Suppression and facilitation of motion perception in humans with single
motion stimulus
Abstract
Research groups use the single motion stimulus design of Dr. D.Tadin
(Tadin, Lappin, Gilroy & Blake, 2003) that allows to putatively assess
spatio-contrast excitatory and inhibitory effects from duration
threshold data of motion perception. The present work presents the
correct neurocomputational model for this experimental design and
analyses issues related to data and model comparisons, among which: (1)
once the full computational model that predicts the psychophysical
results is properly defined, it is shown that two low-level models of
how inhibition acts on neuronal activity, i.e. “divisive inhibition”
and “subtractive inhibition”, predict exactly the same thresholds and
cannot arguably be claimed that one is better than another one, (2) it
is excitatory and inhibitory processes that are the mechanisms shaping
threshold changes, i.e. perceptual “facilitation” and “suppression”
in the behavioural domain, (3) that this experimental design allows a
quantitative comparison and usage of such “contrast–size tuning”
data, (4) that such studies must be carefully designed once the model is
correctly understood and applied, because of the rather large parameter
space (~10-12 variables) necessary to explain the
behavioural measures even in such simple experiments.