Investigating the relationship between individual differences and island sensitivity

Abstract

It is well-attested that native speakers tend to give low acceptability ratings to sentences that involve movement from within islands, but the source of island effects remains controversial. The grammatical account posits that island effects result from syntactic constraints on wh-movement, whereas the resource-limitation view posits that low ratings emerge due to processing-related constraints on the parser, such that islands themselves present processing bottlenecks. We address this debate by investigating the relationship between island sensitivity and individual differences in cognitive abilities, as it has been argued that the two views make distinct predictions regarding whether a relationship should hold. Building directly on Sprouse et al. (2012a), we tested 102 native English speakers on four island types (whether, complex NP, subject, and adjunct islands) using an acceptability judgment task to quantify island sensitivity and three cognitive tasks to capture individual differences in working memory (via reading and counting span tasks) and attentional control (via a number Stroop task). Our results reveal strong island sensitivity effects across all island types. However, we did not find evidence that individual differences in working memory and attentional control modulated island sensitivity, which runs counter to the resource-limitation view (Kluender & Kutas 1993). These results are in line with grammatical accounts of island effects rather than due to processing difficulties.