Non-drug state-shift design

Brain Candy

Brain Candy explores how rhythmic light, recursive visual structure, sound, breath, expectation, and interaction can be composed into adjustable non-drug experiences. It is not a treatment claim, a psychedelic substitute, or a promise that a frequency will optimize the brain. The project is about tunable experience design and measurement: what changes when light rhythm, visual complexity, audio pacing, context, and user agency change (Hewitt et al., 2025; Taylor et al., 2024; Hartogsohn, 2023).

First principles

What is being designed?

A state shift is a temporary change in the texture of experience: attention narrows or widens, visual space becomes more patterned, time feels more rhythmic, the body becomes more salient, or a mood becomes easier to notice. People already seek those shifts through music, breathing, dance, games, meditation apps, light devices, and visualizers. Brain Candy asks how to build such shifts with explicit controls, logged parameters, and safety limits.

The core idea is deliberately modular. Stroboscopic light means light that changes rhythmically over time. Recursive visual structure means imagery that repeats or transforms patterns across scales. Fractal fluency is the idea that many visual systems process some multiscale patterns with relative ease, especially at moderate complexity (Taylor and Spehar, 2011; Taylor et al., 2024). Cyberdelics is the loose public name for media systems that try to shape mind-manifesting experiences through technology, context, and expectation (Hartogsohn, 2023; Smith and Warner, 2022).

Those layers should not be collapsed into one effect. Flicker can change visual phenomenology. Multiscale structure can change how a scene is parsed. Music can change pacing, emotion, and meaning. Breath can add agency and bodily feedback. Expectation can shape interpretation. The research problem is to keep those pieces separate enough to compare.

What gets tuned

  • Temporal rhythm: flicker frequency, pulse shape, phase, and duration
  • Visual intensity: contrast, color, brightness, density, and motion
  • Spatial structure: symmetry, recursion, fractal dimension, and scale
  • Context: instructions, expectation, setting, music, and exit cues
  • Agency: breath pacing, controls, comfort breaks, and stop conditions

What does not follow

  • A vivid session is not evidence of clinical efficacy.
  • A strobe effect is not the same as a psychedelic drug state.
  • A fractal pattern is not automatically restorative.
  • A binaural beat or haptic pulse does not guarantee entrainment.
  • A public project, tool, or company example is not validation.

Connected work

Rhythmic light

Flicker is a real perceptual probe, not a magic frequency

Stroboscopic visual stimulation can produce color, motion, pattern, depth, and geometric effects in some viewers. Any design claim has to begin from controlled stimulus variation, subjective reports, comfort logs, and safety limits, not from guaranteed outcomes.

What flicker can do

In flicker and ganzflicker experiments, people view rhythmic or rapidly changing light fields and report visual effects that are not simply present in the stimulus: colors, grids, ripples, drifting shapes, tunnels, radial forms, and shifting textures. Recent reviews treat these stroboscopically induced visual hallucinations as a useful laboratory model because the stimulus can be controlled while subjective reports can be collected systematically (Hewitt et al., 2025; Bartossek et al., 2021; Reeder, 2022).

The important word is controlled, not guaranteed. The experience depends on frequency, contrast, rhythm, intensity, color, viewing distance, eyes open or closed, attention, fatigue, individual susceptibility, and the way reports are elicited. Amaya and colleagues, for example, compared rhythmic flicker with an arrhythmic control and linked stronger subjective motion, dynamics, and shape ratings to higher-order visual and thalamocortical effects (Amaya et al., 2025).

Measurement is catching up

A good Brain Candy session cannot be evaluated by "felt intense" alone. Current work is building more specific vocabularies for visual content and altered-state reports. The Six-Dimensional Visual Hallucination Questionnaire targets visual qualities across psychedelics and stroboscopic light, while MOSAIC uses topic modelling and language-model-assisted clustering to analyze free-text reports from Dreamachine-style stroboscopic experiences (Hewitt et al., 2026; Beauté et al., 2026). Public efforts such as CIRCE and the Berlin Symposium on Stroboscopic Light show the same shift toward shared instruments and a more explicit research community (CIRCE; BSSL 2026).

Brain Candy's research model is to vary the stimulus, collect structured reports, log comfort and stop events, and resist the temptation to turn a striking percept into a therapeutic claim.

Visual geometry

Why spirals, tunnels, lattices, and honeycombs keep appearing

Altered visual states often contain recurring geometric motifs. Heinrich Klüver called many of them form constants: lattices, cobwebs, tunnels, spirals, and other organized forms. Modern models do not treat those shapes as arbitrary decoration. They ask how the visual cortex maps the retina, how local excitation and inhibition spread, and how patterned activity might become a perceived geometry (Ermentrout and Cowan, 1979; Bressloff et al., 2002).

Rule, Stoffregen, and Ermentrout make the bridge especially useful for Brain Candy because they model flicker-induced geometric phosphenes directly. Their account links flicker frequency and contrast with cortical resonance, inhibitory stabilization, lateral inhibition, and retino-cortical mapping (Rule et al., 2011). Nearby empirical work also shows that perceived color and form vary with stimulation frequency and that whole percepts and percept features have their own frequency dependencies (Becker and Elliott, 2006; Allefeld et al., 2011).

The design lesson is precise but bounded. Geometric models justify taking spirals, radial fields, tunnels, grids, and symmetry seriously as visual-system phenomena. They do not justify claiming that a designer can dial a single frequency and reliably produce the same inner image in every viewer.

Useful variables

  • Frequency and phase of flicker
  • Contrast and brightness envelope
  • Spatial scale and retinal position
  • Symmetry, radial structure, and tiling
  • Eyes-open versus eyes-closed viewing

Model limits

  • Models simplify biology so that mechanisms can be tested.
  • Reports differ across people and sessions.
  • Frequency effects are real but not one-to-one control codes.
  • Geometric plausibility is not evidence of a whole altered state.

Recursive structure

Fractals are a design vocabulary, not a cure

Brain Candy uses recursive and fractal-like imagery because the visual system is sensitive to scale, symmetry, and repeated structure. The claim should stay that specific.

Why multiscale structure matters

A fractal-like pattern contains structure across scales: large shapes echo medium shapes, which echo smaller details. Natural scenes often have this multiscale organization, and several lines of perception research suggest that people often prefer or process moderate-complexity fractal patterns more fluently than either very sparse or highly chaotic stimuli (Taylor and Spehar, 2011; Taylor et al., 2024). Processing fluency is the broader psychological idea that easier-to-process stimuli can feel more familiar, pleasant, or coherent (Reber et al., 2004).

This makes fractal abstraction a serious design material, especially for VR. Barton and colleagues found that abstract fractal-like VR could maintain restoration comparable to nature VR in their study, and that interactive versions improved focus and motivational value (Barton et al., 2024). That is useful for Brain Candy because it separates visual structure from literal nature imagery. It does not mean that fractals heal the brain.

Restorative abstraction needs comparators

Attention-restoration and biophilic-design literatures are relevant because they give Brain Candy comparison conditions: nature VR, abstract fractal-like VR, plain visualizers, blank rest, rhythmic-light conditions, and interactive versus passive versions (Berto, 2005; Taylor, 2021). The point is not to borrow the warmest wellness language from those fields. The point is to ask which visual features make an experience easier to stay with, easier to leave, or easier to describe afterward.

Comparison, not equivalence

Psychedelic visual science gives useful measures

Psychedelic research matters here because it studies altered visual content in detail: geometry, color, depth, motion, imagery, contextual perception, pattern completion, and the felt meaning of sensory change. Aqil and Roseman argue that sensory dimensions help structure psychedelic experience rather than merely decorating it (Aqil and Roseman, 2023). Newer work on psilocybin and visual contextual computation gives a stronger way to discuss altered visual inference without claiming that a screen or strobe reproduces the drug state (Aqil et al., 2025).

Whole-brain dynamics and connectome-harmonic work are also useful as conceptual background. They show how researchers describe large-scale brain activity in terms of spatial modes and altered dynamical repertoires (Atasoy et al., 2016; Atasoy et al., 2017; Luppi et al., 2023). But those papers do not prove that flicker or VR is equivalent to psychedelics. For Brain Candy, they remain vocabulary and hypothesis generators.

Comparison axes

  • Visual geometry: form constants, symmetry, patterning
  • Dynamics: drift, pulsing, flow, motion, rhythm
  • Contextual perception: figure-ground and meaning shifts
  • Embodiment: breath, posture, audio, and presence
  • Appraisal: surprise, comfort, awe, unease, agency

Source types

  • Peer-reviewed papers support evidence claims.
  • Preprints and tools support emerging measurement context.
  • Companies and public programs show design activity, not proof.
  • Community taxonomies can name phenomena, not validate effects.

Set, setting, and media

Context is part of the mechanism

The same visual stimulus can land differently depending on framing, music, mood, fatigue, instruction, and whether the user feels able to stop. Brain Candy has to design that context explicitly.

Cyberdelics and context engineering

Set and setting also matter outside psychedelic clinics. Any altered-perception interface shapes expectation before the first image appears. Hartogsohn's account of cyberdelics, and Smith and Warner's work on context engineering, are useful because they treat onboarding, media, symbolism, social setting, and meaning-making as part of the experience rather than packaging around it (Hartogsohn, 2023; Smith and Warner, 2022; Smith, 2024).

VR work around psychedelic preparation, simulation, social connection, and open-label feasibility makes the same point from the design side. It can help people rehearse, reflect, compare, or communicate altered experiences, but it should be described by study design and outcome rather than by broad broad change language (Sekula et al., 2022; Suzuki et al., 2017; Rastelli et al., 2022; Kaup et al., 2023).

Audio, breath, haptics, and agency

Music is one of the strongest context layers because it changes time, affect, anticipation, and memory. Psychedelic-therapy music research and public projects such as Wavepaths are useful here, but in Brain Candy they become design questions: tempo, density, tension, release, silence, and how much control the listener keeps (Kaelen et al., 2018; Messell et al., 2022; Wavepaths).

Breath and biofeedback add a different ingredient: agency. A breath-coupled interface lets the user participate in pacing instead of receiving a fixed audiovisual stream. Viscereality is one nearby example, using breath-based interaction and coupled oscillator dynamics as a bioresponsive VR design space (Fejer et al., 2025). Binaural-beat and haptic layers should stay in this restrained lane. Reviews and meta-analyses show mixed, parameter-sensitive effects for binaural beats, not a reliable switch for mental states (Garcia-Argibay et al., 2019; Ingendoh et al., 2023; Mallik and Russo, 2022).

Safety and boundaries

Light intensity is part of the design, not a footnote

Flashing light can be risky for people with photosensitive epilepsy or other sensitivities. A responsible Brain Candy interface needs conservative defaults, clear warnings, intensity control, easy stopping, and no pressure to push through discomfort. Web accessibility guidance around flashing content and public epilepsy guidance give practical boundary conditions for the design conversation (W3C WCAG, 2024; Epilepsy Foundation).

Clinical light-stimulation work, including gamma sensory-stimulation research and registered trials, is relevant background for how carefully timed light can be studied. It should not be imported as a claim that Brain Candy treats a condition (Iaccarino et al., 2016; Martorell et al., 2024; MIT GENUS; NCT05637801; NCT06922812).

The evidence boundary is simple: Brain Candy can be designed as a careful perceptual and experiential system. Clinical claims require clinical evidence. Durable psychological change requires its own outcomes. Stronger experiences are not automatically better experiences.

Design requirements

  • Visible pre-session photosensitivity warning
  • Conservative defaults for brightness, contrast, and pulse intensity
  • Immediate pause and stop controls
  • Short sessions before long sessions
  • Plain descriptions of what is known and unknown

Evidence status

  • Strong: flicker can induce patterned visual phenomena in some people.
  • Strong: flashing-light safety needs explicit controls.
  • Moderate: fractal-like abstraction can be a useful restoration comparator.
  • Emerging: psychedelic visual measures can help compare altered percepts.
  • Open: whether the full Brain Candy stack has durable benefits.

Synthesis

The useful claim is smaller and stronger

Brain Candy works best when it says what can be adjusted, what can be measured, and what remains speculative.

For newcomers

The project is not trying to replace drugs, therapy, meditation, or sleep. It is asking whether a designed audiovisual environment can help people explore controllable shifts in perception and attention. The clearest path is to make every design ingredient visible: light rhythm, visual structure, sound, breathing, interaction, expectation, duration, and safety.

For experts

The research value is in separating variables that are often blended in consumer altered-state media. A study can compare rhythmic and arrhythmic light, fractal-like and non-fractal visuals, passive and breath-responsive pacing, neutral and suggestive framing, or closed-eye and open-eye sessions. The outcome should separate visual phenomenology, affect, attention, comfort, adverse events, expectation, and repeatability.

Design tone

The right tone is neither skeptical dismissal nor techno-mystical sales copy. Rhythmic light and recursive imagery can be strong. They also need careful controls. Brain Candy is most credible when it treats vivid experience as a design and measurement problem, not as proof.

References

Source map

These references are grouped by role. Peer-reviewed papers, preprints, public programs, tools, art and design projects, company pages, community indexes, and clinical-trial records do different jobs.

Stroboscopic light and measurement

Flicker modelling and form constants

Fractals, fluency, and restoration

Psychedelic vision and whole-brain dynamics

Context, VR, audio, and breath

Safety and current clinical-light context

Page exports