Processing of code-switched speech: an effortless activity?

Recent research on the processing of code-switched (CS) speech has yielded conflicting results, suggesting that CS can both alleviate cognitive demand and impose a cognitive cost. In light of these seemingly contradictory findings, this study examines the processing of CS utterances by late L1 Bulgarian L2 English bilinguals residing in the UK. Data from 33 Bulgarian-English bilinguals was analysed for reaction times, repetition accuracy, and speech rate. Participants were presented with 8 utterances across 4 conditions: unilingual Bulgarian, unilingual English, alternational CS, and insertional CS, which they had to listen to and rate, and afterwards repeat aloud. Preliminary results show no significant difference in the reaction times to the unilingual conditions and the CS conditions. The accuracy analysis, however, indicates a significantly higher accuracy for the CS conditions compared to the unilingual English condition, and no difference to the unilingual Bulgarian condition. A significantly slower speech rate is observed in the alternational CS condition and unilingual English condition compared to the unilingual Bulgarian condition. The findings from this study do not indicate a processing cost associated with CS activity and demonstrate that CS can be effortless for bilinguals.