| Paper: | PS-1B.3 |
| Session: | Poster Session 1B |
| Location: | Symphony/Overture |
| Session Time: | Thursday, September 6, 18:45 - 20:45 |
| Presentation Time: | Thursday, September 6, 18:45 - 20:45 |
| Presentation: |
Poster
|
| Publication: |
2018 Conference on Cognitive Computational Neuroscience, 5-8 September 2018, Philadelphia, Pennsylvania |
| Paper Title: |
Decision-making through evidence integration at long timescales |
| Manuscript: |
Click here to view manuscript |
| DOI: |
https://doi.org/10.32470/CCN.2018.1058-0 |
| Authors: |
Michael Waskom, Roozbeh Kiani, New York University, United States |
| Abstract: |
When multiple pieces of information bear on a decision, the best approach is to integrate the evidence provided by each one. Signatures of evidence integration have been identified in neuronal responses during decision formation, leading many investigations of how the brain produces complex behavior to focus on these computations. Yet, because evidence integration has most often been studied in simple tasks with short timescales of deliberation, it is unknown whether these models can provide a general account of decision-making. Here, we introduce a novel psychophysical paradigm with a long and discontinuous timescale of evidence availability. We show that choice behavior in this task reflects an evidence integration process that can extend over tens of seconds without loss of information due to memory leak or noise. Our results reveal that temporally-extended decisions approximate the normative computations used for rapid sensory discrimination, validating the generality of the evidence integration framework for modeling human cognition. |
