Explicit and implicit depth-cue integration: Evidence of systematic biases with real objects

Area of Science:

• Visual Perception
• Computational Neuroscience
• Human Factors

Background:

• Previous virtual reality (VR) studies suggested depth perception follows a superadditivity rule, where adding depth cues increases perceived depth.
• The Intrinsic Constraint (IC) model quantitatively explains superadditivity in cue integration.
• Uncertainty remained whether superadditivity in VR was a genuine perceptual phenomenon or an artifact of artificial displays.

Purpose of the Study:

• To investigate if the superadditivity effect in 3D shape estimation generalizes from virtual reality to real-world settings.
• To test the validity of the Intrinsic Constraint (IC) model in naturalistic environments.
• To differentiate genuine depth cue integration from potential biases in virtual environments.

Main Methods:

• Participants judged the 3D shape of real cardboard prisms using a matching task.
• A reach-to-grasp task was included to assess action-based depth perception, analyzing in-flight grip size.
• A novel technique independently controlled binocular and monocular 3D cues for real objects.

Main Results:

• A significant superadditivity effect was observed in both the shape judgment and reach-to-grasp tasks with real objects.
• The magnitude of the observed superadditivity effect was accurately predicted by the Intrinsic Constraint (IC) model.
• The findings demonstrate that superadditivity is a robust feature of human depth estimation.

Conclusions:

• Superadditivity in 3D shape perception is not limited to virtual environments but is an inherent characteristic of human depth estimation.
• The Intrinsic Constraint (IC) model provides a valid framework for understanding depth cue integration in real-world scenarios.
• This research validates the ecological relevance of superadditivity in visual perception.