A COMPUTATIONAL ANALYSIS OF THE RELATIONSHIP BETWEEN NEURONAL AND BEHAVIORAL-RESPONSES TO VISUAL-MOTION

We have documented previously a close relationship between neuronal ac tivity in the middle temporal visual area (MT or V5) and behavioral ju dgments of motion (Newsome et al., 1989; Salzman et al., 1990; Britten et al., 1992; Britten et al., 1996), We have now used numerical simul ations to try to understand how neural signals in area MT support psyc hophysical decisions, We developed a model that pools neuronal respons es drawn from our physiological data set and compares average response s in different pools to produce psychophysical decisions. The structur e of the model allows us to assess the relationship between ''neuronal '' input signals and simulated psychophysical performance using the sa me methods we have applied to real experimental data. We sought to rec oncile three experimental observations: psychophysical performance (th reshold sensitivity to motion stimuli embedded in noise), a trial-by-t rial covariation between the neural response and the monkey's choices, and a modest correlation between pairs of MT neurons in their variabl e responses to identical visual stimuli. Our results can be most accur ately simulated if psychophysical decisions are based on pools of at l east 100 weakly correlated sensory neurons. The neurons composing the pools must include a broader range of sensitivities than we encountere d in our MT recordings, presumably because of the inclusion of neurons whose optimal stimulus is different from the one being discriminated. Central sources of noise degrade the signal-to-noise ratio of the poo led signal, but this degradation is relatively small compared with the noise typically carried by single cortical neurons. This suggests tha t our monkeys base near-threshold psychophysical judgments on signals carried by populations of weakly interacting neurons; these population s include many neurons that are not tuned optimally for the particular stimuli being discriminated.