Turning an object into a scene: buildings activate scene-selective visual cortex independently of visual features

Buildings evoke feature-independent, scene-overlapping neural responses in human visual cortex, with EEG discrimination emerging from 360ms and a ~130ms temporal offset from scene selectivity.

• Why it matters: It remains unclear whether building selectivity reflects shared low-level visual features or instead activates scene representations indirectly, which matters for how ventral temporal cortex is organized.
• What they did: Participants viewed buildings and visually matched boxes while EEG decoding and temporal generalization tested feature dependence; a building-box classifier was also applied to scenes vs chairs, and ultra-fast fMRI (TR=140ms) assessed PPA timing.
• The result: EEG showed discriminable building vs box responses from 360ms, and the building-box classifier generalized to scenes vs chairs; late building-selective activity aligned with earlier scene-selective activity (~130ms offset), and PPA responses were delayed and prolonged for buildings.

AI summary of the abstract · may contain errors · not medical advice · read the paper.