Abstract
Introduction: Contextual effects abound in the real world. A classic example of this is the tilt illusion that results in a target’s orientation appearing repelled from that of the surround. However the opposite effect can also occur (the indirect tilt illusion) and the target’s orientation appears shifted towards that of the surround. These orientation biases have revealed much about the underlying detectors involved in visual processing (low-level effects) but what remains at stake is the role of higher-level cognitive influences. Specifically, it is widely believed that the indirect tilt illusion requires additional input from higher cortical areas where the stimulus is consciously encoded. Methods We designed a novel reverse correlation technique to investigate this. The stimulus consisted of a concentric annular surround (outer diameter 7.8°) containing a 2cpd grating that could have one of 12 possible orientations abutting a circular patch (diameter 1.8°) where a vertical 2cpd grating was presented every 2 seconds. The surround was refreshed on each frame (11.7ms) and each orientation had an equal probability of being selected. The observers’ (n=5) task was to report after each target presentation whether it had appeared tilted clockwise (CW) or counterclockwise (CCW) of a subjective vertical. Results All observers displayed a strong direct tilt illusion with this technique. In the majority of the observers (n=4) we also obtain a robust indirect tilt illusion. Both illusions occur reliably and over a similar time course (roughly ± 60ms around the time of target presentation) despite the lack of conscious access to the surround orientation. Conclusion: These results support the role of a single mechanism underlying orientation biases and refute the need for higher-level conscious processing of the context.
Meeting abstract presented at VSS 2012