Neural mechanisms of multiple choice decision making

Abstract

Studies in decision neuroscience have focused on examining the neural mechanisms of simple decisions where the size of the choice set is small. However, real life decision making often involves choices between many more options, such as shopping for a new laptop or deciding the destination for a honeymoon trip. However, it is broadly unclear whether those simple decision mechanisms are generalizable to decisions with large choice sets. Despite being known as an important region to decision making, the dorsolateral prefrontal cortex (dlPFC) was only reported to be active in a subset of decision making experiments. However, the exact reason why the dlPFC is not involved in other decision making experiments is largely unclear. Apart from decision making, the dlPFC is also implicated in working memory. Hence, it is possible that the dlPFC is particularly important to decision when working memory is particularly demanding (for example when a large number of alternatives are available). Since decision neuroscience studies were largely focused on simple decisions, the role of the dlPFC in multiple choice decision is poorly understood. In addition, another region that is particularly important to both decision making and working memory is the posterior parietal cortex (PPC). The PPC seems to have a similar role to the dlPFC in working memory which makes it another potential candidate that is involved in multiple choice decision making. Hence, I have conducted two studies to examine the roles of dlPFC and PPC in multiple choice decision making. In Study 1, I recruited human participants to perform a multiple choice decision making task, in which they chose between two, four or sixteen options. To test the causal role of dlPFC in multiple choice decision making, I applied either anodal or sham transcranial direct current stimulation (tDCS) over the right dlPFC, in a double-blinded crossover design. I found that in the control sham tDCS session, better decisions were made when there were longer fixations on the better options and also poorer decisions were made when there were longer fixations on the poorer options, regardless whether there were two, four or sixteen options. Interestingly, after the enhancement of dlPFC in the anodal tDCS session, the impact of fixating poorer options on choice accuracy was attenuated on sixteen-option trials, but not on two- and four-option trials. In addition, a follow-up analysis revealed that the impact was strongest on fixations that occurred early, rather than late, during the decision making process. These results suggested that the dlPFC could reduce the influence of the irrelevant information on our choice. In Study 2, an experimental paradigm that was similar to that of Study 1 was adopted. Instead of applying stimulation over the dlPFC, I stimulated the right PPC to test its role in multiple choice decision making. In contrast to the stimulation over the dlPFC, anodal tDCS of the PPC did not alter the impact of fixating the better and poorer options on choice accuracy. However, after the anodal stimulation, the fixations on the options presented on the contralateral side of the stimulation became more influential to whether or not options on the same side were chosen. These results implied that the PPC has a role in processing the option information that is located on the contralateral spatial location. Overall, my results suggested that the dlPFC and PPC have dissociable roles during multiple choice decision making. The dlPFC is related to the processing of decision-relevance of option information. It is particularly important to the filtering of information from poorer options. In contrast, the PPC is related to the processing of spatial representation of decision information to bias decision making. My findings demonstrate that it is possible to enhance the filtering or spatial representation of choice information by using tDCS during multiple choice decision making.