Behavioral and electrophysiological investigations of semantic processing in skilled and less-skilled comprehenders

Theorists of reading comprehension failure are split between two groups: those that posit low-level word reading skills and phonological awareness as underlying factors of poor comprehension ability and those that consider poor comprehension as partially independent of these low-level skills. Several studies with children have now demonstrated that poor comprehenders with adequate decoding skills make up a small but significant proportion of poor readers. One promising hypothesis posits that semantic processing deficits underlie these children's comprehension difficulties. This hypothesis was supported by findings that demonstrated less-skilled comprehenders to show poorer than average performance on a variety of semantic tasks. In order to test whether these findings would generalize to adult poor comprehenders, we evaluated the dissociability of high-level and low-level skills in adults. In addition, we evaluated whether adult less-skilled comprehenders (with adequate decoding abilities) have semantic processing difficulties. A PCA compared the reading skills of large group of college aged readers and found that high level skills such as reading comprehension and vocabulary were partly dissociable from low-level reading skills such as decoding ability. Furthermore, in order to evaluate the semantic processing deficit hypothesis, adult skilled and less-skilled comprehenders were compared on several behavioral and electrophysiological tests of semantic and phonological processing. The findings from these studies revealed that less-skilled comprehenders generated fewer semantic associates in a verbal fluency task and showed reduced categorical priming in an automatic semantic priming task. Additionally, electrophysiological records of less-skilled comprehenders differed from skilled readers during a semantic processing task but no during a phonological processing task. Taken together these findings provide evidence that supports semantic knowledge/semantic processing differences between skilled and less-skilled comprehenders. Implications of these findings are discussed within the construct of an experience based model of semantic knowledge acquisition.