Here the authors derive the mathematical relationship among the key ingredients of the standard neural decision-making model: choice probabilities, read-out weights and correlated variability. This allows them to infer decoding strategies from experimentally measurable quantities and to test whether the organism is using an optimal decoding strategy for a given task, even without knowing the underlying correlations.
- Ralf M Haefner
- Sebastian Gerwinn
- Matthias Bethge
