Interaction of computer synthesized voice with decision style in decision support systems
Michael J. DeHaemer · Medical Entomology and Zoology · 1987
Three different kinds of experimental tasks tested the effect of computer synthesized voice on user performance. Significant interaction between voice and decision style was observed in two of the three experimental tasks. Under the treatment with voice, information provided by the synthetic voice was exactly redundant to that on the microcomputer graphic display through which subjects completed their tasks. 85 subjects were classified using the Myers Briggs Type Indicator as having heuristic, analytic, or neutral decision styles. Response time, error, and the subject's confidence were the measured dependent variables. In a Forecasting Experiment, which was primarily a visual task, a significant effort for decision style was found for response time. An interaction effect was also found, showing improved response time by the analytics under the voice treatment. An Alerting/Memory Experiment required the subjects to recall alerted sectors in a circular display after it had gone blank. Although no improvement in performance was noted under computer synthetic voice, subjects preferred the voice--probably as an aid for short term memory. To complete a complex Shop Scheduling Experiment, subjects were required to follow a step-by-step algorithm to construct a path across the microcomputer screen display. The interaction effect of computer synthetic voice with decision style was dramatic in this experiment. The response time and error for the heuristics increased by over 70 percent, while the response time and error decreased by 28 and 38 percent for the analytics. The results of this experimental series are important because of the numbers of users whose performance and satisfaction may be affected when computer synthetic voice is added to Decision Support Systems (DSS). Some rules for design of DSS are suggested. This thesis is thought to be the first report of the effects of using computer synthesized voice in DSS.