|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
1 AN ELECTROPHYSIOLOGICAL INVESTIGATION OF THE NEURAL CORRELATES OF ATTENTION ASYMMETRIES FOLLOWING TRAUMATIC BRAIN INJURY By TANISHA G. HILL JARRETT A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2013
2 2013 Tanisha G. Hill Jarrett
3 Karleic for their unconditional love, motivation, a nd continuous support
4 ACKNOWLEDGMENTS I thank my mentor, Dr. William M. Perlstein, for his guidance expertise, and mentorship. I would also like to thank Christopher Sozda and Jason Gravano for their assistance with the project and encouragement. Fin ally, I wish to thank my committee, Dr. Vonetta Dotson, Dr. Stephen Boggs, and Dr. Christina McCrae for their commitment and constructive input. This research was supported by the American Psychological Association.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 ABSTRACT ................................ ................................ ................................ ..................... 9 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 11 Overview of the Problem ................................ ................................ ......................... 11 Attention in Traumatic Brain Injury ................................ ................................ .......... 12 Mechanisms of Visuospatial Attention ................................ ................................ .... 14 Alerting ................................ ................................ ................................ ............. 16 Orienting ................................ ................................ ................................ ........... 16 Executive Control ................................ ................................ ............................. 17 The Effects of Traumatic Brain Injury on the Attention Networks ............................ 18 Attentio n Laterality ................................ ................................ ................................ .. 19 Electrophysiological Correlates of Attention ................................ ............................ 22 Rationale for the Current Study and Specific Aims ................................ ................. 25 2 METHODS ................................ ................................ ................................ .............. 32 Participants ................................ ................................ ................................ ............. 32 Clinical and Neuropsychological Assessment ................................ ......................... 33 Cognitive Measures ................................ ................................ .......................... 33 Mood and Sleep Measures ................................ ................................ ............... 35 TBI Behavioral Symptom Measures ................................ ................................ 36 Materials and Procedure ................................ ................................ ......................... 36 EEG Acquisition and Reduction ................................ ................................ .............. 39 Statistical Analyses ................................ ................................ ................................ 40 Demographic and Neuropsychological Data Analyses ................................ ..... 41 Behavioral Data Analyses ................................ ................................ ................ 41 ERP Data Analyses ................................ ................................ .......................... 42 3 RESULTS ................................ ................................ ................................ ............... 53 Behavioral Data Analyses ................................ ................................ ....................... 53 RT Analyses ................................ ................................ ................................ ..... 53 Attention Network Effects ................................ ................................ ........... 53 Overall RT performance ................................ ................................ ............. 55 Error Rate Analyses ................................ ................................ ......................... 56 ERP Data Analyses ................................ ................................ ................................ 58
6 Cue locked Activity ................................ ................................ ........................... 58 N1 (Alerting ) ................................ ................................ ............................... 58 N1 (Orienting) ................................ ................................ ............................ 59 Target locked Activity ................................ ................................ ....................... 60 P3 (Executive Control) ................................ ................................ ............... 60 Neuropsychological Data and Functional Outcome Analyses ................................ 60 Relationships Between Attention Network Scores and TBI Severity I ndices .... 60 Relationships Between A ttention Network Scores and Emotion and Sleep M easures ................................ ................................ ................................ ...... 62 Relationships Between Attention Network Scores and Cognitive P erformance ................................ ................................ ................................ .. 62 4 DISCUSSION ................................ ................................ ................................ ......... 84 Overview of Results ................................ ................................ ................................ 84 Implications ................................ ................................ ................................ ............. 91 Possible Limitations and Future Directions ................................ ............................. 94 LIST OF REFERENCES ................................ ................................ ............................... 98 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 107
7 LIST OF TABLES Table page 2 1 Demographic comparison of control and TBI participants. ................................ 43 2 2 Inju ry characteristics of TBI participants. ................................ ............................ 44 2 3 Clinical and neuropsychological test battery by assessment domain. ................ 45 2 4 Group comparison of m ean and standard deviation scores on measures of mood and sleep function. ................................ ................................ ................... 47 2 5 Neuropsychological performance for control and TBI participants. ..................... 48 3 1 Mean attention ne twork scores for control and TBI participa nts as a function of visual field. ................................ ................................ ................................ ...... 65 3 2 REML ANOVA results of reaction time performance on LANT. .......................... 66 3 3 REML ANOVA results f or error rate (excluding non response trials) performance on LANT. ................................ ................................ ....................... 67 3 4 Number of accepted LANT trials for cue locked N1 comparisons by group. ...... 68 3 5 Number of accepted LANT trials for target locked P3 comparisons by group. ... 68 3 6 REML ANOVA results for N1 amplitude (V ) for orienting comparison. ............. 69 3 7 Hemispheric comparison of mean N1 amplitude (V) for cue locked activity by spatial cue type for TBI and control participants ................................ ............ 70 3 8 Cor relational relationship between hemispheric measures of attention networks and TBI injury severity indices and post injury ratings. ........................ 71 3 9 Correlational relationship between summary reaction time (RT) (correct trials only) and error rate ................................ ................................ ............................. 72 3 10 Correlational relationship between hemispheric measures of attention networks and emotional functioning measures for all participants (n =24). ......... 73 3 11 Correlational relationship between hemispheric measures of attention networks and cognitive performance by assessment domain for all participants (n=24) ................................ ................................ .............................. 74
8 LIST OF FIGURES Figure page 1 1 Anatomy and chemical modulators of the attention networks.. ........................... 30 1 2 Conc eptual model of the right lateralized organization of the orienting system of attention and the predicted effects of specific hemispheric damage ............ 31 2 1 Schematic of the Lateralized Attenti on Network Test (LANT) ........................... 50 2 2 Sensor layout and international 10 10 equivalencies of 64 channel geodesic sensor net.. ................................ ................................ ................................ ......... 51 2 3 Depiction of Lateralized Attention Network Test (LANT) network score calculations.. ................................ ................................ ................................ ....... 52 3 1 Control (n=12) and TBI (n=12) comparison of attention network scores ( SE) by visual field. ................................ ................................ ................................ .... 76 3 2 Mean ( SE) LANT reaction time as a function of cue type, flanker congruency, and target visual field for control and TBI participants.. .................. 77 3 3 LANT error rates (excluding non response trials) as a function of cue type, flanker congruency, and target visual field for control and TBI particpants.. ....... 78 3 4 Group wise means reflecting a significant Group x Cue type x Flanker congruency interaction from REML ANOVA on mean ( SE) error rate.. ........... 79 3 5 Neural comparsion of the a lerting network (no cue vs. double cue) N1 amplitude ( V) by group. ................................ ................................ .................... 80 3 6 Neural comparsion of the left and right orienting network (ce nter cue vs. left spatial cue, center cue vs. right spa tial cue) N1 amplitude ( V) for control and TBI participants. ................................ ................................ .......................... 81 3 7 Comparison of mean ( SE) N1 amplitude ( V) measured across hemispheres for left and right orienting conditions in contro l and TBI participants.. ................................ ................................ ................................ ....... 82 3 8 Neural comparsion of the executive control network (congruent target vs. incongruent target) P3 amplitude ( V) by group.. ................................ ............... 83
9 Abstract of Thesis Prese nted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science AN ELECTROPHYSIOLOGICAL INVESTIGATION OF THE NEURAL CORRELATES OF ATTENTION ASYMMETRIES FOLLOWI NG TRAUMATIC BRAIN INJURY By Tanisha G. Hill Jarrett May 2013 Chair: William M. Perlstein Major: Psychology Attention deficits are among the most commonly reported symptoms following traumatic brain injury (TBI) and are pervasive across all injury seve rity levels. While attention is postulated to be asymmetrically represented within the brain, little is known about the mechanisms underlying this purported imbalance and the efficiency of neural subsystems following disruption. The present study examined the impact of TBI on the hemispheric integrity of three interacting networks of attention alerting, orienting, and executive control. High density scalp recorded brain event related potentials (ERPs) were acquired while 12 moderate to severe TBI and 12 neurologically normal, demographically matched control participants performed the Lateralized Attention Network Test (LANT). Behaviorally, no group differences were found in reaction time benefit from temporal or spatial cueing; however, TBI survivors demo nstrated significant hemispheric differences for the orienting component of attention such that less benefit was derived from left spatial cues. Corresponding to behavioral findings ERP data revealed decreased N1 amplitude recorded over right parietal br ain regions in TBI participants during leftward shifts of attention, reflecting reduced efficiency of the left
10 orienting network. Finally, TBI participants had significantly greater difficulty utilizing spatial cues to overcome conflict created by incongru ent targets than controls, suggesting inefficient communication between the orienting and executive control subsystems. Findings indicate that right hemispheric N1 reduction may be a neural marker of alterations of the orienting system following TBI, and d ecreased effi ciency in communication between attentional networks may occur post injury. Discussion focuses on strategies to overcome orienting deficits and enhance executive functioning within the context of cognitive rehabilitation settings.
11 CHAPTER 1 INTRODUCTION Overview of the Problem Traumatic brain injury (TBI) is an alteration in brain function, or other evidence of brain pathology, caused by insult to the brain from an external mechanical force. It is the leading cause of death and disability i n young children, adolescents, and elderly over 65 years of age, with approximately 1.7 million individuals experiencing a TBI annually (Faul, Wald, & Coronado, 2010). Currently, an estimated 3.17 million TBI survivors live with disabilities in America (Br ain Injury Association of America, 2010). Consequentially, the debilitating nature of TBI impacts level of functioning and community reintegration for survivors (Willemse van Son, Ribbers, Hop, & Stam, 2009), as well as the quality of life for survivors an d caregivers (Arciniegas, Held, & Wagner, 2002). The tremendous heterogeneity of the cognitive and behavioral sequelae of TBI has taxed researchers and clinicians for years and made assessment and treatment efforts a challenge. One of the most devastating outcomes of TBI is the neurocognitive impairment that occurs as a result of the primary acceleration deceleration (i.e coup, contre coup) forces of injury as well as the secondary injury outcomes such as diffuse axonal shearing. It is possible that in s ome instances, TBI patients may be reluctant to seek treatment as they are under the false assumption that the physical limitations of TBI are substantially more debilitating than the cognitive symptoms (Olver, Ponsford, & Curran, 1996). Additionally, the more subtle aspects of cognitive impairment may not be readily apparent until the patient engages in cognitively demanding tasks (Park, Moscovitch, & Robertson, 1999; Stierwalt & Murray, 2002).
12 Although the pattern of neurocognitive impairment following TBI is variable, deficits in attention are amongst the most common post injury symptoms that exist regardless of the magnitude of injury (Auerbach, 1986; Gronwall, 1987; Whyte, Hart, Laborde, & Rosenthal, 2004). Despite the pervasive nature of attentional dysfunction in survivors of TBI, the characteriz ation of attentional impairment remains poor. This is largely attributable to the lack of consensus in the field regarding the multiple subsystems involved in the construct of attention and its hypothesized v arious dimensions (e.g., divided, sustained, selective) that are impacted following TBI. As a established that permits a brief, reliable, and valid measure of the cognitive constr uct of attention. Much research has been invested into understanding the behavioral and neural nature of TBI related attentional impairment. However, few studies take into account the function of laterality i.e., the differential representation of attenti on within each of the cerebral hemispheres. In many models of attention, lateralization remains unaccounted for, which hinders the accurate understanding of the mechanisms influencing post injury attentional functioning. Presently, there is a critical need for well controlled examinations of the hemispheric representation of attentional subsystems to guide the development of novel interventions and cognitive remediation programs that are tailored to the needs of the patient. Attention in Traumatic Brain In jury Attention is a foundational element of the cognitive architecture of the brain, with damage to the attentional system often impacting other cognitive systems. Attention is requisite for other cognitive processes that are vital to everyday functioning such as
13 memory, problem solving, language skills, information processing, and the cognitive control of behavior. As a result, attention dysfunction following TBI is thought to underlie difficulties across other higher order cognitive domains and may impact independent daily functioning as well as academic and vocational status (Brooks & McKinlay, 1987; Priganto, 2005). Thus, it is not uncommon for TBI survivors to experience difficulties in performing previously routine tasks of everyday life with impairment manifesting in various forms. TBI patients commonly present with wide ranging symptoms. Survivors may no longer possess the capacity to attend to a child while performing another cognitively demanding task such as cooking. Maintaining foc us to read for an extended period of time may become too laborious Additionally, TBI survivors may have difficulty focusing exclusively on the road while driving when there are distractions in the surrounding environment. Needless to say, poor attentional function decreases the overall quality of life for survivors of TBI and places them at increased risk for future accidents. Although we are currently unable to predict those who are at risk for attentional disturbance following TBI, specific prognostic f actors have been used to predict individu al cognitive recovery timelines. Current commonly used factors include age (Willemse van Son, Ribbers, Verhagen, & Stam, 2007), premorbid intelligence and educational level (Kesler, Adams, Blasey, & Bigler, 2002), e xtent of white matter damage ( Kinnunen e t al., 2011), and injury severity indices (Bishara, Partridge, Godfrey, & Knight, 1992). It is worth nothing that the nature of these relationships is complex and become s further convoluted when injury site and mecha nism of injury are taken into account.
14 Attention impairment is prevalent regardless of injury severity (Auerbach, 1986; Gronwall, 1987; Whyte et al., 2004). The leading reason that attentional impairment is a frequent occurrence following brain injury is due to the nature of biomechanical forces of injury that makes the brain structures which subse r ve attention especially vulnerable (Stierwalt & Murray, 2002). Primary coup contre coup injuries are produced by the slamming of the brain into the skull (coup) against the opposing surface of the skull (contre coup). Additionally, diffuse axonal injury (DAI), which occurs as a result of the rotational or bending motions of the brain against the skull leads to severe axonal shearing and is a common outcome of TBI (El Sayed, Mota, Fraternali, & Ortiz, 2008). DAI is thought to be a primary contributor to deficits in cognitive performance (Kinnunen et al., 2011; Niogi et al., 2008), and has been found to be a key predict or of functional outcome in terms of death and disability (Medana & Esiri, 2003). Moreover, extensive DAI related white matter damage may result in disconnection between brain regions that are critical for the function of attention Consequentially, this d isconnect may prevent inter regional neural communication and integration necessary for appropriate attentional performance. While attention difficulties do not account entirely for the disabilities commonly found in TBI survivors successful integration b ack into the community, work, and everyday life, in part, relies on the remediation and rehabilitation of attentional impairment. Mechanisms of Visuospatial Attention Visual attention shapes our experience of the environment by prioritizing information rel result, humans have developed the unique capacity to navigate the environment in a logical, goal oriented manner without becoming distracted by their surroundings.
15 Although attention was o nce thought of as a unitary construct, multi component attention models have become standard as they aid in a greater degree of diagnostic accuracy and predictability in clinical disorders where attentional processes go awry. One prominent model, developed by Posner and colleagues (Posner & Petersen, 1990; Posner & Fan, 2004), has conceptualized visuospatial attention as a system comprised of three independent, yet interactive, neural networks each involving specialized functional purpose which include aler ting, orienting, and executive control. Alerting is the generation and maintenance of a vigilant state that allows the processing of an upcoming visual stimulus. The orienting system is responsible for the disengagement, shifting, and reallocation of atten tion to a relevant spatial location. Finally, the executive control network resolves conflict between simultaneous, yet, mutually incompatible competing responses to stimuli and enables selective focus. By treating attention as an organ system, researchers have identified unique neuroanatomical substrates as well as cellular and neurochemical systems that are specific to each network as is shown in Figure 1 1. More recently, it has been found that these networks do not operate independently in all situation s, and that such interactions may be accounted for in terms of the nature of the task at hand (Fan, McCandliss, Sommer, Raz, & Posner, 2002). A computerized Attention Network Test (ANT; Fan et al., 2002) was designed as a measure of the efficiency of the t hree networks and has been widely researched in a variety of healthy and clinical populations since its inception. For the current study, we sought to investigate the construct of attention on the basis of this three system model of attention, with the aim of identifying specific subsystem(s) that are impacted following TBI.
16 Alerting Attention, in its most basic sense, requires the maintenance of a vigilant state. Without the recruitment of an arousal system, one cannot process relevant environmental infor mation. Traditionally, phasic alertness level has been modulated by providing a warning stimulus prior to the occurrence of a target stimulus. The warning detection and response to the pending task relevant stimulus (Petersen & Posner, 2012). This form of alertness is to be distinguished from tonic alertness, which is an internally driven self motivated state of arousal. By activating the phasic alerting system with t he presentation of a preparatory cue, one can readily anticipate the temporal course of the upcoming stimulus. In tasks requiring rapid response to a stimulus, knowing when the stimulus will appear commonly results in reaction time improvement ( i.e. faste r response s ). Functional brain imaging studies have implicated strong thalamic involvement as well as activity within regions of the inferior parietal cortex and the frontal lobe (Fan, McCandliss, Fossella, Flombaum, & Posner, 2005). Moreover, phasic activ ation of norepinephrine contacting neurons of the locus coeruleus, located in the brainstem, has been associated with improvement in sensory processing implicating norepinephrine as neuromodulator of the alerting subsystem of attention (Berridge & Waterhou se, 2003). Orienting stimulus and allocating focus to the selected location. The presentation of spatially relevant cues is thought to elicit the exogenous orienting sys tem and result in bottom up reflexive shifts of attention to the given location. This system is activated in instances
17 where the cue presentation is unexpected or salient (Corbetta & Shulman, 2002). Spatial cueing is thought to convey temporal as well as s patial information, resulting in faster reaction time response s than temporal cueing alone Corresponding cortical functional activation has been found in ventral cortical network which is comprise d of the ventral frontal cortex and the tem p oroparietal ju nction following the presentation of spatial cues (Corbetta & Shulman, 2002; Fan et al., 2005). cholinergic system is critical to the process of orienting. Scopolamine, a cholinergic antagonist, has been shown to impair orienting per formance in rhesus monkeys, highlighting the importance of this neurotransmitter system (Davidson, Cutrell, & Marrocco, 1999). Executive Control The process of coming to conscious awareness of a visual target is guided by the executive control component o f attention. It is thought that the executive control network is responsible for resolving conflict (i.e., overcoming a prepotent response tendency) that is created by competing stimuli, and facilitating the appropriate detection of a target. Eriksen flank er paradigms (Eriksen & Eriksen, 1974) are commonly used to elicit conflict by presenting trials consisting of a central arrow flanked by surrounding arrows that are either congruent (i.e., pointing in the same direction) or incongruent (i.e., pointing in the opposite direction), with the incongruent flankers thought to increase the degree of slower RT response to incongruent compared to congruent targets. In these instan ces of conflict, the executive control system functions as the top down regulator of attention. Anterior cingulate cortex (ACC) as well as dorsolateral prefrontal activity is typically engaged during tasks requiring the executive control system. The ACC is thought to
18 underlie cognitive processes related to conflict monitoring (Carter, Botvinick, & Cohen, 1999), with the dorsolateral prefrontal cortex (dlPFC) recruited in instances of conflict where there are competing responses (Botvinick, Braver, Barch, Ca rter, & Cohen 2001; Carter et al., 1999; MacDonald, Cohen, Stenger, & Carter, 2000). Interestingly, it has more recently been suggested that two executive control systems exist with similar anatomical substrates as aforementioned model; however, they are t hought to serve a different current work and is discussed at length elsewhere (see Dosenbach et al., 2008). Both the ACC and dlPFC receive projections from the ventral tegmental are a of the midbrain, a region of the brain rich with dopaminergic neurons. The Effects of Traumatic Brain Injury on the Attention Networks To date, only one study has investigated the effects of brain injury on the attentional networks (Halterman et al., 200 6). In their study, Halterman and colleagues tested mild TBI patients within two days of injury up until one month post injury using the originally published version of the ANT. TBI patients were found to have significantly reduced function of the orientin g and executive control networks, and spared alerting ability. Executive deficits were found to be longer las ting and remained for at least one month post injury. These results highlight the ability of the ANT to track recovery of attentional function in t he acute phase of injury. A vast range of paradigms are thought to probe similar mechanisms of attention and have been used with TBI patients. Varying degrees of injury and differing lesion sites make comparison of results across studies difficult. Howeve r, compromised executive function is a common outcome following TBI (Boelen, Spikman, Rietveld, & Fasotti, 2009). Findings regarding the alerting system remain mixed, and are likely
19 compounded by differing definitions of alerting (e.g., phasic vs. tonic al ertness). The literature on the orienting system is equally varied. Some studies have shown that, after controlling for generalized slowing in TBI patients, no differences exist in orienting ability (Bate, Mathias, & Crawford, 2001), and others have found differences in basic operations (e.g., disengage, move, engage) underlying the orienting function (Cremona Meteyard, Clark, Wright, & Geffen, 1992). Rios and associates (2004) posit that discrepancies in the literature surrounding TBI and attention may be attributable to the choice of study paradigm, operationalization of the construct of attention as well as the selection of variables for data analysis Not surprisingly the impact of TBI on attentional components is one that is still open to debate. Att ention Laterality The role of attention laterality has been extensively investigated in hemispatial neglect a neurological disorder whereby patients fail to perceive and respond to objects presented in the visual field contralateral to the damaged hemisphe re. While the neural mechanisms of the syndrome remain unknown, spatial neglect has been widely used as a model of cerebral lateralization of function (Corbetta & Shulman, 2011), with some proposing that specific components of the orienting system are affe cted. In particular, neglect has been hypothesized to result in an inability to disengage attention and reengage it to the contralateral space (Losier & Klein, 2001). It is well documented that the effects of left neglect resulting from damage to right pa rietal lesions are often more detrimental, longer lasting, and a more common occurrence than right neglect (Mesulam, 1999; Robertson, 1999). Based on this premise, it has been suggested that the left hemisphere allocates attention primarily to
20 the right vi sual field, whereas the right hemisphere is responsible for more evenly distributed shifts of attention to both fields of vision (Figure 1 2). In this regard, damage to the left hemisphere would result in the engagement of a compensatory mechanism with the right hemisphere controlling the orienting of attention to both visual fields. However, damage to the right hemisphere would leave no mechanism to control orienting to the left visual field (Heilman, 1995; Heilman & Watson 1977). Brain structures suppor ting the orienting function have shown correspondingly asymmetric functional activation patterns in functional neuroimaging studies (Corbetta, Miezin, Shulman, & Petersen, 1993), namely within ventral fronto parietal brain regions (Corbetta & Shulman, 2002 ). Interestingly, in healthy individuals, the extent of self reported attentional impairment is associated with degree of left orienting (right hemispheric) inefficiency (Poynter, Ingram, & Minor, 2010). Less is known about the impact of TBI on the organiz ation of the orienting system, but evidence for leftward orienting deficits have been found in TBI patients performing a sim ple spatial cueing task (Pavlov skaya, Groswasser, Keren, Mordvinov, & Hochstein, 2007). Collectively, findings have led to the commo n belief that the right hemisphere plays a more dominant role in the orienting function of attention. With regard to the alerting system, several imaging studies have revealed discrepant results, with some studies implicating right hemispheric involvement (Sturm & Willmes, 2001). It has also been suggested that the maintenance of an alert state is largely dependent on right hemispheric functioning given that the norepinephrine system has its strongest impact on the right posterior attentional regions (Posn er & Petersen, 1990). Others have found left cerebral hemispheric involvement, namely in
21 anterior intraparietal, inferior parietal, and frontal regions (Fan et al., 2005). These inconsistent results are thought to reflect differences in tonic versus phasi c alertness which are oftentimes referred to as separate cognitive constructs (Posner, 2008; Petersen & Posner, 2012). As mentioned above, tonic alertness is defined as an intrinsically sustained state of vigilance, whereas phasic alertness is stimulus dri ven and often evoked by an external cue. Nevertheless, recent work in healthy individuals shows similar hemispheric alerting capacity (Asanowicz, Marzecova, Jaskowski, & Wolski, 2012). Additionally, comparisons of phasic alertness in left and right brain d amaged patients indicate that the two groups are comparable on measures of alerting efficiency (Audet et al., 2000), which strongly suggests that alerting may be a function that is evenly distributed across cerebral hemispheres. The lesser studied of the networks in terms of lateralization, significant hemispheric differences in behavioral measures of executive control or functional activity patterns of structures that support its function have not been elaborated upon in great detail. Of the few studies t hat exist, findings are mixed, with some describing the executive control component of attention as a bilaterally distributed system comprised of anterior intraparietal, inferior parietal, and frontal regions (Fan et al., 2005), and others implicating righ t hemispheric involvement (Asanowicz et al., 2012). Lateralized Attention Network Test Given the lack of resolution regarding the lateralization of attention function, a Lateralized Attention Network Test (LANT) was developed by Green and associates (20 08) as an extension of the ANT. The LANT was designed to permit measurement of the integrity and efficiency of each subsystem of attention independently within each hemisphere. With the original intention to discern
22 the hemispheric competence of the health y brain, the use of the LANT and its variants have been extended to understanding bilingualism (Marzecova, Asanowicz, Kriva, & Wodniecka, 2012; Tao, Marzecova, Taft, Asanowicz, & Wodniecka, 2011), clinical disorders such as Attention Deficit/Hyperactivity Disorder (Konrad, Neufang, Hanisch, Fink, & Herpertz Dahlmann 2004), and the relationship between progesterone and hemispheric function (Schultheiss, Patalakh, & Rosch, 2012). In general, healthy individuals have demonstrated a right visual field orientin g RT advantage which is thought reflect the notion that rightward exogenous shifts of attention are mediated by both cerebral hemispheres (Figure 1 2) as opposed to the leftward shifts of attention that is solely controlled by the right hemisphere (Asanow icz et al., 2012; Du & Abrams, rightward bias in attention. Electrophysiological Correlates of Attention Electrophysiological recording of brain activity, through mea ns of non invasive scalp recorded electroencephalography (EEG), is especially advantageous in that it provides insights into the neural correlates of underlying sensory and cognitive processes. Often utilized in research settings, EEG is acquired while an individual is performing a computerized experimental task that is designed to elicit isolated cognitive or sensory processes. The superior temporal resolution of EEG that occurs within the timeframe of milliseconds allows for measurement of the slightest n eural changes that may otherwise go undetected by other neuroimaging methods. An event related potential (ERP) is an index of this neural process and is derived from EEG that is time locked to the occurrence of an event (e.g., presentation of a stimulus, r esponse to a stimulus). By averaging EEG over multiple experimental trials for the same time locked
23 event (e.g., stimulus presentation, response) extraneous EEG is reduced, and an event related potential is derived. The resulting ERP component is a wavefor m with polarity that is either positive going (i.e., positive amplitude), or negative going (i.e., negative amplitude). In addition to expressing the waveform in terms of amplitude, ERPs are also described on the basis of latency and scalp distribution (Du ncan, Summers, Perla, Coburn, Mirsky, 2011). ERP waveform latency has been shown to represent the time course of aspects of information processing (Naatanen, Gaillard, & Mantysalo, 1978), and the amplitude is thought to reflect the magnitude of neural reso urces devoted to a given cognitive process in response to a stimulus (Duncan et al., 2011; Duncan, Kosmidis, & Mirsky, 2005; Duncan Johnson & Donchin, 1977). Thus, by acquiring ERP data in the context of LANT performance, we can gain greater information re garding intact and impaired performance associated with different components of attention, and gain gross understanding of their neural bases. To our knowledge, no studies to date have conducted electrophysiological investigations utilizing the LANT. Howev er, the ANT has been successfully utilized in healthy participants to investigate ERPs associated with the attention networks, with results implicating two waveforms in the visual attention processing stream: N1 and P3 (Neuhaus et al., 2010 ). The visual N1 is an early negative going waveform occurring within the 150 200 millisecond (ms) timeframe and is an index attention allocation (Van Voorhis & Hillyard, 1977). In their study of the ANT, Neuhaus et al. observed increases in N1 amplitude during periods of increased phasic alertness that was modulated by the provision of a temporal cue. It has also been shown that the N1, as measured from posterior electrode
24 sites, is elicited in response to spatial cues that are predictive of target location (Ne uhaus et al ., 2010; Salmi, Rinn e, Degerman, & Alho, 2007).Thus, the degree of N1 amplitude enhancement in response to the presentation of temporal and spatial cues is thought to reflect the efficiency of the neural systems underlying the alerting and orienting compon ents of attention, respectively. Currently, no consensus exists regarding the impact of closed head TBI on the visual N1; however, many studies have found significant N1 amplitude reductions in TBI patients in response to the presentation of auditory stimu li (Duncan et al., 2005 ). The most commonly reported impact of TBI on ERPs is a reduction in P3 amplitude (Duncan et al., 2005; Lew, Lee, Pan, & Date, 2004). The P3 waveform is a later occurring positive deflection with onset at approximately 300 ms post s timulus presentation. Of particular interest to the current study is the P3b, which is thought to reflect processes associated with stimulus evaluation and categorization, independent of response selection (Picton, 1992; Polich & Kok, 1995). Previous elect rophysiological studies using the ANT have shown P3 amplitude modulations on the basis of flanker congruity. In particular, P3 amplitude decrements have been found for incongruent targets when compared to congruent This reduction in amplitude to incongrue nt flankers is thought reflect response inhibition that occurs as a result of conflict processing (Neuhaus et al., 2010). Thus, the magnitude of P3 reduction in response to incongruent targets may correspond to the efficiency of conflict processing that is mediated by the executive control system. Taken together, ERP studies suggest that N1 and P3 waveforms are sensitive to measuring the neural resources allocated to the performance of specific aspects of
25 attention as well as the timeframe in which the giv en process is executed. Additionally, the N1 appears to be a complementary neural index of the alerting and orienting networks, and the P3 an index corresponding to cognitive processes that underlie the executive control network. These specific ERP wavefor ms are of particular relevance to us given the potential for changes in latency and/or amplitude to serve as clinical markers of specific attention deficits following TBI. Rationale for the Current Study and Specific Aims The overarching belief motivating the current study is that attention is a construct comprised of functionally dissociable components that may be differentially impacted following traumatic brain injury. Attention difficulties are a common and devastating outcome of TBI and may manifest a s impairments across multiple domains of cognition reflected as difficulties in completing activities of everyday life. Despite the commonly that certain subsystem s of attention more are susceptible to insult, thereby impacting specific constituents of attention. Attention is thought to be a bilaterally distributed system that is asymmetrically represented within the cerebral hemispheres (Greene et al., 2008). In deed, much work has been done to understand the behavioral and neural impact of TBI on attention, but many models of these processes fail to account for such laterality. Furthermore, differences in study design and methodology have made multiple cross stud y comparison cumbersome. We aim to utilize the Lateralized Attention Network Test (LANT; Green et al., 2008) as a measure of the hemispheric integrity of three attentional components alerting, orienting, and executive controls in TBI patients. Our underst anding of the
26 lateralized organization of these systems may permit, with a greater degree of specificity, predictions about directionality of behavioral and neural deficits as well as the manner in which insult to the subsystems impact everyday cognitive f unctioning. Event related potentials may provide a particularly sensitive measure of the subtle changes in attentional components that are likely to occur as a consequence of TBI. As such, the current study is designed to test these assertions with the fol lowing specific aims: Aim 1: Behavioral Performance. Determine the behavioral impact of TBI on the hemispheric representation of attention, as reflected in the alerting, orienting, and executive control component networks. Hypothesis 1: We expect that TBI patients will demonstrate generalized slowing across all conditions, as evidenced by slower reaction times compared to control participants. Regarding the attentional network performance, previous work has shown that the orienting and executive components of visuospatial attention are most susceptible to behavioral performance impairments following TBI (Halterman et al., 2006). We therefore expect TBI to have its greatest impact on the orienting and executive control systems of attention. Regarding laterali zation of function, we predict that TBI survivors will show hemispheric differences in the efficiency of the orienting network. Alerting. As TBI has not been shown to significantly affect the alerting system, we expect the TBI group to demonstrate compara ble alerting effects to healthy controls, and for both groups to have proportionally similar alerting effects across both hemispheres.
27 Orienting. Given the commonly diffuse nature of injury, we predict that both cerebral hemispheres of the brain are likel y to experience damage to anatomical structures that support attention following TBI. However, based up on the known lateralized organization of the orienting system (Figure 1 2), we expect that behaviorally, TBI patients will demonstrate hemispheric redu ctions in the efficiency of the left orienting network when compared to the right orienting network. While the left hemisphere is likely equally impacted following TBI, the right hemisphere is expected to compensate given its ability to equally distribute attention to both visual fields (Figure 1 2). In accordance with these predictions, we expect TBI survivors to demonstrate left visual field/right hemispheric orienting decrements. This will be shown by significantly smaller attention network scores for me asures of left orienting relative to right orienting. Executive control. With regarding to the executive control network, both TBI longer RT s and larger error rate s to incon gruent relative to congruent targets. Given the large body of evidence suggesting marked executive dysfunction following TBI, TBI survivors are expected to show a disproportionally greater conflict effect than healthy, demographically matched comparison pa rticipants. No hemispheric differences for the executive control network are expected within either group. Aim 2: Neural Activity. Exploit the high temporal resolution of scalp recorded ERPs to temporally dissociate the neural activity related to the atte ntion networks and their potential deficits in TBI survivors. We are particularly interested in the impact of TBI on attention asymmetries as examined through cue related N1 (alerting and orienting) and target related P3 (executive control) ERP waveforms.
28 Hypothesis 2: Consistent with our behavioral hypotheses, we expect ERP waveform comparisons to reveal significant between group differences in the neural resources allocated to the orienting and executive control processes of attention. TBI is not expecte d to impact the neural efficiency of the alerting system. We also expect to see hemispheric discrepancies in the neural resources allocated to left versus right orienting within the TBI group only which would indicate reduced leftward orienting performanc e Alerting. For temporally alerting cues, TBI patients are predicted to show N1 amplitude increases of a similar magnitude as control participants, which would suggest that the allocation of resources for the alerting component of attention is comparable across participant groups. Orienting. Regarding the orienting network, we predict that TBI patients will show smaller N1 amplitudes for left spatial cues relative to right spatial cues, indicating reduced left orienting/right hemispheric efficiency As wa s suggested above, TBI may hemisphere and impair its ability to simultaneously control attention to the contralateral visual field given the global depletion of neural res ources that is likely to occur upon injury. Executive control. In accordance with our executive function behavioral hypothesis, both groups are expected to exhibit P3 amplitude attenuation for incongruent relative to congruent flankers. TBI patients are e xpected to have a disproportionately smaller P3 reduction following the presentation of incongruent
29 flankers, which would suggest relative impairment in conflict processing aspects of attention that are mediated by the executive control network. Aim 3: Neu ropsychological Performance and Functional Outcome. We are particularly sensitive to TBI related attentional impairment. Of particular interest is whether the degree of hemispheric specialty for the attentional subsystems in TBI patients is related to present day cognitive function and/or predicted by injury severity factors. Thus, in a series of exploratory analyses, we will examine the relationship between indices of injury severity (e.g., Glasgow Coma Scale (GCS), duration of loss of consciousness (LOC), and length of post traumatic amnesia (PTA), attention network indices, and present day clinical functioning as assessed by neuropsychological measures, mood question naires, and post injury behavioral ratings. Such findings may serve as a platform for the development of future hypotheses.
30 Figure 1 1. Anatomy and chemical modulators of the attention networks. Adapted from Posner & Petersen ( 2012 )
31 Figure 1 2. Conceptual model of the right lateralized organization of the orienting system of attention and the predicted effects of specific hemispheric damage. The right hemisphere is thought to shift attention evenly to both visual fields, whereas the left hemis phere shifts attention to the right visual field. On the basis of this belief, right hemispheric damage may be more detrimental and result in left visual field deficits given the lack of compensation by the left hemisphere.
32 CHAPTER 2 METHODS Participants Twelve moderate to severe TBI participants, and twelve demographically matched neurologically normal comparison participants between 18 and 55 years of age were recruited for the study. All participants were right handed, with the exception of one who ide ntified as ambidextrous. Participants of the TBI group sustained a non penetrating (i.e., closed head) TBI, as defined by Lezak, Howieson, and Loring (2004), more than six months and up to 10 years prior to study enrollment date. Injury severity was determ ined by the following acute indices: duration of loss of consciousness (LOC), duration of post traumatic amnesia (PTA), and Glasgow Coma Scale score (GCS). Moderate TBI was classified as either LOC lasting 30 minutes to 6 hours, PTA between 1 to 7 days, an d/or a GCS score between 9 to 12. Severe TBI was defined as either LOC greater than 6 hours, PTA greater than 7 days, and/or a GCS score between 3 to 8 (Bigler, 1990; Bond, 1986; Lezak, et al., 2004). In the event that injury severity criteria overlapped, the more severe injury criterion was used for classification. On the basis of these criteria, 2 participants were classified as having moderate TBI and 10 were classified as having severe TBI. As shown in Table 2 1, controls were well matched on age, educa tion, and gender. TBI participants were recruited through collaborations with Brooks Rehabilitation Hospital in Jacksonville, the Brain and Spinal Cord Injury Program of Florida, local brain injury support groups, the Florida Brain Injury Association, IRB approved flyers posted in the local Gainesville/Ocala region and the University of Florida, and public service radio announcements. Control participants were recruited by flyers from the University
33 of Florida and the local Gainesville community. Participa nts received course credit or monetary compensation ($10 per hour, up to $50) for participation in the study. All participants provided written informed consent in accordance with the guidelines established by the University of Florida Health Science Cente r Institutional Review Board. Exclusionary criteria for the study included history of major psychopathology, substance abuse, Attention Deficit/Hyperactivity Disorder history, learning disability, uncorrected visual impairment, color blindness as assessed by the Ishiara pseudo isochromatic color plates (Clark, 1924), substance use within two weeks of testing, current antiepileptic medication use, and/or sustained substance abuse within the past year. Individuals were also excluded if they were non native En glish speakers, below 18 or above 55 years of age, exhibited language comprehension deficits, or were presently involved in litigation. Additionally, for TBI participants, individuals with a history of mood or anxiety disorder predating injury, multiple TB Is, a history of penetrating head injury, or neurological disorder unrelated to the brain injury were also excluded. TBI patients with non recurrent post trauma seizures were not excluded from the study if they had not taken antiepileptic medication. Inju ry characteristics of the TBI patients are presented in Table 2 2. Clinical and Neuropsychological Assessment All assessment measures and brief s description of the measured domain s are presented in Table 2 3. Cognitive Measures Participants were adminis tered a brief neuropsychological battery designed to assess primary domains of cognition including: premorbid intelligence, attention,
34 working memory, processing speed, language and language related functions, memory, and executive function. The Rey 15 I tem Test (FIT; Rey, 1964) was used as a basic assessment of effort and memory malingering. Participants were administered the North American Adult Reading Test (NAART; Blair & Spreen, 1989), which is comprised of a 61 item word list, to provide an estimate of premorbid intellectual functioning. Additionally, the Boston Naming Test (BNT; Kaplan, Goodglass, & Weintraub, 2001) was used as a measure of visual confrontational naming ability. The BNT has been shown to reveal word finding difficulties in TBI patie nts of all injury severities (Hellawell, Taylor, & Pentland, 1999; Lezak, 1991). Additionally, the option of providing phonemic cues permits detailed understanding of whether naming impairment results from information retrieval difficulty as opposed to los s of previously stored information. Auditory attention and working memory capacity were assessed using the Digit Span Forward, Digit Span Backward, and Letter Number Sequencing subtests of the Wechsler Adult Intelligence Scale 3rd Edition (WAIS III; The Psychological Corporation, 1997, 2002). Logical Memory Tests I and II of Wechsler Memory Scale III (WMS III; The Psychological Corporation, 1997, 2002) were used as measures of auditory memory and learning. The short form of the Facial Recognition Test (FR T; Levin, Hamsher, & Benton, 1975), originally developed by Benton and colleagues (Benton, Sivan, Hamsher, Varney, & Spreen, 1994), was used to assess facial recognition and visuospatial ability. Given the strong visuospatial component, patients with right posterior lesions tend to perform most poorly on this task (Tranel, Vianna, Manzel, Damasio, & Grabowski, 2009; Warrington & James, 1967) relative to others with focal right hemispheric damage. Trail Making Tests A and B (Reitan, 1955) were
35 administered a s measures of visual scanning and processing speed, with Test B providing additional executive measures of mental flexibility and set shifting. The Wisconsin Card Sorting Task (WCST; Heaton, Chelune, Talley, Kay, & Curtis, 1993), was used to assess cogniti ve flexibility, problem solving ability, and response maintenance. Given the strong executive component required for success on this task, the WCST is sensitive to frontal lobe damage (Milner, B., 1963; Stuss et al 2000) with TBI patients often having pe rformance difficulty The Stroop Test (Golden, 1978) was also used as a measure of global executive function, namely inhibition, processing speed, and the ability to overcome prepotent response tendencies. The Controlled Oral Word Association Test (COWA; l etters F, A, S) and Category Fluency (Animals) tests (Benton & Hamsher, 1976) were administered as measures of verbal and semantic fluency, respectively. Mood and Sleep Measures All participants were initially administered the Mental Health Screening Form III (MHSF III; Carroll & McGinley, 2000) to screen for the presence of an anxiety or mood disorder, or the presence of other major psychopathology. Additionally, the Beck Depression Inventory II (BDI II; Beck, Steer, & Brown, 1996) was used as a measure o f depressive symptomatology and the State Trait Anxiety Inventory (STAI; Spielberger, Gorsuch, Lushene, Vagg, & Jacobs, 1983) assessed situational and dispositional anxiety. Apathy is a frequent outcome of traumatic brain injury that may occur in isolation but is most often comorbid with depression (Kant, Duffy, Pivovarnik, 1998). As such, the Apathy Evaluation Scale (AES; Marin, Biedrzycki, & Firinciogullari, 1991) was used, as it has been shown discriminate apathy from standard measures of depression and anxiety (Marin et al., 1991). The Pittsburgh Sleep Quality Index (PSQI; Buysse,
36 Reynolds, Monk, Berman, & Kupfer, 1989) was used to provide an objective measure of sleep quality and included measures of sleep duration, latency, arousals, and efficiency, a s well as subjective measures such as overall sleep quality. TBI Behavioral Symptom Measures To obtain a comprehensive understanding of current symptoms in light of injury as well as a measure of premorbid function, TBI participants completed additional s elf rating questionnaires (Table 2 2). If a significant other or family member was present during the time of the evaluation, they also completed a caregiver version of the form; however, c aregiver data were not included in the analyses given the low respo nse rate. TBI participants provided ratings on a modified version of the Neurobehavioral Rating Scale (NRS; Levin et al., 1987; Mathias Coats, 1999), which has been used extensively in TBI patients to assess emotional as well as physical symptoms and provi des the additional benefit of assessing for personality changes following injury. TBI participants also completed the Dysexecutive Questionnaire (DEX), which is a part of the Behavioral Assessment of the Dysexecutive Syndrome (BADS; Wilson, Alderman, Burge ss, Emslie, & Evans, 1996) and a 20 item measure of post injury change across the domains of cognition, emotion/personality, motivation, and behavior. Finally, t he Frontal Systems Behavior Inventory (FrSBe; Grace & Malloy, 2001), which is comprised of thre e subscales ( i.e., Apathy, Disinhibit i on, Executive Function) was used to identify and quantify behavior related to frontal lobe dysfunction and to provide an assessment of change relative to premorbid function. Materials and Procedure Upon successful c onsent and enrollment in the study, participants completed a clinical and neuropsychological assessment lasting approximately 90 minutes which
37 was developed to assess individual functioning across domains consisting of mood, sleep, cognition, and behaviora l symptoms. Group outcome and performance across the given m easures are summarized are Tables 2 4 and 2 5. Subsequently, participants performed 384 trials of a computerized version of the Lateralized Attention Network Te st (LAN T; Figure 2 1; Greene et al 2008) while high density, scalp recorded electroencephalogram (EEG) was acquired. The LANT was completed as part of three part experimental protocol, and lasted approximately 30 minutes. Cognitive Task: Lateralized Attention Network Test Developed by Gre ene and colleagues (2008), the LANT is a modified version of the original Attention Network Test (ANT; Fan et al., 2002) and was designed as a brief computerized measure of the efficiency of the attentional networks independently within each hemisphere. Th e task is a variation of 1) a Posner exogenous cueing paradigm (1980), where visual cues are presented to the potential spatial location of a shortly followed target stimulus to facilitate the shift of attention in the corresponding direction, and 2) an Er iksen & (i.e., incongruent) elements that have been shown to require additional conflict processing for appropriate target selection. In the LANT, participants were ini tially shown a warning cue (Figure 2 1a) that is meant to provide information about a pending target stimulus (i.e., a centrally presented arrow). Participants were presented with one of four possible cueing conditions consisting of no cue (no information provided), a center cue (temporally informative, warning participant of pending target stimulus), or a left or right spatial cue (temporally and spatially informative, indicating timing and location of target stimulus). Following the
38 cue presentation, part icipants were then shown the target stimulus, a centrally presented arrow. Targets were presented to one visual hemifield so that the effects of attention laterality are easily discerned. For each trial, the central target stimulus was flanked by either co ngruent arrows (i.e., arrows pointing in the same direction) or incongruent arrows (i.e., arrows pointing in the opposite direction; Figure 2 1b). The presentation of incongruent flankers is thought to elicit a greater degree of conflict given the non corr espondence in the direction of the central arrow relative to its surrounding arrows. This incongruence of stimulus features has been shown to require additional cognitive resources to process the central target (Eriksen & Eriksen, 1974). As shown in Figu re 2 1c, each LANT trial lasted approximately 4.5 seconds and consisted of 1) random variable pre cue fixation (400 1200 ms), 2) cue presentation (100 ms), 3) post cue fixation (150 ms), 4) target arrow appearance flanked by congruent or incongruent arrows ( self terminating after participant response or after 1700 ms elapsed), and 5) post target fixation (850 3350 ms). Participants were instructed to fixate on the central arrow and to respond as quickly and accurately as possible They were exposed to 4 cue types (no cue, center cue, double cue, left spatial cue, right spatial cue), 2 flanker congruency types (congruent, incongruent), 2 target directions (up, down), 2 target visual fields (left, right), and 2 repetitions of this series, yielding 384 total tr ials that were divided into 6 experimental blocks of 64 pseudo randomized trials each. A practice block of 24 trials was given, during which participants received feedback on their accuracy and speed of performance. Stimuli were presented using E Prime so ftware (v.1.0, Psychology Software
39 were made using the middle and index fingers to indicate the direction of the target arrow with a computer mouse, which was rotated 90 deg rees for the buttons to correspond to the upward and downward directions, respectively. To minimize the influence of extraneous variables, the hand that participants initially used to respond was counterbalanced, and subsequently alternated between blocks. EEG Acquisition and Reduction Electroencephalogram (EEG) was recorded from 64 scalp sites using a geodesic sensor net and Electrical Geodesics, Inc., (EGI; Eugene, Oregon) amplifier system (20K gain, nominal bandpass = .10 100 Hz). Electrode sites of int erest were reconfigured from the 64 channel geodesic sensor net to conform to international 10 10 positions (Figure 2 2). Electrode placements enabled recording of vertical and horizontal eye movements that are reflected in electro oculographic (EOG) activ ity. While recording, EEG was referenced to Cz and digitized continuously at 250 Hz with a 16 bit analog to digital converter. A right p osterior electrode served as common ground, suggested by the manufacturer. Upon EEG acquisition, data were mathematically re referenced to an average reference where the electrical activity of each electrode site was reflected as the difference between the electrode of interest and the average of all other recording sites. EEG data were prepared for analysis using Brain Electrical Source Analysis software (BESA version 5.1; Scherg, 1990) and adjusted for motion, electromyographic muscle artifact, electro ocular eye movement, and blink artifacts using BESA derived algorithms. EEG activity was excluded from the dataset using threshold criteria that maximized the number of trials accept ed for each participant. The average voltage
40 cue locked data, and did not differ significantly between TB I and control participants, cue locked t(22) = 0.20, p = .84, target locked t(11) = 1.00, p = .34. Point to point transitions were not Individual subject event related potentials (ERPs) were separately extracted and averaged toge ther from the continuous EEG recording in discrete temporal windows coinciding with each stimulus onset for cue locked (center, double, left spatial, right spatial, and no cue), and target locked (congruent, incongruent) conditions. Epoch duration of cue l ocked epochs lasted1600 ms and target locked epochs were 1400 ms. All averaged ERP epochs were digitally filtered at 30 Hz low pass and baseline corrected using respective pre stimulus windows. The waveform scoring windows for each group were conducted by m ean amplitude and peak latency values for cue locked N1 were measured bilaterally over posterior parietal electrode sites (10 10 international equivalents = PO7, PO8, P9, P10) between 161 191 ms and 177 207 ms windows for control and TBI participants, re spectively. Target locked mean amplitude and peak latency data for the P3 waveform were measured over central posterior electrode sites (10 10 international equivalents = CPz, P1, P2, Pz) between 342 362 ms for healthy controls and 350 370 ms for TBI parti cipants. Statistical Analyses Statistical analyses were conducted in JMP (v.9.0.2, SAS Institute Inc., Cary, NC) and SPSS (v.20, IBM Corp., Armonk, NY). Statistical significance levels were set at tailed tests. Significant interaction eff ects were decomposed using post
41 standard deviations for group and/or condition related effects. Demographic and Neuropsychological Data Analyses Chi square analyses wer e conducted to examine categorical demo graphic variables (e.g., gender ) between groups. Independent sample t tests were used to compare group performance on neuropsychological variables and continuous demographic variables (e.g., age, education). Pearson p roduct correlation coefficients were used in an exploratory manner to investigate the relationships between LANT behavioral performance, neuropsychological test performance, and post injury functional outcome. Behavioral Data Analyses Median correct trial RT (Ratcliff, 1993) and error rates excluding non response conditions (e.g., errors of omission) were analyzed separately using 2 Group (control, TBI) x 4 Cue type (no, center, double, spatial) x 2 Flanker congruency (congruent, incongruent) x 2 Target vi sual field (left, right) mixed model restricted maximum likelihood analyses of variance (REML ANOVAs). Attention network scores were calculated according to the procedure suggested by Fan et al. (2002) using the following median (correct trial) RT subtra ctions: alerting effect = no cue RT minus center cue RT; right orienting effect = center cue RT minus right spatial cue RT; left orienting effect = center cue RT minus left spatial cue RT; executive control/conflict effect = incongruent RT minus congruent RT. A diagram of the calculations is shown in Figure 2 3. Attention network effects were analyzed independently within each network as 2 way (Group x Visual field) mixed model REML ANOVAs.
42 ERP Data Analyses Consiste nt with work done by Neuhaus et al (2 010), attention network ERPs were analyzing according to the following comparisons: alerting effect = no cue amplitude vs. double cue amplitude; left orienting effect = center cue amplitude vs. left spatial cue amplitude; right orienting effect = center c ue amplitude vs. right spatial cue amplitude; executive control/conflict effect = incongruent amplitude vs. congruent amplitude. Regarding statistical analysis of the cue locked ERP amplitude activity for the alerting network, REML ANOVAs included the fol lowing factors: group (control, TBI) and cue type (no cue, double cue). Factors for the orienting network cue locked analysis included: group, cue type (center cue, right spatial cue, left spatial cue), and channel laterality (left hemisphere, right hemisp here). Notably, because we were interested in discerning hemispheric differences specific to the orienting network and hypothesized that there would be differential neural activity within each hemisphere depending on the cue type, both spatial cues (left and right) as well as EEG channel laterality (i.e., left hemisphere channel vs. right hemisphere channel) were entered as factors into the given analysis. For the executive control target locked analysis, factors included group and target flanker congruenc y (congruent, incongruent). Selection of e lectrode sites for cue locked (10 10 international equivalents = PO7, PO8, P9, P10 ) and target locked (10 10 international equivalents = CPz, P1, P2, Pz) EEG data were based on previous findings of posterior N1 mod ulations on the basis of cue type, and parietal P3 modulations on the basis of flanker congruency (Neuhaus et al., 2010). The selected electrode sites are displayed in Figure 2 2.
43 Table 2 1. Demographic comparison of control and TBI participants. Contro ls (n=12) TBI (n=12) Analysis M SD M SD t/X 2 p Age (years) 25.08 9.76 28.67 9.46 0.91 0.37 0.37 Education level (years) 15.58 1.83 14.58 1.88 1.32 0.20 0.54 Male/Female (n) 6/6 7/5 0.17 0.68 -Mother education (years) 15.92 2.19 15.0 8 3.75 0.66 0.51 0.27 Father education (years) 16.17 2.89 15.75 2.38 0.39 0.70 0.16 Average parental education (years) 16.04 2.24 15.42 2.99 0.58 0.57 0.23
44 Table 2 2. Injury characteristics of TBI (n=12) participants. M SD Range Initial Glasgo w Coma Scale Score 6.56 3.84 3 13 Loss of consciousness (days) 34.33 23.93 1 84 Post traumatic amnesia (days) 54.75 63.09 3 168 Time since injury (months) 59.67 24.51 6 89 NRS (self rated) Total Score 66.50 17.40 -DEX (self rated) Total Score 20.08 10.88 -FrSBe Total Score (premorbid) 86.08 17.91 -FrSBe Total Score (post injury) 92.92 20.02 -FrSBe Apathy Subscale (premorbid) 24.00 5.53 -FrSBe Apathy Subscale (post injury) 28.08 8.61 -FrSBe Disinhibition Subscale (premor bid) 27.25 7.26 -FrSBe Disinhibition Subscale (post injury) 28.83 7.83 -FrSBe Executive Function Subscale (premorbid) 34.83 7.27 -FrSBe Executive Function Subscale (post injury) 36.08 8.71 -NRS = Neurobehavioral Rating Scale; DEX = Dysexe cutive Questionnaire; FrSBe = Frontal Systems Behavior Scale (self rated)
45 Table 2 3. Clinical and neuropsychological test battery by assessment domain. Measure Domain Assessed Cognitive Performance North American Adult Reading Test (NAART) Premorbid in telligence REY 15 Item Test (FIT) Effort, memory Boston Naming Test Confrontational naming Digit Span Forward Test (WAIS III) Working memory Digit Span Backward Test (WAIS III), Working memory, mental flexibility Wechsler Memory Scale III (WMS II I) Logical Memory Test I & II Auditory memory & learning Facial Recognition Test (FRT) Recognition of unfamiliar faces, visuospatial ability Letter Number Sequencing (WAIS III) Auditory attention, working memory Trail Making Test A & B (TMT) Processi ng speed, visual tracking, set shifting, mental flexibility Wisconsin Card Sorting Task (WCST) Complex problem solving, cognitive flexibility, response maintenance Stroop Color Word Test Inhibition of automatic processes, executive function Controlled Oral Word Association Test (COWA) FAS & Animals Verbal (letter & semantic) fluency Psychological/Emotional & Sleep Symptoms Mental Health Screening Form III (MHSF III) General screener for major psychopathology or substance abuse/dependence B eck Depression Inventory 2 nd edition (BDI II) Depressive symptomatology Marin Apathy Index (MAI) Cognitive, emotional, and behavioral symptoms of apathy State Trait Anxiety Inventory (STAI) Situational and dispositional anxiety Pittsburgh Sleep Qua lity Index (PSQI) Sleep quality and disturbances over the past month Mental Health Screening Form III (MHSF III) General screener for major psychopathology or substance abuse/dependence Marin Apathy Index (MAI) Cognitive, emotional, and behavioral s ymptoms of apathy
46 Table 2 3 : Continued Measure Domain Assessed Post TBI Behavioral Function Frontal Systems Behavioral Scale (FrSBe) Self & significant other ratings Quantifies behavior related to frontal lobe dysfunction and provides an assessme nt of change relative to premorbid function Neurobehavioral Rating Scale (NRS) Self & significant other ratings Emotional and physical symptoms resulting from injury; assessment of post injury personality change Dysexecutive Questionnaire (DEX) Self rating Post injury change in cognition, emotion/personality, motivation, and behavior Frontal Systems Behavioral Scale (FrSBe) Self & significant other ratings Total score: Quantifies behavior related to frontal lobe dysfunction and provides an assess ment of change relative to premorbid function Apathy subscale: Measures initiation, psychomotor retardation, drive, persistence, energy, concern about self care Disinhibition subscale: Quantifies degree of impulsiveness, hyperactivity, and conformatio n to social convention Executive control subscale: Measures skills related to organization, planning, sequencing, problem solving, insight, and self monitoring of ongoing behavior
47 Table 2 4. Group comparison of mean and standard deviation (SD) scores on measures of mood and sleep function. Controls (n=12) TBI (n=12) Analysis M SD M SD t p BDI II Score 3.58 2.50 8.67 8.98 1.89 0.08 0.77 AES Score 8.75 4.16 11.42 5.09 1.41 0.17 0.57 STAI State Score 27.92 5.73 34.25 11.25 1.73 0.10 0.71 STAI Trait Score* 28.75 6.18 37.58 12.77 2.16 0.05 0.88 PSQI Score* 3.58 1.62 5.58 2.81 2.14 0.05 0.87 *p<.05 BDI II= Beck Depression Inventory 2 nd Edition; AES = Apathy Evaluation Scale; STAI = State Trait Anxiety Inventory; PSQI = Pittsburgh Sleep Quality Index
48 Table 2 5. Neuropsychological performance for control and TBI participants. Controls (n=12) TBI (n=12) Analysis M SD M SD t p Effort FIT 14.83 0.58 14.67 0.89 0.55 0.59 0.21 Premorbid Intelligence NAART FSIQ Estimate 110.08 6.16 109.42 6.49 0.26 0.80 0.10 Attention/Working Memory Digit Span (SS) 10.75 2.96 10.67 1.61 0.09 0.93 0.03 Letter Number Sequencing (SS) 12.42 2.61 10.42 3.18 1.69 0.11 0.69 Processing Speed Trails A Time (T s core)* 56.17 10.71 42.08 17.61 2.37 0.03 0.97 Stroop Color Word Test: Word Reading (T score)* 48.25 6.61 39.83 8.09 2.79 0.01 1.14 Stroop Color Word Test: Color Naming (T score)* 49.83 8.96 41.25 7.47 2.55 0.02 1.04 Language & Language Related COWA Letter fluency FAS (T score) 51.17 9.84 43.17 12.32 1.76 0.09 0.72 Semantic fluency Animals 48.67 8.65 42.83 12.76 1.31 0.20 0.54 Boston Naming Test (T score) 46.25 11.90 38.58 5.23 2.04 0.06 0.83
49 Table 2 5 Continued Controls (n=12) TB I (n=12) Analysis M SD M SD t p Visuoperceptual Ability Facial Recognition Test** 48.58 3.26 44.25 3.60 3.09 0.01 1.26 Memory Logical Memory I Immediate (SS) 12.08 2.50 10.67 3.20 1.21 0.24 0.49 Logical Memory II Delayed (SS ) 12.67 1.61 10.5 3.55 1.92 0.07 0.79 Logical Memory Recognition 26.83 1.90 26.17 3.41 0.59 0.56 0.24 Executive Function Trails B Time (T score)* 52.50 11.48 42.58 11.09 2.15 0.04 0.88 Stroop Color Word Test: Color Naming (T score)* 54.00 11.4 3 51.33 10.77 0.59 0.56 0.24 WCST (Total errors) 12.25 9.66 15.92 19.47 0.58 0.57 0.24 WCST (Perseverative errors) 5.67 2.81 9.08 9.85 1.16 0.27 0.47 WCST (# of failure to maintain set errors) 0.42 1.16 0.08 0.29 0.96 0.35 0.40 *p<.05, **p<.01 ; T sc ore (Mean = 50, SD = 10); SS = Scaled Score (Mean = 10, SD = 3). FIT = Rey 15 Item Test; NAART = North American Adult Reading Test; FSIQ = Full Scale Intelligence Quotient; COWA = Controlled Oral Word Association; WCST = Wisconsin Card Sorting Test.
50 Fi gure 2 1. Schematic of the Lateralized Attention Network Test (LANT) as modified by Greene and associates (2008) A) The four cue c onditions presented in the LANT B) Central target arrows (circled) flanked by congruent and incongruent arrows, respective ly C) Example left spatial cue trial of the LANT. Participant is shown a pre cue fixation, left spatial cue, pre target fixation, target arrow flanked by congruent arrows, and post target fixation.
51 Figure 2 2. Sensor layout and international 10 10 e quivalencies of 64 channel geodesic sensor net. Electrodes circled with solid line represent central posterior sites CPz, P1, P2, and Pz averaged for P3 waveform (executive control networks). Electrodes circled with dashed line represent posterior parieta l sites PO7, PO8, P9, and P10 averaged for N1 waveform (alerting and orienting networks).
52 Figure 2 3. Depiction of Lateralized Attention Network Test (LANT) network score calculations. Network values were derived using median reaction times (RT) for trials with correct response. A ) Alert ing = No cue RT Double cue RT B) Right orienting = Center cue RT Right spatial cue RT Left orienting = Center cue RT Left spatial cue RT C) Conflict/Executive control = Incongruent target RT Congruent target RT A B C
53 CHAPTER 3 RESULTS Behavioral Data Analyses Reaction times (RT) and error rates (excluding errors of omission) for the LANT were negatively correlated for both healthy controls, r (10) = 0.45, and TBI participants, r (10) = 0.12. While the direction of this relationship indicated that faster responses were more prone to error in general ( i.e. speed accuracy trade off), the relationship was not significant in either group ( p s > .10), indicating that a speed accuracy trade o ff was not overly influential in the following results. RT Analyses Attention Network Effects Each of the three hemispheric attention network effects were examined separately using 2 way (Group x Visual field) REML ANOVAs on RT difference scores for each network. Table 3 1 and Figure 3 1 show mean attention network scores for each group as a function of target visual field. Alerting Network. The alerting network was calculated as the no cue minus double cue RT difference, the latter of which contains the added feature of combined temporal and non specific spatial information. This difference provides an index of the saving in RT associated with knowing when the target arrow would appear. As predicted, analysis of the alerting network revealed no significa nt main effects of Group, F (1, 22) = 0.54, p = 0.47, or Target visual field, F (1, 70) = 0.01, p = 0.91. The Group x Target visual field interaction, F (1, 70) = 1.31, p = 0.26 was similarly not significant,
54 indicating that the groups did not significantly differ in hemispheric measures of the alerting component. Orienting Network. The orienting network was calculated by subtracting spatial cue RT from center cue RT, and reflects the RT benefit afforded by provision of spatial cues (left or right) compared to the center cue. Because both cues provide temporal as well as spatial information regarding the target, the difference in RT between these two conditions probes the benefit in RT associated with knowing when and where the target arrow would appear. A nalyses of orienting network scores revealed a main effect of Target visual field, F (1, 70) = 5.08, p < .03, with all participants demonstrating a right visual field bias (i.e., significant RT benefit from spatial cues presented to the right visual field). No significant effects involving Group, F (1, 22) = 0.05, p = 0.82, or Group x Target visual field, F (1, 70) = 1.59, p = 0.21 were found. Though the Group x Target visual field interaction was not significant, post hoc contrasts were conducted to examine w ithin group hemispheric difference s nevertheless, as we hypothesized that TBI patients would show smaller orienting effects for leftward shifts of attention. These contrasts showed that control participants did not differ in their orienting network scores as a function of visual field, t (11) = 0.84, p = .42, but TBI patients had a significantly smaller orienting effect for left compared to right visual field targets, t (11) = 3.06, p < .01. Executive Control Network. The executive control network is calcul ated on the basis of RT difference between incongruent versus congruent flanked targets. This RT difference score reflects the ability to ignore the surrounding distractor arrows and respond to the central target arrow, with the expectation that incongrue nt targets elicit a
55 greater degree of conflict and result in longer RTs in comparison to congruent targets. Results of this analysis revealed no significant main effects of Group, F (1, 22) = 1.84, p = 0.19, or Target visual field, F (1, 166) = 2.25, p = 0.1 4; the Group x Target visual field interaction was similarly not significant, F (1, 166) = 0.93, p = 0.34. Summary of Attentio n Network Behavioral Findings. Overall, TBI patients did not differ from controls on any of the attention network scores; however, TBI patients demonstrated hemispheric alterations that were specific to the orienting network of attention. Both groups showed greater right orienting efficiency, which is consistent with the right hemispheric bias that has been found in healthy individua ls. As predicted, TBI participants showed significantly reduced orienting performance that was specific to the left visual field, whereas controls did not demonstrate these hemispheric discrepancies for the orienting component of attention. Overall RT perf ormance To examine the factors influencing RT performance, a 2 Group x 4 Cue type x 2 Flanker congruency x 2 Target visual field REML ANOVA was performed on median correct trial RTs. Mean RTs as a function of group, cue type, flanker type, and target visu al field are shown in Figure 3 2. Results of the ANOVA are reported in Table 3 2 Statistical analyses revealed a significant main effect of Cue type, F (3, 330) = 36.22, p < 0.0001, with both groups demonstrating reaction time benefit as cues became more s patially informative (no cue RT > center cue RT > double cue RT > spatial cue RT). A main effect of Flanker congruency was observed, F (1, 330) = 501.60, p presence of inco ngruent flankers relative to congruent flankers) in both TBI and control participants. Additionally, a Cue type x Flanker congruency interaction, F (3, 330) = 2.90,
56 p < 0.04, was found, with greater RT benefit observed for congruent targets following the pr esentation of center cues relative to double cues. The opposite pattern (i.e., greater RT benefit for double vs. center cues) was observed for incongruent targets. Analyses also revealed a main effect of Group F (1, 22) = 9.18, p < 0.007, with TBI participa nts responding significantly slower than controls, which supports our generalized slowing hypothesis. Finally, a Group x Flanker congruency interaction was found, F (1, 330) = 8.57, p < 0.004, with TBI participants responding disproportionately slower to t argets with incongruent flankers relative to congruent flankers, suggesting that the TBI group had greater difficulty resolving conflict. Notably, no RT differences were observed for Target visual field, F (1, 330) = 0.01, p = 0.92. Moreover, performance for each visual field did not vary by group, Group x Target visual field, F (1, 330) = 0.50, p = 0.48. Error Rate Analyses A 2 Group x 4 Cue type x 2 Flanker congruency x 2 Target visual field REML ANOVA on mean error rates (excluding non response trials ) was also conducted. A summary of LANT error rates as a function of group, cue type, flanker congruency, and target can be found in Figure 3 3. ANOVA effects are reported in Table 3 3. Results of the analyses revealed a significant main effect of Cue type F (3, 330) = 4.14, p < 0.007, with greater error rate reduction for spatially informative cues relative to temporal (i.e., center and double) cues. Paralleling RT outcome, a main effect of Flanker congruency, F (1, 330) = 63.74, p < 0.0001, revealed that b oth groups committed significantly more errors on conditions requiring a greater degree of conflict processing (i.e., incongruent vs. congruent flankers). A Cue type x Flanker congruency
57 interaction, F (3, 330) = 5.87, p < 0.0007, indicated that performance accuracy for congruent flanker trials were comparable irrespective of the cue type; however, for incongruent flanker trials, greater error rates were observed following the presentation of center and double cues. A main effect of Group was also found, F (1 22) = 6.31, p < 0.02, indicating unexpectedly that, overall, healthy controls committed significantly more errors than TBI participants. A Group x Cue type interaction was found, F (3, 330) = 3.41, p < 0.02. As temporal and spatial cues became more inform ative, control participants correspondingly reduced error rate, whereas TBI patients did not show this pattern as they were unable to benefit from double cues to the same degree. A Group x Cue type x Flanker congruency interaction, F (3, 330) = 2.73, p < 0. 05, revealed that in both groups, the temporally informative (i.e., center and double) cues were not successful in reducing the conflict created by the incongruent flanker. Interestingly, for control participants only, the spatial cue was successful in re ducing conflict, as evidenced by comparable error rates for congruent and incongruent targets. This three way interaction is depicted in Figure 3 4. Simple effects post hoc contrasts were conducted for both groups to compare the mean error between congruen t and incongruent targets for spatial cue trials only. Analyses showed that the error rate discrepancy between congruent and incongruent targets was not significant for the control group, suggesting a reduction in the conflict effect following the presenta tion of spatial cues, F (1, 330) = 0.51, p = 0.48. However, the TBI group was not able to utilize spatially informative cues to overcome conflict, as evidenced by the significant difference that remained between error rate for congruent and incongruent targ ets, F (1, 330) = 4.60, p < 0.04. Finally, a Group x Target
58 visual field interaction approached significance, F (1, 330) = 3.82, p = 0.05, with controls committing significantly more errors for left visual field targets compared to TBI participants. ERP Dat a Analyses Cue locked Activity We first examined if control and patient groups differed in the number of trials included in calculating ERP waveforms as a function of each relevant condition involved in the alerting and orienting effects to show that the two groups had an equal signal to noise ratio for detecting potential condition related differences. The mean number of trials contained per group for each condition is shown in Table 3 4. Control and TBI groups did not significantly differ in the number o f trials for calculating per subject ERP averages on any of the relevant conditions, t p s > .23. Thus, ERP subject averages across the two groups for the conditions relevant to calculating alerting and orienting effects did not differ. N1 (Aler ting: No cue vs. Double cue) The Group x Cue type REML ANOVA on N1 amplitude yielded a significant effect of Cue type, F (1, 162) = 57.51, p < 0.0001, with a larger N1 amplitude for no cue conditions relative to double cue conditions, suggesting that both groups benefitted from the presentation of temporal cueing and yielded an alerting effect. Results also revealed a Group x Cue type interaction, F (1, 162) = 4.73, p < 0.04, with both groups demonstrating comparable N1 amplitude for no cues, but controls ha ving a greater N1 amplitude for the double cue, indicating a disproportionately larger alerting effect for the control participants (Figure 3 5). No significant main effect of Group was found, F (1, 22) = 1.45, p = 0.24.
59 N1 (Orienting: Center cue vs. Left s patial cue & Center cue vs. Right spatial cue) Results of the Group x Cue type x Channel laterality REML ANOVA (Table 3 6) revealed a main effect of Cue type, F (2, 248) = 16.61, p < 0.0001, with greater N1 amplitude in both groups for center cues followe d by right and left spatial cues, respectively (Figure 3 6). A significant Cue type x Channel laterality interaction was found, F (2, 248) = 21.13, p < 0.0001. Follow up post hoc analysis revealed no amplitude difference between left and right hemispheric e lectrode sites for center cue conditions, F (1, 248) = 0.13, p = 0.71. However, as expected, N1 amplitude was greatest for recordings over left hemispheric electrode sites following the presentation of right spatial cues, and greatest over right hemisphere electrodes for left spatial cue conditions. Additionally, a Group x Cue type x Channel laterality interaction was found in support of our original hypothesis, F (2, 248) = 6.70, p < 0.002. As shown in Figure 3 7, controls demonstrated the expected pattern where N1 amplitude was greatest in hemispheric electrode sites contralateral to the location of the spatial cue, suggesting appropriate allocation of neural resources while engaging in the orienting aspect of attention. This N1 amplitude pattern was not up held in TBI patients. Post hoc contrasts were conducted to compare right hemispheric N1 amplitude for left versus right spatial cues for the TBI group. For right hemispheric electrode site recordings, one would expect a more negative N1amplitude for left s patial cues in comparison to right spatial cues. Interestingly, post hoc contrasts comparing spatial cues for right hemispheric recordings revealed that TBI patients did not show these amplitude differences by cue type. This outcome was in accordance with our predictions and strongly suggests that TBI patients exhibited neural alterations in orienting processes of attention that were
60 specific to the right hemisphere during leftward shifts of attention. Group comparison of mean cue locked N1 amplitudes by sp atial cue type is reported in Table 3 7. Target locked Activity We examined if control and TBI groups differed in the number of trials included in calculating ERP waveforms as a function of each relevant condition involved in the executive control network (i.e., congruent vs. incongruent targets), to ensure adequate signal to noise ratio in detecting the predicted effects. Table 3 5 shows the mean number of trials accepted per condition for each group. No significant between group differences were found in the number of trials used for calculating ERP subject averages on the given conditions, t (22) p s > .22., which indicates that the two groups did not differ in the number of trials utilized per subject in deriving ERP averages. P3 (Executive Control: Congruent vs. Incongruent) Analyses revealed that, in general, controls had significantly greater P3 amplitude across both congruent and incongruent flankers (Figure 3 8), resulting in a main effect of Group, F (1,22) = 5.14 p < 0.04. Notably, a neural reflection of the conflict effect (i.e., P3 reduction for incongruent relative to congruent ta rgets) was not found across groups, Flanker congruency F (1, 166) = 1.37, p = 0.24. Additionally the predicted group differences in neural resources allocated to conflict processing were not found, Group x Flanker congruency, F (1, 166) = 0.02, p = 0.90. Neu ropsychological Data and Functional Outcome Analyses Relationships Between Attention Network Scores and TBI Severity I ndices Pearson product correlations were conducted, for TBI patients only, in an exploratory manner to examine whether commonly used cli nical indices of TBI injury
61 severity as well as measures of current degree of emotional, physical, and behavioral dysfunction resulting from injury were related to LANT performance. TBI patients who rated themselves as having more severe neurobehavioral di sturbance following injury on the NRS were found to have greater difficulty resolving conflict (i.e., larger executive control network scores), irrespective of the hemisphere to which target stimuli were presented, left visual field, r (12 )= 0.61, p < 0.04; right visual field, r (12)= 0.63, p < 0.03. A positive association was found be tween the self rated Executive C ontrol subscale on the FrSBe and the executive control network for the left visual field, r (12)= 0.64, p < 0.03. In particular, TBI participants who reported having a greater degree of executive function impairment had more difficulty overcoming the conflict created by incongruent targets presented to the le ft visual field. A significant relationship was found between hospital time and alerting, wi th patients who remained hospitalized for a longer duration having poorer right visual field alerting performance, r (12) = 0.66, p< .03. Notably, none of the commonly used injury severity indices were related to our attention network measures (Table 3 8). Given such outcomes, we looked at the relationship between summary measures of performance (i.e., reaction time and error rate), irrespective of attentional capacity, and hospital indices of injury severity (e.g., Glasgow Coma Scale Score, duration of loss of consciousness, post traumatic amnesia). The number of days that TBI patients remained unconscious from injury was strongly related to reaction time to target stimuli on the LANT paradigm. Patients with a longer duration unconsciousness had slower react ion times to targets presented to the left, r (12)=0.81, p< 0.01, and right visual field, r (12) = 0.79, p < 0.05. Results also yielded a significant relationship between error rate
62 that was specific to right visual field targets and the duration of post trau matic amnesia, r (12) = 0.52, p < 0.01, with direction of the association unexpected. In general, TBI participants who experienced a greater length of amnesia for events following injury committed less commission errors on the LANT for target stimuli that were presented to the right visual field. Neither initial Glasgow Coma S cale score, duration of hospitalization following injury, nor time since injury was related to global error rate or reaction time measures on the LANT (Table 3 9). Relationships B et wee n Attention Network Scores and Emotion and Sleep M easures We were also interested in the nature of emotional and sleep function and attention network scores across both participant groups. There was a significant positive correlation between levels of dep ressive symptomatology on the BDI II and difficulty overcoming conflict (i.e., larger executive control network scores) for both left, r (24) = 0.65, p < 0.002, and right, r (24) = 0.68, p < 0.001, executive control network measures. Situational anxiety was positively associated with the left executive control network scores, r (24) = 0.47, p < 0.03, with those reporting higher levels of anxiety having greater conflict resolution difficulty for left visual field targets. Additionally, apathy was associated wit h increasing difficulty overcoming conflict. Neither trait anxiety nor sleep quality was related to hemispheric measures of attention network performance (Table 3 10). Relationships Between Attention Network Scores and Cognitive P erformance Scores from the neuropsychological battery measuring performance across cognitive domains of intelligence, attention/working memory, processing speed, language and language related function, visuoperceputal skill, memory, and executive
63 function were also correlated wi th LANT behavioral performance scores for all participants, with several salient relationships emerging (see Table 3 11 for an overview of correlational relationships), as described below With regard to alerting network performance, left visual field aler ting capacity was associated with better Digit Span Forward performance, r (24) = 0.50, p < 0.02. Left alerting was also positively associated with the difference score for Trails B versus Trails A, which is thought to isolate set shifting ability while acc ounting for cognitive processing speed, r (24) = 0.41, p < 0.05. Larger left visual field alerting scores were associated with more overall performance errors on the WCST. Unexpectedly, participants who performed well on a facial recognition task had poore r right visual field alerting, r (24) = 0.42, p < 0.05. Investigation of behavioral performance for the orienting network revealed that participants with higher left orienting scores had better recognition memory on WMS III Logical Memory, r (24) = 0.45, p < 0.03. Additionally a strong relationship between left visual field orienting proficiency and the WCST, a cognitive measure of process related to executive function (namely set shifting, planning, and inhibition), was found, with larger left orienting ca pacity associated with more categories completed on the WCST, r (24)= 0.63, p< 0.002, less overall errors, r (24) = 0.73, p < 0.001, less perseverative errors on the task, r (24) = 0.72, p < 0.001. Right orienting efficiency was also associated with less ove rall errors on the WCST, r (24)= 0.48, p< 0.02, and less perseverative errors, r (24)= 0.47, p< 0.03. Several counterintuitive relationships emerged. In particular, those who scored higher on Digit Span Forward demonstrated poorer orienting in response to le ft, r (24) = 0.42, p < 0.02, and right, r (24) = 0.48, p <
64 0.02, visual field targets. More efficient set shifting (Trails B time; Trails B A time) was associated with better left visual field orienting, r (24) = 0.43, p < 0.04; r (24) = 0.42, p < 0.05, re spectively. Interestingly, LANT executive control network scores for either hemisphere were not significantly related to paper and pencil measures of executive control; however, relationships were observed between the executive control network and tests t hat are thought to measure verbal fluency, visuoperceputal ability, working memory, and processing speed. In particular, higher Letter Number Sequencing scores were associated with better conflict resolution when targets were presented to the right visual field (i.e., lower right hemisphere executive control network scores), r (24) = 0.49, p < 0.02. With regard to processing speed, participants who were faster at naming colors and reading the names of colors on the Stroop Color Word Task were more skilled at right hemispheric conflict resolution, Stroop Color Naming, r (24) = 0.45, p < 0.03; Stroop Word Reading, r (24) = 0.46, p < 0.03. Within the domain of visuoperceptual ability, those who were better at identifying and matching pictures of unfamiliar fac es were also better at overcoming conflict, particular for items presented to the left visual field, r (24) = 0.44, p < 0.04. Finally, participants who were more skilled at letter fluency had better conflict resolution for left, r (24) = 0.56, p < 0.005 an d right, r (24) = 0.49, p < 0.02, visual field targets. Notably, our measures of premorbid intellectual functioning, category (animal) fluency, confrontational naming, immediate memory, and delayed memory were not significantly related to any of the LANT network scores.
65 Table 3 1. Mean attention network scores (difference in ms) for control and TBI participants as a function of visual field. Left visual field Right visual field Analysis M SD M SD t p d Control (n=12) Alerting effect 18.77 38.66 28.40 43.43 0.57 0.58 0.23 Orienting effect 37.94 39.24 45.17 33.17 0.84 0.42 0.20 Executive control effect 87.21 48.50 90.80 43.33 0.34 0.74 0.08 TBI (n=12) Alerting effect 36.02 43.92 24.21 54.01 0.82 0.43 0.24 Orienting effect 25.92 57.00 51.5 40.80 3.06* 0.01 0.52 Executive control effect 107.51 69.65 124.02 85.30 1.11 0.29 0.21 *p<.05
66 Table 3 2. REML ANOVA results of reaction time performance on LANT. p <0.05 ** p <0.01 *** p <0.005 **** p <0.001 df F Cue type 3, 330 36.22**** Flanker congruency 1, 33 0 501.60**** Target visual field 1, 330 0.92 Cue type x Flanker congruency 3, 330 2.90* Cue type x Target visual field 3, 330 1.69 Flanker congruency x Target visual field 1, 330 1.21 Cue type x Flanker congruency x Target visual field 3, 330 0.85 Gr oup 1, 22 9.18** Group x Cue type 3, 330 0.23 Group x Flanker congruency 1, 330 8.57*** Group x Target visual field 1, 330 0.50 Group x Cue type x Flanker congruency 3, 330 0.40 Group x Cue type Target visual field 3, 330 0.40 Group x Flanker congrue ncy x Target visual field 1, 330 0.50 Group x Cue type x Flanker congruency x Target visual field 3, 330 0.04
67 Table 3 3. REML ANOVA results for error rate (excluding non response trials) performance on L ANT. p <0.05 ** p <0.01 *** p <0.005 ** ** p <0.001 df F Cue type 3, 330 4.14** Flanker congruency 1, 330 63.74**** Target visual field 1, 330 0.47 Cue type x Flanker congruency 3, 330 5.87*** Cue type x Target visual field 3, 330 1.18 Flanker congruency x Target visual field 1, 330 0.88 Cue type x Flanker congruency x Target visual field 3, 330 1.50 Group 1, 22 6.31* Group x Cue type 3, 330 3.41* Group x Flanker congruency 1, 330 1.65 Group x Target visual field 1, 330 3.82 Group x Cue type x Flanker congrue ncy 3, 330 2.73* Group x Cue type Target visual field 3, 330 0.26 Group x Flanker congruency x Target visual field 1, 330 0.91 Group x Cue type x Flanker congruency x Target visual field 3, 330 0.89
68 Table 3 4. Number of accepted LANT trials for cue locked N1 comparisons by group. Control (n=12) TBI (n=12) Analysis M SD M SD t p No cue 60.08 19.66 68.25 15.78 1.12 0.27 Center cue 58.58 22.41 68.50 16.50 1.24 0.23 Double c ue 58.03 22.30 67.67 14.56 1.25 0.23 Left spatial cue 29.33 11.25 34.08 7.73 1.21 0.24 Right spatial cue 28.92 11.80 33.17 8.11 1.03 0.32 Table 3 5. Number of accepted LANT trials for target locked P3 comparisons by group. Control (n=12) TBI (n=12) Analysis M SD M SD t p Congruent 124.92 33.95 142.00 31.56 1.28 0.22 Incongruent 123.67 32.39 139.00 29.10 1.22 0.24
69 Table 3 6. REML ANOVA results for N1 amplitude (V) for orienting (center vs. left spatial cue vs. rig ht spatial cue) comparison. p <0.01 ** p <0.0001 df F Cue type 2, 248 16.61** Channel laterality 1, 249 0.39 Cue type x Channel laterality 2, 248 21.13** Group 1, 22 0.68 Group x Cue type 2, 248 2.88 Group x Channel laterality 1, 249 0.06 Group x Cue type x Channel laterality 2, 248 6.70*
70 Table 3 7. Hemispheric comparison of mean N1 amplitude (V) for cue locked activity by spatial cue type for TBI and control participants Left spatial cue Right spatial cue Analysis M SD M SD t p Control (n = 12) Left hemisphere 0.71 3.13 2.00 2.61 6.05* 0.00 1.04 Right hemisphere 1.46 2.99 0.80 1.43 2.45* 0.03 1.58 TBI (n = 12) Left hemisphere 0.10 1.10 1.08 2.52 2.08 0.06 0.47 Right hemisphere 0.21 1.37 0.04 1.4 7 0.32 0.76 0.12 p <.01
71 Table 3 8. Correlational relationship between hemispheric measures of attention networks and TBI injury severity indices and post injury ratings. LVF = left visual field; RVF = right visual field; DEX = Dysexecutive Questionnaire ; N RS = Neurobehavioral Rating Scale ; FrSBe = Frontal Systems Behavioral Scale p <.05 ** p <.01 Alerting Orienting Executive Control LVF RVF LVF RVF LVF RVF Initial Glasgow Coma Scale Score ( n =9) 0.45 0.44 0.52 0.26 0.32 0.01 Loss of consciousness (days) ( n =9) 0.35 0.09 0.48 0.52 0.11 0.21 Post traumatic amnesia (days) ( n =12) 0.05 0.02 0.20 0.11 0.10 0.03 Hospital time (days) ( n =11) 0.20 0.66* 0.28 0.07 0.32 0. 24 Time since injury (months ( n =12) 0.12 0.16 0.12 .10 0.11 0.13 DEX Self ( n =12) 0.30 0.29 0.12 0.23 0.24 0.25 NRS Self ( n =12) 0.12 0.22 0.06 0.05 0.61* 0.63* FrSBe Apathy Self (post injury) ( n =12) 0.24 0.22 0.03 0.19 0.53 0.48 FrSBe Disinhibition Self (post injury) ( n =12) 0.14 0.53 0.17 0.40 0.02 0.15 FrSBe Executive Control Self (post injury) ( n =12) 0.26 0.45 0.14 0.09 0.64* 0.41
72 Table 3 9. Correlational relationship between summary reaction time (RT) (correct trials only) and error rate (excluding non response trials) performance and TBI injury severity indices. p <.05 ** p <.01 Mean RT Mean Error Rate LVF RVF LVF RVF Initial Glasgow Coma Scale Score ( n =9) 0.33 0.33 0.06 0.19 Loss of consciousness (days) ( n =9) 0.81** 0.79* 0.12 0.19 Post traumatic amnesia (days) ( n =12) 0.34 0.29 0.5 2** 0.01 Hospital time (days) ( n =11) 0.29 0.40 0.05 0.43 Time since injury (months ( n =12) 0.21 0.28 0.16 0.40
73 Table 3 10. Correlational relationship between hemispheric measures of attention networks and emotional functi oning measures for all participants ( n =24). LVF = left visual field; RVF = right visual field p <.05 Alerting Orienting Executive Control LVF RVF LVF RVF LVF RVF BDI II Score 0.13 0.09 0.24 0.00 0.65** 0.68** AES Score 0.15 0.18 0.01 0.26 0.48* 0.46* STAI State Score 0.11 0.26 0.01 0.25 0.47* 0.3 4 STAI Trait Score 0.01 0.17 0.07 0.19 0.32 0.40 PSQI Score 0.04 0.00 0.14 0.08 0.28 0.23
74 Table 3 11. Correlational relationship between hemispheric measures of attention networks and cognitive performance by assessment domain for all participants (n=24). Alerting Orienting Executive Control LVF RVF LVF RVF LVF RVF Premorbid Intelligence NAART FSIQ 0.12 0.05 0.14 0.13 0.23 0.13 Attention/Working Memory Digit Span Forward total 0.50* 0.11 0.42* 0.48* 0.06 0.01 Digit Span Backward total 0.06 0.38 0.02 0.21 0.02 0.20 Letter Number Sequencing total 0.17 0.20 0.15 0.15 0.30 0.50* Processing Speed Trails A (time) 0.34 0.39 0.37 0.34 0.04 0.35 S troop Color Word Test (Word Reading) 0.03 0.25 0.12 0.22 0.28 0.46* Stroop Color Word Test (Color Naming) 0.19 0.13 0.11 0.29 0.26 0.45* Language & Language Related Letter fluency (FAS) 0.13 0.08 0.23 0.20 0.56** 0.49* Semantic flue ncy (Animals) 0.11 0.04 0.01 0.03 0.38 0.32 Boston Naming Test 0.24 0.13 0.19 0.02 0.15 0.30 Visuoperceptual Ability Facial Recognition Test 0.16 0.42 0.02 0.02 0.44* 0.40 Memory Logical Memory I Immediate 0.25 0.01 0.26 0.21 0.05 0.01 Logical Memory II Delayed 0.33 0.02 0.27 0.16 0.02 0.03 Logical Memory Recognition 0.13 0.21 0.45* 0.35 0.26 0.33
75 Table 3 11 Continued. p <.05 ** p <.01 Alerting Orienting Executive Control LVF RVF LVF RVF LVF RVF Executive Function Trai ls B (Time) 0.40 0.23 0.43* 0.39 0.09 0.32 Trails B A (Time) 0.41* 0.14 0.42* 0.39 0.15 0.16 WCST (# of categories completed) 0.40 0.08 0.63** 0.31 0.12 0.08 WCST (Total errors) 0.45* 0.12 0.73** 0.48* 0.05 0.12 WCST (Perseverative errors) 0 .40 0.14 0.72** 0.47* 0.18 0.30 WCST (# of failure to maintain set errors) 0.04 0.27 0.30 0.02 0.11 0.04
76 A B C Figure 3 1 Control (n=12) and TBI (n=12) comparison of attention network scores ( SE) by visual field. A) A lerting network B) Orienting network C ) Executive c ontrol n etwork VF = visual field.
77 A B Figure 3 2. Mean ( SE) LANT reaction time as a function of cue type, flanker congr uency, and target visual field A ) C ontrol participants B) TBI participants. LVF = left visual f ield ; RVF = right visual field.
78 A B Figure 3 3. LANT error rates (excluding non response trials) as a function of cue type, flanker congruency, and target visual field A) C ontrol participants. B ) TBI particpants LVF = left visual field; RVF = right visual field.
79 A B Figure 3 4. Group wise means reflecting a significant Group x Cue type x Flanker congruency interaction from REML ANOVA on mean ( SE) error rate. Both groups demonstrate increased conflict (i.e., greater error rate for inco ngruent relative to congruent targets) following alerting (i.e., center and double) cues. A) Control participants were able to utilize spatial cues to overcome conflict and reduce error rate for trials comp rised of incongruent targets, making error rate performance comparable to that of congruent targets. B) TBI participants were unabl e to use spatial cues to their advantage, and demonstrated a significant discrepancy between congruent and incongrent targets for spatial cue trials.
80 A B Fig ure 3 5. Neural comparsion of the alerting network (no cue vs. double cue) N1 amplitude ( V) by group. A ) Cue locked grand averaged event related potentials (ERPs) over posterior parietal sites (10 10 international electrode equivalents: PO7, PO8, P9, and P10 ). B) Mean ( SE) N1 amplitude ( V) for alerting network by group comparison
81 A B Figure 3 6. Neural comparsion of the left and right orienting network (center cue vs. left spatial cue, center cue vs. right spatial cue) N1 amplitude ( V) Cue locked grand averaged event related potentials (ERPs) over posterior parietal sites (10 10 international electrode equivalents: PO7, PO8, P9, and P10 ). A ) Control participants. B) TBI participants.
82 A B Figure 3 7. C omparison of mean ( SE) N1 amplitude ( V) measured across hemispheres for left and right orienting conditions A) Cont rol participants exhibit the predicted neural pattern. B) TBI participants demonstrate attenuation of right hemispheric N1 amplitude following the presentation of left spatial cues.
83 A B Figure 3 8. Neural comparsion of the executive control network (congruent target vs. incongruent target) P3 amplitude ( V) by group. A) Target locked grand averaged event related potentials (ERPs) over central posterior sites (10 10 international electrode equivalents: CPz, P1, P2, and Pz ). B) Mean ( SE) P3 amplitude ( V) for executive control network by group comparison.
84 CHAPTER 4 DISCUSSION Overview of Results Recent models have described attention as a system comprised of fu nctionally dissociable components, but have often failed to account for the lateralized organization of attentional processes across the cerebral hemispheres. We sought to investigate the impact of moderate to severe TBI on the hemispheric representation o f three attentional networks and the relationship of their efficiency to post injury behavioral, emotional, and cognitive functioning. Our prediction that TBI patients would respond significantly slower to target stimuli than controls was upheld, and is t hought to reflect a generalized slowing of information processing that commonly occurs following TBI (Spikman, Timmerman, van Zomeren, & Deelman, 1999; Zahn & Mirsky, 1999). Interestingly, none of our hemispheric measures of the attention networks were rel ated to injury severity indices; however, several relationships emerged between overall reaction time and error rate performance. In particular, TBI patients who remained unconscious for a longer period of time following injury had slower reaction times to stimuli presented to either visual field. We also unexpectedly found that those with a longer duration of post traumatic amnesia committed less overall errors on the LANT. The direction of this association is counterintuitive given that post traumatic amn esia is commonly considered the most reliable index of TBI patient outcome (Brooks, Aughton, Bond, Jones, & Rizvi, 1980). Our results may be compounded by the way that post traumatic amnesia was assessed, operationally defined, or reported (e.g., duration of retrograde amnesia vs. anterograde
85 amnesia vs. impaired awareness), which is a common assessment inconsistency in the field (Stuss et al., 1999). With regard to the alerting network, TBI patients performed comparably to controls for both behavioral m easures of alerting capacity within both hemispheres. A lack of hemispheric bias on behavioral measures of the alerting network for both groups supports the concept that phasic alerting is likely a component of attention that is distributed across both hem ispheres (Asanowicz, et a., 2012; Audet et al., 2000). N1 investigation of the alerting network revealed disproportionately larger alerting effects in healthy controls. These findings are consistent with our initial prediction and show that while healthy c ontrols demonstrate a greater magnitude of the alerting effect, TBI participants possess the ability to engage the phasic alerting system to benefit from the presentation of temporal cues. This outcome in our moderate to severe TBI sample is in accordance with previous behavioral measures of alerting that used the ANT to investigate attention in mild TBI (Halterman et al. 2006). Taken collectively, outcomes suggest that the mobilization of arousal that is guided by an external temporal stimulus remains rela tively unaffected in TBI patients regardless of the severity of injury. TBI patients with greater right visual field alerting spent less time in the hospital following injury. Additionally, participants with strong working memory performance (Digit Span F orward) had more efficient left visual field alerting. There were also several unexpected relationships, between the alerting network and paper and pencil measures of executive control ability that warrants additional study. In particular, those who commit ted more errors on the WCST and had poorer set shifting ability (Trails B A time) had better left visual field alerting performance. Interestingly, of the mood measures that
86 were administered in our study, no significant emotional correlates of the alertin g network were found. Furthermore, there was a lack of a relationship between alerting measures and injury severity indices and self rated measures of impairment in our sample of TBI patients. These findings, in conjunction with the behavioral and neural m easures of the alerting network, provide evidence in favor of spared alerting capacity following TBI. Our hypothesis, that the orienting network would be impacted following TBI and that the predominantly right lateralized organization of the orienting sy stem would influence the directionality of deficits, was largely supported across both behavioral and neural measures. On the basis of well established models (Heilman, 1995; Heilman & Watson 1977) which suggest that the right hemisphere shifts attention to both visual fields, we expected that the right hemisphere would compensate for damage to structures of the left hemisphere caused by TBI. In line with these beliefs, we predicted that TBI patients would thereby perform comparably to healthy controls on behavioral and neural measures of left hemispheric/right visual field orienting. This hypothesis was confirmed on the basis of hemispheric comparisons for both N1 amplitude and LANT behavioral performance. Our primary contention, that the lack of orienting compensation from damage to right hemispheric structures would manifest as a reduction in leftward orienting efficiency, was upheld upon direct comparison of left and right orienting network scores in our TBI sample. While a general right visual field bia s (i.e., larger right orienting score) was found across all participants, the orienting effects did not differ by hemisphere for healthy controls. While it is conceivable that our results for TBI patients are a reflection of the right visual field advantag e phenomena, TBI patients
87 demonstrated hemisphere discrepancies in the orienting network that were of a disproportionately greater magn itude than control participants which points toward a performance for survivors of TBI. In alignment with our behavioral findings, striking neural alterations were observed in the TBI group that was also specific to the orienting network of attention. ERP investigation revealed that, for TBI participants, N1 amplitude measures were significantly attenuated over right hemispheric regions during leftward shifts of attention. Corresponding to the reduced behavioral performance for leftward orienting that was found in TBI survivors, neural reflections of left o rienting decrements were found for right hemispheric regions. These findings are in support of work showing that the neural resources involved in the orienting function may be differentially allocated between the hemispheres (i.e., strongly right laterali zed) as the white matter connections of fronto parietal structures that support the orienting network are correspondingly right lateralized (de Schotten et al., 2011). This structural lateralization may result in a general de emphasis of functional activat ion of the left hemisphere as the right hemisphere is able to compensate for performance, therefore negating left hemispheric neural alterations. When taking the behavior findings into consideration, it is possible that right hemispheric N1 reduction for l eftward orienting may serve as a neural marker of the left orienting behavioral impairments found following TBI. With regard to the relationship between orienting ability and our neuropsychological measures, participants with more efficient left orientin g scores were found to have better recognition memory performance (Logical Memory Recognition).
88 Unexpectedly, better attention/working memory (Digit Span Forward) performance on paper and pencil tests was associated with poorer orienting across both hemisp heres. Less surprising, however, was the relationship between left orienting network scores and the neuropsychological measures of executive functioning, namely the WCST, that emerged in both groups. Those with better left orienting performance completed m ore categories on the WCST, committed less perseverative errors, and committed less overall errors. Additionally, those with more efficient set shifting ability (Trails B time, Trails B A time) also demonstrated better left visual field orienting performa nce. We expected to find bilateral impairment on the executive control network of attention in TBI patients. As predicted, hemispheric measures of executive control were equivalent within each group, and both groups demonstrated the expected conflict effe ct resulting in reaction time slowing to incongruent target stimuli. To our surprise, however, no differences were found in TBI patients relative to their counterparts on behavioral measures of the executive control network. P3 examination did not reflect the anticipated conflict effect. We also did not find P3 differences in the magnitude of conflict elicited by incongruent targets during stimulus processing. It is possible that selection of posterior parietal P3 electrode sites contributed to the null fin dings and that a more frontally congruity in individuals performing the ANT (Neuhaus et al., 2010). Of the mood and sleep measures administered, significant relationships emerged only for the executive control network. Amongst all participants, those who identified as having higher levels of depression had more difficulty with resolution of conflict fo r
89 targets that were presented in either visual field. This is consistent with the known functional integration of the affective and executive controls systems. Functional neuroimaging findings have shown ACC functional activity reductions during periods of emotional distraction. ACC activity reduction was also associated with poorer behavioral performance and indicated less efficient conflict resolution during periods of emotional arousal (Hart, Green, Casp, & Belger, 2010). Additionally, those with higher levels of apathy were found to have poorer bilateral executive network performance. Finally, higher levels of situational anxiety were associated with poorer left visual field conflict resolution. Oddly, our executive control network scores were not relat ed to paper and pencil measures of executive control function across all participants. In general, faster processing speed scores (Stroop Color Naming, Stroop Word Reading) and more efficient working memory (Letter Number Sequencing) was associated with be tter right hemispheric conflict resolution. Better letter (FAS) fluency was associated with better executive control network performance across both hemispheres. Finally, participants with better left visual field conflict resolution had greater visuoperce ptual ability with regard to facial recognition (Facial Recognition Test). More severe post injury executive control dysfunction (FrSBe) and frontal behavioral disturbance (NRS) were also significantly related to poorer executive control network performanc e. Unexpectedly, RT and error rate performance provided a more detailed understanding of the interaction between the attention networks following TBI. TBI participants responded disproportionately slower to incongruent flankers relative to congruent flank ers despite the lack of group differences on the executive control
90 network measure (which compares median RT). The disproportionate size of the conflict effect in TBI patients suggests that overcoming conflict created by competing stimuli in may become mor e challenging following TBI. The nature of this conflict resolution impairment, and its influence on other attentional subsystems, was further detangled upon examination of error rate performance. We found an alerting executive control network interaction such that alerting cues resulted in impaired conflict resolution (i.e., resulted in a larger flanker congruency error rate differences) across both groups. This outcome is common in reaction time performance (e.g., Callejas, Lupianez, Funes, Tudela, 2005; Callejas, Lupianez, & Tudela, 2004; Fan et al., 2002) and is thought to reflect the alerting responses and to prevent additional stimulus processing (Callejas et al., 200 4; Callejas et al., 2005; Posner, 1994). Coinciding with this notion, it is not surprising that greater error rates were observed in our study for incongruent relative to congruent stimuli following the presentation of an alerting cue given the additional processing demands required in those circumstances. Even more compelling was the finding that TBI patients were unable to utilize spatial cues to overcome conflict created by incongruent targets. In general, orienting executive control system interaction s have been observed, with the orienting component of attention shown to facilitate the reduction of t he conflict effect (Callejas et al., 2004; Fan et al., 2002). It is thought that the orienting system may serve to enhance executive control (Callejas e t al., 2005). Spatial cues are thought to guide attention to the central target stimulus and engage selective focus making it easier for the observer
91 to ignore incongruent flanking stimuli and respond more quickly (Callejas et al., 2004). While this inter action between the orienting and executive control networks has traditionally benefitted reaction time performance, control participants in our study revealed this pattern of performance with regard to error rate. Whereas controls demonstrated the appropri ate reduction in error rate to incongruent stimuli that were preceded by spatially informative cues, TBI participants did not derive the same benefit. TBI patients were unable to overcome the conflict created by the incongruent target and continued to exhi bit substantially larger error rates for incongruent targets despite the guidance provided by spatial cueing which serves to orient attention. These findings largely suggest that TBI may disrupt neural systems responsible for the communication between the orienting and executive control networks. Implications Regarding our formal behavioral measures of attentional networks (i.e., RT subtractions) that were derived from the LANT, no indication of hemispheric differences in the representation of alerting, ori enting, or executive control networks were present in healthy individuals. Despite these surprising outcomes, more direct comparison of hemispheric network scores resulted in the emergence of behavioral reductions in the efficiency of left orienting for TB I patients. Paralleling the left orienting behavioral decrements were hemispheric N1 reductions in TBI patients that occurred upon presentation of left spatial cues i.e., during the time period that the left orienting network was recruited. We speculat e that these neural alterations which were specific to leftward shifts of attention, reflect the reduced left orienting behavioral performance found in our TBI sample. If so, it is plausible that right hemispheric N1 amplitude may correspond to the
92 degree of left orienting impairment that occurs as a consequence of TBI. With these promising findings comes the possibility that the N1 may also serve as a TBI prognostic marker of orienting system functioning during the course cognitive recovery. Certainly, th e nature of this behavioral neural relationship which appears specific to the left orienting system requires additional detailed study. The impact of TBI on the interaction between the attention networks was best understood by investigating factors that contributed to error rate and reaction time performance. With this line of inquiry, it was found that TBI resulted in a disconnection in the dynamic communication processes that occur between the orienting and executive control attentional systems. These d eficits are likely to appear in everyday life as difficulties incorporating salient environmental information (i.e., cues) to optimize performance related to a task at hand. A reduced ability to shift attention may decrease selective focus particularly in situations where there is equally salient information competing for attention. While healthy individuals are likely capable of utilizing the guidance of environmental cues to ignore the surrounding distractors that compete for attention, individuals with a TBI may be unable to overcome the conflict that is created by these distractors, thereby making them prone to performance error. These deficits may become more apparent in real life situations that are more cognitively demanding. Take, for example, a sit uation where an individual is driving exogenous cue to which the driver refle xively orients attention in that direction. The
93 potentially distracting aspects of the environment (e.g., the radio playing, cars driving in the oncoming lane), the driver is able to engage selective focus to the task at hand, appropriately modify the course of the car, and avoid hitting the ball. For TBI patients, the disjointed interplay between the orienting and executive control networks can result in disastrous mistakes. Our exploratory analyses revealed that the degree of left orienting behavioral impairment was strongly related to performance on a task that requires the integration of executive function skills, which again emphasizes the critical relationship between th e systems occurs as a result of damage to the orienting and executive control networks independent of one another, or whether TBI impacts white matter projections between the two networks. As expected, TBI patients possessed intact phasic alerting ability regardless of the extent of injury. Given the spared alerting capacity, we propose that utilization of alerting cues may help to remediate the lack of interaction between th e orienting and executive control systems, and such techniques can be implemented in cognitive rehabilitation training. In fact, it has been shown that the provision of alerting cues in visuospatial neglect patients influences the hemispheric functioning o f the orienting and executive control networks. In right hemisphere damaged/left neglect patients, when phasic alertness is engaged by auditory cues, improvements in orienting deficits to left sided targets and reduced interference of distracters (i.e., co nflict) in the left visual field are observed (Chica et al., 2012; Robertson, Mattingley, Rorden, & Driver, 1998). As such, the alerting system may modulate the interplay between the orienting and
94 executive control systems, with phasic alerting improving o rienting efficiency, and orienting improvement resultantly reducing conflict processing impairment. Possible Limitations and Future Directions Our findings are to be interpreted in light of several limitations that warrant further discussion. Firstly, the small sample sizes of both groups impacts the generalizability of findings to a broader population base. It is also possible that having a limited number of participants influenced the statistical power to detect significant changes in the current study. Moreover, the heterogeneity of TBI poses additional challenges as grouping TBI patients together subdues potentially meaningful within group variance that may arise from individual differences in lesion site as well as mechanism of injury. Such drawbacks m ay be circumvented in the future by clustering TBI patients according to to post injury c omparison of attention efficiency and ensure comparability across time points. There are also limitations that are inherent to the utilization of electrophysiological techniques to measure of cognitive processes. Specifically, the offset to the superior t emporal resolution of EEG is a reduction in spatial resolution. As such, EEG precludes the precise localization of the neuroanatomical substrates of the attention subsystems that are of interest to the study. Current methods that permit concurrent EEG fMRI acquisition would be best suited for localization of brain regions as well as detection of the neural time course of underlying cognitive processes. Despite such drawbacks, the results of the current study may serve as a basis to guide future theoretic ally driven hypotheses and investigations. To address the behavioral aspect of attention network laterality more definitely, future work may wish to
95 extend the spatial cues and target stimuli more peripherally to ensure adequate sensitivity to detecting su btle laterality effects, particularly within the orienting network. Increasing retinal eccentricity decreases visual acuity and requires additional attention resources for appropriate discrimination of the target stimulus (Asanowicz et al., 2012; Golla, I gnashchenkova, Haarmeier, & Their, 2004). The modification of eccentricity will increase the attentional demands of the observer and enhance the likelihood of capturing the effects of interest. Furthermore, it is debatable whether the current paradigm per mitted pure isolation of the exogenous orienting system. Seminal work conducted by Corbetta and Shulman (2002) has shown that the orienting system can be further divided into a dorsal endogenous task demand. In particular, the physical attributes of the spatial cue has allowed for dissociation of the two orienting systems. Exogenous cues are un expected and may location. Endogenous cues require shifts of attention that are attributable to within person factors (e.g., observer goals), and may be presented centr ally as an arrow indicating the direction that the observer should voluntarily orient attention. Thus, by presenting valid cues i.e., cues that are always predictive of spatial location learned contingences develop and these resulting internal expectations drive the endogenous orienting system (Chica, Bartolomeo, & Valero Cabre, 2011). Typically, exogenous attention is evoked by manipulating cue validity (e.g., presenting invalid cues that are not predictive of spatial location), thereby leaving the observe r unable to develop an
96 internal probability representation of target location. It is quite possible that the valid cues presented in the LANT elicited the endogenous orienting system. Given that in the LANT, the cue always predicted target location begs th e question of whether (Corbetta and Shulman, 2002). Manipulating the validity of the s patial cues in future work may help to ensure pure isolation of the exogenous orienting system of attention, bias on participant response. The removal of within pe the endogenous orienting system, such as expectation bias, is integral to the appropriate segregation of the exogenous orienting system. Given the possibility that the N1 may serve as a neural marker of the degree of eff iciency of the left orienting network in TBI survivors, future work may focus on longitudinal tracking the recovery of orienting function on the basis of N1 amplitude restoration. To address our TBI orienting th more certainty, we are currently looking at the nature of white matter changes in frontoparietal projections that occur as an outcome of TBI, and the relationship of white matter integrity to LANT performance. Diffusion imaging of in vivo white matter c onnectivity is a promising research and clinical methodology as its sensitivity to detecting brain injury in both acute and chronic phases surpasses current clinical imaging methods, and it has been found to predict injury severity as well as functional ou tcome (Betz, Zhuo, Roy, Shanmuganathan, & Gullapalli, 2012; Kraus et al., 2007).
97 purported modulatory effect on the orienting executive control system dynamic interaction. Prov ision of a sensory stimulus to evoke the phasic alerting system may be outcomes of such investigations have the potential to guide future cognitive rehabilitation interve ntions, and the tracking of recovery of function in TBI survivors. While much work lies ahead with regard to these beliefs, it is possi ble that utilizing TBI intact alerting capacity may be a critical component to the improvement of impairment in the orienting and executive control constituents of visuo spatial attention.
98 LIST OF REFERENCES Arciniegas, D. B., Held, K., & Wagner, P. (2002). Cognitive impairment following traumatic brain injury. Current Treatment Options in Neurology, 4, 43 57. Asa nowicz, D., Marzecova, A., Jaskowski, P., & Wolski, P. (2012). Hemispheric asymmetry in the efficiency of attentional networks. Brain and Cognition, 79, 117 128. Audet, T., Mercier, L., Collard, S., Rochette, A., & Hebert, R. (2000). Attention deficits: I s there a right hemisphere specialization for simple reaction time, sustained attention, and phasic alertness? Brain and Cognition, 43, 17 27. Auerbach, S.H. (1986). Neuroanatomical correlates of attention and memory disorders in traumatic brain injury: A n application of neurobehavioral subtypes. Journal of Head Trauma Rehabilitation, 1, 1 12. Bate, A.J., Mathias, J.L., & Crawford, J.R. (2001). The covert orienting of visual attention following severe traumatic brain injury. Journal of Clinical and Experim ental Neuropsychology, 23, 386 398. Beck, A. T., Steer, R. A., & Brown, G. K. (1996). Beck Depression Inventory Second Edition (BDI II) USA: The Psychological Corporation. Benton, A. & Hamsher, K. (1976). Multilingual aphasia examination. Iowa City, IA : University of Iowa. Benton, A. L., Sivan, A. B., Hamsher, K. de S., Varney, N. R., & Spreen, O. (1994). Contributions to neuropsychological assessment (2nd ed.). New York: Oxford University Press. Berridge, C.W. & Waterhouse, B.D. (2003). The locus coeru leus noradrenergic system: Modulation of behavioral state and state dependent cognitive processes. Brain Research Reviews, 42, 33 84. Betz, J., Zhuo, J., Roy, A., Shanmuganathan, K., Gullapalli, R.P. (2012). Prognostic value of diffusion tensor imaging p arameters in severe traumatic brain injury. Journal of Neurotrauma, 29, 1292 1305. Bigler, E.D. (1990). Neuropathology of traumatic brain injury. In. E.D. Bigler (Ed.), Traumatic Brain Injury Austin, TX: Pro ed. Bishara, S.N., Partridge, F.M., Godfrey, H. P., & Knight, R.G. (1992). Post traumatic amnesia and Glasgow Coma Scale related to outcome in survivors in a consecutive series of patients with severe closed head injury. Brain Injury, 6, 373 380.
99 Blair, J.R., & Spreen, O. (1989). Predicting premorbid IQ : A revision of the National Adult Reading Test. The Clinical Neuropsychologist, 3, 129 136. Boelen, D.H.E., Spikman, J.M., Rietveld, A.C.M., & Fasotti, L. (2009). Executive dysfunction in chronic brain injured patients: Assessment in outpatient rehabilita tion. Neuropsychological Rehabilitation, 19 625 644. Bond, M.R. (1986). Neurobehavioral sequelae of closed head injury. In I. Grant & K. M. Adams (Eds.), Neuropsychological assessment of neuropsychological disorders (pp. 347 373), New York: Oxford Univers ity Press. Botvinick, M.M, Braver, T.S., Barch, D.M., Carter, C.S., & Cohen, J.D. (2001). Conflict monitoring and cognitive control. Psychological Review, 108, 624 652. Brain Injury Association of America, DE: Brain Injury Association of America, 2010. B rooks, D.N., Aughton, M.E., Bond, M.R., Jones, P., & Rizvi, S. (1980). Cognitive sequelae in relationship to early indices of severity of brain damage after severe blunt head injury. Journal of Neurology, Neurosurgery, and Psychiatry, 43, 529 534. Brooks, D.N., & McKinlay, W. (1987). Return to work within the first seven years of severe head injury. Brain Injury, 1 5 15. Buysse, D.J., Reynolds,C.F., Monk,T.H., Berman,S.R., & Kupfer, D.J. (1989). The Pittsburgh Sleep Quality Index (PSQI): A new instrument f or psychiatric research and practice. Psychiatry Research, 28 193 213. Callejas, A., Lupianez, J., Funes, M.J., & Tudela, P. (2005). Modulations among the alerting, orienting, and executive control networks. Experimental Brain Research, 167, 27 37. Calle jas, A., Lupianez, J., & Tudela P. (2004). The three attentional networks: On their independence and interactions. Brain and Cognition, 54, 225 227. Carroll, J.F.X. & McGinley, J.J. (2000). Mental Health Screening Form III (MHSF III). Unpublished screenin g instrument, Project Return Foundation, Inc., New York, NY. Carter, C.S., Botvinick, M.M., & Cohen, J.D. (1999). The contribution of the anterior cingulate cortex to executive processes in cognition. Reviews in Neuroscience, 10, 49 57. Chica, A.B., Barto lomeo, P., & Valero Cabre, A. (2011). Dorsal and ventral parietal contributions to spatial orienting in the human brain. The Journal of Neuroscience, 31, 8143 8149.
100 Chica, A.B., de Schotten, M.T., Toba, M., Malhotra, P., Lupianez, J., & Bartolomeo, P. (20 12). Attention networks and their interactions after right hemisphere damage. Cortex, 48, 654 663. Cohen, J. (1988). Statistical power analysis for the behavioral sciences. Hillsdale, NJ: Erlbaum Associates. Corbetta, M., Miezin, F.M., Shulman, G.L., & Pe tersen, S.E. (1993). A PET Study of visuospatial attention. The Journal of Neuroscience, 13, 1202 1226. Corbetta, M., & Schulman, G.L. (2002). Control of goal directed and stimulus driven attention in the brain. Nature Reviews Neuroscience, 3, 201 215. Cor betta, M. & Shulman, G.L. (2011). Spatial neglect and attention networks. Annual Review of Neuroscience, 34, 569 599. Clark, J.H. (1924). The Ishihara Test for Color Blindness. American Journal of Physiological Optics 5 269 276. Cremona Meteyard, S.L., Clark, C.R., Wright, M.J., & Geffen, G.M. (1992). Covert orientation of visual attention after closed head injury. Neuropsychologia, 30, 123 132. Davidson, M.C., Cutrell, E.B., & Marrocco, R.T. (1999). Scopolamine slows the orienting of attention in prima tes to cued visual targets. Psychopharmacology, 142, 1 8. D.G.M., & Catani, M. (2011). A lateralized brain network for visuospatial attention. Nature Neuroscience, 14, 1245 1246. Dosenbach, N.U.F., Fair, D.A., Cohen, A.L., Schlaggar, B.L., & Petersen, S.E. (2008). Trends in Cognitive Sciences, 12, 99 105. Duncan, C.C., Kosmidis, M.H., & Mirsky, A.F. (2005). Closed head injury related information processing: An event related potential analysis. International Journal of Psychophysiology, 58, 133 157. Duncan, C.C., Summers, A.C., Perla, E.J., Couburn, K.L., & Mirsky, A.F. (2011). Evaluation of traumatic brain injury: Brain potentials in diagnosis, function, and prognosis. Inter national Journal of Psychophysiology, 82, 24 40. Duncan Johnson, C.C., & Donchin, E. 1977. On quantifying surprise: the variation of event related potentials with subjective probability. Psychophysiology, 14, 456 467. El Sayed, T., Mota, A., Fraternali, F., & Ortiz, M. (2008). Biomechanics of traumatic brain injury. Computational Methods in Applied Mechanics and Engineering, 197, 4692 4701.
101 Eriksen, B.A., & Eriksen, C.W. (1974). Effects of noise letters upon the identification of a target letter in a no nsearch task. Perception and Psychophysics, 16, 143 149. Fan, J., McCandliss, B.D., Fossella, J., Flombaum, J.I., & Posner, M.I. (2005). The activation of attentional networks. NeuroImage, 26, 471 479. Fan, J., McCandliss, B.D., Sommer, T., Raz, A., & Po sner, M.I. (2002). Testing the efficiency and independence of attentional networks. Journal of Cognitive Neuroscience 14 340 347. Faul, M., Xu, L., Wald, M.M., & Coronado, V.G. Traumatic brain injury in the United States: Emergency department visits, hos pitalizations, and deaths. Atlanta (GA): Centers for Disease Control and Prevention, National Center for Injury Prevention and Control; 2010. Golla, H., Ignashchenkova A., Haarmeier, T., & Thier, P. (2004). Improvement of visual acuity by spatial cueing: A comparative studyin in human and non human primates. Vision Research, 44, 1589 1600. Golden, C.J., (1978). Stroop Color and Word Test: A manual for clinical and experimental uses. Chicago, IL: Stoelting Co. Grace, J., & Malloy, P.F. (2001). Frontal Syst ems Behavior Scale Professional Manual Lutz, FL: Psychological Assessment Resources, Inc. Greene, D.J., Barnea, A., Herzberg, K., Rassis, A., Neta, M., Raz, A., & Zaidel, E. (2008). Measuring attention in the hemispheres: The lateralized attention network test (LANT), Brain and Cognition, 66, 21 31. Gronwall, D. (1987). Advances in the assessment of attention and information processing after head injury. In H.S. Levin, J. Grafman, & H.M. Eisenberg (Eds.), Neurobehavioral recovery from head injury (pp. 355 371). New York: Oxford University Press. Halterman, C.I., Langan, J., Drew, A., Rodriguez, E., Osternig, L.R., Chou, L., & van Donkelaar, P. (2006). Tracking the recovery of visuospatial attention deficits in mild traumatic brain injury. Brain 129, 747 75 3. Hart, S.J., Greene, S.R., Casp, M., & Belger, A. (2010). Emotional priming effects during Stroop task performance. NeuroImage, 49, 2662 2670. Heaton, R.K., Chelune, G.J., Talley, J.L., Kay, G.G., & Curtis, G. (1993). Wisconsin Card Sorting Test (WCST) m anual, revised and expanded. Odessa, FL: Psychological Assessment Resources. Heilman, K. M. (1995). Attentional asymmetries. In R. J. Davidson & K. Hugdahl (Eds.), Brain asymmetry (pp. 217 234). Cambridge: The MIT Press.
102 Heilman, K.M. & Watson, R.T. (1977 ). Mechanisms underlying the unilateral neglect syndrome. Advances in Neurology, 18 93 106. Hellawell, D.J., Taylor, R., & Pentland, B. (1999). Cognitive and psychosocial outcome following moderate or severe traumatic brain injury. Brain Injury, 13, 489 5 04. Kant, R., Duffy, J.D., & Pivovarnik, A. (1998). Prevalence of apathy following head injury. Brain Injury, 12, 87 92. Kaplan, E. F., Goodglass, H., & Weintraub, S. (2001). The Boston Naming Test (2nd ed.). Philadelphia, PA: Lippincott Williams & Wilkins Kesler, S.R., Adams, H.F., Blasey, C.M., & Bigler, E.D. (2003). Premorbid intellectual functioning, education, and brain size in traumatic brain injury: An investigation of the cognitive reserve hypothesis. Applied Neuropsychology: Adult, 10, 153 162. Ki nnunen, K.M., Greenwood, R., Powell, J.H., Leech, R., Hawkins, P.C., Bonnelle, V., Patel, M.C., Counsell, S.J., & Sharp, D.J. (2011). White matter damage and cognitive impairment after traumatic brain injury. Brain, 134, 449 463. Konrad, K., Neufang, S., H anisch, C., Fink, G.R., & Herpertz Dahlmann, B. (2004). Dysfunctional attentional networks in children with attention/deficit hyperactivity disorder: Evidence from an event related functional magnetic resonance imaging study. Biological Psychiatry, 2006, 6 43 651. Kraus, M.F., Susmaras, T., Caughlin, B.P., Walker, C.J., Sweeney, J.A., & Little, D.M. (2007). White matter integrity and cognition in chronic traumatic brain injury: A diffusion tensor imaging study. Brain, 130, 2508 2519. Levin, H.S., Hamsher, K. de S., & Benton, A.L. (1975). A short form of the Test of Facial Recognition for clinical use. Journal of Psychology, 91, 223 228. Levin, H. S., High, W. M., Goethe, K. E., Sisson, R. A., Overall, J. E., Rhoades, H. M., Eisenberg, H. M., Kalisky, Z., & Gary, H. E. (1987). The Neurobehavioral Rating Scale: Assessment of the behavioral sequelae of head injury by the clinician: Journal of Neurology, Neurosurgery, and Psychiatry, 50, 183 193. Lew, H.L., Lee, E.H., Pan, S.S. & Date, E.S. (2004). Electrophys iologic abnormalities of auditory and visual information processing patients with traumatic brain injury. American Journal of Physical Medicine and Rehabilitation, 83, 428 433. Lezak, M.D. (1991). Emotional impact of cognitive inefficiencies in mild head trauma [abstract]. Journal of Clinical and Experimental Neuropsychology 13, 23. Lezak, M.D., Howieson, D.B., & Loring, D.W. (2004). Neuropsychological Assessment (4 th edition). New York: Oxford University Press.
103 Losier, B.J.W., & Klein, R.M. (2001). A r eview of the evidence for a disengage deficit following parietal lobe damage. Neuroscience and Biobehavioral Reviews, 25, 1 13. MacDonald, A.W. III, Cohen, J.D., Stenger, V.A., & Carter, C.S. (2000). Dissociating the role of the dorsolateral prefrontal an d anterior cingulate cortex in cognitive control. Science, 288, 1835 1838. Marin R.S., Biedrzycki, R.C., & Firinciogullari, S. (1991). Reliability and validity of the Apathy Evaluation Scale. Psychiatry Research 38, 143 62. Marzecova, A., Asanowicz, D., Kriva, L., & Wodniecka, Z. (2012). The effects of bilingualism on efficiency and lateralization of attentional networks. Bilingualism: Language and Cognition 1 16. Mathias, J.L., & Coats, J.L. (1999). Emotional and cognitive sequelae to mild traumatic brain injury. The Journal of Clinical and Experimental Neuropsychology 21, 200 215. Medana, I.M., & Esiri, M.M. (2003). Axonal damage: A key predictor of outcome in human CNS diseases. Brain, 126, 515 530. Mesulam, M.M. (1999). Spatial attention and n eglect: parietal, frontal, and cingulate contributions to the mental representation and attentional targeting of salient extrapersonal events. Philosophical Transactions of the Royal Society B: Biological Sciences, 354, 1325 1346. Milner, B. Effects of di fferent brain lesions on card sorting: The role of the frontal lobes. (1963). Archives of Neurology, 9, 100 110. Naatanen, R., Gaillard, A.W., & Mantysalo, S. (1978). Early selective attention effect on evoked potential reinterpreted. Acta Psychologica, 42 313 329. Neuhaus, A.H., Urbanek, C., Opgen Rhein, C., Hahn, E., Ta, T., Koehler, S., Gross, M., & Dettling, M. Event related potentials associated with Attention Network Test. (2010). International Journal of Psychophysiology, 76, 72 79. Niogi, S.N., Muk herjee, P., Ghajar, J., Johnson, C., Kolster, R.A., Sarkar, R., Lee, H., Meeker, M.R., Zimmerman, R.D., Manley, G.T., & McCandliss, B.D. (2008). Extent of microstructural white matter injury in postconcussive syndrome correlates with impaired cognitive rea ction time: A 3T diffusion tensor imaging study of mild traumatic brain injury. American Journal of Neuroradiology, 29, 967 973. Nobre, A.C., Sebestyen, G.N., & Miniussi, C. (2000). The dynamics of shifting visuospatial attention revealed by event related potentials. Neuropscyhologia, 38, 964 974.
104 Olver, J.H., Ponsford, J.L., & Curran, C.A. (1996). Outcome following traumatic brain injury: A comparison between 2 and 5 years after injury. Brain Injury, 20, 841 848. Park, N.W., Moscovitch, M., & Robertson, I.H. Divided attention impairments after traumatic brain injury. Neuropsychologia, 37, 1119 1133. Pavlovskaya, M., Groswasser, Z., Keren, O., Mordvinov, E., & Hochstein, S. (2007). Hemispheric visual atentional imbalance in patients with traumatic brain i njury. Brain and Cognition, 64, 21 29. Petersen, S.E. & Posner, M.I. (2012). The attention system of the human brain: 20 years after. Annual Review of Neuroscience, 35, 73 98. Picton, T.W. (1992). The P300 wave of the human event related potential. Journ al of Clinical Neurophysiology, 4, 456 476. Polich, J. & Kok, A. (1995). Cognitive and biological determinants of P300: An integrative review. Biological Psychology, 41 103 146. Posner, M.I. (1980). Orienting of attention. The Quarterly Journal of Exper imental Psychology, 32, 3 25. Posner, M. I. (1994). Attention: The mechanisms of consciousness. Proceedings of the National Academy of Sciences, 91, 7398 7403. Posner, M.I. (2008). Measuring alertness. Annals of the New York Academy of Sciences, 1129: 19 3 199. Posner, M.I. & Fan, J. (2004). Attention as an organ system. James R. Pomerantz and Michael C. Crair (Eds.), Topics in Integrative Neuroscience: From Cells to Cognition Cambridge: Cambridge University Press. Posner, M.I. & Petersen, S.E. (1990). The attention system of the human brain. Annual Review of Neuroscience, 13 25 42. Poynter, W., Ingram, P., & Minor, S. (2010). Visual field asymmetries in attention vary with self reported attention deficits. Brain and Cognition, 72 355 361. Priganto, G. P. (2005). Disturbances of self awareness and rehabilitation of patients with traumatic brain injury: A 20 year perspective. Journal of Head Trauma Rehabilitation, 20, 19 29. Ratcliff, R. (1993). Methods for dealing with reaction time outliers. Psycholog ical Bulletin, 114 510 532. Reitan, R.M. (1955). The relation of the Trail Making Test to organic brain damage. Journal of Consulting Psychology, 19, 393 394.
105 Rey, A. (1964). Paris: Presses Universitaires de France. Rios, M., Perianez, J.A., & Munoz Cespedes, J.M. (2004). Attentional control and slowness of information processing after severe traumatic brain injury. Brain Injury, 18, 257 272. Robertson, I.H. (1999). Cognitive rehabilitation: attention and neglect. Trends i n Cognitive Sciences, 3, 10, 385 393. Robertson, I. H., Mattingley, J. B., Rorden, C., & Driver, J. (1998). Phasic alerting of neglect patients overcomes their spatial deficit in visual awareness. Nature,395 169 175. Scherg, M. (1990). Fundamentals of dip ole source potential analysis. In F. Grandori & M. Hoke (Eds.), Auditory evoked magnetic fields and electric potentials. Advances in audiology (Vol. 6, pp. 65 78). Basel: Karger. Schultheiss, O.C., Patalakh, M., & Rosch, A.G. (2012). Salivary progesterone is associated with reduced coherence of attentional, cognitive, and motivational systems. (2012). Brain and Cognition, 80, 214 222. Salmi, J., Rinne, T., Degerman, A., & Alho, K. (2007). Orienting and maintenance of spatial attention in audition and vision : An event related brain potential study. European Journal of Neuroscience, 25, 3725 3733. Spielberger, C.D., Gorusch, R.L., Lushene, R., Vagg, P.R., & Jacobs, G.A. (1983). Manual for the State Trait Anxiety Inventory Palo Alto, CA: Consulting Psychologi sts Press. Spikman, J.M., Timmerman, M.E., van Zomeren, A.H., & Deelman, B.G. (1999). Recovery versus retest effects in attention after closed head injury. Journal of Clinical and Experimental Neuropsychology, 21, 585 605. Stierwalt, J.A.G. & Murray, L.L. (2002). Attention impairment following traumatic brain injury. Seminars in Speech and Language, 23, 129 138. Sturm, W. & Willmes, K. (2001). On the functional neuroanatomy of intrinsic and phasic alertness. NeuroImage, 14, 76 84. Stuss, D.T., Binns, M.A ., Carruth, F.G., Levine, B., Brandys, C.E., Moulton, R.J., Snow, W.G., & Schwartz, M.L. (1999). The acute period of recovery from traumatic brain injury: Posttraumatic amnesia or posttraumatic confusional state? Journal of Neurosurgery, 90, 635 643.
106 Stuss D.T., Levine, B., Alexander, M.P., Hong, J., Palumbo, C., Hamer, L., Murphy, K.J., & Izukawa, D. (2000). Wisconsin Card Sorting Test performance in patients with focal frontal and posterior brain damage: Effects of lesion location and test structure on s eparable cognitive processes. Neuropsychologia, 38, 388 402. Tao, L., Marzecova, A., Fafta, M., Asanowicz, D., & Wodniecka, Z. (2011). The efficiency of attentional networks in early and late bilinguals: The role of age acquisition. Frontiers in Psychology 2, 1 19. The Psychological Corporation. (1997). WAIS III WMS II technical manual. San Antonio, TX: The Psychological Corporation. The Psychological Corporation. (2002). WAIS III WMS II technical manual. San Antonio, TX: The Psychological Corporation. Tra nel, D., Vianna, E., Manzel, K., Damasio, H., & Grabowski, T. (2009). Neuroanatomical correlates of the Benton Facial Recognition Test and Judgment of Line Orientation Test. Journal of Clinical and Experimental Neuropsychology, 31, 219 233. Van Voorhis, S. & Hillyard, S.A. (1977). Visual evoked potentials and selective attention to points in space. Perception and Psychophysics, 22, 54 62. Warrington, E.K. & James, M. (1967). An experimental investigation of facial recognition in patients with unilateral c erebral lesions. Cortex, 3, 317 326. Whyte, J., Hart, T., Laborde, A., & Rosenthal, M. (2004). Rehabilitation issues in traumatic brain injury. In: J.A. DeLisa, B.M. Gans & W.L. Bockenek (Eds.), Rehabilitation medicine: principles and practice (3rd ed.) (pp. 1677 1713). Pennsylvania: Lippincott Williams & Wilkins. Willemse van Son, A.H.P., Ribbers G.M., Hop, W.C.J., & Stam, H.J. (2009). Factors that lead to successful community integration following severe traumatic brain injury. Journal of He ad Trauma Rehabilitation, 21, 8 21. Willemse van Son, A.H.P., Ribbers G.M., Verhagen, A.P., & Stam, H.H. (2007). Prognostic factors of long term functioning and productivity after traumatic brain injury: A systemic review of prospective cohort studies. Cli nical Rehabilitation, 21, 1024 1037. Wilson, B. A., Alderman, N., Burgess, P. W., Emslie, H., & Evans, J. J. (1996). Behavioural assessment of the dysexecutive syndrome Bury St. Edmunds, UK: Thames Valley Test Company. Zahn, T.P., & Mirsky, A.F. (1999). Reaction time indicators of attention deficits in closed head injury. Journal of Clinical and Experimental Neuropsychology, 21, 352 367.
107 BIOGRAPHICAL SKETCH Tanisha G. Hill Jarrett graduated with a Bachelor of Science degree in p sychology from the Univers ity of Pittsburgh in 2010. Following a one year post baccalaureate research program at the University of Pittsburgh, she began her graduate education at the University of Florida in 2011. She will receive her Master of Science from the University of Florid a in 2013, and pl ans to pursue her Doctorate in c linical p sych ology, with a concentration in n europsychology.