Visual illusions as a tool to hijack numerical perception: Disentangling nonsymbolic number from its continuous...

Area of Science:

• * Cognitive Psychology
• * Visual Perception
• * Psychophysics

Background:

• * Ongoing debate exists regarding the visual mechanisms underlying nonsymbolic number processing.
• * Previous models proposed quantity extraction via primitive visual segmentation, but recent theories suggest low-level features (e.g., density, area) are confounded with numerosity.
• * Disentangling discrete units from continuous variables in visual numerosity is crucial for understanding perception.

Purpose of the Study:

• * To investigate whether visual numerosity processing operates on discrete units or continuous variables using visual illusions.
• * To differentiate the roles of illusory contours (ICs) from other low-level visual features in numerosity perception.
• * To determine if findings are task-dependent.

Main Methods:

• * Stimuli were created using open inducers to generate Kanizsa-like illusory contours (ICs), simulating connections between objects.
• * Experiments manipulated the number of ICs (0, 2, or 4) while controlling low-level visual features.
• * Participants performed tasks including selecting the larger set and estimating numerosity.

Main Results:

• * Increasing IC connections systematically led to underestimation of numerosity in a comparison task (Experiment 1).
• * This effect was absent with closed inducers that prevented illusory lines (Experiments 2A, 2B).
• * Numerosity underestimation persisted even when ICs and continuous features (convex hull) were independently manipulated (Experiment 3) and across different tasks (Experiment 4).

Conclusions:

• * Findings suggest numerosity perception relies on basic visual-segmentation mechanisms, particularly those involving discrete units.
• * Illusory contours significantly influence numerosity judgments, indicating a role for discrete perceptual signals.
• * Future theoretical frameworks must integrate both continuous and discrete perceptual number signals to fully explain numerosity perception.