How Brain Stimulation Reveals the Secrets of Visual Imagination
A flash of light appears without any light actually entering your eyes. This curious phenomenon, known as a phosphene, is helping scientists unravel one of the most enduring mysteries of human consciousness.
Imagine closing your eyes and picturing a loved one's face. The detailed image that forms in your mind, despite no light actually reaching your eyes, represents one of the most fascinating capabilities of the human brain. For centuries, philosophers and scientists have wondered how we can generate such vivid mental images without any visual input. Today, using an innovative technology called transcranial magnetic stimulation (TMS), researchers are beginning to unravel this mystery. Through the study of phantom light flashes called phosphenes induced by TMS, we're gaining unprecedented insights into the neural machinery behind visual mental imagery—and discovering that the line between seeing and imagining is surprisingly thin.
Transcranial magnetic stimulation (TMS) is a non-invasive brain stimulation technique that uses magnetic fields to safely and temporarily activate specific brain regions. When a TMS coil is placed against the scalp, it generates brief magnetic pulses that pass through the skull and induce small electrical currents in the underlying brain tissue.
When TMS is applied to the visual cortex at the back of the brain, people often report seeing flashes of light or geometric patterns, even though their eyes are closed. These perceptions are called phosphenes—visual experiences that occur without actual light entering the eye.
The sensory recruitment framework proposes that when we form mental images, we're essentially "seeing" in the absence of visual input by reactivating our visual system 4 .
Research has shown that phosphenes are most reliably induced when stimulating early visual areas (V1, V2) and dorsal visual areas (V3d, V3a) close to the interhemispheric cleft 7 .
| Brain Area | Location | Phosphene Susceptibility | Typical Phosphene Location |
|---|---|---|---|
| V1 (Primary Visual Cortex) | Medial occipital lobe | High | Contralateral visual field |
| V2d (Dorsal V2) | Dorsal medial occipital | High | Lower contralateral visual field |
| V2v (Ventral V2) | Ventral medial occipital | High | Upper contralateral visual field |
| V3d/V3a | Dorsal occipital | Moderate to High | Lower visual field/periphery |
| hMT+/V5 | Lateral occipital | Low | Variable, often moving phosphenes |
| LOC (Lateral Occipital Complex) | Lateral occipital | Low | Complex forms (rare) |
The visual cortex is organized retinotopically, meaning specific areas correspond to different parts of the visual field. TMS stimulation of these areas produces phosphenes in corresponding locations.
In 2002, a pioneering TMS study conducted by researchers in Germany provided some of the first causal evidence that visual mental imagery directly involves the early visual cortex . This experiment was groundbreaking because it moved beyond simply observing brain activity during imagery tasks to actively testing how stimulating the visual cortex would affect perception during mental imagery.
This study demonstrated that visual mental imagery increases the excitability of the visual cortex, making it more responsive to TMS stimulation.
Landmark Study
Researchers first determined each individual's phosphene threshold (PT)—the minimum TMS intensity needed to reliably generate phosphenes when stimulating their visual cortex.
Participants created detailed visual mental images of various shapes or patterns while researchers delivered TMS at various intensities.
To contrast with visual imagery, participants engaged in a non-visual task involving listening to and processing sounds.
After each TMS pulse, participants reported whether or not they had seen a phosphene, allowing researchers to track changes in phosphene perception thresholds.
The findings were striking. During the visual imagery tasks, participants reported seeing phosphenes at significantly lower TMS intensities compared to their baseline thresholds. This threshold-lowering effect occurred regardless of where in their visual field they positioned their mental images. In contrast, during the auditory control task, no significant change in phosphene threshold occurred.
Significant decrease in phosphene threshold
No significant change in phosphene threshold
| Experimental Condition | Effect on Phosphene Threshold | Interpretation |
|---|---|---|
| Baseline (no task) | Established baseline PT | Individual reference point |
| Visual Imagery Task | Significant decrease in PT | Visual cortex excitability increased |
| Auditory Control Task | No significant change | Effect specific to visual processing |
| Imagery across different visual field locations | Consistent decrease in PT | Effect independent of specific mental image location |
This study provided compelling evidence for the sensory recruitment framework, suggesting that the early visual cortex is not just a passive processor of incoming visual information but actively participates in generating mental images. As one review noted, "TMS does not simply induce a 'virtual lesion' but interacts with cortical state" 6 , highlighting how the brain's current activity level dramatically influences how it responds to stimulation.
The 2002 study was among the first to demonstrate what has become a fundamental principle in TMS research: the effects of stimulation are highly dependent on the initial state of the targeted neural population. When the visual cortex is actively engaged in imagery, its excitability changes, which in turn alters how it responds to TMS 1 6 .
A 2024 study demonstrated that visual adaptation—the reduced responsiveness to constant stimulation—can actually shield exogenous (involuntary) attention from being disrupted by TMS 1 . This suggests that adaptation and attention are interactive processes in the early visual cortex.
| Factor | Impact on TMS Effects | Practical Implications |
|---|---|---|
| Stimulus Onset Asynchrony (SOA) | Optimal suppression at 50-150 ms; effects vary nonlinearly with timing | Critical to carefully time TMS relative to visual stimuli |
| TMS Intensity | Lower intensities may enhance perception; higher intensities typically suppress | Intensity should be calibrated to desired effect |
| Visual Angle/Eccentricity | Larger, peripheral stimuli more susceptible to TMS effects | Stimulus design affects outcomes |
| Cortical State (e.g., during imagery) | Increased excitability lowers phosphene thresholds | Brain state at time of stimulation matters significantly |
| Coil Type and Placement | Different coils have different focal properties; placement affects which neural populations are targeted | Technical parameters significantly influence results |
Meta-analysis of TMS studies shows optimal suppression occurs at specific time intervals after visual stimulus presentation 8 .
Visual neuroscience research using TMS relies on several specialized tools and methodologies:
This personalized measure determines the minimum TMS intensity required to generate phosphenes in an individual .
Used to identify individual visual areas through retinotopic mapping for precise TMS targeting 7 .
Essential for monitoring eye position during experiments, since eye movements affect phosphene perception 5 .
Sometimes used simultaneously with TMS to measure direct brain responses to stimulation 5 .
Computational models that simulate prosthetic vision by predicting phosphene patterns 3 .
The convergence of evidence from TMS studies, particularly research involving phosphenes and visual mental imagery, points to a profound conclusion: the same neural machinery that processes actual visual information is recruited when we generate mental images. The early visual cortex is not merely a camera capturing external reality but an active participant in constructing our internal visual world.
Understanding how to modulate visual cortex excitability has implications for developing visual cortical prostheses for the blind 3 , treating conditions like visual snow syndrome 5 , and rehabilitating visual processing deficits following brain injury.
As one research team noted, we're steadily progressing "towards biologically plausible phosphene simulation for the differentiable optimization of visual cortical prostheses" 3 —a technical way of saying we're learning to speak the visual brain's language well enough to potentially restore sight through artificial means.
The next time you close your eyes and picture a beautiful scene, remember: you're not just imagining—you're briefly activating the very same visual circuits that allow you to see the world, demonstrating that the boundary between perception and imagination is far more permeable than we ever realized.