Abstract
Investigations of spoken language comprehension often employ subsidiary task paradigms, such as word- and phoneme-monitoring procedures. In the present experiment we investigated what kinds of effects such secondary tasks might have on the course of simultaneous comprehension processing. Subjects listened to texts either without any secondary task demands (i.e., unconstrained comprehension), under word-monitoring instructions, or under phoneme-monitoring instructions. Comprehension was evaluated immediately after the presentation of each text with verification statements probing both the representation of surface structure and, also, the representation of high-level propositions, low-level propositions, and macropropositions from the text's meaning structure. Relative to unconstrained comprehension, we found an overall facilitation of comprehension due to word monitoring, and no effects as a result of phoneme monitoring. Moreover, the pattern of results across propositional levels was consistent with predictions derived from W. Kintsch and T. A. van Dijk's (Psychological Review, 1978, 85, 363-394) theory of text comprehension. This facilitation of simultaneous comprehension processing is inconsistent with the notion of a "fixed-capacity decision mechanism," to which differences in monitoring effects are typically attributed. We argue that subsidiary task paradigms can be used for investigations of spoken text comprehension, but only when accompanied with probes for the well-established characteristics of text representation.