Figure 2.

Firing rates during repetition of natural stimuli

(A) Firing-rate responses averaged over all sites of monkey K in one example session. Responses to different stimuli are shown with different symbol-color combinations.

(B) Black line: responses as in (A) but averaged over all stimuli, sessions, sites, and animals. Green line: same analysis after removing all immediate repetitions of a given stimulus. Vertical line separates “early” and “late” repetitions. Inset: repetition-related linear slope for early (red) and late (cyan) trials (bars show permutation-based two-sided significance thresholds for p < 0.01).

(C) Same as black line in (B), with separate lines for each stimulus and animal (line color approximates stimulus color).

(D) Grand average time-resolved firing-rate responses for the indicated repetitions. Inset: zoom on first 250 ms.

(E) Repetition-related slopes for early (red) or late (cyan) trials as function of peri-stimulus time.

(F) Scatterplot of repetition-related response change (slope) as a function of response strength (intercept), for one example animal. Each circle corresponds to the combination of a recording site with a visual stimulus. Dashed line indicates linear fit.

(G) Spearman correlation coefficients, averaged over animals, separately for slopes/intercepts fit to all, early, and late trials, respectively, computed as in (F).

(B–E) All error regions indicate bootstrap estimates of ±2 SEM (see STAR Methods), shown for illustration, whereas statistical inferences were based on non-parametric permutation tests, including correction for multiple comparison if necessary.