Semi-orthogonal subspaces for value mediate a tradeoff between binding and generalization

TL;DR

This study shows that the brain uses semi-orthogonal neural subspaces to bind and encode the values of options in a decision-making task, balancing binding accuracy and generalization.

Contribution

It introduces the novel concept that neural populations employ semi-orthogonal subspaces for value binding, supported by empirical data from macaque brain recordings.

Findings

Neural populations encode offer values in distinct subspaces based on spatial location.

Orthogonal subspaces encode sequential offer values, aiding binding.

Behavioral errors correlate with neural misbinding, especially for extreme offer values.

Abstract

When choosing between options, we must associate their values with the action needed to select them. We hypothesize that the brain solves this binding problem through neural population subspaces. To test this hypothesis, we examined neuronal responses in five reward-sensitive regions in macaques performing a risky choice task with sequential offers. Surprisingly, in all areas, the neural population encoded the values of offers presented on the left and right in distinct subspaces. We show that the encoding we observe is sufficient to bind the values of the offers to their respective positions in space while preserving abstract value information, which may be important for rapid learning and generalization to novel contexts. Moreover, after both offers have been presented, all areas encode the value of the first and second offers in orthogonal subspaces. In this case as well, the orthogonalization provides binding. Our binding-by-subspace hypothesis makes two novel predictions borne out by the data. First, behavioral errors should correlate with putative spatial (but not temporal) misbinding in the neural representation. Second, the specific representational geometry that we observe across animals also indicates that behavioral errors should increase when offers have low or high values, compared to when they have medium values, even when controlling for value difference. Together, these results support the idea that the brain makes use of semi-orthogonal subspaces to bind features together.