FIG. 4.

(Color online) Deconvolved pupil size (mean ± 1 standard error across subjects) for (a) reverberant versus anechoic trials, (b) talker gender-match versus -mismatch trials, and (c) maintain- versus switch-attention trials, with trial schematics showing the time course of stimulus events (compare to Fig. 1). Hatched region shows temporal span of statistically significant differences between time series. The onset of statistically significant divergence (vertical dotted line) of the maintain/switch conditions is in close agreement with the end of the cue. a.u. = arbitrary units (see Sec. II A 5 for explanation of effort).