Cognitive and Neural Processes of Constucted Preferences
Neuroscience and Neurobiology
Past research has identified many ways in which decisions are influenced by the context of the decision environment. In this dissertation, we use eye-tracking and functional magnetic resonance imaging (fMRI) to investigate the processes and strategies individuals use in different decision contexts. The first two chapters use eye-tracking to extract fine-grained process data that uncover the different strategies that individuals employ in response to changes in the decision environment. Chapter 2 investigates how decision processes change in response to different modes of preference elicitation. We find that preferences for gambles shift between choice and bid elicitation. More importantly, these preference reversals are accompanied by differential attention to gamble attributes (probability of winning versus amount to win), suggesting that these reversals are due to differential weighting of these attributes in the two contexts. Chapter 3useseye-tracking to investigate how decision strategies shift as the number of options and time pressure increases. In response to an increasing number of choice options, all subjects consider only a subset of the options before making a decision, in which the value of a single attribute (the probability of winning) has the largest effect on whether an option is considered. Critically, those experiencing time pressure tended to then use a simpler attribute-based strategy to choose amongst the subset of options considered, while those not under time pressure tended to use a compensatory tradeoff-based strategy.Chapter 3 highlights how eye-tracking can be especially useful in detecting subtle changes in decision strategies. In the final study, we use fMRI data and multi-voxel pattern analysis to understand how decision strategies are represented in the brain. Overall, we find that different forms of task complexity recruit similar neural regions across the frontal and parietal cortices, which have previously been implicated as responding in a non-specific manner to general cognitive demands. However, inconsistent with a general or non-specific role in cognitive demand, multi-voxel pattern analysis can distinguish different forms of task complexity in all of these regions. Given that different forms of task complexity evoke different decision strategies, this suggests that fMRI could potentially be used to decode or predict the decision strategies individuals are using. Taken together, this work furthers our understanding of how decision processes changes in response to changes in the task context and suggests important future directions for using eye-tracking and fMRI to identify the processes and strategies individuals use to make a decision.