<%BANNER%>

Apraxia of Speech

Permanent Link: http://ufdc.ufl.edu/UFE0042956/00001

Material Information

Title: Apraxia of Speech Change in Error Consistency Following a Multimodal Intensive Treatment (MMiT)
Physical Description: 1 online resource (59 p.)
Language: english
Creator: SZELES,DANA M
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2011

Subjects

Subjects / Keywords: APHASIA -- APRAXIA -- INCONSISTENCY -- REHABILITATION -- SPEECH -- STROKE
Clinical and Health Psychology -- Dissertations, Academic -- UF
Genre: Psychology thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Post-stroke apraxia of speech (AOS) is a motor speech disorder thought to result from disrupted motor programming for speech production and may greatly limit employment and educational opportunities, as well as functional communication. Error inconsistency, or variability in productions has been considered a potential feature of the disorder, but has not been investigated following intervention. Previous research with a Multimodal intensive Treatment (MMiT) for a post-stroke reading disorder revealed subjective changes in speech production potentially related to AOS. Similarly, we found that following MMiT, three participants with post-stroke AOS showed reduced error inconsistency across productions and increased consistency of location for errors that remained. Preliminary findings support previous reports of improved speech production in adults with AOS and evidence change in a clinically relevant speech pattern following this treatment approach. Further research may reveal maintenance of changes following intervention and individual participant characteristics which may influence treatment outcomes.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by DANA M SZELES.
Thesis: Thesis (M.S.)--University of Florida, 2011.
Local: Adviser: Conway, Tim.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2013-04-30

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2011
System ID: UFE0042956:00001

Permanent Link: http://ufdc.ufl.edu/UFE0042956/00001

Material Information

Title: Apraxia of Speech Change in Error Consistency Following a Multimodal Intensive Treatment (MMiT)
Physical Description: 1 online resource (59 p.)
Language: english
Creator: SZELES,DANA M
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2011

Subjects

Subjects / Keywords: APHASIA -- APRAXIA -- INCONSISTENCY -- REHABILITATION -- SPEECH -- STROKE
Clinical and Health Psychology -- Dissertations, Academic -- UF
Genre: Psychology thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Post-stroke apraxia of speech (AOS) is a motor speech disorder thought to result from disrupted motor programming for speech production and may greatly limit employment and educational opportunities, as well as functional communication. Error inconsistency, or variability in productions has been considered a potential feature of the disorder, but has not been investigated following intervention. Previous research with a Multimodal intensive Treatment (MMiT) for a post-stroke reading disorder revealed subjective changes in speech production potentially related to AOS. Similarly, we found that following MMiT, three participants with post-stroke AOS showed reduced error inconsistency across productions and increased consistency of location for errors that remained. Preliminary findings support previous reports of improved speech production in adults with AOS and evidence change in a clinically relevant speech pattern following this treatment approach. Further research may reveal maintenance of changes following intervention and individual participant characteristics which may influence treatment outcomes.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by DANA M SZELES.
Thesis: Thesis (M.S.)--University of Florida, 2011.
Local: Adviser: Conway, Tim.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2013-04-30

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2011
System ID: UFE0042956:00001


This item has the following downloads:


Full Text

PAGE 1

1 APRAXIA OF SPEECH: CHANGE IN ERROR CONSISTENCY FOLLOWING A MULTIMODAL INTENSIVE TREATMENT (MMiT) By DANA M. SZELES A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIR EMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2011

PAGE 2

2 2011 Dana M. Szeles

PAGE 3

3 To Grandma Szeles, who taught me that communication is both a gift and an art, to be cherished deeply and practiced frequently May she rest in peace

PAGE 4

4 ACKNOWLEDGMENTS I want to tha nk my mentor, Dr. Tim Conway for his support, guidance and encouragement throughout this project and during my time thus far at the University of Florida. I also want to thank Dr. Stacy Harnish for her dire ction, support and insight throughout this project, as well as her readiness to explore this exciting and controversial area of speech research. I would like to thank the BRRC staff for their direct cont ributions to the therapy explore d in this study and their ongoing investment in the care and rehabilitation of post stroke adults. I also want to thank Dr. John Rosenbek for sharing his experience and insight regarding the nature and study of apraxia of speech, as well as Dr. Amy Rodriguez, Tiffany Fernande s, and Dr. Lori Altmann for their valuable efforts in rating, transcribing, and helping to arrange methods of speech evaluation for this study. Also, I would like to thank my family and friends for their ongoing support and encouragement through my ongoing academic and research endeavors. Last, but certainly not least, I would like to recognize the members of my supervisory committee, Dr. Glenn Ashkanazi, Dr. Vonetta Dotson, and Dr. David Janicke fo r committing the ir time and effort to reviewing the scienti fic rigor of this project.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURES ................................ ................................ ................................ .......... 8 LIST OF ABBREVIATIONS ................................ ................................ ............................. 9 ABSTRACT ................................ ................................ ................................ ................... 10 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 13 Apraxia of Speech Definition, Significance, and Diagnosis ................................ ..... 13 Error Inconsistency in Apraxia of Speech ................................ ............................... 15 Multimodal Intensive Treatment (MMiT) in Apraxia of Speech ................................ 20 MMiT Overview ................................ ................................ ................................ 20 MMiT Stages of Treatment ................................ ................................ ............... 21 2 METHODS ................................ ................................ ................................ .............. 24 Study Design ................................ ................................ ................................ .......... 24 Study Inclusion and Exclusion Criteria ................................ ................................ .... 24 Participant Demographics ................................ ................................ ....................... 24 Apraxia of Speech Diagnosis ................................ ................................ .................. 26 Participant Speech Characteristics ................................ ................................ ......... 26 Assessment of Comorbid Language Disorders ................................ ....................... 27 Assessment of Comorbid Motor Speech Disorders ................................ ................ 28 Dependent Measure ................................ ................................ ............................... 29 Phonetic Transcription ................................ ................................ ...................... 29 Rater Reliability ................................ ................................ ................................ 30 Total Number of Errors ................................ ................................ ..................... 30 Total Error Inconsistency across Repetitions ................................ ................... 30 Consistency of Error Loca tion ................................ ................................ .......... 31 Consistency within sessions ................................ ................................ ...... 32 Consistency across sessions ................................ ................................ ..... 33 Tre atment Progress ................................ ................................ ................................ 34 Language Improvement ................................ ................................ .......................... 34 3 RESULTS ................................ ................................ ................................ ............... 36 Rater Reliability ................................ ................................ ................................ ....... 36 Total Number of Errors ................................ ................................ ........................... 36

PAGE 6

6 Error Inconsistency across Repetitions ................................ ................................ ... 36 C onsistency of Error Location ................................ ................................ ................. 37 Consistency within Sessions ................................ ................................ ............ 37 Consistency across Sessions ................................ ................................ ........... 37 Treatment Progress ................................ ................................ ................................ 38 Greatest Level Achieved ................................ ................................ .................. 38 Time Required for Each Treatment Stage ................................ ........................ 39 Language Improvement ................................ ................................ .......................... 39 4 DISCUSSION ................................ ................................ ................................ ......... 41 Research Question 1: Generalization to Accuracy of Prod uctions .......................... 41 Research Question 2: Error Inconsistency across Repeated Productions .............. 42 Research Question 3: Consistency of Error Location ................................ ............. 43 Consistency within Sessions ................................ ................................ ............ 43 Consistency across Sessions ................................ ................................ ........... 44 Research Question 4: Treatment Progress and Improved Consistency ................. 45 Research Question 5: Language Improvement and Improved Consistency ........... 46 Infl uence of Comorbid Language and Motor Speech Disorders ............................. 48 Limitations ................................ ................................ ................................ ............... 51 Future Directions ................................ ................................ ................................ .... 52 APPENDIX : SUBJECTIVE RATINGS OF SENSORY AWARENESS ........................... 54 LIST OF REFERENCES ................................ ................................ ............................... 55 BIOGRAPHICAL SKETCH ................................ ................................ ............................ 59

PAGE 7

7 LIST OF TABLES Table page 1 1 Participant demographic information ................................ ................................ .. 25 1 2 ABA 2 subtest severity ratings ................................ ................................ ............ 26 1 3 Western Aphasia Battery total scores and classification ................................ .... 27 1 4 Assessment of comorbid motor speech disorders ................................ .............. 28 2 1 Treatment hours completed and maximum levels of stimuli achieved ................ 39 2 2 Language measure outcomes ................................ ................................ ............ 40

PAGE 8

8 LIST OF FIG URES Figure page 1 1 Magnetic resonance images for Participant1 (top row), Participant2 (middle row), and Participant3 (bottom row). ................................ ................................ ... 25 2 1 Example of consistency of error location within sessions ................................ ... 33 2 2 Example of calculation for overall consistency of error location across sessions ................................ ................................ ................................ ............. 34 2 1 Consistency of error location for all three participants at pre and post treatment ................................ ................................ ................................ ............ 37

PAGE 9

9 LIST OF ABBREVIATION S ABA 2 Apraxia Battery for Adults, second edition Aphasia Language impairment Ap raxia of speech A motor speech disorder, caused by disrupted motor planning and programming for speech movements AQ Aphasia Quotient; a measure of overall severity of language impairment obtained in the Western Aphasia Battery CTOPP Comprehensive Test of P honological Processing Dysarthria A motor speech disorder, caused by neuromuscular deficiency LAC 3 Lindamood Auditory Conceptualization Test, third edition MMiT Multimodal Intensive Treatment MSE Motor Speech Examination Oral apraxia A motor speech disord er, characterized by impaired production of non speech oral movements Phonological alexia A post stroke reading disorder, characterized by impaired ability to SLP Speech language pathologist WAB Western Aphasi a Battery

PAGE 10

10 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science APRAXIA OF SPEECH: CHANGE IN ERROR CONSISTENCY FOLLOWING A MULTIMOD AL INTENSIVE TREATMENT (MMiT) By Dana M. Szeles May 2011 Chair: Tim Conway Major: Psychology Post stroke apraxia of speech is a motor speech disorder thought to result from disrupted motor planning and/or programming (Darley, 1969; Ogar, Willock, Baldo & Wilkings, 2006; McNeil, Doyle, & Wambaugh, 2000) and often seen in combination with other language or motor speech disturbances. Among originally implicated features of the disorder is error inconsistency, or variability in performance across repeated attempts at the same utterance (Johns & Darley, 1970; LaPointe & Johns, 1975; Deal & Darley, 1972; Mloch, Darley, & Noll, 1982; Wertz, LaPointe, & Rosenbek, 1984 ), which has been recently challenged by several investigators ( Odell, McNeil, Rosenbek, & Hunt er, 1990; Mauszycki, Dromey, & Wambaugh, 2007; Wambaugh, Nessler, Bennett, & Mauszycki, 2004 McNeil, Odell, Miller, & Hunter 1995). During previous intervention efforts with patients enrolled in a Multimodal intensive Treatment (MMiT) for phonological al exia, a post stroke reading disorder, subjective changes in verbal praxis were reported (Conway, Heilman, Gonzalex Rothi, Alexander, Adair, Crosson, & Heilman, 1998; Kendall, Conway, et al., 2003; Kendall, Rodriguez, et al., 2006; Kendall, Rosenbek, et al. 2008) This treatment intensively trains the unique

PAGE 11

11 multimodal features of phonemes and works to implement learned phonological principles through extensive sensorimotor practice Thus, it was proposed that MMiT might provide a mechanism for improved ora l motor programming in individuals with apraxia of speech and likewise, skills for more consistent and accurate production. As patterns of error consistency in apraxia of speech remain largely unresolved the influence of theoretically based interventions on resulting patterns of error has not yet been explored. Therefore, the purpose of this Phase I research project was to examine change in error consistency in both number and location across successive repetitions of the same target words, before and afte r MMiT. In the context of a single subject repeated measures design, three participant s with apraxia of speech and aphasia underwent 120 hours of MMiT. We administered the Repeated Trials subtest of the Apraxia Bat tery for Adults, second edition (ABA 2, Da bul, 2000) at pre and post treatment as the dependent variable. Consistency in total number and location of errors across successive repetitions of the same utterance was calculated at each time point. Following MMiT, the total number of errors produced a cross all 30 responses (three responses for each of 10 target items) decreased for two participants One participant did not show this reduction, demonstrating a total of 5 errors at both pre and post treatment. Small and medium effect sizes were seen fol lowing MMiT for reduced error inconsistency (change in total errors across successive repetitions) in all three participants, and all three demonstrated increased consistency of error location. Rates of consistency in error location across sessions were mi nimal for all three participants.

PAGE 12

12 Preliminary findings support previous reports of improved verbal praxis in adults with post stroke apraxia of speech By incorporating extensive motor practice of learned phonological principles during training, MMiT may s trengthen skills for formulating, accessing and implementing associated oral motor programs. In turn t hese treatm ent approaches may also strengthen the consistenc y and/or accuracy of performance. Additional participant and follow up data is needed to dete rmine the viability of MMiT for recovery in apraxia of speech and identify patient specific characteristics which might influence treatment outcomes.

PAGE 13

13 CHAPTER 1 INTRODUCTION Apraxia of Speech Definition, Significance, and Diagnosis Apraxia of speech, or ver bal apraxia, is a motor speech disorder characterized by nonfluent articulation and marked difficulty in the production of speech sounds. Deficits associated with the disorder may severely restrict the consistency, quality, and intelligibility of speech an d may limit both educational and employment opportunities, as well as communication with friends and loved ones. While apraxia of speech may resul t from brain trauma, tumors, or n eurodegenerative disease (Ogar, Slama, Dronkers, Amici, & Gorno Tempini, 2005 ), it is most often seen following damage to the left hemisphere as caused by vascular occlusion, or stroke (Duffy, 1995). Strokes in this population occur most often within the middle cerebral artery, which subserves the majority of sites in the brain ass ociated with language comprehension, organization, and production. It has been theorized that apraxia of speech is due to a disruption in the motor planning and programming required for speech (Darley, 1969; Ogar et al., 2006; McNeil et al., 2000), at a le vel between language (or the underlying message) and execution (or production of the message through overt communication). This is distinct from including sentence stru or peripheral biomechanic apraxia of speech represents a disruption in the ability to select and implement the

PAGE 14

14 proper oral motor programs for movements within the lips, jaw, tongue,etc. In turn, patients with apraxia of spee ch in the absence of other speech or language deficits may struggle to position their mouth appropriately during volitional speech, despite knowledge of the message they wish to convey and capacity for the required strength and range of articulatory moveme nts. Since the initial characterization of apraxia of speech by Darley in 1969, contro versy over differential diagnosi s continues due to a range of features that exist in other language and motor speech disturbances. Such features include noticeable gropi ng or attempts to control the oral articulators during speech, slowed or effortful speech, attempts to self correct speech sound errors during production, greater difficulty articulating words of increased length or complexity, and relatively intact automa tic, but diminished volitional speech (Knollman Porter, 2008; Wambaugh, Duffy, McNeil, Robin, & Rogers, 2006). Though many of these symptoms have been traditionally attributed to apraxia of speech, they may also be the result of muscle weakness or diminish ed muscular control (dysarthria), impaired language comprehen sion, poor language processing or deficient phonological planning (aspects of aphasia), or deficits in motor control for non speech oral movements (oral apraxia) (West, Bowen, Hesketh, & Vail, 2008). Thus, while apraxia of speech is not due to underlying language or neuromuscular deficits, widespread brain damage due to vascular occlusion results in high rates of apraxia of sp eech ar e extremely rare and findings for these individuals may not generalize to most clinical samples (Duffy, 2005). These complications make the

PAGE 15

15 objective assessment of apraxia of speech within this context a challenging, but clinically ne cessary endeavor. Error Inconsistency in Apraxia of Speech Among originally implicated features of apraxia of speech is the pattern of error inconsistency, or variability in performance across repeated productions of the same utterance (Johns & Darley, 197 0; LaPointe & Johns, 1975; Deal & Darley, 1972; Mloch et al., 1982; Wertz et al., 1984). Although originally considered a core trait of the disorder, recent investigations have called this feature into question, suggesting that errors may instead be highl y predictable for specific phonemes across various target words ( Wambaugh et al., 2004; Mauszycki et al., 2007; Odell et al., 1990), across repeated productions of the same target word (McNeil et al., 1995) and across samp ling occasions (Mauszycki, Wambaug h, & Cameron 2010). Despite earnest efforts to understand and predict error patterns, recent investigations have raised as many questions about the nature and clinical manifestation of speech sound errors in apraxia of speech as they have clarified. Attem pts to demonstrate the consistency of performance across time for example, have produced little agreement in the literature Mauszycki and Wambaugh (2006) for example, reported changes in the total number and position of errors (initial, medial, or fi nal) across sampling sessions. Similarly, greater predictability of errors across sampling sessions has been shown for certain phonemes, with greater change occurring for other phonemes across repeated sessions (Mauszycki et al., 2007). As these studies were l imited to observation s conducted with a single participant Mauszycki et al. (2010) conducted a group analysis of 11 individuals with apraxia of speech and aphasia in an attempt to replicate findings This study instead revealed a

PAGE 16

16 high degree of consistenc y in the total number of errors and patterns of performance for specific phonemes across repeated sampling. Thus, it remains unclear whether the location of errors (presence of errors for certain phonemes) generally varies over repeated occasions for indiv iduals with apraxia of speech, or if changes in each Additional debate remains in the literature regarding whether error consistency (for specific phonemes across different exemplars) is influenced b y blocked versus randomized conditions of presentation for target phonemes. During blocked /bat/, /bam/, ban/). Random presentation may instead intersperse presentat ion of exemplars for various phonemes (i.e. /mad/, /bop/, /net/) throughout treatment While Wambaugh et al. (2004) proposed that errors were more consistent during blocked presentation in a single participant with apraxia of speech and aphasia, group anal ysis by Mauszycki et al. (2010) again found that neither the blocked nor random presentation conditions influenced predictability of errors over time. This further highlights the inherent variability that may exist between individuals with apraxia of speec h and the lack of any one, agreed upon pattern of performance for all indivi duals who exhibit this disorder due to methodological differences across studies. In effect, it appears that ongoing lack of agreement in the literature may be attributed to differ ing methods of error analysis (individual versus group). A s previously noted, a mong recent studies which propose that errors in apraxia of speech may be consistent and predictable for specific phonemes (Wambaugh et al., 2004; Mauszycki et al., 2007), obser vations were made regarding a single speaker with apraxia of speech

PAGE 17

17 and aphasia. Furthermore, intra individual variability in these studies for error type across phonemes ranged from 0 to 58 % (Wambaugh et al., 2004) and 0 to 25% (Mauszycki et al., 2007), r espectively. These rates appear to coincide with early research by LaPointe and Horner (1976), who observed average variability (percent of repeated productions which differed from the preceding response across 10 repetitions of the same word) of 23 50% ac ross seven subjects with apraxia of speech and aphasia. In turn, it is possible that variability at the individual level may occasionally go unrecognized in group analysis (Mauszycki et al, 2010). This possibility considered, these studies also highlight the danger of attributing specific deficits and patterns of performance (i.e. consistency of errors) in single subjects to all individuals with apraxia of speech. studies, this has been quantified as the degree of consistency in error type and location across successive, repeated productions of the same target word (Johns & Darley, 1970; LaPointe & Johns, 1975; McNeil et al., 1995). Mloch et al. (1982) investigated consist ency in the productions of various target words across successive repetitions of sample readings. Mauszycki et al. (2007) examined error patterns for target phonemes across various exemplars (i.e. consistency of errors for the /b/ sound across exemplars, / bat/, /bad/, and /ban/). Still others hav e examined consistency of errors for specific phonemes across various target words over time (Mauszycki et al., 2010; Mauszycki err ors make it difficult to define general patterns of performance in apraxia of speech.

PAGE 18

18 These conflicting findings cannot be directly compared of course, due to notable differences in comorbid diagnoses among sample participants and methods of speech elicita tion across studies. However, ongoing debate over the presence of error inconsistency in apraxia of speech perhaps defines just what recent efforts have tried to clarify, predict, and better understand the very presence of error inconsistency within and between speakers with apraxia of speech. That is, individuals with apraxia of speech may indeed evidence greater difficulty or selective impairments in certain motor programs, and in turn greater predictability of errors across successive repetitions a nd over time. However, it does not appear that these selective differences occur predictably and reliably for the same phonemes across individuals. Moreover, despite suggestions that inconsistent errors may be limited, inconsistent performance on other pho nemes may evidence inadequate, but not absent motor programs for these additional speech sounds. Data is needed to definitively answer these questions. The possi bility that error consistency may be a defining characteristic of apraxia of speech has importa nt clinical implications. Mauszycki et al. (2010) report ed that patterns of predictable errors may allow for more efficient and focused treatment approaches, e.g. training phonemes that are predominantly produced with errors. Wambaugh, Kayliny ak Fliszar, West, and Doyle have applied these principles to treatment approaches for apraxia of speech, using phonemes which are consistently in error as treatment stimuli (1998). Despite improvements for a restricted set of trained phonemes, overgeneralization, or p roduction of trained sounds in place of untrained sounds has been reported ( Raymer, Haley, & Kendall, 2002; Wambaugh, Martinez, McNeil, & Rogers, 1999). Therefore, considering the likelihood that errors may instead

PAGE 19

19 be largely inconsistent both within and across individuals, treatment across a broad range of phonemes may be warranted. As proposed by Mauszycki et al. (2010), practice with specific, problematic phonemes may foster skills for creating and accessing novel oral motor programs for those phonemes which may be consistently in error. Yet additional practice with phonemes which are produced correctly and unpredictably on occasion may allow the refinement of existing, but inadequate motor programs for phonemic errors which are inconsistent on repeated attempts. By training sensorimotor and tactile kinesthetic features of a broad range of phonemes, this type of treatment may facilitate skills for identifying perceptual and production differences between sounds which vary slightly in both place (/b/ vers us /t/) and manner (/b/ versus /p/) of production. Further, extensive motor practice with multiple phonemes during phonological awareness/processing, as well as segmenting and blending activities may additionally train skills for selecting appropriate moto r sequences and self correcting production errors. Such explicit training may decrease the potential for overgeneralization. While error patterns may be expected to change following intervention and inform methods for evaluating treatment outcomes, there h ave been no studies to date examining change in patterns of error consistency for untrained target words following a broad, multimodal treatment of phonemes. Therefore, the purpose of the present study was to examine change in the number and location of er rors across repeated productions of the same utterance, and examine change following a multimodal treatment in participants with apraxia of speech and aphasia.

PAGE 20

20 Multimodal Intensive Treatment (MMiT) in Apraxia of Speech MMiT Overview During prior interventi on efforts for adults with phonological alexia enrolled in a Multimodal intensive Treatment (MMiT), subjective changes were observed in (Conway, et al., 1998; Kendall, et al, 2003; 2006; 2008) MMiT is based on a treatment progr am that was originally designed to train phonological awareness, reading, spelling and speech in children and adults, and later modified to retrain similar aspects of language and speech in adults with phonological alexia (Conway et al., 1998). Treatment f ollows a neurodevelopmental model, hierarchically training speech and language skills based on the typical sequence of acquisition for individual phonemes in children (Alexander & Slinger, 2004). As this treatment involves the integration of both language perception and processing skills as well as sensorimotor practice, it was therefore proposed that MMiT might jointly improve features of both speech and language components of communication. On examining differences in performance based on mode of stimulus presentation, Johns and Darley (1970) demonstrated that participants were able to produce more accurate articulations when visual monitoring was combined with auditory stimuli, versus either stimulus mode alone. Likewise, MMiT involves training multimodal features of phonemes and feedback of performance throughout levels of instruction. The multimodal features include opportunities for visual and auditory modeling by the therapist, feedback and comparison activities, and exercises to develop phonological a wareness, as well as motor practice and implementation of these learned skills. These activities may in turn develop learned associations between phonemes, novel phoneme sequences and their relevant motor programs. Further, MMiT uses pseudoword s

PAGE 21

21 ( phoneme c ombinations that follow English conventions but have no semantic meaning ) as target words for treatment. This allows participants to practice various phoneme combinations without added demands on, or aid from lexical or semantic processing networks. MMiT a lso advocates semi random versus blocked practice of phonemes. Random practice of various phonemes and phonemic sequences has previously been shown to improve production and enhance retention of treated phonemes in individuals with apraxia of speech and ap hasia (Knock, Ballard, Robin, & Schmidt, 2000). Thus, phoneme groups are introduced during Stage I of MMiT one at a time, and practiced until mastery before a new set of phonemes is introduced. As additional groups are incorporated into treatment, all pho nemes introduced to that point are prac ticed randomly. MMiT Stages of Treatment Stages of treatment for MMiT are hierarchically organized, and require mastery of each stage before progressing to subsequent levels. This ensures a strong foundation for highe r level principles in more advanced stages of treatment Stages of treatment progress as follows: Stage I Multimodal discovery of phonemes Training associations between visual, auditory, oral kinesthetic, oral tactile, & metalinguistic features of phonem /p/) Stage II Train phonological awareness Phonological problem solving tasks where participants must identify a single change between two pseudoword s in a series (e.g. a change of phoneme number, order, or i dentity by adding, omitting, substituting, shifting, or repeating phonemes) Stage III Train generalization Generalization training of Stage II phonological awareness to reading

PAGE 22

22 (blending phonemes) and spelling (segmenting pseudoword s) accuracy, and self correction of errors during these tasks MMiT is heavily guided by methods of Socratic q uestioning during each stage of treatment, in which participants are questioned regarding the accuracy of their productions. By encouraging participants to attend to var ious aspects of speech this method of questioning promotes skills for self monitoring and correction. Therapists may provide statements which provide auditory feedback and opportunity for technique also provides visual information by allowing participants to compare the position of their mouth with a mirror to that of their therapist. Further, sensory feedback may be provided, in which the participant must discriminate betw een various aspects of Finally, the manner of production (movements required for transitioning between sounds) may be examined through extensive practice with a variety of phoneme combinations and techniques for identifying sensation within th e articulators during rapid movement from one sound to the next Together, these training activities may help develop skills for novel oral motor programming. The present study sough t to answer the following research questions with respect to the multimodal treatment: Research Question 1: Does MMiT generalize to improved accuracy of productions during repetition of untrained real words? Research Question 2: Does MMiT reduce inconsiste ncy in the total number of errors across successive repetitions of the same target word? Research Question 3: Does MMiT increase consistency of error location across successive repetitions of the same target word, and are errors consistent in location acro ss sessions?

PAGE 23

23 Research Question 4: Is treatment progress related to the degree of improvement on measures of error consistency? Research Question 5: Do changes in language impairment relate to the degree of improvement on measures of error consistency?

PAGE 24

24 CH APTER 2 METHODS Study Design This project was appended to an ongoing study of MMiT in phonological alexia, a post stroke reading disorder. M easures of apraxia of speech were therefore added to existing measures of language perception and production to furt her examine treatment outcomes for apraxia of speech following this intervention. This was a single subject repeated measures design, with replication across three participants with apraxia of speech and aphasia. Treatment was provided two to four hours pe r day, over four to five days per week, over approximately 12 weeks, incorporating a total of 120 hours of MMiT per participant. Study Inclusion and Exclusion Criteria Participants were recruited through existing resources in the Brain Rehabilitation Resou rce Center (BRRC) of the Malcom Randall VA Medical Center, which have been well established for an ongoing study of phonological alexia a post stroke reading impairment To ensure that the period of spontaneous recovery had subsided, all participants wer e at least 6 months post stroke at the time of enrollment. Participants were not diagnosed with depression or other psychiatric illness, degenerative neurological illness, chronic medical illness, and/or severely impaired vision or hearing, which might co mpromise their ability to understand and adhere to treatment demands. Participant Demographics Participants were three right handed, monolingual English speakers who presented with left hemisphere stroke (see Figure 1 1 .) and comorbid aphasia. Participant demographic information is shown in Table 1 1.

PAGE 25

25 Figure 1 1. Magnetic resonance images for Participant1 (top row), P articipant2 (middle row), and Participant3 (bottom row). Table 1 1. Participant demographic i nformation Age Gender Education (years) Time Post Onset (months) Participant 1 60 M 15 113 Participant 2 43 M 12 19 Participant 3 60 M 16 12

PAGE 26

26 Apraxia of Speech Diagnosis All three participants had a diagnosis of apraxia of speech as documented in their initial BRRC speech language and behavioral neurology screening. Once participants met initial study criteria, we administered the Apraxia Battery for Adults, second edition, ( ABA 2 ) (Dabul, 2000) to document severity of a range of features associated with apraxia of speech and to elicit profiles o f individual participant deficits. The ABA 2 is a normed and standardized clinical measure, designed to provide a rough estimate of severity for a range of features implicated in apraxia of speech. Severity ratings for ABA 2 subtests are listed in Table 1 2. Table 1 2 AB A 2 subtest severity r atings ABA 2 Subtest Participant 1 Participant 2 Participant 3 Diadochokinetic Rate Mild Mild Mild Increasing Word Length A Mild None None Increasing Word Length B Severe Moderate Moderate Limb Apraxia None Non e None Oral Apraxia None None None Latency and Utterance Time Severe None None Repeated Trials Moderate Moderate Mild An expert speech language pa thologist (SLP) with more than 3 5 years of experience in speech language pathology and extensive work in the field of apraxia of speech research was blinded to outcomes on ABA 2 measures confirmed apraxia of speech diagnosis for all three participants This diagnosis was confirmed using audio recorded speech samples from select ABA 2 subtests. Participant Spe ech Characteristics In addition to meeting perceptual criteria as defined by ABA 2 severity scores, Participant1 also demonstrated visible and audible searching during speech production (unsuccessful attempts at finding the appropriate oral mot or position to

PAGE 27

27 produce a target sound) numerou s and varied off target productions of target phonemes, increasing number of phonemic (sound level) errors with words of increased length and complexity, difficulty initiating speech, and an awareness of error s but inability to self correct. Participant2 also demonstrated visible and audible searching behaviors increasing number of phonemic errors with words with increased length and complexity, and fewer errors in automatic versus volitional speech. Participa speech was additionally characterized by visible and audible searching behaviors numerous and varied off target productions of target phonemes during speech difficulty initiating speech, fewer errors in automatic versus volitional speech, and an aw areness of errors but inability to self correct. Assessment of Comorbid Language Disorders Degree of comorbid language impairment was obtained with the Western Aphasia Battery (WAB) (Kertesz, 1982). WAB scores and aphasia classification are listed in Table 1 3. Table 1 3. W estern Aphasia Battery total scores and c lassification Participant 1 Participant 2 Participant 3 Fluency (max 20) 13 14 17 Comprehension (max 10) 7.4 9.55 10 R epetition (max 10) 7.10 8.5 9.4 Naming (max 10) 4.8 7.2 8.6 Aphasia Quot ient (max 100) 64.6 93.2 90 Aphasia Classification Anomic Anomic Anomic Note: Totals listed for participants on WAB subtests are weighted raw scores Aphasia Quotient Weighted composite of subtest scores on the Western Aphasia Battery indicating overall severity of language impairment; a score of 100 indicates no impairment; scores of 93.8 and higher are considered normal. Anomic aphasia is characterized by scores of 5 10 on Fluency, 7 10 on Comprehension, 7 10 on Repetition, and 0 9 on Naming, as demons trated in all three participants.

PAGE 28

28 Assessment of Comorbid Motor Speech Disorders To account for possibly comorbid motor speech disorders, the Motor Speech Examination (MSE, see Freed, 2000) and a lab developed measure of sensory awareness ( see Appendix) were administered to determine the possible presence and severity of dysarthria. The MSE is a clinician rating tool designed to evaluate the integrity of the mouth and facial muscles first at rest (i.e. pre sence of droop or asymmetry in the tongue, lips, or face), then during non speech oral movement (i.e. range and amplitude of motion in lips, tongue, etc.), and finally during speech (i.e. phonation). The lab Sensory Awarene change in oral sensation and motor control. Results from these measures are listed in Table 1 4. Table 1 4. Assessment of c omorbid motor s peech d isorders Participant 1 Participant 2 P articipant 3 MSE Summary Mild deviation of mouth to the right (mandible, tongue) and deviation of mouth during production of /e/ sound Mild deviation of tongue to right; weak glottal stop, cough Mild breathiness, harshness, and pitch breaks during phonat ion Subjective Ratings of Sensory Awareness No Subjective loss of sensation; some difficulty controlling muscles in lips, inadequate jaw strength during biting (while eating only) No Subjective loss of sensation; difficulty closing right eye completely, no difficulty projecting voice or producing sufficient speaking volume Reduced sensation in right cheek/neck/inner ear, problems winking right eye; no noticeable problems in or around mouth

PAGE 29

29 The Limb and Oral Apraxia subtest of the ABA 2 was also administ ered to determine the presence and severity of limb apraxia and oral apraxia in the context of apraxia of speech. These subtests measure the accuracy and control of various non speech oral and limb movements (stick out your tongue, throw a ball, etc.). Non e of the three participants demonstrated measurable deficits for limb or oral apraxia on these subtests. Dependent Measure We administered the Repeated Trials subtest of the ABA 2 at pre and post treatm ent as the dependent variable. In this task, partici pants are asked to repeat an of words without error. Variability scores may also be c alculated to demonstrate change in total number of errors across succ essive repetitions of each target word. Phonetic Transcription Audio recorded speech samples for the Repeated Trials subtest of the ABA 2 were broadly transcribed by a trained transcripti onist who was blinded to assessment time points, using the International Phonetic Alphabet. Due to the need for qualitative assessment of speech characteristics for adequate transcription, the identity of the participants could not be masked on audio recor ded speech samples. T he transcriptionist was however, blinded to treatment progress, pre treatment speech and unique profile of performance. Two Ph.D. level speech language pathologists (SLPs) blinded to participant identities and time points used transcriptions to rate total number and location of errors

PAGE 30

30 on each of three repeated response s for all 10 target words. This included o th er disrupted speech behaviors and signs of abnormal prosody (i.e. false starts, perseverations, incorrect stress assignment). A third, blinded SLP provided a consensus rating for discrepant ratings ( participant response s which were rated differently by the two primary SLPs). Rater Reliability Relia bility of transcription ratings was calculated as percent agreement in the total number of errors for each session, location of errors (percent of phonemes scored as incorrect by both raters), and item to item agreement (percent of all target words which m atched in both number and location of errors for both raters). Total Number of Errors To answer Research Question 1 (do treatment effects generalize to accuracy of productions during repetition of untrained real words?), the total number of errors was calc ulated before and after MMiT. This was totaled across all 30 responses (t hree repetitions for each of 10 target words ) for each pa rticipant. C hange following MMiT (increase or decrease in the total number of errors) was then determined It was expected tha t following MMiT, participants would be more accurate in their productions and therefore produce fewer errors at post treatment. Total Error Inconsistency across R epetitions To answer Research Question 2 (does MMiT reduce inconsistency in the number of pho nemic d was used (Busk & Serlin, 1992) to measure effect size for change in the total number of errors for each response. As noted by Beeson and Robey (2006) this variation may be calculated

PAGE 31

31 wi th representative data points ( i.e. target words within a session) as a measure of change in untrained items Effect Size was calculated as follows: where A 1 and A 2 represent the standard deviation (SD) for the total number of errors across success ive re petitions for each target word during pre and post treatment periods respectively. S A1 is the standard deviation of calculated SD values across all target words at pre treatment. Standard deviation was calculated for each target word, and the average was calculated in this way to determine an overall inconsistency score for each participant and time point. This value could then be compared between pre and post treatment for the evaluation of change and effect size. Effect sizes for change in each partici were classified as small (.2 to .5), medium (.5 to .8) and large (>.8) (Cohen, 1988) It was expected that following MMiT, participants would be more consistent in their productions ac ross successive repetitions. In turn, inconsi stency scores (change in the number of errors across successive repetitions of each target word as measured by change in the average standard deviation across all target words ) would be reduced. Consistency of Error Location Research Question 3 (does MMiT increase consistency in error location within and between sessions?) was performed to replicate and extend previous research on typical former study suggested that indivi xia of speech who demonstrate a production error on a given phonem e are likely to demonstrate errors in the same

PAGE 32

32 word location (or on the same target phoneme) on subsequent responses of the same target word onsistency of error loca tion colleagues therefore refers to the percentage of phonemic (sound level) errors which occur repeatedly across successive repetitions of the same target word. This is c alculated for given phonemes within each target word, as oral motor demands may differ based on surrounding phonemes. For example, the /m/ sound set of oral motor demands on the range and accuracy of movement than the /m/ in Consistency within session s To answer th e first part of Research Question 3 (does MMiT increase consistency in error location within sessions?), c onsistency of error location within a session was calculated This was achieved by first totaling the number of phonemic errors present across all 30 responses (three repetitions for each of 10 target words). Next, phonemic errors occurring two or three times on the same target phoneme across the three Overall c onsistency in error location for the session was calculated as total number of (three repetitions for each of 10 target words). For example, in a participant computer, fumputa the initial position and /a/ in the final position). Of those three phonemic errors, two occur for the same phoneme across the three repetitio ns (/w/ and /f/ in place of the first phoneme). Therefore, for this first target word, two phonemic errors are consistent in

PAGE 33

33 error location. Overall consistency of error location is therefore calculated at 66.7%. This ex ample is illustrat ed in Figure 2 1 below. Target word computer Repetition 1 w omputer Repetition 2 computer Repetition 3 f omput a Total # of consistent errors 2 Total # of errors 3 Consistency of error location 66.7 % Figure 2 1. Example of consistency of e rror l ocatio n within s ession s Consistency across s essions To answer the second part of Research Question 3 (does MMiT reveal consistency in error location between sessions?), c onsistency of error location between sessions was obtained This was performed as an extensi on of previous research to further compare performance before and after MMiT. First, t he total number of errors present in the same word location (for the same phoneme) at both time point s was totaled. This was then divided by the total number of phonemic errors produced during both pre and post treatment. For instance, the presence of an error for a target phoneme during any of the three successive repetitions at one time point would be considered to be consistent acro ss sessions if it was also present du ring the other time point An example of this calculation is illustrated for one target word in Figure 2 2 below. In this example, the /a/ produced in place of the final phoneme for computer is considered consistent across sessions, as it appears for at least one repetition at pre treatment and also for at least one repetition at post treatment. This measure provides an estimate of error predictability for certain phonemes over time.

PAGE 34

34 Pre treatment Post treatment Target word computer computer Repetition 1 w omputer comput a Repetition 2 computer comput a Repetition 3 f omput a comput a Total # of consistent errors 4 Total # of errors 6 Consistency across sessions 66.67% Figure 2 2. Example of calculation for overall c onsistency of e rror l ocation a cro ss s essions Treatment Progress To answer Research question 4 (does treatment progress predict degree of improvement on measures of error consistency? ), t ime spent in each stage of treatment and highest levels of training achieved w ere recorded for each par ticipant. This wa s intended to highlight potential patterns of performance based on progress through treatment. It was expected that participants who spent less time in initial stages of oral awareness training and advanced to more complex stimuli through treatment would also demonstrate more accurate and consistent performance on measures of error consistency at post treatment. Language Improvement To answer Research question 5 (does improvement on measures of language impairment predict degree of improvem ent on measures of error consistency?) measures of phonological perception and production were administered at pre and post treatment. This was measured via the Lindamood Auditory Conceptuallization Test, third edition (LAC 3 ) ( Lindamood & Lindamood, 200 4 ) and the Comprehensive Test of Phonological Processing (CTOPP ) ( Wagner et al., 1998 ) for phonological

PAGE 35

35 perception and production, respectively. Additionally, the WAB AQ was acquired at post treatment to determine change in overall severity of aphasia. It was expected that participants who demonstrated more severe language impairment at pre treatment, as evidenced by lower scores on these measures, might also demonstrate a greater number of total errors at both time points Additionally, it was proposed tha t cognitively target and manipulate motor components of speech during MMiT; therefore, it was expected that these participants would also demonstrate poorer performance on measures of err or consistency (greater inconsistency in both number and location of errors across successive repetitions of the same target words) at both time points Finally it was expected that participants who demonstrated gains on language measures would also d emon strate gains on measures of consistency in error location and number across successive repetitions of the same target words.

PAGE 36

36 CHAPTER 3 RESULTS Rater Reliability Percent agreement for error ratings on the Repeated Trials subtest was calculated at 96% for to tal number of errors. Agreement in error location (percentage of phonemes scored as incorrect by both raters) was 95%. Raters demonstrated 91% overall item to item agreement. Total Number of Errors Research Question 1 asked whether MMiT would generalize to accurate production during repetition of untrained real words. Indeed, total number of errors produced across all 30 responses decreased from pre to post treatment assessment for Participant1 (65 to 38) and Participant2 (14 to 7). Participant3 did not sho w this reduction, demonstrating a total of 5 errors at both pre and post treatment. Thus, MMiT did generalize to accurate productions during repetition of untrained words for two participants. Error Inconsistency across Repetitions Research Question 2 ask ed whether MMiT would reduce inconsistency in the number of errors across successive repetitions of the same target word Indeed, Participant1 showed reduced error inconsistency (change in total number of errors acr oss successive repetitions of target word s ) from .58 to d =.24, small d = .78, medium d = .39, small effect). Thus, all three participants showed a treatment effect for reduced error inconsistency across repetitions, as expected.

PAGE 37

37 Consistency of Error Location Research Question 3 asked whether MMiT would increase consistency in error location (percentage of errors occurring on the same target phoneme within a target word, across more than one repetition) within and between sessions. C onsistency w ithin Sessions All three participants demonstrated increase d consistency of error location at post treatment. While participants showed a large range in performance at pre treatment (40 77%), a smaller range was demonstrated at post treatm ent (60 86%) These data are shown in Figure 2 1. Thus, following MMiT, all participants showed increased consistency of error location and also a smaller range in per formance was seen across participants. Figure 2 1. Consistency of e rror l ocation for a ll three p articipants at p re and post t reatment Consistency a cross S essions Measures of consistency in error location across sessions (total percentage of phonemic er rors which were present on a given sound segment at both pre and post 0 10 20 30 40 50 60 70 80 90 100 Participant 1 Participant 2 Participant 3 Percentage of errors consistent in location Consistency of Error Location PRE POST

PAGE 38

38 and post treatment time points, occ urred cons istently in the same location within a target word. Thirt y eight percent of errors occurred in the same location, across pre and post treatment time points. None of the errors produced at post treatment for Participant3 were consistent across both time points. Thus, a relatively small percentage of total errors at both time points were specific to a certain phoneme. This was seen across all three participants. Treatment Progress Research Question 4 asked whether treatment progress was related to degree of improvement on measures of error consiste ncy. A ll three participants demonstrated cons stage. Greatest Level Achieved Participant2 and Participant3 achieved the same maximum phoneme combination for single syllabl e pseudowords during training and the same maximum number of syllables for multisyllable pseudowords. Participant1 achieved a slightly reduced maximum number of phonemes for single syllable pseudowords, but did not progress to the multisyllable pseudowords level of training in MMiT. Partcicipant1, who reached the least complex levels of practice for pseudoword stimuli also demonstrated the greatest number of errors at pre and post treatment. While this participant also demonstrated the weakest gains in con sistency of error location from pre to post treatment, he also demonstrated the greatest initial level of consistency in error location at pre treatment.

PAGE 39

39 Time Required for Each Treatment Stage Participant1 took slightly more time to complete oral awarene ss training and spent the majority of treatment in training simple syllable phonological processing, reading and spelling. In comparison, Participant3 completed simple syllable training quite readily, spending the majority of time with more complex syllabl e words. Unfortunately, treatment data for Participant2 was not available for review for simple stages of syllable production during MMiT due to a therapist error with treatment records. However, it appears that Participant2 spent more time in more advance d stages of treatment with complex syllable pseudowords relative to Participants 1 and 2 Time to complete each stage of treatment and maximum level of training achieved are listed in Table 2 1. Table 2 1. Treatment h ours c ompleted and m aximum levels of st imuli a chieved Participant 1 Participant 2 Participant 3 Oral Awareness Training 5 4 CV/VC/CVC Treatment 115 29 Complex Syllables (CCV/VCC/CVCC/CCVC/ CCVCC) 0 74 51 Greatest number of phonemes 3 phonemes 4 phonemes 4 phonemes Greatest number of s yllables None 5 syllable pseudoword s 5 syllable pseudoword s Notes: CV Consonant Vowel P attern denotes complexity of phonemic combinations for pseudoword stimuli Greatest number of phonemes Number of phonemes incorporated into target items during pse udoword segmenting and blending activities, for single syllable pseudoword s only Greatest number of syllables Total number of syllables incorporated into m ultisyllable pseudoword targets Language Improvement Research Question 5 asked whether improvement s on language measures predicted degree of improvement on measures of error consistency. All three participants showed modest, but measurable improvements on each of three measures of language, including overall severity, phonological perception, and phono logical

PAGE 40

40 production. Pre and post treatment scores for all three participants are listed in Table 2 2. Table 2 2. Language measure o utcomes Participant 1 Participant 2 Participant 3 WAB AQ pre 64.6 93.2 90 WAB AQ post 65.6 94.8 91.8 LAC 3 pre 55 5 9 63 LAC 3 post 62 76 85 CTOPP pre 55 52 55 CTOPP post 58 61 58 Notes: WAB AQ Aphasia Quotient from the Western Aphasia Battery ; max 100 LAC 3 S tandar d scores on the Lindamood Auditory Conceptualization test third edition CTOPP S tandar d sc ores on the Comprehensive Test of Phonological Processing

PAGE 41

41 CHAPTER 4 DISCUSSION Despite a wealth of studies measuring error consistency within target words and across repeated sampling occasions in adults with post stroke apraxia of speech there have be en no studies to date examining change in error consistency following a multimodal treatment for a broad range of phonemes, including phonemes that may or may not be produced in error prior to treatment. Thus, i t was proposed that MMiT, which trains disti nctive sensory and motor features of phonemes through problem solving activities (phonological processing, phoneme blending, and phoneme segmenting) with pseudowords (phoneme sequences that follow the linguistic conventions of English, but do not have a s emantic meaning ) might improve the accuracy and/or consistency of repeated productions of real words in adults with post stroke apraxia of speech. Therefore, the purpose of the present study was to examine change in error patterns following MMiT in a sampl e of participants with post stroke apraxia of speech and aphasia. Resea rch Question 1: Generalization to Accuracy o f Productions It was proposed that through semi random practice of a broad range of phonemes and implementation of learned phonological princ iples through various oral motor speech tasks, MMiT might increase the accuracy of phoneme production. Indeed, improved accuracy of productions for two participants during the repetition of untreated, real word stimuli appears to provide initial support fo r treatment generalization. As these ability to accurately identify phoneme combinations in target words, and access oral motor plans and sequences necessary to accurately say a target word. Thus, it appears

PAGE 42

42 that following MMiT, participants may apply skills targeted during activities with pseudowords (phoneme combinations with no semantic meaning) to real word stimuli as well. It appears that b y targeting only those phoneme s which are consistently in error prior to treatment, response generalization to untrained sounds remains limited (Wambaugh et al., 1998 Raymer et al., 2002 ) despite benefits for the production of phoneme exemplars (words that contain the target phoneme) In turn these findings support an approach which targets a broad range of phonemes through a variety of novel phoneme combinations, thereby allowing participants to integrate skills for blending and segmenting phonemes to untreated words for phonemes w hich may or may not be consistently in error prior to treatment While the present investigation did not control for level of exposure or continued practice outside of treatment with the specific real words used during the Repeated Trials task, these resul ts offer preliminary support for the generalization of MMiT to real word repetition Follow up studies may examine the ecological validity of the repetition task, and the possible relationship to additional measures of real world communication. Research Q u estion 2: Error Inconsistency a cross Repeated Productions Beyond measures of overall improved accuracy, this project sought to examine the influence of MMiT on error inconsistency in the total number of phonemic errors produced across repetitions. Indeed, it appears that MMiT may reduce inconsistency across repeated productions of the same utterance Improvements in these participants indicate that following MMiT, successive repetitions within an item are produced more similarly amongst each other than at pre treatment. This may suggest that once individuals have selected an appropriate motor program, they are better able to

PAGE 43

43 maintain the selected motor sequence across repeate d productions of the same word. Despite a similar direction of change, the particip ant with the smallest relative treatment effect for reduced error inconsistency also presented with a high initial severity of apraxia of speech and aphasia a nd consequently, relatively limited gains during MMiT. Additionally, improvements in error incons istency may have been obscured by the far greater number of errors present at bo th pre and post treatment in this participant, compared to others in the sample. In turn, gains in the production of certain target w ords may have been largely obscured by ong oing phonemic errors within other words in the target series. Resea rch Question 3: Consistency of Error L ocation Consisten cy w ithin Session s Measures of consistency in error location were calculated to determine t he predictability of errors for specific p honemes within target words This measure was apraxia of speech (McNeil et al., 1995). Following MMiT, gains were observed for increased consistency of error location in a ll participants. This uniform direction of change appears to suggest that while participants are generally more accurate in their productions and produce fewer phonemic errors overall, the explicit and broad phoneme and word level training in MMiT (includi ng a problem solving approach to phonological pro cessing, reading and spelling) may have trained the participants in a structured and systematic approach to word production overall. In turn, phonemic errors at post treatment appear to be more predictable f or certain speech sounds across repeated, oral productions of words following intervention

PAGE 44

44 Interestingly, results from this analysis appeared to diverge somewhat from previous work by McNeil et al. (1995). In this prior investigation, four speakers with location, with a small range (86 94%). In contrast to these findings, results from the present investigation demonstrated average consistency in error location for participants of about 53.3% prior to treatment, with a fairly large range (40 77%). Following MMiT, while results were more consistent in error location for all three participants, they still presented with a lower average consistency of about 72.7% and range of 60 86% Divergent finding s this former study, and the potential influence of comorbid language and motor speech deficits in the participants examined in the present investigation. Therefore, w hile th is study was not able to replicate previous findings, it instead provides preliminary evidence of pattern s of error consistency for participants with apraxia of speech and aphasia. Further, it provides evidence of change in this error pattern following int ervention. Cons istency a cross Sessions It has been suggested that the presence of consistent phonemic errors across time may reflect specific oral motor programming deficits for certain phonemes (Mauszycki et al., 2010). However, it appeared that a relativ ely small proportion of phonemic errors were present within a given word at both time points in this study This appeared to reflect both a decrease in the total number of phonemic errors at post treatment (few pre treatment errors remained for specific ph onemes following intervention overall ) and also the unexpected

PAGE 45

45 multiple productions of the same target word at post treatment (consistent errors of production for phonemes produced correctly prior to intervention) Inconsistency in error location across sessions in this case may also reflect the intervening role of MMiT Through extensive oral awareness training and methods of guided questioning during treatment partici pants are routinely instructed to consider the way their mouth is positioned while they produce each phoneme, and thoughtfully plan the necessary sequence of movements before an attempt is made. This may alter their approach to real word combinations of ph onemes, and may evidence some degree of overgeneralization of the treatment approach ( Raymer et al., 2002; Wambaugh et al., 1999) In effect, while MMiT trains participants to apply skills for self monitoring and correction to phonemes which are produced i n error, participants may apply this approach to the production of all phonemes, whether or not oral motor programs for those phonemes are already well established. T his study does not have data to determine the exact nature of errors that appear ed for pho nemes which were produc ed correctly prior to treatment. However, it does suggest that the intervention may have resolved some specific motor programming deficits for the majority of phonemes (as evidenced by fewer errors overall at post treatment) Researc h Q uestion 4: Treatment Progress and Improved C onsistency To determine the degree of improved consistency based on progress through treatment, hours spent in each level of treatment and highest level of training achieved for pseudoword stimuli was determin ed for each participant. Interestingly, w hile gains were observed for increased consistency of error location in all participants following treatment the participant who demonstrated the least treatment progress also showed

PAGE 46

46 the greatest consistency in er ror location at both pre and post treatment. This observed error pat tern is particularly surprising in light of previous findings from Mloch et al. (1982), who found that consistency in the misarticulation of certain words, but not for specific phonemes, was associated with greater severity of apraxia of speech. However, findings must be considered in light of concurrent improvements in overall level of phonemic accuracy. In turn, a high degree of consistency in error location for a participant with substa ntially more errors might highlight the presence of errors at post treatment and the potential role of a more systematic approach to word production in less impaired individuals may instead reflect treatment benefit. Therefore, while increased consistency of error location for a small subset of remaining errors may reflect overall improved oral motor programming, those who progress slowly through treatment and continue to demonstrate a large number of errors may evidence ongoing deficits for motor programming and production. Research Question 5: Language Improvement and Improved Consistency The participant who was most impaired at pre treatment on measures o f aphasia se verity (as evidenced by comparatively lower WAB, LAC 3, and CTOPP scores) also demonstrate d the least treatment progress. On measures of accuracy and consistency across repeated productions, this participant demonstrated the greatest number of phonemic err ors at both time points, and the smallest relative treatment effect for decrease d error inconsistency across repetitions. These findings highlight the likely role of initial language impairments in the degree of treatment progress, as well as on measures o f accuracy and consistency of productions following treatment.

PAGE 47

47 Following MMiT, all participants improved on all three language measures. Concurrent improvements for all individuals on both the LAC 3 and CTOPP also demonstrate improved phonological processi ng at the level of both perception (identifying features and differences between phonemes) and production (with added demands of executing motor programs when producing words or pseudowords). These changes were seen along with increased consistency of err or location for all participants, and a decrease in total number of errors. Again, one participant did not show concurrent improvements in language and the total number of errors produced, which may have been due to near ceiling performance prior to treatm ent. In turn, improvements in both language function a nd consistency of repeated real word productions suggest that as language impairment begins to resolve, residual motor programming deficits may become more apparent, and therefore more easily targeted i n therapy. Likewise, if MMiT were modified for the treatment of apraxia of speech rathe r than phonological alexia, it might initially train improved phonological processing for all phonemes, as currently employed in MMiT for phonological alexia. Fol lowing these initial stages, it might then explicitly target oral motor production and practice for a broad variety of combinations for target phonemes. Extensive oral motor practice with a variety of exemplars (various phoneme combinations which include the tar get phoneme) during treatment has shown benefits for the production of phonemes trained specifically during therapy ( Wambaugh et al., 1998 ). Thus this added component of treatment may provide a range of oral motor practice for a broad range of phonemes in the context of many different surrounding phonemes.

PAGE 48

48 Influence of Comorbid Language and Motor Speech Disorders As noted earlier several additional language and motor speech disorders must be co nsidered in relation to changes seen for patterns of error con sistency following MMiT. O ral apraxia, or deficits in oral motor control for non speech oral movements is one motor speech disorder whi ch may influence treatment outcomes While this may not be expected to impact performance on measures of word repetition as evaluated by dependent measures ability to adhere to task demands, particularly during initial oral awareness training in MMiT, Stage I As this stage involves extensive oral ongue, pucker your lips, touch the roof of your mouth, etc.) difficulties in this area acquire the necessa ry skills for more advanced stages of treatment. Based on outcomes from the Oral Apraxia subtest of the ABA 2 at pre tr eatment evaluation however, it appears that none of the participants in this sample experienced deficit s that might impair performance during treatment (Table 1 2.) and at least two of the participants advanced through oral awareness training within the fi rst four to five hours of treatment. In turn, it is unlikely that oral apraxia contributed significantly to initial patterns of error consistency in number and location, as none of the participants demonstrated oral apraxia deficits at a measurable level. D ysarthria, a motor speech disorder due to neuromuscular weakness or incoordination may present more similarly to apraxia of speec h during performance on dependent measures (multiple repetitions of real words) Some mild dysarthria appeared evident for al l three participants, as evidenced by some observed deviation of the articulators at rest and some observed abnormalities during phonation. None of the participants in this sample endorsed subjective loss of sensation in or around the

PAGE 49

49 mouth, which presumab ly allowed for unhindered completion of Stage I Oral awareness training during MMiT. It should be noted that o ne participant did endorse experiencing some difficulty controlling the lip muscles and some reduced jaw strength while eating Additionally, this participant demonstrated the most severe apraxia of speech and language impairment at pre treatment, and fewer gains on measures of error consistency. However, low consistency of error location for this participant across sessions suggests that errors wer e not produced exclusively for specific phonemes As observed by McNeil et al. (1995), participants with dysarthria appear to show a high degree of consistency in error location across repeated productions of the same target word s In contrast the partici pants in the present study demonstrated a relatively low percentage of errors which were consistent for specific phonemes across both pre and post treatment time points. Thus, the pattern of errors observed for participants in the current study suggest t hat phonemic errors were likely not due exclusively to comorbid dysarthria. While this provides evidence for persistent deficits in the production of certain phonemes within a session for participants with dysarthria this prior study did not examine produ ction across repeated sessions. Therefore, the extent to which dysarthria contributed to change s in error consistency in the current sample appears minimal, but cannot definitively be ruled out at this time. C omorbid aphasia, or language impairment, may al so have contribute d to resulting patterns of speech error s Likewise, McNeil et al. (2008) have suggested that initial reports of error inconsistency in apraxia of speech may have resulted from the inclusion of individuals who actually demonstrate phonemic paraphasias. Phonemic paraphasias

PAGE 50

50 are productions which are largely intelligible, but may include various substitutions, omissions, and transpositions of phonemes which differ within the target word for lan, & Barresi 2001). While often perceptually similar to apraxia of speech, phonemic paraphasias may reflect language level disturbances in the selection of target phonemes versus deficits in motor programming for an intact phonological sequence ( Mauszyc ki et al., 2010 ). While this variability may arguably be attributed to comorbid aphasia diagnoses, it of speech. McNeil et al. (1995) reported consistency in error lo cation and type with little intra individual variability for specific phonemes across successive item repetitions. This was noted in contrast to observed inconsistency in error location across successive repetitions across four participants with conduction aphasia. However, while certain words may also be consistently in error across successive contextual readings in those there remains great variability in the specific errors produced within each word and word location (Mloch et al., 1982). The extent to which aphasic impairments may have contributed to observed error patterns was examined in this study using pre treatment evaluations for aphasia severity and classification. WAB scores in turn revealed classification of ano mic aphasia for all three participants. Largely intact ability on Re petition in this battery argues strongly against a diagnosis of conduction aphasia, as examined by McNeil and colleagues in relation hasics level versus sound level errors. None of the participants in this study demonstrated word level errors during the

PAGE 51

51 Repeated Trials task, providing further support for the presence of a predominant p honetic motoric impairment versus linguistic deficiency during pre and post treatment performance on the dependent measure (McNeil et al., 1995). As phonemic paraphasias may be features of a variety of aphasic syndromes, the possibility remains that these types of language level errors may have occurred to some extent in the present sample. speech is extremely rare (Duffy, 2005) and it is therefore difficult to generalize findings in these limited studies to most clinical populations. Therefore, while this project accounts for the nature of some presenting errors, overall it contributes to a growing understanding of typical patterns of error production in a clinically relevant sample and demonstrates the i mpact that MMiT may have on speech and language deficits Limitations For purposes of this preliminary investigation, broad phonetic transcription was employed to first determine the potential for change in error consistency following MMiT. Findings from t his level of error analysis provide preliminary support for post treatment improve ment in consistency of speech productions in this sample. However, narrow transcription may provide additional information on error type (specifically differences between err ors of substitution and distortion) and provide data on the variability of productions for specific phonemes within and between sessions. Generalization of findings is limited by the small sample size and limited number of words for measurement of word rep etition. Future work may consider selecting test stimuli based on the frequency of occurrence for various phonemes (Mines et al., 1978), and ensure a broad range of phoneme combinations for testing. Finally, findings are limited to isolated repetition of r eal word stimuli, and may not equate to performance variability in other

PAGE 52

52 contexts, such as reading or spontaneous speech. Despite study limitations, this investigation reveal s post treatment change s in error patterns on a measure typically used in clinical settings to guide treatment plans (ABA 2). Future Directions Together, these findings further reveal the inherent variability that may exist within and between individuals with apraxia of speech. This is evidenced by o bserved differences within and among participants in this study and the influence of individual participant qualities (i.e. treatment progress and initial severity of speech and language impairments) on treatment outcomes for dependent measures. Moreover, these findings demonstrate the necess ity and value of a single subject research designs in this population As highlighted by Mauszycki et al. (2010), group analysis may overshadow important differences that might exist between individuals in the total number and location of errors produced across repeated sampling sessions. In turn, future research may investigate error analysis at both group and individual levels to better characterize patterns of error inconsistency between and within speakers with apraxia of speech. In summary, these resu lts support subjective reports of improved verbal praxis following MMiT and suggest it may improve the consistency and/or accuracy of speech in adults with post stroke apraxia of speech. Additional participant and follow up data is needed to determine the influence of MMiT on additional features of apraxia of speech, maintenance of observed gains, and patient specific characteristics which might influence treatment outcomes. Lastly, if MMiT is implemented to solely treat apraxia of speech then several modi fications to the treatment approach seem relevant to consider. A modified version of the MMiT for apraxia of speech may involve the following treatment components:

PAGE 53

53 1) Broad, multimodal training for phonological processing and perception 2) Extensive, semi random oral motor practice with a broad range of phonemes 3) Semi random practic e of trained phonemes within a greater number of exemplars (phoneme combinations which incorporate the target phoneme for training) and various speech production contexts (repeated prod uctions of the same target word, words of increasing length, spontaneous and automatic speech, etc.) 4) Tracking treatment outcomes through additional measures of oral motor production and generalization

PAGE 54

54 APPENDIX: SUBJECTIVE RATINGS O F SENSORY AWARENESS 1. Are t here any areas of your face where you feel less sensitive to touch, temperature, or pain? 2. Are there any areas of your face where it is more difficult for you to control your muscles? 3. Do you find it hard to make certain facial movements? a. Smiling b. Laughing c. Sh owing someone you are angry d. Opening your eyes widely e. Puckering your lips 4. Do you have any difficulty biting, chewing, or otherwise moving your jaw? 5. Do you find it difficult to keep your teeth strongly clenched? 6. Is it hard for you to keep your jaws or lips t ogether when you are not eating or speaking? 7. Can you feel your tongue against the roof of your mouth? 8. Can you feel your tongue between your teeth? 9. Can you feel your tongue between your lips? 10. Can you feel your tongue against the inside of your cheek? 11. Do you have any difficulty drinking or swallowing? 12. Is it difficult for you to produce the volume of sound that you would like? 13. Do you find it difficult to project your voice?

PAGE 55

55 LIST OF REFERENCES Alexander, A W & Slinger Constant A (2004). Current Sta tus of T reatments for Dyslexia: Critical Review J ournal of Child Neurol ogy 19 744 758. Busk, P.L. & Serlin, R. (1992). Meta analysis for single case research. In T.R. Kratochwill & J.R. Levin (Eds.), Single case research design and analysis: New directio ns for psychology and education (pp. 187 212 ). Hillsdale, NJ: Lawrence Erlbaum Associates. Beeson, P.M. & Robey, R.R. (2006). Evaluating Single Subject Treatment Research: Lessons Learned from the Aphasia Literature. Neuropsychol ogical Rev iew 16 161 169. Cohen J. (1988). Statistical Power Analysis for the Behavioral Sciences ( 2nd ed.). Hillsdale, NJ: Lawrence Erlbaum Associates. Conway, T.W., Heilman, P., Gonzalez Rothi, L.J., Alexander, A.W., Adair, J., Crosson, B., & Heilman, K.M. (1998). Treatment of a case of phonological alexia with agraphia using The Auditory Discrimination in Depth (ADD) program. Journal of the International Neuropsychological Society, 4 608 620. Dabul, B. (2000). Apraxia battery for adults (2nd ed ) Austin, TX: PRO ED Inc. Darley F.L (1969 November ). The classification of output disturbances in neurogenic communication disorders Paper presented at the annual m eeting of the American Speech and Hearing Association Chicago IL. Deal, J.L., & Darley, F. L. (1972). The influence of ling uistic and situation variables on phonemic accuracy in apraxia of speech. Journal of Speech and Hearing Research 15 639 653. Duffy J. R. (1995). Motor Speech Disorders St. Louis MO : Elsevier Mosby Duffy, J. R. (2005). Motor speech disorders: Substrates, differential diagnosis, and management (2nd ed.). St. Louis, MO: Elsevier Mosby. Freed, D. (2000). Motor Speech Disorders: Diagnosis and Treatment San Diego, CA: Singular. Goodglass, H., Kaplan, E., & Barresi, B. (2001). Assessment of aphasia and related di sorders (3rd ed ) Philadelphia, PA: Loppincott Williams & Wilkins. Haley, K.L., Bays, G.L. & Ohde, R. N. (2001). Phonetic properties of aphasic apraxic speech: A modified narrow transcription anal ysis. Aphasiology, 15, 1125 1142. Itoh M, Sasanuma S, Ushi jima T. (1979). Velar movements during speech in a patient with apraxia of speech. Brain and Lang uage, 7, 227 239

PAGE 56

56 Johns, D.F. & Darley, F.L. (1970). Phonemic variability in apraxia of speech. J ournal of Speech and Hear ing Res earch 13 556 83. Kendall, D.L., Conway, T.W., Rosenbek, J.C., & Gonzalez Rothi, L.J. (2003). Case study Phonological rehabilitation of acquired phonologic alexia. Aphasiology 17, 1073 1095. Kendall, D.L., Rodriguez, A.D., Rosenbek, J.C., Conway, T.W., Gonzalez Rothi, L.J. (2006) Influence of intensive phonomotor rehabilitation on apraxia of speech. Journal of Rehabilitation Research & Development 43 409 418. Kendall, D.L., Rosenbek, J.C., Heilman, K.M., Conway, T.W., Klenberg, K., Gonzalez Rothi, L.J., & Nadeau SE. (2008). Ph oneme based Rehabi litation of Anomia in Aphasia. Brain and Language, 105, 1 17. Kertesz, A. (1982). The Western Aphasia Battery New York: Grune & Stratton. Knock, T.R., Ballard, K.J., Robin, D.A., & Schmidt, R.A. (2000). Influence of order of stimulus pre sentation on speech motor learning: A principled approach to treatment for apraxia of speech. Aphasiology, 14 653 668. Knollman Porter, K. (2008). Acquired Apraxiaof Speech: A Review. Topics of Stroke Rehabilitation 15 484 493. LaPoin te, L. L., & Horner, J. (1976). Repeated t rials of words by patients with neurogenic phonological selection sequencing impairment (apraxia of speech). Clinical Aphasiology 6 261 277. LaPointe, L.L., & Johns, D. F. (1975). Some phonemi c characteristics of apraxia of speech. J ournal of Communication Disorders 8 259 269. Lindamood, P.C, & Lindamood, P.D. (2004). The Lindamood Auditory Conceptualization Test (3rd ed ) Austin, TX: PRO ED Inc. Mauszycki, S.C., Dromey, C., & Wambaugh, J.L. (2007). Variability in apraxia of speech: A perceptual, acoustic and kinematic analysis of stop consonants. Journal of Medical Speech Language Pathology 15 223 242. Mauszycki, S.C., & Wambaugh, J. L. (2006). Perceptual analysis of consonant production in multisyllabic words in apraxia of speech : A comparison across repeated sampling times. Journal of Medical Speech Language Pathology 14 263 267. Mauszycki, S.C., Wambaugh, J.L., & Cameron, R.M. (2010). Variability in apraxia of speech: Perceptual analysis of monosyllabic word productions across repeated sampling times. Aphasiology 24 838 855.

PAGE 57

57 McNeil M.R., Doyle P.J., & Wambaugh J. (2000). Apraxia of speech: a treatable disorder of motor planning and programming. In: E. Nadeau, L.J. Gonzalez Rothi, B. Crosson. (Eds.). Aphasia and language: theo ry to pract ice (pp. 221 66) New York: The Guilford Press. McNeil, M.R., Odell, K., Miller, S. B., & Hunter, L. (1995). Consistency, variability, and target approximation for successive speech repetitions among apraxic, conduction aphasic, and ataxic dysar thria speakers. Clinical Aphasiology 23 39 55. McNeil, M. R., Robin, D. A., & Schmidt, R. A. (1997). Apraxia of speech: Definition, differ entiation, and treatment. In M. R. McNeil (Ed.), Clinical management of sensorimotor speech disorders ( pp. 311 344 ). New York: Thieme. McNeil, M.R., Robin, D.A., & Schmidt, R. A. (2008). Apraxia of speech: Definition, differentiation, and treatment. In M. R. McNeil (Ed.), Clinical management of sensorimotor speech disorders ( 2nd ed., pp. 249 268 ). New York: Thieme. Mines, M., Hanson, B., & Shoup, J. (1978). Frequency of occurrence of phonemes in Conversational English. Language and Speech 21 221 241. Mlcoch, A G. and Darley, F L. and Noll, J. D. (1982). Articulatory Consistency and Variability in Apraxia of Speech. In R.H Brookshire (Ed.), Clinical Aphasiology Conference Proceedings (pp. 235 238). Minneapolis, MN: BRK Odell, K., McNeil, M.R., Rosenbek, J. C., & Hunter, L. (1990). Perceptual characteristics of consonant production by apraxic speakers. Journal of Speech and Hearing Disorders 55 345 359. Ogar J., Slama H., Dronkers N., Amici, S., & Gorno Tempini, M.L. (2005). Apraxia of speech: an overview. Neurocase 11, 427 432. Ogar J, Willock S, Baldo J, & Wilkings D. (2006). Clinical and anatomical correlates of apraxia of speech. Brain and Lang uage 97 343 350. Peach, R.K. & Tonkovich, J.D. (2003). Phonemic characteristics of apraxia of speech resulting from subcortical hemorrhage Journal of Communication Disorders 37 77 90. Raymer, A.M., Haley, M.A., Kendall, D. (2 002) Over generalization in Treatment for Severe Apraxia of Speech: A Case Study. Journal of Medical Speech Pathology 10, 313 317. Schmidt, R.A., & Lee, T. D. (1999). Motor control and learning: A behavioral emphasis (3rd e d .) Champaign, IL: Human Kinetics

PAGE 58

58 Square P A Roy A.E., Martin R.E. (1997). Apraxia of speech: Another form of praxis disruption. In: LJG, Rothi, K.M., Heilman, (Eds.) Apraxia: The neuropsychology of action (pp. 173 206) East Sus sex: Psychology Press Wagner, R.K., Torgesen, J.K., A lexander, C. (1999). The Comprehensive Test of Phonological Processing Austin, TX: PRO ED Inc. Wambaugh, J.L., Martinez, A.L., McNeil, M. R & Rogers, M. A. (1999). Sound production treatment for apraxia of speech: ove rgeneralization and maintenance effects Aphasiology 13 821 837. Wambaugh, J.L., Duffy, J.R., McNeil, M.R., Robin, D. A., & Rogers, M. (2006). Treatment guidelines for acquired apraxia of spe ech: A synthesis and evaluation of the evidence. Journal of Medical Speech Language Pathology 14 15 3 3. Wambaugh J L Kaylinyak Fliszar M M West J E Doyle P J. (1998). Effects of treatment for sound errors in apraxia of speech. Journal of Speech Language Hearing Research 41, 725 743. Wambaugh, J. L., Nessler, C ., Bennett, J., & Mauszycki, S. C. (2 004). Variability in apraxia of speech: A perceptual and VOT analysis of stop consonants. Journal of Medical Speech Language Pathology 12 221 227. Wertz, R.T., LaPointe, L.L., & Rosenbek, J. C. (1984). A praxia of speech in adults: The disorder and its man agement Orlando, FL: Grune & Stratton. West C, Bowen A, Hesketh A, Vail A. (2008). Interventions for motor apraxia following stroke. Cochrane Da tabase of Systematic Reviews 1.

PAGE 59

59 BIOGRAPHICAL SKETCH Dana Szeles graduat ed with h onors from the Universit y of Connecticut in 2007 with a B.S. in c ognitive n euroscience. Her senior t hesis was conducted in the Behavioral Neuroscience lab at UConn under the direction of Dr. Roslyn Holly Fitch, and explored recovery from auditory processing deficits in an animal model of hypoxia i schemia. Following graduation, she worked for two years as a clinical research assistant at the Olin Neuropsychiatry Research Center in Hartford, Connecticut, studying fMRI of major d epr essive d isorder in adolescents and adults. Dana is now a neuropsychology graduate student in the Department of Clinical and Health Psychology at the University of Florida, studying under the direction of Dr. Tim Conway in the Brain Imaging Rehabilitation and Cognition lab. She is interested in the progres sion and course of recovery for individuals with acquired language and motor speech disorders S h e i s currently w o r k i n g on a project through the Brain Rehabilitation Research Center at the Ma lcom Randall VA Medical Center explorin g patterns of brain reorganization in phonological alexia using fMRI. T h r o u g h t h i s e x p e r i e n c e s h e h a s a l s o b e e n a b l e t o work closely with Dr. Stac y Harnish and other experts in speech and l an guage p athology exploring change in apraxia of speech following language intervention. Dana c o m p l e t e d her M.S in the spring of 2011 t h r o u g h t h e Department of Clinical and Health Psychology a t the University of Florida