly read aloud as many of the printed colors as possible in 45 seconds. Finally, the color-word section comprises the color words from the word page printed in incongruent colors from the color page. For example, the wo rd red is shown printe d in blue ink. Subjects are again provided 45 seconds and instructed to name as many of the colors the words are printed in as possible (rather than the word that is pr inted). The Stroop Color an d Word Test is thought to be a test of an individuals ability to volitionally inhibit au tomatic word reading to produce the required response. Our findings on these test s and their interpretation follow.

PAGE 78

78 FAS Test Separate Wilcoxon signed-rank tests were conducted for each stimuli item of the FAS (i.e., the letters F, A, and S) in the on versus off DBS conditi on. No statistical differences in phonemic verbal fluency between DBS conditions we re found for any of the three stimuli letters (F, p = 0.36, A, p = 0.25; S, p = 0.99), or across all stimuli (p = 0.99). For the stimulus letter F, the mean number of correct response across su bjects was 10.92 (SD = 4.34, range = 15) in the on DBS condition and 11.92 (SD = 4.54, range = 16) off DBS. Mean number of correct productions for the stimulus letter A was 9.75 (SD = 4.85, range 18) on DBS and 8.58 (SD = 3.63, range = 12) off DBS. A mean of 10.92 (SD = 5.16, range = 19) correct productions were observed on DBS compared to 11.42 (SD = 4.38, ra nge = 15) off DBS for stimulus letter S. When these values were combined across all three stimulus letters, the mean number of correct productions on DBS was 31.58 (SD = 13.12, ra nge 51) and 31.92 (SD = 11.40, range = 41) off DBS. A pre-post decline in phonemic and sema ntic verbal fluency is perhaps the most common neuropsychological finding after GPi and STN DBS surgery (Dan iele et al., 2003; De Gaspari et al., 2006; Morrison et al., 2004; Parsons, Rogers, Braaten, Woods, & Troster, 2006; Rothlind, Cockshott, Starr, & Marks, Jr., 2007; Saint-Cyr, Trepanier, Kumar, Lozano, & Lang, 2000; Trepanier, Kumar, Lozano, Lang, & Saint-Cyr, 2000). In contrast to this well-established, persistent decline in verbal fluency pre-post DBS surgery, the influence of post-operative DBS state (i.e., on and off stimulation) on this measure has received very little attention. However, Schroeder and colleagues (2003) provide some guidance in their study of phonemic verbal fluency in seven subjects on and off stimulation. Phonemic verbal fluency was found to significantly decline in the on versus the off DBS condition. PET results revealed this decline

PAGE 79

79 was accompanied by decreased regional cerebral blood flow in several area s including the right orbitofrontal cortex, the left inferior temporal gyrus, and the left inferior frontal/insular cortex (Schroeder et al., 2003, p. 447). The neurophysio logical mechanism proposed by these authors for this decline with STN DBS was decreased ac tivation of a left-sided networkincorporating the inferior frontal cortex, the insular cortex, and the temporal cortex (Schroeder et al., 2003, p. 447). Our data are not consistent with these fi ndings but other recent wo rk supports our findings. Witt et al. (2004) studied the effects of STN DBS on verbal flue ncy (including phonemic fluency) in 23 subjects with PD and found no change in verbal fluency between the on and off DBS states. The differences between thes e two previous studies might be explained by differences in the cognitive status of the partic ipants. That is, Schroede r et al. (2003) did not appear to screen or assess cognitive stat us, while Witt and colleagues (2004) excluded participants with evid ence of cognitive dysfunction, as did our current study. Regardless, the effects of DBS state on verbal fluency in indi viduals with PD demands further systematic attention. Stroop Color and Word Test Separate Wilcoxon signed-rank tests were conducted for each of th e subtests of the Stroop in the on versus off DBS conditions. No statistical diffe rences in t-scores were found for the word section (p = 0.49), the color secti on (p = 0.30), or the color-word section (p = 0.28). Further, no significant differences in interference score were f ound (p = 0.56). Overall, GPi and STN DBS have generally been found to be well to lerated procedures in terms of associated cognitive decline besides the common and persiste nt decline in verbal fluency described above (Daniele et al., 2003; Limousin et al., 1998; Pillon et al., 2000). However, meta-analysis has revealed significant, albeit small, declines in executive functions a nd verbal learning and

PAGE 80

80 memory associated with STN DBS (along with moderate declinesin semantic and phonemic verbal fluency (Parsons et al., 2006, p. 578). Studies which attemp t to determine the effects of the on and off DBS state ar e emerging. Jahanshahi et al. (2 000) reported that bilateral GPi and STN DBS improved Stroop control trial performance on versus off DBS, while the on STN DBS state worsened performance on the in terference portion of the Stroop. Pillion and colleagues (2000) found that bilate ral STN DBS improved performanc e in the word and the color portions of the Stroop, though more errors were not ed on DBS in the interference condition of the Stroop color test. Comparing the exact findings between these two studies and our own is made challenging by the fact that each study used a different version of the Stroop. Insufficient power does not appear to explain our nonsignificant findings as Pillon et al. found significant group differences in a similar group of 13 subject s (six with GPi DBS & with seven with STN DBS). Regardless, many more data are needed un derstand he effects of DBS state on measures of cognitive function, including response inhibition. Strengths/Weaknesses This study has several strengths which a llow it to make a contribution to the understanding of the speech effects of DBS in PD, particularly at the level of motor planning/programming. The experi mental design was rigorous and controlled for many threats to internal and external va lidity. Particular strengths of the design include the use of double blind testing, a thorough washout for both stimulation and medication effects, rigi d inclusion/exclusion criteria, and the use of an objective measuremen t approach for determining speech effects (i.e., SRT). These strengths are not insi gnificant. To our knowledge, double blind testing has not previously been conducted in the literature whic h has been focused on the speech effects of DBS

PAGE 81

81 in patients with PD. The use of thorough washouts of DBS is also an impo rtant aspect of this work. Many previous data are collected from studi es with inadequate wa shouts and we utilized two two-hour washouts before the DBS condition was implemented. The strict inclusion/exclusion criteria a llowed us to obtain a homogenous and representative sample of individuals with PD. Finally, ou r use of a SRT paradigm allowe d us to measure the primary variable of interest using an instrument al, objective measurement tool. Weaknesses of the study are also present. The small sample size (n = 12) is a clear limitation which highlights the preliminary nature of our findings. However, the sample size appeared to be sufficient to answer the primar y research questions. The strength of the study is also compromised by the fact that the exact implantation site is unknown in 75% (9/12) of participants. Although all participants underwent un ilateral or staged bila teral procedures, threequarters of participants were recruited from a la rger surgical trial which seeks to compare the effects of GPi and STN DBS in a blinded fash ion. Although the exact surgical sites will be known upon completion of the larger trial, at the pr esent, determining differ ential effects of GPi or STN DBS on SRT is impossible. Our incomplete knowledge regarding the exact processes involved in speech motor planning/programming and their neurophysiological correlates is another limitation. For example, the anatomical lo cations of the speech sensorimotor store and buffer have not been established experimentally. Another limitation of the study is the quality of the digital recordings obtained during the appropriate portions of the screening and test sessions. These recordings were sufficient for the purposes of the present study, which include aiding in the description of study particip ants by determining linguistic competency, intelligibility scores, and speech diagnosis and severity, as well as for determining response accuracy. However, due to the fact that the r ecordings were made during screening and testing sessions conducted at

PAGE 82

82 patients homes or in a clinic environment, they are of insufficient quality to use for additional, more precise analyses of the speech signal, such as temporal measurement of word length or acoustic analyses. Alternative Explanations Alternative explanations fo r our findings must be considered. Chief among these alternative considerations must be the notion that the findings are due to a different process than speech motor planning/programming. A variety of cognitive influences could be used to explain our findings including global c ognitive function, attention an d concentration, and working memory. However, the MMSE was used to sc reen for global cognitive dysfunction and participants with MMSE < 26 were exclude d. Mean MMSE was 28.67 (SD = 0.98). Changes in overall cognitive function thus appear unlikely to explain our findings for SRT as global cognitive function was intact for our participan ts. The forward and backward portions of the WMS-III Spatial Span were used to screen fo r disorders of attention and working memory. Individuals with standard scores < 7 on either subtest were excl uded. Mean Spatial Span forward standard score was 10.17 (SD = 2.04) and Spatial Span backward standard score was 10.83 (SD = 1.99). These means are both within the normal range and thusly, changes in attention or working memory appear unlikely to explain our findings. The response priming procedure we utilized also does not seem to support cognitive mechanisms such as working memory to explain our findings. This testing procedure was designed to make little demand on cognitive function in general. Additionally, we argue that this paradigm is not a test of working memory, as subjects were provided with the target immediately upon command for execution.

PAGE 83

83 Another alternative explanation for these findings is that th e differences we found are due to execution level speech deficits, rather than the deficits in speech motor planning/programming. This explan ation does not appear plausible, as the primary dependent measure of SRT was calculated before speech was executed. The response accuracy data also support deficits primarily at the level of motor pla nning/programming, alt hough deficits at the level of execution also appeared to be present. This notion is based on the perceptual assessment that sound distortions occurred duri ng production of words in the response priming test on 4% of all trials. Sixty-five sound distor tions were perceived in 1,526 trials, 38 while subjects were on DBS and 27 off DBS. Although Van der Merwe ( 1997) suggests that distortions may occur due to programming level deficits, it is traditional to attribute this type of error to execution level deficits. Therefore, it appears plausible that the hypokinetic dysarthria of our subjects was influenced by deficits in both pl anning/programming and execution. Discussion Conclusion We conducted an experiment in subjects with DBS and PD in which participants completed a response priming protocol in two pr iming conditions (i.e., switch and no switch) both on and off DBS. SRT was found to be significantly different across the priming conditions in that subjects produced a word more quickly in the no switch versus the switch condition. Our participants were also found to pr oduce a word more quickly in the no switch condition when on versus off DBS. The proposed mechanism of this improvement is an increased ability to maintain the motor program in the buffer prior to the command for execution. SRT was not significantly different in the switc h condition across DBS st ates, suggesting that DBS has little influence on mechanisms involved in switching of speech motor programs (i.e.,

PAGE 84

84 inhibition of unwanted motor programs or retrie val of new subprograms from the sensorimotor store). Traditionally, the speech deficits in individuals with hypoki netic dysarthria (and indeed all dysarthria types) have been considered to be execution level deficits (Darley, Aronson, & Brown, 1969a, b, 1975; Duffy, 2005; York ston et al. 1999). However, this conceptualization of dysarthria as strictly an execution level disord er has been questioned by Kent and associates (Kent & Rosenbek, 1982; Kent et al. 1997), as well as more recen t experimental findings from Spencer & Rogers (2005; see also Spencer 2006 ). Our present findings also support the notion that individuals with PD and hypoki netic dysarthria have speech de ficits at the level of motor planning/programming. Furthermore, our finding s suggest that these planning/programming deficits can be measured using a SRT paradigm. Finally, DBS of the GPi and/or STN appears to differentially influence the motor planning/prog ramming processes required in the different priming conditions of our experiment. In othe r words, our findings suggest that DBS is associated with an improvement in the maintena nce of the speech motor program in the buffer but not the multiple processes involved in switching of speech motor programs.

PAGE 85

85 Figure 5-1. Distribution of errors differentiated among error type. PR = premature response, PHONO = phonological error, LS = lexical /semantic error, PPP = production of previous prime, PPT = production of previous ta rget, ME = multiple errors (i.e., two or more of other error types), ISR = initial sound repetition, NR = no response, P = production of the prime in switch trial.

PAGE 86

86 CHAPTER 6 FUTURE WORK Our experience with this resear ch project suggests many avenues for future work in this area. Experiments to further determine how robust the positive influences of DBS on maintaining speech motor programs in the buffer are warrante d. This can be accomplished in a number of ways, such as by varying the length of delay or with the use of articula tory suppression between presentation of the prime and target. Further investigations into the laterality e ffects of DBS in PD on speech motor program maintenance and switching are also warranted. As pr eviously discussed in Chapter 5, our lack of significant differences in the switch condition be tween the on and off DBS states may have been due to the fact that subjec ts with unilateral left DBS were targeted for recruitment due to the critical nature of the left hemisphere in speech and language. Although subjects with bilateral DBS also participated, right DBS was turned of f for the entirety of th e experiment. Since the inhibition process involved in the switch condition may rely heavily on right hemisphere cerebral circuitry, a comparison between PD subj ects with right and left DBS may assist in further determining the effects of DBS on the sw itching of speech motor programs. It might be expected that right hemisphere DBS would improve performance in the s witch condition due to an improved ability to inhibit unwanted motor programs. Such a paradigm would also allow a comparison on the effects of left and right DBS on main tenance of speech motor programs. Subsequent work in this area may be impr oved by collecting execution level speech data in addition to the motor planni ng/programming variable s studied in the present experiment. For example, data such as the duration of movement during target speech pr oductions would target the level of execution and complement the SRT planning/programming data we collected. We attempted to complete post-hoc analysis of move ment time in the present experiment, but were

PAGE 87

87 unable to reliably make these measurements due to the presence of extraneous noise in the acoustic signal. This was presumably due to the environments the digital recordings were made (i.e., subjects homes and clini cal environments). A lthough this was convenient for participants and aided in recruitment, it woul d be a significant improvement to make acoustic recordings in a sound treated room to more purely capture the sp eech signal. This would also provide the highquality digital recordings necessary for acoustic analysis of the speech signal which would allow for additional insights at the level of execution. Although acoustic analysis of the speech effects of DBS in subjects with PD has been comple ted previously (Dromey et al., 2000; HoffmanRuddy et al. 2001, Wang et al., 2003), these studi es suffer methodological limitations such as unspecified stimulation washouts and small sample sizes. Another interesting ar ea for future study would be modification of the paradigm in order to compare limb planning/programming with spe ech planning/programming. This is important because of the differential responses across th e corticospinal and corticobulbar systems to treatments for PD (e.g., DBS, levodopa) that are co mmonly reported in the li terature. However, data from Adams and colleagues (2004) suggest that the reported differential response of these systems to levodopa, for example, may be due to differences in the measurement approaches used rather than true differenc es. If appropriate ly modified, the employed experimental paradigm would allow for comparisons across these two sy stems using the same measurement approach (i.e., RT). Such an approach could facilitate further understanding of how treatments for PD influence different movements. Finally, overall, the participants in our study were judged to have only mild-moderate dysarthria on average. Only two of the 12 were on the more severe end of the severity spectrum with moderate-severe dysarthria. This may have caused a ceiling effect in terms of speech

PAGE 88

88 improvements with DBS. Further study in pati ents with more severe dysarthria would be beneficial to more completely understand the sp eech effects of DBS in individuals with PD.

PAGE 89

89 APPENDIX A TRAINING SESSION STIMULI

PAGE 90

90 APPENDIX B EXPERIMENTAL STIMULI

PAGE 91

91 LIST OF REFERENCES Adams, S. G., Jog, M., Eadie, T., Dykstra, A., Gauthier, G., & Ve rcher, J. L. (2004). Jaw and finger movements during visual and auditory motor tracking in Parkinson disease. Journal of Medical Speech-Language Pathology, 12, 125-130. Alexander, G. E., DeLong, M. R., & Stric k, P. L. (1986). Parallel organization of functionally segregated circuits linking basal ganglia and cortex. Annual Review of Neuroscience, 9, 357-381. Aron, A. R., Fletcher, P. C., Bullmore, E. T ., Sahakian, B. J., & R obbins, T. W. (2003). Stop-signal inhibition disrupted by damage to right inferior frontal gyrus in humans. Nature Neuroscience, 6, 115-116. Aron, A. R., Robbins, T. W., & Poldrack, R. A. (2004). Inhibition a nd the right inferior frontal cortex. Trends in Cognitive Sciences, 8, 170-177. Benabid, A. L. (2003). Deep brain s timulation for Parkinsons disease. Current Opinion in Neurobiology, 13, 693-706. Benabid, A. L., Pollak, P., Gao, B., Hoffmann, D., Limousin, P., Gay, E., et al. (1996). Chronic electrical stimulation of the ventralis intermedius nucleus of the thalamus as a treatment of movement disorders. Journal of Neurosurgery, 84, 203-214. Benabid, A. L., Pollak, P., Gross, C., Ho ffmann, D., Benazzouz, A., Gao, B., et al. (1994). Acute and long-term effects of subthalami c nucleus stimulation in Parkinsons disease. Stereotactic and Functional Neurosurgery, 62, 76-84. Benecke, R., Rothwell, J. C., Dick, P. R., Day, B. L., & Marsden, C. D. (1987). Disturbance of sequential movements in patients with Parkinsons disease. Brain, 110, 361-379. Benton, A. L., & Hamsher, K. (1976). Multilingual aphasia examination: Manual of instruction. Iowa City: University of Iowa. Bergman, H., & Deuschl, G. (2002). Pathophys iology of Parkinsons disease: From clinical neurology to neuroscience and back. Movement Disorders, 17, S28-S40. Bloxham, C. A., Mindel, T. A., & Frith, C. D. (1984). Initiation and execution of predictable and unpredictable move ments in Parkinson's disease. Brain, 107, 371-384.

PAGE 92

92 Brooks, V. B. (1986). The neural basis of motor control New York: Oxford University Press. Brown, R. G., Dowsey, P. L., Brown, P., Jahans hahi, M., Pollak, P., Benabid, A. L., et al. (1999). Impact of deep brain stimulation on upp er limb akinesia in Parkinson's disease. Annals of Neurology, 45, 473-288. Chambers, C. D., Bellgrove, M. A., Stokes, M. G., Henderson, T. R., Garavan, H., Robertson, I. H., et al. (2006). Executive "brake failure" following deactivation of human frontal lobe. Journal of Cognitive Neuroscience, 18, 444-455. Contreras-Vidal, J. L., & Stelmach, G. E. (1996). Effects of Pa rkinsonism on motor control. Life Sciences, 58, 165-176. Daniele, A., Albanese, A., Contar ino, M. F., Zinzi, P., Barbier, A., Gasparini, F., et al. (2003). Cognitive and behavioural effects of chroinc stimulation on cognitive function in Parkinsons disease. Journal of Neurology, Neur osurgery, & Psychiatry, 74, 175-182. Darley, F. L., Aronson, A. E., & Brown, J. R. (1969a). Clusters of deviant speech dimensions in the dysarthrias. Journal of Speech and Hearing Research 12, 462-496. Darley, F. L., Aronson, A. E., & Brown, J. R. (1969b). Differential diagnostic patterns of dysarthria. Journal of Speech and Hearing Research, 12, 246-269. Darley, F. L., Aronson, A. E., & Brown, J. R. (1975). Motor Speech Disorders Philadelphia: W. B. Saunders Company. Davis, K. D., Taub, E., Houle, S., Lang, A. E., Dosdrovsky, J. O., Tasker, R. R., et al. (1997). Globus pallidus stimula tion activates the cortical moto r system during alleviation of parkinsonian symptoms. Nature Medicine, 3, 671-674. De Gaspari, D., Siri, C., Di Gi oia, M., Antonini, A., Isella, V., Pizzolato, A., et al. (2006). Clinical correlates and cognitive underpinnings of verbal fluency impairment after chronic subthalamic stimulation in Parkinsons disease. Parkinsonism and Related Disorders, 12, 289-295. The deep-brain stimulation for Parkins on's disease study group. (2001). Deep-brain stimulation of the subthalamic nucleus or the pa rs interna of the globus pallidus in Parkinson's disease. New England Journal of Medicine, 345, 956-963. DeLong, M. R. (1990). Primate models of move ment disorders of basal ganglia origin. Trends in Neuroscience, 13, 281-285. DeLong, M. R., & Wichmann, T. (2007). Circui ts and circuit disorders of the basal ganglia. Neurological Review, 64, 20-24.

PAGE 93

93 Desbonnet, L., Temel, Y., Visser-Vande walle, V., Blokland, A., Hornikx, V., & Steinbusch, H. W. (2004). Prem ature responding following bilate ral stimulation of the rat subthalamic nucleus is amplit ude and frequency dependent. Brian Research, 1008, 198-204. Dirnberger, G., Reumann, M., Endl, W., Li ndinger, G., Lang, W., & Rothwell. J. C. (2000). Dissociation of motor preparation from memory and attenti onal processes using movement-related co rtical potentials. Experimental Brain Research, 135, 231-240. Donovan, N. J., Velozo, C. A., & Rosenbek, J. C. (in press) The Communicative Effectiveness Survey: Investigati ng its item-level psychometrics. Journal of Medical Speech Pathology. Draper, I. T., & Johns, R. J. (1964). The diso rdered movement in parkinsonism and the effect of drug treatment. Bulletin of the Johns Hopkins Hospital, 115, 465-480. Dromey, C., Kumar, R., Lang, A. E., & Loza no, A. M. (2000). An investigation of the effects of subthalamic nucleus stimul ation on acoustic measures of voice. Movement Disorders, 15, 1132-1138. Duffy, J. R. (2005). Motor speech disorders: Substrates, differential diagnosis, and management (2nd edition). St. Louis: MO: Elsevier Mosby. Ellis, C., Okun, M. S., Gonzalez-Rothi L. J., Crosson, B., Rogalski, Y., & Rosenbek, J. C. (2006). Expressive langua ge use after PD: Deficits in use but not form. Movement Disorders, 21, 97-98. Ellis, C., & Rosenbek, J. C. (2007). The basal ganglia and expressive language: A review and directions for future research. Communicative Disorders Review, 1, 1-15. Esselink, R. A. J., de Bie, R. M. A., de Haa n, R. J., Lenders, M. W. P. M., Nijssen, P.C. G., Staal, M. J., et al. (2004). Unilateral palidot omy vs bilateral subthalamic nucleus stimulation in PD: A randomized trial. Neurology, 62, 201-207. Evarts, E. V., Teravainen, H., & Calne, D. B. (1981). Reaction time in Parkinson's disease. Brain, 104, 167-186. Gale, J. T., Amirnovin, R., Williams, Z. M., Flaherty, A. W., & Eskander, E. N. (in press). From symphony to cacophony: Pathophysiology of the human basil ganglia in Parkinson disease. Neuroscience and Biobehavioral Reviews. Garavan, H., Ross, T. J., & Stein, E. A. (1999). Right hemispheric dominance of inhibitory control: An even t-related functional MRI study. Proceedings of the National Academy of Sciences of the Unit ed States of America, 96, 8301-8306.

PAGE 94

94 Gentil, M., Chaubin, P., Pinto, S., Pollak, P., & Benabid, A. L. (2001). Effect of bilateral stimulation of the subthalamic nucleus on parkinsonian voice. Brain and Language, 78, 233-240. Gentil, M., Garcia-Ruiz, P., Pollak, P., & Bena bid, A.-L. (1999). Effect of stimulation of the subthalamic nucleus on oral control of patients with parkinsonism. Journal of Neurology, Neurosurgery, & Psychiatry, 67, 329-333. Gentil, M., Garcia-Ruiz, P., Pollak, P., & Bena bid, A.-L. (2000). Effect on bilateral deepbrain stimulation on oral control of patients with parkinsonism. European Neurology, 44, 147-152. Gentilucci, M., & Negrotti, A. (1999a). The control of an action in Parkinson's disease. Experimental Brain Research, 129, 269-277. Gentilucci, M., & Negrotti, A. (1999b). Planning and executing an action in Parkinson's disease. Movement Disorders, 14, 69-79. Ghika, J., Villemure, J.-G., Fankhauser, H., Fa vre, J., Assal, G., & Ghika-Schmid, F. (1998). Efficiency and safety of bilateral cont emporaneous pallidal stimulation (deep brain stimulation) in levodopa-responsive patients with Parkinsons disease with severe motor fluctuations: A 2 year follow-up review. Journal of Neurosurgery, 89, 713-718. Golden, C. J., & Freshwater, S. M. (2002). The Stroop color and word test: A manual for clinical and experimental uses. Wood Dale, IL: Stoelting Co. Grill, W. M., Snyder, A. N., & Miocinovic, S. (2004). Deep brain stimulation creates an informational lesion of the stimulated nucleus. Neuroreport, 15, 1137-1140. Gross, R. E. (2004). Deep brain stimulat ion in the treatment of neurological and psychiatric disease. Expert Review of Neurotherapeutics, 4, 465-478. Gueye, L., Viallet, F., Legallet, E., & Trouc he, E. (1998). The use of advance information for motor preparation in Parkinsons disease: Effects of cueing and compatibility between warning and imperative stimuli. Brain and Cognition, 38, 66-86. Harrington, D. L., & Haaland, K. Y. (1991) Sequencing in Parkinson's disease: Abnormalities in programming and controlling movement. Brain, 114, 99-115. Henry, F. M., & Rogers, D. E. (1960). In creased response latency for complicated movements and a memory drum theory of neuromotor reaction. The Research Quarterly, 31, 448-458. Hikosaka, O. (2007). GABAergic out put of the basal ganglia. Progress in Brain Research, 160, 209-226.

PAGE 95

95 Hoffman-Ruddy, B., Schulz, G., Vitek, J., & Ev att, M. (2001). A preliminary study of the effects of subthalamic nucleus deep brain stim ulation on voice and speech characteristic in Parkinsons disease. Clinical Linguistics & Phonetics, 15, 97-101. Inzelberg, R., Plotnik, M., Flash, T., Schechtman, E., Shahar, I., & Korczyn, A. D. (2001). Mental and motor switching in Parkinson's disease. Journal of Motor Behavior, 33, 377-385. Jahanshahi, M., Ardouin, C. M., Brown, R. G ., Rothwell, J. C., Obeso, J., Albanese, A., et al. (2000). The impact of deep brain stimula tion on executive function in Parkinson's disease. Brain, 123, 1142-1154. Jones, H. N., Kendall, D. L., Sudhyadhom, A ., & Rosenbek, J. C. (in press). The effects of lesion therapy and deep brain stimulation on speech function in patients with Parkinsons disease. Communicative Disorders Review Jones, D. L., Phillips, J. G., Bradshaw, J. L., Iansek, R., & Bradshaw, J. A. (1992). Programming of single movements in Parkinso n's disease: Comparison with Huntington's disease. Journal of Clinical and Expe rimental Neuropsychology, 14, 762-772. Keller, E. (1987). The cortical representation of motor processes of speech. In E. Keller & M. Gopnik (Eds.). Motor and sensory processes of language (pp. 125-162). Hillsdale, NJ: Lawrence Erlbaum Associates. Kent, R. D., Kent, J. F., & Rosenbek, J. C. (1987). Maximum performance tests of speech production. Journal of Speech and Hearing Disorders, 52, 367-387. Kent, R. D., Kent, J. F., Rosenbek, J. C., Vorperian, H. K., & Weismer, G. (1997). A speaking task analysis of the dys arthrias in cereb ellar disease. Folia Phoniatrica et Logopaedica, 49, 63-82. Kent, R. D., & Rosenbek, J. C. (1982). Pr osodic disturbance and neurologic lesion. Brain and Language, 15, 259-291. Klapp, S. T. (2003). Reaction time analysis of two types of motor preparation for speech articulation: Action as a sequence of chunks. Journal of Motor Behavior, 35, 135-150. Koller, W. C., Pahwa, R., Lyons, K. E., & Albanese, A. (1999). Su rgical treatment of Parkinsons disease. Journal of the Neurological Sciences, 167, 1-10. Konishi, S., Nakajima, K., Uchida, I., Kikyo, H., Kameyama, M., & Miyashita, Y. (1999). Common inhibitory mechanism in human in ferior prefrontal cortex revealed by eventrelated functional MRI. Brain, 122, 981-991.

PAGE 96

96 Krack, P., Batir, A., Van Blercom, N., Chabardes, S., Fraix, V., Ardoun, C., et al. (2003). Five-year follow-up of bilateral st imulation of the subthalamic nuc leus in advanced Parkinsons disease. New England Journal of Medicine, 349, 1925-1934. Kropotov, J. D., & Etlinger, S. C. (1999). Selection of actions in the basal gangliathalamocortical circuits : Review and model. International Journal of Psychophysiology, 31, 197-217. Kumar, R., Lozano, A. M., Kim, Y. J., Hutchison, W. D., Si me, E., Halket, E. et al. (1998a). Evaluation of subthalamic nucleus deep brain stimulation in advanced Parkinsons disease. Neurology, 51, 850-855. Kumar, R., Lozano, A. M., Montgomery, E ., & Lang, A. E. (1998b). Pallidotomy and deep brain stimulation of the pallidum and subthalamic nucleus in advanced Parkinsons disease. Movement Disorders, 13, 73-82. Kumru, H., Summerfiel d, C., Valldeoriola, F., & Valls -Sole, J. (2003). Effects of subthalamic nucleus stimulation on characterist ics of EMG activity underlying reaction time in Parkinson's disease. Movement Disorders, 19, 94-100. Labutta, R. J., Miles, R. B., Sanes, J. N ., & Hallett, M. (1994). Motor program memory storage in Parkinson's disease patients tested with a delayed response task. Movement Disorders, 9, 218-222. Lashley, K. S. (1917). The accuracy of moveme nt in the absence of excitation from a moving organ. American Journal of Physiology, 43, 169. Leung, H. C., & Cai, W. (2007). Common and differential ventrolateral prefrontal activity during inhibition of hand and eye movements. Journal of Neuroscience, 27, 9893-9900. Levelt, W. J. & Wheeldon, L. (1994). Do speakers have access to a mental syllabary?. Cognition, 50, 239-269. Liddle, P. F., Kiehl, K. A., & Smith, A. M. (2001). Event-related fMRI study of response inhibition. Human Brain Mapping, 12, 100-109. Limousin, P., Krack, P., Pollak, P., Benazzouz, A., Ardoun, C., Hoffmann, D. et al. (1998). Electrical stimulation of the subthalamic nucleus in a dvanced Parkinsons disease. The New England Journal of Medicine, 339, 1105-1111. Lozano, A. M., Dostrovsky, J., Chen, R., & Ashby, P. (2004). Deep brain stimulation for Parkinson's disease: Disrupting the disruption. Lancet Neurology, 1, 225-231.

PAGE 97

97 Marks, Jr., W. J. (2005). Programming thalamic and palatal deep brain stimulators for the treatment of movement disorders (CD-ROM). 2005 American Academy of Neurology Marsden, C. D., & Obeso, J. A. (1994). The functions of the basil ganglia and the paradox of stereotaxic surgery in Parkinsons disease. Brain, 117, 877-897. Mink, J. W. (1996). The basal ganglia: Focu sed selection and inhibition of competing motor programs. Progress in Neurobiology, 50, 381-425. Mink, J. W. (2003). The basal ganglia and involuntary movements. Neurological Review, 60, 1365-1368. Mink, J. W., & Thach, W. T. (1993). Basal gang lia circuits and their role in behavior. Current Biology, 3, 950-957. Montgomery, Jr., E. B., Baker, K. B., Lyons, K. & Koller, W. C. (2000). Motor initiation and execution in essential tremor and Parkinsons disease. Movement Disorders, 15, 511-515. Morrison, C. E., Borod, J. C., Perrine, K., Beric, A., Brin, M. F., Rezai A., et al. (2004). Neuropsychological functioning following bila teral subthalamic nucleus stimulation in Parkinsons disease. Archives of Clinic al Neuropsychology, 19, 165. Muller, T., Eising, E., Khun, W., Buttner, T., Coenen, H. H., & Przuntek, H. (1999). Delayed motor response correlate s with striatal degenerati on in Parkinsons disease. Acta Neurologica Scandinavica, 100, 227-230. Obwegeser, A. A., Uitti, R. J., Witte, R. J., Lucas, J. A., Turk, M. F., & Wharen, Jr., R. E. (2001). Quantitative and qualitative outcome meas ures after deep brain stimulation to treat disabling tremors. Neurosurgery, 48, 274-284. Okun, M. S., & Foote, K. D. (2005). Subthalamic nucleus versus globus pallidus interna deep brain stimulation, the rematch: Will pallidal deep brain stimulation make a triumphant return?. Archives of Neurology, 62, 533-536. Ostergaard, K., Sunde, N., & Dupont, E. (2002) Effects of bilatera l stimulation of the subthalamic nucleus in patients with severe Parkinsons disease and motor fluctuations. Movement Disorders, 17, 693-700. Parsons, T. D., Rogers, S. A., Braaten, A. J., Woods, S. P., & Troster, A. J. (2006). Cognitive sequelae of subthalamic nucleus deep brain stimulation in Parkinsons disease: A meta-analyses. Lancet Neurology, 5, 578-588. Pascual-Leone, A., Valls-Sole, J., Brasil-Neto, J. P., Cohen, L. G., & Hallett, M. (1994). Akinesia in Parkinson's disease. I. Shortening of simple reaction time with focal, single-pulse transcranial magnetic stimulation. Neurology, 44, 884-891.

PAGE 98

98 Pillon, B., Ardouin, M. A., Damier, P., Krack, P., Houete, J. L., Klinger, H., et al. (2000). Neuropsychological changes between off and on STN of GPi stimulation in Parkinsons disease. Neurology, 55, 411-418. Pinto, S., Gentil, M., Fraix, V., Benabid, A.-L ., & Pollak, P. (2003). Bilateral subthalamic stimulation effects on oral force control in Parkinsons disease. Journal of Neurology, 250, 179-187. Pinto, S., Thobois, S., Costes, N., Le Bars, D., Benabid, A.-L., Broussole, E., et al. (2004). Subthalamic nucleus stimulation in dys arthria in Parkinsons disease: A PET study. Brain, 127, 602-615. Robertson, C., & Flowers, K. A. (1990). Motor set in Park insons disease. Journal of Neurology, Neurosurgery, and Psychiatry, 53, 583-592. Rodriguez-Oroz, M. C., Obeso, J. A., Lang, A. E., Houeto, J.-L., Pollak, P., Rehncrona, S., et al. (2005). Bilate ral deep brain stimulation in Parkin sons disease: A mu lticentre study with 4 years follow-up. Brain, 128, 2240-2249. Romero, D. H, Van Gemmert, A. W., Adler, C. H., Bekkering, H., & Stelmach, G. E. (2003). Altered aiming movements in Parkinson' s disease patients and elderly adults as a function of delays in movement onset. Experimental Brain Research, 151, 249-261. Romito, L. M., Scerrati, M., Contarino, M. F., Iacoangeli, M., Bentiboglio, A. R., & Albanese, A. (2003). Bilateral high frequency subt halamic stimulation in Parkinsons disease: Long-term neurological follow-up. Journal of Neurological Sciences, 43, 119-128. Rothlind, J. C., Cockshott, R. W., Starr, P. A., & Marks, Jr., W. J. (2007). Neuropsychological performance following staged b ilateral pallidal or subthalamic nucleus deep brain stimulation for Parkinsons disease. Jouraml of the International Neuropsychological Society, 13, 68-79. Rousseaux, M., Krystkowiak, P., Kozlowski, O., Ozsancak, C., Blond, S., & Destee, A. (2004). Effects of subthalamic nucleus stimul ation on parkinsonian dysarthria and speech intelligibility. Journal of Neurology, 251, 327-334. Roy, E. A., Saint-Cyr, J., Taylor, A., & La ng, A. (1993). Movement sequencing disorders in Parkinson's disease. International Journal of Neuroscience, 73, 183-194. Rubchinsky, L. L., Kopell, N., & Sigvardt K. A. (2003) Modeling facilitation and inhibition of competing motor programs in basal ganglia subthalamic nucleus-pallidal circuits. Proceedings of the National Academy of Scie nces of the United States of America, 100, 14427-14432.

PAGE 99

99 Saint-Cyr, J. A., Trepanier, L. L., Kumar, R., Lozano, A. M., & Lang A. E. (2000). Neuropsychological consequences of chronic bilate ral stimulation of the subthalamic nucleus in Parkinsons disease. Brain, 123, 2091. Santens, P., De Letter, M., Van Borsel, J., De Reuck, J., & Caemaert, J. (2003). Lateralized effects of subthalamic nucleus stimulation on different aspects of speech in Parkinsons disease. Brain and Language, 87, 253-258. Schmidt, R. A. (1975). A schema theory of discrete motor skill learning. Psychological Review, 82, 225-260. Schroeder, U., Kuehler, A., Lange, K. W., Ha slinger, B., Tronnier, V. M., Krause, M., et al. (2003). Subthalamic nucleus stimulation a ffects a frontotemporal network: A PET study. Annals of Neurology, 54, 445-450. Schubert, T., Volkmann, J., Muller, U., Sturm, V ., Voges, J., Freund, H. J., et al. (2002). Effects of pallidal deep brain stimulation a nd levodopa treatment on reaction-time performance in Parkinson's disease. Experimental Br ain Research, 144, 8-16. Schulz, G. M. (2002). The effects of speech therapy and pharmacological treatments on voice and speech in Parkinsons disease: A review of the literature. Current Medicinal Chemistry, 9, 1359-1366. Schulz, G. M., & Grant, M. K. (2000). Effects of speech therapy and pharmacologic and surgical treatments on voice and speech in Park insons disease: A review of the literature. Journal of Communication Disorders, 33, 59-88. Schupbach, W. M. M., Chastan, N., Welter, M. L., Houeto, J. L., Mesnage, V., Bonnet, A. M., et al. (2005). Stimulation of the subthala mic nucleus in Parkinsons disease: A 5 year followup. Journal of Neurology, Neur osurgery, & Psychiatry, 76, 1640-1644. Sheridan, M. R., Flowers, K. A., & Hurre ll, J. (1987). Programming and execution of movement in Parkinsons disease. Brain, 110, 1247-1271. Smiley-Oyen, A. L., Lowry, K. A., & Kerr, J. P. (2007). Planning and control of sequential rapid aiming in adults with Parkinson's disease. Journal of Motor Behavior, 39, 103-114. Smiley-Oyen, A. L., & Worringham, C. J. ( 2001). Peripheral constraint versus on-line programming in rapid aimed sequential movements. Acta Psychologica, 108, 219-245. Smith, D. J. (2004). Motor progra mming. Retrieved October 8, 2008 from http://www.smithsrisca.demon.co.uk/motor-programming.html

PAGE 100

100 Spencer, K. A. (2006, March). Effect of medi cation withdrawal on response preparation in Parkinsons disease: Preliminary evidence. Poster presentation at the Speech Motor Control conference, Austin, TX. Spencer, K. A., & Rogers, M. A. (2005). Speech motor programming in hypokinetic and ataxic dysarthria. Brain and Language, 94, 347-366. Stelmach, G. E., Garcia-Colera, A., & Mar tin, Z. E. (1989). Force transition control within a movement sequence in Parkinson's disease. Journal of Neurology, 236, 406-410. Sternberg, S., Knoll, R. L., Monsell, S., & Wright, C. E. (1988). Motor programs and hierarchical organization in the control of rapid speech. Phonetica, 45, 175-197. Sternberg, S., Knoll, R. L., & Turock, D. L. (1990). Hierarchical cont rol in the execution od action sequences: Tests of two invariance properties In M. Jeannerod (Ed.) Attention and performance XIII: Motor representation and control (pp. 3-55). Hillsdale, NJ: Lawrence Erlbaum Associates. Sternberg, S., Monsell, S., Knoll, R. L., & Wright, C. E. (1978). The latency and duration of rapid movement sequences: Comparisons of speech and typewriting. In G. E. Stelmach (Ed.) Information processing in motor control and learning (pp. 117-152). New York: Academic Press. Sternberg, S., Wright, C. E., Knoll, R. L., & Monsell, S. (1980). Motor programs in rapid speech: Additonal evidence. In R. A. Cole (Ed.) Perception and production of fluent speech (pp.507-534). Hillsdale, NJ: Lawrence Erlbaum Associates. Temel, Y., Blokland, A., Ackermans, L., Boon, P., van Kranen-Mastenbroek, V. H. J. N., Beuls, E. A. M., et al. (2005). Differential effect s of subthalamic nucleus stimulation in advanced Parkinson disease on reaction time performance. Experimental Brain Research, 169, 389-399. Temel, Y., Visser-Vandewalle, V., Aendekerk, B., Rutten, B., Tan, S., Scholtissen, B., et al. (2005). Acute and separate modulation of mo tor and cognitive performa nce in parkinsonian rats by bilateral stimulation of the subthalamic nucleus. Experimental Neurology, 193, 43-52. Thobois, S., Mertens, P., Guenot, M., Hermier, M., Mollion, H., Bouvard, M., et al. (2002). Subthalamic nucleus stimulation in Park insons disease: Clin ical evaluation of 18 patients. Journal of Neurology, 249, 529-534. Tombaugh, T. N., Kozak, J., & Rees, L. ( 1999). Normative data stratified by age and education for two measures of verbal fluency: FAS and animal naming. Archives of Clinical Neuropsychology, 14, 167-177. Tornqvist, A. L., Schalen, L., & Rehncrona, S. (2005). Effects of difference electrical parameter settings on the intelligibility of speech in patients with Parkinsons disease treated with subthalamic deep brain stimulation. Movement Disorders, 20, 416-423.

PAGE 101

101 Trepanier, L. L., Kumar, R., Lozano, A. M., Lang, A. E., & Saint-Cyr, J. A. (2000). Neuropsychological outcome of GPi pallidotomy and GPi or STN deep brain stimulation in Parkinsons disease. Brain Cognition, 42, 324. Van der Merwe, A. (1997). A theoretical framework for the characterization of pathological speech sensorimotor control. In M. R. McNeil (Ed.) Clinical management of sensorimotor speech disorders (pp. 1-25). New York: Thieme. van den Wildenberg, W. P., van Boxtel, G. J., van der Molen, M. W., Bosch, D. A., Speelman, J. D., & Brunia, C. H. (2006). Stimula tion of the subthalamic region facilitates the selection and inhibition of motor responses in Parkinson's disease. Journal of Cognitive Neuroscience, 18, 626-636. Vink, M., Kahn, R. S., Raemaekers, M., van den Heuvel, M., Boersma, M., & Ramsey, N. F. (2005). Function of striatum beyond inhibition and execution of motor responses. Human Brain Mapping, 25, 336-344. Vitek, J. L., & Walter, B. L. (2005). Surgic al treatments for Parkinsons disease (CDROM). 2005 American Academy of Neurology. Volkmann, J. (2004). Deep brain stimulation for the treatment of Parkinsons disease. Journal of Clinical Neurophysiology, 21, 617. Wang, E., Metman, L. V., Bakay, R., Arzbaecher, J., & Bernard, B. (2003). The effect of unilateral electrostimula tion of the subthalamic nucleus on re spiratory/phonatory subsystems of speech production in Parkinsons disease A preliminary report. Clinical Linguistics & Phonetics, 17, 283-289. Weiss, P., Stelmach, G. E., & Hefter, H. (1997). Programming of a movement sequence in Parkinson's disease. Brain, 120, 91-102. Whelan, B. -M., Murdoch, B. E., Theodoros, D. G., Silburn, P., & Hall, B. (2005). Beyond verbal fluency: Investigating the long-term effects of b ilateral subthalamic (STN) deep brain stimulation (DBS) on la nguage function in two cases. Neurocase, 11, 93-102. Weismer, G. (2007). Neural perspectives on mo tor speech disorders. In G. Weismer (Ed.) Motor speech disorders (pp. 57-91). San Diego, CA : Plural Publishing. Witt, K., Pulkowski, U., Herzog, J., Lorenz, D., Hamel, W., Deuschl, G., et al. (2004). Deep brain stimulation of the subthalamic nucleus improves cognitive flexibility but impairs response inhibition in Parkinsons disease. Archives of Neurology, 61, 697-700. Yorkston, K. M., Beukelman, D. R., St rand, E. A., & Bell, K. R. (1999). Management of Motor Speech Disorders in Children and Adults (2nd Edition). Austin, TX: Pro-Ed Inc.

PAGE 102

102 BIOGRAPHICAL SKETCH Harrison N. Jones completed his B.A. in comm unication disorders at North Carolina State University in 1996, followed by his M.A. in co mmunication disorders at Appalachian State University in 1998. Upon completion of his masters degree, Mr. Jones started his clinical fellowship as a speech-language pathologist, which he completed in 1999. Since beginning his fellowship, he has continued to practice as a spe ech-language pathologist with expertise in the evaluation and treatment neuroge nic communication and swallowing disorders in adults. He has practiced at a variety of institutions including Duke University Medical Center and Shands Hospital at the University of Florida. Mr. Jones enrolled at the University of Florida in pursuit of his Ph.D. in rehabilitation scienc e in 2004. His broad area of research interest is in motor speech disorders in patients with neurol ogical disease. He is particul arly interested in preparatory aspects of speech production (e.g., motor planning/programming) and speech disorders in individuals with Parkinsons disease. Following co mpletion of his Ph.D., Mr. Jones will return to Duke University to join th e academic faculty as an assistant professor. His primary responsibilities will be to conduct a programmatic line of research in hi s areas of scientific interest and provide evaluation a nd treatment services to adults with neurogenic speech and swallowing disorders.