Dynamical aspects of decision making under uncertainty

---

Tezin Türü: Doktora

Tezin Yürütüldüğü Kurum: Orta Doğu Teknik Üniversitesi, Enformatik Enstitüsü, Sağlık Bilişimi Anabilim Dalı, Türkiye

Tezin Onay Tarihi: 2016

Öğrenci: KEMAL TAŞKIN

Danışman: DİDEM GÖKÇAY

Özet:

Decision making is defined as the mental process of choosing among a set of alternatives. Although several aspects of decision making have been investigated so far in cognitive science, dynamics of this process as a whole remained to be studied in detail. In this context, we investigated the effects of consecutive decisions on the current decision in a decision making task under uncertainty. In the attempt of elaborating these effects, we analyzed responses of participants to risky choices in two experiments by two different approaches. In the first experiment, in order to understand participants’ physiological expressions under risky choices, we utilized TOBII T120 eye-tracker and collected pupillary responses of participants. In the second experiment, we collected participants’ response times to given decisions. We analyzed our results with a slightly modified version of Balloon Analog Risk Task (BART). Participants were also subjected to a survey, DOSPERT, prior to the experiments to monitor and distinguish their individual risk taking attitudes. In addition, in two supplementary experiments using Cups Task and Cambridge Gambling Task (CGT), particular aspects of the proposed system were shown to generalize beyond BART. Finally, a simulation was also developed and run in order to elaborate on whether participants’ decision strategies indicated a learning of the task. Our thorough analysis of participants’ responses and the results of the simulation indicated a dynamic system consisting of momentary risk taking and risk aversive states. Participants’ pupil dilation magnitudes were found to be predictable from this dynamical model, abstracted from their consecutive decisions. Natural risk attitudes, extracted from the survey had no statistically significant effect on the results. Our study indicated that risk taking states have important roles on the understanding of decision making tasks. In summary, our findings suggests a model that fuses emotional and cognitive aspects within risky uncertain decisions.