Integration of contour and surface information in shape detection

Area of Science:

• Visual perception
• Cognitive psychology
• Computational neuroscience

Background:

• Shape perception research often isolates contour detection from surface segregation.
• Stable shape perception requires integrating contour and surface information.
• The interplay between contour and surface cues in shape detection is not fully understood.

Purpose of the Study:

• To investigate how human observers integrate contour and surface information for shape detection.
• To quantify the benefit of combining contour and surface cues in a visual search task.
• To compare empirical results with models of sensory cue combination.

Main Methods:

• Used a two-alternative forced-choice (2AFC) task with arrays of oriented Gabor elements.
• Manipulated shape saliency based on contour alignment and interior Gabor isolinearity.
• Measured shape detectability using contour cues, surface cues, and combined cues.

Main Results:

• Demonstrated a significant double-cue benefit, indicating observers utilize both contour and surface information.
• Shape detection performance with combined cues exceeded that predicted by a minimum rule (probability summation).
• Observed performance was consistent with an integration rule (information summation), with no evidence against it.

Conclusions:

• Human visual system effectively integrates contour and surface cues for robust shape perception.
• Cue combination in shape detection follows a principle more complex than simple probability summation.
• Findings support models of information integration in sensory processing.