Aphantasia research topples the 'reverse vision' theory of the mind's eye

A new review in Consciousness and Cognition argues that the dominant model of mental imagery — Joel Pearson’s 2019 idea that visualizing something runs ordinary perception backward, repainting the picture onto the primary visual cortex at the back of the brain — no longer holds up. The authors, two of whom cannot form mental images themselves, marshal clinical and imaging evidence against it. Patients whose primary visual cortex was destroyed can still summon images, brain-imaging meta-analyses find visualization reliably lights up the prefrontal cortex and a face-and-word-recognition strip in the left temporal lobe while the rear visual cortex often stays quiet, and one architect lost his imagery after a stroke to that temporal region rather than the back of the brain. The team allows only a demoted role for the primary visual cortex, as a dimmer switch that sharpens an image assembled elsewhere.

In place of the backward-seeing account, the authors propose that a mental image is an emergent property of several regions — prefrontal cortex, the fusiform strip, and a movement-planning area near the top of the head — firing in coordination, like a chord that needs multiple strings sounding at once. Because such a network can break in many places, the framework explains why some people with aphantasia lack only pictures while others report no inner sound, smell, taste, or touch at all, and why some still dream in images. It also reframes claims of hidden or ‘subliminal’ imagery: since imagery means the felt experience itself, unconscious images are treated as a contradiction in terms.

The condition shows measurable physical signatures — people with aphantasia don’t sweat at scary text the way others do (though they react normally to scary pictures), and they show weaker connectivity between distant brain regions — suggesting words normally trigger the body’s fear response by way of an image. The authors, who describe experiencing spatial knowledge as ‘facts’ rather than pictures, caution against tests that treat skills like Tetris or mental rotation as proof of visualization, since those can be solved by comparing features rather than seeing. Their broader point for the field: there is still no objective test for aphantasia, and researchers should stop pretending otherwise.