The Role of Basal Ganglia Reinforcement Learning in Lexical Priming and Automatic Semantic Ambiguity Resolution
- Jose Ceballos, Department of Psychology, University of Washington, Seattle, Washington, United States
- Andrea Stocco, Psychology, University of Washington, Seattle, Washington, United States
- Chantel Prat, Psychology, University of Washington, SEATTLE, Washington, United States
AbstractThe current study aimed to elucidate the contributions of the subcortical basal ganglia to human language by adopting the view that these structures engage in a basic neurocomputation that may account for its involvement across a wide range of linguistic phenomena. Specifically, we tested the hypothesis that basal ganglia reinforcement learning mechanisms may account for variability in semantic selection processes necessary for ambiguity resolution. To test this, we used a biased homograph lexical ambiguity priming task that allowed us to measure automatic processes for resolving ambiguity towards high frequency word meanings. Individual differences in task performance were then related to indices of basal ganglia functioning and reinforcement learning, which were used to group subjects by learning style: primarily from choosing positive feedback (Choosers), primarily from avoiding negative feedback (Avoiders), and balanced participants who learned equally well from both (Balanced). The pattern of results suggests that balanced individuals, whom learn from both positive and negative reward equally well, had significantly lower access to the subordinate homograph word meaning. Choosers and Avoiders, on the other hand, had higher access to the subordinate word meaning even after a long delay between prime and target. Experimental findings were then tested using an ACT-R computational model of reinforcement learning that learns from both positive and negative feedback. Results from the computational model confirm and extend the pattern of behavioral findings, and provide a reinforcement learning account of lexical priming processes in human linguistic abilities, where a dual-path reinforcement learning system is necessary for precisely mapping out word co-occurrence probabilities.