Striatal action-value neurons reconsidered

Abstract

It is generally believed that during economic decisions, striatal neurons represent the values associated with different actions. This hypothesis is based on a large number of studies, in which the neural activity of striatal neurons was measured while the subject was learning to prefer the more rewarding action. Here we show that these publications are subject to at least one of two critical confounds and that most are subject to both. First, we show that even weak temporal correlations in the neuronal data may result in an erroneous identification of action-value representations. We demonstrate this by erroneously identifying action-value representations, both in simulations and in the neural activity recorded in unrelated experiments. Second, we show that the experiments and analyses designed to dissociate action-value representation from the representation of other decision variables cannot do so. Specifically, we show that neurons representing policy may be erroneous identified as representing action-values. We suggest different solutions to identifying action-value representation that are not subject to these confounds. Applying one of these solutions to previously identified action-value neurons in the basal ganglia we fail to detect action-value representations there. Thus, we conclude that the claim that striatal neurons encode action-values must await new experiments and analyses.