Individual differences in reading ability

© 1985 Dr. Linda CupplesThe present investigation comprised two experiments that were concerned with individual differences in reading-comprehension ability in relatively skilled adults. Experiment One investigated the relationship that exists between reading-comprehension skill and linguistic awareness when the latter was assessed on three levels - the semantic, the syntactic, and the phonological. It was found that individuals who achieved different levels of comprehension accuracy on a passage-reading test, also performed differently on a task that required them to bring their knowledge about form-class properties of words to conscious awareness. Thus, skilled comprehenders responded more accurately than less skilled comprehenders in a task that required them to judge whether two words served the same grammatical function in sentences. On the other hand, less skilled comprehenders did not perform below the level of skilled comprehenders in two other word-judgement tasks that required them to attend to the meanings or initial sounds of words. Experiment Two was designed to investigate the relationship between reading-comprehension skill and syntactic processes in greater detail. The semantic and syntactic word-judgement tasks from Experiment One were included, as well as two sentence-judgement tasks - meaning and grammaticality classification. Furthermore, in order to determine whether group differences would be observed when individuals did not have to make explicit judgements concerning specific linguistic properties of stimuli, subjects were also required to perform a single-sentence question-answering task. To provide more direct evidence concerning possible differences in the on-line syntactic processes of skilled and less skilled comprehenders, the sentences used in these tasks varied systematically as a function of several different types of syntactic complexity. It was hoped that this would throw some light on any differential increase in processing difficulty that the reader groups experienced as a result of (a) an increase in the number of clauses contained in sentences, (b) a lack of correspondence between the surface order of words and their underlying grammatical roles, and (c) the omission of optional grammatical function words. The results obtained in this experiment replicated those of Experiment One in that skilled comprehenders performed more accurately than less skilled comprehenders in the syntactic but not the semantic word-judgement task. In accordance with this finding, it was also true that reliable ability-related differences were more likely to be detected when individuals were instructed to focus on the grammatical well-formedness of word strings than when they were asked to attend to their meaningfulness. However, group differences were even more pronounced when individuals were required to process single sentences in sufficient detail to answer subsequent comprehension questions. In addition, although no evidence was obtained to suggest that less skilled comprehenders were affected more than skilled comprehenders by increasing the number of clauses contained in sentences, there was some evidence to suggest that skilled comprehenders were better able to cope with increases in syntactic complexity that occur when optional grammatical function words are omitted. However, this ability-related difference was only observed when sentences deviated from the standard S-V-O word order of English, and when the meanings of the individual words that they contained did not constrain the way in which those words could be combined to form a meaningful sequence (i.e., when semantically reversible object relative-clause sentences served as stimuli). On the basis of the findings from both of these experiments, it was concluded that skilled adult comprehenders were better able than less skilled adults to bring their syntactic knowledge to conscious awareness, and also that they were more likely to spontaneously perform a complete syntactic analysis of text whilst reading