Fig. 2: Psychophysical results.
A Audiovisual sequences. Stimuli were sequences of five clicks and flashes whose temporal statistics were manipulated so as to optimize the model predictions for the temporal correlation and for the temporal lag (see Methods). B The MCDCORR activity significantly predicts the causality judgments in the causality judgment task (mixed-effect logistic regression, p < 10−15). C The MCDLAG activity significantly predicts the temporal order judgments in the temporal order judgment task (mixed-effect logistic regression, p < 10−15). Each point corresponds to a different audiovisual stimulus. Given that the same six audiovisual stimuli were used for both tasks, points with the same colors in the left and right panels represent the participants’ responses to the same stimuli in the two tasks. Error bars represent 2 s.e.m. (standard error of the mean, N = 13). Note that values of MCDCORR and MCDLAG differ from participant to participant because of the fitting procedure, but the error bars are smaller than the size of the points.
