Pupil dilation deconvolution reveals the dynamics of attention at high temporal resolution

Abstract

The size of the human pupil increases as a function of mental effort. However, this response is slow, and therefore its use is thought to be limited to measurements of slow tasks or tasks in which meaningful events are temporally well separated. Here we show that high-temporal-resolution tracking of attention and cognitive processes can be obtained from the slow pupillary response. Using automated dilation deconvolution, we isolated and tracked the dynamics of attention in a fast-paced temporal attention task, allowing us to uncover the amount of mental activity that is critical for conscious perception of relevant stimuli. We thus found evidence for specific temporal expectancy effects in attention that have eluded detection using neuroimaging methods such as EEG. Combining this approach with other neuroimaging techniques can open many research opportunities to study the temporal dynamics of the mind's inner eye in great detail.

Fig. 1.

The AB task. (A) An example of a dual-target trial, which required the reporting of two unspecified target letters (T1 and T2) in a stream of distractors (digits). At the end of the trial, a comma or dot appeared to keep attention focused on the stream. The task was to report whether a comma or dot had been presented and then to report the targets in order of appearance. However, responses in either order were counted as correct in the analyses. (B) Accuracy scores of T1 and T2|T1 (i.e., T2 accuracy on trials where T1 is correctly reported) as a function of the temporal interval between the targets. A binominal mixed-effects model revealed a substantial decrement in performance at lag 3 compared with performance at lag 15 (P < 0.001), revealing a robust AB effect. Lag 0 refers to performance in the one-target condition.

Fig. 2.

Averages of normalized pupil dilation (A, C, and E) and deconvolved attentional pulses. The underlying attentional pulses (B, D, and E). The strengths of the attentional pulses for each condition of interest were compared using permutation tests. In each panel, the data depicted are time-locked on the presentation of the first target. (A and B) Comparison between no-blink and one-target trials. (C and D) Comparison between no-blink and blink trials. (E and F) Comparison between the one-target and no-target trials. In the blink and no-blink trials, the second target’s onset was always at 300 ms following T1. To correct for multiple comparisons, a single threshold test was used in each permutation test (24). The single threshold statistic t for panels B, D, and F was 2.654, 2.682, and 2.677, respectively. Significant differences (α = 0.05) in the strength of attentional pulses are denoted by asterisks.