Dependence of perceptual choice on number of response alternatives and fidelity of evidence

The study of perceptual decisions has been developed as a substitute for investigating more complex multiple attribute decisions. However, little attention has been paid to the similarity of results between the two literatures. Four separate behavioral experiments and a secondary trial- by-trial analysis investigated the sensitivity of perceptual decisions. Results were compared to both previous perceptual decision research and that of multiple attribute decisions in an effort to bridge the divide. The first experiment examined the effect of increasing the similarity of available response alternatives on accuracy and reaction time. The results suggest that high levels of similarity can begin to degrade the decision process by lowering accuracy and slowing reaction time; however these changes may be dependent on the extent to which the alternatives use overlapping neuronal pools. The second experiment examined the effect of increasing the number of response alternatives available for a single decision. The results suggest that increasing the number of alternatives may not affect performance until some critical point (in this case, eight alternatives). The third experiment examined how delay in the presentation of evidence compared to the start of the decision process affects the overall accuracy and reaction time once information is given. The results demonstrate that as the decision process extends in time, decisions are made faster and less accurately. Finally, the fourth experiment examined how the interrupting the incoming stream of information with either highly informative or highly misleading evidence would affect the decision. The results illustrated a complicated picture in which highly informative evidence accelerated decisions but misleading evidence failed to slow decisions. In addition to the individual aims, a secondary analysis investigated potential trial-by- trial variation in performance. There was some evidence that participants made ongoing adjustments to their strategy dependent on performance, but only when feedback was available; previous trial status (e.g., highly informative vs. highly misleading evidence) did not affect current trial performance. In sum the results demonstrate that perceptual decisions do show high levels of sensitivity to a variety of manipulations, but fail to replicate many of the results from more complicated multiple attribute decisions