Interaction of bottom-up and top-down processes in the perception of ambiguous figures

Abstract

Ambiguous figures reverse their appearance during prolonged viewing and can be perceived in two (or more) available interpretations. Both physical stimulus manipulations and cognitive control influence the perception of ambiguous figures, but the underlying mechanisms remain poorly understood. In the current study, the perception of an ambiguous figure was manipulated by adaptation to unambiguous figures and/or placing the ambiguous figure into a context of unambiguous figures. Our results indicate that both adaptation and context can effectively modulate perception of the ambiguous figure. When manipulated together, adaptation and context processes showed additive effects upon the perception of the ambiguous figure implying the independent mechanisms. Thus, top-down and bottom-up processes seem to influence the perception of the ambiguous figures independently and neither seems to be uniquely responsible for the generation of perceptual changes.

Circular inference in bistable perception

“…Consistent with the present study, some authors have focused on how these various effects are combined [51][52][53]. Moreno-Bote et al showed that cue combination in a bistable display can be well explained by a multiplicative law (their predictions are similar to the NB model described here) [54], whereas Zhang and colleagues demonstrated that different types of priors are effectively combined [4].…”

Common contextual influences in ambiguous and rivalrous figures

“…A good example of this is the study by Sobel and Blake [12] described in the Introduction. On the other hand in previous ambiguous figure studies, the spatial context was an independent figure, albeit one that also biased the target’s percept towards that of the context [10]; [11]). In the present study, we demonstrated that an independent context, i.e.…”

“…We argue that the high correlations found in this study are due to the overriding influence of the context figures. There are good examples in the literature where contextual influences overcome competing manipulations such as in the study described by [10]. They demonstrated that the influence of spatial context, which biases the interpretation of an ambiguous figure towards that of context, overcame any adaptation effects which on their own bias the interpretation of the target figure away from that of the adaptor figure.…”

“…Not surprisingly, simultaneous contextual influences also alter the perceptual interpretation of both ambiguous and rivalrous figures [3,10–12], and are thus considered to be valuable tools for probing their underlying mechanisms [13]. Intaitė et al [10], using a display of overlapping skeleton squares which alternated between being perceived as stacked upwards or stacked downwards, showed that when the squares were flanked by unambiguous coherently stacked versions, observers’ first percepts were biased towards that of the flanking context.…”

“…Intaitė et al [10], using a display of overlapping skeleton squares which alternated between being perceived as stacked upwards or stacked downwards, showed that when the squares were flanked by unambiguous coherently stacked versions, observers’ first percepts were biased towards that of the flanking context. The influence of spatial context on ambiguous figures has also been revealed through the introduction of backgrounds favoring a particular percept [14], or when the context itself is ambiguous [15].…”