Interpretation of individual differences in computational neuroscience

Computational neuroscience offers an excellent opportunity to understand the neural underpinnings of behavior. However, interpretation of individual, for example developmental, differences in these underpinnings is less straightforward. We illustrate this in studies examining individual differences in reinforcement learning. In these studies, a computational model yields an individual-specific prediction error regressor to model activity in a brain region of interest. Individual differences in the resulting regression weight are typically interpreted as individual differences in neural coding. We first demonstrate that, as individual differences are already captured in the individual-specific regressor, the absence of individual differences in neural coding is not concerning. We then review that the presence of individual differences is typically interpreted as individual differences in use of brain resources. However, we illustrate in simulations that they may also have originated in: standardization of the prediction error, individual differences in brain networks outside the region of interest, individual differences in duration of the prediction error response, individual differences in outcome valuation, and in neglected individual differences in computational model parameters or the type of computational model. To disambiguate interpretations, we provide several recommendations. In this manner we hope to advance interpretation of individual differences in computational neuroscience.