Fig. 1: Computing autocorrelations of spiking activity in V4 columns during fixation and attention tasks.

a In the fixation task (FT), the monkey was rewarded for fixating a central fixation point (FP) on a blank screen for 3 s on each trial. b In the attention task 1 (AT1), monkeys were trained to detect an orientation change in one of four peripheral grating stimuli, while an attention cue indicated which stimulus was likely to change. Monkeys reported the change with a saccade to the stimulus opposite to the change (black arrow). The cued stimulus was the target of covert attention (yellow spotlight), while the stimulus opposite to the cue was the target of overt attention. c In the attention task 2 (AT2), the monkey was rewarded for detecting a small luminance change in one of two grating stimuli, directed by an attention cue. The monkey responded by releasing a bar. The brown frame shows the blank screen in the pre-stimulus period. In all tasks, epochs marked with brown frames were used for analyses of spontaneous activity and epochs marked with orange frames were used for the analyses of stimulus-driven activity. The cue was either a short line (AT1) or two small dots (AT2) indicating the covert attention target. The dashed circle denotes the receptive field locations of recorded neurons (V4 RFs) and was not visible to the monkeys (see Supplementary Fig. 1 for details). d Multi-unit spiking activity (black vertical ticks) was simultaneously recorded across all cortical layers with a 16-channel linear array microelectrode. The autocorrelation of spike counts in 2 ms bins was computed from the spikes pooled across all channels (green ticks). e The autocorrelation (AC) computed from the pooled spikes on an example recording session. Multiple slopes visible in the autocorrelation in the logarithmic-linear coordinates indicate multiple timescales in neural dynamics. Source data for panels d and e are provided as a Source Data file.