Decision making under uncertainty in a quasi realistic binary decision task – An fMRI study

Abstract Approaching real life decision making through Quasi Realistic Decision Making may increase the ecological validity of decision making experiments. This could help narrow the gap between laboratory settings and observations in real world contexts and thus allow for generalization of laboratory results to everyday life. A binary decision task with quasi realistic context and stimuli was created to investigate neural processing of certain and uncertain decision making, using functional Magnetic Resonance Imaging. On the basis of behavioral data (consistency of decisions in identical contexts), trials with uncertain and certain decision making were identified. This allowed for comparing uncertain and certain conditions, and contrasting each condition with a low level baseline (i.e., between trial fixation dot). A Conjunction analysis between contrasts of uncertainty versus baseline and certainty versus baseline indicated a large overlap of neural network recruitment distributed in bilateral middle frontal, medial frontal, inferior parietal, occipito-temporal, and medio-temporal areas, and the cingulate cortex. While basic neural processing principles in uncertain and certain contexts were comparable, the direct contrast revealed activation foci in middle cingulate and in frontal and parietal areas. The quasi realistic approach revealed a common network for decision making which is modulated by uncertainty.