Effects of instruction on yes-no responses to L2 collocations

Abstract

The lexical decision task (LDT), in which a participant makes dichotomous judgments on target letter strings, is an established method in psycholinguistic research to investigate the mental lexicon. With the expansion of research interests from single lexeme to collocations, second language (L2) researchers have started to use a similar judgment task at a phrasal level (referred to as a phrasal decision task or PDT in this paper). However, unlike the LDT, the PDT has not yet established a standard form of prompt, and variation has been observed in previous L2 studies. Hence, the purpose of this study was to examine effects of varying instructions on PDT performance. Three instructions (acceptable/ commonly used/natural) were tested with Japanese university students and native speakers of English, who were asked to make judgments on English word combinations. Examining responses to congruent (felicitous both in Japanese and English), incongruent (felicitous only in Japanese or English), and baseline items, the study identified some effects of instruction differences. However, these effects were not so strong as to obscure the expected cross-linguistic congruency effect. Therefore, this result has led to the conclusion that researchers have more freedom of instruction selection in the PDT, at least among the three examined in this study and to the extent that the congruency effect was measured by accuracy scores.