A computational analysis of the relationship between neuronal and behavioral responses to visual motion

• DOI: 10.1523/jneurosci.16-04-01486.1996
• Publication Date: 2018-04-02T19:35:42Z
• Authors: MN Shadlen; KH Britten; WT Newsome; JA Movshon
• Keywords: Stimulus (psychology)

We have documented previously a close relationship between neuronal activity in the middle temporal visual area (MT or V5) and behavioral judgments of motion (Newsome et al., 1989; Salzman 1990; Britten 1992; 1996). now used numerical simulations to try understand how neural signals MT support psychophysical decisions. developed model that pools responses drawn from our physiological data set compares average different produce The structure allows us assess "neuronal" input simulated performance using same methods we applied real experimental data. sought reconcile three observations: (threshold sensitivity stimuli embedded noise), trial-by-trial covariation response monkey's choices, modest correlation pairs neurons their variable identical stimuli. Our results can be most accurately if decisions are based on at least 100 weakly correlated sensory neurons. composing must include broader range sensitivities than encountered recordings, presumably because inclusion whose optimal stimulus is one being discriminated. Central sources noise degrade signal-to-noise ratio pooled signal, but this degradation relatively small compared with typically carried by single cortical This suggests monkeys base near-threshold populations interacting neurons; these many not tuned optimally for particular