Supplementary Materials.zip

The present study explores the role of individual characteristics such as frequency of code-switching and dominance, on early (N400) and late (LPC) neural responses to switched vs. non-switched targets. Most ERP code-switching studies to date have used emotionally neutral stimuli, but including emotional valence might help in characterizing the impact of code-switching habits on neural responses, especially since the effects of emotional valence have been observed on both the N400 and LPC components. Forty-seven English-Spanish/Spanish-English bilinguals were presented with English sentences ending with high-expected, emotionally congruent or low-expected, emotionally incongruent words that were either non-switched (English) or switched (Spanish) and made a meaningfulness judgment task on the final target. ERP results showed that both dominance and switching habits modulated code-switching costs. Habitual code-switchers recorded smaller switch costs than non-habitual code-switchers in the 300-500 ms time-window. In addition, bilinguals more dominant in English obtained larger switching costs in both the 300-500 ms and 600-800 ms time-windows, unlike balanced bilinguals who were more proficient in Spanish. The electrophysiological effects of expectancy and switching diverged in the N400 and LPC time-windows. Only the switching effect, but no expectancy, was present at the N400 component, and only expectancy, but no switching, at the LPC component. Valence significantly modulated early and late neural responses to Spanish and English targets. Positivity bias was found for English and negativity bias for Spanish in the N400 time -window and the reverse pattern of positivity bias for Spanish and negativity bias for English in the LPC time-window. In addition, switch costs for positive and negative words differed as a function of semantic expectancy, suggesting that emotional congruency can affect participants’ language control.