Abstract
High-level face identity aftereffects require awareness (e.g., Moradi et al., 2005), whereas low-level tilt aftereffects occur without awareness (e.g., He et al., 2001). Here we demonstrate that intermediate-level aspect-ratio aftereffects require awareness. During adaptation, we presented an ellipse with a tall or flat aspect ratio to one eye. In the unaware condition, a dynamic-masking pattern was dichoptically presented to prevent awareness of the adaptor ellipse. In the aware (control) condition, the dynamic-masking pattern was monoptically superimposed over the adaptor ellipse so that both were visible. This control condition allowed us to determine the degree to which preventing awareness reduced aspect-ratio aftereffects over and above local masking effects. During adaptation (2000 ms), participants reported the aspect ratio of the adaptor ellipse (if it was visible) so that we could verify the awareness manipulation. After adaptation, participants reported the aspect ratio of a briefly-flashed (73 ms and backward-masked) test ellipse using the method of adjustment. In the aware condition, the aspect ratio of the flashed ellipse appeared distorted away from that of the adaptor (e.g., adaptation to a flat ellipse made a circle appear tall). No aftereffect occurred in the unaware condition. Lack of awareness rather than low-level local masking is likely to be the crucial factor because local inhibitory interactions in V1 would have been stronger in the monoptic-masking than dichoptic masking condition (at least for neural spike rates; Macknik & Martinez-Conde, 2004). Furthermore, these aftereffects arise from adaptation in global aspect-ratio coding and not local curvature coding, because they showed substantial binocular transfer, and no aftereffects occurred for the component curved segments using the same paradigm. These results suggest that adaptation of aspect-ratio coding requires high-level and/or recurrent processes that generate conscious awareness, similar to face identity coding and different from local orientation coding.
NSF BCS0643191, NIH R01EY018197-02S1.