Binocular interaction: contrast matching and contrast discrimination are predicted by the same model

Area of Science:

• Vision Science
• Neuroscience
• Perception

Background:

• Previous models of binocular vision assumed linear summation of monocular signals.
• Recent work challenged these models, proposing a two-stage approach with early linear transduction and later gain control.

Purpose of the Study:

• To thoroughly investigate the two-dimensional stimulus space of left- and right-eye contrasts.
• To compare contrast discrimination and contrast matching tasks within a unified theoretical framework.

Main Methods:

• Utilized 25 stimuli with varying left and right eye contrasts (0.3-32%) for horizontal gratings.
• Employed a two-alternative forced-choice (2IFC) discrimination task and a contrast matching task.
• Applied a two-stage computational model with fixed parameters from prior research.

Main Results:

• The two-stage model successfully predicted both discrimination and matching data.
• The model demonstrated superior performance compared to linear and quadratic binocular summation models.
• Discrimination performance correlated with the gradient of the internal contrast-response function, while matching related to response magnitude.

Conclusions:

• The two-stage model provides a robust framework for understanding binocular contrast vision.
• Perceptual tasks like contrast discrimination and matching can be explained by the same theoretical model.
• This research refines our understanding of how visual information is processed across both eyes.