Enhancing reinforcement learning models by including direct and indirect pathways improves performance on striatal dependent tasks

A major advance in understanding learning behavior stems from experiments showing that reward learning requires dopamine inputs to striatal neurons and arises from synaptic plasticity of cortico-striatal synapses. Numerous reinforcement learning models mimic this dopamine-dependent synaptic plasticity by using the reward prediction error, which resembles dopamine neuron firing, to learn the best action in response to a set of cues. Though these models can explain many facets of behavior, reproducing some types of goal-directed behavior, such as renewal and reversal, require additional model components. Here we present a reinforcement learning model, TD2Q, which better corresponds to the basal ganglia with two Q matrices, one representing direct pathway neurons (G) and another representing indirect pathway neurons (N). Unlike previous two-Q architectures, a novel and critical aspect of TD2Q is to update the G and N matrices utilizing the temporal difference reward prediction error. A best action is selected for N and G using a softmax with a reward-dependent adaptive exploration parameter, and then differences are resolved using a second selection step applied to the two action probabilities. The model is tested on a range of multi-step tasks including extinction, renewal, discrimination; switching reward probability learning; and sequence learning. Simulations show that TD2Q produces behaviors similar to rodents in choice and sequence learning tasks, and that use of the temporal difference reward prediction error is required to learn multi-step tasks. Blocking the update rule on the N matrix blocks discrimination learning, as observed experimentally. Performance in the sequence learning task is dramatically improved with two matrices. These results suggest that including additional aspects of basal ganglia physiology can improve the performance of reinforcement learning models, better reproduce animal behaviors, and provide insight as to the role of direct- and indirect-pathway striatal neurons.Author summary: Humans and animals are exceedingly adept at learning to perform complicated tasks when the only feedback is reward for correct actions. Early phases of learning are characterized by exploration of possible actions, and later phases of learning are characterized by optimizing the action sequence. Experimental evidence suggests that reward is encoded by the dopamine signal, and that dopamine also can influence the degree of exploration. Reinforcement learning algorithms are machine learning algorithms that use the reward signal to determine the value of taking an action. These algorithms have some similarity to information processing by the basal ganglia, and can explain several types of learning behavior. We extend one of these algorithms, Q learning, to increase the similarity to basal ganglia circuitry, and evaluate performance on several learning tasks. We show that by incorporating two opposing basal ganglia pathways, we can improve performance on operant conditioning tasks and a difficult sequence learning task. These results suggest that incorporating additional aspects of brain circuitry could further improve performance of reinforcement learning algorithms.