The Assessment of Code Switching Experience Survey

A STUDY OF THE RELATIONSHIP BETWEEN CODE SWITCHING AND THE BILINGUAL ADVANTAGE: EVIDENCE THAT LANGUAGE USE MODULATES NEURAL INDICES OF LANGUAGE PROCESSING AND COGNITIVE CONTROL

APPROVED BY SUPERVISING COMMITTEE: ________________________________________ Nicole Wicha, Ph.D., Chair ________________________________________ Todd Troyer, Ph.D. ________________________________________ Charles Wilson, Ph.D. ________________________________________ Brian Derrick, Ph.D. ________________________________________ Kara Federmeier, Ph.D. Accepted: _________________________________________ Dean, Graduate School

Copyright 2013 Angélique Michelle Blackburn All Rights Reserved

DEDICATION This dissertation is dedicated to my Bans Finden, who carried me emotionally through this process (and carried me physically out of the Grand Canyon in a blizzard so I could get back to work alive), and to my Dad, who was nice enough to pass along his math skills and provided a comfortable childhood in which I was allowed to read and “play science” rather than doing my chores.


by ANGÉLIQUE MICHELLE BLACKBURN, M.S.

DISSERTATION Presented to the Graduate Faculty of The University of Texas at San Antonio in Partial Fulfillment of the Requirements for the Degree of DOCTOR OF PHILOSOPHY IN BIOLOGY WITH A CONCENTRATION IN NEUROBIOLOGY

THE UNIVERSITY OF TEXAS AT SAN ANTONIO College of Sciences Department of Biology December 2013

UMI Number: 3607533

All rights reserved INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if material had to be removed, a note will indicate the deletion.

UMI 3607533 Published by ProQuest LLC (2014). Copyright in the Dissertation held by the Author. Microform Edition © ProQuest LLC. All rights reserved. This work is protected against unauthorized copying under Title 17, United States Code

ProQuest LLC. 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, MI 48106 - 1346

ACKNOWLEDGEMENTS “Nos esse quasi nanos, gigantium humeris insidentes... sed quia in altum subvenimur et extollimur magnitudine gigantea.” It is here that I have the opportunity to thank those giants on whose shoulders I currently stand. This work would not have been possible without the pioneering work on lexical co-activation by Judith Kroll and Susan Bobb. I am personally grateful to these women for igniting my passion for bilingual research and setting me on this amazing career path. I must also thank my advisor, Nicole Wicha, for allowing me the freedom to develop and complete these experiments. Having the resources and flexibility to pursue my research interests as they evolved is a rare and coveted circumstance for a graduate student and I am very grateful for that opportunity. I also thank my committee – each time I stood before them, I knew I was standing before greatness. I am extremely grateful for their guidance and for the guidance of my lab-mates, Shukhan Ng and Amanda Martinez-Lincoln, who answered my million questions and entertained my need to discuss every possible research design flaw. I also thank my benefactors, Marty and Roni Gillespie, for their advice and assistance during my first year. For emotional support, I thank Lucie Burg, Charlotte, Buddy, Bert and Gabrielle Blackburn – thanks for accompanying me on the midnight walks back to my car as I tried to scare away the wild hogs, coyotes, and bobcats! Finally, I wish to thank Dr. Owens, my mentor who laid the foundation for a successful graduate experience. This work would not have been possible without the 2446 participants to whom I will be eternally grateful. I would thank you all by name, but the Institutional Review Board protects your anonymity.

iv

This work was supported in part by the Specialized Neuroscience Research Program Fellowship and NICHD/NIGMH HD060435 and received computational support from Computational System Biology Core, funded by the National Institute on Minority Health and Health Disparities (G12MD007591) from the National Institutes of Health. Statistical support was provided by the Statistics Consulting Center at UTSA. I would also like to thank Christopher Hwozdek, Viridiana Estrada, Svetlana Anthony, Stephanie Paez, Crystal Escobales, and Estefania Ciscernos for help in screening thousands of research participants and for technical assistance. This Doctoral Dissertation was produced in accordance with guidelines which permit the inclusion as part of the Doctoral Dissertation the text of an original paper, or papers, submitted for publication. The Doctoral Dissertation must still conform to all other requirements explained in the “Guide for the Preparation of a Master’s Thesis/Recital Document or Doctoral Dissertation at The University of Texas at San Antonio.” It must include a comprehensive abstract, a full introduction and literature review, and a final overall conclusion. Additional material (procedural and design data as well as descriptions of equipment) must be provided in sufficient detail to allow a clear and precise judgment to be made of the importance and originality of the research reported. It is acceptable for this Doctoral Dissertation to include as chapters authentic copies of papers already published, provided these meet type size, margin, and legibility requirements. In such cases, connecting texts, which provide logical bridges between different manuscripts, are mandatory. Where the student is not the sole author of a manuscript, the student is required to make an explicit statement in the introductory material to that manuscript describing the student’s contribution to the work and acknowledging the contribution of the other author(s). The signatures of the Supervising Committee which precede all other material in the Doctoral Dissertation attest to the accuracy of this statement.

December 2013 v


Angélique Michelle Blackburn, Ph.D. The University of Texas at San Antonio, 2013

Supervising Professor: Nicole Wicha, Ph.D. Bilinguals sometimes outperform age-matched monolinguals on non-language tasks involving cognitive control. But the bilingual advantage is not consistently found in every experiment and may reflect specific attributes of the bilinguals tested. The goal of this dissertation was to determine if the way in which bilinguals use language, specifically switching between languages within a conversation (code switching) or refraining from this behavior, plays a role in the sporadic bilingual advantage. The bilingual advantage may arise from managing interference from one language in order to stay in the other, or from inhibiting one language in order to switch into the other. If language switching engages a general inhibitory mechanism to stop speaking in one language and switch to the other, bilinguals who frequently code switch (“switchers”) might outperform bilingual “non-switchers” on non-language inhibition tasks. Alternatively, if the bilingual advantage results from frequent inhibition of interference from one language to stay in the other, non-switchers might outperform switchers. The Assessment of Code-Switching Experience Survey (ACSES) was created to obtain an objective, rapid, and reliable measure for Spanish-English bilinguals' code-switching vi

frequency. Then, event-related potentials (ERPs) of bilinguals with varied code-switching frequency were measured during a language task, lending validity to the survey. Semantic violations in code-switched and non-switched words were embedded in sentences as probes to determine how code-switches are processed and whether processing of code-switches is modulated by code-switching experience. The amplitude of a code switching positivity (typically elicited to a code-switch), but not the amplitude of the N400 (typically elicited to a semantic anomaly), was modulated by code-switching frequency. This indicates that codeswitching experience affects processing of a code-switch but not semantic retrieval. During Simon and Flanker inhibition tasks, which require suppression of interference from incongruent stimulus cues, a larger N2 ERP was elicited for incongruent versus congruent trials. The amplitude of this N2 congruity effect is typically correlated with inhibitory control ability. The effect was carried by the non-switchers; switchers and monolinguals did not differ. Thus, a general bilingual advantage was not found. Rather, non-switchers manifested a larger neural response linked to inhibitory control of interference, which may result from inhibiting interference in one language in order to stay in the other. However, no difference between switchers, non-switchers, and monolinguals was found on inhibition tasks that required simple response inhibition in the absence of interference. Thus, language production may differentially tap into existing cognitive control mechanisms that are specific to the way in which we are using language. I have interpreted the non-switcher interference suppression advantage in terms of recent studies demonstrating switching benefits in switchers and spatial benefits in bilinguals who use sign language. I suggest refining the definition of the bilingual advantage to include multiple aspects of cognition that are differentially affected by language experience.

vii

TABLE OF CONTENTS Acknowledgements ................................................................................................................... iv Abstract ..................................................................................................................................... vi List of Tables ........................................................................................................................... ix List of Figures ........................................................................................................................... xi Chapter One: Introduction .......................................................................................................... 1 Chapter Two: The Assessment of Code-Switching Experience Survey ......................................12 Chapter Three: The Effect of Code-Switching Experience on the Neural Response Elicited to a Sentential Code Switch ..................................................................................................56 Chapter Four: Effects of Code Switching on Non-Language Inhibitory Control Tasks ...............91 Chapter Five: Conclusions - Redefining the Bilingual Advantage(s) ........................................ 128 Appendices...................................................................................................................................143 References ............................................................................................................................... 202 Vita

viii

LIST OF TABLES CHAPTER TWO Table 1

Study 1, Mean language proficiency self-ratings (standard deviations in parentheses) ....................................................................................................23

Table 2

Study 1, Components Yielded on Short Version of Survey. Component loadings greater than 0.4 are displayed ..........................................................................25

Table 3

Table 3. Study 1, Component Correlation Matrix.............................................28

Table 4

Test/retest correlations of component scores on Short Version. Correlations greater than 0.7 are acceptable. ........................................................................29

Table 5

Percentage of individuals who consider themselves to be code-switchers in each of the scoring categories for Component 1. ......................................................30

Table 6

Study 2, Mean language proficiency self-ratings (standard deviations in parentheses).....................................................................................................35

Table 7

Study 2, Final Factor Solution. Component loadings greater than 0.3 are displayed .........................................................................................................37

Table 8

Study 3, Demographics Analysis for Spanish/English (S/E) bilinguals. Variance explained by components obtained for each S/E bilingual demographic... ........45

Table 9

Study 3, Demographics Analysis for Spanish/English (S/E) Bilinguals. Reliability measured as Cronbach's alpha for components in each S/E bilingual demographic. ...................................................................................................45

Table 10

Study 3, Mean (and Standard Deviation) and Cohen's d Effect Sizes (d) of Each Component Score by Spanish/English Bilingual Demographic Group. High scores for a given component are bolded. ........................................................46 ix

CHAPTER FOUR Table 1

Mean (SD) scores for switchers and non-switchers on each component of ACSES on a Likert Scale (1=Never, 7=Always) ............................................ 101

Table 2

Mean values of language and non-language factors used to match groups...... 102

x

LIST OF FIGURES CHAPTER TWO Figure 1

Steps towards the creation of ACSES. .............................................................19

Figure 2

Percentage of Participants in each Frequency of Code-Switching Category (n = 408). ................................................................................................................31

Figure 3

Percentage of Participants in each Frequency of Code-Switching Category (n = 730).................................................................................................................40

Figure 4

Percentage of Texans (n = 585) and Non-Texans (n = 142) in each Frequency of Code-Switching Category................................................................................48

CHAPTER THREE Figure 1

ERP waveforms. Raw waveforms are presented for each condition, time-locked to the onset of each stimulus. ...........................................................................66

Figure 2

Differences waves of the code switch and semantic effects. Difference waves for the code switch effect without a semantic violation represent a point-bypoint subtraction of the grand averaged response to a code switch (CS) minus the control; difference waves for the code switch effect with a semantic violation represent a point-by-point subtraction of the grand averaged response to a code switched semantic violation (CS and SemV) minus the semantic violation alone (SemV). Difference waves for the semantic effect without a code switch are calculated as the response to a semantic violations (SemV) minus the control; the semantic effect with a code switch is calculated as the response to a code switched semantic violation (CS and SemV) minus the response elicited to code switches alone (CS). These waves indicate that codexi

switches elicit a positivity with a latency onset around 320 ms. This onset overlaps with the first half of the LPC (LPCa) elicited to a semantic violation, but not the LPCb Figure 3

……………………………………………………………70

Electrophysiological correlates of code switching. A code-switching positivity can be observed from 320-650ms post stimulus onset. A negativity is also observed over left anterior lateral sites (*p < 0.05) ..........................................71

Figure 4

Interactions of code-switching and semantic congruity ....................................74

Figure 5

Correlations of frequency of code-switching with the amplitude semantic congruity effect on the N400 (left panel) and the amplitude of the codeswitching positivity (right panel) .....................................................................78

CHAPTER FOUR Figure 1

Inhibitory Control Tasks. ............................................................................... 104

Figure 2

Mean response times (a) and percent accuracy (b) on the Flanker task (n=24/group) ................................................................................................. 108

Figure 3

Mean response times (a) and percent accuracy (b) on the Simon task (n=24/group) ................................................................................................. 108

Figure 4

N2 Congruity Effect. Raw waveforms of monolingual (a), switcher (b), and non-switcher (c) responses to congruent and incongruent trials. Non-switchers have a significantly larger N2 congruity effect, as shown in the difference waves, subtraction of congruent waveforms from incongruent waveforms for each group (d). Top Panel: Congruity effects for each group during the Flanker task. Bottom Panel: Congruity effects for each group during the Simon task…… ........................................................................................................ 111 xii

Figure 5

N2 Nogo Effect. Raw waveforms of monolingual (a), switcher (b), and nonswitcher (c) responses to go and nogo trials. No interaction of Group by N2 Nogo Effect was found, as shown in the difference waves, subtraction of go waveforms from nogo waveforms for each group .......................................... 112

xiii

CHAPTER ONE: INTRODUCTION Like many other environmental factors, living in a bilingual environment can have longterm effects on cognitive ability. In a landmark study, Bialystok and colleagues (2004) showed that bilinguals out-perform monolinguals on nonlinguistic tasks that require cognitive control, i.e. ”the ability to flexibly adapt behavior to current demands…while dealing with interference and competition” (Festman, Rodriguez-Fornells, & Münte, 2010, p.1). When presented with stimuli that evoke two competing responses, bilinguals were faster than monolinguals to manage the conflict according to the task demands and inhibit the incorrect response. This “bilingual advantage” over monolinguals is thought to reflect increased cognitive control related to the ability to inhibit interference or competing responses (i.e. inhibitory control). Because the bilingual advantage, which is largest in older adults, may function as a compensatory strategy to offset the signs of dementia and neurodegenerative disease, it is important to understand its underlying cause (Bialystok, Craik, & Freedman, 2007; Craik, Bialystok, & Freedman, 2010; Schweizer, Ware, Fischer, Craik, & Bialystok, 2012). The bilingual advantage in inhibitory control has substantiated Green’s Inhibitory Control model1 (Bialystok, et al., 2004). According to the model, a bilingual’s two languages are both active, meaning that either language can be accessed and spoken at a given moment (Kroll, Bobb, & Wodniekca, 2006). While one language is in use, words in the other language interfere and compete for selection, and are thought to be actively inhibited (Costa, 2005; Kroll, Bobb, Misra, & Guo, 2008; Kroll, et al., 2006; A. Rodriguez-Fornells et al., 2005; Spivey & Marian, 1999). For instance, when a bilingual says “table”, the Spanish translation “mesa” competes for

1

Although recent evidence suggests that bilinguals may be advantaged in general executive functioning which may include inhibition, attention, and conflict monitoring, we have focused specifically on inhibitory control tasks that involve conflict. Distinctions between conflict and inhibition and implications for Green’s Model are discussed in Chapter 4.

1

selection and “mesa” must be inhibited so “table” can be produced.2 Bilinguals may develop increased inhibitory control because they must constantly manage interference from one language in order to stay in the other. Green’s Inhibitory Control model yields the prediction that bilinguals who are better at managing interference in order to refrain from switching between languages will exhibit enhanced inhibitory control. However the bilingual advantage is not consistently found in every experiment and may reflect specific properties of the bilinguals tested (Hilchey & Klein, 2011), such as age (Bialystok, Craik, & Ruocco, 2006; Bialystok, Craik, & Ryan, 2006; Bialystok, Martin, & Viswanathan, 2005), socioeconomic status (Mindt et al., 2008; Morton & Harper, 2007), environmental factors (Bialystok, 2006; Bialystok et al., 2005; Bialystok & DePape, 2009), community membership (Green, 2011), and individual differences in language use (Festman, et al., 2010). In particular, code-switching, the effortless and intentional alternation between two languages within a given conversation (Bullock & Toribio, 2009; Gardner-Chloros, 2009), may involve inhibition. It takes longer to produce a word in a language that has recently been used then disengaged than one that has not recently been used (Philipp & Koch, 2009). It is thought that this delay in production occurs because switching out of a language requires backward inhibition of the previously used language (e.g.Mayr & Keele, 2000). In addition, when rapidly switching between two languages, the languages interfere with each other more than when they are used in isolation (Christoffels, Firk, & Schiller, 2007). Constantly engaging an inhibitory control mechanism in order to switch between languages and manage interference that arises as a result of keeping both languages available during code-switching may strengthen general inhibitory control. Therefore, rather than using inhibition to suppress interference from the one 2

Global inhibition (of the language) and lexical inhibition (of individual words within the language) are treated together throughout this dissertation.

2

language in order to stay in the other language, inhibitory control ability in bilinguals may be improved by using backward inhibition to switch between languages. It is hypothesized here that it is the way in which language is used – rather than simply knowing more than one language – that confers a benefit to inhibitory control mechanisms. Specifically, the bilingual advantage may arise from managing interference from one language in order to stay in the other, or from inhibiting one language in order to switch into the other. Code-switching frequency, which differs across bilingual social groups, provides an ideal way to determine which aspect(s) of bilingualism contribute to the bilingual advantage (Green, 2011; Poplack, 1980, 1988). To determine whether inhibition during code switching modulates inhibitory control, “switchers” who tend to code switch within a conversation and “non-switchers” who tend to keep their languages separate (e.g. speaking one at home and the other at work) were compared on tasks which measure non-language inhibitory control. If non-switchers use inhibition to manage interference from the language not in use, it was expected that they would demonstrate an advantage over switchers and monolinguals on tasks that require inhibition. Conversely, if switchers use inhibition to switch frequently between languages, they should demonstrate an inhibitory control advantage. These are not necessarily opposing theories; it is possible that both switchers and non-switchers engage inhibition and a general bilingual advantage may be observed. In order to examine the cognitive modifications that may arise as a result of codeswitching, three main aims were addressed. The first was to create an efficient and reliable survey to measure how frequently an individual code-switches. The second was to validate the survey by determining whether code-switching frequency modulates processing of sentential code switches. As part of the second study, it was first necessary to determine which processing 3

stages differ for code-switches and non-switches. Finally, the validated survey was used to distinguish switchers and non-switchers in order to test whether frequency of code-switching affects inhibitory control during non-language tasks. Aim 1: Creation of the Assessment of Code-Switching Experience Survey (ACSES) Bilinguals code-switch for a variety of social, situational, and individual reasons (Appel & Muysken, 2005; Grosjean, 1982; Gumperz, 1982), and the frequency of switching varies across communities, situations, and individuals (Poplack, 1980, 1988). All healthy bilinguals possess the ability to switch, although the degree to which they engage in this behavior varies from nearly never to quite frequently (Bullock & Toribio, 2009; Poplack, 1988). The motivation, frequency, and patterns of code switching differ drastically between communities, and even between individuals within a community (Blom & Gumperz, 1972; Poplack, 1988). Given this heterogeneity, the bilingual research community needs a tool that can be used across a variety of studies to reliably assess multiple aspects of an individual’s code-switching experience. The Assessment of Code-Switching Experience Survey (ACSES) was created to measure how often a bilingual intentionally code-switches, and distinguishes between different motivations for intentional switching (e.g. to compensate for weak linguistic access (Gumperz, 1982) or to express social, emotional, or semantic factors that are better expressed in one language than the other (Grosjean, 1982; Heredia & Altarriba, 2001; Poulisse & Bongaerts, 1994)). In Chapter 1, I present the development and rigorous testing of ACSES using a three-step process. During the first study, a long version of the survey was created by compiling questions from experts in the field and using focus group input. The survey was administered to 408 participants and the factor analytic solution was used to create an internally consistent short 4

version of the survey. Second, the short version was administered to a new population and the factor analytic solution was confirmed. In a third study, a demographics analysis was performed to ensure the survey was generalizable to Spanish/English bilinguals of different ages, order of language acquisition, and geographical locations. The survey was tested for factorial validity, test-retest reliability, and internal consistency. However, before using the survey to test performance on non-language tasks, it was necessary to validate the survey in a code-switching experiment. Aim 2: Does code-switching experience modulate processing of a sentential code switch? Although many studies have investigated code switching from the sociolinguistic perspective, few have explored the cognitive changes that may arise from residing in a codeswitching environment. One such potential consequence is the ease of processing a code switch. Processing passages containing code switches incurs a reading time delay compared to passages in only one language, while comprehension is ultimately preserved (Kolers, 1966; Macnamara & Kushnir, 1971). Similar delays found for producing a code switch are attenuated in individuals who code-switch frequently in daily life compared to those who do not (Prior & Gollan, 2011). Likewise, I hypothesized that bilinguals who code switch frequently would process code switches more easily than bilingual non-switchers. In Chapter 3, I present a study in which a sentence reading paradigm with embedded code switches was used to test whether codeswitching frequency, as measured by ACSES, impacts processing of a code switch, and what processing stages are affected. This study provided a crucial validation of ACSES on a language task involving code-switching. Reading delays for code-switched words may be due to processing costs at various levels of processing, such as word recognition, semantic retrieval, and integration of a word into the 5

preceding context (Grainger & Beauvillain, 1987; Grainger & Beauvillain, 1988; Grainger & O'regan, 1992). Event-related potentials (ERPs) provide a fine-grained temporal measure of processing that allow us to determine the stages of processing affected by encountering a written code switch. Two ERP components are known to be elicited by unexpected events during sentence reading: the N400, which is linked to semantic retrieval, and the late positive complex (LPC), which is linked to integration of a word into the preceding sentence context. The N400 is a centro-parietal negativity that peaks around 400ms after the onset of a semantically meaningful stimulus and is typically larger for words that are more difficult to integrate into the context than words that are more plausible or predicted (for a review see Kutas & Federmeier, 2011; Kutas & Hillyard, 1980b). In contrast, late positivities are linked to reprocessing and repair of problematic or unexpected sentences and integration of the word into the sentence (Hagoort, Brown, & Groothusen, 1993; Kutas & Hillyard, 1980a; Osterhout & Holcomb, 1992). In the first ERP study of sentential code switching, Moreno et al. (2002) proposed that the code-switching cost may not occur at the semantic level, but rather during a working memory process needed to integrate the code switch into the preceding sentence structure (reflected by a left anterior negativity around 400ms post stimulus onset) and during a decision-related stage of processing (reflected by a LPC to a code switch). This LPC has been interpreted as a taskreconfiguration processes needed to activate the other language in order to integrate the code switch into the preceding context (see E. M. Moreno, Rodríguez-Fornells, & Laine, 2008). However, because semantic violations never occurred on code switched words in the Moreno et al. study, it was not clear if code switching also incurred a cost on accessing the meaning of individual lexical items. Conversely, Proverbio et al. (2004) and van der Meij et al. (2011) both observed an N400 effect in response to a code switch, but only van der Meij et al. reported a 6

code switching effect on the LPC (Proverbio, Leoni, & Zani, 2004; Van Der Meij, et al., 2011). Direct comparison between these studies is made difficult by methodological differences. Thus before testing the effects of frequency of code-switching on processing of a code switch, it is necessary to identify the electrophysiological correlates of code switching and whether reading a code switch affects access to meaning. Therefore, the sentential code-switching study had two goals. First, semantic violations were used as a probe of the sensitivity to semantic access of a code-switched word. I implemented a fully crossed design, which manipulated semantic congruity and the occurrence of a code switch on the same word, allowing me to isolate the ERP components elicited by each process and determine at what stages in comprehension, if any, these processes would interact. Once ERP correlates of processing a code-switch were identified, I could address the second key question whether processing a code switch is modulated by frequency of code-switching. There is some evidence that expertise increases the amplitude of the LPC during a given task. Likewise, I predicted the amplitude of the LPC elicited in response to a code-switch would be positively correlated with frequency of code-switching. Aim 3: Language Use Modulates Inhibitory Control The sentence reading study in Chapter 3 showed that code-switching experience, as measured by ACSES, modulated the neural response to processing a sentential code switch. It may be possible that frequent code switching can also affect cognitive processes that overlap functionally and perhaps anatomically with code switching (Festman, 2012; Festman & Münte, 2012; Festman, et al., 2010; Green, 2011; Luk, Green, Abutalebi, & Grady, 2012; Prior & Gollan, 2011; Prior & MacWhinney, 2009). The final aim of this dissertation as presented in Chapter 4 was to determine whether this code switching experience also translates to non7

language tasks involving inhibition. If the bilingual advantage occurs as a result of using inhibition frequently to switch between languages, switchers should demonstrate an inhibitory control advantage. Alternatively, if the advantage results from using inhibition to manage interference from the language not in use, non-switchers should demonstrate an advantage over switchers and monolinguals on tasks that require inhibition. Finally, inhibition engaged during both switching and staying may play a role in the bilingual advantage, leading to a general benefit for bilinguals over monolinguals. To test these predictions, switchers, non-switchers, and monolinguals performed two tasks thought to measure inhibitory control: the Simon and Flanker tasks. In the Simon task, participants indicate the color of a stimulus whose spatial location is either congruent or incongruent with the response hand. On incongruent trials, participants must inhibit conflicting spatial and color stimulus cues to select the correct response (see Lu & Proctor, 1995 for a review; Simon & Rudell, 1967). In the Flanker task, participants must inhibit conflicting information from distracters (Eriksen & Eriksen, 1974). Specifically, a target chevron which indicates which button the participant should push is flanked by either congruent chevrons pointing in the same direction as the target (e.g.