Individual Differences in Proactive and Reactive Control in Bilinguals Alexandre Chauvin A Thesis in The Department of Psychology Presented in Partial Fulfillment of the Requirements For the degree of Master of Arts (Psychology) at Concordia University Montréal, Québec, Canada August, 2015 ©Alexandre Chauvin, 2015 CONCORDIA UNIVERSITY School of Graduate Studies This is to certify that the thesis prepared By: Alexandre Chauvin Entitled: Individual Differences in Proactive and Reactive Control in Bilinguals and submitted in partial fulfillment of the requirements for the degree of Master of Arts (Psychology) complies with the regulations of the University and meets the accepted standards with respect to originality and quality. Signed by the final Examining Committee: ____________________________ Chair: Dr. Andrew Chapman ____________________________ Examiner: Dr. Norman Segalowitz ____________________________ Examiner: Dr. Karen Li ____________________________ Supervisor: Dr. Natalie Phillips Approved by _____________________________________________________________ Dr.Virginia Penhune, Chair of Department _____________________________________________________________ Dean of Faculty of Arts and Science: Dr. André Roy Abstract This study investigated individual differences in bilinguals’ use of proactive and reactive control processes during an executive control task (the AX-CPT) in relation to aspects of the bilingual experience (e.g., second language proficiency). Participants were presented with cue- target letter pairs, one letter at a time (AX, AY, BX, or BY; B and Y are any letter other than A or X) and were instructed to press the “yes” button for AX pairs and the “no” button for any other pair. They completed three blocks which varied in terms of the most frequent trial type (AX-70% vs. AY-70% vs. BX-70%). Event-related brain potentials (ERPs) were recorded from 15 young adult bilinguals during the AX-CPT. The N2, an ERP related to conflict detection, was analyzed in conjunction with behavioural performance. Individual variations in cognitive control strategy were differentially associated with aspects of bilingualism in the AX-70 and AY-70 blocks. In the AX-70 block, greater engagement of proactive control was associated with shorter overall reaction times (RTs), lower accuracy, and enhanced conflict detection. In the AY-70 block, a proactive strategy was associated with lower accuracy, but similar RTs compared to a reactive strategy. Different patterns of association were found between self-reported language-switching behaviours and cognitive control strategy in the AX-70 block compared to the AY-70 block. The results support the idea of individual differences in the relative use of proactive and reactive mechanisms in bilinguals. These differences were related to aspects of language- switching which is an important source of interindividual variability among bilinguals. iii Acknowledgments This research project was supported in great part by grants awarded to Dr. Natalie Phillips from the Natural Sciences and Engineering Research Council of Canada (NSERC; grant number 203751). Additional financial support was provided by Concordia University, and the Centre for Research on Brain, Language and Music. I would like to express my most sincere gratitude to Dr. Natalie Phillips for her patience, her guidance, and many stimulating conversations over the course of this project. I would also like to thank current and past members of the Cognitive, Aging and Psychophysiology Laboratory for their support both moral and instrumental. In particular, I want to thank Julia Carvalho, Alexandra Covey, Kristina Coulter, and Camille Williams for their assistance with recruitment and data collection. Additionally, I would like to extend heartfelt thanks to Samantha Bishundayal for her help and support throughout the entire project. A most sincere thank you to the members of my thesis review committee Dr. Karen Li and Dr. Norman Segalowitz for their input and the constructive advice they have given me since the very beginning of this project. A final thank you goes to my friends and family for their encouragement and their support. In particular, I would like to thank my partner Jillian Budd for motivating me and continuing to believe in me when I most needed it. iv Table of Contents List of Figures ............................................................................................................................... vii List of Tables .................................................................................................................................. ix Introduction ..................................................................................................................................... 1 Why a bilingual advantage in executive control? ............................................................... 3 The bilingual advantage in executive control ..................................................................... 9 Dual modes of cognitive control ....................................................................................... 16 The state of the bilingual advantage hypothesis ................................................................ 19 Assessing individual differences ....................................................................................... 20 Methods ......................................................................................................................................... 26 Participants ........................................................................................................................ 26 Bilingual Language-Switching Questionnaire .................................................................. 27 Stimuli ............................................................................................................................... 27 Procedure ........................................................................................................................... 30 EEG Data Acquisition ....................................................................................................... 32 Results ........................................................................................................................................... 34 Behavioural results ............................................................................................................ 34 Electrophysiological results .............................................................................................. 36 Correlational analyses ....................................................................................................... 40 Strategy in the AX-70 block .................................................................................. 40 Strategy in the AY-70 block .................................................................................. 43 Strategy in the BX-70 block .................................................................................. 45 Discussion ..................................................................................................................................... 45 AX-70 block ...................................................................................................................... 46 AY-70 block ...................................................................................................................... 49 Limitations and future research ......................................................................................... 52 Conclusion ......................................................................................................................... 54 References ..................................................................................................................................... 56 Tables ............................................................................................................................................ 65 Figures ........................................................................................................................................... 76 Appendices .................................................................................................................................... 90 List of Figures Figure 1. Reaction time on all four trial types (AX, AY, BX, BY) across three blocks (AX-70, AY-70, BX-70) .............................................................................................................................. 76 Figure 2. Accuracy on all four trial types (AX, AY, BX, BY) across three blocks (AX-70, AY-70, BX-70) ........................................................................................................................................... 77 Figure 3. Target-locked electrophysiological activity for all four trial types in the AX-70 block at site FCz .......................................................................................................................................... 78 Figure 4. Target-locked electrophysiological activity for all four trial types in the AX-70 block at site CPz .......................................................................................................................................... 79 Figure 5. Target-locked electrophysiological activity for all four trial types in the AY-70 block at site FCz. ......................................................................................................................................... 80 Figure 6. Target-locked electrophysiological activity for all four trial types in the AX-70 block at site CPz .......................................................................................................................................... 81 Figure 7. Target-locked electrophysiological activity for all four trial types in the AY-70 block at site FCz. ......................................................................................................................................... 82 Figure 8. N2 amplitude on all four trial types (AX, AY, BX, BY) across three blocks (AX-70, AY-70, BX-70) .............................................................................................................................. 83 Figure 9. N2 amplitude on all four trial types (AX, AY, BX, BY) for five electrodes (Fz, FCz, Cz, CPz, Pz) collapsed across the three blocks ............................................................................. 84 Figure 10. N2 latency on all four trial types (AX, AY, BX, BY) across three blocks (AX-70, AY- 70, BX-70) ..................................................................................................................................... 85 Figure 11. Correlation between participants’ Accuracy Contrast value (BX – AY) and their average RT in the AX-70 block collapsed across trial types ........................................................ 86 Figure 12. Correlation between participants’ Accuracy Contrast (BX – AY) value and their average accuracy in the AX-70 block collapsed across trial types ............................................... 86 vii Figure 13. Correlation between participants’ RT Contrast (BX – AY) value and the average amplitude of the N2 component at site CPz for AY trials in the AX-70 block ............................ 87 Figure 14. Correlation between participants’ RT Contrast (BX – AY) value and the average latency of the N2 component at site CPz for AY trials in the AX-70 block ................................. 87 Figure 15. Partial correlation between participants’ RT Contrast (BX – AY) value in the AX-70 block and their score on the Unwanted Switches factor of the Bilingual Language Switching Questionnaire controlling for L2 proficiency ................................................................................ 88 Figure 16. Correlation between participants’ Accuracy Contrast (BX – AX) value and their average accuracy in the AY-70 block collapsed across all trial types. ......................................... 88 Figure 17. Partial correlation between participants’ RT Contrast (BX – AX) value in the AY-70 block and their score on the L2 to L1 Switches factor of the Bilingual Language Switching Questionnaire controlling for L2 proficiency ................................................................................ 89 Figure 18. Partial correlation between participants’ RT Contrast (BX – AX) value in the AY-70 block and their score on the Contextual Switches factor of the Bilingual Language Switching Questionnaire controlling for L2 proficiency ................................................................................ 89 viii List of Tables Table 1. Demographic data for all participants ............................................................................. 65 Table 2. The Bilingual Language Switching Questionnaire and its four factors .......................... 66 Table 3. Probabilities in the three blocks of the AX-CPT ............................................................ 67 Table 4. RT and Accuracy contrasts in the AX-70 block for all participants ............................... 68 Table 5. Correlations between cognitive control strategy and behavioural performance in the AX-70 block ............................................................................................................................ 69 Table 6. Correlations between cognitive control strategy and N2 amplitude on AY trials in the AX-70 block ........................................................................................................................ 69 Table 7. Correlations between cognitive control strategy and N2 latency on AY trials in the AX-70 block ............................................................................................................................ 70 Table 8. Partial correlations between code-switching and cognitive control strategy controlling for L2 proficiency in the AX-70 block ....................................................................... 71 Table 9. RT and Accuracy contrasts in the AY-70 block for all participants ............................... 72 Table 10. Correlations between cognitive control strategy and behavioural performance in the AY-70 block ........................................................................................................................ 73 Table 11. Correlations between cognitive control strategy and N2 amplitude on AY trials in the AY-70 block ............................................................................................................... 74 Table 12. Correlations between cognitive control strategy and N2 latency on AY trials in the AY-70 block ........................................................................................................................ 75 Table 13. Partial correlations between code-switching and cognitive control strategy controlling for L2 proficiency in the AY-70 block ....................................................................... 75 ix 1 Individual Differences in Proactive and Reactive Control in Bilinguals Executive control (also referred to as cognitive control) is an integral part of the cognitive system and is involved in the regulation and coordination of one’s cognitive processes (Koechlin, Ody, & Kouneiher, 2003). Some examples of cognitive control processes include working memory (i.e., the capacity to maintain, manipulate, and update information for brief periods of time in the presence of concurrent distractions or processing demands), problem solving, inhibiting irrelevant information, planning, and the ability to flexibly alternate between tasks. Evidence suggests that bilingualism influences cognitive control, with bilinguals faring better than monolinguals on various tasks that involve inhibition (e.g., Bialystok et al., 2005), task switching (e.g., Prior & Macwhinney, 2009), and conflict resolution (e.g., Costa, Hernández, & Sebastián-Gallés, 2008). However, both the nature (e.g., Kroll & Bialystok, 2013) and the very existence (e.g., de Bruin, Treccani, & Della Sala, 2014; Hilchey & Klein, 2011; Klein, 2014, Paap & Greenberg, 2013) of this bilingual advantage are heavily debated. In an attempt to clarify the nature of the advantage and explain the inconsistent empirical findings, this thesis explored individual differences in aspects of bilingualism (e.g., proficiency, age of acquisition) and their relation to cognitive control. Prior to examining the reasons behind the emergence of a bilingual advantage in cognitive control, one must review cognitive abilities in general. Cognitive abilities are characterized by their capacity to adapt. This propensity for change and adaptation is often referred to as plasticity (Kolb & Whishaw, 1998), a term which has seen many different definitions over the years (for a brief review, see Lövdén, Bäckman, Lindenberger, Schaefer, & Schmiedek, 2010). However, it is generally accepted that cognitive abilities can be influenced or modified through extended use and learning experiences.