Citation
Attention and Emotion

Material Information

Title:
Attention and Emotion Extent and Time Course of Competition between Task Stimuli and Affective Pictures in Human Visual Cortex
Creator:
Deweese, Elizabeth M McGinnis
Place of Publication:
[Gainesville, Fla.]
Florida
Publisher:
University of Florida
Publication Date:
Language:
english
Physical Description:
1 online resource (85 p.)

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Psychology
Committee Chair:
KEIL,ANDREAS
Committee Co-Chair:
LANG,PETER J
Committee Members:
BRADLEY,MARGARET M
KAAN,EDITH
Graduation Date:
8/9/2014

Subjects

Subjects / Keywords:
Amplitude ( jstor )
Amplitude modulation ( jstor )
Anxiety ( jstor )
Fear ( jstor )
Mental stimulation ( jstor )
Signals ( jstor )
Snake phobias ( jstor )
Snakes ( jstor )
Time windows ( jstor )
Visual cortex ( jstor )
Psychology -- Dissertations, Academic -- UF
attention -- emotion -- fear -- ssvep
Genre:
bibliography ( marcgt )
theses ( marcgt )
government publication (state, provincial, terriorial, dependent) ( marcgt )
born-digital ( sobekcm )
Electronic Thesis or Dissertation
Psychology thesis, Ph.D.

Notes

Abstract:
Cues signaling threat attract attentional resources based on their inherent stimulus significance, which may be at the cost of processing task-related information. Of central importance is how the visual system resolves competition for processing resources among stimuli differing in motivational salience. The experiments described in this dissertation were designed to specifically analyze inter-individual differences regarding the extent and time course of distraction effects observed in participants high in snake fear, relative to controls, and to quantify levels of competition in visual cortex, using electrophysiology. To examine processing of specific fear cues at the level of lower-tier visual cortical processing, we employed steady-state visual evoked potentials (ssVEPs), which measure the amplitude of neural activity elicited by a task-relevant stimulus flickering at a known rate; competition by a distractor can therefore be measured as an attenuation of the task-evoked processing. Results support the notion that emotionally arousing cues compete for limited resources at the level of the visual cortex, and further provides a foundation for future research investigating visuocortical trade-off effects in high fear populations. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis:
Thesis (Ph.D.)--University of Florida, 2014.
Local:
Adviser: KEIL,ANDREAS.
Local:
Co-adviser: LANG,PETER J.
Statement of Responsibility:
by Elizabeth M McGinnis Deweese.

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Resource Identifier:
968786082 ( OCLC )
Classification:
LD1780 2014 ( lcc )

Downloads

This item has the following downloads:


Full Text

PAGE 1

ATTENTION AND EMOTION: EXTENT AND TIME COURSE OF COMPETITION BETWEEN TASK STIMULI AND AFFECTIVE PICTURES IN HUMAN VISUAL CORTEX By ELIZABETH MENTON MCGINNIS DEWEESE A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE U NIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2014

PAGE 2

© 2014 Elizabeth Menton McGinnis Deweese

PAGE 3

To my parents, Jim and Cindy McGinni s, to my husband, Matthew, and to my family and friends, for their enduring love and continued support

PAGE 4

4 ACKNOWLEDGMENTS It is with the utmost level of gratitude and respect that I acknowledge my mentor, Andreas Keil, for his guidance, encouragement, an d support throughout my graduate career. I would also like to express sincere thanks to Margaret Bradley and Peter Lang for their expertise and co mentorship. In addition, I thank Edith Kaan for her valuable input on my dissertation project. I am also grat eful to all former and current members of the Center for the Study of Emotion and Attention, who have enriched my daily life in ways not limited to scientific endeavors. This research was supported by the National Institute of Mental Health Grants R01 MH08 4932 02 and R01 MH097320, and by a grant from the Spanish Government (I+D+i PSI2009 07066), awarded to Andreas Keil.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURES ................................ ................................ ................................ .......... 8 ABSTRACT ................................ ................................ ................................ ..................... 9 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 10 2 SUSTA INED COMPETITIVE BIASES TO VISUAL SNAKE FEATURES: HYPERVIGILANCE WITHOUT AVOIDANCE ................................ ......................... 17 Method ................................ ................................ ................................ .................... 19 Participants ................................ ................................ ................................ ....... 19 Stimuli and Procedure ................................ ................................ ...................... 20 EEG Recording ................................ ................................ ................................ 23 EEG Reduction and Analyses ................................ ................................ .......... 23 Steady state Visual Evoked Potential Analyses ................................ ............... 24 Statistical Analyses: Behavioral data and SAM ratings ................................ .... 25 Statistical Analysis: ssVEP time course ................................ ........................... 25 Results ................................ ................................ ................................ .................... 27 Behavioral data and SAM ratings ................................ ................................ ..... 27 ssVEPs ................................ ................................ ................................ ............. 27 Discussion ................................ ................................ ................................ .............. 29 3 COMPETITION EFFECTS OF EMOTIONALLY AROUSING DISTRACTORS ON A CONCURRENT TASK ................................ ................................ .................. 41 Methods ................................ ................................ ................................ .................. 42 Participants ................................ ................................ ................................ ....... 42 Stimuli and procedure ................................ ................................ ...................... 43 EEG Recording ................................ ................................ ................................ 44 EEG Reduction and Analyses ................................ ................................ .......... 44 Steady state Visual Evoked Potential Analyses ................................ ............... 45 Statistical Analyses: Behavioral data and SAM ratings ................................ .... 45 Statistical An alysis: ssVEP time course ................................ ........................... 46 Coherent motion detection (dot) task ................................ ............................... 46 Picture Distractors ................................ ................................ ............................ 47 Competition analysis ................................ ................................ ........................ 47 Results ................................ ................................ ................................ .................... 48 Behavioral data and SAM ratings ................................ ................................ ..... 48

PAGE 6

6 ssVEPs ................................ ................................ ................................ ............. 49 Coherent motion detection (dot) task ................................ ......................... 49 Picture distractors ................................ ................................ ...................... 49 Competition analysis ................................ ................................ ........................ 50 Discussion ................................ ................................ ................................ .............. 51 4 COMPETITION EFFECTS IN HIGH AND LOW SNAKE FEA RFUL PARTICIPANTS ................................ ................................ ................................ ...... 60 Methods ................................ ................................ ................................ .................. 6 1 Participants ................................ ................................ ................................ ....... 61 High fear group ................................ ................................ .......................... 61 Low fear group ................................ ................................ ........................... 62 Stimuli and procedure ................................ ................................ ...................... 63 EEG Recording ................................ ................................ ................................ 63 EEG Reduction and Analyses ................................ ................................ .......... 63 Steady state Visual Evoked Potential Analyses ................................ ............... 64 Statistical Analyses: Behavioral data and SAM ratings ................................ .... 64 Statistical Analysis: ssVEP time course ................................ ........................... 65 Between group analysis: Dot stimulus ................................ ....................... 66 Between group analysis: Picture stimulus ................................ .................. 66 Results ................................ ................................ ................................ .................... 67 Behavioral data and SAM ratings ................................ ................................ ..... 67 ssVEPs ................................ ................................ ................................ ............. 68 Between group comparison: Dot stimulus ................................ .................. 68 Between group comparison: Picture stimulus ................................ ............ 68 Discussion ................................ ................................ ................................ .............. 68 5 GENERAL DISCUSSION ................................ ................................ ....................... 73 Emotional Stimuli Act as Strong Competitors for Visual Processing Resources ..... 74 Specificity of the Visuocortical Response to Fear Cue s ................................ .......... 76 Perceptual Enhancement of Motivationally Relevant Stimuli in Visual Cortex ........ 78 Conclusions ................................ ................................ ................................ ............ 79 LIST OF REFERENCES ................................ ................................ ............................... 80 BIOGRAPHICAL SKETCH ................................ ................................ ............................ 84

PAGE 7

7 LIST OF TABLES Table page 2 1 Mean ssVEP amplitude (change from baseline) for pleasant (erotica, cat) and neutral (work, cows) distractors for high and low fear participants. .................... 33 3 1 Average ssVEP amplitude values used to calculate competition index. ............. 53

PAGE 8

8 LIST OF FIGURES Figure page 2 1 Time sequence for a single trial. ................................ ................................ ......... 34 2 2 Grand average time domain , topogr aphical distribution , and frequency spectra across all participants and conditions. ................................ ................... 35 2 3 B ehavioral performance data . ................................ ................................ ............. 36 2 4 H ilbert transformed data for high and low fear participants for all conditions. ..... 37 2 5 Permutation corrected waveforms. ................................ ................................ ..... 38 2 6 Me an ssVEP amplitude averaged across high and low fear participants, for pleasant, neutral, and unpleasant distractors. ................................ .................... 39 2 7 Mean ssVEP amplitude calcula ted as the change from baseline for mut ilation and snake categories for all participants. ................................ ........................... 40 3 1 Time sequence for a single trial. ................................ ................................ ......... 54 3 2 Grand average d frequency spec tra across all low fear participants. .................. 55 3 3 Behavioral performance . ................................ ................................ .................... 56 3 4 Grand average Hilbert transform. ................................ ................................ ....... 57 3 5 Average ssVEP amplitude for flickering dot (gray circles) and picture (black squares) stimuli relative to the baseline. ................................ ............................. 58 3 6 C ompetition i ndex plots . ................................ ................................ ..................... 59 4 1 Time sequence for a single trial. ................................ ................................ ......... 70 4 2 Grand averaged frequency spectra and r epresentation of electrode pla cement. ................................ ................................ ................................ .......... 71 4 3 Average ssVEP amplitude modulation to the flickering dot stimulus for low fear (left) and high fear (right) participants. ................................ ......................... 72

PAGE 9

9 Abstract of Disse rtation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy ATTENTION AND EMOTION: EXTENT AND TIME COURSE OF COMPETITION BETWEEN TASK STIMULI AND A FFECTIVE PICTURES IN HUMAN VISUAL CORTEX By Elizabeth Menton McGinnis Deweese August 2014 Chair: Andreas Keil Major: Psychology Cues signaling threat attract attentional resources based on their inherent stimulus significance, which may be at the cost of processing task related information. Of central importance is how the visual system resolves competition for processing resources among stimuli differing in motivational salience. The experiments conducted here were designed to specifically analyze inte r individual differences regarding the extent and time course of distraction effects for participants with self reported high fear of snakes, compared to low fear individuals, and to quantify levels of competition in visual cortex using electrophysiology. To examine processing of specific fear cues at the level of lower tier visual cortical processing, we employed steady state visual evoked potentials (ssVEPs), which measure the amplitude of neural activity elicited by a task relevant stimulus flickering at a known rate; competition by a distractor is measured as an attenuation of the task evoked processing. Results support the hypothesis that emotionally arousing cues compete for limited resources at the level of the visual cortex, and further provide a fou ndation for future research investigating visuocortical trade off effects in high fear individuals .

PAGE 10

10 CHAPTER 1 INTRODUCTION In a constantly changing environment, observers are flooded with a barrage of sensory information, of which only a subset can be pro cessed in depth. The combination of high throughput sensory traffic and limited capacity results in a daunting computational challenge for the visual system. In an environment in which multiple stimuli are competing for attentional resources, the visual sy stem is unable to fully process each object in parallel and therefore institutes competition for neural representation. Attention leads to a prioritization of information that is most relevant to the behavioral goal at hand, while simultaneously suppressin g distracting information (James, 1890; Desimone and Duncan, 1995; Knudsen, 2007). Many studies have demonstrated competition effects in recordings from visual neurons, which display pronounced cost effects when multiple stimuli, preferred and non prefer red, occupy their receptive fields (Reynolds, 1999). Only a subset of incoming information is processed in the visual cortex, highlighting the need for a selection mechanism that resolves competition and enables access of relevant information to limited ca pacity systems. One dominant view described by Desimone and Duncan (1995), is biased competition, in which the allocation of attention to competing representations is influenced, and ultimately determined by, a combination of stimulus properties (such as i ntensity and saliency) and attentional top down signals, which can bias competition toward stimuli that are important for behavior (Desimone and Duncan, 1995; Reynolds and Heeger, 2009). In this vein, emotionally arousing stimuli have been hypothesized to automatically attract attentional resources based on inherent stimulus significance, resulting in optimized sensory processing. Of central importance is how the

PAGE 11

11 visual system resolves competition for processing resources among stimuli differing in motivati onal salience, and in particular, the ways in which emotionally arousing stimuli refers to the engagement of macroscopic populations of visual neurons: thus, in discussing metaphorical currency for a biological one (see also Franconeri et al., 2013). Cues that signal threat or reward automatically attract attentional resources based on their inherent stimulu s significance, optimizing their processing in sensory systems (Bradley, et al., 2003), which may be at the cost of processing concurrent information (Ihssen et al. 2007). Empirically, task irrelevant appetitive and aversive stimuli (as distractors) interf ere with a variety of perceptual and cognitive tasks (Schimmack and Derryberry, 2005) suggesting that they are powerful competitors for limited capacity. Evidence for distraction by the presence of emotional stimuli in the visual domain comes from a broad base of behavioral, physiological, electrocortical, and neuroimaging studies (Müller et al., 2008; Öhman et al., 2001; Sabatinelli et al., 2005; Shafer, et al., 2012; Wangelin et al., 2011; Wieser et al., 2012). A number of different theories have attemp ted to explain why certain stimuli biased competition theory suggests arousal modulates the strength of (i.e., biases) competing mental representations beginning in p erception and continuing into memory consolidation. Early models of emotion proposed a biphasic organization of motive systems associated with either preservative or protective reflexive behaviors (Konorski, 1967), and were considered to be the behavioral basis of emotions. More recent

PAGE 12

12 theoretical advances of emotion and attention are rooted in theories highlighting the role of emotional stimuli as action cues, prompting preparation of the organism for action (Frijda, 1986). An extensive body of work by Lan g and colleagues (Lang et al., 1990) has examined behavioral and physiological responses during affective picture processing assessing the theoretical view that expressed emotions are founded on evolutionarily old motivational circuits in the brain that de veloped to ensure the survival of an organism. This body of research has shown that emotional stimuli attract attention involuntarily, resulting in prioritized responses to survival relevant information when measured with physiological, electrocortical, or neuroimaging indices (for a review, see Lang and Bradley, 2010). In fact, attentional biases toward threatening information have been implicated as an important factor in the etiology and maintenance of disorders in the fear and anxiety spectrum (Becker et al., 2001; see Bar Haim et al., 2007 for review). Many theoretical models of anxiety implicate maladaptive visuo spatial attentional processing of fearrelated information in the onset and maintenance of symptoms. In anxious individuals, for example, at tention to threatening information is highly prioritized over neutral or positive information and has been established in many studies (e.g., Öhman et. al, 2001; Sabatinelli et. al, 2005). Much of the research investigating the extent to which threatening information affects attentional resource allocation has been interpreted in terms of hypervigilance avoidance, which posits that perception of threat relevant stimuli in anxious individuals is characterized by initial hypervigilance and subsequent defensiv e avoidance (Mogg et al., 1997). However, more recent findings have been mixed (e.g., Bögels and Mansell, 2004; Wieser et al., 2009).

PAGE 13

13 Paralleling the results of hemodynamic imaging (Sabatinelli, 2007) and electrophysiological (see Bradley, Keil and Lang, 2012 for review) studies, there has been evidence that the steady state visual evoked potential (ssVEP; Regan, 1989) shows a similar modulation when processing emotional cont ent, compared to neutral (Müller et al . , 2008). For example, in an overlapping sti mulus display, Müller et al. (2008) observed a reduction in ssVEP amplitude to flickering (7.5 Hz) dots when pleasant and unpleasant pictures served as distractors, compared to neutral. To examine processing of specific fear cues at the level of lower tier visual cortical processing, we employed steady state visual evoked potentials, which measure the amplitude of neural activity elicited by a task relevant stimulus that is flickering at a known rate; competition by a distractor can therefore be measured as an attenuation of the task evoked processing. Because ssVEPs possess high signal to noise ratios and excellent psychometric reliability, they may be particularly suitable for studying individual differences in terms of perception or attention, for instanc e differences between high and low fearful individuals. For example, using the ssVEP, Wieser et al. (2012) observed a prominent visual competition effect for threatening faces observed only in individuals with high social anxiety. An additional advantage o f ssVEPs is that they are readily converted to a time varying measure of neural population activity in visual cortex, providing an opportunity to quantify the time course of competition for processing resources in a neurophysiologically meaningful way (Mül ler et al., 2008). By studying the time varying ssVEP amplitude for distractors and task arrays separately, it is possible to assess fluctuations in visual cortical engagement both within and between high and low fearful individuals, and across conditions.

PAGE 14

14 Capitalizing on the strengths of this approach, two experiments were designed to address a) the extent and time course of task evoked ssVEP amplitude modulation to emotional compared to neutral distractors, and b) inter individual differences regarding b oth the time course and the neurophysiological locus of distraction effects seen in observers high in snake fear. Task relevant stimuli were moving dots, flickering at 8.57 Hz, which moved randomly or with brief intervals of coherent motion; in both experi . Distractor pictures were presented in the background of the moving dot display (i.e. overlapping) and included snake pictures, as well as emotionally arousing content depic ting erotica and violence, and control pictures. We expected to replicate previous findings (e.g., Müller et al., 2008) of attenuated ssVEP amplitude for the moving dot task when emotional, compared to neutral, pictures served as distractors (Experiment 1). Thus, for instance, we predicted that both high and low fear individuals will show reduction of ssVEP amplitude when pictures of violence are presented as distractors. If anxious individuals show an attentional bias toward fear relevant stimuli, we exp ected ssVEP amplitude to show more attenuation when snake pictures served as distractors, compared to pictures of violence, whereas the opposite pattern was expected for non fearful participants. Experiment 2 1 was designed to examine the extent to which the reduction of the task evoked ssVEP signal when emotional (specifically snakes in Experiment 2b) stimuli serve as distractors is accompanied by enhancement of the visual processing of the 1 Experiment 2 has been divided into two separate chapters to discuss general competition effects in low fear participants (Experiment 2, Chapter 3) and competition effects of fear relevant stimuli in high fear, compared to low fear, participants (Experi ment 2b, Chapter 4).

PAGE 15

15 emotional stimulus. To this end, both the task relevant dot stimu lus (8.57 Hz) and the examine whether emotionally arousing, specifically fear relevant distractors in detection task. Recent work investigating competition between two simultaneously presented stimuli has resulted in two distinct patterns of attentional resource allocation: Studies supporting trade off (resource sharing) accounts of attention have demons trated costeffects associated with task relevant and task irrelevant stimuli, where emotionally arousing distractors are processed at the cost of the competing task relevant stimulus (Hindi Attar et al., 2012; Wieser et al. 2012). Alternately, Wieser et al . (2011) observed additive effects of attention, in which an increase in ssVEP amplitude when individuals high in social anxiety viewed angry faces did not interfere with the electrocortical processing of a spatially separated concurrent face. Thus, additi onal attentional resources are recruited which are not at the cost of a competing stimulus. It is important to note, however, that the majority of literature reporting additive effects of attention has been demonstrated in paradigms in which the stimuli ha ve been spatially separated (e.g., Keil et al., 2005; Wieser et al., 2011), whereas trade off effects have been r eported in spatially overlapping paradigms (e.g.; Müller et al., 2008; Hindi Attar et al., 2012). In line with recent literature demonstratin g a trade off between task and spatially overlapping distractor stimuli when viewing unpleasant distractors in healthy subjects (Hindi Attar et al., 2012), we predicted reciprocal amplitude effects in the snake fearful group. Specifically, we expected atte nuation of the dot evoked ssVEP amplitude when

PAGE 16

16 emotional, compared neutral, pictures served as distractors, as reflected by enhanced ssVEP amplitude to the flickering emotional distractors. Thus, a decrease in ssVEP amplitude to the flickering dot stimulus (8.57 Hz) and subsequent increase in ssVEP amplitude to the flickering background stimulus (12 Hz) would support a trade off effect of attention. Under a competition hypothesis, we expected ssVEP amplitude to show greater attenuation to the moving dot tas k and a greater reciprocal increase in ssVEP amplitude to the flickering background picture when emotional, compared to neutral pictures served as distractors (Experiment 2), and for this effect to be enhanced for participants in the high fear group when t he fear relevant stimulus (snake) served as the distractor (Experiment 2b) .

PAGE 17

17 CHAPTER 2 SUSTAINED COMPETITIVE BIASES TO VISUAL SNAKE FEATURES: HYPERVIGILANCE WITHOUT AVOIDANCE For several decades, clinical and translational research has examined the role of heightened attention to fear related stimuli (hypervigilance) in patients diagnosed with disorders of fear and anxiety (Öhman and Mineka, 2001). Hypervigilance, or attentional bias to threat cues, has been implicated as an important factor in the etiol ogy and maintenance of disorders in the fear and anxiety spectrum (Becker et al., 2001; see Bar Haim et al., 2007 for review). Recently, training procedures have been developed building on these findings, and have demonstrated that reducing hypervigilance has been effective in reducing symptom report in anxiety patients (Amir, et al., 2009; McNally, 2007). Studies investigating the mechanisms mediating hypervigilance have consistently found heightened sensitivity to visual features associated with specific phobic objects in paradigms as diverse as categorical perception and visual search (Kolassa et al., 2007a; Straube et al., 2007; Lipka et al., 2011; see Heeren et al., for review). Alternatively, a hypervigilance avoidance hypothesis of fear processing ( e.g., Mogg et al., 1989; Mathews, 1990;) posits a sequence in which initial hypervigilance is subsequently followed by perceptual avoidance of the fear relevant stimulus in observers high in trait anxiety. Evidence of fear related avoidance (Bögels and Man sell, 2004) has been less consistently found but has been most robust in studies of eye tracking (Mogg et al., 2000, Garner et al., 2006). For instance, Wieser at al. (2009) presented angry, happy, and neutral faces to students characterized by high or low fear of negative evaluation and found that high fear participants initially looked more at emotionally arousing faces than to neutral faces (hypervigilance), but that this pattern

PAGE 18

18 reversed in the second half of the viewing period (avoidance). Similarly, P flugshaupt and colleagues (Pflugshaupt et al., 2005) demonstrated that compared to controls, observers with spider phobia initially fixated at locations closer to spiders, but subsequently fixated at locations further away from spiders. Another approach to quantify attentional processing across time is to examine attention at the level of the visual cortex: To assess hypervigilance avoidance patterns of attention on the level of lower tier visual cortical processing, we employ the steady state visual evok ed potential (ssVEP), which has favorable properties for exploring the mechanisms by which fear related content impairs concurrent task processing on the level of sensory registration. In Experiment 1, we examined the extent to which individual differences in self reported fear of snakes modulates ssVEP amplitude across an interval in which fear relevant images serve as distractors. Task relevant stimuli were moving dots, flickering at 8.57 Hz, which moved randomly or with brief intervals of coherent motion dots. Distractor pictures were presented in the background of the moving dot display (i.e. overlapping) and included snake pictures, as well as emotionally arousing content depicting erotica and violence, as well as control pictures. We expected to replicate previous findings of attenuated ssVEP amplitude for the moving dot task when emotional, compared to neutral, pictures served as distractors. Thus, for instance, we expected both hi gh and low fear participants to show reduction of ssVEP amplitude when pictures of violence were distractors. Of central importance was ssVEP task evoked amplitude when snake pictures served as distractors: we expected ssVEP amplitude to show more attenu ation than

PAGE 19

19 pictures of violence for snake fearful participants, whereas the opposite was expected for low fear participants. To test hypotheses regarding whether hypervigilance and/or avoidance are supported by a direct neural index of time varying visual cortical engagement, the following predictions were made: If sustained hypervigilance in the presence of the fear cue presentation characterizes phobic processing (Kolassa et al., 2007; Wieser et al., 2011; Wieser et al., 2012), the data would indicate red uced ssVEP amplitude for dots superimposed over snake pictures that persists across the presentation interval for high fear participants. Alternatively, a hypervigilance avoidance hypothesis predicts that an initial reduction in ssVEP amplitude for dots ap pearing over fear relevant stimuli for high fear participants will be followed by enhanced ssVEPs, indicating perceptual avoidance. Method Participants Forty one female right handed undergraduate students at the University of Florida provided written co nsent following the guidelines proposed by the University of (20 USD) for their participation. All participants were screened for photic epilepsy. Half of the sample was selected based on a prescreening inventory given to 561 undergraduate students participating in the Introduction to Psychology course: Students scoring within the top 15% on the 30 item s n ake fear questionnaire (SNAQ; Klorman et al., 1974) were contacted and recruited into the high fear group. Participants in this group scored well above the 85 th percentile in the SNAQ (mean 20.58, SD 3.53, 85 th percentile for female students = 17). The remaining participants were unselected and reported normal levels of snake fear (SNAQ mean = 6.31, SD = 3.78). All participants

PAGE 20

20 included in the final analysis performed the coherent motion detection task at a minimal level of 60% or better (average: 72.9% correct, range: 61.68 98.3% correct), which is consistent with accura cy thresholds used in similar paradigms employing a challenging motion detection task (Müller et al., 2008; Hindi Attar et al., 2010). Three of the forty one students participating in the experiment performed the task at less than 60% correct, and were exc luded from the final analysis. The data from thirty eight participants (age range: 18 25, mean age: 18.8) with normal to corrected to normal vision were included in the final analysis. Due to incomplete responses from three participants, stimulus rating da ta was included only for thirty five participants . Stimuli and Procedure Stimuli were presented centrally on a 23 inch Samsung SyncMaster SA950 LED monitor, set at a resolution of 1680 x 1050 with a refresh rate of 120 frames per second (i.e., 8.33 ms refr esh interval). Erotic couples, neutral people at work, and mutilated human bodies composed the pleasant, neutral, and unpleasant stimulus categories, respectively. To evaluate individual differences between participant groups with respect to snake fear, sn ake pictures were added to the unpleasant stimulus set. Categories of kitten (pleasant) and cow pictures (neutral) were included to provide animal content as an added control for the snake pictures. Accordingly, each hedonic (distractor) content included a human and an animal category (pleasant: erotica, kittens; neutral: people at work, cows; unpleasant: mutilation, snakes), with 20 pictures in each subset totaling to 120 pictures. Pictures were selected from the International Affective Picture System (IA PS; Lang et al., 1997) based on normative valence and arousal ratings using the Self Assessment Manikin (SAM; Bradley and Lang, 1994) 9 point scale. Additional images

PAGE 21

21 were selected from the public domain to complete balanced human and animal picture catego ries. Valence and arousal ratings for neutral pictures from the IAPS were 6.28 and 4.15, for pleasant 7.01 and 5.84, and for unpleasant images 2.84 and 5.9, respectively. All stimuli were controlled for visual complexity, measured as jpeg size, and were ma tched for luminance using scripts from the MATLAB® image processing toolbox. Luminance was then measured for the entire screen using a Gossen (Nürnberg, Germany) luminance meter and was 80.02 cd/m 2 on average. Picture stimuli were circular in nature, and w ere cropped and adjusted such that the defining element of each picture was positioned at the center of a circle (see Figure 2 1). Each trial began with a 1 second presentation of an IAPS image with individual pixels scrambled, to avoid contamination of the ssVEP with transient responses to the luminance gradient created by stimulus onset. Next, a total of 150 yellow dots (each 0.3 x 0.3 degrees of visual angle) were superimposed upon the scrambled image for 2917 ms. The scrambled background picture was t hen replaced by either a pleasant, neutral, or unpleasant picture which remained on the screen for the duration of the trial (5834 ms; Figure 2 1). All picture stimuli were grayscale pictures subtending a viewing angle of 6.9° at a viewing distance of 170 cm. Dots were distributed randomly across pictures and flickered at a rate of 8.57 Hz. Background images and the overlapping flickering dots remained inside the circle (6.9° visual angle) at all times. 8 frames. All dots remained in continuous motion throughout the trial and each dot changed its position by 0.04 degrees in a random direction with every ssVEP cycle (i.e. 8.57 times/sec). In a random subset of 50% of the trials, 100% of the dots moved cohe rently in the same direction

PAGE 22

22 (target), and participants were instructed to respond to coherent motion events with a mouse click, as quickly and as accurately as possible. Coherent motion of the targets occurred in one of four diagonal directions (45°, 135° , 225°, 315°) at random. In an effort to produce a difficult and demanding detection task, coherent motion lasted for only 4 successive cycles of 8.57 Hz (i.e., 466.64 ms). Targets occurred unpredictably once (in 58 of the 120 trials) or twice (in 4 of the 120 trials) in a given trial, with the remaining 58 trials consisting of random movement of the dots. The first possible coherent motion event was at 1169 ms (i.e., 10 cycles) after stimulus onset and the last coherent motion event was at 7000 ms (i.e., 6 0 cycles). Targets occurring during the scrambled image were not included in the behavioral analysis. Trials with dual targets were inserted to ensure attention was directed to the task for the entire duration of the trial, and such trials were not include d in the final analysis. As a result, only single target or no target trials occurring during picture presentation were included in the final analysis. Participants were instructed to click the mouse as soon as coherent motion was detected. Each trial last ed for 9751 ms, with inter stimulus intervals randomly varying between 3000 and 5000 ms. Fixation was facilitated by presenting a white fixation dot at the center of the screen (i.e. circle). Prior to the experiment, all participants completed the SNAQ a nd performed 15 practice trials to become familiar with the stimulation and task. In the training session, 8 of the trials contained a target (coherent motion of the flickering dots), with one of those targets being a double target. Following the experimen t, participants rated each of the 120 affective picture stimuli used in the experiment in pseudo randomized order on the

PAGE 23

23 dimensions of affective valence and arousal, using a paper and pencil version of the SAM. EEG Recording Electrophysiological data were collected from the scalp using a 257 sensor net Keil et al. 2014). The EEG was collected cont inuously with a sampling rate of 250 Hz (16 bit resolution) and were band pass filtered online in the 0.1 90 Hz frequency range using a hardware elliptical filter. The vertex electrode (Cz) was used as the recording reference. Further processing and filter ing was performed offline. EEG Reduction and Analyses Continuous data were low pass filtered offline at a frequency (3dB point) of 40 Hz (12th order Butterworth filter with 24 dB / octave roll off implemented in MATLAB®) prior to segmenting. Single epochs of 9200 ms in length (400 ms pre and 8800 ms post dot onset) were then extracted from the continuous EEG signal. Using the artifact rejection procedure proposed by Junghöfer et al. (2000) trials with artifacts were identified based on the distribution of statistical parameters of EEG epochs (absolute value, standard deviation, maximum of the differences) and were extracted across time points and channels. Sensors contaminated with artifacts were replaced by statistically weighted, spherical spline interpo lated values, and a maximum of 25 channels was set for interpolation. Trials with spatially concentrated bad sensors were excluded as well, as these would invalidate interpolation for approximated sensors (see Junghöfer et al., 2000, for a more detailed de scription). As a result, each of the six picture conditions retained an average of 14 trials (SD = 0.28), which did not differ by condition ( p > 0.59).

PAGE 24

24 To the extent that the signal of interest (the ssVEP) is concentrated in one specific frequency band, pr evious work has established that stable estimates of the time varying ssVEP amplitude are possible with trial counts between 10 and 20 (e.g., Keil et al., 2003; Wieser et al., 2011). To ensure satisfactory signal quality, we submitted each ta to the circular T square statistic (Mast and Victor, 1991), which formally tests the temporal stability of the entrained brain signal at a given driving frequency. To this end, the entire ssVEP viewing epoch for each experimental condition was segmented in non overlapping epochs containing 4 cycles each, and then submitted to the circular T square algorithm. This algorithm can be used to test for the presence of an evoked signal at the frequency of interest, taking both phase and amplitude information in to account. All participants included in this study showed reliable (defined as p < 0.05 for the Chi square distributed circular T square at site Oz and its nearest neighbors) evoked oscillations at the driving frequency. This suggests satisfactory signal to noise ratios with the trial counts available in this experiment. Steady state Visual Evoked Potential Analyses Artifact free epochs of the voltage data were averaged for the six picture distractors , by group. Time varying amplitude at the stimulation fr equency of 8.57 Hz was extracted by means of a Hilbert transformation of the time domain averaged data using in house MATLAB® scripts: Data were filtered with a 10 th order Butterworth band pass filter having a width of .5 Hz around the center frequency of 8.57 Hz. Then the time varying amplitude was extracted as the complex conjugate of the band pass filtered signal and the Hilbert transformed analytic signal, for each time point. Data were then temporally smoothed applying a linear moving average, with a w indow length of

PAGE 25

25 420 ms, corresponding to the time resolution of the time varying amplitude, which was 421 ms full width at half maximum of the filter's impulse response. Statistical Analyses : Behavioral data and SAM ratings The percentage of correctly ide ntified targets (hits) and false alarms were calculated for each distractor condition and participant. Button presses occurring after 150 ms post target onset (coherent motion) were accepted as correct detections. False alarms were calculated as the percen tage of trials in which a response was made (e.g., clicking the mouse) in the absence of a coherent motion target. Behavioral sensitivity, a measure of detection sensitivity (Macmillan and Creelman, 2004), was calculated by subtracting the number of false alarms from the number of hits for each condition comparison. Differences among conditions were evaluated by means of omnibus repeated measures analysis of variance (ANOVA) that included distractor hedonic content (pleasant, neutral, unpleasant), semantic category (humans, animals) and fearfulness (high snake fear, low snake fear). SAM ratings were averaged across participants for each of the six picture conditions by fearfulness, and were tested using planned comparisons (paired t tests) separately for val ence and arousal. Statistical Analysis: ssVEP time course Permutation controlled t tests for each sampling point and scalp location were conducted to assess the time course of ssVEP amplitude using the sample by sample data which has the best temporal res olution for assessing time varying ssVEP amplitude. In these analyses, t tests were calculated at each EEG sensor and sampling point for each distractor, which compared ssVEP amplitude between high snake fear and control participants. Significance threshol ds for each comparison were calculated by computing 500 electrode by time point matrices of t tests based on random

PAGE 26

26 permutations of the existing data, i.e., with group membership shuffled. The statistic for each topography and time point entered a referen ce distribution, whose 2.5% tails served as the criterion for statistical significance (see McGinnis and Keil, 2011 for a similar procedure). Comparisons reaching significance (as determined by the permutation corrected t tests) indicated scalp locations a nd sampling points at which the high fear group was significantly different from the control group, for each condition separately. To assess the full factorial statistical model of the mixed design of the present study, and to complement time course anal yses, ssVEP amplitudes were averaged in an occipito parietal cluster of electrodes including Oz and its 24 nearest neighbors (EGI sensors: 116 119 123 127 135 139 147 150 157 159; see Figure 2 2), resulting in posterior regional mean amplitudes. These ampl itudes were then subjected to a z transformation to address inter individual variability in amplitude. To evaluate modulations in ssVEP amplitude for different distractors, ssVEP amplitude was deviated from a baseline period ( 1500 300 ms pre stimulus onset ) in four time windows following picture onset (T1: 380 1100 ms, T2: 1100 2100 ms, T3: 2100 3100 ms, T4: 3100 4100 ms). These time windows were selected based on visual inspection of the data, and to capture the temporal dynamics related to ssVEP amplitud e modulation of background distractors, across the viewing epoch. The initial 700 ms time window was selected to capture immediate effects related to picture onset (expected based on earlier research; Müller et al., 2008), and was followed by three 1000 ms time windows selected to sample any effects related to sustained picture viewing. A linear mixed model analysis (implemented in SPSS ©) was conducted with fixed effects of distractor (pleasant,

PAGE 27

27 neutral, unpleasant), semantic category (human, animal) and t ime, and subject as a random variable, nested within fear group. For effects involving repeated measures, the Huynh Feldt procedure was used to correct for violations of sphericity. Results Behavioral data and SAM ratings Performance on the coherent m otion detection task was affected by the hedonic properties of the distractor ( F 1,33 = 4.33, p = 0.017), with a significant quadratic trend in behavioral sensitivity indicating that, compared to neutral distractors, emotionally engaging pictures (both plea sant and unpleasant) prompted poorer performance ( F (1, 36) = 5.411, p = 0.024; see Figure 2 3). As expected, high fear participants rated pictures of snakes as more unpleasant ( t 33 = 3.89, p < 0.0001) and more arousing ( t 33 = 2.99, p = 0.005) than did low fear participants. High fear participants also rated pictures of erotica ( t 33 = 2.28, p = 0.03) and mutilation ( t 33 = 3.23, p = 0.004) as slightly more unpleasant than did the low fear participants, and rated pictures of mutilation ( t 33 = 2.29, p = 0.03) as more arousing, than did low fear participants . ssVEPs The flickering dots in the task relevant stream reliably evoked steady state responses at the expected frequency of 8.57 Hz, as illustrated in Figure 2 2, with the greatest overall ssVEP amplitudes across all experimental conditions occurring for sensor Oz and its nearest neighbors. The grand mean time varying energy of the signal over occipital sensors as quantified by the Hilbert transform is shown in Figure 2 4. As illustrated in Figure 2 5, permu tation t tests conducted on each individual sampling point

PAGE 28

28 (critical value of t = 2.8 at the p perm < 0.05; dashed line) identified when ssVEP amplitudes were statistically different between high and low fear participants for each distractor type. As illust rated in Figure 2 5, high fear participants differed significantly in ssVEP amplitude from the low fear group only when snake pictures served as distractors, and this difference persisted across the distractor interval. Differences in ssVEP amplitude betwe en high fear participants and low fear individuals were most pronounced later in the distractor interval, which is inconsistent with a hypothesis of initial vigilance followed by avoidance. A mixed model linear analysis assessed differences in ssVEP ampli tude as a function of the background distractor for each group. Significant modulation of ssVEP amplitude varied as a function of distractor, F (2,828) = 18.4, p < 0.0001, semantic category, F (1, 828) = 3.9, p = 0.048 and time, F (3,828) = 4.9, p = 0.002. R eplicating previous research, over all distractor contents, dot evoked ssVEP amplitudes were attenuated when emotionally evocative (either pleasant or unpleasant), compared to neutral pictures, served as distractors (Pleasant vs. Neutral, p = 0.002; Unplea sant vs. Neutral, p < 0.0001; Figure 2 6). These main effects were qualified by a three way interaction of distractor, semantic category, and fearfulness, F (2,828) = 6.5 , p = 0.002, indicating that ssVEP amplitude was differentially modulated as a funct ion of distractor content and semantic category for high and low fear participants. Follow up tests of this interaction assessed effects of fearfulness and semantic category separately for each distractor type. Most importantly, a significant interaction o f fearfulness and semantic category, F (1, 252) = 12.3, p = 0.001 indicated that for high fear participants, snake distractors prompted a

PAGE 29

29 larger reduction in ssVEP amplitude compared to mutilation distractors (t18 = 2.15, p = 0.045), as illustrated in Figur e 2 7. The low fear group did not show this difference, t < 0.6, n.s. For pleasant pictures, as expected, high and low fear participants did not differ in their pattern of ssVEP modulation as a function of distractor, p > .3. For neutral pictures, a signif icant interaction of fearfulness and semantic category F (1, 252) = 4.7, p = 0.03 indicated that pictures of people at work showed slightly less ssVEP reduction than cows (p = 0.056) for high fear participants. Means for pleasant and neutral category compar isons can be found in Table 2 1. Discussion Experiment 1 investigated the extent to which ssVEP amplitude and the associated time course of visual competition was modulated by task irrelevant fearful stimuli and a foreground task. For all participants, a reduction in ssVEP amplitude for task related stimuli was found when task irrelevant, but emotionally arousing distractors were present, consistent with previous studies (e.g., Müller et al., 2008; (Hindi Attar et al., 2010; Müller et al., 2008)). This su pports the notion that emotionally arousing cues compete for limited resources at the level of the visual cortex (Keil et al. 2005; Hajcak et al., 2013). Importantly, for participants reporting high levels of snake fear, the reduction in task related ssVEP amplitude was greatest when snakes served as distractors, even when compared to other aversive content. In addition, there was no evidence that initial hypervigilance shifts to perceptual avoidance later in the interval for high fear participants. Rather, the reduction in task evoked ssVEP amplitude was most pronounced later in the distractor interval, consistent with sustained hypervigilance. Preferential and sustained cortical processing of fear relevant cues in the absence of avoidance has been demons trated in several studies assessing attentional

PAGE 30

30 capture by phobia relevant stimuli (McTeague et al., 2011, Wieser et al., 2011, Wieser et al., 2012), whereas patterns of hypervigilance avoidance have been less consistent, and most robust in eye tracking st udies (Pflugshaupt et al., 2005; Wieser et al., 2009). To the extent that ocular control is a complex mechanism which involves widespread cortical communication along with coordinated activity in several cortical and subcortical structures (Sommer and Wurt z, 2008), it is conceivable that a dynamic sequence of hypervigilance and avoidance is specific to eye movements. Importantly, oculomotor data do not inform on covert shifts of attention, whereas the ssVEP technique is sensitive to covert attention process es, providing a potential explanation for sustained hypervigilance in the absence of avoidance in studies using the ssVEP. Previous studies examining competition effects in healthy observers have shown that modulation of perceptual arousal begins around 4 00 ms after stimulus onset and persist for a few hundred milliseconds (Müller et al., 2008). This parallels the pattern of ssVEP amplitude modulation demonstrated by the low fear group in the present experiment (see Figure 2 4, right). Group differences b etween anxious and non anxious observers, however, emerged later in the viewing epoch. These differences in ssVEP amplitude began 1100 ms after picture onset, and reliably indicated that the greatest amplitude reduction occurred specifically in the snake c ondition for fearful observers, an effect that persisted as long as the fear relevant image remained on screen. Thus, while all participants showed early sensitivity to distraction, only snake fearful observers displayed sustained interference, induced spe cifically by the snake pictures. This finding is consistent with work by Wieser et al. (2012), who observed a prominent visual competition effect for threatening faces, observed only in individuals with high social

PAGE 31

31 anxiety. Specifically, these authors foun d that participants high in social anxiety showed diminished visual engagement for task stimuli that were superimposed on an angry, but not neutral or happy faces, which was maintained throughout the viewing period (~3000 ms). Thus, automatic attentional b ias toward a threatening stimulus seems to reliably persist as long as the threat stimulus remains visible. In the clinical and translational research literature, hypervigilance in the context of fear relevant cues is seen as dysfunctional attentional and perceptual processes that may cause or contribute to the maintenance of phobias (McNally, 2007). Thus, objective and quantitative measures of selective attention to threat have important applications in diagnosis and treatment of anxiety disorders (Bö gels and Mansell, 2004; Amir et al., 2009) as well as other disorders that involve the dysregulation of affective processing (Kemp et al., 2004; Donaldson et al., 2007). To the extent that the ssVEP technique employed here is sensitive to electrocortical m odulation related to inter individual differences (e.g., differential amplitude modulation of a fear relevant stimulus in a fearful participant versus a control), applications in the clinical research arena are conceivable: In assessment, these methods cou ld be used to objectively identify patients with dysfunctional attentional resource allocation to fear related stimuli. In addition, inter individual differences in the time course of hypervigilance versus perceptual avoidance could be a novel way to assign patients to individualized treatment and in predicting treatment outcome. Specifically, modulatory responses of distraction by or interference of fear related stimuli over the course of treatment may be examined and quantified using the ssVEP, which may h ave implications for therapeutic treatments aiming to direct attention away from a threat cue. It should be noted that replication of these

PAGE 32

32 findings in a clinically anxious population is an essential next step before this measure may be used as a reliable diagnostic tool. However, previous studies assessing emotional reactivity to fear relevant stimuli have found significant differences using both fMRI (Sabatinelli et al., 2005) and EEG (Wieser et al., 2012) in sub clinical populations. In future research, anxiety patients diagnosed with small animal phobia may aid in supporting and generalizing these findings. In conclusion, the results from Experiment 1 provide support for the hypothesis that task irrelevant but emotionally engaging stimuli act as strong competitors, interfering with the visual processing of a concurrent task stimulus. For high fear participants, task evoked ssVEP reductions were larger when fear cues served as distractors, compared to other aversive content, and compared to low fear part icipants. Moreover, the attenuation of task evoked ssVEPs when fear cues were present in the array was sustained across the temporal interval for participants reporting high snake fear, consistent with a hypothesis of sustained hypervigilance .

PAGE 33

33 Table 2 1. Mean ssVEP amplitude ( change from baseline ) for pleasant (erotica, cat) and neutral (work, cows) distractors for high and low fear participants. Fearfulness Semantic Category Low fear High fear Erotica 0.15 (0.09) 0.32 (0.06) Cats 0.24 (0.09) 0.30 (0.07) Work 0.09 (0.08) 0.20 (0.08) Cows 0.12 (0.08) 0.15 (0.07) Note: Parentheses denote the standard error of the mean.

PAGE 34

34 Figure 2 1 . Time sequence for a single trial: intervals of coherent motion could occur be tween 1169 7000 ms (target window, gray box) post stimulus onset. Each trial lasted 9751 ms with a variable 3 5 second inter trial interval.

PAGE 35

35 Figure 2 2 . A) Grand average time domain of the ssVEP signal averaged across al l stimulus conditions and participants, plotted at sensor Oz. B) top: Grand average topographical distribution of ssVEP amplitude across all participants and conditions in the time window between 2200 and 4200 ms (red box) after stimulus onset. B) bottom: Grand average of the frequency spectra across all participants and conditions, demonstrating a reliable peak at the driving frequency of the flickering dots, 8.57 Hz. C) Representation of electrode placement; sensors used in the cluster analyses are highli ghted by red circles.

PAGE 36

36 Figure 2 3 . Group averages of behavioral performance (hits false alarms) for high fear (dark gray) and low fear (light gray) participants for pleasant, neutral, and unpleasant distractor contents.

PAGE 37

37 Figure 2 4. Hilbert transformed data for high and low fear participants for all conditions, averaged over a cluster of occipito parietal electrodes; averaged amplitudes were calculated for sequential 600 ms time bins. The light gray panel indicates pr esentation of the flickering dots and scrambled image (2900 ms before stimulus onset) and the dark gray panel indicates simultaneous presentation of the flickering dots and the background image (total duration of 5800 ms). Note different scales.

PAGE 38

38 Figure 2 5. Permutation corrected waveforms for unpleasant (left), neutral (center), and pleasant (right) conditions. The critical value for conditions reaching significance at the p perm = 0.05 level is indicated by the dotted black line ( t = 2.8). Conditions ab ove the significance level indicate group differences. The time scale refers to the simultaneous onset of the dots and the background picture (dark gray box).

PAGE 39

39 Figure 2 6. Mean ssVEP amplitude (calculated as the change from baseline) averaged across hig h and low fear participants, for pleasant, neutral, and unpleasant distractors.

PAGE 40

40 Figure 2 7. Mean ssVEP amplitude (calculated as the change from baseline) for mutilation and snake categories for high fear (dark gray) and low fear (light gray) particip ants.

PAGE 41

41 CHAPTER 3 COMPETITION EFFECTS OF EMOTIONALLY AROUSING DISTRACTORS ON A CONCURRENT TASK An interesting question prompted by the results of Experiment 1 is the extent to which the reduction of the task evoked ssVEP signal when emotionally arousing sti muli serve as distractors is accompanied by enhancement of the visual processing of those stimuli, resulting in a trade off of attention. The ssVEP offers a unique solution to address hypotheses concerning the relative amount of cortical processing of conc urrently presented stimuli: when two stimuli flickering at different rates are simultaneously presented in an overlapping spatial array, the electrocortical signature of the processing of each separate stimulus can be calculated and separated in the freque ncy domain, providing a continuous measure of visual resource allocation to a specific stimulus amid competing visual cues. Frequency tagging, then, allows researchers to quantify responses to multiple visual objects separately, and as a result, is ideally suited for the investigation of competition between objects simultaneously present in the field of view. To examine whether attention involves inhibition in addition to enhancement of cortical neural responses to sensory input, Chen et al., (2003) used magnetoencephalography (MEG) to measure frequency tagged steady state visual evoked responses of two superimposed images. The authors concluded that the power of the MEG signal could be either increased or decreased depending on the attention condition, ev en when the visual input remained unchanged. Although these results could not definitively establish inhibitory mechanisms of attention, they served as a basis for the discussion of possible trade off effects of attention as quantified by the ssVEP. Import antly, Keitel et al. (2010) have demonstrated that competition in

PAGE 42

42 frequency tagging paradigms is not limited to a specific range or combination of frequencies. A growing body of research investigating competition in visual cortex converges to suggest that there may be strong biases toward a target stimulus resulting in a relative loss of neural signal to competing stimuli in a given location or across time (Müller and Hübner, 2002; Wang, Clemenz, and Keil, 2007; Müller, Andersen, Keil, 2008; Hindi Attar, An dersen, Müller, 2010; Hindi Attar and Müller, 2012). It is well established that the ssVEP is modulated by the emotional content of a visual stimulus (Keil et al., 2003) and increases with attention (Müller et al., 1998; Keil et al., 2008). Thus, frequen cy tagging of both task relevant and distractor stimuli represents a promising avenue for examining hypotheses regarding the allocation of attentional resources to stimuli differing in motivational salience. Experiment 2 was designed to assess potential c ompetition effects of attention related to processing emotionally arousing distractors in a frequency tagging paradigm. If emotionally arousing distractors capture and hold attentional resources, then the time varying amplitude of the ssVEP evoked by the m otion detection task is expected to show significant reduction when viewing emotionally arousing, compared to neutral, resources from the detection task, we predict an increas e in ssVEP amplitude to the emotionally background distractors (at the cost of the flickering dot stimulus), supporting a trade off effect of attention . Methods Participants Twenty three female right handed undergraduate students at the University of Flor ida provided written consent following the guidelines proposed by the University of

PAGE 43

43 (20 USD) for their participation. All participants were screened for photic epilepsy. Participants included in the final analysis performed the coherent motion detection task at a minimal level of 60% or better (average: 75.9% correct, range: 61.1 94.6% correct). The data from twenty three participants (age range: 18 27, mean age: 19.2) wi th normal to corrected to normal vision were included in the final analysis, with the exception of three students who performed the motion detection task at less than 60% correct, and were excluded only from the behavioral analysis. Due to incomplete respo nses from four participants, stimulus rating data was included for nineteen participants . Stimuli and procedure Experiment 2 was designed to examine the extent to which the reduction of the task evoked ssVEP signal when emotional stimuli serve as distra ctors is accompanied by electrocortical enhancement of the emotional stimulus. To this end, we have frequency tagged both the task relevant dot stimulus (8.57 Hz) and the distractor processing resources from the foreground detection task. Task irrelevant distractor stimuli were identical to the IAPS images used and described in Chapter 2 for Experiment 1. For the present experiment, the six distractor contents described in Exper iment 1 were collapsed across pleasant (erotica and cats), neutral (people at work and cows) and unpleasant (human mutilation and snakes) distractor contents, with 40 pictures in each combined semantic category of picture distractors totaling to 120 pictur es. Each trial began with a 1 second presentation of an IAPS image with individual pixels scrambled, to avoid contamination of the ssVEP with transient responses to the

PAGE 44

44 luminance gradient created by stimulus onset. Next, a total of 150 yellow dots (each 0 .3 x 0.3 degrees of visual angle) were superimposed upon the scrambled image for 1750 ms. The scrambled background picture was then replaced by either a pleasant, neutral, or unpleasant picture which remained on the screen for the duration of the trial (64 16 ms; Figure 3 1). The first possible coherent motion event was at 1170 ms (i.e., 10 cycles) after onset of the flickering dots and the last coherent motion event was at 7000 ms (i.e., 60 cycles). Each trial lasted for 10,166 ms. The remaining informati on regarding stimulus presentation and experimental procedure was identical to the methods described in Chapter 2 for Experiment 1 . EEG Recording Electrophysiological data were collected in the same laboratory, using the same equipment, as Experiment 1. EE G data were collected from the scalp using a 257 sensor net (EGI, Eugene, OR) for which scalp impedance for each sensor was kept rate of 250 Hz (16 bit resolution) and w ere band pass filtered online in the 0.1 90 Hz frequency range using a hardware elliptical filter. The vertex electrode (Cz) was used as the recording reference. Further processing and filtering was performed offline. EEG Reduction and Analyses Continuous data were low pass filtered offline at a frequency (3dB point) of 40 Hz (12th order Butterworth filter with 24 dB / octave roll off implemented in MATLAB®) prior to segmenting. Single epochs of 10,500 ms in length (400 ms pre and 10,100 ms post trigger o nset) were extracted from the continuous EEG signal. Artifact rejection procedures were identical to those implemented in Experiment 1. Each of the three distractor contents retained an average of 14.48 trials (SD = 0.46), which did not differ

PAGE 45

45 by condition (p > 0.93). As in Experiment 1, all participants included in Experiment 2 showed reliable (defined as p < 0.05 for the Chi square distributed circular T square at site Oz and its nearest neighbors) evoked oscillations at the driving frequency, suggesting satisfactory signal to noise ratios with the trial counts available in this experiment. Steady state Visual Evoked Potential Analyses Artifact free epochs of the voltage data were averaged for each of the three picture distractors. Time varying amplitude a t the stimulation frequencies of 8.57 Hz (dots) and 12 Hz (pictures) was extracted by means of a Hilbert transformation of the time domain averaged data using in house MATLAB® scripts: Data were filtered with a 10 th order Butterworth band pass filter havin g a width of .5 Hz around the center frequency of 8.57 Hz and 12 Hz for the dot and picture conditions, respectively. Then the time varying amplitude was extracted as the complex conjugate of the band pass filtered signal and the Hilbert transformed analyt ic signal, for each time point. Data were then temporally smoothed applying a linear moving average, with a window length of 400 ms, for the data between 2000 10100 ms. Statistical Analyses: Behavioral data and SAM ratings The percentage of correctly ident ified targets (hits and correct rejections) was calculated for each distractor condition and participant. Mean target detection rates were calculated for early (1580 4284 ms) and late (4290 7000 ms) time windows to assess changes in behavioral performance during early versus late periods of the picture viewing epoch. Only trials in which a coherent motion event occurred during the simultaneous presentation of a background distractor were included in the early late analysis (see Figure 3 1); double targets a nd targets occurring during the scrambled

PAGE 46

46 period were omitted. Behavioral performance was first evaluated by means of omnibus repeated measures analysis of variance (ANOVA) with distractor (pleasant, neutral, unpleasant) and time (early, late) as factors, and followed by planned comparisons (paired t tests). SAM ratings were averaged across participants for each of the three picture conditions, and were tested using planned comparisons separately for valence and arousal. Statistical Analysis: ssVEP time cou rse ssVEP amplitudes were averaged in an occipital cluster of electrodes including Oz and its 12 nearest neighbors (EGI sensors: 117 118 125 126 127 136 137 138 139 147 148 149 150), resulting in occipital regional mean amplitudes (Figure 3 2). To evaluat e modulation of the ssVEP signal across time, ssVEP amplitude was deviated from a baseline period (2000 2600 ms) in three time windows following picture onset (T1: 3200 4400 ms, T2: 4600 5800 ms, T3: 6000 7800 ms). These time windows were selected based o n visual inspection of the data, and to capture the temporal dynamics related to ssVEP amplitude modulation across the viewing epoch. The initial 1200 ms time window was selected to capture immediate effects related to picture onset (expected based on the results of Experiment 1 and findings from previous research, e.g., Müller et al., 2008), and was followed by two additional time windows selected to sample any effects related to sustained picture viewing. Coherent motion detection (dot) task Initial anal yses were conducted for each of the flickering stimuli in isolation to assess ssVEP amplitude modulation of the task relevant dot stimulus and task irrelevant picture stimulus with respect to the hedonic content of the distractor stimulus.

PAGE 47

47 For the flickeri ng dot task, a repeated measures ANOVA was conducted with distractor (pleasant, neutral, unpleasant), and time (baseline, T1, T2, T3) as factors . Picture Distractors A repeated measures ANOVA was conducted for the flickering picture stimulus with distracto r (pleasant, neutral, unpleasant), and time (baseline, T1, T2, T3) as factors. Paired t tests were used to follow up significant interactions. Competition analysis To quantify neural competition between the flickering dot and picture stimuli, a competition index was calculated that represented the difference in ssVEP power measured for the flickering dot task and picture distractor. The mean ssVEP amplitudes were averaged across participants for each distractor content and each time point, and was deviated from the baseline separately for the driving frequency associated with the dot task and the distractor picture. For the coherent motion task, this measure was typically negative (i.e. less power than when no picture distractor was present); for the picture distractor, this measure was typically positive (i.e. more power than for the scrambled (baseline) picture). Thus, by adding the absolute values of these two measures, this competition index estimates the total difference in steady state power between the flickering dot task and picture distractors. These measures were analyzed in a repeated measures ANOVA with factors of distractor (pleasant, neutral, unpleasant) and time window (T1, T2, T3). Significant effects were followed by repeated measures ANOVA s retaining the factor of distractor, conducted separately for each time point. Effects in those models were followed by paired samples t tests.

PAGE 48

48 For all effects involving repeated measures, the Huynh Feldt procedure was used to correct for violations of s phericity. Results Behavioral data and SAM ratings Interference of emotionally arousing background pictures on the coherent motion detection task was reflected in the behavioral performance data, as illustrated in Figure 3 3. Behavioral performance was s ignificantly worse for early versus late occurrences of coherent motion events (Time, F (1,19) = 13.31, p = 0.002). Task performance was also significantly worse for pleasant ( t 19 = 2.63, p = 0.017) and neutral ( t 19 = 2.84, p = 0.011) distractors occurring in the early time window, compared to the late time window. However, performance was not significantly different between early and late time windows for unpleasant stimuli ( t < 1.5, n.s.), suggesting that while participants were better able to identify per iods of coherent motion later in the viewing period for pleasant and neutral conditions, unpleasant pictures consistently interfered with behavioral performance throughout the duration of the trial. Using the 9 point SAM scale, participants rated unpleasa nt ( t 18 = 7.62, p < 0.0001) and pleasant ( t 18 = 5.87, p < 0.0001) distractors as significantly more arousing compared to neutral distractors. There was no significant difference in arousal ratings between pleasant and unpleasant distractors ( t < 1.6, n.s.) . Valence ratings differed significantly between each distractor content ( t 18 = 4.61 [neutral vs. pleasant], t 18 = 11.42 [neutral vs. unpleasant], and t 18 = 14.45 [pleasant vs. unpleasant], p < 0.0001 for all comparisons), with unpleasant pictures being ra ted as most unpleasant .

PAGE 49

49 ssVEPs The flickering dot task and picture distractors reliably evoked steady state responses at the expected frequencies of 8.57 Hz and 12 Hz, respectively, with the greatest overall ssVEP amplitudes across all experimental conditi ons occurring for sensor Oz and its nearest neighbors (Figure 3 2, A). The grand mean time varying energy of the signal over occipital sensors as quantified by the Hilbert transform is also shown in Figure 3 4 for both the dot and the picture stimuli . Coh erent motion detection (dot) task The task relevant flickering dot stream was affected by the hedonic content of the distractor stimulus (Distractor Content, F (2,44) = 3.507, p = 0.039), with a significant quadratic trend indicating that emotionally arousi ng distractors (both pleasant and unpleasant), resulted in a reduction in ssVEP amplitude to the flickering dots ( F (1,22) = 7.926, p = 0.01), compared to neutral distractors (Figure 3 5, gray circles). The ssVEP amplitude of the flickering dot stimulus was also affected by time ( F (3,66) = 5.373, p = 0.012), for which an overall reduction in ssVEP amplitude was observed during the picture viewing interval, compared to the baseline segment . Picture distractors The ssVEP amplitude of the task irrelevant flick ering picture stream increased reliably across time ( F (3,66) = 6.263, p = 0.01) and was modulated by the content of the flickering distractor (Time x Distractor ( F (6,132) = 3.233, p = 0.005)). This interaction was driven by significantly smaller ssVEP ampl itude for pleasant distractors compared to unpleasant distractors ( t 22 = 3.312, p = 0.003) in the first time window (T1, 3200 4400 ms; Figure 3 5, black squares). No other comparisons reached significance ( t < 1.832, n.s.) .

PAGE 50

50 Competition analysis A repeated measures ANOVA assessed differences in ssVEP amplitude relative to baseline. Figure 3 4 (inset) illustrates that, when the distractor picture was present, ssVEP amplitude decreased for the flickering dot task (left panel) and increased for the flickering picture distractor (right panel; Tagged Stimulus, F (1,22) = 5.3, p = 0.031), demonstrating the expected suppression of ssVEP amplitude for the motion detection task and reciprocal enhancement of ssVEP amplitude to the flickering distractor, compared to bas eline (scrambled image). A three way interaction of tagged stimulus (dot, pictures), distractor content and time ( F (6,132) = 2.385, p = 0.032), indicated that ssVEP amplitude varied as a function of both distractor type and tagged stimulus, across the pre sentation interval. The three way interaction was therefore followed by analysis of the competition index, which quantified the visual cortical activity evoked by each of the two concurrent flickering stimuli (see Table 3 1). The ANOVA confirmed main eff ects of distractor content ( F (2,44) = 6.366, p = 0.004) and time ( F (2,22) = 121.168, p < 0.0001). Follow up ANOVAs revealed a main effect of distractor content in both T1 ( F (2,44) = 4.839, p = 0.016) and T2 ( F (2,44) = 4.932, p = 0.012) time windows (Figure 3 6); there was no difference at T3 ( F < 1.86, n.s.). In T1, the main effect of distractor was due to greater competition for unpleasant ( t 22 = 3.130, p = 0.005) compared to neutral distractors (pleasant was not different from neutral, t < 1.5, n.s.). The greatest amount of competition was observed in T2, where both pleasant ( t 22 = 2.675, p = 0.014) and unpleasant ( t 22 = 2.612, p = 0.016) distractors resulted in greater levels of competition between the two flickering stimuli, compared to neutral distracto rs. As illustrated in Figure 3 6, these competition effects, which are largest when a reduction in ssVEP

PAGE 51

51 amplitude to the flickering dot stimulus is accompanied by an increase in ssVEP amplitude to the distracting picture stimulus, suggest a trade off of a ttentional processing resources that change across time and are modulated by emotional content . Discussion Experiment 2 was designed to investigate the extent to which a reduction of the task evoked ssVEP signal when emotionally arousing stimuli serve as d istractors is accompanied by a reciprocal electrocortical enhancement of the emotional stimulus, as afforded by frequency tagging. The pattern of ssVEP amplitude modulation demonstrated here is consistent with the results of Experiment 1 and previous liter ature (e.g., Müller et al., 2008; Hindi Attar et al., 2010), reporting a reduction in ssVEP amplitude to the task relevant stream while emotionally arousing pictures were presented as distractors. Competitive effects of the overlapping dot and picture st ream revealed cost effects for the motion detection task when unpleasant pictures were presented as distractors in the time window between 3200 4400 ms, where an increase in ssVEP amplitude to the flickering picture stimulus was at the cost of (i.e., a red uction in) ssVEP amplitude to the flickering dot stimulus (Figure 3 5, T1 interval). Cost effects were generalized to all emotionally arousing content during the second time window, between 4600 5800 ms, where the greatest amount of competition was evident for conditions in which pleasant and unpleasant pictures, compared to neutral, served as distractors. Competition between emotionally arousing pictures and moving dots began almost immediately after picture onset (450 ms post picture onset) and persisted for an additional 2600 ms. This temporal pattern is consistent with the findings from Wieser et al. (2012), who demonstrated a sustained perceptual bias toward threatening faces in

PAGE 52

52 observers with social anxiety. In line with published literature in health y observers (Hindi Attar et al., 2010; Müller et al., 2008), interference by emotional distractors was also reflected in poorer behavioral performance for coherent motion events occurring earlier versus later in the trial sequence .

PAGE 53

53 Table 3 1. Average s sVEP amplitude values used to calculate competition index. Mean ssVEP Amplitudes Baseline corrected Means Competition Index Dots, 8.57 Hz BSL T1 T2 T3 Mean BSL T1 T2 T3 Absolute Deviation from BSL T1 T2 T3 Pleasant 0.63 0.52 0.57 0.56 0.64 0.12 0.07 0.07 | P d + P p | 0.16 0.18 0.18 Neutral 0.64 0.57 0.61 0.58 0.06 0.03 0.06 | N d + N p | 0.12 0.10 0.14 Unpleasant 0.64 0.51 0.57 0.55 0.12 0.07 0.09 | U d + U p | 0.21 0.17 0.19 Pictures, 12 Hz BSL T1 T2 T3 Mean BSL T1 T2 T3 Pleasant 0.36 0.39 0.47 0.46 0.35 0.04 0.11 0.11 Neutral 0.36 0.41 0.43 0.43 0.05 0.08 0.08 Unpleasant 0.34 0.44 0.46 0.46 0.08 0.1 0.1 Note. Average ssVEP amplitude values used to calculate a competition index that quantified how the visual cortical activity evoked by the two flickering stimuli interacted for each distractor content and time window. Mean ssVEP amplitudes were deviated from the baseline and averaged across participants for each tagged stimulus (dots, pictures), distrac tor content (pleasant, neutral, unpleasant), and time point (baseline, T1, T2, T3). The resulting competition indices are plotted in Figure 3 6 .

PAGE 54

54 Figure 3 1. Time sequence for a single trial: intervals of coherent motion could occur between 1169 7000 ms (target window, gray box) post stimulus onset. Each trial lasted 9751 ms with a variable 3 5 second inter trial interval. Stimuli were overlapping, and flickered at 8.57 Hz (dots) and 12 Hz (picture).

PAGE 55

55 Figure 3 2. Grand averaged frequency spectra across all (low fear) participants and conditions, demonstrating a reliable peak at the driving frequency of the flickering dots, 8.57 Hz and also the flickering pictures, 12 Hz.

PAGE 56

56 Figure 3 3. Behavioral performance is represented as percent correct for early (1580 4284 ms) and late (4290 7000 ms) coherent motion events occurring during a given trial for pleasant (gray), neutral (white) and unpleasant (black) distractors.

PAGE 57

57 Figure 3 4. Grand average Hilbert transform (gray waveform) across all stimulus cond itions for the flickering dot (left) and picture (right) stimuli. Gray boxes indicate the three time windows of interest, including the baseline segment (bsl: 2000 2600 ms, T1: 3200 4400 ms, T2: 4600 5800 ms, T3: 6000 7800 ms). Onset of the flickering pict ure stimulus is indicated by the vertical dotted black line (3750 ms). Insets: Average ssVEP amplitude modulation for the flickering dot (left) and picture stimuli (right) for pleasant (blue), neutral (black) and unpleasant (red) distractor contents, acros s the four time windows of interest.

PAGE 58

58 Figure 3 5 . Average ssVEP amplitude for flickering dot (gray circles) and picture (black squares) stimuli relative to the baseline by distractor content for each time window (T1: 3200 4400 ms, T2: 4600 5800 ms, T3: 6 000 7800 ms).

PAGE 59

59 Figure 3 6 . Plots of the competition index, which reflects how visual cortical activity evoked by the two flickering stimuli was modulated for pleasant (gray), neutral (white) and unpleasant (black) distractor contents, across time.

PAGE 60

60 CHA PTER 4 COMPETITION EFFECTS IN HIGH AND LOW SNAKE FEARFUL PARTICIPANTS Experiment 2b examines the extent to which the reduction of the task evoked steady state visual evoked potential (ssVEP) signal when snakes serve as distractors is accompanied by enhanc ement of the visual processing of the snake stimulus in a small snake fearful group, compared to low fear individual s . It has been suggested that anxiety enhances sensory sensitivity (Lang, Davis and Öhman, 2000) and biases attention toward threatening in formation (see Heeren et al., 2013 for review). The ssVEP, which is known to reliably vary with levels of emotional arousal (Keil et al., 2003), has recently been used as a measure to study electrocortical processing of anxiety inducing stimuli in anxious populations. For example, McTeague et al. (2011) demonstrated sustained amplitude enhancement for emotional compared to neutral faces for high socially anxious observers, using the ssVEP. Wieser et al. (2012) used frequency tagging to examine cost effects of fear relevant stimuli (threatening faces) relative to a change detection task in socially anxious observers. Compared to controls, the authors reported a perceptual bias toward the social threat cue at the cost of the task stimulus for observers high in social anxiety, an effect that persisted throughout the duration of the viewing period. Chapter 4 aims to address the extent to which inter individual differences affect ssVEP modulation of a task evoked ssVEP signal when overlapping fear relevant stimu li serve as distractors. As described in Chapter 3, frequency tagging was used to quantify the relative amount of cortical processing of two concurrently presented stimuli by assigning each stimulus a distinct frequency which can later be separated in the frequency domain. With this approach, it is possible to a) quantify and separate the

PAGE 61

61 visual cortical responses to task and distractor evoked steady state signals, and b) to explore possible inter individual differences related to processing fear relevant distractors in the high fear group. In line with the results of Experiments 1 and 2, we expect a reduction in ssVEP amplitude to the task evoked stimulus for all emotionally arousing distractors. Snake distractors are expected to elicit sustained attenua tion of task evoked ssVEP amplitude in high fear participants, more so than the attenuation prompted by other unpleasant arousing content. If emotionally arousing, specifically fear relevant distractors, do oherent motion detection task, we predict an reciprocal increase in ssVEP amplitude to emotionally arousing background distractors, and expect this effect to be enhanced for snake distractors in high fear participants. Methods Participants To explore inte r individual differences with respect to ssVEP amplitude modulation, individuals reporting a high level of self reported snake fear were compared to the same group of (low fear) participants described in Experiment 2. Importantly, data for both participant groups was collected in an identical fashion (same laboratory, paradigm, experimenters, student population and semester). High fear group Fourteen female right handed undergraduate students at the University of Florida provided written consent following Institutional Review Board and received either course credit or compensation (20 USD) for their participation. All participants were screened for photic epilepsy. Participants

PAGE 62

62 included in the final ana lysis performed the coherent motion detection task at a minimal level of 60% or better (average: 80.1% correct, range: 62.59 97.27% correct). The data from fourteen participants (age range: 18 20, mean age: 18.3) with normal to corrected to normal vision we re included in the final analysis. Stimulus rating data was included for all 14 participants. Participants were selected as part of the high fear group based on a prescreening inventory given to 830 undergraduate students enrolled in the Introduction to P sychology course at the University of Florida: Students scoring within the top 15% on the 30 item snake fear questionnaire (SNAQ; Klorman, 1974) were contacted and recruited into the high fear group. Participants in this group scored well above the 85 th pe rcentile on the SNAQ (mean 20.95, SD 2.96, 85 th percentile for female students = 16). Low fear group Participants included in this group are the same participants included in Experiment 2, Chapter 3: The twenty three female participants included in the fin al analysis performed the coherent motion detection task with rate of 75.9% correct, on average (range: 61.1 94.6% correct). The data from twenty three participants were included in the final analysis, with the exception of three students who performed the motion detection task at less than 60% correct, and were excluded only from the behavioral analysis. Stimulus rating data was included for nineteen participants. The twenty three participants were also enrolled in the Introduction to Psychology course at the University of Florida, and were unselected from the undergraduate research participant pool. Participants in the low fear group reported common levels of snake fear (SNAQ mean = 5.52, SD = 3.62).

PAGE 63

63 Stimuli and procedure The aim of this analysis is to assess possible differences in ssVEP amplitude modulation with respect to individual differences in self reported levels of snake fear, therefore the unpleasant picture category has been divided into the two picture contents of which it is composed: human mutilation and snake pictures. Pleasant and neutral picture contents remain collapsed across subcategories of erotica and cats, and people at work and cows, respectively, for all proceeding analyses. EEG Recording Electrophysiological data were collected i n the same laboratory, using the same equipment, as was used in Experiment 1; The analysis conducted in this chapter is an extension of Experiment 2, thus all EEG recording procedures are identical to those described in Chapter 3. EEG data were collected f rom the scalp using a 257 sensor net (see Keil et al. 2014). The EEG was collected continuously with a sampling rate of 250 Hz (16 bit resolution) and were band pass filtered online in the 0.1 90 Hz frequency range using a hardware elliptical filter. The vertex electrode (Cz) was used as the recording reference. Further processing and filtering was performed offline. EEG Reduction and Analyses Data reduction and analysis proce dures were identical to those described for Experiment 2 in Chapter 3, with the exception of dividing the unpleasant picture distractors into semantic categories of human mutilation and snake pictures. Each of the four distractor contents retained an avera ge of 14.48 trials (SD = 0.46) for low fear participants and 13.63 (SD = 0.14) for high fear participants, which did not differ by condition (Low fear: p > 0.93, High fear: p > 0.98) or between participant groups (p >

PAGE 64

64 0.96). As in Experiments 1 and 2, all participants included in Experiment 2b showed reliable (defined as p < 0.05 for the Chi square distributed circular T square at site Oz and its nearest neighbors) evoked oscillations at the driving frequency, suggesting satisfactory signal to noise ratios with the trial counts available in this experiment. Steady state Visual Evoked Potential Analyses Artifact free epochs of the voltage data were averaged for each of the four picture distractors. Time varying amplitude at the stimulation frequencies of 8.57 Hz (dots) and 12 Hz (pictures) was extracted by means of a Hilbert transformation of the time domain averaged data using in house MATLAB® scripts: Data were filtered with a 10 th order Butterworth band pass filter having a width of .5 Hz around the center frequency of 8.57 Hz and 12 Hz for the dot and picture conditions, respectively. As described in Chapter 3, the time varying amplitude was then extracted as the complex conjugate of the band pass filtered signal and the Hilbert transformed analytic signal, for each time point. Data were then temporally smoothed applying a linear moving average, with a window length of 400 ms, for the data between 2000 10100 ms . Statistical Analyses: Behavioral data and SAM ratings The percentage of correctly identified targ ets (hits and correct rejections) was calculated for each distractor condition and participant. Mean target detection rates were calculated for early (1580 4284 ms) and late (4290 7000 ms) time windows to assess changes in behavioral performance during ear ly versus late periods of distractor picture presentation. Only trials in which a coherent motion event occurred during the simultaneous presentation of a background distractor were included in the early late analysis (see Figure 4 1); double targets and t argets occurring during the scrambled period were omitted.

PAGE 65

65 Behavioral performance data were first evaluated for between group differences using a repeated measures analysis of variance (ANOVA) with distractor (pleasant, neutral, mutilation, snakes) and t ime (early, late) as factors, and fearfulness as a between subjects factor. Follow up comparisons were conducted using independent sample t tests. A second analysis was conducted for fearful and low fear groups separately, followed by paired sample t tests (for each group, separately) where necessary. Each of the four picture distractors (pleasant, neutral, mutilation, and snakes) retained an average of 4.6 early and 3.6 late trials in high fear participants, and 4.4 early and 4.1 late trials in low fear pa rticipants. Due to the small number of retained trials during coherent motion events 1 for the early and late conditions, conclusions regarding behavioral performance as an effect of time should be interpreted with caution. Behavioral performance data were also assessed SAM ratings were averaged across participants for each of the four picture conditions by fear group, and were tested for group differences using independent samples t tests, separately for valence and arousal . Statistical Analysis: ssVEP tim e course ssVEP amplitudes were averaged in a large occipital cluster of electrodes including Oz and its neighbors (EGI sensors: 101 105:110 113:119 120:130 133:142 145:153 156:162 165:171 174:178 187; see Figure 4 2, B), resulting in occipital regional mea n amplitudes. A large cluster of electrodes was selected to account for a small 1 Because coherent motion events occurred in only half of the trials per distractor content (the remainder of the trials consisted of random motion), there was a maximum of 10 trials (5 early, 5 late) per distractor content when cohe rent motion events were possible.

PAGE 66

66 sample size (N = 14) with more between subject variability. Normalization of the data was required to minimize variability (Song and Keil, 2013). As such, the mean ssVEP amplit ude for each picture distractor was normalized to the neutral condition on a within subject basis . Between group analysis: Dot stimulus A repeated measures ANOVA (implemented in SPSS ©) was conducted with distractor (pleasant, neutral, mutilation, snakes) and time (baseline, T1, T2, T3) as within subjects factors, and fearfulness (high fear, low fear) as a between subjects factor. As the primary interest in this analysis was investigating if ssVEP amplitude modulation to the unpleasant distractor varied bet ween participant groups, significant interactions were followed with an ANOVA testing unpleasant content (mutilation versus snakes) separately for each time window (T1, T2, T3) using a between subjects factor of fearfulness. Independent samples t tests wer e conducted to follow up differences in mean amplitude elicited by snake versus mutilation distractors, between the groups for the time points of interest. As such, difference scores (average ssVEP amplitude modulation for snake distractors minus average s sVEP amplitude modulation for mutilation distractors) were calculated for each time point and group . Between group analysis: Picture stimulus A repeated measures ANOVA (implemented in SPSS ©) was conducted with distractor (pleasant, neutral, mutilation, snakes) and time (baseline, T1, T2, T3) as within subjects factors, and fearfulness (high fear, low fear) as the between subjects factor to assess varying levels of ssVEP modulation as a function of the flickering distractor stimulus.

PAGE 67

67 For effects involvin g repeated measures, the Huynh Feldt procedure was used to correct for violations of sphericity. Results Behavioral data and SAM ratings A main effect of time ( F (1,32) = 23.997, p < 0.0001) indicated that behavioral performance was significantly worse f or target events occurring early versus later in a trial. A three way interaction of time, fearfulness, and distractor content, ( F (3,96) = 6.13, p = 0.002) was mainly due to behavioral performance when mutilation pictures served as distractors, in which lo w fear participants did not show a difference in performance for coherent motion events occurring early versus late in a trial. For all other distractor contents, performance was worse for early, relative to late, presentations of coherent motion events fo r both low and high fear participants. A repeated measures ANOVA assessing effects of the background distractor on detection of early versus late coherent motion events for each group separately resulted in a main effect of time for both high fear ( F (1,13 ) = 12.156, p = 0.004) and low fear ( F (1,19) = 11.022, p = 0.004) participants. As expected, high fear participants rated pictures of snakes as more unpleasant ( t 29 = 7.055, p = 0.005) and more arousing ( t 29 = 5.205, p < 0.042) than did low fear particip ants. For participants in both groups, emotionally arousing pictures were rated as more pleasant (for erotica and cats), more unpleasant (for human mutilation and snakes), and more arousing than neutral distractors (all p < 0.025).

PAGE 68

68 ssVEPs T he task evoked steady state response was reliably evoked at the expected frequencies of 8.57 Hz and 12 Hz for the dots and pictures, respectively, for the high fear group (Figure 4 2, A; see Figure 3 2, A for the low fear group). Between group comparison: Dot stimulus For the task relevant flickering dot stream, ssVEP amplitude was differentially modulated between high and low fear participants as a function of distractor content and time (Distractor x Time x Fearfulness: F (9,315) = 2.403, p = 0.024; Figure 4 3). Unplea sant distractors elicited opposite patterns of ssVEP amplitude reduction for high and low fear participants at T1 and T3: a Distractor Content x Fearfulness interaction at T1 (3200 4400 ms; F (1,35) = 4.784, p = 0.035) and T3 (6000 7800 ms; F (1,35) = 4.912, p = 0.033) indicated that ssVEP amplitude for high fear participants was smaller during the T1 time interval when mutilation pictures serves as distractors and smaller during the T3 time interval when snake pictures served as distractors, whereas the oppo site pattern of ssVEP amplitude was observed for the low fear individuals. ssVEP amplitude did not differ for unpleasant distractors during T2 (F< 0.36, n.s.) . Between group comparison: Picture stimulus The flickering distractor stream was reliably modula ted by the background distractor (Distractor, F (3,105) = 3.202, p = 0.03) across time (Time F (3,105) = 14.95, p < 0.0001) for both groups, but did not yield any significant between group differences. Discussion In Experiment 2b, high and low fear and pa rticipants differed only in the way in which ssVEP amplitude was modulated by unpleasant distractors during the T1 and T3 time intervals: For high fear participants, an initial attenuation of ssVEP amplitude when

PAGE 69

69 mutilation pictures serve as distractors du ring T1 was subsequently followed by an attenuation of ssVEP amplitude when snakes serve as distractors during T3, whereas the opposite pattern of ssVEP amplitude modulation was observed for low fear individuals. Although the reduction in ssVEP amplitude t o the flickering dots when snakes served as distractors was not observed to be greater than the ssVEP amplitude reduction observed for all other emotional contents (as in Experiment 1) for the high fear participants in Experiment 2b, we did replicate the l ate (6000 7800 ms) reduction in ssVEP amplitude to the flickering dot stream when snakes serve as distractors at T3, compared to the other distractor contents. We also observed a temporally sustained, albeit modest, reduction in ssVEP amplitude when snakes served as distractors during the T1 T3 time windows for high fear participants, although this reduction in ssVEP amplitude was not greater than the reduction elicited by pleasant and mutilation distractors, as in Experiment 1. It is important to note th at in Experiment 1 (N = 20), the observed sustained reduction in ssVEP amplitude was driven mainly by a subgroup of participants with the highest levels of self reported snake fear. Thus, it seems plausible that in a smaller sample with an overall lower ra nge of self reported snake fear (Experiment 1, range: 17 28, Experiment 2b, range: 17 24), it may be difficult to replicate the robust reduction in ssVEP amplitude observed in Experiment 1 when snakes served as distractors. Modulation of the flickering p icture stimulus was affected by picture distractors and time, but in contrast to our hypothesis, no group differences were observed in ssVEP amplitude modulation to the flickering picture stimulus .

PAGE 70

70 Figure 4 1. Time sequence for a single trial: interv als of coherent motion could occur between 1169 7000 ms (target window, gray box) post stimulus onset. Each trial lasted 9751 ms with a variable 3 5 second inter trial interval. Stimuli were overlapping, and flickered at 8.57 Hz (dots) and 12 Hz (picture).

PAGE 71

71 Figure 4 2. A) Grand averaged frequency spectra across all (high fear) participants and conditions, demonstrating a reliable peak at the driving frequency of the flickering dots, 8.57 Hz and flickering pictures, 12 Hz. B) Representation of electrode p lacement; sensors used in the cluster analyses are highlighted by red circles.

PAGE 72

72 Figure 4 3. Average ssVEP amplitude modulation to the flickering dot stimulus for low fear (left) and high fear (right) participants, averaged across distractor contents f or pleasant (blue), neutral (black), mutilation (red) and snake (green) conditions, across the four time windows of interest (bsl: 2000 2600 ms, T1: 3200 4400 ms, T2: 4600 5800 ms, T3: 6000 7800 ms).

PAGE 73

73 CHAPTER 5 GENERAL DISCUSSION In summary, the results o f Experiments 1, 2 and 2b provide further support for the hypothesis that task irrelevant but emotionally engaging stimuli act as strong competitors in visual cortex, and effectively interfere with the visual processing of a concurrent task stimulus. Thes e findings are consistent with previous literature, which suggests emotionally arousing cues compete for limited resources at the level of the visual cortex (Keil et al. 2005; Hajcak et al., 2013). The extent to which competition in visual cortex results i n trade off effects of attention was examined in Experiment 2, where cost effects were observed from 3200 5800 ms the first 2600 ms of the picture viewing period for emotional, compared to neutral, distractors. Amplification effects of ssVEP amplitude modulation to the flickering distractor stimulus occurred later in the viewing epoch for emotional compared to neutral distractors, beginning at 4600 ms and persisting for an additional 1200 ms. In all experiments, interference by emotional distractors was reflected as a decrease in task performance for all participants when emotional, compared to neutral, pictures served as distractors. For participants reporting high levels of snake fear, a sustained reduction of ssVEP amplitude to the task relevant strea m was observed for high fear participants in Experiment 1, but was only replicated in the last 6000 7800 ms of picture viewing in Experiment 2b. This series of experiments provides a basis for further investigating the time course of competition for atte ntional processing resources in human visual cortex. Building on the results of Experiments 1, 2, and 2b, future research may take advantage of the temporal pattern and visuocortical trade off effects established here to objectively identify patients with dysfunctional attentional resource allocation to fear related stimuli,

PAGE 74

74 which may inform therapeutic treatments aiming to direct attention away from a threat cue (Amir et al., 2009). Emotional Stimuli Act as Strong Competitors for Visual Processing Resource s Evidence from functional magnetic resonance imaging studies (Kaster and Ungerleider, 2000) as well as single cell recording studies in non human primates (Reynolds and Chelazzi, 2004) provides evidence for modulation of neural activity in visual cortex b y biasing attention toward stimuli important for current behavioral goals (Desimone and Duncan, 1995; Shuler and Bear, 2006). The biased competition framework has since been adopted to account for large scale neural effects of human visual attention (Dunca n et al., 1997, Desimone, 1998), for which competition for a competition for distinct populations of visual cortical neurons (see Franconeri et al., 2013), as quantified here u sing the ssVEP. Experiments 1 and 2 reliably demonstrated modulation of task evoked ssVEP amplitude when emotional compared to neutral pictures served as distractors. These findings are consistent with the literature suggesting emotionally arousing stimuli receive prioritized processing in sensory systems (Bradley et al., 2003) at the cost of concurrently presented information (Keil et al., 2005; Ihssen et al., 2007). In healthy observers, interference by an emotionally arousing distractor has been demons trated to be relatively early but brief (500 1000 ms; Müller et al., 2008; Hindi Attar et al., 2010). While all participants in Experiment 1 showed early sensitivity to distraction, participants reporting high levels of snake fear demonstrated the most pro nounced reduction in task evoked ssVEP amplitude when snakes served as distractors, an effect which was maintained throughout the viewing period (5800 ms).

PAGE 75

75 Although an extended viewing period (compared to previously published literature, e.g., Müller et al ., 2008; Wieser et al., 2012) contributed to a better understanding of the temporal dynamics associated with competition of an emotionally arousing distractor, Experiment 1 was limited in that perceptual sensitivity was observed only in reference to the fl ickering dots, and thus no reciprocal measure of enhancement or reduction of the competing stimulus could be quantified. According to the biased competition model of attention (Desimone and Duncan, 1995), preference for motivationally relevant cues will be strongest when competing cues overlap in space. To this aim, Experiment 2 used frequency tagging to measure the extent to which the reduction of the task evoked ssVEP signal when emotionally arousing stimuli serve as distractors is accompanied by enhanc ement of the visual processing of those stimuli. With respect to stimulus competition, the results of Experiment 2 support the hypothesis that emotionally arousing stimuli receive preferential processing and involuntarily capture and divert attentional re sources away from concurrently presented stimuli. As evidenced by the competition indices in Figures 3 3 and 3 4, competition between a motion detection task and task irrelevant emotional distractor appears approximately 450 ms after onset of the distracto r stimulus (T1), and persists for approximately 2600 ms. Further, Experiment 2 supports the hypothesized trade off effect of attention between task and distractor stimulus, such that a relative decrease in task evoked ssVEP amplitude results in a reciproca l enhancement of ssVEP amplitude to the flickering distractor stimulus. Similar findings were observed in a recent study examining competition of a threatening stimulus and concurrent task in socially anxious participants, where an involuntary perceptual b ias toward threatening faces was

PAGE 76

76 maintained throughout the viewing period (Wieser et al., 2012). However, we failed to observe sustained threat specific differences in ssVEP amplitude modulation to fear cues in our analysis of inter individual differences, and observed threat specific differences only in the last 1800 ms of the picture viewing epoch (Experiment 2b). Although group differences emerged only for the way in which ssVEP amplitude modulation to unpleasant stimuli changed across time, Figure 4 3 i llustrates a similar, albeit modest, attenuation of task evoked ssVEP amplitude to the snake distractor in the high fear group. However, these results are based on ssVEP amplitude modulation of a small snake fearful sample (N = 14), and warrant replication . Specificity of the Visuocortical Response to Fear Cues Recent approaches to emotional perception suggest that the motivational significance, or biological relevance, of emotional stimuli modulates sensory processing. The majority of this research has be en conducted in the visual domain, where sensory responses to stimuli bearing biological importance are facilitated (Shuler and Bear, 2006) when presented as intense, rare, or unexpected stimuli (Koch and Ullman, 1985). It is well established that in anxio us individuals, threatening information is highly prioritized over neutral or positive information (e.g., Öhman et al., 2001; Sabatinelli et al., 2005; McTeague et al., 2011). It is unclear, however, if hypervigilance is specific to threat features, or if sensory information is generally amplified when presented in a threatening context. It is plausible, for instance, that in an experiment where threatening cues are interspersed with non threatening cues, that a hypervigilant pattern of responding is genera lized to all incoming stimuli once a threat cue is detected (Michalowski et al., 2009). Recent attempts to explore effects of fear on perception have utilized electrocortical measures to further understand the stimulus specificity of

PAGE 77

77 neurophysiological mar kers of hypervigilance. Thus, an interesting question emerges regarding the electrocortical response elicited by fear relevant cues in high fear participants: is modulation in visual cortex is specific to the feared stimulus, or generalized to all stimuli? In a visual search task, Weymar, Keil, and Hamm (2013) recorded event related potentials for observers high or low in spider fear, to examine the extent to which specific fear modulates afferent sensory processing in lower tier visual cortex. Enhanced am plitude of the C1, the earliest component of visual processing in primary visual cortex, was reliably evoked for high fear participants only. Importantly, modulation of the C1 in high fear participants was not specific to spider cues, but was generalized t o all target stimuli, suggesting an overall enhanced sensitivity to all visual stimuli, irrespective of content. A general state of hypervigilance toward all visual stimuli in settings in which a threatening stimulus is likely to appear seems to be the fir st step of defensive activation: Michalowski et al. (2009) observed threat specific responses in fearful participants only after an initial period of general hypervigilance in a context in which fear relevant stimuli may exist. The authors demonstrated an overall enhancement of the P1 to all emotional stimuli (general hypervigilance) followed by enhanced early posterior negativity and late positive potential specific only to fear relevant stimuli in the fearful group. This explanation is further supported b y the defense cascade model (Lang et al., 1997; Fanselow, 1994), which posits that in a context in which a threat may occur, an organism displays general hypervigilance to all stimuli in the environment, and once a threat is detected, attention is selectiv ely captured by that cue, resulting in a shift of attention (and active defensive responding) toward the threat

PAGE 78

78 stimulus. Thus, when the possibility exists that a fear relevant stimulus could appear, fearful participants show increased sensitivity to all s timuli, which may in turn amplify signals in early visual cortex. Data from the high fear participants in Experiments 1 and 2b demonstrate a similar pattern of responding to the fear relevant snake distractor: In Figure 2 4, an initial decrease in ssVEP am plitude during the T1 time window is observed for all background distractors in the high fear group, and similarly for T1 in Experiment 2b (Figure 4 3). This brief decrease (general hypervigilance) in task evoked ssVEP amplitude to all distractors is follo wed by a subsequent return to baseline amplitude for all distractors, with the exception of the snake distractor. These findings provide further evidence of an involuntary perceptual bias to fear relevant stimuli in fearful participants, which is supported by notions regarding dysfunctional attentional processes in anxious populations (Bar Haim et al., 2007; Yiend, 2010). Perceptual Enhancement of Motivationally Relevant Stimuli in Visual Cortex The general findings of the experiments presented in this di ssertation are in line with the notion that threat cues reflexively activate motivational circuits, leading to heightened attention and facilitated orienting to stimuli important for survival (Lang et al., 1997). These experiments are limited, though, in t hat there is no direct measure of motive system engagement. However, stimulus rating data have been included for all experiments as a proxy of self reported emotional relevance (e.g., Lang et al., 1993; Bradley, 2009; Lang and Bradley, 2010). In this vein, it has been proposed that re entrant modulation of the visual cortex is a critical process for enhancing perception of emotionally arousing visual stimuli, effected through afferent modulation of occipital cortex by anterior cortical and subcortical struc tures (Lang et al., 1997). As the ssVEP is a measure of ongoing stimulus processing in visual cortex, it is likely that the ssVEP

PAGE 79

79 may reflect biasing signals of higher order cortical regions (Kastner and Ungerleider, 2000), which in turn alter the sensitiv ity of neurons in favor of features associated with emotionally arousing content (e.g., Keil et al., 2012) Conclusions In sum, the data presented here are consistent with the view that the processing capacity of the visual system is limited, and must ther efore rely on cues that activate motivational systems that inherently capture attention and utilize a greater percentage of attention resources at the cost of concurrent stimuli. This pattern is reflected as competition in distinct neural circuits, in whic h re entrant modulation may lead to more effective connectivity among local circuits in the occipital lobe over time (Gruber et al., 2004), and result in facilitated detection and identification of motivationally relevant features (Keil et al., 2003; Morat ti et al., 2004; Keil et al., 2007; Keil et al., 2012).

PAGE 80

80 LIST OF REFERENCES Amir N., Beard C., Taylor C.T., Klumpp H., Elias J., Burns M. and Chen X., Attention training in individuals with generalized social phobia: a randomized controlled trial, Journa l of Consulting and Clinical Psychology 77 , 2009,961 973. Bar Haim Y., Lamy D., Pergamin L., Bakermans Kranenburg M.J. and van IJzendoorn M.H., Threat related attentional bias in anxious and nonanxious individuals: a meta analytic study, Psychological Bull etin 133 (1), 2007,1 24. Becker E.S., Rinck M., Margraf J. and Roth W.T., The emotional Stroop effect in anxiety disorders: general emotional or disorder specificity?, Journal of Anxiety Disorders 15 (3), 2001, 147 159. B ö gels S.M. and Mansell W., Attentio n processes in the maintenance and treatment of social phobia: hypervigilance, avoidance and self focused attention, Clinical Psychology Review 24 (7), 2004, 827 856. Bradley M.M. and Lang P.J., Measuring emotion the self assessment manikin and the semanti c differential, Journal of Behavior Therapy and Experimental Psychiatry 25 , 1994, 49 59. Bradley M.M., Sabatinelli D., Lang P.J., Fitzsimmons J.R., King W. and Desai P., Activation of the visual cortex in motivated attention, Behavioral Neuroscience 117 , 2 003, 369 380. Desimone R., Visual attention mediated by biased competition in extrastriate visual cortex, Philosophical Transactions of the Royal Society of London. Series B, Biological sciences 353 , 1998, 1245 1255. Desimone R. and Duncan J., Neural mecha nisms of selective visual attention, Annual Review of Neuroscience 18 , 1995, 193 222. Donaldson C., Lam D. and Mathews A., Rumination and attention in major depression, Behavior Research and Therapy 45 (11), 2007, 2664 2678. Garner M., Mogg K. and Bradley B.P., Orienting and maintenance of gaze to facial expressions in social anxiety, Journal of Abnormal Psychology 115 (4), 2006,760 770. Hajcak G., MacNamara A., Foti D., Ferri J. and Keil A., The dynamic allocation of attention to emotion: simultaneous and independent evidence from the late positive potential and steady state visual evoked potentials, Biological Psychology 92 (3), 2013, 447 455.

PAGE 81

81 Heeren A., De Raedt R., Koster E.H. and Philippot P., The (neuro)cognitive mechanisms behind attention bias modif ication in anxiety: proposals based on theoretical accounts of attentional bias, Frontiers in Human Neuroscience 7 , 2013, 119. Hillyard S.A., Hink R.F., Schwent V.L. and Picton T.W., Electrical signs of selective attention in the human brain, Science 182 , 1973, 177 180. Hindi Attar C., Andersen S.K. and M ü ller M.M., Time course of affective bias in visual attention: convergent evidence from steady state visual evoked potentials and behavioral data, Neuroimage 53 , 2010, 1326 1333. Hindi Attar C. and M ü ller M .M., Selective attention to task irrelevant emotional distractors is unaffected by the perceptual load associated with a foreground task, PLoS One 2012,DOI: 10.1371/journal.pone.0037186. Ihssen N., Heim S. and Keil A., The Costs of emotional attention: aff ective processing inhibits subsequent lexico semantic analysis, Journal of Cognitive Neuroscience 19 (12), 2007, 1932 1949. Jungh ö fer M., Elbert T., Tucker D.M. and Rockstroh B., Statistical control of artifacts in dense array EEG/MEG studies, Psychophysio logy 37 , 2000,523 532. Keil A., Debener S., Gratton G., Jungh ö fer M., Kappenman E.S., Luck S.J., Luu P.,Mille r G.A. and Yee C.M., Committee report: publication guidelines and recommendations for studies using electroencephalography and magnetoencephalograp hy, Psychophysiology 51 (1), 2014, 1 21. Keil A., Gruber T., M ü ller M.M., Moratti S., Stolarova M., Bradley M.M. and LangP.J., Ea rly modulation of visual perception by emotional arousal: evidence from steady state visual evoked brain potentials, Cognitive, Affective, and Behavioral Neuroscience 3 (3), 2003, 195 206. Keil A., Moratti S., Sabatinelli D., Bradley M.M. and Lang P.J., Additive effects of emotional content and spatial selective attention on electrocortical facilitation, Cerebral Cortex 15 (8), 20 05, 1187 1197. Kemp A.H., Silberstein R.B., Armstrong S.M. and Nathan P.J., Gender differences in the cortical electrophysiological processing of visual emotional stimuli, NeuroImage 21 (2), 2004, 632 646. Klorman R., Weerts T.C., Hastings J.E., Melamed B.G . and Lang P.J., Psychometric description of some Specific Fear Questionnaires, Behavior Therapy 5 , 1974, 401 409. Kolassa I.T., Buchmann A., Lauche R., Kolassa S., Partchev I., Miltner W.H. and Musial F., Spider phobics more easily see a spider in morphed schematic pictures, Behavior and Brain Functions 3 , 2007b, 59.

PAGE 82

82 Kolassa I., Kolassa S., Musial F. and Miltner W.H., Event related potentials to schematic faces in social phobia, Cognition and Emotion 21 , 2007a,1721 1744. Lang P.J., Bradley M.M. and Cuthber t B.N., Motivated attention: affect, activation, and action, Attention and Orienting: Sensory and Motivational Processes, 1997, Lawrence Erlbaum Associates;Hillsdale, N.J, 97 135. Lipka J., Miltner W.H. and Straube T., Vigilance for threat interactions wit h amygdala responses to subliminal threat cues in specific phobia, Biological Psychiatry 70 (5), 2011, 472 478. Macmillan N.A. and Creelman C.D., Detection Theory: A User s Guide, 2, Revised ed., 2004, Taylor & Francis. Mathews A., Why worry? the cognitiv e function of anxiety, Behaviour Research and Therapy 28 (6), 1990, 455 458. McGinnis E.M. and Keil A., Selective processing of multiple features in the human brain: effects of feature type and salience, PloS One 6 , 2011, e16824. McNally R.J., Mechanisms o f exposure therapy: how neuroscience can improve psychological treatments for anxiety disorders, Clinical Psychology Review 27 , 2007, 750 759. McTeague L.M., Shumen J.R., Wieser M.J., Lang P.J. and Keil A., Social Vision: sustained perceptual enhancement o f affective facial cues in social anxiety, NeuroImage 54 , 2011, 1615 1654. Mogg K., Mathews A. and Weinman J., Selective processing of threat cues in anxiety states: a replication, Behavior Research and Therapy 27 , 1989,317 323. Mogg K., Millar N. and Brad ley B.P., Biases in eye movements to threatening facial expressions in generalized anxiety disorder and depressive disorder, Journal of Abnormal Psychology 109 (4), 2000, 695 704. M ü ller M.M., Andersen S.K. and Keil A., Time course of competition for visua l processing resources between emotional pictures and foreground task, Cerebral Cortex 18 , 2008, 1892 1899. Ö hman A., Flykt A. and Esteves F., Emotion drives attention: detecting the snake in the grass, Journal of Experimental Psychology: General 130 , 2001, 466 478. Ö hman A. and Mineka S., Fears, phobias, and preparedness: toward an evolved module of fear and fear learning, Psychological Review 108 , 2001, 483 522. Pflugshaupt T., Mosimann U.P., von Wartburg R., Schmitt W., Nyffeler T. and M ü riR.M., Hypervigil ance avoidance pattern in spider phobia, Journal of Anxiety Disorders 19 (1), 2005, 105 116.

PAGE 83

83 Regan D., Human Brain Electrophysiology: Evoked Potentials and Evoked Magnetic Fields in Science and Medicine, 1989, Elsevier; New York. Reynolds J.H. and Heeger D .J., The normalization model of attention, Neuron 61 (2), 2009, 168 185. Sabatinelli D., Bradley M.M., Fitzsimmons J.R. and Lang P.J., Parallel amygdala and inferotemporal activation reflect emotional intensity and fear relevance, Neuroimage 24 , 2005, 1265 1 270. Schimmack U. and Derryberry D., Attentional interference effects of emotional pictures: threat, negativity, or arousal?, Emotion 5 , 2005, 55 66. Shafer A.T., Matveychuk D., Penney T., O Hare A.J., Stokes J. and Dolcos F.,Processin g of emotional distra ction is both automatic and modulation by attention: evidence from an event related fMRI investigation, Journal of Cognitive Neuroscience 2012, 1233 1252. Sommer M.A. and Wurtz R.H., Visual perception and corollary discharge, Perception 37 , 2008, 408 418. S traube T., Mentzel H.J. and Miltner W.H., Waiting for spiders: brain activation during anticipatory anxiety in spider phobics, Neuroimage 37 (4),2007, 1427 1436. Wangelin B.C., Low A., McTeague L.M., Bradley M.M. and Lang P.J., Aversive picture processing: effects of a concurrent task on sustained defensive system engagement, Psychophysiology 48 , 2011, 112 116. Wieser M.J., McTeague L.M. and Keil A., Sustained preferential processing of social threat cues: bias without competition? Journal of Cognitive Neur oscience 2011, 1973 1986. Wieser M.J., McTeague L.M. and Keil A., Competition effects of threatening faces in social anxiety, Emotion 12 , 2012, 1050 1060. Wieser M.J., Puli P., Wyers P., Alpers G.W. and Muhlberger A., Fear of negative evaluation and the hy pervigilance avoidance hypothesis: an eye tracking study, Journal of Neural Transmission 116 (6), 2009, 717 723. Victor J.D. and Mast J., A new statistic for steady state evoked potentials, Electroencephalography and Clinical Neurophysiology 78 (5), 1991,3 78 388 .

PAGE 84

84 BIOGRAPHICAL SKETCH Menton Deweese is originally from Meridian, MS. She earned her Bachelor of Science in psychology with a minor in geology and concentration in human services from Millsaps College in May 2009. She received a Master of Science in the area of Behavioral and Cognitive Neurosciences in December 2011 from the Department of Psychology at the University of Florida, and received her Ph.D. in Psychology in August of 2014. r high school, where she worked extensively with mentally and physically handicapped adults and children. As an undergraduate, Menton began her research career studying developmental processes in infants and toddlers under the tutelage of Dr. Melissa Kelly , and developed an interest in cognitive neuroscience working in the lab of Dr. Melissa Lea. Internships and volunteer positions at behavior therapy clinics and developmental centers throughout the course of her college career were valuable learning exper iences, and contributed to a yearning to further understand the processes underlying behavior on a deeper level. Eager to receive training in electrophysiology, Menton was accepted into the graduate program of the Department of Psychology at the Universi ty of Florida in 2009, under the direct mentorship of Dr. Andreas Keil, and co mentorship of Drs. Margaret emotional stimulus processing, where she has investigated the role of feature salience in perception and attention, using event related potentials and steady state visual ev oked potentials. Menton has also received training in basic experimental and technical skills involved in functional magnetic resonance imaging, and has collaborated

PAGE 85

85 with members of the Biomedical Engineering department at UF on a simultaneous EEGfMRI proj ect. Upon completion of her doctorate, Menton moved to Houston, TX, where she is currently a postdoctoral fellow under the tutelage of Dr. Francesco Versace in the Department of Behavioral Science at MD Anderson Cancer Center.



PAGE 1

ReviewarticleAsystematicreviewoftheabilityofurineconcentrationtodistinguish antipsychotic-frompsychosis-inducedhyponatremiaWanlopAtsariyasinga,MorrisB.Goldmanb ,naDepartmentofPsychiatry,FacultyofMedicineSirirajHospital,MahidolUniversity,Bangkok,ThailandbDepartmentofPsychiatryandBehavioralSciences,FeinbergSchoolofMedicine,NorthwesternUniversity,446EastOntario, Suite7-100Chicago,IL60611,USAarticleinfoArticlehistory: Received24October2013 Receivedinrevisedform 31January2014 Accepted20March2014 Availableonline29March2014 Keywords: Waterintoxication Schizophrenia Vasopressin Resetosmostat SIADHabstractLife-threateninghyponatremiainpsychoticpatientsiscommonandtypicallyisattributabletoeither antipsychoticmedicationortoacutepsychosisinthosewiththepolydipsia-hyponatremiasyndrome. Thepreferredtreatmentforonesituationmayworsenthehyponatremiaifcausedbytheothersituation. Henceitiscriticaltodistinguishbetweenthesetwopossibilities.Casereportsandserieswereidenti ed throughelectronicdatabases.Fifty-fourcasesofhyponatremiawithoutrecognizedcausesinpsychotic patientsweredividedintothosewithdilute( o plasmaosmolality)orconcentrated( 4 plasma osmolality)urine.Thedistributionofurineconcentrationandmeasureslikelytobeassociatedwith psychoticillnessanditstreatmentwerecomparedinbothgroups.Naranjo'sscalewasutilizedto determinetheprobabilityhyponatremiawasdrug-induced.Urineosmolality tabimodaldistribution (intersection219mOsm/kg)betterthanaunimodaldistribution. ‘ Probable ’ drug-inducedcasesoccurred 6.8(95%CI ¼ 1.6 – 28.9)timesmoreofteninthosewithconcentratedurine.Acutepsychoticexacerbations occurred4.5(95%CI ¼ 0.4 – 54.1)timesmoreofteninthosewithdiluteurine.These ndings,aswellas severalothertrendsinthedata,indicatethatmeasuresofurineconcentrationcanhelpdistinguish betweenantipsychotic-inducedandpsychosis-inducedhyponatremia. & 2014ElsevierIrelandLtd.Allrightsreserved. Contents 1.Introduction........................................................................................................129 2.Methods...........................................................................................................130 2.1.Datasourcesandstudyselection.................................................................................130 2.2.Dataextraction...............................................................................................131 2.3.Dataanalysis.................................................................................................131 2.4.Statisticalanalysis.............................................................................................131 3.Results............................................................................................................132 4.Discussion.........................................................................................................132 Con ictofinterest.......................................................................................................133 AppendixA.Supplementaryinformation..................................................................................133 References.............................................................................................................1331.Introduction Life-threateninghyponatremia(waterintoxication)isarelatively commonconditioninpatientswithpsychoticillnesses( Renneboog etal.,2006;Hawkenetal.,2009;Meulendijksetal.,2010;Williams andKores,2011 ).Oftentheelectrolyteimbalanceisattributableto prescribedmedications,particular lydiureticsorpsychotropicssuch asanticonvulsantmoodstabilizers,serotoninreuptakeinhibitors, tricyclicantidepressantsandantipsychoticmedications( Liamisetal., 2008;Letmaieretal.,2012 ).Inaddition,nicotineaswellascigarette smokepersecanimpairwaterbalance( Robertson,2006 ).Thisis especiallyapttocontributetohyponatremia( Blum,1984;Ismail Contentslistsavailableat ScienceDirect journalhomepage: www.elsevier.com/locate/psychresPsychiatryResearchhttp://dx.doi.org/10.1016/j.psychres.2014.03.021 0165-1781/ & 2014ElsevierIrelandLtd.Allrightsreserved. nCorrespondingauthor.Tel.: þ 13126952089;fax: þ 17083836387. E-mailaddress: m-goldman@northwestern.edu (M.B.Goldman). PsychiatryResearch217(2014)129 – 133

PAGE 2

etal.,2010 )becausesomanyschizophre nicpatients,particularly thoseatriskofhyponatremia( deLeonetal.,1996 ),smokeheavily. Hyponatremiatypicallyresolvessoonaftertheoffendingagentis stopped,butinthecaseofantipsychoticsthismayhaveaseemingly paradoxicaleffectifthehyponatremiaisduetoacutepsychosis ( Goldmanetal.,1997;Meulendijksetal.,2010 ).Thisisbecauseacute psychosishasalsobeenassociatedwithimpairedwaterexcretion ( Targowla,1923 ),primarypolydipsia( HoskinsandSleeper,1933 )and waterintoxication( Barahal,1938 )wellbeforeantipsychoticswere discovered.Aftertheirintroduction,theliteratureisrepletewith reportsindicatingantipsychoticmedicationbothcontributesto ( Meulendijksetal.,2010 )andameliorates( Dubovskyetal.,1973; Hariprasadetal.,1980;CanusoandGoldman,1999;Liamisetal., 2008 )hyponatremia.Currentlyperhaps40%ofpsychoticpatients admittedwithunexplainedhyponatremiaarenottakingantipsychoticmedication( WilliamsandKores,2011 ). Bothantipsychoticmedicationandacutepsychosisinduce dilutionalhyponatremia,whichoccurswhen uidintakeoverwhelmsrenaldilutingcapacity.Theretainedwaterswellsallbody tissuesthusproducingcerebraledemaandultimatelycerebral compressionbytheskull(waterintoxication).Undernormal circumstances,hyponatremiacanbepreventedbyvaryingurine dilution.Urineconcentrationisprimarilyregulatedbytheantidiuretichormone,argininevasopressin(AVP),whichissecreted fromthebrainandthenactsinthekidney.Decreasesinplasma osmolalityinhibitAVPsecretion,inwhichcasethenormalkidney excreteshugeamountsofwater( 20L/day). Dilutionalhyponatremiatypicallyoccursduetoeitherarelatively xedimpairmentintheabilitytoproducediluteurine,ora variableimpairmentwhichlessensasthehyponatremiaworsens. Dependingontheplasmaosmolalityatwhichurinebeginstobe concentrated(osmoticsetpoint),thepatientmayeitherpresent withsymptomaticorasymptomatichyponatremia( Robertson, 2006 ).Hyponatremiawithotherpsychotropicshasbeenassociatedwithboth xedandvariableimpairments( Meulendijks etal.,2010 ),whileacutepsychosishasbeenconsistentlyassociated withthevariableimpairment( Hariprasadetal.,1980;Viewegetal., 1986;Goldmanetal.,1988;Kishimotoetal.,1989;Delvaetal.,1990; Ohsawaetal.,1993 ).Themechanismappearstobeattributableto psychosisloweringtheosmoticsetpointforAVPsecretion ( Goldmanetal.,1997 )perhapsduetoastressdiathesisthatis associatedwiththeunderlyingpsychiatricillness( Goldmanetal., 2007,2011;Goldman,2009 ).Thecontributionoftheelevated AVPtothehyponatremiaisdemonstratedbyitsrapidreversalwith AVPantagonists( Josiassenetal.,2008,2012 ).Allclassesof antipsychoticdrugsareassociatedwithhyponatremia( Mannesse etal.,2010 ),andcurrentlytherearenopublishedguidelinesto aidclinicianswhethertoincreaseordecreaseantipsychoticmedication.Intheabsenceofcontrolledstudies,weconducteda systematicreviewofpublishedcasesofmedicatedpsychotic patientswithunexplainedhyponatremiawhohadconcurrent measuresofurinedilution.2.Methods 2.1.Datasourcesandstudyselection Onehundredandthirty-sixarticleswereidenti edfromtheMEDLINEdatabase from1960toSeptember2012usingthefollowingMeSHterms:hyponatremia, inappropriateADHsyndrome,antipsychoticagentsandEnglishlanguage(seeFlow Diagram Fig.1 ).Ninety-onemoreweredrawnfromarecentcomprehensivereviewof antipsychotic-inducedhyponatremiaauthoredby Meulendijksetal.(2010) .Forty-one duplicateswereidenti ed,leaving186tobescreened.Nonewerecontrolledstudies, andallwereeithersingleoraseriesofcasereports.Eighty-oneofthesearticleswere excludedonthebasisoftheabstractbecausetheydidnotdescribepsychoticpatients. Articles identified through MEDLINE database (n=136) Screening Included Eligibility Identification Articles identified through Meulendijks et al. 2010 (n=91) Articles after 41 duplicates removed (n=186) Articles screened (n=186) Articles excluded by review of abstract (n=81) Full-text articles examined for eligibility (n=105) Off topic (n=17) No urine data (n=31) Not 1° psychotic disorder (n=21) Others (n=3) 44 articles (64 cases) extracted data 39 articles (54 cases) included in analysis Full-text articles examined for eligibility (n=116) Additional full-text articles from references of these articles (n=11) 5 articles (10cases) with recognized causes of hyponatremia excluded Fig.1. FlowDiagram. Source :AdaptedfromthePRISMA2009FlowDiagram( Moheretal.,2009 ). W.Atsariyasing,M.B.Goldman/PsychiatryResearch217(2014)129 – 133 130

PAGE 3

Theremaining105articlesaswellas11moredrawnfromthesearticles'references wereevaluated.Seventeenofthese116articleswereexcludedbecausetheywerenot relevant,31becausetheydidnotcontainconcurrentindicesofurinedilution(neither urineosmolalitynorurinespeci cgravity),21becausethepsychiatricdiagnosiswas notaprimarypsychoticdisorder(schizophrenia,schizophreniform,schizoaffective,or psychosisNOS),andthreeforotherreasons,yielding64psychotichyponatremic patients(44articles)onantipsychotictherapyandwithconcurrentmeasuresofurine dilutiononpresentation( SupplementalTableS2 ). 2.2.Dataextraction Serumsodium,plasmaosmolality,urineosmolality,urinespeci cgravityand urinesodiumwererecordedaswereconcurrentmedicationsthatcouldpotentially accountforhyponatremia( Liamisetal.,2008 ).Variablesdrawnfromguidelinesfor adverseeventreporting( Kellyetal.,2007 ),andhypothesizedmeasuresdistinguishingdrug-inducedfrompsychosis-inducedcaseswereidenti edineachreport. Tencases(sixonthiazidediuretics,oneoncarbamazepine,oneonSSRI,twoon tricyclicantidepressants)weretakingothermedicationsrecognizedtocause hyponatremialeavinga nalsampleof54inpatients(39articles).Assessmentof whetheracasewasanadversedrugreactionwasbasedon Naranjoetal's.(1981) s probabilityscale.Thisassessmentwastheprimaryoutcomemeasuresinceit couldbescoredinallcases.Scoresonthisscalerangefrom 4to þ 13andare categorizedas ‘ unlikely ’ ( 4to0), ‘ possible ’ (1 – 4), ‘ probable ’ (5 – 8)and ‘ certain ’ (9 – 13).Secondaryoutcomeswere:antipsychoticdose(transformedtode ned-daily dose(DDD))( ATC/DDDIndex,2012 );presenceofacutepsychoticexacerbationon admission;presenceofpolydipsia(perauthor);timefrominitiationofcurrent antipsychotictoonsetofhyponatremia;andtheeffectofantipsychoticdiscontinuation ( ‘ dechallenge ’ )andofantipsychoticrechallengewiththesamemedicationon hyponatremia.Thesemeasureswerenotavailableforallcases.Assessmentswere madebyoneoftheauthors(WA)whosoughtinputfromtheremainingauthoronthe rareoccasionthatthecorrectresponsewasnotevident. 2.3.Dataanalysis Subjectsweregroupedaccordingtowhethertheirurinewasdiluteor concentratedatpresentation,i.e.whetherurineosmolalitywaslowerorhigher thanconcurrentplasmaosmolality.Whilethedivisionissomewhatarbitrary,itis physiologicallysigni cantandbias-free( Robertson,2006 ).Eightofthe54caseshad onlyurinespeci cgravity,andifspeci cgravitywas o 1.010weplacedthecaseinthe dilutegroupsincethisisacommoncutofffordiluteurine( Voinescuetal.,2002 ).Inany event,urinespeci cgravitywas o 1.004forsevenofthesecasesandwas1.015forthe eighthcase.Threecaseshadonlyplasmaso diumavailable,soplasmaosmolalitywas estimatedas2 plasmasodium þ 10( RossandChristie,1969 ).Theassociatedurine osmolalityinallfourinstanceswaseither30mOsm/kgmoreorlessthanthis calculatedplasmaosmolality,sogroupingwaslikelytobeaccurate. 2.4.Statisticalanalysis WeusedaGaussianmixturemodelto t1-and2-gaussianstotheurineosmolality values,andutilizedBayesianandAkaikeinformationcriterion(AIC)todetermine whetherunimodalorthebimodal twasmoreconsistentwiththedataandAkaike weightstode neconditionalprobabilities( WagenmakersandFarrell,2004 ).Continuousoutcomeandothermeasureswerecomparedwithtwotailedindependent Table1 Hyponatremicpatientswithconcentrated( 4 p Osm)versusdilute( o p Osm)urineaCharacteristicsConcentratedDilute P value No.ofcases20340.06 Age,mean(S.D.),years46.1(12.1)41.5(11.4)0.18 Gender Male,no.11210.62 Female,no.913 Plasmasodium,mean(S.D.),mmol/L116.2(6.3)115.4(6.7)0.67 Plasmaosmolality,mean(S.D.),mOsm/kg246.1(14.9)( n ¼ 18)239.0(16.2)( n ¼ 27)0.13 Urineosmolality,mean(S.D.),mOsm/kg443(160)( n ¼ 19)138.8(59.5)( n ¼ 27) o 0.001 Urinespeci cgravity,mean(S.D.)1.012(0.005)( n ¼ 6)1.004(0.003)( n ¼ 14) o 0.001 Urinesodium,mean(S.D.),mEq/L73.4(55.1)( n ¼ 13)31.0(23.5)( n ¼ 16)0.02 WaterloadingtestbImpaired,no.200.14cNormal,no.12 PolydipsicdYes,no.8200.61 No,no.23 Durationofmentaldisorder,mean(S.D.),years13.6(9.8)( n ¼ 12)13.3(5.7)( n ¼ 10)0.93 Psychoticexacerbation Yes,no.690.20 No,no.31 DDD-equivalentdosage,mean(S.D.)3.6(8.0)( n ¼ 16)1.8(1.5)( n ¼ 14)0.40 Currentdrugduration,mean(S.D.),days46.4(45.6)( n ¼ 10)388.2(1023)( n ¼ 10)0.30 Antipsychotic Firstgeneration,no.1729 Secondgeneration,no.330.447 Combinedorunclassi ed,no.02 DechallengeeNegative,no.11120.50 Positive,no.21 Rechallenge Positive,no.500.17 Negative,no.01 AdverseDrugReaction Probable,no.820.003cPossible,no.1232aThe10caseswithrecognizedcausesofhyponatremiawereexcludedfromthisanalysis.bAbilitytoexcreteanoralwaterloadperauthor'scriteria.cAdjustedforexpectedcellsizebyFishersexacttest.dPerauthor'sassessmentonly.eWhetherstoppingmedicationledtoworseningofhyponatremia. W.Atsariyasing,M.B.Goldman/PsychiatryResearch217(2014)129 – 133 131

PAGE 4

sample t -tests,whilePearson'sChiSquarewasconductedforcategoricaldata,applying Fisher'sexacttestifexpectedcellcountswerelessthan ve.Toaddressthepossible roleofraterbiasinscoringwhetheracasewasanadversedrugreaction,wealso comparedgroupsonNaranjomeasuresforcasesonlyratedby Meulendijksetal(2010) .3.Results Datafor64hyponatremicpsychoticinpatientsreceiving antipsychotictherapy,eachwithconcurrentlymeasuredurine concentration,wereextractedfrom44articles( Fig.1 ).Tenhad recognizedcausesfortheirhyponatremiaandwereexcludedfrom furtheranalysis( SeeSupplementaryTableS1 ).Thirty-fourofthe remaining54caseshaddiluteurineand20hadconcentrated urine( P ¼ 0.06)( Table1 ).Thetwogroupsresembledeachotheron mostdemographic,clinicalandplasmameasures.Inparticular, meanageandplasmaosmolalityweresimilar(concentrated group:246,dilutegroup:239mOsm/kg)andthevastmajorityof patientsinbothgroupswerenotedtobepolydipsic.Urinesodium concentrationwashigherinsubjectswithconcentratedurine consistentwiththeirmorepersistentAVPactivity,valuesinthose withdiluteurinewerestillelevatedsuggestiveofresidualAVP activity. Abimodalmodel tthedistributionofurineosmolalities (AkaikeInformationCriteria:604.7,BayesianInformationCriteria: 613.9)betterthanaunimodalmodel(AIC:615.3,BIC:619.0; weight0.0049)withanintersectionat219mOsm/kg( Fig.2 ), suggestingtwodifferentprocesseswereresponsibleforthe hyponatremia.Concentratedurinewasmorefrequentlyassociated withadversedrugreactionsthandiluteurine( Table1 ).Indeed, theoddsofhaving ‘ probable ’ versus ‘ possible ’ drug-induced hyponatremiawas6.8(95%CI(1.6 – 28.9)timeshigherinthose withconcentratedurine.Thisdifferenceremainedsigni cant whenonlythecasesratedby Meulendijksetal.(2010) were assessed(2¼ 4.22,d.f. ¼ 1, P ¼ 0.05).Ontheotherhand,a psychoticexacerbationonadmissionassociatedwithdiluteurine (OR4.5,95%CI ¼ 0.4 – 54.1).Thosewithconcentratedurine,on average,weretakingtwiceashighadoseofneurolepticbutwere onitforonly1/8thaslongasthosewithdiluteurine.Becauseof markedintersubjectvariationthesedifferencesdidnotapproach signi cance(i.e. P 4 0.30)( Table1 ).All veofthepatientswith concentratedurineredevelopedhyponatremiawhenrechallenged, whilethesinglepatientwithdiluteurinedidnot.Becauseofthe smallnumbersinvolved,thisonlyshowedatrendtowardsigni cance( P ¼ 0.17).Noneoftheselatterdifferences,however, reachedstatisticalsigni cance.Whileveryfewpatientsineither groupredevelopedseverehyponatremiawhentheirantipsychotic medicationwasstopped,thesepatientsweretypically uid restrictedduringtheirhospitalization. 4.Discussion Life-threateninghyponatremiawas rstidenti edinpsychotic patientsinthe1930s( Barahal,1938 )andoverthesubsequent yearshasbeenassociatedwithbothantipsychoticmedicationand exacerbationsoftheunderlyingpsychiatricdisorder( Meulendijks etal.,2010 ).Nopublishedguidelinesfordistinguishingthesetwo possibilitieshavebeenreported.Intheabsenceofcontrolledtrials, wecompletedasystematicreviewofcasereportsandcaseseries intheEnglishliterature.Weeliminatedcaseswithrecognized causesofhyponatremialeaving54casesofunexplainedhyponatremiainmedicatedpatientswithaprimarypsychoticdisorder, eachhavingconcurrentmeasuresofurineconcentration( Fig.1 ). Wefoundthatthedistributionofurineosmolalitiesacross thesecases tabimodaldistributionwhichintersectedat 219mOsm/kg,suggestiveoftheirbeingtwodistinctprocesses ( Fig.2 ).Asweanticipated( Targowla,1923;Viewegetal.,1986; Goldmanetal.,1988,1997;Kishimotoetal.,1989;Delvaetal., 1990;Ohsawaetal.,1993 ),probabledrug-inducedhyponatremia occurredsixtimesasoftenincaseswithconcentratedurineand psychoticexacerbationsoccurredfourtimesasoftenincaseswith diluteurine( Table1 ).Whiledurationofdrugtreatmentwasabout one-eighthaslong,medicationdosagewastwiceashigh,and recurrenceofhyponatremiauponrechallengeoccurredexclusively inthosewithconcentratedurine,thedifferencesbetweenthose withconcentratedanddiluteurinedidnotreachsigni cance.Still, takenasawholethesedataallpointtoagreaterrolefor antipsychoticmedicationinthosewithconcentratedurine.In contrastpolydipsiawasaboutasfrequentlyobservedinboth categoriesconsistentwithothers' ndingsthatthisisanunreliablemeasureofincreasedintake( deLeonetal.,1996 ). Thereareseverallimitationstothisstudy.Theresultsarebased onareviewofcasereports,raisingissuesaboutthegeneralizability,aswellasreliabilityandcomparabilityofthedata.For instance,samplingofurinelikelyvariedinrelationshiptoconcurrentmeasuresofplasmaconcentration.Theconclusionthat urineconcentrationisbi-modalassumesthepooled ndingsfrom thesecasereportsconstituteasampleofthepopulationof interest.Theassessmentofwhetherapsychoticexacerbationor polydipsiawaspresentonadmissionassumesthatauthorswere equallyreliableanddiligentinmakingthisassessment.These assumptionsareundoubtedlyviolated,butwhatmaybereasonableistoassertthereisunlikelytobeasystematicbiasfavoring onepossibilityortheother. Whilethemajorityofcasesof ‘ probable ’ antipsychotic-induced hyponatremiaoccurredwithconcentratedurine,twocasesdid occurwithdiluteurine( Table1 ).Wesuspectthathyponatremia mayinfactarisefromantipsychotictreatmentinsomepatientswithdiluteurine.Possiblemechanismsareanticholinergicinducedpolydipsiasecondarytodrymouth( Mannesseetal., 2010 )inapatientwithresetosmostat,ormedication-induced resetosmostatduetohypotensioninapatientwithprimary polydipsia( Robertson,2006;Meulendijksetal.,2010 ).Indeed,in oneofthetwocasesthepatientwasreceivingthioridazinewhich haspotentanticholinergiceffectswhichcouldinducepolydipsia. Intheotherthepatientwasreceivingloxapine,whichisanalphaadrenergicantagonicthatinduceshypotensionandpotentially Fig.2. TheresultsofaGaussianmixturemodel tof2-gaussianstotheurine osmolalitydatainmedicatedpsychoticpatientswithunexplainedhyponatremia. Theunimodalmodelisshownintheinset.Thelikelihoodthebimodal tthedata betterwas o 0.005. W.Atsariyasing,M.B.Goldman/PsychiatryResearch217(2014)129 – 133 132

PAGE 5

resetosmostat.Hence,urineconcentrationisunlikelytobea completelyreliablemarker,andotherclinicalhistoryand ndings shouldbeconsideredaswell. Werecommendthatclinicianspresentedwithanacutely hyponatremicpsychoticpatienttakeacarefulhistory,determine concurrenturineandplasmaosmolality,imposetargeted uid restrictionandrigorouslysearchforreversiblecauses( Viewegand Leadbetter,1997;Siegel,2008 ).Whilemethodologicissuessummarizedabovelimittheapplicabilityofthe ndings,theydo providesomeguidanceformanagingpatientswithhyponatremia whichcannotbeattributedtootherrecognizedcauses.Hence,we recommendifurineisconcentratedthecurrentantipsychoticbe switchedtoanotherstructuralclass(withoutanticholinergic properties).Ifurineisdiluteandtheantipsychoticdoseis subtherapeutic,werecommendthedosebeincreased.Ifthedose istherapeuticanotheragentshouldbeprescribed.Ifhyponatremia doesnotrapidlyresolve,theclinicianshouldreviewthedifferentialdiagnosisandconsiderclozapinetreatment( Canusoand Goldman,1999 ).Patientsshouldbeobservedcloselyuntilthey tolerateadlibaccesstowaterwithoutdevelopingsymptomatic hyponatremia. Con ictofinterest Dr.GoldmanandDr.Atsariyasingcontributedequallytothis work.Dr.GoldmanandDr.Atsariyasinghadfullaccesstoallofthe datainthestudyandtakeresponsibilityfortheintegrityofthe dataandtheaccuracyofthedataanalysis.Dr.Goldmanwasa recipientofgrantfundsfromOtsukaforstudieswithtolvaptan (Samsca).Dr.Atsariyasingreportsnocon ictofinterest. AppendixA.Supplementaryinformation Supplementarydataassociatedwiththisarticlecanbefoundin theonlineversionat http://dx.doi.org/10.1016/j.psychres.2014.03.021 . ReferencesATC/DDDIndex,2012.WHOCollaboratingCenterforDrugStatisticsMethodology. www.whocc.no/atc_ddd_index/ (accessed16.10.12.). Barahal,H.S.,1938.Waterintoxicationinamentalcase.PsychiatryQuaterly12, 767 – 771 . Blum,A.,1984.Thepossibleroleoftobaccocigarettesmokinginhyponatremiaof long-termpsychiatricpatients.JournaloftheAmericanMedicalAssociation 252,2864 – 2865 . Canuso,C.M.,Goldman,M.B.,1999.Clozapinerestoreswaterbalanceinschizophrenicpatientswithpolydipsia-hyponatremiasyndrome.JournalofNeuropsychiatryandClinicalNeurosciences11,86 – 90 . deLeon,J.,Dadvand,M.,Canuso,C.,Odom-White,A.,Stanilla,J.,Simpson,G.M., 1996.Polydipsiaandwaterintoxicationinalong-termpsychiatrichospital. BiologicalPsychiatry1,28 – 34 . Delva,N.J.,Crammer,J.L.,Lawson,J.S.,Lightman,S.L.,Sribney,M.,Weier,B.J.,1990. Vasopressininchronicpsychiatricpatientswithprimarypolydipsia.British JournalofPsychiatry157,703 – 712 . Dubovsky,S.L.,Grabon,S.,Berl,T.,Schrier,R.W.,1973.Syndromeofinappropriate secretionofantidiuretichormonewithexacerbatedpsychosis.Annalsof InternalMedicine79,551 – 554 . Goldman,M.B.,Luchins,D.J.,Robertson,G.L.,1988.Mechanismsofalteredwater metabolisminpsychoticpatientswithpolydipsiaandhyponatremia.New EnglandJournalofMedicine318,397 – 403 . Goldman,M.B.,Robertson,G.L.,Luchins,D.J.,Hedeker,D.,Pandey,G.N.,1997. Psychoticexacerbationsandenhancedvasopressinsecretioninschizophrenic patientswithhyponatremiaandpolydipsia.ArchivesofGeneralPsychiatry54, 443 – 449 . Goldman,M.B.,Gnerlich,J.,Hussain,N.,2007.Neuroendocrineresponsestoacold pressorstimulusinpolydipsichyponatremicandinmatchedschizophrenic patients.Neuropsychopharmacology32,1611 – 1621 . Goldman,M.B.,2009.Themechanismoflife-threateningwaterimbalancein schizophreniaanditsrelationshiptotheunderlyingpsychiatricillness.Brain ResearchReviews61,210 – 220 . Goldman,M.B.,Wang,L.,Wachi,C.,Daudi,S.,Csernansky,J.,Marlow-O'Connor,M., Keedy,S.,Torres,I.,2011.Structuralpathologyunderlyingneuroendocrine dysfunctioninschizophrenia.BehaviouralBrainResearch218,106 – 113 . Hariprasad,M.K.,Eisinger,R.P.,Nadler,I.M.,Padmanabhan,C.S.,Nidus,B.D.,1980. Hyponatremiainpsychogenicpolydipsia.ArchivesofInternalMedicine140, 1639 – 1642 . Hawken,E.R.,Crookall,J.M.,Reddick,D.,Millson,R.C.,Milev,R.,Delva,N.,2009. Mortalityovera20-yearperiodinpatientswithprimarypolydipsiaassociated withschizophrenia:aretrospectivestudy.SchizophreniaResearch107, 128 – 133 . Hoskins,R.G.,Sleeper,F.H.,1933.Organicfunctionsinschizophrenia.Archivesof NeurologyandPsychiatry(AMA)30,123 – 140 . Ismail,Z.,Syms,J.,Blumberger,D.,George,T.P.,2010.Vareniclineinducedpolydipsiaandhyponatremiainapatientwithschizophrenia.Schizophrenia Research119,268 . Josiassen,R.C.,Goldman,M.,Jessani,M.,Shaughnessy,R.A.,Albazzaz,A.,Lee,J., Ouyang,J.,Orlandi,C.,Czerwiec,F.,2008.Double-blind,placebo-controlled, multicentertrialofavasopressinV2-receptorantagonistinpatientswith schizophreniaandhyponatremia.BiologicalPsychiatry64,1097 – 1100 . Josiassen,R.C.,Curtis,J.L.,Shaughnessy,R.A.,Filmyer,D.M.,Geboy,A.G.,Skuban,N., Czerwiec,F.,2012.Vaptans:apotentialnewapproachfortreatingchronic hyponatremiainpsychoticpatients.ClinicalSchizophreniaandRelatedPsychoses6,21 – 26. Kelly,W.N.,Arellano,F.M.,Barnes,J.,Bergman,U.,Edwards,R.I.,Fernandez,A.M., Freedman,S.B.,Goldsmith,D.I.,Huang,K.A.,Jones,J.K.,McLeay,R.,Moore,N., Stather,R.H.,Trenque,T.,Troutman,W.G.,vanPuijenbroek,E.,Williams,F., Wise,R.P.,2007.Guidelinesforsubmittingadverseeventreportsforpublication.DrugSafety30,367 – 373 . Kishimoto,T.,Hirai,M.,Ohsawa,H.,Terada,M.,Matsuoka,I.,Ikawa,G.,1989. Mannersofargininevasopressinsecretioninschizophrenicpatients — with referencetothemechanismofwaterintoxication.JapaneseJournalofPsychiatryandNeurology43,161 – 169 . Letmaier,M.,Painold,A.,Holl,A.K.,Vergin,H.,Engel,R.,Konstantinidis,A.,Kasper,S., Grohmann,R.,2012.Hyponatraemiaduringpsychopharmacologicaltreatment: resultsofadrugsurveillanceprogramme.InternationalJournalofNeuropsychopharmacology15,739 – 748 . Liamis,G.,Milionis,H.,Elisaf,M.,2008.Areviewofdrug-inducedhyponatremia. AmericanJournalofKidneyDiseases52,144 – 153 . Mannesse,C.K.,vanPuijenbroek,E.P.,Jansen,P.A.F.,vanMarum,R.J.,Souverein,P.C., Egberts,T.C.G.,2010.Hyponatraemiaasanadversedrugreactionofantipsychoticdrugs:acase-controlstudyinVigiBase.DrugSafety33,569 – 578 . Meulendijks,D.,Mannesse,C.K.,Jansen,P.A.F.,vanMarum,R.J.,Egberts,T.C.G.,2010. Antipsychotic-inducedhyponatraemia:asystematicreviewofthepublished evidence.DrugSafety33,101 – 114 . Moher,D.,Liberati,A.,Tetzlaff,J.,Altman,D.G.,2009.ThePRISMAGroup(2009). Preferredreportingitemsforsystematicreviewsandmeta-analyses:the PRISMAstatement.AnnalsofInternalMedicine151,264 – 269 . Naranjo,C.A.,Busto,U.,Sellers,E.M.,Sandor,P.,Ruiz,I.,Roberts,E.A.,Janecek,E., Domecq,C.,Greenblatt,D.J.,1981.Amethodforestimatingtheprobabilityof adversedrugreactions.ClinicalPharmacologyandTherapeutics30,239 – 245 . Ohsawa,H.,Kishimoto,T.,Shimayoshi,N.,Matsumura,K.,Tahara,K.,Kitera,K., Higashiura,N.,Noriyama,Y.,Matsumoto,H.,Hirai,M.,1993.Atrialnatriuretic peptideandargininevasopressinsecretioninschizophrenicpatients.Acta PsychiatricaScandinavica88,130 – 134 . Renneboog,B.,Musch,W.,Vandemergel,X.,Manto,M.U.,Decaux,G.,2006.Mild chronichyponatremiaisassociatedwithfalls,unsteadiness,andattention de cits.AmericanJournalofMedicine119,71,e1-71.e8 . Robertson,G.L.,2006.Regulationofargininevasopressininthesyndromeof inappropriateantidiuresis.AmericanJournalofMedicine119(7Suppl.1), S36 – S42 . Ross,E.J.,Christie,S.B.,1969.Hypernatremia.Medicine48,441 – 473 . Siegel,A.J.,2008.Hyponatremiainpsychiatricpatients:updateonevaluationand management.HarvardReviewofPsychiatry.16(1),13 – 24 . Targowla,R.,1923.Destroublesfonctionneldureindanslesmaladiesmentales. L'excretiondel'eau(Kidneymalfunctionandmentalillness:waterexcretion). BulletinsetmémoiresdelaSociétémédicaledeshôpitauxdeParis47, 1711 – 1715 . Vieweg,W.V.,David,J.J.,Rowe,W.T.,Peach,M.J.,Veldhuis,J.D.,Spradlin,W.W.,1986. Correlationofcigarette-inducedincreaseinserumnicotinelevelswitharginine vasopressinconcentrationsinthesyndromeofself-inducedwaterintoxication andpsychosis(SIWIP).CanadianJournalofPsychiatry31,108 – 111 . Vieweg,W.V.,Leadbetter,R.A.,1997.Polydipsia-hyponatraemiasyndrome:epidemiology,clinicalfeaturesandtreatment.CNSDrugs7,121 – 138 . Voinescu,G.C.,Shoemaker,M.,Moore,H.,Khanna,R.,Nolph,K.D.,2002.The relationshipbetweenurineosmolalityandspeci cgravity.AmericanJournalof theMedicalSciences323,39 – 42 . Wagenmakers,E.J.,Farrell,S.,2004.AICmodelselectionusingAkaikeweights. PsychonomicBulletinandReview11,192 – 196 . Williams,S.T.,Kores,R.C.,2011.Psychogenicpolydipsia:comparisonofacommu-nitysamplewithaninstitutionalizedpopulation.PsychiatryResearch187, 310 – 311 . W.Atsariyasing,M.B.Goldman/PsychiatryResearch217(2014)129 – 133 133