Differential encoding of safe and risky offers
Maisson, D. J.-N.; Yoo, S. B. M.; Wang, M. Z.; Cash-Padgett, T. V.; Zimmermann, J.; Hayden, B.
Show abstract
Common currency theories in neuroeconomics hold that neurons in specific brain regions specifically encode subjective values of offers and not stimulus-specific information. The rationale behind these theories is that abstract value encoding lets the decision maker compare qualitatively different options. Alternatively, expectancy-based theories hold that the brain preferentially tracks the relationship between options and their outcomes, and thus does not abstract away details of offers. To adjudicate between these theories, we examined responses of neurons in six reward regions to risky and safe offers while macaques performed a gambling task. In all regions, responses to safe options are unrelated to responses evoked by equally preferred risky options. Nor does any region appear to contain a specialized subset of value-selective neurons. Finally, in all regions, responses to risky and safe options occupy distinct response subspaces, indicating that the organizational framework for encoding risky and safe offers is different. Together, these results argue against the idea that putative reward regions carry abstract value signals, and instead support the idea that these regions carry information that links specific options to their outcomes in support of a broader cognitive map.
Matching journals
The top 3 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Preferences reveal separable valuation systems in prefrontal-limbic circuits 96%
- The representation and valuation of subgoals in the human brain during model-based hierarchical behavior 96%
- Temporally-specific patterns of neural activity across interconnected corticolimbic structures during reward anticipation 96%
Similar papers in this journal
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.