The anxious entrepreneur who lies awake imagining a failed pitch, a confrontation with a co-founder, or a business collapse is not simply worrying. They are running those scenarios through neural circuits that respond to them as if they are happening — producing cortisol, training threat responses, and updating memory representations of how the world works. The distress is real not because the imagination is powerful in some metaphorical sense, but because the brain’s perceptual and learning systems cannot fully distinguish between vivid imagination and actual experience.

The neural architecture that makes imagination feel real

Neuroimaging studies have consistently found that when participants visualise an object, place, or scenario, the cortical regions that process actual visual perception activate — not identically, but substantially and in the same functional networks. More than two decades of evidence established that visual mental imagery activates early and mid-level visual cortex, as well as the higher-level temporal regions associated with object recognition, in patterns that significantly overlap with those produced by actual perception of the same stimuli.

The most mechanistically precise recent evidence comes from Dijkstra and colleagues’ 2025 fMRI study with 26 participants, which identified the specific brain region that monitors whether a signal is imagined or real: the fusiform gyrus. Activity in this mid-level visual region was the key determinant of whether participants categorised an experience as real or imagined. When imagery was sufficiently vivid to produce fusiform activity levels comparable to those generated by actual perception, participants made reality errors — treating the imagined experience as a real one. The frontal network encoded the binary judgment of reality, but that judgment was downstream of the sensory signal strength. The architecture is not a malfunction. It is the same capacity for internal simulation that enables planning, creativity, and future thinking. Its cost is that sufficiently vivid imagined scenarios produce sensory signals strong enough to trigger the same downstream emotional, physiological, and learning responses as real events.

Imagined threat conditions real fear responses — and imagined safety removes them

Reddan, Wager and Schiller’s 2018 controlled study produced the most direct demonstration of imagination’s equivalence to real experience in the threat domain. Participants underwent real auditory fear conditioning — a neutral tone was paired with an unpleasant shock until it reliably produced a measured fear response. They were then randomised to three extinction conditions: real extinction (hearing the tone without the shock), imagined extinction (vividly imagining hearing the tone without the shock), or a control condition. Imagined and real extinction were equally effective in reducing threat-related neural patterns and physiological responses when participants were subsequently re-exposed to the real conditioned stimulus.

The neural mechanism was specific: imagined extinction, like real extinction, engaged the ventromedial prefrontal cortex as a central hub — the same vmPFC that real extinction learning requires. The amygdala’s threat response was down-regulated by imagined exposure through the same neural pathway as actual exposure. The finding establishes the bidirectionality of the mechanism with unusual clarity. Just as imagined safety can reduce a real fear response, imagined threat scenarios can produce and maintain real fear responses. The anxious person who repeatedly imagines a catastrophic outcome is conditioning their threat response system through the same associative learning mechanisms that real aversive experiences use.

Imagination updates memory as if events actually happened

Research by Benoit, Paulus and Schlichting, published in 2025, demonstrated that imagining interactions with another person changes how the brain represents that person in memory and how participants subsequently evaluate and behave toward them — in ways that parallel the effects of real social experiences. Participants who imagined positive interactions with a stranger liked that stranger more and showed changed neural memory representations of them. The hippocampal-cortical memory encoding mechanisms engaged during imagined interactions overlapped substantially with those engaged during real interactions. The researchers summarised the finding directly: “You can paint the world black just by imagining it.”

The implication for anxiety is specific: the anxious person who repeatedly imagines social rejection, business failure, or catastrophic outcomes is not merely experiencing distress in the moment. They are updating their hippocampal-cortical memory representations of those scenarios, building expectation structures that treat the imagined outcomes as experienced data about how the world works. The catastrophic imagination is active learning that makes the feared outcome feel more familiar, more probable, and more real — because neurologically, it has been experienced, repeatedly.

Mental practice improves performance — and mental rehearsal of failure degrades it

A meta-analysis of 35 studies with over 3,200 participants found that mental practice — deliberately imagining performing a physical or cognitive skill — improved subsequent actual performance, with an effect size of d = 0.48. The mechanism established in subsequent neuroimaging work is that motor imagery activates the supplementary motor area, premotor cortex, and cerebellum — the same regions involved in actual motor execution — producing functional strengthening of the neural circuits the skill requires.

The inversion of this finding for anxiety is direct. If imagining performing a skill improves actual performance through shared neural activation, then imagining performing a failure degrades the neural circuits associated with confident execution. The anxious entrepreneur mentally rehearsing a catastrophic pitch is not preparing for difficulty — they are functionally practising the neural patterns associated with the feared outcome, making those patterns more accessible and more likely to activate under real performance conditions.

The same mechanism is the solution

Clinical research on imagery rescripting established that deliberately modifying the content of distressing intrusive imagery — changing the imagined outcome of a feared scenario rather than trying to suppress the imagery — produces significant reductions in emotional distress and anxiety symptom severity. The mechanism is the same as the Reddan and colleagues imagined extinction finding: deliberately constructing an alternative imagined outcome engages the vmPFC-amygdala extinction pathway, reducing the threat response associated with the original conditioned stimulus.

This is the architecturally precise version of what “thinking positively” loosely gestures at. Attempting to suppress a catastrophic image does not engage the extinction pathway — it requires active effort and does not produce durable amygdala down-regulation. Deliberately constructing and vividly simulating an alternative outcome does engage the pathway, because it uses the same perceptual learning mechanism that real exposure would activate. Imagination is not only the problem. It is the treatment — using the same neural machinery in a different direction.

Virtual reality exposure therapy confirms this architecture in clinical practice. Patients with specific phobias and social anxiety who undergo exposure in explicitly artificial virtual environments — which they know are not real — show genuine physiological fear responses and, following repeated exposure, genuine extinction that generalises to real-world contexts. The brain’s threat response system does not require the patient to believe the scenario is real. It requires only sufficient sensory signal strength to cross the threshold identified by the fusiform gyrus research. The VR environment crosses that threshold even when the patient knows it is not real, because the reality-monitoring judgment is downstream of the sensory signal, not upstream of it.

Book worth reading on this

Anxious by Joseph LeDoux is the most neurologically precise treatment of why imagination triggers the same threat circuits as real experience. LeDoux’s distinction between the survival circuit — the system that processes threats, real and imagined, before conscious awareness is involved — and the conscious experience of fear is the foundational conceptual tool for understanding why anxiety cannot be resolved by the simple instruction to stop worrying. The threat system is not responding to conscious belief about whether something is real or dangerous. It is responding to signal strength in the perceptual circuits. For any entrepreneur trying to understand why anxiety persists despite rational knowledge that the feared outcome is unlikely, LeDoux provides the neurological account that makes sense of the experience.

If the dynamics described here are significantly affecting your wellbeing, speaking with a psychologist is the right next step. UK: Samaritans (116 123, free, 24/7). Mind (0300 123 3393). BACP: bacp.co.uk/search/Therapists. Crisis Text Line — text HOME to 741741 (US, UK, Canada, Ireland). International: internationaltherapistdirectory.com.

This article is for educational and informational purposes only. Sources: Dijkstra, N. et al. (2025), A Neural Basis for Distinguishing Imagination from Reality, Neuron. Kosslyn, S.M., Thompson, W.L. & Ganis, G. (2006), The Case for Mental Imagery, Oxford University Press. Reddan, M.C., Wager, T.D. & Schiller, D. (2018), Attenuating Neural Threat Expression with Imagination, Neuron, 100(4), 994–1005. Benoit, R.G., Paulus, P.C. & Schlichting, M.L. (2025), Imagined Social Experiences Update Memory Representations, CU Boulder/Max Planck Institute. Schacter, D.L., Addis, D.R. & Buckner, R.L. (2007), Remembering the Past to Imagine the Future, Nature Reviews Neuroscience, 8(9), 657–661. Holmes, E.A. & Mathews, A. (2010), Mental Imagery in Emotion and Emotional Disorders, Annual Review of Clinical Psychology, 6, 163–187. Driskell, J.E., Copper, C. & Moran, A. (1994), Does Mental Practice Enhance Performance?, Journal of Applied Psychology, 79(4), 481–492. Arntz, A. (2012), Imagery Rescripting as a Therapeutic Technique, Journal of Experimental Psychopathology, 3(2), 189–208. LeDoux, J. (2015), Anxious, Viking. van der Kolk, B. (2014), The Body Keeps the Score, Viking. Burns, D. (1980), Feeling Good, William Morrow.