The mechanisms and neural signature of average numerosity perception

The human brain is endowed with an intuitive sense of number allowing to perceive the approximate quantity of items in a scene, or "numerosity." This ability is not limited to items distributed in space, but also to events unfolding in time and to the average numerosity of dynamic scenes. How the brain computes and represents the average numerosity over time however remains mostly unclear. Here we investigate the mechanisms and electrophysiological (EEG) signature of average numerosity perception. To do so, we used dynamic stimuli composed of 3-12 arrays presented for 50 ms each, and asked participants to judge the average numerosity of the sequence. Our results first show that the weight of different arrays in the sequence in determining the judgement is subject to both primacy and recency effects, depending on the length of the sequence. Moreover, we show systematic perceptual adaptation effects across trials, with the bias on numerical estimates depending on both the average numerosity and length of the preceding stimulus. The EEG results show numerosity-sensitive brain responses starting very early after stimulus onset, and that activity around the offset of the sequence can predict both the accuracy and precision of judgments. Additionally, we show a neural signature of the adaptation effect at around 300 ms, whereby the amplitude of brain responses can predict the strength of the bias. Overall, our findings support the existence of a dedicated, low-level perceptual mechanism involved with the computation of average numerosity, and highlight the processing stages involved with such process.