Reading tasks from PALPA: How do controls perform on visual lexical decision, homophony, rhyme, and synonym judgements?

Abstract

Background: PALPA (Psycholinguistic Assessments of Language Processing in Aphasia; Kay, Lesser, & Coltheart, 1992) is a resource widely used by both clinicians and researchers. However, several of the subtests lack data regarding the performance of proficient English language speakers on these tasks.Aims: This paper investigates factors affecting the speed and accuracy of performance of young control participants on four assessments from PALPA: Visual lexical decision (subtest 25); synonym judgements (subtest 50); rhyme judgements (subtest 15); and homophone judgements (subtest 28).Methods and Procedures: Data are presented regarding both speed and accuracy of performance on each of the four tasks, and statistical analysis of those factors that influence performance within each test is carried out, for the participants as a group and also for the individuals within the group.Outcomes and Results: Visual lexical decision showed significant effects of frequency on response latency and accuracy, and of lexicality and imageability on response latency alone; synonym judgements showed significant effects of imageability on response latency; significant effects of word type were found on response latency for homophone judgements; for rhyme judgements there was a significant effect of rhyme for both accuracy and latency, and a significant interaction between rhyme and visual similarity.Conclusions: For the clinician seeking to interpret the performance of the person with aphasia on the tasks we have described here, we have presented data that provide some indication of the speed and accuracy of performance of young controls on these tasks. It is clear that ceiling effects in accuracy mask effects of psycholinguistic variables on normal performance that become apparent when speed of response is considered. However, performance is far from at ceiling for all the tasks described—some participants perform close to chance on some conditions. Finally, these data highlight the fact that comparison of the pattern of performance of individual participants with that of a group of controls can be problematic given the variability of control patterns of performance.