From Intention to Action: Mental Processes Underlying Movement Planning and Execution.

Abstract

The processes that immediately precede the start of a movement often cannot be voluntarily inhibited. These processes might be thought of as occurring after movement preparation has passed the "point of no return". An important question for temporal models of human performance concerns the nature of these late ballistic processes. The counterm and ing procedure and race model provide the means of studying these processes. The term "counterm and ing" comes from the fact that subjects are sometimes required to cancel a response to a previously issued comm and . In the procedure, subjects are instructed to make a designated response to a "go" signal as quickly as possible, but to attempt to avoid responding if they should subsequently detect a "stop" signal. The race model treats subjects' reactions to the two signals as a race between two independent processes to a point of no return. Whether a response will be successfully inhibited depends upon the outcome of this race. The procedure and model are used together to measure the effects of experimental manipulations on the durations of processes before and after the point of no return separately. These measurements, in turn, are used to infer the nature of each set of processes. The dissertation develops this approach, and then applies it to the study of motor programming. Some of the basic assumptions underlying the application of the race model to the counterm and ing procedure are tested in a series of experiments. The approach is used to discover which set of processes is affected by changes in response complexity. By determining whether response complexity affects ballistic processes, the dissertation provides further insight into the relation between motor programming and the overt execution of movement. The results support a model of response preparation in which (1) response inhibition depends upon the outcome of a race between independent excitatory and inhibitory processes, and (2) reaction time is the sum of the durations of at least two stages, separated by the point of no return. The ballistic processes appear to be affected by the repetition of stimulus-response pairs, but not by the physical or semantic properties of the stimulus. It also seems that they can be primed without being triggered. Finally, the effect of response complexity occurs entirely before the point of no return. From this it is argued that much of the motor programming that precedes even a single rapid movement occurs at a relatively central level and does not necessarily lead to overt movement. (Abstract shortened with permission of author.)