Optimal decision-making between multiple objects requires accurate recall of each object’s value. Value recall is known to be affected by a number of factors including the training duration and the number of objects, etc. Here, we hypothesized that the level of discrete values that has to be learned while other factors being equal impact recall. In this study, participants (n = 10) learned to associate abstract fractal objects with monetary rewards. The objects were divided into three groups with two, three or five reward levels, respectively. Importantly, the number of objects in each group, the dynamic range of values and the training duration were the same across the three groups and the low-level visual features were randomized between value categories. By the end of value learning, subjects were asked to indicate the value of each object using a sliding bar (unitary choice trials). Subjects’ performance for all three groups were similar and not significantly different (2 level: 86.2%, 3 level: 87.7%, 5 level: 88.4%, p>0.1). Importantly, value memory tested around 2 hours later using the same unitary choice trials showed lower recall for the objects that belonged to groups with more reward levels (2 level: 84.6%, 3 level: 77.8%, 5 level: 64.2%, p<0.05). Our results suggest that all else the same, value resolution can affect value recall. It remains to be seen whether exposure to contexts with different value resolutions shapes subsequent choice behavior which may be suggestive of a framing phenomenon in our future studies.

Authors:
Zahra Naghdabadi, Sharif University of Technology, Iran
Mehran Jahed, Sharif University of Technology, Iran
Ali Ghazizadeh, Sharif University of Technology, Iran

About the Presenter(s)
Ms. Zahra Naghdabadi is currently a PhD candidate at Department of Electrical Engineering at Sharif University of Technology, doing her thesis under supervision of Dr. Jahed and Dr. Ghazizadeh in neuroscience.

See this presentation on the full schedule – On Demand Schedule

Bookmark