Grammatical complexity and comprehension of compressed speech.

Abstract

In order to investigate the effect of rate of presentation on the compression of materials that differed in grammatical complexity, an experiment was designed in which the comprehension-test portions of Forms A and B of the Nelson-Denny Reading Test were rewritten in an attempt to reduce grammatical complexity so as to make two difficulty levels available. The rewriting resulted in grammatically simplified versions of the original material. The variables, arranged in a multifactor independent-groups design, were: which of two levels of grammatical complexity was used, high or low; which of four rates of presentation was used, 175, 275, 325, or 375 words per minute; and which of two forms of the material was used, Form A or Form B. The dependent variable was the number of csorrect responses to test questions. An analysis of covariance was used to test the statistical significance of the effects. Verbal Scholastic Aptitude Test score was the adjusting variable. The following effects were statistically significant: (a) level of grammatical complexity—the grammatically simplified versions of both forms of the material resulted in greater average comprehension than the original versions; (b) rate of presentation—generally, comprehension remained at approximately the same level as rate of presentation increased from 175 wpm to 325 wpm, but from 325 wpm to 375 wpm, comprehension dropped off sharply; (c) form of material—Form B of the material resulted in greater average comprehension than Form A; (d) form of material by level of grammatical complexity—comprehension varied as a function of grammatical complexity when Form A of the test was used but did not vary when Form B was used.