First- and second-order transformational apparent motion rely on common shape representations

Area of Science:

• Visual Perception
• Cognitive Neuroscience
• Computational Vision

Background:

• Transformational apparent motion (TAM) is an illusion where observers perceive continuous shape change when one object replaces another at the same location.
• Previous research suggests TAM relies on shape representations, but the level of processing (low-level vs. high-level) remains debated.
• Understanding the basis of TAM is crucial for deciphering visual object recognition and motion perception mechanisms.

Purpose of the Study:

• To investigate whether transformational apparent motion (TAM) perception depends on a common, high-level shape representation.
• To determine if TAM is perceived similarly for first-order objects (defined by luminance contrast) and second-order objects (defined by texture contrast).
• To test the hypothesis that TAM is driven by high-level shape information rather than low-level visual attributes.

Main Methods:

• Comparison of transformational apparent motion (TAM) perception between first-order (luminance-defined) and second-order (texture-defined) stimuli.
• Systematic variation of stimulus onset asynchrony (SOA) to assess the temporal dynamics of TAM perception for both stimulus types.
• Qualitative and quantitative assessment of the perceived motion percept in both first- and second-order TAM displays.

Main Results:

• A compelling motion percept, characteristic of transformational apparent motion (TAM), was observed for second-order displays.
• The perceived motion in second-order TAM displays was comparable to that observed in first-order TAM displays.
• Both first- and second-order TAM exhibited similar patterns of perception across a range of stimulus onset asynchronies.

Conclusions:

• The findings support the high-level shape account of transformational apparent motion (TAM).
• TAM perception appears to be driven by segmentation mechanisms that utilize high-level shape information, independent of low-level visual features like luminance or texture contrast.
• This suggests that the visual system employs abstract shape representations for processing apparent motion, irrespective of the defining visual attribute.