Information Processing on Simultaneously Presented Auditory Tasks
Louis M. Herman, John F. Corso · The Journal of the Acoustical Society of America · 1961
This study primarily investigated the concept of a single limited-capacity channel in information processing and the role of long-term storage (memory) in controlling inputs to that channel. In a monotic pursuit situation, S's (four groups of ten) tracked a 1000-cps tone for 20 identical trials (53 sec each) during which the auditory course changed in intensity over a 35-db range, approximately. Within each trial, the course followed a random pattern, but the pattern was repeated from trial to trial. During two intervals (8 sec each) which occurred randomly within each trial, S's were required to judge simultaneously the longest in a series of five tonal pulses. The tonal pulses (four equal and one unequal in duration) were presented in S's opposite ear at either 350 or 2250 cps and were relatively easy or relatively difficult to discriminate. It was found that in the simultaneous-task situation, practiced S's transmitted more information on the auditory tracking task and less on the tonal-pulse task than did unpracticed S's. When the tracking task was of low information input, S's were able to maintain the performance level achieved on that course in the single task situation and were able to process additional information on the tonal pulse task. However, S's tracking the course of high information input were not able to maintain their prior performance level on the tracking task during the simultaneous task situation. For the more difficult discrimination of the tonal pulses, less information was transmitted on that task, but no negative effects were observed on the amount of information transmitted on the tracking task. The results of this research provide support for the single-channel hypothesis. Practice on a task results in long term storage and appears to determine the preference for selecting inputs from that task rather than from some other task. Broadbent's model requires the addition of an intertask decision loop if it is to account for the findings of this study.