Seeing number using texture: How summary statistics account for reductions in perceived numerosity in the visual...

Area of Science:

• Visual perception
• Cognitive neuroscience
• Computational vision

Background:

• Peripheral visual perception involves reduced information due to representational constraints.
• Visual crowding, a consequence of excessive integration in the visual periphery, is often explained by summary-statistic pooling models.
• These models suggest texture-like processing aids information reduction in peripheral vision.

Purpose of the Study:

• To extend summary-statistic models of peripheral vision to explain perceived numerosity.
• To investigate the model's ability to account for reduced numerosity in peripherally viewed dot arrays.
• To explore the model's predictions regarding clustering effects and element shape on peripheral numerosity.

Main Methods:

• Utilized a summary-statistic model of peripheral appearance.
• Examined perceived numerosity in peripherally viewed arrays of randomly placed and clustered dots.
• Assessed the model's predictions concerning element shape's influence on numerosity estimation.

Main Results:

• The summary-statistic model successfully accounted for reduced numerosity in random dot arrays.
• The model did not explain the effects of dot clustering on perceived numerosity.
• The model predicts numerosity estimation is sensitive to element shape in peripheral vision.

Conclusions:

• Summary-statistic models offer a limited explanation for peripheral numerosity perception, particularly concerning clustering.
• The model's prediction regarding shape sensitivity presents a novel aspect of peripheral numerosity phenomenology.
• Further model extensions are needed to encompass a broader range of peripheral visual behaviors.