PROCESSING RESOURCE CAPACITY AND THE COMPREHENSION OF GARDEN PATH SENTENCES

Three experiments explored the relationship between verbal working mem ory capacity and the comprehension of garden path sentences. In Experi ment 1, subjects with high, medium, and low working memory spans made acceptability judgments about garden path and control sentences under whole sentence and rapid serial visual presentation (RSVP) conditions. There were no significant differences between subjects with different working memory spans in the comprehension of garden path sentences in either condition. In Experiments 2A and 2B, subjects with high and lo w working memory spans were tested on the same materials at three RSVP rates. There were no significant differences between subjects with di fferent working memory spans in the magnitude of the effect of garden path sentences at any presentation rate. The results suggest that work ing memory capacity, as measured by the Daneman and Carpenter (1980) r eading span task, is not a major determinant of individual differences in the processing of garden path sentences.