Making ambiguous displays unambiguous: The influence of real colors and colored aftereffects on perceptual alternation

Abstract

The relationship between orientation-contingent colored aftereffects (CAEs) and perceptual alternation of ambiguous displays was investigated in three experiments. In all experiments, the ambiguous test display consisted of vertical and horizontal contours that either could be perceived as separate surfaces (the diamond organization) or could be combined to form upright and inverted Us (the rectangles organization). In Experiments 1 and 2, observers inspected the test display when it was achromatic, when it was colored in a manner consistent with the appearance of CAEs, and when it was colored in a non-CAE manner. In Experiment 3, the test display was inspected monocularly before and after monocular color-orientation adaptation. The achromatic viewing conditions of Experiments 1 and 2, both monocular preadaptation achromatic viewing conditions of Experiment 3, and the postadaptation achromatic viewing condition involving the nonadaptation eye in Experiment 3 produced essentially the same results: reports of approximately equal duration for the diamond and rectangles organizations and steady rates ofperceptual alternation. The non-CAE color conditions of Experiment 2 produced a similar pattern of results. However, the color condition of Experiment 1 (i.e., simulated CAEs), and the postadaptation condition involving the adaptation eye in Experiment 3 produced a decrement in the perceptual alternation rate, with the diamond organization being reported almost exclusively. These results indicate that ambiguous displays can be made unambiguous by altering the appearance of the displays either through the addition of appropriate colors or by exposing observers to appropriate adaptation conditions. These results are consistent with stimulus-bound explanations of both CAEs and perceptual alternation.The two displays in Figure 1 are perceptually ambiguous. That is, although the physical properties of the displays (e.g., the orientation, contrast, and number of contours within the displays) do not change during a period of observation, perception of the displays does change. On some occasions, perception of Figure la (the Schroeder staircase) is such that the point farthest from the observer appears to be the upper right-hand comer. On other occasions, the perceived staircase is such that this same point appears to be the nearest point relative to the observer. Similarly, Figure ib may be seen as consisting of a centrally located, diamond-shaped figure made up of horizontal bars superimposed on a background of A preliminary presentation of the results reported here was made at

Eye movements and the associative basis of contingent color aftereffects: A comment on Siegel, Allan, and Eissenberg (1992).

“…McCollough (1965a) noted that CAEs induced with vertical (and horizontal) edges generalized to the vertical (and horizontal) orientation components of different complex stimuli, the apparent color of vertical components in a pattern of concentric circles being the same as the apparent color of vertical components in a pattern of radial lines. Considered in conjunction with other investigations of CAEs and pattern organization (e.g., Foreit & Ambler, 1978), McCollough's observations are consistent with subsequent conclusions that CAEs are contingent on orientation components and not on higher order (spatiotopic) relationships between these components (Broerse & Crassini, 1986). In these terms, simple stimulus dimensions such as contour orientation (or direction of motion; see Stromeyer, 1978) have enjoyed a privileged status in theoretical attempts to account for these phenomena (e.g., neural adaptation of edge-sensitive mechanisms).…”

Scanning and form-contingent color aftereffects.

“…We indeed presented our subjects with no instruction as to viewing mode, and did not report this fact. It would be tedious to elaborate everything the subject is not told, and CAE experiments frequently do not report whether subjects were (or were not) told to scan (or fixate; e.g., Broerse & Crassini, 1986; McCollough, 1965).…”

“…We indeed presented our subjects with no instruction as to viewing mode, and did not report this fact. It would be tedious to elaborate everything the subject is not told, and CAE experiments frequently do not report whether subjects were (or were not) told to scan (or fixate; e.g., Broerse & Crassini, 1986;Mc-Collough, 1965). Humphrey et al (1994) "speculate that [the Siegel et al, 1992] subjects may have tended to fixate at the same position on the patterns during induction and the posttest" (p. 89).…”

A functional theory of the McCollough effect.

“…S. Harris, 1980; Over, 1977; Skowbo, Timney, Gentry, & Morant, 1975; Stromeyer, 1978). Attempts have been made to show that MEs may depend on perceptual organization (Jenkins & Ross, 1977; Uhlarik, Pringle, & Brigell, 1977), on imagery (Finke & Schmidt, 1977, 1978), and on other cognitive factors; such claims have mostly not been verified (Broerse & Crassini, 1981a, 1981b, 1984, 1986). One cannot be dogmatic, but on present evidence MEs are best explained in terms of a neuropsychological mechanism that is biased by the deliberate pairing of pattern with color without the intervention of “higher,”cognitive principles.…”