Precise attention filters for Weber contrast derived from centroid estimations.

Area of Science:

• Visual perception
• Cognitive psychology
• Computational neuroscience

Background:

• Selective attention allows focusing on specific visual stimuli.
• Understanding attention filters is crucial for visual processing models.

Purpose of the Study:

• To quantify the effectiveness of selective attention to contrast polarity in visual tasks.
• To model observers' attention filters based on dot intensity.

Main Methods:

• Observers estimated dot cloud centroids under different attention instructions (brighter, darker, or all dots).
• Quantitative attention filters were derived based on dot intensity weighting.
• Performance accuracy was compared across attention conditions.

Main Results:

• Attended dots received significantly higher weights than unattended dots (approx. 4x).
• Observers were more accurate when attending to all dots compared to selective attention to one contrast polarity.
• Individual observers exhibited distinct computational parameters for centroid estimation.

Conclusions:

• Selective attention to contrast polarity can reduce accuracy in centroid estimation tasks.
• Attention filters are measurable and comparable to physical filters.
• Individual differences in visual computation exist despite similar attention filter structures.