Abstract
There may be no property of human communication more widely recognized than the fact that interpersonal messages function in the service of multiple social goals. Despite this, relatively little is known of the encoding processes that give rise to such messages. One effort to specify these processes is found in the work of Greene and Lindsey (1989) who draw on action assembly theory in positing a model of the structures and processes involved in production of multiple‐goal messages. These authors suggest that the increased cognitive load accompanying multiple‐goal messages arises from difficulties in assembling, or integrating, incompatible message features. This paper reports two empirical investigations designed to examine the plausibility of Greene and Lindsey's account. Experiment 1 revealed that, relative to messages aimed at accomplishing two compatible social ends, messages produced in pursuit of incompatible goals were characterized by greater pause‐phonation ratios and longer average pause durations. In Experiment 2, goal compatibility was found to be less important in its effect on these variables than number of goals (i.e., single versus multiple). The results of these studies are taken to indicate that the account offered by Greene and Lindsey must be amended, and that in addition to the demand on time and processing capacity associated with assembling incompatible message features, multiple‐goal messages are characterized by heavier demand on processing capacity associated with maintaining more complex message‐relevant specifications while they are executed.