Comparing young children's comprehension and production in discourse

Abstract

This study used a verbal communication (discourse) framework to compare comprehension (receptive language) and production (expressive language). The two processes were studied using a test paradigm with the child in the responder role and the adult in the initiator role.;Comparable receptive and expressive tests were designed. A communication model developed by Blank, Rose and Berlin (1978a, 1978b) was used which allows one to control an essential aspect of discourse, i.e., the cognitive complexity level(s) of demands posed by the adult. The conceptual complexity of the adult's demands are placed on a four level scale with successive levels representing increasing degrees of cognitive complexity. The ability of the child to respond is assessed by a scale focused on adequacy of response.;Each test mode (R;E) contained 60 items; 15 at each of the four complexity levels. Middle class children, aged four to six, were individually tested in three sessions. There were four groups of 32 children each: two"single mode" groups to assess test-retest effects (R-R; E-E) and two "dual mode" groups (E-R; R-E) to allow for the comparison of the two processes as well as to assess order effects. To obtain an estimate of intellectual functioning all children were individually administered the Slosson Intelligence Test.;The hypotheses were that in a structured test paradigm, comprehension and production would be equivalent processes when the demands in the two modes were of equivalent level(s) of complexity.;The results supported the hypotheses in that when equated for conceptual difficulty, there appeared to be an overall comparability of the two systems. The results indicated that ability to comprehend and produce seems to be more dependent on the degree of complexity of a demand rather than on the mode of assessment. The one difference between modes occurred at Level I. As anticipated, age significantly affected performance; the four groups of skills progressed in difficulty as predicted by the discourse model; order of testing in the "dual mode" groups was found to affect performance differentially; and there was a differential patterning of Levels I-IV as related to IQ level.;The implications of these findings for children with atypical language patterns are explored.