Activation theory and task design: An empirical test of several new predictions.

Predictions from activation theory concerning task design are presented. Activation theory is modified to (a) account for the frequent empirical finding that relatively nonstimulating tasks cause elevated arousal levels and (b) allow for effects of individual differences on task-characteristics/taskresponse relationships. An experiment was conducted to test the hypotheses that (a) a relatively nonstimulating task would cause more negative reactions (increased arousal, decreased satisfaction) than a moderately high-stimulating task; (b) electrodermal lability and extroversion would moderate task-stimulation-level/task-response relationships; and (c) there would be inverted-U relationships between activation levels and performance. Modest support for each of the hypotheses was obtained. Activation theory is discussed as an explanation for the process that intervenes between task stimulation and responses of task performers. Activation theory has been used to explain the effects of variations in task design on responses of task performers (Scott, 1966), but little empirical research has evaluated activation theory's utility as a task design theory. This shortage of research may be due to a belief that activation theory allows only vague predictions (Hackman & Oldham, 1980; Steers & Mowday, 1977), or does not consider effects of individual differences (Hackman & Oldham, 1980; Hulin & Blood, 1968), in explaining task-characteristic/task-response relationships. The present study addresses these issues in an empirical test of activation theory. Activation theory is based on the concept of activation level, denned here as the degree of neural activity in the reticular activation system (RAS), a major part of the central nervous system (Fiske & Maddi, 1961; Scott, 1966). Activation level is monotonically related to the total stimulation impact on the RAS from external (e.g., noise), internal (e.g., heart rate), and cerebral cortex (e.g., cognitions) sources. Because activation level is difficult to measure directly, it is typically inferred from other measures of central nervous system activation (e.g., electroencephalograms) and/or physiological arousal (e.g., skin conductance).