Citation
Monolinguals and Bilinguals' Attentional Control in the Presence of Cognitive and Emotional Distraction

Material Information

Title:
Monolinguals and Bilinguals' Attentional Control in the Presence of Cognitive and Emotional Distraction
Creator:
Pelham, Sabra D
Place of Publication:
[Gainesville, Fla.]
Florida
Publisher:
University of Florida
Publication Date:
Language:
english
Physical Description:
1 online resource (104 p.)

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Psychology
Committee Chair:
ABRAMS,LISE
Committee Co-Chair:
KEIL,ANDREAS
Committee Members:
KERTES,DARLENE
KAAN,EDITH
Graduation Date:
8/9/2014

Subjects

Subjects / Keywords:
Attention ( jstor )
Bilingualism ( jstor )
Business executives ( jstor )
Cognition ( jstor )
Emotional expression ( jstor )
Experimentation ( jstor )
Factorials ( jstor )
Memory ( jstor )
Mental stimulation ( jstor )
Words ( jstor )
Psychology -- Dissertations, Academic -- UF
attention -- bilingualism -- cognitive -- control -- emotion
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:
Past research has shown that bilinguals are better able to control attention in the presence of distraction than monolinguals. However, distractors in previous studies have always been cognitive in nature; the targets require a particular response, but the distractors also cue a response that may match or differ from the response required by the target. The purposes of the present study were (1) to investigate whether bilinguals also have an advantage over monolinguals in controlling attention in the presence of emotional distractors, and (2) to assess the degree to which bilinguals control attention as a function of the type of distraction. Monolingual and bilingual participants completed cognitive tasks in the presence of cognitive distraction (Experiment 1), emotional distraction (Experiment 2), and combined cognitive and emotional distraction (Experiment 3). Two tasks, a flanker task and a digit parity task, were used in each experiment, and the tasks were parallel across the different experiments such that only the nature of the distractors differed. Monolinguals' and bilinguals' performance were compared via the speed and accuracy of their responses as well as the size of their conflict effects. For cognitive distraction, a conflict effect is the difference in response times (RTs) between incongruent trials (trials with high distraction) and congruent trials (trials with low distraction). For emotional distraction, a conflict effect is the difference in RTs between emotional trials (trials with an emotional picture present) and neutral trials (trials with a neutral picture present). A bilingual advantage in controlling attention in the presence of cognitive distraction emerged on the flanker task in Experiment 1, where bilinguals' RTs were faster overall than monolinguals. In Experiment 2, when distractors were only emotional pictures, there was no bilingual advantage on either task, as both groups' RTs were equivalently slowed when emotional pictures were present. When distractors were cognitive and emotional in Experiment 3, bilinguals manifested an advantage in the flanker task such that they were unaffected by the presence of the emotional distractor pictures in contrast to monolinguals, and bilinguals had smaller conflict effects than monolinguals when trials were accompanied by emotional distractor pictures. These results demonstrate that bilinguals have an advantage over monolinguals in controlling attention in the presence of emotional distractors when the simultaneous demands on cognitive control are high. These findings are interpreted within a conflict-monitoring framework, where dorsal anterior cingulate cortex activation in response to information processing conflicts is necessary for the emergence of bilingual advantages. ( 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: ABRAMS,LISE.
Local:
Co-adviser: KEIL,ANDREAS.
Statement of Responsibility:
by Sabra D Pelham.

Record Information

Source Institution:
UFRGP
Rights Management:
Copyright Pelham, Sabra D. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
968786263 ( OCLC )
Classification:
LD1780 2014 ( lcc )

Downloads

This item has the following downloads:


Full Text

PAGE 1

1 MONOLINGUALS AND BIL INGUALS' ATTENTIONAL CONTROL IN THE PRESENCE OF C OGNITIVE AND EMOTION AL DISTRACTION By SABRA D. PELHAM A DISSERTATION PRESENTED TO THE GRA DUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL F ULFILLMEN T OF THE REQUIREMENTS FOR THE DEGREE OF DOCTORATE OF PHILOSO PHY UNIVERSITY OF FLORIDA 2014

PAGE 2

2 © 201 4 Sabra D. Pelham

PAGE 3

3 To my husband, whose support has been unwavering, whose pride in me has been unmatched, and whose belief in me has never failed

PAGE 4

4 ACKNOWLEDGMENTS I want to thank my mentor, Dr. Lise Abrams, for all her support , insight, and encouragement th r ough this process , my committee members for their suggestions, which improved this project, and the many research assistants who helped to co llect data for the project.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ ........... 4 LIST OF TABLES ................................ ................................ ................................ ...................... 7 LIST OF FIGURES ................................ ................................ ................................ .................... 8 ABSTRACT ................................ ................................ ................................ ............................... 9 CHAPTER 1 OVERVIEW ................................ ................................ ................................ ...................... 11 Executive Function in Bilinguals ................................ ................................ ........................ 12 Processing Emotional Stimuli ................................ ................................ ............................. 14 Emotional Language ................................ ................................ ................................ ... 15 Emotional Pictures ................................ ................................ ................................ ...... 17 Duration of Emotional Effects ................................ ................................ ..................... 18 Emotional Processing in Bilinguals ................................ ................................ ............. 19 The Dorsal Anterior Cingulate Cortex ................................ ................................ ................ 20 Tasks with Cognitive and/or Emotional Distraction ................................ ............................ 23 2 HYPOTHESES ................................ ................................ ................................ .................. 27 Hypothesis 1: Cognitive distraction in a cognitive task ................................ ....................... 27 Hypothesis 2: Emotional distraction in a cognitive task ................................ ...................... 28 Hypothesis 3: Emotional and cognitive distraction in a cognitive task ................................ 29 Hypothesis 4: Carryover effects from emotional trials ................................ ........................ 30 3 EXPERIMENT 1 COGNITIVE DISTRACTIO N ................................ ............................ 31 Method ................................ ................................ ................................ ............................... 31 Pa rticipants ................................ ................................ ................................ ................. 31 Materials ................................ ................................ ................................ ..................... 33 Procedure ................................ ................................ ................................ .................... 35 Results ................................ ................................ ................................ ............................... 36 Flanker Task ................................ ................................ ................................ ............... 36 Response times ................................ ................................ ................................ ..... 36 Accuracy ................................ ................................ ................................ .............. 37 Digit Parity Task ................................ ................................ ................................ ......... 38 Response times ................................ ................................ ................................ ..... 38 Accuracy ................................ ................................ ................................ .............. 38 Discussion ................................ ................................ ................................ ................... 39 4 EXPERIMENT 2 EMOTIONAL DISTRACTIO N ................................ .......................... 46

PAGE 6

6 Method ................................ ................................ ................................ ............................... 46 Participants ................................ ................................ ................................ ................. 46 Materials ................................ ................................ ................................ ..................... 47 Procedure ................................ ................................ ................................ .................... 48 Results ................................ ................................ ................................ ............................... 49 Participants' Valence Ratings ................................ ................................ ...................... 49 Emotional Arrows Task ................................ ................................ ............................... 50 Response times ................................ ................................ ................................ ..... 50 Accuracy ................................ ................................ ................................ .............. 50 Emotional Digit Parity Task ................................ ................................ ........................ 51 Response times ................................ ................................ ................................ ..... 51 Accuracy ................................ ................................ ................................ .............. 51 Valence Carryover Effects on Neutral Filler Trials ................................ ...................... 52 Emotional arrows task ................................ ................................ .......................... 53 Emotional digit parity task ................................ ................................ .................... 53 Discussion ................................ ................................ ................................ .......................... 54 5 EXPERIMENT 3 COGNITIVE AND EMOTIO NAL DISTRACTION ........................... 63 Method ................................ ................................ ................................ ............................... 63 Participants ................................ ................................ ................................ ................. 63 Materials ................................ ................................ ................................ ..................... 6 4 Procedure ................................ ................................ ................................ .................... 64 Results ................................ ................................ ................................ ............................... 64 Participants' Valence Ratings ................................ ................................ ...................... 64 Flanker Task with Emotional Distractors ................................ ................................ ..... 65 Response times ................................ ................................ ................................ ..... 65 Accuracy ................................ ................................ ................................ .............. 68 Digit Parity Task with Emotional Distractors ................................ ............................... 69 Response times ................................ ................................ ................................ ..... 69 Accuracy ................................ ................................ ................................ .............. 70 Valence Carryover Effects on Neutral Filler Trials ................................ ...................... 71 Experiment 3 flanker task with emotional distractors ................................ ......... 71 Experiment 3 digit parity task with emotional distractors ................................ ... 72 Discus sion ................................ ................................ ................................ .......................... 72 6 GENERAL DISCUSSION ................................ ................................ ................................ . 84 APPENDIX ................................ ................................ ................................ .............................. 95 LIST OF REFERENCES ................................ ................................ ................................ .......... 97 BIOGRAPHICAL SKETCH ................................ ................................ ................................ ... 104

PAGE 7

7 LIST OF TABLES Table page 3 1 Monolinguals' and bilinguals' self reported language proficiency and exposure ............. 42 3 2 Monolinguals and bilinguals' demographic characteristics and cognitive tests ................ 42 3 3 Experiment 1 Cognitive flanker task response times (in ms) ................................ ........ 43 3 4 Experiment 1 Cognitive flanker task accuracy (in %) ................................ .................. 43 3 5 Experiment 1 Cognitive digit parity task response times (in ms) ................................ . 43 3 6 Experiment 1 Cognitive digit parity task accuracy (in %) ................................ ............ 43 4 1 Monolinguals' and bilinguals' self reported language proficiency and exposure ............. 59 4 2 Monolinguals and bilinguals' demographic characteristics and cognitive tests ................ 59 4 3 Mean valence for pictures in each picture set ................................ ................................ . 60 4 4 Mean arousal for pictures in each picture set ................................ ................................ .. 60 4 5 Experiment 2 Emotional flanker task response times (in ms) ................................ ....... 60 4 6 Experiment 2 Emotional flanker task accuracy (in %) ................................ ................. 60 4 7 Experiment 2 Emotional digit parity task response times (in ms) ................................ 60 4 8 Experiment 2 Emotional digit parity task accuracy (in %) ................................ ........... 61 5 1 Monolinguals' and bilinguals' self reported language proficiency and exposure ............. 77 5 2 Monolinguals and bilinguals' demographic characteris tics and cognitive tests ................ 77 5 3 Experiment 3 Cognitive and emotional flanker task response times (in ms) ................. 78 5 4 Experiment 3 Conflict effects for each level of distractor picture valence (in ms) ........ 78 5 5 Experiment 3 Cognitive and emotional flanker task accuracy (in %) ........................... 78 5 6 Experiment 3 Cognitive and emotional digit parity task response times (in ms) .......... 79 5 7 Experiment 3 Cognitive and emotional digit parity task accuracy (in %) ..................... 79 6 1 Summary of results ................................ ................................ ................................ ........ 94

PAGE 8

8 LIST OF FIGURES Figure page 3 1 Target and distractors for Exp eriment 1, cognitive ................................ ......................... 44 3 2 Time course for digit parity task ................................ ................................ .................... 44 3 3 Bilinguals' and monolinguals' flanker task RTs in Experimen t 1 ................................ .... 45 4 1 Targets and distractors for Experiment 2, emotional distraction ................................ ..... 62 5 1 Targets and distractors for Experiment 3, cogn itive and emotional distraction ............... 80 5 2 Monolinguals' flanker task RTs in Experiment 3 ................................ ............................ 81 5 3 Bilinguals' flanker task RTs in Exp eriment 3 ................................ ................................ . 82 5 4 The interaction of digit distractor type and distractor picture valence on accuracy in Experiment 3. ................................ ................................ ................................ ................ 83 5 5 Exa mple of a Gabor patch ................................ ................................ .............................. 83

PAGE 9

9 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Doctorate of Philosophy MONOLINGUALS AND BI LINGUALS' ATTENTIONA L CONTROL IN THE PRESENCE OF C OGNITIVE AND EMOTION AL DISTRACTION By Sabra D. Pelham August 2014 Chair: Lise Abrams Major : Psychology Past research has shown that bilinguals are better able to control attention in the presence of di straction than monolinguals. However, distractors in previous studies have always been cognitive in nature; the targets require a particular response, but the distractors also cue a response that may match or differ from the response required by the target . The purposes of the present study were (1) to investigate whether bilinguals also have an advantage over monolinguals in controlling attention in the presence of emotional distractors, and (2) to assess the degree to which bilinguals control attention as a function of the type of distraction. Monolingual and bilingual participants complete d cognitive tasks in the presence of cognitive distraction (Experiment 1), emotional distraction (Experiment 2), and combined cognitive and emotional distraction (Experi ment 3). Two tasks, a flanker task and a digit parity task, were used in each experiment, and the tasks were parallel across the different experiments such that only the nature of the distractors differ ed . Monolinguals' and bilinguals' performance were com pared via the speed and accuracy of their responses as well as the size of their conflict effects. For cognitive distraction, a conflict effect is the difference in response times (RTs) between incongruent trials (trials with high distraction) and congruen t trials (trials with low distraction).

PAGE 10

10 For emotional distraction, a conflict effect is the difference in RT s between emotional trials (trials with an emotional picture present) and neutral trials (trials with a neutral picture present). A bilingual advant age in controlling attention in the presence of cognitive distraction emerged on the flanker task in Experiment 1 , where bilinguals ' RT s were faster overall than monolinguals. In Experiment 2, when distractors were only emotional pictures, there was no bil ingual advantage on either task, as both groups' RTs were equivalently slowed when emotional pictures were present. When distractors were cognitive and emotional in E xperiment 3 , bilinguals manifested an advantage in the flanker task such that they were un affected by the presence of the emotional distractor pictures in contrast to monolinguals , and bilinguals had smaller conflict effects than monolinguals when trials were accompanied by emotional distractor pictures. These results demonstrate that bilingual s have an advantage over monolinguals in controlling attention in the presence of emotional distractors when the simultaneous demands on cognitive control are high. These findings are interpreted within a conflict monitoring framework , where dorsal anterio r cingulate cortex activation in response to information processing conflicts is necessary for the emergence of bilingual advantages.

PAGE 11

11 CHAPTER 1 OVERVIEW , people who speak different languages are coming into co ntact increasingly often. Bilingual or multilingual people have an advantage over monolinguals in that they can communicate with more people simply because they are not limited to communicating in only one language. However, the ability to communicate more broadly is not the only advantage that accompanies bilingualism. One finding in recent research is that bilinguals often exhibit executive function benefits on nonverbal attention tasks relative to monolinguals. Tasks that require participants to respond to one aspect of a stimulus while ignoring distraction from other stimuli demonstrate a bilingual advantage: Bilinguals often respond more quickly and accurately and are less slowed by the distraction than monolinguals (Bialystok, Craik, Klein, & Viswanath an, 2004; Bialystok, Craik, & Luk, 2008; Carlson & Meltzoff, 2008, Costa, Hernández, & Sebastián Gallés, 2008; Linck, Hoshino, & Kroll, 2008; Luk, De Sa, & Bialystok, 2011; Martin Rhee & Bialystok, 2008). This bilingual advantage in attentional control in the presence of distraction has only been demonstrated when the distraction is cognitive in nature. However, emotional stimuli have also been shown to strongly engage attention and can serve as potent distractors (Algom, Chajut, & Lev, 2004; Arnell, Killma n, & Fijavz, 2007; Ashley & Swick, 2009; Bertels, Kolinsky, Pietrons, & Morais, 2011; Colbeck & Bowers, 2012; Dolcos & McCarthy, 2006; Fernandes, Koji, Dixon, & Aquino, 2011; MacKay et al., 2004; Mather et al., 2006; Schimmack, 2005; Thomas, 2006; Yamanski , LaBar, & McCarthy, 2002). The present research test ed Specifically, the research question of interest wa s whether bilinguals show an advantage over monolinguals in controlling attention only in the presence of cognitive distraction or whether that advantage extends to controlling attention in the presence of emotional distraction.

PAGE 12

12 Executive Function in Bilinguals Bilinguals' advantage in attentional control has been documented in various tasks that involve cognitive distraction, including the Stroop task, Simon task, and flanker tasks. In the Stroop task (e.g., Stroop, 1935), participants are required to name the color in which a target word is printed, and distraction occurs when the written wo rd spells a color different from the ink color. In the Simon task (e.g., Simon & Ruddell, 1967), participants press a button with their left or right hand to indicate the direction in which a target arrow is pointing. The distraction occurs when the locati on of the target (the side of the computer screen on which it appears) is incongruent relative to the response hand, such as a left pointing arrow on the right side of the computer screen. In flanker tasks, targets are typically a central chevron (< or >) or arrow that is surrounded by "flanker" chevrons or arrows , and participants use a left or rig ht handed response to indicate the target 's direction . The nature of the distraction is whether the flankers point in the same (congruent) or opposite (incongrue nt) direction as the target . The impact of cognitive distraction in these tasks is typically measured by calculating conflict effects , which provide a measure of interference from incongruent trials . One measure of conflict effects involves comparing respo nses to incongruent trials (e.g., a left pointing chevron surrounded by right pointing flankers) with responses to congruent trials, i.e., where the distractor information is consistent with the target (e.g., a left pointing chevron surrounded by other lef t pointing flanking chevrons ; e.g., Nieuwenhuis et al., 2006 ). A second measure of conflict effects compares incongruent trials to neutral trials (e.g., ). Conflict effects consistently emerge in these tasks such that people respond more slowly on incongruent trials than either neutral or congruent trials, showing that the presence of a cognitive distractor impaired performance.

PAGE 13

13 Research has shown that bilinguals often have smaller conflict effects than monolinguals due to their being less slowed t han monolinguals on incongruent trials, and they also frequently respond faster overall than monolinguals on all trial types. Both findings are thought to be indicative of a bilingual advantage, a better ability to control attention in the presence of cogn itive distraction than monolinguals , which speeds bilinguals' performance . Bialystok et al. (2008) found that bilinguals had significantly smaller Stroop effects than monolinguals because they were less slowed on incongruent trials. Similarly, Bialystok et al. (2004) and Bialystok et al. (2008) showed that bilinguals also had smaller conflict effects than monolinguals on a Simon task. Using a modified Attentional Network Task (ANT), which is a type of flanker task (Fan, McCandliss, Sommer, Raz, & Posner, 20 02), bilinguals again had smaller conflict effects than monolinguals (Costa et al., 2008; Pelham & Abrams, 2012). Other researchers have found no evidence of smaller conflict effects for bilinguals but instead found th at bilinguals responded faster overall on all trial types than monolinguals on tasks like the flanker task (Emmorey, Borinstein, Thompson, & Gollan, 2008 ) and the Simon task ( Martin Rhee & Bialystok, 2008 ) . In their review of studies investigating bilingual advantages, Costa et al. (2009) note d that the overall speed advantage for bilinguals is actually more common than smaller conflict effects. While both of these bilingual advantages occur for RTs , it is rare to find any effect of bilingualism on accuracy , which is likely due to ceiling effec ts. A ccuracy for both monolinguals and bilinguals is typically very high on tasks used to investigate the effect of bilingualism on control of attention, so accuracy is not usually an informative measure. It is important to note that the foregoing discuss ion of bilingual advantages has treated bilingualism as if it were a discrete and uniform variable, i.e., as if one is either fully bilingual or not at all bilingual all, and as if all bilinguals are the same. Furthermore, mu ch research

PAGE 14

14 investigating bilin gual advantages has defined bilinguals as those who learned both of their languages in childhood and who speak bot h of their languages regularly. However, bilinguals are extremely diverse and possess individual differences that might affect their cognitive function. Proficiency level is one obvious dimension along which bilinguals vary, but they may also vary in how often they speak each of their languages (how balanced they are), at what age they began learning and/or became proficient in their second lang uage, how often they switch between languages, in what contexts they use their two languages, and whether or not those contexts overlap or are distinct. The emergence of a n overarching bilingual advantage in attentional control is impressive given these di fferent ways in which bilinguals are bilingual and the different ways biliingualism has been defined across studies . However, w hile these findings demonstrate that bilinguals are better than monolinguals at controlling attention in the presence of cognitiv e distraction, no research has investigated whether they are better able to control attention in the presence of emotional distraction, i.e., when the stimuli that serve as distractors arouse an emotional response. In the next section, I review research re lated to emotional stimuli and how these stimuli have been shown to engage attentional resources, independent of bilingualism. Processing Emotional Stimuli According to the Motivated Attention theory of emotion, emotions are actually predispositions to ac tion (Lang, 1995), and as such they prepare individuals to respond to stimuli in the environment that are relevant for survival. Emotions prepare one for action by engaging one of two motivational systems. The appetitive system, which is engaged by positiv e and pleasant stimuli, creates a predisposition to approach such stimuli. The aversive system, which is engaged by negative and unpleasant stimuli, creates a predisposition to withdraw. This distinction between positive and negative stimuli is referred to as valence (Lang, Bradley, &

PAGE 15

15 Cuthbert, 2008). Another dimension along which emotional stimuli can vary is arousal, ranging from low to high, where low arousal is equated with feeling sleepy or relaxed, and high arousal is associated with feeling excited. Hence, a single stimulus is characterized both in terms of its valence and its arousal. For example, a picture of a garbage can will have a negative valence and will be low in arousal, while a picture of a person being stabbed will have a negative valence and will be high in arousal. Valence and arousal norms have been established for a large number of pictures in the International Affective Picture System, or IAPS (Lang et al., 2008), and similar norms for words in English have been established in the Affe ctive Norms for English Words, or ANEW (Bradley & Lang, 1999). Because emotional stimuli are relevant for survival, evolution has favored attentional systems that are easily and quickly engaged by such stimuli, and as a consequence, emotional stimuli ten d to capture and hold attention. That is, in any given environment, if stimuli appear that engage the appetitive or aversive motivational systems (i.e., emotional stimuli), such stimuli gain preferential access to conscious perception, attracting attention over (and often at the expense of) non emotional stimuli. The mo re arousing the stimulus , the more strongly it engages and hold s attention. Research with young adults using a variety of paradigms has demonstrated this capacity of emotional stimuli to attr act and hold attention. Emotional L anguage With respect to emotional words, participants are faster to indicate that a letter string is a word on lexical decision tasks if it is an emotional word than a neutral word (Kousta, Vinson, & Vigliocco, 2009), s uggesting that emotional words gain attention faster than neutral words. Participants also demonstrate superior recall for emotional words relative to neutral words on incidental recall tasks (Buchanan, Etzel, Adolphs, & Tranel, 2006; Jay, Caldwell Harris, & King, 2008), suggesting that emotional words hold attention longer than neutral words, which leads to

PAGE 16

16 better recall. Parallel findings have been demonstrated using an emotional Stroop task. Participants named the font color of neutral or emotional words , and they took longer to name the color of emotional words (Algom et al., 2004; MacKay et al., 2004), including taboo words (Bertels et al., 2011), an effect of emotion that occurs independently of whether trials are blocked or mixed (Ashley & Swick, 2009 ). Participants are also slower to name pictures in the presence of taboo distractor words in a picture word interference task (Dhooge & Hartsuiker, 2011). These results again suggest that the emotional words hold attention longer than neutral words, slowi RT s to name a word's font color or to name a picture. Research using rapid serial visual presentation (RSVP) also demonstrates the power of emotional words to capture attention via the attentional blink phenomenon (Anderson, 2005; Arnell e t al., 2007; Keil & Ihsshen, 2004; Colbeck and Bowers, 2012). In a RSVP task, words are presented on a computer screen in rapid sequence at a rate of 8 10 words per second, and participants are asked to identify particular target words (e.g., words written in green letters). The attentional blink refers to a failure to detect the second of two targets (T2) when it appears close in time to the first target (T1). The attentional blink has been interpreted as indicating that ongoing processing of T1 interferes attentional resources are limited: T2 cannot be perceived when presented within a few words after T1 because there are insufficient attentional resources to process T2 until processing of T1 is complete. Research has demonstrated that emotion interacts with the attentional blink such that when the T2 is an emotionally arousing positive, negative, or taboo word (and T1 is neutral), T2 is detected better than neutral words (e.g., Anderson, 2005; K eil & Ihssen, 2004). Conversely, Arnell et al. (2007) and Colbeck and Bowers (2012) found that when T1 is an emotionally

PAGE 17

17 arousing word, detection of neutral T2 words is worse than when T1 is a neutral word. Together, these studies demonstrate that when emo tional words serve as a second target, they capture target, they act like distractors relative to the second target, holding attention and worsening detection of T2. Emotional P ictures Research using emotionally arousing pictures has demonstrated that such pictures also capture and hold attention. Similarly to emotional words, they are remembered better than neutral stimuli when they serve as targets (Bradley, Greenwald, Petry, & Lang, 1992; Palom b a, Agrilli, & Mini, 1997), and this advantage in recalling emotional pictures is long lasting, evident even four weeks after the pictures are presented (Ham an n , Ely, Grafton, & Kilts, 1999). Also like emotional words, emotional pictures impair performance when they are presented as distractors in various cognitive tasks (Fernandes et al., 2011; Schimmack, 2005; Yamanski et al., 2002). With respect to memory tasks, emotional pictures decreased accuracy in a response del ayed working memory task (Dolcos & McCarthy, 2006) as well as a source memory task (Mather et al., 2006). Even in cognitive tasks that do not involve a memory component, emotional pictures still seem to capture attention, which affects task performance. Sc himmack (2005) found that the presence of emotional stimuli slowed performance when the task was cognitively simple (identify the location of a horizontal line, either above or below a distractor picture) or complex (decide which of two multiplication math problems has the larger product). Fernandes et al. (2011) used a digit parity task, where participants had to press one button if the digits matched in terms of parity (both odd or both even) and another button if they did not match. When an emotional pic ture appeared between the digits, performance on the digit parity task was

PAGE 18

18 impaired. All of these results demonstrate that emotional pictures engage attentional resources, resulting in a decrement in behavioral performance on a cognitive task. Duration of Emotional E ffects Electrophysiological experiments also demonstrate that emotional pictures capture attention and have the capacity to hold attention for a long time. Hajcak and Olvet (2008) showed participants negative, positive, and neutral pictures and measured their ERP recordings, specifically late positive potentials (LPP), which typically have a mid parietal distribution and are considered an index of attention to emotional stimuli. Analyses on each 200 ms time window beginning with presentation of e motional pictures showed that enhanced LPPs endured for 3000 ms following the offset of negative pictures and 2800 ms following the offset of positive pictures, indicating that not only do emotional pictures capture attention, but they also hold attention long after the pictures have disappeared. Behavioral results with the emotional Stroop task provide converging evidence that the effect of emotional stimuli in capturing attention is long lasting (Ashley & Swick, 2009; Bertels et al., 2011). Ashley and Sw ick (2009) used a block of trials where the neutral and emotional (negative) words were intermixed, presented in a 7 word sequence so that the word in the first position was either emotional or neutral and the following six words were all neutral. This man ipulation allows for an evaluation of the effect of an emotional word on responses to subsequent neutral words. Both young and older participants responded more slowly in naming the ink color of the second (neutral) word when the previous word was emotiona l relative to neutral, demonstrating a "carryover" effect of emotion to a subsequent nonemotional trial. For young adults, this carryover effect lasted even longer, occurring for all subsequent neutral words that followed the emotional words (except the on e in the 5th position ) , with both types of emotional words slowing color naming on a subsequent neutral word. While these studies show

PAGE 19

19 carryover effects of emotional words, I am unaware of any behavioral studies that have used emotional pictures to investi gate whether emotional stimuli have carryover effects to subsequent trials. Emotional P rocessing in B ilinguals Almost all of the research investigating the effect of emotion on attention has been conducted without consideration of whether participants wer e monolingual or bilingual. There are a few studies investigating the effect of emotion in bilinguals, which have demonstrated that García, Fraga, Sánchez Casa s, & Molero, 2010), an affective Simon type task using emotional words (Altarriba & Bainsight Brown, 2010), and an emotional Stroop task (Sutton, Altarriba, Gianico, & Bainsight Brown, 2007). However, these studies cannot conclusively determine the degree al. (2010) and Sutton at al. (2007) did not include a monolingual group for comparison, so whether emotional words hold attention to the same degree in bilingual s and monolinguals cannot be determined from these experiments. While Altarriba and Bainsight Brown (2010) found equivalent conflict effects for both monolinguals and bilinguals, bilinguals responded more slowly on all trials than monolinguals, which is in dicative of a lexical access deficit (Bialystok et al., 2008; Gollan, Montoya, Cera, & Sandoval , 200 8 ; Ivanova & Costa, 2008). The disadvantage in lexical access may hav e negated their executive function advantage in handling attention, resulting in equivalent conflict effects for bilinguals and monolinguals (instead of smaller conflict effects for bilinguals). Lastly, none of the studies used pictures (all used words), s o the influence of emotional distraction from pictures on bilinguals' cognitive performance is unknown. Using evidence from neuroimaging studies, described in the next section, I

PAGE 20

20 hypothesize that bilinguals will possess an advantage in controlling attentio n in the presence of emotional pictures. The Dorsal Anterior Cingulate Cortex Research across several domains provides reason to suspect that the bilingual advantage in attention may extend beyond inhibiting/suppressing interference from cognitive distract ion to inhibiting or suppressing interference from emotional distractors. Of particular relevance in this regard is the dorsal anterior cingulate cortex ( d ACC). The dACC is critical for controlling attentional demands from both cognitive and emotional stim uli. However, the dACC is also a brain region engaged in bilinguals under conditions of language conflict, i.e., when words in both languages are active. Because bilinguals use this area more regularly than monolinguals as a function of regular use of two languages, they may have an advantage in emotional processing by better suppression of interference from irrelevant emotional stimuli. First, research in the literature on attention has shown that the dACC is activated when completing effortful attention t asks, i.e., tasks that require top down control of attention engaged to focus on task relevant stimuli (Crottaz Herbette & Menon, 2006; Davis, Hutchinson, Lozano, Tasker, & Dostrovsky, 2000), and tasks that involve response conflict, i.e., tasks where irre levant stimuli cue a response that differs from the response required by target stimuli (Barch, Braver, Sabb, & Noll, 2000; Botvinick, Cohen, & Carter, 2004 ). For example, Davis et al. (2000) performed single cell recordings from individual dACC neurons as participants carried out four different cognitive tasks: mental arithmetic, semantic fluency, letter fluency, and a Stroop task. They found that while individual dACC these tasks, these neurons did not respond to painful stimuli, which might also be expected to draw attention. This pattern of results supports the idea that the dACC is involved in allocation of top down, effortful attention rather than simply automatic, reactive attention. Botvinick ,

PAGE 21

21 Nyst rom, Fissell , Carter , & Cohen (1999) had participants complete a flanker task during fMRI, where flankers where either congruent or incongruent with the direction of the target arrow. These researchers found greater dACC activation on incongruent trials th an on congruent trials, suggesting that dACC activation is most prominent under conditions of conflict. It is worth noting that all of the above studies were completed with participants who were not distinguished for their monolingual or bilingual status. Only one recent study (Abutalebi et al., 2012) compared monolinguals and bilinguals (German Italian) on a flanker task during fMRI, and not only did bilinguals show a reduction in the size of conflict effects over the course of the study while monolinguals did not do so, they also showed less activation of the dACC than monolinguals on incongruent trials. In addition, bilinguals had thicker gray matter in the dACC than monolinguals , which was negatively correlated with conflict effect sizes on a flanker tas k (more gray matter was associated with smaller conflict effects) , but only for bilinguals. These results indicate that bilingual ism increases cortical dACC density and shows that bilingual s engage the dACC more efficiently than monolinguals, resulting in better behavioral performance while using fewer neural resources. For tasks involving language, numerous studies have shown dACC activation in bilinguals, such as tasks involving translation (Price, Green, & von Studnitz, 1999), language switching (Abutale bi et al., 2007), and language conflict (Abutalebi & Green, 2007; Rodriguez Fornells et al., 2005; van Heuven, Schriefers, Dijkstra, & Hagoort, 2008). With respect to translation, Price et al. (1999) used PET scanning to show increased activation in German English bilinguals' dACC when they translated visually presented words. In terms of language switching, Abutalebi et al. (2007) had French Italian bilinguals listen to a narrative that was read predominantly in one language but contained unpredictable swi tches into the other language. Results from fMRI scans

PAGE 22

22 showed that these language switches resulted in increased dACC activation. Lastly, language different responses . Rodriguez Fornells et al. (2005) used fMRI and ERP to measure brain activation in monolingual and bilingual German Spanish speakers during a phoneme monitoring task. Participants were shown pictures, and their task was to press a button or to withhold re German or Spanish. Half of the pictures depicted objects that began with a vowel in German and a consonant in Spanish, which created language conflict for bilingua ls by requiring opposite responses in German and Spanish. Results showed that bilinguals responded more slowly and activated the dACC more than monolinguals for these pictures. Engagement of the dACC in the bilingual group suggests that bilinguals engage t his brain region when they are resolving language conflict, whereas monolinguals do not need to engage the dACC for this task. Similar results suggesting that language conflict activates the dACC were found by van Heuven et al. (2008), who had Dutch Engli sh bilinguals perform lexical decisions on letter strings to decide if they were English words. Some of the trials contained letters strings that were interlingual homographs (i.e., words in both Dutch and English), which create a language conflict. While interlingual homographs creating conflict comes from the finding that participants w ere slower to make lexical decisions for interlingual homographs than for non homograph words. Furthermore, results revealed significant dACC activation only for trials containing interlingual homographs relative to other trials that did not have language conflict (e.g., letter strings that were only words in English).

PAGE 23

23 Finally, the literature on emotion and attention indicates that the dACC serves as an interface between cognitive and emotional processes (Allman, Hakeem, Erwin, Nimchinsky, & Hof, 2001; Egn er, Etkin, Gale, & Hirsch, 2008), such that the dACC is engaged to prioritize allocation of attentional resources when emotional and cognitive stimuli both capture attention. Egner et al. (2008) had participants view happy or fearful male or female faces w ith congruent or incongruent words written across the faces, during fMRI. The words were either emotional (happy or fear) or non emotional (male or female), creating congruent and incongruent conditions for both the emotional and non emotional picture dime nsions. Participants completed two blocks of trials, wherein they identified the gender of the faces (with congruent and incongruent gender labels) in one block and the emotion of the faces (with congruent and incongruent emotion labels) in the other block . Results showed that a rostral region of the ACC was activated in the presence of emotional conflict but not in the presence of non emotional (i.e., cognitive) conflict. However, a dorsal area of the ACC was activated for both emotional and cognitive conf lict, suggesting that dorsal ACC (dACC) is involved in mediating attention in the presence of both emotional and cognitive conflict. Tasks with Cognitive and/or Emotional Distraction Given the role of the dACC in control of attention to cognitive and emot ional stimuli and in bilingual language processing, it is possible that by virtue of regular engagement of the dACC for control of language, bilinguals may be better able than monolinguals to suppress distraction from irrelevant emotional stimuli as well a s cognitive stimuli. To investigate this possibility, two tasks were selected for the present research, which allow the manipulation of different types of distractors (cognitive only, emotional only, or both emotional and cognitive) across three experiment s: (1) a digit parity task, and (2) a flanker task.

PAGE 24

24 Digit parity tasks require that participants determine whether two num bers do or do not match on a particular dimension, usually odd or even status. Such tasks were first used to study the representation of number concepts (e.g., Berch, Foley, Hill, & Ryan, 1999; Deheane, Bossini, & Giraux, 1993 ). More recently, digit parity tasks have been used to explore the effect of emotional distraction on cognitive processing (Aquino & Arnell, 2007; Fernandes et al., 2011; Thomas & Hasher , 2006). This research has shown that when participants complete digit parity tasks, the presence of various types of emotional stimuli impairs parity judgments. For example, the presence of sexually explicit words (Aquino & Arnell, 2 007) and emotionally arousing negative words (Thomas, 2006) slowed digit parity judgments relative to neutral words, and the presence of highly arousing negative pictures resulted in slower and less accurate judgments than highly arousing positive pictures (Fernandes et al., 2011). These results suggest that emotional stimuli presented during the digit parity task take attentional resources from that task, resulting in slower digit parity judgments. Since digit parity judgments are impaired in the presence of both emotional words and pictures, such a task appears well suited for investigating the effect of emotional distraction on cognitive processing. However, as with most research on attention, participants in these studies were not screened for the number of languages they spoke, nor were they identified as monolinguals. Carrying out similar experiments with separate monolingual and bilingual participant groups is necessary to determine whether or not monolingualism versus bilingualism affects attentional allocation/control in such paradigms. The digit parity task is particularly appropriate for use with bilinguals since research by Fias, Reynvoet, and Brysbaert ( 20 01 ) indicates that visually presented Ara bic numerals activate their semantic properties dir ectly without lexical/verbal mediation, i.e., without activating the word for the number they represent. As discussed earlier, using a nonverbal cognitive task is

PAGE 25

25 important when comparing monolinguals and bilinguals because monolinguals are known to perfor m some verbal tasks better than bilinguals (e.g. , Bialystok et al., 2008; Gollan et al. , 200 8 ; et al., 2005; Ivanova & Costa, 2008). B executive function benefits, making monolingual and bilingual groups ap pear to have equivalent executive function capacities even if they do not. Fias et al. (2001) suggest that because digit parity judgments are completed without accessing the lexical words for the numbers represented by the digits, the effect of distraction in this task can be measured independently from lexical access. Under this assumption, u sing emotional pictures or other digits as distractors ensure d that the stimuli for the entire task wa s nonverbal, and since all responses were made with a button pres s, all responding was nonverbal, thus assuring that the task wa s nonverbal in every way that c ould be experimentally controlled. Three separate digit parity tasks were used. In the cognitive distraction version, two digits were presented with a centrally presented, intervening digit to serve as a distractor. In the emotional distraction version, an intervening emotional picture served as a distractor. In the cognitive and emotional distractor version, both a digit and an emotional picture served as distrac tors. While digit parity judgments have been shown to be slowed in the presence of emotional distractor pictures, this task has never been used to test the effect of cognitive distractors on parity judgments. Hence, it was desirable to also use a task that has been shown to be affected by the presence of cognitive distractors, for comparison. For this purpose, a flanker task implemented like the ANT (Fan et al., 2002) was a lso administered. The ANT was originally designed to test attentional control in the presence of cognitive distraction, which is achieved by manipulating the congruency of the distractors; distraction occurs from incongruent trials, where flanker arrows point in a direction opposite from the target

PAGE 26

26 arrow. Modifications of this task resulte d in three different versions. In the cognitive distraction version, a flanker task was used to evaluate the effect of cognitive distraction on completion of a cognitive task. In the emotional distraction v ersion, the task measure d attentional control in t he presence of emotional distraction, without cognitive distraction, by presenting an emotional picture in the center of the screen along with a single target arrow, above or below the picture, but with no flankers. In the cognitive and emotional distracto r version, the task had both cognitive and emotional distraction by presenting both target and flanking arrows along with a centrally presented emotional picture. As mentioned previously, flanker tasks have been used to compare monolinguals and bilinguals, and results from studies using this paradigm have revealed faster RT s (Costa et al., 2008; Costa et al., 2009) and reduced conflict effects for bilinguals, resulting from bilinguals being less slowed on incongruent trials than monolinguals (Costa et al., 2008; Pelham & Abrams, 2012). However, no version of the flanker task has been used with emotional distractors, reinforcing the necessity for having both the digit pa rity task , which has shown an effect of emotional distraction on RT s, and a flanker task, which reliably reveals the effect of cognitive distraction on RT s.

PAGE 27

27 CHAPTER 2 HYPOTHESES The design of the three experiments included subsets of the following factors depending on the analysis: 1. A between subjects factor, Group, representing the bilingual status of the participants. The two levels of Group were bilinguals and monolinguals. 2. A within subjects factor, Flanker Type (in the flanker task) and Digit Distractor Type (in the digit parity task), which reflects the trial type in terms of the target di stractor relationship. Both tasks had three levels of neutral, congruent, and incongruent, while the digit parity task had an additional level, ambiguous, which is described in the method. 3. A within subjects factor, Distractor Picture Valence, which reflect s the emotional nature of the picture. There were three levels: positive, negative, and neutral. 4. A within subjects factor, Previous T rial D istractor V alence , representing the emotional nature of the picture on the preceding trial. There were three levels : positive, negative, neutral . Hypothesis 1 : Cognitive distraction in a cognitive task I hypothesize d would allow them to control interference from cognitive distractors more efficiently than monolinguals. If thi s hypothesis is supported, then a main effect of group would emerge for RT s due to bilinguals responding faster overall than monolinguals, as well as a Group X Flanker Type/Digit Distractor Type interaction, resulting from bilinguals having smaller conflic t effects than monolinguals because of smaller interference effects from incongruent trials compared to congruent trials. These results would confirm the bilingual advantage in cognitive control shown previously on various tasks such as the flanker task an d extend that finding to a cognitive task not previously tested with bilinguals, i.e., digit parity. This is important because it establishes digit parity as a more globally useful task beyond just testing emotional distraction. As noted previously, signif icant effects were only expected for RT s, but accuracy was also assessed to determine the occurrence of any speed accuracy tradeoffs.

PAGE 28

28 An alternate outcome is that bilinguals w ould show a cognitive advantage only on the flanker task and not on the digit par ity task. If digits do result in lexical access of the word for the digits, contrary to Fias et al. (2001) , this might slow responding in a way that obscures any executive function advantage that bilinguals might have. If this is the case, it would also be possible that (as with the Altarriba & Bainsight RT s for digit parity judgments would be slower than monolinguals across all task conditions. These slower RTs would only be expected to emerge on the digit pari ty task and not the flanker task because the flanker task does not involve lexical access. Hypothesis 2 : Emotional distraction in a cognitive task I hypothesize d that bilinguals would manifest better attentional control than monolinguals more broadly, w hich would enable them to better suppress interference from emotional distractors than monolinguals. If this hypothesis is supported, a Group x Distractor Picture Valence interaction would reveal that bilinguals had smaller interference effects produced by emotional trials relative to neutral trials, compared to monolinguals. These results would indicate demonstrated when distractors were cognitive in nature, extends to c ontrolling attention in the presence of emotional distractors. Alternatively, bilinguals may not show better attentional control on tasks that involve emotional distractors, and if so, both groups would be equivalently slowed and error prone following emot ional distractors. These findings would indicate that emotional distraction supersedes cognitive control of attention and any bilingual benefits associated with that control, capturing and holding attention equally in both groups. This might occur because the relevance of emotional stimuli to survival has resulted in evolution of attentional systems that are easily and strongly engaged by such stimuli, and the ability to override attention to emotion may have

PAGE 29

29 been maladaptive and therefore not selected for. Hence, no amount of practice in using brain regions associated with controlling attention (such as those by bilinguals) would be sufficient to override the effect of millions of years of evolution on attention to emotional stimuli. Hypothesis 3 : Emotional and cognitive distraction in a cognitive task I hypothesize d would enable them to suppress interference from combined cognitive and emotional distractors better than monolinguals. Furthermore, I hypothesize d th monolinguals would emerge in the condition where both cognitive and emotional distractors are present. This hypothesis would be supported if the analysis revealed a Group X Flanker Type/Digi t Distractor Type X Distractor Picture Valence interaction, such that bilinguals were less slowed than monolinguals in conditions where distractors we re both cognitive and emotional relative to neutral trials. As in Experiments 1 and 2, bilinguals would b e less slowed than monolinguals on trials that involve only cognitive distraction (e.g., for the flanker : incongruent flanker trials with neutral pictures relative to neutral flanker trials with neutral pictures) and trials that only involve emotional distr action (e.g., neutral flanker trials with emotiona l pictures relative to neutral flanker trials with neutral pictures). Furthermore, the greatest magnitude of would be on trials with both types of distractors were pr esent, relative to one type of distractor present. This result would likely emerge because monolinguals were most slowed on trials involving multiple distractors compared to only one type of distractor. In contrast, bilinguals' enhanced ability to control attention would enable them to handle the distraction independent of the number of distractors. These results would affirm results from the first two experiments and would extend the finding to indicate that the greater the level of distraction, the greate r the size of the bilingual advantage in controlling attention.

PAGE 30

30 Alternatively, bilinguals may show an attentional advantage only on trials with cognitive distractors. This alternative would be supported by a significant Group x Flanker Type/Digit Distracto r Type x Distractor Picture Valence interaction such that bilinguals are less slowed than monolinguals on cognitive conflict trials only when those trials are accompanied by neutral pictures. This finding would indicate that the bilingual advantage in cont rolling attention in the presence of cognitive distraction is eliminated by the presence of emotional distractors, thus underscoring the attentional priority that emotional stimuli receive. Hypothesis 4: Carryover effects from emotional trials I hypothesi ze d that participants would experience carryover effects of experimental trials with emotional distractor pictures , slowing subsequent filler trials. I further hypothesize d that would enable them to suppress interfer ence from carryover effects of emotional pictures on filler trials better than monolinguals . This hypothesis w ould be supported if the analysis reveals a Group X Previous Trial Distractor Valence interaction in Experiments 2 and 3 , such that bilinguals are less slowed than monolinguals on filler trials preceded by emotional distractor pictures.

PAGE 31

31 CHAPTER 3 EXPERIMENT 1 C OGNITIV E DISTRACTION Method Participants Sixty participants, 3 0 bilinguals fluent in English and Spanish ( 66.7 % female; 9 3 .4% Hispanic, 6 .6% Caucasian) and 3 0 monolinguals ( 73.3 % female; 73.4 % Caucasian , 23.3 % African American, 3.3% Hispanic ) fluent only in English , were recruited from graduate and undergraduate courses at the University of Florida as well as via newspaper advertisements i n the university newspaper. Participants enrolled in courses with a research requirement received partial course credit for participation, while participants who were not in a class for which they had a research requirement were paid $10 per hour for their participation. All participants were pre screened for their language status through use of the Language Experience and Proficiency Questionnaire (LEAP Q; Marian, Blumenfeld & Kaushanskaya, 2007). To be eligible for participation, monolinguals were require d to have a self rated proficiency of 3 (fair) or lower in speaking and understanding a language other than English, rated on a 0 (none) to 10 (better than the average native speaker) scale. Bilinguals were required to have a self rated proficiency of 7 (g ood) or greater in speaking and understanding both English and Spanish on the same 0 to 10 scale. Additionally, all bilingual participants were also screened for proficiency in Spanish with an oral interview conducted by a Spanish speaking research assista nt whose first language was Spanish (see Appendix for the script used as a template for the interview) . Participants who were unable to carry on a fluent conversation with a native Spanish speaker were not allowed to complete the experiment. Because resear ch has demonstrated that proficient bilinguals, irrespective of age of L2 acquistion, manifest advantages in attentional control relative to monolinguals (e.g. Pelham & Abrams, 2013; Tao, Marzecová, Taft, Asanowicz, & Wodniecka,

PAGE 32

32 2011 ), neither age of onset of second language acquisition nor age at which native like proficiency was attained were used as criteria to define the bilinguals for the present study. Table 3 1 shows the mean proficiency ratings and exposure for both groups to their dominant language (English) and their non dominant language (Spanish for bilinguals; any language other than English for monolinguals). T he groups did not differ in their self rated dominant language (English) proficiency in speaking or understanding , t s< 1 , but they did d iffer in their non dominant language proficiency in speaking, t ( 58 ) = 26.79 , p < .001, and understanding, t ( 58 ) = 29.70 , p < .001 , with bilinguals reporting greater proficiency than monolinguals (who didn't have a non dominant language per se) . In addition , the groups differed from one another in their exposure to their dominant , t (5 8) = 7.92 , p < .001, and non dominant , t (57 ) = 8.25 , p < .001, languages , as expected, with bilinguals reporting less exposure to their dominant language and more exposure to th eir non dominant language than monolinguals . Monolinguals' and bilinguals' demographic characteristics as well as performance on several cognitive tests are presented in Table 3 2 . The demographics questionnaire was given orally before the experimental ta sks in Spanish for the bilinguals and in English for the monolinguals. The forward and backward digit span and vocabulary tasks were given in English after the experimental tasks were completed. The groups differed significantly in terms of age, t ( 5 8) = 2. 29 , p = .0 3 , with the bilingual group being older than the monolingual group. However , the groups did not differ in years of formal education, t < 1 , self reported college GPA, t < 1 , self reported health rating, t (5 8) = 1. 25 , p = .2 1 , gaming hours per mon th, t < 1, backward digit span, t < 1, or vocabulary score, t < 1. The groups did differ on their forward digit span, t ( 58 ) = 2. 19 , p = .03, with mono linguals having a longer forward digit span ( M = 7. 90 , SD = 1. 4 ) than bi linguals ( M = 7.07 , SD = 1. 1 ) .

PAGE 33

33 Ma terials For the flanker task (see Figure 3 1, top panels) , the target was always a central arrow pointing in either the left or right direction. Distractors were either flanking arrows (incongruent or congruent) or flanking horizontal lines that were the same size as the arrows (neutral). For incongruent trials, the flankers provided cognitive distraction in the form of a task irrelevant directional response cue. The direction of the flankers indicated a different response from the target's direction, crea ting cognitive distraction because the direction of the flankers must be ignored in order to provide a correct response. In contrast, on congruent trials, the target and flankers matched direction, making it unnecessary to ignore the flankers to respond co rrectly and thus creating little cognitive distraction. On neutral trials, the flankers provided no directional cue and again created minimal cognitive distraction in responding to the target. Participants completed three blocks of 126 trials with 42 of ea ch distractor type in each block. Distractor type was semi random with no more than five of the same type of trial occurring consecutively. For the digit parity task, the digits 1 through 9 (excluding 5) were used as targets. The number 5 was excluded bec ause its magnitude makes it neutral in terms of the Spatial Numerical Association Response Codes (SNARC) effect. The SNARC effect is a phenomenon where people who read from left to right have a tendency to respond more quickly in a digit parity task with a correct left handed response when the number on the left is small and faster with a correct right handed response when the number on the right is large (Berch et al., 1999; Deheane et al., 1993). To counterbalance any possible SNARC effects across all tri al types, we only included numbers that could be categorized as big or small, whereas 5 is neutral. Using the above digits results in 64 unique digit pairs, and for reasons explained in Experiment 2, the digit pair 1,1 was excluded, resulting in 63 digit p airs that were used . For each trial, a pair of target digits appear ed ,

PAGE 34

34 one to the left and the other to the right of a central fixation, while a single distractor digit appear ed in the center of the screen. Distractors were either a digit from the set of target digits or a neutral X (see upper panels of Figure 3 1 bottom panels ). When the distractor was a digit, it could be congruent , incongruent , or ambiguous with the targets' parity, which determines the degre e of cognitive distraction. When the target digits match ed in parity and the distractor was the opposite parity, it provide d cognitive distraction because the response elicited by comparison of the distractor to either of the target digits was incongruent with the response required by the targets. The distractor wa s categorized as ambiguous if the target digits did not match in parity because then the distractor necessarily match ed one of the targets but not the other, creating a situation where the response elicited by comparison of the distractor w ith one of the targets elicit ed a response that conflict ed with the correct response , while comparison of the distractor with the other of the targets elicit ed a response that wa s in accord with the correct response . T hus , ambiguous distractors also provi de d a cognitive distraction but different from incongruent trials . In contrast, for trials where the target digits match ed and distractor match ed in parity , the response elicited by the comparison of the distractor to the targets was congruent with the res ponse required by the target digits, making it unnecessary to ignore the distractor and imposing minimal cognitive distraction. In the case of a neutral distractor X, the level of cognitive distraction was also low. Neutral distractors provide d a baseline measure without distract ion to assess whether congruent, incongruent , and ambiguous trials facilitate d or impede d responding. Participants complete d three blocks, each with 12 6 trials containing 32 incongruent and incongruent trials and 31 neutral and ambi guous trials in each block . Order of distractor type was semi random with the stipulation that no more than five trials of the same type could occur in a row.

PAGE 35

35 Procedure Participants first completed the LEAP Q and demographics questionnaires. They then co mplete d both a flanker task and a digit parity task , with order of the two tasks counterbalanced across participants. Prior to the beginning of each task, participants complete d 12 practice trials. Each trial for the task s last ed 3000 ms and began with a c appearance of the targets and distractor, a variable delay (D1) of between 400 and 1300 ms elapsed . For the flanker task, the ed on the screen for the entire trial, and following D1 , the ta rget and flankers appear ed either above or below the fixation equally side of the keyboard for left ide of the keyboard for right pointing targets. Participants had 1500 ms to respond, and once they respond ed or when 1500 ms had elapsed, the target and distractor disappear ed . There w as a second variable delay (D2), which lasted for 3000 ms minus D1 minus response time . Participants w er e allowed to take a break after each block (see Figure 3 2 for a graphical illustration of the time course of trials in the two tasks). Variable delays were included as part of the time course for each trial fo r the following reason. W ithout variable delays, each new trial would begin at a predictable delay following a response, and that predictability might serve as a n implicit cue to the appearance of the upcoming target. Since cuing prior to the appearance of a target speeds response times (Fan et al, 2004) and affects monolinguals and bilinguals differently (Costa et al., 2008), we wanted to ensure that there was no possibility that the timing of the trials would inadvertently serve as a cue to participants. Figure 3 2 shows a single incongruent digit parity trial. The time course for the digit parity task was identical to that used in the flanker task shown in Figure 3 started each trial and remained on the screen

PAGE 36

36 during the f irst delay, D1. Then , the target and distractor digits appeared horizontally centered on the screen , with the distractor digit appearing in the center of the screen and replacing the central fixation . Participants were told to ignore the distractor in the middle and respond as quickly as task, and which button co rresponded to matching and non matching parity was counterbalanced across participants. Results Flanker Task Response t ime s Before ANOVAs were performed on the response time data for the f lanker task, incorrect responses and practice trials were excluded from analysis. After excluding these trials, the data were trimmed of outliers that were more than 2 standard deviations above or below each participant's mean for congruent, incongruent, and neutral trials. Trimming resulted in equivalent data loss across groups (5.1% for monolinguals and 5.1% for bilinguals) and across flanker types (4.7% for congruent, 6.3% for incongruent, and 5.0% for neutral). Because language balance is important for the emergence of bilingual advantages ( Tao et al., 2011 ; Zied, et al., 2004) and because regular video game play is known to improve (i.e., reduce) people's Response Time s (Dye, Green, & Bavelier, 2009; Strobach, Frensch, & Schubert, 2012), language balance and video game play, or gaming, were entered as covariates in al l analyses. Language balance was determined using the percentage of exposure participants reported for languages they spoke or had studied. A score ranging from 0 (completely unbalanced) to 1 (completely balanced) was computed by dividing the percentage of exposure for the language they were exposed to less by the percentage of exposure for the language they

PAGE 37

37 were exposed to more. For example, if a participant reported being exposed to English 50% of the time and Spanish 50% of the time, their language balan ce score would be 1, whereas if a participant reported being exposed to English 100% of the time and any other language 0% of the time, their language balance score would be 0. This allowed for calculation of language balance scores in a uniform metric for both monolingual and bilingual participants. Participants also reported their average daily, weekly, and monthly hours of video game play, and participants' total monthly hours of video game play was entered as a measure of gaming. A 2 (Group: monolingua l, bilingual) x 3 (Flanker Type: neutral, congruent, incongruent) mixed factorial ANOVA was conducted on mean Response Time s, shown in Table 3 3 . Results from the analysis revealed a main effect of flanker type, F (2, 112) = 222.57, MSE = 242.09, p < .001, 2 = .80, such that Response Time s to targets with neutral flankers ( M = 610.41 SD = 44.32) were faster than targets with congruent flankers ( M = 616.29, SD = 44.03), which were faster than targets with incongruent flankers ( M = 714.98, SD = 44.32), all p ' s < .001. There was also a main effect of group, F (1, 56) = 7.34, MSE = 5171.26, p 2 = .12, such that bilinguals responded faster ( M = 627.67, SD = 52.71) than monolinguals ( M = 672.20, SD = 38.57). However, there was no interaction, F < 1. These results are shown in Figure 3 3. Accuracy Accuracy was computed on all experimental trials, and incorrect responses and trials where participants failed to respond in the allotted time were considered inaccurate. A 2 (Group: monolingual, bilingual) x 3 ( Flanker Type: neutral, congruent, incongruent) mixed factorial ANOVA was conducted on mean proportion of accurate responses, shown in Table 3 4 . Results revealed a main effect of flanker type, F (2, 112) = 11.47, MSE = .001, p 2 = .17, such that participants responded more accurately to congruent trials ( M = 99.3%, SD = 0.2%), than neutral trials ( M = 98.7%, SD = 0.2%), p = .04, which in turn were more accurate than

PAGE 38

38 incongruent trials ( M = 95.6%, SD = 0.5%), p < .001. There was no main effect of group or interaction, F 's< 1. Digit Parity Task Response t ime s Prior to analyzing data, incorrect responses and practice trials were excluded. After excluding these trials, the data were trimmed of outliers that were more than 2 standard deviations above or below each participant's mean for trials with congruent, incongruent, and neutral, and ambiguous trials. Trimming resulted in minimal data loss for both groups (3.9% for monolinguals and 3.8% for bilinguals) and approximatel y equivalent data loss for congruent (4.4%), incongruent (3.2%), neutral (4.0%), and ambiguous (3.5%) digit distractors. A 2 (Group: monolingual, bilingual) x 4 (Digit Distractor Type: congruent, incongruent, neutral, ambiguous) mixed factorial ANOVA was conducted on mean Response Time s with language balance and gaming as covariates. Response times for each group in each digit distractor condition are shown in Table 3 5 . Results revealed a significant main effect of digit distractor type, F (3, 168) = 40.06 , MSE = 820.48, p 2 = .42, such that Response Time s to targets with congruent distractors were faster ( M = 905.99, SD = 79.29) than targets with neutral distractors ( M = 935.90, SD = 80.22), p < .001, which were faster than targets with incongruen t distractors ( M = 947.95, SD = 80.82), p = .03, which were faster than trials with ambiguous distractors ( M = 1015.82, SD = 96.67), p < .001. There was no main effect of group, F (1, 56) = 1.92, MSE = 25759.20, p 2 = .03, and no interaction, F (1, 5 6) = 1.17, MSE = 820.48, p = 2 = .02. Accuracy Accuracy was computed on experimental trials, and as with the flanker task, incorrect responses and trials where participants failed to respond in the allotted time were considered

PAGE 39

39 inaccurate. A 2 (Grou p: monolingual, bilingual) x 4 (Digit Distractor Type: congruent, incongruent, neutral, ambiguous) mixed factorial ANOVA with language balance and gaming as covariates was conducted on mean proportion of accurate responses, shown in Table 3 6 . Results did not reveal a main effect of distractor type, F (3, 168) = 1.05, MSE = .003, p = .37 2 = .02, but the main effect of group was marginally significant, F (1, 56) = 2.98, MSE = .02, p 2 = .05, with bilinguals responding more accurately ( M = 92.1%, SD = 6.3%) than monolinguals ( M = 87.9%, SD = 8.7%). There was, however, no interacti on, F < 1. Discussion The purpose of Experiment 1 was to replicate findings of a bilingual advantage in controlling attention on a flanker task and to potentially extend that finding to a new task involving attention, digit parity. The hypothesis was that bilinguals would manifest an advantage over monolinguals on both the flanker task and the digit parity task by responding faster overall and by having smaller conflict effects. On the flanker task, bilinguals responded faster than monolinguals overall, con sistent with the most commonly found bilingual advantage observed in previous studies ( Costa et al., 2009; Hilchy & Klein, 2011 ). However, they did not have smaller conflict effects than monolinguals, which may not be surprising given that there are a numb er of other studies investigating bilingual advantages that shown bilinguals to have an overall speed advantage but no bilingual advantage in conflict effects ( e.g., Bialystok & Viswanathan, 2009; Emmorey et al., 2008; Martin Rhee & Bialystok, 2008 ). Thus , these results partially replicate findings of a bilingual advantage in the flanker task with respect to speed of responding. No group differences occurred in terms of accuracy , consistent with past research ( Bialystok & Viswanathan, 2009; Costa et al., 2 008; Costa et al., 2009; Pelham & Abrams, 2013 ), although both groups seem to have been using a speed accuracy tradeoff for congruent trials, which were slower than neutral trials but also more accurate. This tradeoff did not help performance on

PAGE 40

40 incongruen t trials, which were slower and less accurate than neutral trials, demonstrating their greater susceptibility to interference. Results from the digit parity task revealed no differences between the groups in terms of RT s, so the results do not extend find ings of a bilingual speed advantage to a new task. Both groups were similarly sensitive to the digit distractor type, with the more interfering distractors (incongruent and ambiguous ) slowing RT s and the congruent distractors speeding RT s. While the digit parity task has been claimed to be a non verbal task ( Fias et al., 2001 ) , the lack of a bilingual advantage found here could instead suggest that the task is requiring lexical access of the words representing the digits. Since bilinguals often experience l exical access deficits relative to monolinguals ( Bialystok et al., 2008 ; Gollan et al., 2008; Gollan, Montoya, & Werner, 2002, Ivanova & Costa, 2008 ), lexical access of the digits' words could slow bilinguals' responding and thereby obscure any executive f unction advantage in speed that bilinguals might have (see also Altarriba & Bainsight Brown, 2010) . Specifically, while both groups experience d interference from the digit distractors and we re slowed by this, bilinguals' advantage in controlling distractio n should have resulted in them being less slowed than monolinguals. However, RT s may not have revealed any bilingual advantages if the task also indirectly involved lexical access, a process on which bilinguals experience slowing relative to monolinguals b ecause of interference from activation of words in multiple languages. Consequently, bilinguals' speed advantage with the digit distractors may have been offset by the slowing from lexical access , resulting in the groups having similar RTs on the digit par ity task. This occurred exclusively for t he digit parity task because the flanker task does not offer a possibility for lexical access to occur . D espite equivalent RTs, bilinguals were marginally more accurate than monolinguals on the digit parity task , su ggesting a slight advantage on this

PAGE 41

41 measure. Even if the task involved lexical access as suggested above, bilinguals generally experience a disadvantage in lexical access only in terms of speed of responding, not accuracy (Gollan et al., 2008; Gollan et al ., 2002; Ivanova & Costa, 2008; Portocarrero, Burright, & Donovick, 2007), making it possible for a bilingual advantage to emerge on digit parity when accuracy is the dependent measure. However, the marginality of this effect, along with the lack of a bili ngual advantage in speed of responding, does not provide strong evidence for a bilingual advantage. Speed, not accuracy , is general ly where bilingual advantages or disadvantages emerge in most studies , most likely because of ceiling effects in accuracy . In sum, t hese results establish that there are some bilingual advantages in managing cognitive distraction, consistent with previous research, and these will serve as a baseline of comparison for Experiment 3, when emotional distraction is added along with c ognitive distraction . Results from the next experiment, Experiment 2, will establish whether bilingual advantages similarly emerge solely in the presence of emotional distraction via positive and negative pictures.

PAGE 42

42 Table 3 1. Monolinguals' and bilinguals ' self reported language proficiency and exposure Group Monolinguals Bilinguals Variable Mean SD Mean SD * % Exposure to Dominant Language 97.5 5.6 69.2 18.7 * % Exposure to Non Dominant Language 2.1 3.2 33.7 20.4 Proficiency Speaking D ominant Language (0 10 scale) 9.2 1.9 9.5 .81 * Proficiency Speaking Non Dominant Language ( 0 10 scale) 1.7 .91 8.3 .98 Proficiency Understanding Dominant Language (0 10 scale) 9.5 .68 9.5 .73 * Proficiency Understanding Non Dominant Languag e (0 10 scale) 1.8 .87 9.0 1.0 * p < .001 Table 3 2. Monolinguals and bilinguals' demographic characteristics and cognitive tests Group Monolinguals Bilinguals Variable Mean SD Mean SD * Age (years) 19.1 1.4 20.0 1.8 Education (years) 13.4 2 .2 13.5 3.9 College GPA (out of 4.0) 3.4 .58 3.5 .42 Health (1 10 scale) 8.2 1.5 8.6 1.1 Gaming Hours per month 22.5 23.6 28.3 29.9 Vocabulary (out of 25) 14.0 3.9 13.7 3.7 * Forward Digit Span 7.9 1.4 7.1 1.1 Backward Digit Span 5.2 1.4 5.2 1. 4 * p < .05

PAGE 43

43 Table 3 3. Experiment 1 Cognitive flanker task response times (in ms) Group Monolinguals Bilinguals Flanker Type Mean SD Mean SD Congruent 632.5 36.7 616.3 49.6 Incongruent 720.4 42.5 709.5 57.5 Neutral 616.1 36.5 604.7 51.0 Table 3 4. Experiment 1 Cognitive flanker task accuracy (in %) Group Monolinguals Bilinguals Flanker Type Mean SD Mean SD Congruent 98.9 2.4 99.3 0.9 Incongruent 95.9 4.8 95.3 4.4 Neutral 99.1 3.2 99.1 1.0 Table 3 5. Experi ment 1 Cognitive digit parity task response times (in ms) Group Monolinguals Bilinguals Digit Distractor Type Mean SD Mean SD Congruent 900.3 86.6 911.7 72.3 Incongruent 950.6 85.0 945.3 77.8 Neutral 937.1 86.8 934.7 74.5 Ambiguous 10 16.2 106.7 1015.3 87.3 Table 3 6. Experiment 1 Cognitive digit parity task accuracy (in %) Group Monolinguals Bilinguals Digit Distractor Type Mean SD Mean SD Congruent 90.7 8.3 91.8 4.7 Incongruent 87.0 14.1 83.8 10.7 Neutral 91.7 5.6 90.7 4.7 Ambiguous 92.4 6.7 92.1 5.3

PAGE 44

44 Experiment 1 Targets and Distractors Congruent Incongruent Neutral Flanker Task Digit Parity Figure 3 1. Target and distract ors for Experiment 1, cognitive Figure 3 2. Time course for digit parity task 3 7 1 3 4 1 3 X 1 + + +

PAGE 45

45 Figure 3 3 . Bilinguals' and monolinguals' flanker task RTs in Experiment 1

PAGE 46

46 CHAPTER 4 EXPERIMENT 2 E MOTIONAL D ISTRACTION Method Participants Sixty participants, 3 0 bilinguals fluent in English and Spanish ( 73.3 % female; 100% Hispanic ) and 3 0 monolinguals ( 73.3 % female; 80.0 % Caucasian, 10.0 % African American, 3.3 % Hispanic , 6.7% Asian ) fluent only in English , were recruited and compensated as described in Experiment 1. Additionally , particip a nts were pre screene d for their language status and categorized as monolingual or bilingual based upon the same criteria as was used in Experiment 1. B oth the monolingual and bilingual groups reported being highly proficient in their dominant language (English) , with equivale nt proficiency in understanding , t ( 58 ) = 1.59 , p = . 12, but unlike Experiment 1, monolinguals reported higher self rated proficiency in speaking English than bilinguals , t ( 58 ) = 3.39 , p = . 02 . As in Experiment 1, bilinguals reported greater proficiency in speaking, t ( 57 ) = 31.69 , p < .001, and in understanding, t ( 57 ) = 29.48 , p < .001 , their non dominant language relative to monolinguals . Also as in Experiment 1, bilinguals reported less exposure to their dominant language, t (5 8) = 5.03 , p < .001, and more exposure to their non dominant language, t (57 ) = 6.87 , p < .001 than monolinguals . Language demographics for the groups are presented in Table 4 1. Monolinguals' and bilinguals' demographic characteristics as well as performance on several cognitive tests are presented in Table 4 2. The groups differed significantly in terms of age, t ( 5 8) = 4.72 , p < . 0 01 , with the bilingual group being older than the monolingual group. The groups also differ ed in their years of formal education, t (58) = 2.07, p = . 04 , with bilinguals reporting more education . However, the groups did not differ in self reported college GPA, t <

PAGE 47

47 1 , health rating, t < 1 , gaming hours per month, t (12) = 1.21 , p = .2 5, vocabulary score, t < 1 , forward digit span, t ( 58 ) = 1.29 , p = . 20, or bac kward digit span, t < 1. Materials A modified flanker task and a digit parity task, both of which are described below, were used in this experiment such that neither task contain ed any cognitive distractors. Rather, d istractors for this experiment consist ed of 378 pictures selected from IAPS (Lang et al., 2008), 126 emotionally arousing positive pictures (M valence = 7.1, SD valence = 0.43; M arousal = 5.8, SD arousal = 0.57) , 126 emotionally arousing negative pictures (M valence = 2.2, SD valence = 0.39; M arous al = 6.0, SD arousal = 0.55) , and 126 neutral pictures with a mid range valence and low arousal (M valence = 4.9 , SD valence = 0.52; M arousal = 3.3, SD arousal = 0.59 ). An ANOVA on valence rating indicated that the pictures in each valence category differed, F (1, 351 ) = 4.25 , MSE = .179 , p < .00 1 2 = . 96, with negative pictures being rated lower than neutral pictures, p < .001, which were rated lower than positive pictures, p < .001. A n ANOVA on arousal rating revealed that all three picture types also differed on this dimension, F (1, 351 ) = 790 .80 , MSE = .294 , p < .00 1 2 = . 82, with negative pictures being more arousing than positive pictures, p = . 0 1, which were more arousing than neutral pictures, p < .00 1. The ranges of valence and arousal ratings were also examined, separately for males an d females, to ensure that three valence categories were similarly non overlapping for both genders. A n additional two sets of 93 neutral filler pictures (31 for each block) were selected from IAPS , one for the arrows task, and one for the digit parity task . The neutral filler picture sets were counterbalanced so that no picture was ever shown more than one time in an experimental session. Tables 4 3 and 4 4 provide the IAPS valence and arousal ratings for positive, negative, and neutral pictures in each pic ture set. The modified flanker task showed only a centrally presented target arrow , and the distractors consisted of 63 positive, 63 negative, and 63 neutral pictures . Participants completed

PAGE 48

48 three blocks of trials with an opportunity to take a break betwee n blocks. The blocks consisted of 126 trials, 63 experimental trials with positive, negative, and neutral distractors (21 of each type ), alternating with 63 neutral filler trials in each block that were used to measure carryover effects from emotionally a rousing pictures on subsequent non emotional trials. Half of the f iller trials presented the target arrow with either a scrambled non emotional picture that wa s the same size as the distractors, whereas the remaining filler trials contained a neutral pictu re. The choice of two types of filler trials was made because the scrambled pictures allowed for measurement of a baseline for responding in the absence of an actual picture. However, if only scrambled picture fillers had been used, participants would quic kly have realize d that every other trial contained a picture. Having half the fillers be neutral pictures and half the fillers be scrambled neutral pictures made it difficult for participants to recognize a pattern to the trials and prevented participants from anticipating a (possibly emotionally arousing) distractor picture on an upcoming trial. For the digit parity task, targets were the same as those in Experiment 1. E ach digit pair was presented once with a positive distractor, once with a negative dis tractor, and once with a neutral distractor, with each type of digit pair distractor type counterbalanced across three blocks of trials. Each block contain ed 126 trials, 63 digit pairs presented with positive, negative, and neutral distractors (21 of each type), and 63 with neutral or scrambled filler distractors. Procedure For the arrows task, the only change from Experiment 1 is that the flankers (cognitive distractors) were not presented, and instead an emotional or neutral distractor picture appeared in place of the central fixation. Participants were instructed to ignore the picture and respond as pointing arrows. In the digit parity task, the time cours

PAGE 49

49 responses were identical to Experiment 1, but an emotional or neutral distractor picture replace d the fixation instead of having a distractor digit . After both tasks were completed all participants rated the v alence of a subset of 48 pictures used in the two tasks (16 positive, 16 negative, and 16 neutral pictures, 8 each from picture set A and picture set B) . T he rating system was the same as that used in the development of the IAPS pictures (see Lang et al., 2008 for details) . The purpose of this assessment was to ensure that the valence of the distractor pictures was perceived by b oth monolinguals and bilinguals in the same categories as the IAPS ratings . Results Participants' Valence Rating s To evaluate whet her monolingual and bilingual participants perceived the valence of the pictures in the same way, a 2 (Group: monolingual, bilingual) x 3 (Valence: positive, negative, neutral) mixed factorial ANOVA compared the mean valence ratings given by participants. Results revealed a main effect of valence, F ( 2, 116 ) = 650. 38 , MSE = . 56 , p < . 001, 2 = .92 , such that negative pictures were rated lower than neutral pictures, p < .001, which were rated lower than positive pictures, p < .001, consistent with IAPS ratings for the pictures, but there was no main effect of group, F ( 1, 58 ) = 1.72 , MSE = . 33 , p = .20 2 = .03, However, there was a marginally significant interaction, F ( 2, 116 ) = 2.64 , MSE = .56 , p = .07 2 = .04, where bilinguals rated the positive pictures ( M = 7.2, SD = .84) higher than monolinguals ( M = 6.8, SD = .68), p = .02. However, the groups did not differ in their valence ratings for negative pictures, p = .34, or neutral pictures, p = .48.

PAGE 50

50 Emotional Arrows Task Response t ime s Incorrect responses and practice trials were excluded from before ANOVAs were performed on the RT data f or the emotional arrows task. Data were then trimmed of outliers that were more than 2 standard deviations above or below each participant's mean for trials with positive, negative, and neutral pictures. Trimming resulted in approximately equivalent data l oss across for both groups (5.3% for monolinguals and 4.6% for bilinguals) and for trials accompanied by positive (5.5%), negative (6.0%), and neutral (4.2%) picture distractors. A 2 (Group: monolingual, bilingual) x 3 (Distractor Picture Valence: positiv e, negative, neutral) mixed factorial ANOVA with language balance and gaming as covariates was conducted on mean RT . This analysis revealed a main effect of distractor picture valence, F (2, 112) = 26.25, MSE = 230.46, p < 2 = 0.31, such that participants responded more quickly to trials with neutral pictures ( M = 680.96, SD = 53.03) than trials with positive pictures ( M = 692.26, SD = 53.50), which were faster than trials with negative pictures ( M = 709.33, SD = 59 .01), all p ' s < .001. However, there was no main effect of group, F (1, 56) = 1.67, MSE = 8180.73, p 2 = 0.03, nor was there an interaction, F < 1. RT s for each group are shown in Table 4 5 . Accuracy Accuracy was computed on all experimental trial s, and incorrect responses and trials where participants failed to respond in the allotted time were considered inaccurate. A 2 (Group: monolingual, bilingual) x 3 (Distractor Picture Valence: positive, negative, neutral) mixed factorial ANOVA was conducte d on mean proportion of accurate responses with language balance and gaming as covariates. Results revealed no main effect of distractor picture valence, F (2, 112) = 1.11, MSE < .001, p 2 = .02, no main effect of group, F (1, 56) = 1.57, MSE =

PAGE 51

51 .002, p 2 = .03, and no significant interaction, F < 1. Table 4 6 shows mean accuracy for the groups in for each distractor type. Emotional Digit Parity Task Response t ime s After ex cluding incorrect responses and practice trials, the data were trimmed of outliers that were more than 2 standard deviation above or below each participant's mean for trials with positive, negative, and neutral distractor pictures. Trimming resulted in equ al data loss for both groups (3.9% for monolinguals and 4.2% for bilinguals) and for trials accompanied by positive (4.1%), negative (3.8%), and neutral (4.2%) picture distractors. A 2 (Group: monolingual, bilingual) x 3 (Distractor Picture Valence: positi ve, negative, neutral) mixed factorial ANOVA was conducted on mean RT s with language balance and gaming as covariates. Results revealed a main effect of distractor picture valence, F (2, 112) = 17.65, MSE = 953.08, p 2 = .24, such that RT s to targe ts with neutral distractors were responded to more quickly ( M = 937.53, SD = 74.94) than targets with positive distractors ( M = 963.44, SD = 78.59), p = .01, which were faster than responses to targets with negative distractors ( M = 980.89, SD = 87.31), p < .001. However, there was no main effect of group, F < 1, and no interaction between distractor picture valence and group, F (2, 112) = 1.95, MSE = 953.08, p 2 = .03. Table 4 7 shows the mean RT s for both groups for each type of distractor picture. Accuracy Proportion of accurate responses was computed on experimental trials such that incorrect responses and trials where participants failed to respond in the allotted time were considered inaccurate. A 2 (Group: monolingual, bilingual) x 3 (Distractor Picture Valence: positive, negative, neutral) mixed factorial ANOVA was conducted on mean proportion of accurate

PAGE 52

52 responses. Results revealed a main effec t of distractor picture valence, F (2, 110) = 4.00, MSE = .002, p 2 = .07, such that participants responded most accurately to trials with neutral picture distractors ( M = 90.4%, SD = 7.0%), which were more accurate than trials with negative picture distractors ( M = 88.6%, SD = 8.7%), p = .03, or positive pictur e distractors ( M = 87.8%, SD = 9.3%), p = .002. Accuracy in responding to trials with negative and positive picture distractors did not differ, p = .35. There was no main effect of group, F (1, 55) = 1.86, MSE = .02, p 2 = .03, but there was a margi nally significant interaction, F (2, 110) = 2.45, MSE = .002, p 2 = .04. Monolinguals' and bilinguals' mean accuracy for each type of picture distractor are shown in Table 4 8 . Following up this interaction by examining the effect of distractor pict ure valence separately for each group, bilinguals responded less accurately on trials with either positive valence distractors ( M = 86.54%, SD = 10.05%) and negative valence distractors ( M = 87.58%, SD = 9.48%) , which did not differ, p = .28 , than trials w ith neutral valence distractors ( M = 90.58%, SD = 6.52%), p = .001 and p = .009 respectively. In contrast, monolinguals' accuracy was equivalent for all types of distractors, those accompanied by positive ( M = 89.24%, SD = 8.34%), negative ( M = 89.78%, SD = 7.67%), and neutral ( M = 90.47%, SD = 7.64%) distractors, all p 's > .31. Looking at the effect of group within each level of distractor picture valence did not reveal any differences between monolinguals and bilinguals, p s > .26. Valence Carryover Effec ts on Neutral Filler Trials In this experiment , every other trial was a filler trial accompanied by a neutral valence distractor picture. The purpose of this was to ensure that any lingering effects of valence from one experimental trial would have subside d prior to the onset of a new trial. This design also allowed for analysis of whether the valence of the distractor on an experimental trial carried over to the subsequent filler trial. All RT data for these analyses were trimmed of outliers as

PAGE 53

53 previously described, and incorrect responses and trials where participants failed to respond in the allotted time were considered inaccurate for accuracy analyses. Because the only new factor in these analyses is Previous Trial Distractor Valence, only effects relev ant to this factor are reported below. Emotional arrows t ask Response t ime s . A 2 (Group: monolingual, bilingual) x 3 (Previous trial distractor valence: positive, negative, neutral) mixed factorial ANOVA was conducted on mean RT s for filler trials. There was no main effect of previous trial distractor valence, F (2, 112) = 2.13, MSE = 134.20, p 2 = .04 and no interaction with this variable, F < 1. Accuracy . A 2 (Group: monolingual, bilingual) x 3 (Previous trial distractor valence: positive, negative, neutral) mixed factorial ANOVA was conducted on the mean proportion of accurate respon ses. There was a main effect of previous trial distractor valence, F (2, 112) = 4.45, MSE = .0002, p 2 = .07, such that participants responded marginally more accurately to filler trials preceded by a trial with a negative valence distractor ( M = 98 .8%, SD = 2.4%) than to those preceded by positive valence distractors ( M = 98.3%, SD = 4.01%), p = .10, or neutral valence distractors ( M = 98.3%, SD = 3.9%), p = .10, which did not differ, p = .81. However, there was no main effect of group, F < 1, signi ficant interaction, F (2, 112) = 1.08, MSE = .0002, p 2 = .02. E motional digit parity t ask Response t ime s . A 2 (Group: monolingual, bilingual) x 3 (Previous trial distractor valence: positive, negative, neutral) mixed factorial ANOVA was conducted on mean RT s for the trimmed filler trial data, and t here were no main effect of previous distractor trial valence or interaction, all F 's < 1.

PAGE 54

54 Accuracy . Accuracy was computed on filler trials for all participants, and incorrect responses and trials where participants failed to respond in the allotted tim e were considered inaccurate. A 2 (Group: monolingual, bilingual) x 3 (Previous trial distractor valence: positive, negative, neutral) mixed factorial ANOVA was conducted on the mean proportion of accurate responses. Results revealed a main effect of prev ious trial distractor valence, F (2, 112) = 7.07, MSE = .001, p 2 = .11, such that participants responded more accurately to filler trials preceded by a trial with a positive valence distractor ( M = 90.3%, SD = 7.8%) or a negative valence distracto r ( M = 90.3%, SD = 6.7%), than to those preceded by a neutral valence distractors ( M = 88.5%, SD = 8.1%) , p = .005 and p = . 01 respectively . However, accuracy did not differ between filler trials preceded by trials with positive and negative valence distr actors, p = .88. There was no interaction with group , F (2, 112) = 1.91, MSE = .001, p = .15 2 = .03. Discussion The purpose of Experiment 2 was to compare monolinguals' and bilinguals' ability to control attention in the presence of emotionally distracting pictures and to test the hypothesis that bilinguals would have an advantage over monolin guals in controlling attention in the presence of emotionally distracting stimuli. The hypothesis was not supported by either RTs or accuracy in either the arrows or digit parity task. With respect to RTs, monolinguals' and bilinguals' RT s did not differ f or either task, although both groups' RTs were sensitive to the presence of emotional pictures, demonstrating slower RTs following positive and negative pictures relative to neutral pictures, and therefore the results provide no support for the idea the bi linguals were better at controlling distraction from the emotional pictures. In contrast to Experiment 1, bilinguals experienced no advantages in terms of RTs for either task in this experiment, and bilinguals responded less accurately for digit parity. W hile one explanation for the present findings is that bilingual advantages are limited to attentional tasks

PAGE 55

55 involving cognitive rather than emotional distraction, it is important to note that nature of the cognitive distraction in Experiment 1 and emotiona l distraction in Experiment 2 are inherently different in terms of their similarity to the to be performed task because distractors in the Experiment 1 were task relevant, and those in Experiment 2 were task irrelevant. For example, the distractors in the flanker task were flanking arrows, and the distractors in the digit parity task were digits, making it critical to ignore the distraction for accurate performance on both tasks. Distractors are particularly susceptible to interference when they cue a respo nse that is not the correct one, such as incongruent flankers and incongruent and ambiguous distractor digits. In contrast, the distractors in Experiment 2 were emotional pictures, which have no relationship to either the target arrows or digits, and their potential for interference comes from their physiological arousal and not information that could cue an inappropriate response. One could therefore argue that the tasks in Experiment 2 were cognitively less demanding than the tasks in Experiment 1 and m ay not have engaged the cognitive control mechanisms that are necessary for bilingual advantages to occur. Research has shown that increases in dACC activation occur on tasks that induce response conflict possibilities ( Barch et al., 2000; Botvinick, Brave r , Barch , Carter, & Cohen, 1999 ), and this brain region has also been hypothesized to be crucial in the emergence of bilingual advantages (Abutalebi et al.. 2012). Furthermore, the dACC is thought to be involved in deployment of cognitive control resources by monitoring for conflicts in information processing ( Botvinick et al., 2004 ), and d etection of conflict increases dACC activation, which in turn signals the need for enhanced cognitive control . Since the tasks in Experiment 2 did not induce any response conflicts, the cognitive control network from which bilingual advantages are thought to arise may not have been strongly engaged for either group, in the presence of task irrelevant emotional distraction.

PAGE 56

56 In terms of accuracy, group differences emerged on ly on the digit parity task but in the direction contrary to the hypothesis: Bilinguals responded less accurately to trials with positive and negative valence distractors than to those with neutral distractors, while monolinguals' accuracy was equivalent f or trials independent of distractor picture valence. This result suggests that bilinguals experienced greater distraction from the negative and positive pictures than monolinguals. The emergence of a bilingual disadvantage was unexpected, but the fact that it showed up only in one specific task on one specific measure (digit parity accuracy) means that caution should be used in interpreting its significance. In particular, the potential confound in the digit parity task (lexical access) may be playing a n un anticipated role, and as previously mentioned, accuracy is not an ideal measure upon which to base claims about bilin gual advantages or disadvantages given that neither advantages nor disadvantages generally emerge. Furthermore, there were contradictory re sults for accuracy on the digit parity task between Experiment 1, where bilinguals manifested a marginal advantage over monolinguals, and Experiment 2, where they had a disadvantage relative to monolinguals, suggesting that future replication of these find ings is warranted. The purpose of the analyses involving filler trials was two fold: 1) to test the hypothesis that participants would experience carry over effects on filler trials, i.e., slowing and/or reduced accuracy on filler trials preceded by exper imental trials with positive or negative valence distractor pictures compared to filler trials preceded by experimental trials with neutral distractor pictures, and 2) to test the hypothesis that bilinguals would experience less interference from carryover effects than monolinguals. Results from both tasks provided some support for the idea that the valence of a distractor on a preceding trial carries over to a response on a subsequent trial , but in the opposite direction such that distractor valence facili tate d rather than impaired

PAGE 57

57 responding. P articipants were more accurate in responding to filler trials preceded by negative (arrows and digit parity task) and positive (digit parity task) distractor pictures than to trials preceded by neutral distractor pic tures. Despite the evidence for carryover effects, there were no group differences in any of the analyses of carryover effects , contrary to the hypothesis that bilinguals would be less susceptible to carryover effects. This may be due to the fact that carr yover effects facilitated rather than interfered with responding. Because the valence of distractors on preceding trials did not serve as distractors when responding on subsequent filler trials, there was no necessity for the emergence of a bilingual advan tage in controlling distraction, i.e., no distraction needed to be controlled on filler trials, and hence no group differences. In sum, the results with respect to RTs in Experiment 2 were consistent in both tasks, with no evidence of a bilingual advantag e. Two possible reasons for the lack of a bilingual advantage in RTs in Experiment 2 have been proposed. One is that bilinguals are unable to control attention to emotional stimuli better than monolinguals. The other is that because the distractors in thes e tasks did not cue a conflicting response (or any response), the tasks failed to engage cognitive control sufficiently for either group, resulting in both groups being equally distracted by the pictures. The tasks in Experiment 3, which include d both irre levant emotional picture distractors and cognitive distractors that can cue a (sometimes conflicting) response, can adjudicate between these explanations. If the first explanation is correct, then bilinguals and monolinguals should both be equally influenc ed by the emotional distractor pictures in Experiment 3. If the second explanation is correct, then a bilingual advantage in controlling attention in the presence of emotional distractors should emerge in Experiment 3. Experiment 3 also explore d whether th e

PAGE 58

58 facilitative carryover effects observed in this experiment replicate under conditions of simultaneous cognitive and emotional distraction.

PAGE 59

59 Table 4 1 . Monolinguals' and bilinguals' self reported language proficiency and exposure Group Monolingu als Bilinguals Variable Mean SD Mean SD ** % Exposure to Dominant Language 94.5 18.1 67.1 23.7 ** % Exposure to Non Dominant Language 2.1 3.5 31.5 21.7 * Proficiency Speaking Dominant Language (0 10 scale) 9.5 .63 9.0 1.0 ** Proficiency Sp eaking Non Dominant Language (0 10 scale) 1.2 .74 8.6 1.0 Proficiency Understanding Dominant Language (0 10 scale) 9.5 .73 .9.2 .89 ** Proficiency Understanding Non Dominant Language (0 10 scale) 1.6 .94 8.4 1.4 * p < .05, ** p < .001 Table 4 2 . Monolinguals and bilinguals' demographic characteristics and cognitive tests Group Monolinguals Bilinguals Variable Mean SD Mean SD ** Age (years) 18.8 1.2 20.3 1.3 * Education (years) 12.8 2.4 14.3 3.0 College GPA (out of 4.0) 3.5 . 28 3.5 .37 Health (1 10 scale) 8.2 1.4 8.1 1.5 Gaming Hours per month 11.8 10.5 28.5 40.2 Vocabulary (out of 25) 13.3 3.6 13.5 4.0 Forward Digit Span 7.1 1.4 6.6 1.4 Backward Digit Span 4.8 1.4 4.9 1.5 * p < .05, ** p < .001

PAGE 60

60 Table 4 3 . Mean v alence for pictures in each picture set Picture Set Set A Set B Distractor Picture Type Mean SD Mean SD Negative 2.2 .42 2.2 .35 Neutral 4.9 .45 4.9 .45 Positive 7.1 .45 7.1 .43 Table 4 4 . Mean arousal for pictures in each picture set Picture Set Set A Set B Distractor Picture Type Mean SD Mean SD Negative 6.0 .57 6.0 .52 Neutral 3.4 .48 3.4 .52 Positive 5.8 .56 5.8 .58 Table 4 5. Experiment 2 Emotional flanker task response times (in ms) Group Monolinguals Bilinguals Distractor Picture Valence Mean SD Mean SD Positive 686.5 45.5 698.0 60.7 Negative 701.8 48.5 716.9 67.9 Neutral 674.5 40.4 687.4 63.2 Table 4 6 . Experiment 2 Emotional flanker task accuracy (in %) Group Monolinguals Bil inguals Distractor Picture Valence Mean SD Mean SD Positive 98.2 2.1 97.8 3.1 Negative 98.2 1.9 98.1 3.4 Neutral 98.9 1.5 97.9 4.1 Table 4 7. Experiment 2 Emotional digit parity task response times (in ms) Group Monolinguals Bilingu als Distractor Picture Valence Mean SD Mean SD Positive 965.2 75.4 961.7 82.9 Negative 991.9 88.5 969.8 86.2 Neutral 941.7 70.1 933.3 80.5

PAGE 61

61 Table 4 8. Experiment 2 Emotional digit parity task accuracy (in %) Group Monolinguals Biling uals Distractor Picture Valence Mean SD Mean SD Positive 89.3 8.3 86.3 10.1 Negative 89.8 7.7 87.4 9.6 Neutral 90.4 7.6 90.4 6.6

PAGE 62

62 Experiment 2 Targets and Distractors Positive Negative Neutral Figure 4 1 . Targ ets and distractors for Experiment 2, emotional distraction 3 1 3 1 3 1 Arrows Task Digit Parity

PAGE 63

63 CHAPTER 5 EXPERIMENT 3 COGNITIVE AND EMOTIO NAL D ISTRACTION Method Participants Sixty participants, 3 0 bilinguals fluent in English and Spanish ( 60.0 % female; 90.0% Hispanic , 10.0% Caucasian ) and 3 0 monolinguals ( 70.0 % female; 8 0.0 % Caucasian, 13.3 % African American, 6.7 % Asian ) fluent only in English , were recruited and compensated as for Experiments 1 and 2. Participants were also pre screened for their language status and categorized as mon olingual or bilingual based upon the same criteria as was used in Experiments 1 and 2. The groups did not differ in their proficiency in speaking or understanding their dominant language, t s< 1. Consistent with both experiments, bilinguals reported greater non dominant language proficiency in speaking, t ( 57 ) = 27.00 , p < .001, and understanding, t ( 57 ) = 27.91 , p < .001 , than monolinguals . Bilinguals also reported less exposure to their dominant language, t ( 58 ) = 4.71 , p < . 001, and more exposure to their no n dominant language, t ( 57 ) = 7.79 , p < .001 , than monolinguals . Language demographics for the groups are presented in Table 5 1. Monolinguals' and bilinguals' demographic characteristics as well as performance on several cognitive tests are presented in Ta ble 5 2. The groups did not differ in terms of age, t < 1 , years of formal education, t ( 5 8) = 1.08 , p = 28, self reported college GPA, t (38) = 1.23 , p = .2 5 , health rating , t ( 5 8) = 1.08 , p = 28 , gaming hours per month, t ( 24 ) = 1. 56 , p = . 13 , vocabulary sc ore, forward digit span, t < 1 , p = . 20, or backward digit span, t < 1 , t ( 5 8) = 1.06 , p = 29 .

PAGE 64

64 Materials The materials were identical to those in Experiment 1, where distractors were cognitive in nature. Identical picture sets from Experiment 2 were used here for emotional and neutral distractors as well as fillers. Procedure The flanker task was the same as Experiment 1 , except for the addition of an emotional or neutral distractor (picture) along with the fl anker distractors on each trial . The digit pari ty task was identical to Experiment 1, except for the inclusion of an emotional or neutral distractor (picture) on each trial along with the cognitive distractor (digit ) distractor on the digit parity task . Half of the time, the target digits and distracto r appeared above the picture distractor, and for the other half of the trials, the target digits and distractor appeared below the distractor picture. See Figure 5 1 for examples of the stimuli for both the task s in this experiment . As in Experiment 2, all participants rated the valence of a subset of the pictures used in the tasks. Results Participants' Valence Rating s To evaluate whether monolinguals and bilinguals perceived the valence of the pictures similarly, a 2 (Group: monolingual, bilingual) x 3 (V alence: positive, negative, neutral) mixed factorial ANOVA compared the mean valence ratings given by participants, and results did not reveal a main effect of group, F < 1, or an interaction, F ( 2, 116 ) = 1.06 , MSE = .49 , p = .35 2 = .02. However, the ma in effect of valence was significant, F ( 2, 116 ) = 715.65 , MSE = .49 , p < .001 2 = .92, such that negative pictures were rated lower than neutral pictures, p < .001, which were rated lower than positive pictures, p < .001, consistent with the valence cate gories suggested by IAPS ratings.

PAGE 65

65 Flanker Task with Emotional Distractors Response t ime s Incorrect responses and practice trials were excluded from analysis , and the data were trimmed of outliers that were more than 2 standard deviations above or below ea ch participant's mean for each of the 9 trial types . Trimming resulted in similar data loss across groups ( 4 . 1 % for monolinguals and 4.4 % for bilinguals ) and for congruent trials (positive distractor: 4 . 1 %; negative distractor: 5.1 %; neutral distractor: 3 . 0%), incongruent trials (positive distractor: 4.4 %; negative distractor: 5.0 %; neutral distractor: 4.2 %), and neutral trials (positive distractor: 4 . 0 %; negative distractor: 4. 9 %; neutral distractor: 3. 8 %). Despite trimming the data, the data remained quit e skewed as indicated by a standardized z score for skew , which was 21.08 , compared to 4.9 for Experiment 1 and 7.6 for Experiment 3. To reduce th e skew , participants whose mean RT s on incongruent trials with negative distractors (the condition that elici ted the slowest RT s ) that were more than 2 standard deviations above or below the mean for their groups were removed. This resulted in removal of 2 monolingual and 2 bilingual participant s . Removal of these four participants reduced the skew of the data to a z score of 4. 9 but did not substantially change the means for either group. A 2 (Group: monolingual, bilingual) x 3 (Flanker Type: neutral, congruent, incongruent) x 3 (Distractor Picture Valence: positive, negative, neutral) mixed factorial ANOVA was conducted on mean RT s with language balance and gaming added as covariate s. While there was no main effect of group, F (1, 52) = 1.16, MSE = 39436.45, p 2 = .02, there was a main effect of flanker type, F (2, 112) = 85.96, MSE = 1943.45, p 2 = .61, such that neutral flanker trials ( M = 764.95 , SD = 63.66 ) were responded to marginally faster than congruent flanker trials ( M = 770.02, SD = 62.62), p = .08, which were faster than incongruent flanker trials ( M = 859.72 , SD = 78.42 ), p < .001 . Ther e was also a main effect of distractor picture valence,

PAGE 66

66 F (2, 112) = 8.14, MSE = 836.98, p 2 = .13, such that participants responded faster to trials accompanied by neutral distractor pictures ( M = 785.84 , SD = 63.55 ) than to those accompanied by positive ( M = 801.17 , SD = 66.92 ) or negative ( M = 807. 68 , SD = 71.56 ) distractor pictures , p 's < .001 . RT s to trials with positive distractor pictures were marginally faster than trials with negative distractor, p = . 07 . These main effects were qualified by a marginally significant Flanker Type x Distractor Picture Valence interaction, F (4, 20 8) = 2.24, MSE = 764.39, p = .07, 2 = .04 , a marginal Group x Flanker Type interaction, F (2, 104) = 2.37, MSE = 1468.57, p 2 = .04, and a marginal three way interaction of group x flanker type x distractor picture valence , F (4, 208) = 2.04, MSE = 764.39, p 2 = .04. Mean R T s for each group is broken down by flanker type and distractor picture valence in Table 5 3 . Following up on the highest order, three way interaction within each group , a 3 (Flanker Type: neutral, congruent, incongruent) x 3 (Distractor Picture Valence: positive, negative, neutral) mixed factorial ANOVA for monolinguals revealed a significant interaction for the monolingual group shown in Figure 5 2 , F (4, 100) = 2.50, MSE = 657.56, p = .05 2 = .09 . While the effect of distractor picture valence was significant for monolinguals for all flanker types, i.e., congruent, F (2, 50) = 12.69, MSE = 518.47, p 2 = .34, incongruent, F (2, 50) = 3.23, MSE = 979.45, p 2 = .11, and neutr al, F (2, 50) = 5.27, MSE = 472.83, p 2 = .17, the basis of the interaction stemmed from positive valence pictures. In particular, for incongruent trials with positive distractor pictures , RTs were particularly slowed such that they were equivalent to RTs for trials with negative pictures , whereas for congruent and neutral trials, RTs following positive pictures were faster than trials with negative pictures. Indeed, comparing the groups directly on incongruent trials when distractors were positive showed that bilinguals responded marginally faster ( M = 839.6, SD = 86.0) than monolinguals ( M = 895.9, SD = 98.2),

PAGE 67

67 F (3, 52) = 2.76, MSE = 8584.25, p = .10 2 = .05 . Trials with positive valence distractors were also slowed relative to trials with neutral valence distractors when flankers were congruent and incongruent, but not when flankers were neutral . In contrast, the Flanker Type x Distractor Picture Valence ANOVA for the bilingual group showed no interaction, F (4, 100) = 1.83, MSE = 892.24, p 2 = .06, and no effect of distractor picture valence, F < 1. There was only a main effect of flanker type, F (2, 50) = 26.01, MSE = 1201.01, p 2 = .51, such that bilinguals responded faster to neutral ( M = 757.3, SD = 52.35) and congruent ( M = 762.3, SD = 50.87) trials than to incongruent trials ( M = 848.2, SD = 65.99), p < .001, but did not differ for neutral and congruent trials, p = .20. Figure 5 3 shows this nonsignificant interaction separately for bilinguals. The three way interaction was also followed up by looking at between group differences for conflict effects within each level of valence. C onflict effects arising from incongruent trials were evaluated in two ways : (1) by comparing incongruent to neutral RTs ( i.e., termed neutral conflict effect) , and (2) by comparing incongruent to congruent RTs ( i.e., termed congruent conflict effect) . To evaluate neutral conflict effects, three 2 (Grou p: monolingual, bilingual) x 2 (Flanker Type: neutral, incongruent) repeated measures ANOVAs were conducted on participants' mean RTs within each level of valence . Similarly, t o evaluate congruent conflict effects, three additional 2 (Group: monolingual, b ilingual) x 2 (Flanker Type: congruent, incongruent) repeated measures ANOVAs, one for each level of valence, were conducted on participants' mean RTs. A significant group x flanker type interaction in these analyses indicates that the groups differed in t he size of their conflict effects. Results of these analyses with neutral valence distractor pictures revealed no group x flanker type interaction, F s < 1, indicating equivalent neutral and congruent conflict effects for

PAGE 68

68 both groups . However, when distrac tor pictures were positive, there was a significant group x flanker type interaction for the neutral conflict effect, F (1, 52) = 5.00, MSE = 2015.86 , p = . 0 3, 2 = .0 9, such that interference effects from incongruent trials relative to neutral trials were smaller for bilinguals ( M conflict = 80.1, SD conflict = 58.0) than for monolinguals ( M conflict = 131.8, SD conflict = 65.5 ) . Furthermore, w hen distractors were negative, there was a marginally significant group x flanker type interaction for the congruent conf lict effect, F (1, 52) = 3.52, MSE = 858.01 , p = . 07 2 = .0 6, such that conflict effects as measured by congruent versus incongruent trials were smaller for bilinguals ( M conflict = 63.2 , SD conflict = 43.4 ) than for monolinguals ( M conflict = 91.5 , SD conflict = 41.7 ) . Table 5 4 shows the size of the neutral and congruent conflict effects for each level of distractor valence. N o other group x flanker type interaction s were significant , p s > .18 . Accuracy Accuracy was computed on all experimental trials with incorrect responses and trials where participants failed to respo nd in the allotted time considered inaccurate. A 2 (Group: monolingual, bilingual) x 3 (Flanker Type: neutral, congruent, incongruent) x 3 (Distractor Picture Valence: positive, negative, neutral) mixed factorial ANOVA with language balance and gaming as covariates was conducted on mean proportion of accurate responses. Results revealed a main effect of flanker type, F (2, 1 04 ) = 11.09 , MSE = .0 04 , p 2 = . 18 , such that participants responded m ore accurately to congruent trials ( M = 9 7. 5 % SD = 8.2 %) and neutral trials ( M = 97.0 %, SD = . 7.5 % ) than to inc ongruent trials ( M = 91.0 %, SD = 10.4 %), p's < .001, but did not differ with respect to congruent and neutral trials, p = .1 9 . There was no main effect of distractor picture valence , F (2 , 104) = 1.21, MSE = .002, p 2 = .02, or of group, F (1, 52) = 1.42, MSE = .06, p 2 = .03 . None of the interactions were significant, p s > 28 . Table 5 5 shows the mean accuracy for each group for each flanker type and distractor valence.

PAGE 69

69 Digit Parity T ask with Emotiona l Distractors Response t ime s Only experimental trials where participants responded correctly were included in the analyses. After excluding practice trials and incorrect trials, the data were trimmed of outliers that were more than 2.5 standard deviations above or below each participant's mean for trials with positive, negative, and neutral distractor pictures. Trimming resulted in minimal data loss for both groups ( 3 . 6 % for monolinguals and 4 .0% for bilinguals ) and for congruent trials ( positive distracto r: 2 . 4 % ; negative distractor: 1.3 %; neutral distractor: 3.0 % ), incongruent trials (positive distractor: 2.1%; negative distractor: 2.1% ; neutral distractor: 4.2 %), neutral trials (positive distractor: 3.4%; negative distractor: 2.8 %; neutral distractor: 4. 3 %), and ambiguous trials (positive distractor: 3.9 %; negative distractor: 3.4 %; neutral distractor: 4.8 %) . A 2 (Group: monolingual, bilingual) x 4 (Digit Distractor Type: congruent, incongruent, neutral, ambiguous) x 3 (Distractor Picture Valence: positiv e, negative, neutral) mixed factorial ANOVA was conducted on mean RT s . Results revealed a main effect of digit distractor type, F (3, 159) = 16.93, MSE = 3646.46, p 2 = .24, such that RT s to trials with congruent ( M = 963.82, SD = 87.12) and neutra l distractors ( M = 964.87, SD = 78.74), which did not differ, p = .84, were faster than RT s to trials with incongruent distractors ( M = 991.59, SD = 87.88), p < .001, which were faster than responses to trials with ambiguous digit distractors ( M = 1037.28, SD = 85.46) p s = .001. No other effects were significant, p s > .34 . Table 5 6 provides a breakdown of mean RT s for each group in each digit distractor and picture valence condition. The se results did not change when participants were excluded as in the fl anker task , i.e., when the monolingual and bilingual participants with the slowest RT s and the monolingual and bilingual participants with the fastest RT s on incongruent trials with negative distractors were excluded.

PAGE 70

70 Accuracy Accuracy was computed on ex perimental trials for all participants , and incorrect responses and trials where participants failed to respond in the allotted time were considered inaccurate. A 2 (Group: monolingual, bilingual) x 4 ( Digit Distractor Type: neutral, congruent, incongruen t, ambiguous) x 3 (Distractor Picture Valence: positive, negative, neutral) mixed factorial ANOVA was conducted on mean proportion of accurate responses. Results revealed a main effect of digit distractor type, F (3, 1 68 ) = 10.56 , MSE = .0 3 , p 2 = . 1 6 , such that participants responded more accurately to ambiguous trials ( M = 89.0 % , SD = 10.7%) than to neutral trials ( M = 85. 7 % , SD = 10.6%) , p < .001 and congruent trials ( M = 84.0% , SD = 15.4%), p = .00 2 , both of which were more accurate than inco ngruent trials ( M = 8 0.5 %, SD = 16.12 %) , p < .0 0 1 and p = .01 respectively . There was a lso a marginally significant main effect of distractor picture valence, F (2, 11 2 ) = 2. 45 , MSE = .0 1 , p = .0 9 2 = .0 4 , such that participants responded more accurately to trials accompanied by neutral distractor pictures ( M = 85.7 %, SD = 11.0% ) than those accompanied by negative distractor pictures ( M = 83.8 %, SD = 13.3 %) , p = .02 . T rials accompanied by positiv e distractors ( M = 84.8 %, SD = 1 3.0%) did not differ from either negative distractors, p = .25, or neutral distractors, p = .21 . Table 5 7 shows the means for each group in each digit distractor and picture valence condition. T here was no significant effec t of group, F (1, 56) = 1.46, MSE = .18, p = .2 2 = .02, and there were no significant interactions involving group, all F 's < 1.8, all p 's > .15. There was, however, a significant interaction between digit distractor type and distractor picture valence, F (6, 336) = 2.62, MSE = .01, p 2 = . 04. The interaction was followed up by looking at distractor picture valence within each level of distractor digit type , and results revealed that the basis of the interaction arose from incongruent digit distractors. T here was an effect of distractor pict ure valence only for trials accompanied by incongruent distractor digits,

PAGE 71

71 F (2, 118) = 3.23, MSE = .01, p = .04 2 = .05. When digit distractors were incongruent, participants responded more accurately to trials accompanied by neutral valence pictures ( M = 83.4%, SD = 13.7% ) than to those accompanied by negative valence pictures ( M = 78.3%, SD = 19.6% ), p = .01, a nd marginally more accurately than trial s accompanied by positive valence pictures ( M = 79.8%, SD = 22.1% ), p = .07, while trials with positive and neutral valence distractors did not differ in accuracy , p = .49 . No effect of distractor picture valence eme rged when digit distractors were congruent, p = .61 , neutral, p = .95 , or ambiguous, p = .40 . Figure 5 4 shows this interaction. Valence Carryover Effects on Neutral Filler Trials As with Experiment 2 , every other trial in this experiment was a filler tria l accompanied by a neutral valence distractor picture , which allowed for analysis of whether the effect of valence of the distractor on an experimental trial carried over to the subsequent filler trial. RT s were trimmed of outliers as previously described, and incorrect responses and trials where participants failed to respond in the allotted time were considered inaccurate for accuracy analyses . Only effects relevant to Previous Trial Distractor Valence are reported below. Experiment 3 f lanker t ask with e motional d istractors Response times. The filler trials were trimmed of outliers as described above, and a 2 (Group: monolingual, bilingual) x 3 (Flanker type: congruent, incongruent, neutral) x 3 (Previous trial distractor valence: positive, negative, ne utral) mixed factorial ANOVA was conducted on mean RT s. There was no main effect of previous trial distractor valence, F (2, 104) = 1.66, MSE = 592.62 , p = . 19 2 = .03, and no significant interactions with this factor , F s < 1. Accuracy . Accuracy was computed on filler trials for all participants, and incorrect responses and trials where participants failed to respond in the allotted time were considered inacc urate, and a 2 (Group: monolingual, bilingual) x 3 (Flanker type: congruent, incongruent,

PAGE 72

72 neutral) x 3 (Previous trial distractor valence: positive, negative, neutral) mixed factorial ANOVA was conducted the mean proportion of accurate responses. T here was no main effect of previous trial distractor valence or any interactions with this factor , F s < 1. Experiment 3 digit parity t ask with emotional d istractors Response times. A 2 (Group: monolingual, bilingual) x 4 (Digit distractor type: congruent, incon gruent, neutral, ambiguous) x 3 (Previous trial distractor valence: positive, negative, neutral) mixed factorial ANOVA was conducted on mean RT s for filler trials, and results revealed a main effect of previous trial distractor valence, F ( 2 , 112 ) = 5.78 , M SE = 3492.37 , p < .001 2 = . 09, such that participants responded faster to filler trials preceded by a trial with a positive valence distractor picture ( M = 944.1, SD = 92.9) or a negative valence distractor picture ( M = 948.9, SD = 78.5) than to filler trials preceded by a t rial with a neutral valence picture distractor ( M = 976.6, SD = 79.0), p < . 001 , but RT s for filler trials preceded by trials with positive and negative distractor pictures did not differ, p = .18. However, there were no interactions with this factor , all F s < 1. Accuracy . A 2 (Group: monolingual, bilingual) x 4 (Digit distractor type: congruent, incongruent, neutral, ambiguous) x 3 (Previous trial distractor valence: positive, negative, neutral) mixed factorial ANOVA was conducted on mean propo r tion of a ccurate responses , and results revealed only a marginal main effect of previous trial distractor valence , F (2, 112) = 2.36 , MSE = .009 , p = .10 2 = .0 4 , but this marginal effect did not reveal and significant pairwise comparisons , and there were no inte ractions with this factor, F < 1 . Discussion The purpose of Experiment 3 was to test the hypothesis that bilinguals would control attention in the simultaneous presence of cognitive and emotional distractors better than monolinguals. Results from the fla nker task provide support for this hypothesis in several ways.

PAGE 73

73 First, only monolinguals were affected by distractor picture valence, responding more slowly in the presence of negative and positive pictures relative to neutral pictures, whereas bilinguals' RTs were equivalent regardless of distractor picture valence. Second, bilinguals experienced smaller conflict effects than monolinguals on incongruent flanker trials accompanied by positive and negative distractor pictures. Furthermore, bilinguals also res ponded marginally faster than bilinguals for incongruent trials with emotional (positive) distractors, circumstances that offered the most distraction. These findings demonstrate an advantage over monolinguals in controlling attention to cognitive distract ors, specifically in the presence of emotional distraction, and they support the hypothesis that bilinguals would control attention in the presence of joint cognitive and emotional distraction better than monolinguals. Results from the digit parity task , o n the other hand, provide no support for the hypothesis , as monolinguals and bilinguals equivalently experienced interference from incongruent trials in conjunction with emotional distraction, evidenced by slower RTs overall and reduced accuracy specifical ly for negative and positive pictures relative to neutral pictures. The lack of a bilingual advantage for the digit parity task is consistent with the findings of Experiment 1, as was the sensitivity of both groups to the digit distractor type, with incong ruent trials and ambiguous trials slowing RTs relative to the other conditions. However, unlike Experiment 1, ambiguous trials had the highest accuracy, suggesting a speed accuracy tradeoff for ambiguous trials that did not occur for incongruent trials. A bilingual advantage may again have failed to emerge in this experiment because of participants accessing the words for the digits, contrary to some claims that the digit parity task is non lexical ( Fias et al., 2001 ). Because bilinguals are slower in lexic al access than monolinguals, this slowdown would offset any speed advantage that bilinguals might have in controlling distraction.

PAGE 74

74 These results also extend the results from Experiment 2, where no bilingual advantage was found when distractors were emotio nal distractor pictures, although both groups did experience slowing in the presence of those distractors. When cognitive distraction (flankers) was added along with the emotional distractors in Experiment 3, a bilingual advantage in managing these distrac tions emerged, supporting an interpretation where high levels of cognitive control are necessary for a bilingual advantage in managing emotional distraction to occur. The addition of the cognitive distractors allowed for an unpredictable response cue, one that was either consistent or conflicted with the correct response. The flanker task would therefore be expected to activate the dACC , consistent with evidence from previous research (e.g., Abutalebi et al., 2012), which in turn would have signaled the nee d for increases in cognitive control for both monolinguals and bilinguals. Based on the present results, bilinguals' better ability to control attention may be limited to contexts when cognitive control is up regulated, which allowed them to completely sup press interference from the emotional pictures in Experiment 3 so that RTs were equivalent regardless of distractor picture valence. In contrast, monolinguals experienced decrements in performance due to interference from the emotional distractor pictures. Thus, bilinguals can control attention in the presence of emotional distractors more effectively than monolinguals but only when demands on attentional control are high. While the results in terms of accuracy on the flanker task replicated the difficulty of incongruent trials found in Experiment 1, where incongruent trials were responded to less accurately, and extended that difficulty in this experiment to the digit parity task, there was no evidence of a bilingual advantage in either task. Whereas the f lanker task consistently demonstrated no group differences in accuracy across all three experiments, the digit parity task exhibited inconsistent group effects in accuracy across experiments, again underscoring the need

PAGE 75

75 for interpreting these results with caution. Nonetheless, independent of group, the digit parity task was sensitive to detecting an effect of valence in accuracy that was specific to incongruent trials, where accuracy was lower in the presence of emotional relative to neutral distractors . Th is interaction suggests that when the demand on cognitive control was highest , i.e., due to the presence of an incongruent distractor digit, controlling distraction from the emotional pictures was especially difficult . This finding supports the idea that t he combined sources of distraction makes the digit parity task more difficult, similar to the flanker task, but that this increase in difficulty did not dissociate bilinguals' and monolinguals' performance unlike the flanker task . In terms of carryover ef fects on filler trials, RT s on the digit parity task revealed an effect of previous trial valence. P articipants responded faster on filler trials preceded by trials with a positive or negative valence distractor than to those preceded by a trial with a neu tral valence distractor . Similar to the carryover effects on accuracy in Experiment 2, these f acilitation on RTs were equivalent for bilinguals and monolinguals, again providing counterevidence to the hypothesis being tested. These facilitation effects may be related to the difference between emotional distractors that are words versus pictures. Studies showing inhibitory carryover effects have typically involved emotional words (e.g., Ashley & Swick, 2009; Bertels et al., 2011 ), whose lexical and semantic properties get activated upon presentation. Even after a word is no longer visible, residual activation in the semantic network associated with an emotional word can continue to provide distraction ( e.g., Ashley & Swick, 2009 ) , which could result in an inh ibitory carryover effect. When the distractors are emotional pictures as in the present experiments, the visual nature of the distractors may become less distracting once the picture disappears and can even be helpful to processing. Research by Phelps, Lin g, and Carrasco (2006) supports this interpretation by showing that emotional pictures facilitate perception and attention, i.e.,

PAGE 76

76 identification of the orientation of a Gabor patch shown in Figure 5 5 , following the offset of those pictures. The mechanism underlying facilitation following emotional pictures maybe due to the lingering effects of arousal that such pictures produce. In addition to emotional arousal heightening perceptual sensitivity (Lane, Chau, & Dolan, 1999; Phelps et al., 2006), research indicates that the late positive potential (LPP), which indexes attention to emotional pictures, results in physiological arousal that persists for 6 seconds after the offset of arousing emotional pictures and is associated with emotional arousal (Cuthbert , Schupp, Bradley, Birbaumer & Lang, 2000).

PAGE 77

77 Table 5 1. Monolinguals' and bilinguals' self reported language proficiency and exposure Group Monolinguals Bilinguals Variable Mean SD Mean SD * % Exposure to Dominant Language 92.2 19.7 71. 5 14.5 * % Exposure to Non Dominant Language 3.4 4.9 26.6 15.3 Proficiency Speaking Dominant Language (0 10 scale) 9.5 .57 9.4 .81 * Proficiency Speaking Non Dominant Language (0 10 scale) 1.4 .87 8.4 1.1 Proficiency Understanding Dominant Lan guage (0 10 scale) 9.5 .63 9.4 .76 * Proficiency Understanding Non Dominant Language (0 10 scale) 1.6 .94 8.8 1.0 * p < .001 Table 5 2. Monolinguals and bilinguals' demographic characteristics and cognitive tests Group Monolinguals Bilingua ls Variable Mean SD Mean SD Age (years) 18.6 3.5 19.2 1.3 Education (years) 13.4 2.4 12.8 4.2 College GPA (out of 4.0) 3.34 .47 3.51 .41 Health (1 10 scale) 8.1 1.2 8.4 1.2 Gaming Hours per month 22.2 28.6 9.9 11.7 Vocabulary (out of 25) 13.0 3.7 13.9 3.1 Forward Digit Span 7.2 1.4 6.7 1.3 Backward Digit Span 4.7 1.0 4.7 1.4

PAGE 78

78 Table 5 3 . Experiment 3 Cognitive and e motional f lanker task response times (in ms) Group Monolinguals Bilinguals Flanker Type x Distractor Valence Mean SD Mean SD Congruent Positive 781.8 70.6 767.0 64.4 Negative 787.0 78.3 779.6 56.8 Neutral 762.8 70.5 7 6 0 . 9 52.0 Incongruent Positive 881.2 106.2 851.5 89.9 Negative 873.2 98.8 852.1 79.4 Neutral 859.4 92.1 841.0 56.4 Neutral Positive 769.2 63.9 756.5 45.8 Negative 785.8 84.9 766. 6 66.4 Neutral 765.2 73.1 748.8 49.7 Table 5 4 . Experiment 3 Conflict effects for each level of distractor picture valence (in ms) Group Monolinguals Bilinguals Conflict Effect Distractor Valence Mean SD Mean SD Neutral Positive 131.8 65.5 80.1 58.0 Negative 94.0 40.1 77.1 36.0 Neutral 93.5 39.1 92.0 29.8 Congruent Positive 108.0 58.2 79.7 55.0 Negative 91.5 41.7 63.2 44.4 Neutral 120.9 53.4 100.1 48.8 Table 5 5 . Experiment 3 Cognitive and emotional flanker task accuracy (in % ) Group Monolinguals Bilinguals Flanker Type x Distractor Valence Mean SD Mean SD Congruent Positive 98.4 3.7 96.6 10.6 Negative 98.7 2.6 96. 6 9.6 Neutral 98.7 2.3 95.7 14.3 Incongruent Positive 89.8 15.8 93.1 15.3 Negative 91.4 14.2 90.0 14.1 Neutral 89.6 14.2 91.9 13.7 Neutral Positive 96.8 4.2 95.9 9.3 Negative 98.6 2.7 95.6 11.8 Neutral 97.4 4.9 9 6.9 11.0

PAGE 79

79 Table 5 6 . Experiment 3 Cognitive and emotional digit parity task response times (in ms) Group Monolinguals Bilinguals Flanker Type x Distractor Valence Mean SD Mean SD Congruent Positive 959.7 92.2 954.7 122.4 Negative 984.2 1 09.8 987.6 102.4 Neutral 962.1 95.2 933.7 107.1 Incongruent Positive 1005.1 86.4 990.7 106.7 Negative 1019.3 111.4 992.3 127.9 Neutral 973.2 109.9 968.4 89.5 Neutral Positive 966.5 91.0 957.8 77.3 Negative 977.8 72.8 978.6 118.0 Neutral 968.3 91.1 940.0 72.1 Ambiguous Positive 1040.7 92.7 1055.6 88.0 Negative 1048.4 88.7 1024.6 99.6 Neutral 1035.5 98.5 1018.9 80.9 Table 5 7 . Experiment 3 Cognitive and emotional digit parity task accuracy (in % ) Group Monolinguals Bilinguals Flanker Type x Distractor Valence Mean SD Mean SD Congruent Positive 84.5 21.1 85.3 15.6 Negative 85.0 22.2 80.9 21.2 Neutral 84.0 12.6 84.5 14.6 Incongruent Positive 82.5 18.4 77.2 26. 3 Negative 80.7 15.0 75.9 24.2 Neutral 83.0 12.7 83.9 15.4 Neutral Positive 85.7 12.7 85.9 13.0 Negative 84.5 11.9 87.1 13.1 Neutral 85.1 10.9 85.8 11.3 Ambiguous Positive 88.7 13.2 88.9 12.8 Negative 89.5 10.6 8 6.9 14.0 Neutral 90.5 9.8 89.2 12.2

PAGE 80

80 Positive Negative Neutral Flanker Task Congruent Incongruent Neutral Digit Parity Congruent Incongruent Neutral Ambiguous Figure 5 1 . Targets and distractors for Experiment 3, cognitive and emotional distraction 4 3 1 3 7 1 3 7 1 3 7 1 3 4 1 3 4 1 3 4 1 3 X 1 3 X 1 3 X 1 4 3 1 4 3 1

PAGE 81

81 Figure 5 2 . Monolinguals' flanker task RTs in Experiment 3

PAGE 82

82 Figure 5 3 . Bilinguals' flanker task RTs in Experiment 3

PAGE 83

83 Figure 5 4. T he interaction of digit distractor type and distractor picture valence on accuracy in Experiment 3. Figure 5 5 . Example of a Gabor patch

PAGE 84

84 CHAPTER 6 GENERAL DISCUSSION Findings of bilingual advantages in controlling attention have previously been limited to contexts in which distractors were cognitive in nature. The present study sought to determine whether bilinguals are better than monolinguals in controlling attention more broadly, specifically in the presence of emotional distraction and in the presence of joint cognitive and emotional distraction. Two tasks were used to investigate these questions: the flanker task (and modifications thereof) and the digit parity task. The flanker task was chosen because it consistently ha s shown an effect of cognitive distraction on RT s ( Costa et al., 2008; Eriksen & Eriksen, 1974; Fan et al., 2002; Nieuwenhuis et al., 2006; Pelham & Abrams, 2013 ), an effect which has been useful in dissociating bilinguals and monolinguals ' sensitivity to this distraction. R esults of studies that have used the flanker task have often revealed a bilingual advantage , either in smaller conflicts or overall faster responding (e.g., Abutalebi et al., 2012; Costa et al., 2008, Pelham & Abrams, 2013) . The digit par ity task was selected because it has been shown to be sensitive to an effect of emotional distraction, specifically distraction from emotional pictures, on RT s ( Fernandes et al., 2011 ), but this issue had not previously been assessed in bilinguals. The res ults of these experiments , summarized in Table 6 1, demonstrated some instances of a bilingual advantage, namely in RTs on the flanker task, and revealed that bilinguals are better able to control attention in the presence of emotional distraction than mon olinguals when there are simultaneous cognitive demands. Most explanations of bilingual advantages are predicated on the necessity of managing two languages. Bilinguals engage domain general executive function to monitor conflict between their two language s, in particular the dACC ( Price et al., 1999; Abutalebi & Green, 2007; Rodriguez Fornells et al., 2005; van Heuven, Schriefers, Dijkstra, & Hagoort, 2008 ), which in

PAGE 85

85 turn improves bilinguals' EF, resulting in their responding faster overall and/or having s maller conflict effects than monolinguals on tasks that require executive functioning such as the flanker task ( Abutalebi et al., 2012 ). Our results are consistent with this assertion, as a bilingual advantage on the flanker task emerged . I n Experiment 1, bilinguals manifested an overall speed advantage relative to monolinguals, i.e., bilinguals responded more quickly than monolinguals on all trials (congruent, incongruent, and neutral). This bilingual advantage in speed of responding in the presence of cog nitive distraction replicates previous studies ( Bialystok & Viswanathan, 2009; Emmory et al., 2008 ), although no bilingual advantage emerged in terms of conflict effects. This is not uncommon, as the finding of an overall speed advantage is more consistent ly observed than the bilingual advantage of smaller conflict effects (e.g., as noted in Costa et al., 2008; Hilchey & Klein, 2011). In contrast, no bilingual advantage of either type emerged in Experiment 2 when distractors were solely emotional pictures. A possible interpretation of this finding, i.e., bilinguals are not better than monolinguals at controlling attention to distracting emotional stimuli, was ruled out by the findings of Experiment 3 , where emotional pictures slowed RTs only for monolinguals and not bilinguals, and bilinguals had smaller conflict effects (less interference from incongruent trials) than monolinguals when emotional distractors were present . Thus, bilinguals are able to exhibit an advantage in the presence of emotional pictures. Why did a bilingual advantage in the presence of emotional distraction emerge in Experiment 3 but not Experiment 2? The difference must be attributed to the one variable that changed between the two experiments, cognitive distraction . The presence of cogn itive distractors in Experiment 3 sometimes (33% of the time) cued a conflicting response possibility , which was hypothesized to increase dACC activation, consistent with previous resea r ch

PAGE 86

86 ( Abutalebi et al., 2012 ) , and in turn would signal the need for inc reased cognitive control ( Botvinick et al., 2004 ). Conversely, without conflict response options as in Experiment 2, dACC was not increased, cognitive control was not upregulated, and thus no bilingual advantage emerged. This interpretation is consistent w ith research showing that dACC activation is associated with conflicts in information processing and with other research that clearly links bilingual advantages with dACC activation (Abutalebi et al., 2012). The present results suggest that having distract ors that cue response conflicts is necessary for the emergence of a bilingual advantage in suppressing emotional distraction to occur. I t is important to note that independently of the bilingual monolingual comparisons, the flanker task was sensitive to th e manipulations of both cognitive and emotional distraction. Experiment 1 results revealed a robust influence of flanker type, with incongruent trials slowing RTs relative to congruent and neutral trials, as is typically the case with flanker tasks ( Fan et al., 2002 ). Experiment 2 showed a parallel influence of distractor picture valence, where positive and negative pictures slowed participants' RTs compared to neutral pictures in a simplified arrows task, consistent with findings of the effect of emotional pictures in other simple cognitive tasks, such as identifying the location of a horizontal line as being above or below a distractor picture ( Schimmack, 2005 ). Experiment 3 replicated both effects under conditions that simultaneously presented both types of distraction, demonstrating that cognitive and emotional distraction have independent, interfering influences. Consequently, any differences (or lack there of) between monolinguals and bilinguals cannot be explained by the distractors faili ng to have the intended impact. Turning to the digit parity task, the results were less informative in shedding light on bilinguals' advantage in controlling attention under conditions of distraction , revealing group

PAGE 87

87 differences only for accuracy, not speed, which were i nconsistent across experiments . There was a bilingual advantage in Experiment 1 , a bilingual disadvantage in Experiment 2 , and no group differences in Experiment 3. Furthermore, t he lack of group differences in RTs, even overall RTs, on the digit parity ta sk in all three experiments could be indicative of the task necessitating lexical access of the target and distractor digits . This was unexpected because of Fias et al's ( 2001 ) claim that visually presented digits do not activate lexical properties. They m ade this assertion based on the fact that in a type of picture word interference task, simultaneously presenting an Arabic numeral (e.g. , 5) and a written word corresponding to a different number ( e.g., four) did not slow participants ' naming times for the written word but did slow participants' naming times for the numerals. Hence, they concluded that Arabic numerals are processed like pictures, which do not result in lexical access when they serve as distractors . However, their findings may have been spec ific to the particular task they used. In the present study, retrieving semantic aspects of numerals (i.e., whether they were odd or even) was necessary to successfully perform the digit parity task, suggesting that lexical access is likely to occur. Becau se bilinguals accomplish lexical access more slowly than monolinguals ( Gollan et al., 2008; Ivanova & Costa, 2008; Pelham & Abrams, 2013 ) , their lexical access deficit may have negated any potential executive function benefit in a task that also requires l exical access . In the present study, bilinguals were neither faster nor slower than monolinguals and exhibited equivalent conflict effects on the digit parity task. These findings leave open the possibility that bilinguals experienced an RT advantage on th e part of the digit parity task involving control of attention but that this advantage was offset by their disadvantage in lexical access. Altarriba and Bainsight Brown ( 2010 ) suggested a similar explanation for their findings of a bilingual disadvantage w hen they compared monolinguals and bilinguals on an emotional

PAGE 88

88 Simon task. In that task, participant s pressed a key on the right or left side of the keyboard depending on whether or not a word was positive (e.g., hug, party) or negative (war, grave), and wo rds could appear on the right or left side of the computer screen, making them congruent or incongruent with the correct response hand. Contrary to a bilingual advantage, bilinguals responded more slowly than monolinguals in all conditions , and they had la rger conflict effects. Altarriba and Bainsight Brown proposed that bilinguals' disadvantages in lexical access may have contributed to their findings since the task required participants to access the semantic meaning of the words in order to evaluate whet her they were positive or negative. T o empirically test the hypothesis that digit parity judgments result in lexical access , future research c ould compare neural activation patterns when participants name digits , when they name pictures, and when they make digit parity judgments to see if the these tasks are underpinned by overlapping neural substrates. Since naming digits and naming pictures clearly require lexical access, those condition s would provide a template for neural activation when lexical access occurs . Comparing digit naming to picture naming and judgments of digit parity would allow for identification of neural activity associated with digits as opposed to non digits. Finally, i f digit parity involves lexical access, subtracting neural activity associated only with digits from both digit naming and digit parity judgments, and comparing the remaining neural activation patterns to picture naming , should reveal a common pattern of activation associated with lexical access apart from semantics associ ated with any particular type of word. If participants do not lexically access the words for the digits when they perform the digit parity task, then neural activation common to digit and picture naming should be absent from digit parity judgments. While the digit parity task was insensitive to differences between the groups in the present study, the task was sensitive to the effects of both types of distraction, similar to the flanker task.

PAGE 89

89 Participants in the present study were influenced by the presence of the digit distractor types, where incongruent and ambiguous distractors slowed participants' RTs relative to neutral and congruent distractors, demonstrating a novel influence of a distractor in digit parity tasks. In addition, RT s were slowed in the p resence of positive and negative pictures relative to neutral pictures , replicating findings from previous research with the digit parity task ( Aquino & Arnell, 2007; Fernandes et al., 2011; Thomas & Hasher 2006 ). The effects of valence were particularly ro bust in the digit parity task, emerging both for RTs and for accuracy, with participants responding more accurately on trials with neutral distractor pictures than those with positive (Experiment 2) and negative (Experiments 2 and 3) distractor pictures. It is worth noting that while the presence of emotional pictures had inhibitory effects on responses for experimental trials, the carryover effects of valence on subsequent filler trials was facilitative in nature . I n Experiment 2, negative distractor pict ures on a preceding trial increased accuracy in responding to filler trials on the arrows task relative to neutral distractor pictures , and on the digit parity task, both positive and negative pictures had this effect. In Experiment 3, when filler trials w ere preceded by positive or negative pictures, participants responded faster than if a filler trial was preceded by a trial with a neutral distractor picture. These effect s likely arise from the fact that emotional pictures increase physiological arousal ( Bradley, 2009). There are a number of indexes of such arousal, but one reliable index is galvanic sk in conductance ( GSC, aka sweating), which is controlled by the sympathetic nervous system. Codispoti, Mazzetti, and Bradley (2009) demonstrated that present ation of positive, negative, and neutral pictures for durations as short as 80 msec reliably induced increases in galvanic skin conductance . I nitial increase s in GSC are thought to be related to novelty and occur regardless of valence , whereas only positiv e and negative pictures result in increased GSC when they are no

PAGE 90

90 longer novel, such as with repeated presentation (Bradley, 2009) . Furthermore, Codispoti et al. (2009) demonstrated that even for novel pictures , the effect of increased GSC lasted longer and was greater for positive and negative pictures (6 seconds) than neutral pictures (< 1 second) , showing that positive and negative pictures have a sustained effect on physiological arousal compared to neutral pictures . Carryover effects in the present stud y may therefore reflect the influence of this sustained arousal on subsequent filler trials. An alternative possibility is that participants may experience a reactive increase in cognitive control on trials with highly distracting emotional pictures, and t his increase may persist into the next trial and thereby facilitate responding on that trial. Facilitation in the present study following the offset of emotional pictures is consistent with the findings of Phelps et al. (2006) , where such pictures improve d perceptual sensitivity following the offset of those pictures (e.g., Phelps et al., 2006) . However, the effects of arousal are not always beneficial, as highly arousing (positive and negative) stimuli can also hinder performance on subsequent trials as d emonstrated on the emotional Stroop task ( Ashley & Swick, 2009; Bertels et al., 2011 ) . Future research should attempt to define under what conditions valence and arousal exert inhibitory versus facilitative influences and whether or not those influences ho ld across different types of valenced stimuli (e.g., words versus pictures). O ne l imitation of the present study is that the emotional distraction was always irrelevant to the task . When stimuli are irrelevant, participants do not need to attend to the m to complete the task, and this irrelevance may make them easier to ignore than when they are relevant to the task. It is possible that bilinguals only manifest an advantage in controlling attention to emotional stimuli when those stimuli are task irrelevan t . When the emotional stimuli are relevant, as would be the case in an emotional Simon task where participants would have to pres s

PAGE 91

91 a button to indicate whether an image were positive or negative, attention to the emotional stimuli would be required. Direct ing attention to emotional stimuli might make suppressing interference from that stimuli much more difficult for both monolinguals and bilinguals and eliminate the bilingual advantage exhibited here. In this regard, a n interesting follow up to the present study would involve creating an experimental paradigm where emotional stimuli were task relevant , as in the emotional Simon task suggested above. P articipants would see four types of stimuli presented on the left or right side of a computer screen : positiv e pictures, negative pictures, a "+" sign, or a " " sign. Positive pictures and "+" signs would require a keypress with one hand , while negative pictures or a " " sign would require a keypress with the other hand. This paradigm includes a response conflict possibility in that the position of the picture or symbol on the left or right cues a prepotent left or right handed response , which can be congruent or incongruent with the correct response. B oth participant group s would be expected to respond more slowl y when pictures cued a response than when symbols cued a response because emotional pictures typically slow responding (e.g., Fernandes et al., 2011, Schimmack, 2005; Yamanski et al., 2002 ) by hold ing attention longer than non emotional stimuli (Hajcak & O lvet, 2008 ). However, if bilinguals have an advantag e in controlling attention to em otiona l stimuli when the stimuli are task relevant, then they should be less slowed than monolinguals when emotional pictures cue a response. A modification of this paradig m could also address the question of whether the presence of a possible response conflict is necessary for emergence of bilingual advantages. In this case, the same stimuli would be used to cue responses, but instead of presenting stimuli on the left or ri ght side of the computer screen, all stimuli would be presented in the center of the screen, and thus they would never cue any prepotent response. If

PAGE 92

92 bilingual advantages only emerge when conflicts in information processing occur, than no bilingual advanta ge should arise when the stimuli are centrally presented on the computer screen. Other limitations relate to whether or not these results generalize to all bilinguals. As noted in the introduction, bilinguals are extremely diverse, but in the present study , only Spanish E nglish bilinguals were included when different language combinations might affect attentional control differently. For example, it may be that managing language conflict when two langauges are very similar, as is the case for French and Ita lian, is more attentionally deman d ing for a bilingual than it would be if his/her langauges were very different, e.g., if s/he spoke French and Mandarin. If that is the case, then the two languages that a bilingual speaks would be important in determining whether or not a bilingual advantage emerged and if so, the degree of that advantage. Additionally, while language proficiency is a continuous variable, monolinguals and bilinguals were dichotomized for the purpose of the present study as if bilingualism i s an 'all or none' quality , when that is not at all the case . While it would be ideal to study attentional control at all levels of bilingual proficiency, this approach was not practical in the present study since doing so would potentially require hundred s of participants or more . Having included only participants at the extremes of the proficiency spectrum for the monolingual (low end) and bilingual (high end) groups makes it impossible to address how people with intermediate levels of bilingual proficien cy might be affected by using two languages. Given that, it is impossible to tell at what point along the continuum of second language proficiency bilingual advantages begin to emerge. Still another limitation relates to the fact that the bilingual particip ants' language switching habits were not controlled, something that should be included as a covariate in future research . This is important because research has demonstrated that language switching may be a

PAGE 93

93 key factor in determining whether or not bilingua ls have advantages relative to monolinguals (Prior & Gollan, 2011) . Determining whether or not bilinguals are better able than monolinguals to control distraction from emotional stimuli is important for research on bilingualism, emotion, and attention. T he present study support s the idea that bilinguals control attention to emotional stimuli better than monolinguals , at least under some circumstances . This suggests that being fluently bilingual may have a broader and more far reaching impact on executive function than pr eviously thought and opens up a new world of research questions. For example, does the bilingual advantage in controlling attention in the presence of emotional distraction stem strictly from attentional control, or does it arise due to bil inguals having better ability to control their responses to the emotional stimuli more broadly, i.e., do bilinguals have better emotional control than monolinguals? If so, do bilingual children develop emotional control earlier and/or better than monolingu al children? The present study has provided the first step in understanding bilinguals ' and monolinguals' ability to control attention in the presence of emotional stimuli. However, additional research is needed to determine whether a bilingual advantage o ccurs under more general conditions or whether it is limited to contexts in which the emotional stimuli are irrelevant and the cognitive demands on attention are high.

PAGE 94

94 Table 6 1. Summary of results Task Flanker Digit Parity Distraction Effect RTs Acc uracy RTs Accuracy Experiment 1 Cognitive Distractor N < C < I C > N > I C < N < I < A Group B < M B > M 2 Way Interaction Experiment 2 Emotional Valence NeuPos=Neg Group 2 Way I nteraction B: Neu>Pos=Neg M: Neu=Pos=Neg Experiment 3 Cognitive & Emotional Distractor N < C < I C=N > I C=N < I < A A > N=C > I Valence Neu Neg Group Distractor x Valence I: Neu> Pos=Neg C : Neu, & A: Neu=Pos=Neg 3 Way Interaction M: Distractor x Valence B: No Distractor x Valence Conflict Effects B < M for Pos and Neg B = M for Neu Carryover Effects Exp. 2 Prev. Trial Valence Neg>Neu=Pos Neg=Pos>Neu Group Exp. 3 Prev. Trial Valence Neg=Pos
PAGE 95

95 APPENDIX SAMPLE SCRIPT TO ENGAGE BILINGUALS IN CONVERSATION IN SPANISH For reader clarity, the script is shown in English but was actually administered in Spanish. Questions asked by the experimenter are shown in ita lics. 5. Give the participant time to respond in Spanish. 6. If they do not do so, say in Spanish: . 7. Continue with: f this experiment As they are getting settled, say: 8. experiment 9. Next, ask one of the following questions. If the participant answers with a one word response, elaborate on the question with follow up questions to elicit a more complex response (sug gested questions are indicated below). If that fails to elicit a multi word response, ask another one of the questions below. Are you excited to watch the Gator football/basketball game this weekend? 10. 11. 12. "What year are you?" 13. Follow up wi 14. "What are you studying?"

PAGE 96

96 15. with your degree "Did you have any trouble finding your way here/the room?" 16. 17. Once the participant has demonstrated that s/he can converse fluently in Spani sh, continue by saying in Spanish, informed consent. Please read through the informed consent. If you want to participant, please print your name, sign it, and date the paper. This is the only paper with your name on it. It will Then say: some questions to ask you Then administer the demographic questionnaire nd ask questions that require the participant to elaborate his/her responses. For example: if the participant says s/he was born in Colombia, Say Or I do/see if I ever get a chance

PAGE 97

97 LIST OF REFERENCES Abutalebi, J., Annoni, J. M., Zimine, I., Pegna, A. J., Seghier, M. L., Lee Jahnke, H., Lazeyras,F., Cappa, S. F., & Khateb, A. (2007). Language control and lexical competition in bilingua ls an event related fMRI study. Cerebral Cortex , 18 , 1496 -1505. doi: 10.1093/cercor/bhm182 Abutalebi, J., & Green, D. (2007). Bilingual Language production: The neurocognition of language representation and control. Journal of Neurolinguistics , 20 , 242 27 5. doi: 10.1016/j.jneuroling.2006.10.003 Abutalebi, J., Rosa, P. A. D., Green, D., Hernandez, M., Scifo1, P., Keim,R., Cappa, S. F., & Costa, C. (2012). Bilingualism tunes the anterior cingulate cortex for conflict monitoring. Cerebral Cortex , 22 , 2076 208 6. doi:10.1093/cercor/bhr287 Algom, D., Chajut, E., & Lev, S. (2004). A rational look at the emotional Stroop phenomenon : a generic slowdown, not a Stroop effect. Journal of Experimental Psychology: General , 3 , 323 338. doi: 10.1037/0096 3445.133.3.323 All man, J. M., Hakeem, A., Erwin, J. M., Nimchinsky, E., & Hof, P. (2001). The anterior cingulate cortex. The evolution of an interface between emotion and cognition. Annals of the New York Academy of Sciences , 935 , 107 117. doi: 10.1111/j.1749 6632.2001.tb03476.x Altarriba, J. & Bainsight D. M. (2010). The representation of emotion vs. emotion laden words in English and Spanish in the Affective Simon Task. International Journal of Bilingual ism , 15 , 310 328. doi: 10.1177/1367006910379261 Anderson, A. K. (2005). Affective influences on the attentional dynamics supporting awareness. Journal of Experimental Psychology: General , 134 , 258 281. doi:10.1037/0096 3445.134.2.258258 Aquino, J. M. & Ar nell, A. M. (2007). Attention and the processing of emotional words Dissociating the effects of arousal. Psychonomic Bulletin & Review , 14 , 430 435. doi:10.3758/BF03194084 Arnell, K. M., Killman, K. D., & Fijavz, D. (2007). Blinded by emotion: target misse s follow attention capture by arousing distractors in RSVP. Emotion , 7 , 465 77. doi:10.1037/1528 3542.7.3.465 Ashley, V., & Swick, D. (2009). Consequences of emotional stimuli: age differences on pure and mixed b locks of the emotional Stroop. Behavioral an d Brain Functions , 5 , 1 11. doi:10.1186/1744 9081 5 14 Barch, D. M., Braver, T. S., Sabb, F. W., & Noll, D. C. (2000). Anterior cingulate and the monitoring of response conflict: Evidence from an fMRI study of overt verb generation. Journal of Cognitive Ne uroscience , 12 , 298 309. doi: doi:10.1162/089892900562110

PAGE 98

98 Berch, D. B., Foley, E. J., Hill, R. J., & Ryan, P. M. (1999). Extracting parity and magnitude from Arabic numerals: Developmental changes in number processing and mental representation. Journal of Experimental Child Psychology , 74 , 286 308. doi:10.1006/jecp.1999.2518 Bertels, J., Kolinsky, R., Pietrons, E., & Morais, J. (2011). Long lasting attentional influence of negative and taboo words in an auditory variant of the emotional Stroop task. Emotion , 11 , 27 37. doi:10.1037/a0022017 Bialystok, E., Craik, F. I. M., Klein, R., & Viswanathan, M. (2004). Bilingualism, aging, and cognitive control: evidence from the Simon task. Psychology and Aging , 19, 290 303. doi:10.1037/0882 7974.19.2.290 Bialystok, E. , Craik, F. I. M., & Luk, G. (2008). Cognitive control and lexical access in younger and older bilinguals. Journal of Experimental Psychology: Learning, Memory, and Cognition, 34 , 859 873. doi:10.1037/0278 7393.34.4.859 Bialystok, E., & Viswanathan, M. (20 09). Components of executive control with advantages for bilingual children in two cultures. Cognition , 112 (3), 494 500. doi:10.1016/j.cognition.2009.06.014 Botvinick, M. M., Braver, T. S., Barch, D. M., Carter, C. S., & Cohen, J. D. ( 1999 ). Conflict monit oring and cognitive control. Psychological review , 108 , 624. doi: 10.1037//0033 295X.I08.3.624 Botvinick, M. M., Cohen, J. D., & Carter, C. S. (2004). Conflict monitoring and anterior cingulate cortex: an update. Trends in cognitive sciences , 8 , 539 546. Botvinick, M., Nystrom, L. E., Fissell, K., Carter, C. S., & Cohen, J. D. (1999). Conflict monitoring versus selection for act ion in anterior cingulate cortex. Nature , 402, 179 181. doi: 10.1038/46035 Bradley, M. M., Greenwald, M. K., Petry, M. C., & Lang, P. J. (1992). Remembering pictures: Pleasure and arousal in memory. Journal of Experimental Psychology: Learning, Memory, and Cognition , 18 , 379 390. doi:10.1037//0278 7393.18.2.379 Bradley, M. M., & Lang, P. J. (1999). Affective norms for English words ( ANEW): Instruction manual and affective ratings . Technical Report C 1, The Center for Research in Psychophysiology, University of Florida. Buchanan, T. W., Etzel, J. A., Adolphs, R., & Tranel, D. (2006). The influence of autonomic arousal and semantic rel atedness on memory for emotional words. International Journal of Psychophysiology , 61 , 26 33. doi:10.1016/j.ijpsycho.2005.10.022 Carlson, S. M., & Meltzoff, A. N. (2008). Bilingual experience and executive functioning in young children. Developmental Sci ence, 11, 282 298. doi:10.1111/j.1467 7. 687.2008.00675.x

PAGE 99

99 Colbeck, K. L., & Bowers, J. S. (2012). Blinded by taboo words in L1 but not L2. Emotion , 12 , 217 222. doi10.1037/a0026387 Costa, A., Hernández, M., Costa Faidella, J., & Sebastián Gallés, N. (2009 ). On the bilingual advantage in conflict processing : Now you see it, now you don't. Cognition , 113 , 135 149. doi:10.1016/j.cognition.2009.08.001 Costa, A., Hernández, M., & Sebastián Gallés, N. (2008). Bilingualism aids conflict resolution: Evidence from the ANT task. Cognition, 106 , 59 86. doi:10.1016/j.cognition.2006.12.013 Crottaz Herbette, S., & Menon, V. (2006). Where and when the anterior cingulate cortex modulates attentional response : combined fMRI and ERP evidence. Journal of Cognitive Neuroscien ce , 18 , 766 780. doi:10.1162/jocn.2006.18.5.766 Cuthbert, B. N., Schupp, H. T., Bradley, M. M., Birbaumer, N., & Lang, P. J. (2000). Brain potentials in affective picture processing: covariation with autonomic arousal and affective report. Biological psych ology , 52 , 95 111. doi: 10.1016/S0301 0511(99)00044 7 Davis, K. D., Hutchinson, W. D., Lozano, A. M. Tasker, R. R., & Dostrovsky, J. O. (2000). Human Anterior Cingulate cortex neurons modulated by attention demanding tasks. Journal of Neurophysiology , 83 , 3575 3577. doi:10.1162/jocn.2006.18.5.766 Deheane, S., Bossini, S., & Giraux, P. (1993). The mental representation of parity and number magnitude. Journal of Experimental Psychology: General , 122 , 371 396. doi:10.1037/0096 3445.122.3.371 Dhooge, E., & Hart suiker, R. J. (2011). How do speakers resist distraction Evidence from a taboo picture word interference task. Psychological Science , 22 , 855 859. doi10.1177/0956797611410984 Dolcos, F., & McCarthy, G. (2006). Brain systems mediating cognitive interference by emotional distraction. The Journal of Neuroscience , 26 , 2072 2079. dio:10.1523/JNEUROSCI.5042 05.2006 Dye, M. W. G., Green, C. S., & Bavelier, D. (2009). The development of attention skills in action video game players. Neuropsychologia , 47 , 1780 1789. doi: 10.1016/j.neuropsychologia.2009.02.002 Egner, T., Etkin, A., Gale, S., & Hirsch, J. (2008). Dissociable neural systems resolve conflict from emotional versus nonemotional distracters. Cerebral Cortex , 18 , 1475 1484. doi: 10.1093/cercor/bhm179 Emmorey, K., Borinstein, H. B., Thompson, R., & Gollan, T. H. (2008). Bimodal bilingualism. Bilingualism: Language and cognition , 11 , 43 61. doi: 10.1017/S1366728907003203

PAGE 100

100 Eriksen, B. A., & Eriksen, C. W. (1974). Effects of noise letters upon the identification of a target letter in a nonsearch task. Perception & psychophysics , 16 (1), 143 149. doi: doi.org/10.3758/BF03203267 Fan, J., McCandliss, B.D., Sommer, T., Raz, A., & Posner, M.I. (2002). Testing the efficiency and independence of attentional networks. Journ al of Cognitive Neuroscience , 14 , 340 347. doi:10.1162/089892902317361886 Fernandes, M. A., Koji, S., Dixon, M. J., & Aquino, J. M. (2011). Changing the focus of attention: the interacting effect of valence and arousal. Visual Cognition , 19 , 1191 1211. doi :10.1080/13506285.2011.618151 Ferré, P., García, T., Fraga, I., Sánchez Casas, R., & Morelo, M. (2010). Memory for emotional words in bilinguals: Do words have the same emotional intensity in the first and in the second language? Cognition and Emotion , 24 , 760 785. doi:10.1080/02699930902985779 Fias, W., Reynvoet, B., & Brysbaert, M. ( 20 01 ). Are Arabic numerals processed as pictures in a Stroop interference task? Psychological Research , 65 , 242 249. doi:10.1007/s004260100064 Gollan, T. H., Fennema Notestin e, C., Montoya, R. I., & Jernigan, T. L., (2007). The bilingual effect on Boston Naming Test performance. Journal of the International Neuropsychological Society, 13 , 197 208. doi:10.1017/S1355617707070038 Gollan, T. H., Montoya, R. I., Cera, C., & Sandova l, T. C. (2008). More use almost always means a smaller frequency effect: Aging, bilingualism, and the weaker links hypothesis. Journal of Memory and Language , 58 , 787 814. doi:10.1016/j.jml.2007.07.001 Gollan, T. H., Montoya, R. I., Fennema Notestine, C., & Morris, S. K. (2005). Bilingualism affects picture naming but not picture classification. Memory & Cognition , 33 (7), 1220 1234. doi: 10.3758/BF03193224 Gollan, T. H., Montoya, R. I., & Werner, G. A. (2002). Semantic and letter fluency in Spanish English bilinguals. Neuropsychology , 16 , 562 576. doi : 10.1037//0894 4105.16.4.562 Hajcak, G., & Olvet, D. M. (2008). The persistence of attention to emotion: brain potentials during and after picture presentation. Emotion, 8 , 250 255. doi: 10.1037/1528 3542.8.2.2 50 Hamann, S. B., Ely, T. D., Grafton, S. T., & Kilts, C. D. (1999). Amygdala activity related to enhanced memory for pleasant and aversive stimuli. N ature Neuroscience , 2 , 289 294. doi: doi.org/10.1038/23419 Hilchey, M. D., & Klein, R. M. (2011). Are ther e bilingual advantages on nonlinguistic interference tasks? Implications for the plasticity of executive control processes. Psychonomic Bulletin and Review, 18, 625 658. doi:10.3758/s13423 011 0116 7

PAGE 101

101 Ivanova, I., & Costa A. (200 8 ). Does bilingualism hampe r lexical access in speech production? Acta Psychologica, 127 , 277 288. doi:10.1016/j.actpsy.2007.06.003 Jay, T., Caldwell Harris, C., & King, K. (2008). Recalling taboo and nontaboo words. American Journal of Psychology , 121 , 83 103. doi:10.2307/20445445 Keil, A., & Ihssen, N. (2004). Identification facilitation for emotionally arousing verbs during the attentional blink. Emotion, 4 , 23 35. doi:10.1037/1528 3542.4.1.23 Kousta, S. T., Vinson, D. P., & Vigliocco, G. (2009) Emotion words, regardless of polar ity, have a processing advantage over neutral words . Cognition , 112 , 473 481. doi:10.1016/j.cognition.2009.06.007 Lane, R. D., Chua, P. M., & Dolan, R. J. (1999). Common effects of emotional valence, arousal and attention on neural activation during visual processing of pictures. Neuropsychologia , 37 (9), 989 997. doi: 10.1016/S0028 3932(99)00017 2 Lang, P. J. (1995). The emotion probe: Studies of motivation and attention. American Psychologist , 50 , 372 385. doi:10.1037//0003 066X.50.5.372 Lang, P. J., Brad ley, M. M., & Cuthbert, B. N. (2008). International Affective Picture System (IAPS): Technical Manual and Affective Ratings. Technical Report A 8 . University of Florida, Gainesville, FL. Linck, J. A., Hoshino, N., & Kroll, J. F. (2008). Cross language lexi cal processes and inhibitory control. The Mental Lexicon, 3, 349 374 . doi : 10.1075/ml.3.3.06lin Luk, G., De Sa, E., & Bialystok, E. (2011). Is there a relation between onset age of bilingualism and enhancement of cognitive control? Bilingualism: Language an d Cognition, 14, 588 595. doi:10.1017/S1366728911000010 MacKay, D. G., Shafto, M., Taylor, J. K., Marian, D. E., Abrams, L., & Dyer, J. R. (2004). Relations between emotion, memory, and attention: evidence from taboo Stroop, lexical decision, and immediate memory tasks. Memory and Cognition , 32 , 474 488. doi:10.3758/BF03195840 Marian, V., Blumenfeld, H. K., & Kaushanskaya, M. (2007). The language experience and proficiency questionnaire: Assessing Language Profiles in Bilinguals and Multilinguals . Journal o f Speech, Language, and Hearing Research , 50 , 940 96. doi:10.1044/1092 4388(2007/067) Martin Rhee, M. M., & Bialystok, E. (2008). The development of two types of inhibitory control in monolingual and bilingual children. Bilingualism: Language and Cognition , 11, 81 93. doi:10.1017/S1366728907003227 Mather, M., Mitchell, K. J., Raye, C. I., Novak, D. L., Greene, E. J., & Johnson, M. K. (2006). Emotional arousal can impair feature binding in working memory. Journal of Cognitive Neuroscience , 18 , 614 625. doi: 10.1162/jocn.2006.18.4.614

PAGE 102

102 Nieuwenhuis, S., Stins, J. F., Posthuma, D., Polderman, T. J., Boomsma, D. I., & de Geus, E. J. (2006). Accounting for sequential trial effects in the flanker task: Conflict adaptation or associative priming?. Memory & cognition , 34 (6), 1260 1272. doi: 10.3758/BF03193270 Palo m b a, D., Agrilli, A., & Mini, A. (1997). Visual evoked potentials, heart rate responses and memory to emotional pictorial stimuli. International Journal of Psychophysiology , 27 , 55 67. doi: 10.1016/S0167 8760 (97)00751 4 Pelham, S. D., & Abrams, L. (2013). Cognitive advantages and disadvantages in early and late bilinguals. doi: 10.1037/a0035224 Phelps, E. A., Ling, S., & Carrasco, M. (2006). Emotion facilitates perception and potentiates the perceptual benefi ts of attention. Psychological science , 17 , 292 299. doi: 10.1111/j.1467 9280.2006.01701.x Portocarrero, J. S., Burright, R. G., & Donovick, P. J. (2007). Vocabulary and verbal fluency of bilingual and monolingual college students. Archives of Clinical Neu ropsychology , 22 , 415 422. doi:10.1016/j.acn.2007.01.015 Price, C., Green, D. W., & von Studnitz, R. (1999). A functional imaging study of translation and language switching. Brain , 122 , 2221 2235. doi:10.1093/brain/122.12.2221 Prior, A., & Gollan, T. H. ( 2011). Good language switchers are good task switchers: Evidence from Spanish English and Mandarin English bilinguals. Journal of the International Neuropsychological Society , 17 , 682 691. doi:10.1017/S1355617711000580 Rodriguez Fornells, A., van der Lugt, A. Rotte, M. Britti, B., Heinze, H., & Munte, T. F. (2005). Second Language Interferes with Word Production in Fluent Bilinguals: Brain Potential and Functional Imaging Evidence . Journal of Cognitive Neuroscience, 17, 422 433. doi:10.1162/0898929053279559 Schimmack, U. (2005). Attentional interference effects of emotional pictures: threat, negativity, or arousal. Emotion , 5 , 55 66. doi: 10.1037/1528 3542.5.1.55 Simon, J. R., & Rudell, A. P. (1967). Auditory S R compatibility: The effect of an irrelevant c ue on information processing. Journal of Applied Psychology , 51 , 300 304. doi:10.1037/h0020586 Strobach, T., Frensch, P. A., & Schubert, T. (2012). Video game practice optimizes executive control skills in dual task and task switching situations. Acta Psyc hologica , 140 , 13 24. doi: 10.1016/j.actpsy.2012.02.001 Stroop, J. R. (1935). Studies of interference in serial verbal reactions. Journal of Experimental Psychology , 18 , 643 662. doi:10.1037//0096 3445.121.1.15 Sutton, T. A., Altarriba, J., Gianico, J. L., & Bainsight Brown, D.M. (2007). The automatic access of emotion: Emotional Stroop effects in Spanish English bilingual speakers. Cognition and Emotion, 21 , 1077 1090. doi: 10.1080/02699930601054133

PAGE 103

103 Tao, L., Marzecová, A., Taft, T., Asanowicz, D., Wodnieck a, Z. (2011). The efficiency of attentional networks in early and late bilinguals: the role of age of acquisition. Frontiers in Psychology , 2 . doi: 10.3389/fpsyg.2011.00123 Thomas, R. C . & Hasher, L. . (2006). The influence of emotional valence on age diffe rences in early processing and memory. Psychology and Aging , 21 , 821 825. doi:10.1037/0882 7974.21.4.821 doi : 10.1037/0882 7974.21.4.821 Van Heuven, W. J., Schriefers, H., Dijkstra, T., & Hagoort, P. (2008). Language conflict in the bilingual brain. Cerebr al Cortex , 18 , 2706 2716. doi: 10.1093/cercor/bhn030 van Veen, V., Cohen, J. D., Botvinick, M. M., Stenger, V. A., & Carter, C. S. (2001). Anterior cingulate cortex, conflict monitoring, and levels of processing. Neuroimage , 14 , 1302 1308. doi:10.1006/nimg .2001.0923 Yamanski, H., LaBar, K. S., & McCarthy, D. (2002). Dissociable prefrontal brain systems for attention and emotion. Proceedings of the National Academy of Sciences, 99 , 11447 11451. doi: 10.1073/iti3312109 Zied, K.M., Phillipe, A., Karine, P., Val erie, H T., Ghislaine, A.,Arnaud, R., & Didier, L.G. (2004). Bilingualism and adult differences in inhibitory mechani sms: Evidence from a bilingual S troop task. Brain and Cognition , 54 , 254 256. doi: 10.1016/j.bandc.2004.02.036

PAGE 104

104 BIOGRAPHICAL SKETCH Sabra Pel ham was born in New Orleans, Louisiana but grew up in Lawrence, Kansas where she attended the University of Kansas . She s in ling uistics and anthropology and a Master of Arts degree in applied linguistics from KU, and s hortly afte r completing the course work for her she relocated to Florida and began teaching English as a second language at Daytona State College (D S C) . She received tenure in her third year at D S C and was promoted from the status of faculty to assis tant and then associate professor . During her time at D S C, she pursued independent research in child language acquisition and began taking courses in psychology with the view to returning to graduate school to pursue a Ph.D. in psychology. After eight year s teaching at D S C, she returned to graduate scho ol at the University of Florida where she received a Master of Science degree in cognitive psychology in the spring of 2012 and her doctorate in cognitive psychology in the summer of 2014 . In the fall of 2014 , Sabra took a position as an Assistant Professor of P sychology at Spokane Falls Community College in Spokane, Washington.