Optimal Response in Decision Making: An Experimental Investigation of Decision Strategies
MetadataShow full item record
A decision process can be conceptually separated into a perceptual process and a decision strategy. The former includes all the different mechanisms that contribute to accumulate information relevant to the decision, whereas the decision strategy determines when enough information has been accumulated and a decision can be taken. Although perceptual processes have been extensively investigated in the last decades, decision strategies have received comparatively little attention. The main aim of this work is to fill this gap by analysing four decision strategies with two different experimental paradigms. We also focus on ancillary decision-making topics, such as the effect of stimulus intensity, foreperiod duration, payoff manipulation, and the response distributions in the rate domain. We initially performed a qualitative analysis of decision strategies by using a classic reaction time tasks on human participants while assuming the Drift Diffusion Model, one of the many models used for simple and fast decisions, as the perceptual process. We found that increasing the time of the trial does not have a relevant effect on the response, which is in contrast with some of the decision rules considered here. However, this approach is limited by the implicit assumption of a perceptual model that would result in different prediction for the decision strategies. We suggest the use of a different experimental design, called the EXACT Paradigm, which allows us to analyse decision strategies without having to assume any perceptual process. We tested the feasibility of such approach and applied it to several experimental studies, including a direct comparison with a classic reaction time task. Overall, two of the four decision strategies (modified Reward Rate and Reward/Accuracy) appeared to model the data satisfactorily. We discuss several ways in which the EXACT Paradigm can be used for expanding our knowledge in the field of decision-making.