Perception as a hierarchical competition: a model that differentiates imagined, veridical, and hallucinated percepts

Abstract

AbstractMental imagery is a process by which thoughts become experienced with sensory characteristics. Yet, it is not clear why mental images have diminished sensory features compared to veridical images, nor how mental images are phenomenologically distinct from hallucinations, another type of non-veridical sensory experience. Current evidence suggests that imagination and veridical perception both depend on similar neural systems. If so, they should interfere with one another. Here we propose that considering how externally-generated stimuli (i.e. sensory input) and internally-generated stimuli (i.e. thoughts) might compete can sufficiently differentiate veridical, imaginary, and hallucinatory perception. To predict the perceptual consequences of competition, we constructed a computational model of a serially-connected, hierarchical system with bidirectional information flow, emulating the primate visual system. We simulated mental imagery as high-level generative feedback, both with and without competition from bottom-up sensory input. Without competition, imagined stimuli dominated representation at each level of the hierarchy. Yet, when competing with sensory input, imagined stimuli were outcompeted at lower levels while remaining dominant at higher levels, with an equilibrium point in between. We interpret our findings under a framework whereby low-level, high-dimensional stimulus information (e.g. in the early visual cortex) contributes most to the sensory aspects of perceptual experience, while high-level, low-dimensional stimulus information (e.g. towards temporal regions) contributes most to the abstract aspects of perceptual experience. Our findings suggest that ongoing bottom-up sensory input during waking life may prevent imagined stimuli from overriding veridical sensory experience. In contrast, internally-generated stimuli may be hallucinated when sensory input is dampened or eradicated. Our model can explain individual differences in mental imagery quality, as well as aspects of daydreaming, hallucinations, and mental imagery in multiple sensory modalities.