Abstract
Vertical disparities are used by the visual system to interpret slant from horizontal disparities. Different combinations of horizontal and vertical disparities can produce the same apparent slant. Adaptation to apparently unslanted surfaces produced in this way gives rise to slant aftereffects (Berends & Erkelens, 2001). It is not clear if the mechanism that measures vertical disparity, per se, contributes to this aftereffect. We used a novel stimulus, in which different display elements carried the horizontal and vertical disparity signals, to determine whether adaptation occurs in a mechanism that measures vertical disparity. Horizontal disparity (in the form of horizontal size ratio, HSR) was carried by a horizontal row of dots, and vertical disparity (vertical size ratio, VSR) was carried by horizontal lines above and below the row of dots. Interestingly, for this stimulus, changing the VSR changes the apparent slant of the dot row, without changing the apparent slant of the inducing lines. Observers adapted to stimuli that contained only the lines, not the dot row (and thus, a VSR signal, but not an HSR signal). The adapting stimuli appeared unslanted at all VSR values. When a test stimulus contained both the horizontal dot row and the lines, there was virtually no adaptation aftereffect. However, when the test stimulus contained only the horizontal dot row, there was a large aftereffect; this is striking because the adapting stimulus contained only vertical disparity, and the test stimulus contained only horizontal disparity. One interpretation of these data is based on the observation that eye position, like VSR, can be used to interpret horizontal disparity as slant: prolonged exposure to a discrepancy between eye position and VSR may cause recalibration of the mechanism that uses eye position to estimate slant from horizontal disparity.
This work was supported by NIH grant EY013988 (BTB, IO, HQ) and by the Natural Sciences and Engineering Research Council of Canada (PAD).