What is it about?
When you picture an apple in your mind, what is your brain actually doing? The dominant assumption has long been that mental imagery works like perception in reverse — that signals flow "backwards" down the visual system, firing up neurons in early visual areas the same way actually seeing an apple would. This paper challenges that assumption. Reviewing the neurophysiological and behavioural evidence, the authors argue that feedback connections in the brain — the pathways that carry signals from higher to lower brain areas — are unlikely to generate the kind of strong neural firing ("spiking") in early visual areas that real vision does. Instead, they propose a new framework: the spontaneous activity reshaping hypothesis. Rather than switching neurons on, mental imagery works by selectively switching neurons off — suppressing activity in neurons that code for features not relevant to the imagined object, effectively carving a mental image out of the brain's constant background neural "chatter." Think of it less like drawing a picture, and more like sculpting one by removing what doesn't belong.
Featured Image
Photo by Milad Fakurian on Unsplash
Why is it important?
Understanding how the brain generates mental images matters far beyond basic neuroscience. Mental imagery is central to memory, planning, creativity, and emotional processing — and is increasingly used in clinical interventions for conditions ranging from PTSD to phobias. This new framework resolves a long-standing puzzle: why mental images feel less vivid and more fleeting than real perception. It also offers a fresh explanation for the full spectrum of imagery ability — from aphantasia (the inability to form mental images) to hyperphantasia (unusually vivid imagery) — suggesting these differences may reflect variations in the brain's inhibitory control rather than its excitatory drive. The hypothesis generates concrete, testable predictions that can be examined with non-invasive brain imaging techniques, opening new directions for research.
Perspectives
This work invites a rethink of a foundational assumption in cognitive neuroscience. The brain has long been characterised as a prediction machine, constantly sending top-down signals to refine perception — but the role those signals play at the earliest stages of visual processing has remained contested. By proposing that imagination operates through inhibition rather than excitation in early visual areas, this paper aligns mental imagery research with emerging models of how feedback shapes brain function more broadly. A deeper implication concerns the question of control over our own mental images. If imagery is sculpted from the brain's spontaneous background activity — rather than generated purely by top-down intent — then what we picture in our minds is never entirely of our own choosing. Our volition shapes the image, but the restless, self-generated activity of early visual areas shapes it too. This helps explain why mental images feel unstable and hard to hold steady, and why unwanted visual thoughts can arise and persist against our wishes — as in the intrusive imagery experienced in PTSD and related conditions. More broadly, it raises profound questions about the nature of mental content and agency: if the raw material of imagination is spontaneous neural noise, the boundary between what we choose to think and what simply arises in the mind may be far blurrier than we assume.
Roger Koenig-Robert
University of New South Wales
Read the Original
This page is a summary of: Spiking the mind: Rethinking the role of cortical feedback in visual mental imagery., Psychological Review, April 2026, American Psychological Association (APA),
DOI: 10.1037/rev0000621.
You can read the full text:
Contributors
The following have contributed to this page
