Stroboscopic Light Stimulation Safety Within and Beyond Laboratory Settings: Observational Evidence and Practical Guidance

Stroboscopic light stimulation (SLS) on closed eyes reliably evokes vivid geometric visual phenomena, and in some contexts, altered-state experiences, motivating its increasing use across research, public installations, recreational use, and exploratory clinical contexts. The main medical risk for SLS is a photosensitive epileptic response. In contrast, non-epileptic sensitivities (e.g., migraine/photophobia, anxiety, autism/ADHD, psychosis) typically induce discomfort or distress rather than medical emergencies. Here, we integrate a multi-laboratory safety survey (24 studies; n = 1,070 participants, including cohorts with major depressive disorder), and operational data from two commercial SLS providers, with a focused review of SLS and sensory-intolerance evidence, to refine safety screening and characterise adverse events. Across laboratory studies, there were 20 minor side-effects (e.g., early withdrawal due to discomfort; ~18.7 per 1,000), with no severe incidents requiring medical attention. From two commercial datasets (~4.2 million closed-eye SLS sessions), 14 major adverse events requiring medical attention were documented (~3.3 per million); minor reactions were not systematically captured. Synthesising these strands, we developed an evidence-based tool: the Sussex Strobe Safety Screening Questionnaire (4SQ), which uses non-specialist language to identify known sensitivities to SLS, alongside practical recommendations for risk mitigation across research, clinical, commercial, and public contexts. Together, these findings suggest that the absolute risk under SLS is low but non-zero. Screening tools can help exclude pre-existing conditions, while graded exposure (e.g., a short ‘taster’ session, after which participants may opt out) and trained staff can mitigate the risk of first-episode events. Nonetheless, such events should be anticipated and minimised through clear screening pathways and their impact reduced through rehearsed on-site response plans. . CC-BY-NC 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted November 19, 2025. ; https://doi.org/10.1101/2025.11.17.25340398doi: medRxiv preprint 1.0 Introduction Stroboscopic light stimulation (SLS), usually delivered to closed eyes at frequencies between ~5-50 Hz, evokes stroboscopically induced visual hallucinations (SIVHs) comprising dynamic geometric patterns and colours, with occasional reports of complex visual hallucinations and altered-state experiences 1–3. Comparable phenomena occur with open-eye stimulation using diffusion goggles 4. As SLS use expands beyond specialist research laboratories, clear safety frameworks are required. Contemporary SLS research has so far addressed multiple, partly overlapping aims, such as investigating the neural correlates of SIVHs using fMRI and EEG, particularly in relation to neural entrainment and thalamocortical dynamics 1,5–13. Many of these studies combine electrophysiological and phenomenological measures, linking entrainment and thalamocortical dynamics to the subjective features of SLS-induced visual experiences. SLS has also been used to examine altered states of consciousness (ASCs; Bartossek et al., 2021; Beauté et al., 2025; Montgomery et al., 2024; Schwartzman et al., 2019), and to model simple visual hallucinations 7,16,17. Interest has grown in the potential therapeutic applications of SLS, particularly at gamma (~40 Hz) and alpha (9–11 Hz) frequencies, for conditions such as Alzheimer’s disease and depression 18–21. However, the evidence for any benefit remains preliminary and inconclusive 18,22. Against this backdrop, we aim to synthesise cross-laboratory practice and operational data from two commercial SLS providers, alongside a focused review of SLS and sensory-intolerance evidence1. Our goal is not only to characterise risk, but to develop a broadly applicable, evidence-based screening tool that reflects best practices across research, clinical, commercial and public domains. As use broadens, understanding who is at risk—and why— becomes increasingly important. We therefore begin by outlining the principal medical risks and discomfort-related sensitivities relevant to SLS. 1 Several of the present authors worked with Collective Act Ltd (CAL) on a large-scale public-oriented experience called Dreamachine, which used SLS and music to induce SIVHs in a large number of people. DJS, AKS, JWS, and TH provided scientific input; FM provided philosophical input; see Hewitt et al., 2025 and https://dreamachine.world. The collaboration with CAL informed aspects of our approach to SLS safety advocated here. However, this collaboration brings some legal considerations, so we’re not able to report the Dreamachine data at present. . CC-BY-NC 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted November 19, 2025. ; https://doi.org/10.1101/2025.11.17.25340398doi: medRxiv preprint 1.1 Medical risks and sensitivities associated with SLS Primary risk factor: photosensitive seizure The main medical risk associated with SLS is photosensitive convulsive seizures (PSE). While the lifetime risk of any epilepsy is relatively common (~1 in 26 in developed countries), PSE is comparatively uncommon, representing only a small proportion of epilepsy diagnoses 23. General-population estimates indicate an annual incidence of ~1 per 100,000 and a lifetime prevalence of ~1 in 10,000 24. Importantly, not all people with epilepsy are at risk from SLS exposure. Photophobia, defined by a sensitivity or intolerance to bright light, is relatively common in people with epilepsy, but true PSE remains rare. Given the rarity of this risk factor, SLS can be delivered with minimal risk when appropriate precautions are taken for PSE, along with the following secondary risk factors. In practice, the principal concern is undetected PSE, first-episode reactions in individuals who screen negative, since most known PSE cases (typically in adolescence) are aware and self-exclude. We examine this topic in greater depth in the Discussion. Several factors, however, may increase the risk of a participant having PSE or suffering from a seizure more generally. The most notable of these factors are family history, age, sleep deprivation, and pregnancy, recent alcohol or recreational drug use, and medications that lower seizure threshold. PSE is most concentrated in younger age groups—up to six times more common in children and adolescents (7– 19 years) than in adults 23. Epilepsy, including the photosensitive variant, is heterogeneous but widely recognised as heritable, often occurring in multiple members of the same family 25. Sleep deprivation is a well-established factor that increases the risk of seizures in people with epilepsy 26,27. Mechanistically, observational and experimental work indicates that reduced sleep increases both cortical excitability and the frequency of epileptiform discharges, particularly in idiopathic generalised epilepsies 26–29. Diary- based and prospective studies show that acute sleep loss can raise seizure risk by as much as threefold 30,31. Similarly, several prescribed medications, such as smoking cessation drugs (e.g., bupropion), antipsychotics, opioid analgesics, and stimulants, are associated with increased seizure susceptibility (see Supplementary Material for a summary list). Pregnancy involves physiological, hormonal, and psychological changes that may influence seizure threshold across trimesters 32, with additional factors, including altered drug metabolism, disrupted sleep, stress, and fatigue, potentially increasing susceptibility to seizure provocation. Given this potential for greater vulnerability, SLS exposure during pregnancy is not recommended. . CC-BY-NC 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted November 19, 2025. ; https://doi.org/10.1101/2025.11.17.25340398doi: medRxiv preprint Secondary risk factors: sensory sensitivities and photophobia Setting PSE aside, several conditions may increase an individual’s sensitivity to SLS, not because they are typically linked to serious medical events, but because they can heighten discomfort or distress under high-intensity sensory stimulation. These include anxiety, migraine, autism spectrum disorder (ASD), attention-deficit hyperactivity disorder (ADHD), and psychosis. Here, we summarise these sensitivities along with their prevalence. Anxiety. In the U.K., medical consultations for anxiety have recently increased markedly 33, with prevalence estimates ranging from 3.8% to 25% of the population 34. Individuals with heightened anxiety may be more susceptible to discomfort from specific lighting conditions, and SLS can be anxiety-inducing 35,36. Migraine. Current estimates put the global prevalence of migraine at ~14–15%; in terms of burden, migraine accounts for about ~4.9% of global years lived with disability 37. Photophobia is common, and bright or flickering light is frequently reported as a trigger. However, it has been argued that much of this ‘provocation’ reflects photophobia during the premonitory phase rather than an independent external trigger (Schulte et al., 2015). Whether such stimuli act as true triggers or reflect premonitory hypersensitivity, high-contrast SLS can exacerbate discomfort in people with migraine. ASD. Prevalence estimates vary considerably depending on age, diagnostic criteria, and methods. The National Institute for Health and Care Excellence estimates the global prevalence of ASD in children to be around 1-2%, and ~1.1% of U.K. adults 39. ASD is often associated with sensory hypersensitivity, heightened arousal, and elevated physiological stress responses, though these traits are heterogeneous and not universal 40,41. For some, high-intensity visual stimuli, such as bright light or SLS, may induce a stress response, particularly in those with heightened sensory reactivity 41. ADHD. ADHD affects roughly 3–4% of U.K. adults (NICE, 2021), and photophobia appears markedly over-represented, with ~69% of adults with ADHD reporting light sensitivity versus ~28% of controls 42. While ADHD per se is not a contraindication for SLS exposure, self-reported photophobia suggests a higher likelihood of discomfort under SLS. Psychosis. Psychotic disorders, including schizophrenia, are relatively uncommon (global prevalence of schizophrenia ~0.28% 43; broader psychotic disorders ~0.3%- . CC-BY-NC 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted November 19, 2025. ; https://doi.org/10.1101/2025.11.17.25340398doi: medRxiv preprint 0.7% 44). Because SLS can evoke vivid visual phenomena, interactions with active positive symptoms may increase distress or confusion, warranting additional caution. 1.2 Public Incidents Shaping Perceived Risk and Regulation Several widely publicised cultural incidents have shaped risk perception and regulation of SLS exposure, largely due to them unintentionally triggering photosensitive epileptic seizures. The infamous 1997 “Pokémon incident” in Japan involved rapidly alternating red–blue flashes at ~12 Hz shown on television to an audience mainly of children. There were seizures in hundreds of viewers, many without prior epilepsy history 45. In response, Japan introduced broadcasting regulations in 1999, limiting flash frequency to <3 Hz, restricting high-intensity flashes, and prohibiting saturated-red flicker 45. Similar concerns have arisen elsewhere: a Dutch cohort study found seizures were over three times more likely at night-time electronic dance music events with intense stroboscopic lighting compared with daylight events 46. In the U.K., Ofcom guidance, operationalised via the Harding Flash and Pattern Analyser (‘Harding test’), which algorithmically screens broadcast images for luminance, colour, and spatial pattern flicker likely to provoke photosensitive seizures, to restrict flashing imagery. Even with these tests the London 2012 Olympic logo promo had to be withdrawn after seizure concerns 47. 1.3 Current Screening Practice Across contexts, screening protocols vary widely in scope, design, and their evidence base. Laboratory studies typically use bespoke self-report questionnaires to exclude known PSE and other neurological/psychiatric vulnerabilities. In contrast, public installations often rely on simplified health declarations or verbal briefings with varying degrees of formality. Both approaches have their limitations: many focus narrowly on epilepsy, overlooking other clinically relevant sensitivities, and protocols are often context-specific and insufficiently documented to allow replication or cross-study comparisons. To address these gaps, we integrate three strands: (i) a cross-laboratory SLS safety survey, (ii) operational data from two commercial providers, and (iii) a focused review of photic-stimulation and sensory-intolerance research. Together, these strands quantify absolute and operational risk under SLS, identify where risk concentrates, and translate the evidence into practical, proportionate guidance. For each safety consideration, we assess currently available data and specify screening and operating procedures that prioritise phenotypic risk markers (photophobia, migraine/visual aura, seizure history, sensory intolerance, light-triggered discomfort) over diagnostic labels. We condense these into a simple, implementable . CC-BY-NC 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted November 19, 2025. ; https://doi.org/10.1101/2025.11.17.25340398doi: medRxiv preprint pathway suitable for research, commercial use, clinical feasibility, and public installations—including the Sussex Strobe Safety Screening Questionnaire (4SQ) for pre-screening and clear escalation/opt-out steps. Our aim is a standardised, platform-agnostic safety architecture that enables safe inclusion at scale while acknowledging that absolute risk of adverse events is low but non-zero. 2.0 Methods 2.1 Cross-laboratory and commercial data acquisition on SLS safety and practice A structured Qualtrics portal was created to capture SLS safety practices and outcomes across four research laboratories and two commercial providers. The portal recorded: (i) submitter identifiers; (ii) reporting scope (whole group/platform vs. individual studies, programmes, or product sessions); (iii) date range and total number of experiments conducted; (iv) technical parameters of the SLS used (e.g., device type, stimulation mode, eye state, frequency range, brightness, duty cycle, music synchrony); (v) participant/user counts; (vi) counts of major incidents (e.g., epileptic seizures, panic attacks, events with medical implications) and minor side- effects (e.g., discomfort leading to early withdrawal without medical consequences); (vii) number of epileptic seizures; (viii) participant/user demographics (where available); (ix) supporting documents (safety protocols, participant information, incident reports). Laboratory and supplier datasets were provided in anonymised form, and their identities are anonymised throughout the paper. Research laboratories Data were contributed by four research laboratories, based in the UK, Germany, the Netherlands, and Australia, which conducted SLS studies between 2020 and 2025 and included healthy and clinical participants. Lab A provided data from one study (N = 20); Lab B contributed twelve studies (N = 461), including two with participants with Major Depressive Disorder (MDD; N = 105); Lab C contributed six studies (N = 480); and Lab D provided five studies (N = 109). Where available, laboratories provided summary information on screening procedures and adverse-event logging. SLS characteristics. Across laboratories, SLS was delivered using commercial and custom hardware. Devices included commercial stroboscopes (Lucia N003, Light Attendance GmbH, Innsbruck, Austria; Roxiva RX1, roXiva Ltd); custom-built stroboscopes (one constructed by Lumenate Growth Ltd, Bristol, UK, and another developed at the University of Sussex); a projector (Epson EB-G7400U); and standard computer monitors at maximum brightness. . CC-BY-NC 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted November 19, 2025. ; https://doi.org/10.1101/2025.11.17.25340398doi: medRxiv preprint Commercial SLS providers Data were also obtained from two commercial providers of SLS. Commercial Provider A is a mobile-app platform that delivers closed-eye SLS experiences using the smartphone flashlight. Frequency, brightness, and duty cycle are modulated in synchrony with audio content, with many sessions designed to complement music and enhance user-reported impact. The sessions comprise dynamic SLS sequences with frequencies of 3–13.5 Hz and duty cycles of ~0.0–0.7. The platform has been in operation since 2021 and, to date, has delivered ~3.8 million closed-eye SLS sessions to a large international user base. Commercial Provider B is a manufacturer of a dedicated SLS device designed to support wellbeing and to reliably induce vivid SIVHs. The device delivers controlled SLS via high-intensity LEDs, with configurable frequency, brightness, and duty cycle parameters, and is used in both group settings and individual sessions. Users can create their own SLS sessions within a frequency envelope of 0.01–200.00 Hz and a duty cycle of 1–99% or select from the lamp’s built-in pre-set sessions. Some SLS sessions are tuned to align with music, with rhythmic modulation of light parameters designed to enhance the subjective experience. According to the company's operational figures, ~400,000 closed-eye SLS sessions have been conducted over the past five years. This conservative total excludes additional usage from consumer (non-professional) units and likely underestimates total exposures. Commercial providers developed their own safety screening criteria (see Table 1 for a summary). 3.0 Results: laboratory-based studies 3.1 Laboratory-based SLS survey Most studies (n = 21) used fixed-frequency SLS with reported frequencies spanning 1–80 Hz; all studies used a square-wave SLS signal. Stimulation protocols typically used frequencies within the alpha frequency range (8–12 Hz; Mauro et al., 2015), with one lab also experimenting with 15 Hz stimulation. A subset (n = 3; total n = 95) used dynamic frequency shifts (3–15 Hz). Flash duration and inter-flash interval (duty cycle) varied: most protocols adopted a 50% duty cycle (equal flash/inter-flash), with values ranging from 25% to 75%. Brightness levels varied by device: mean luminous flux for stroboscope-based setups was ~5,373 lumens (SD = 326); the projector operated at 447.2 cd/m²; monitor setups typically delivered ~300 cd/m². Because luminous flux (lumens) and luminance (cd/m²) quantify different properties, these figures are not directly comparable and are reported as provided by each laboratory. . CC-BY-NC 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted November 19, 2025. ; https://doi.org/10.1101/2025.11.17.25340398doi: medRxiv preprint Session duration ranged from brief exposures (8 s) to extended stimulation (≤ 25 min), with some protocols including ramp-up periods and rest intervals. 3.2 Participants Across studies, 1,070 unique participants took part (mean age = 25.04 years, SD = 2.68). Of these, 389 (36.4%) identified as male, 680 (63.6%) as female, and 1 (<0.1%) as non-binary. 105 participants self-reported symptoms consistent with MDD, assessed using the Patient Health Questionnaire-9 (PHQ-9). 3.3 SLS Safety Screening Across Laboratories Screening and risk-management procedures varied across groups, reflecting differences in populations, aims, and institutional requirements (Table 1). We collated safety protocols from all four laboratories (Labs A-D) and identified a common “minimum set”, alongside areas of divergence. All laboratories set a minimum age criterion (18 or 21 years) for participation, reflecting the higher incidence of photosensitive epilepsy in those aged 7–18 23 and consent and safeguarding requirements for minors. Every protocol excluded a personal history of epilepsy, particularly photosensitive subtypes, and most (A, B, D) also screened for first-degree family history as a precaution. All laboratories assessed sensitivity to bright and flashing light (including migraine history or photophobia), and pregnancy was a common exclusion in these research settings. Despite convergence on core exclusions, laboratories differed on several dimensions: Anxiety. All laboratories acknowledged potential discomfort in individuals with heightened anxiety, but screening approaches varied. Lab A used the State–Trait Anxiety Inventory, Trait version (STAI-T; Spielberger, 2012) with an exclusion cut- off >50. Lab B excluded self-reported severe anxiety within the past 3 years. Lab C used a broad “severe psychiatric condition” criterion. Lab D did not screen specifically for anxiety. Psychiatric disorders. Labs A and C applied broad exclusions (“any psychiatric disorder”; “any severe psychiatric condition”). Lab D excluded psychosis or a first- degree family history of schizophrenia. Lab B excluded self-reported severe anxiety or psychosis within 3 years, with less emphasis on broader or lifetime history. Importantly, these broad exclusions reflect standard good practice in laboratory- based experiments to reduce heterogeneity and outcome variability, rather than SLS- specific risk. . CC-BY-NC 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted November 19, 2025. ; https://doi.org/10.1101/2025.11.17.25340398doi: medRxiv preprint Alcohol and recreational drug use. Labs A, C, and D asked about recent use and instructed participants not to attend under the influence to reduce adverse-reaction risk and preserve data quality. Lab B did not formally assess recent substance use. Psychotropic medication use. Labs A, C, and D excluded current psychoactive medications (e.g., antipsychotics, anxiolytics), to avoid agents that may lower seizure threshold or alter SLS responsiveness. Lab B did not specify medication status. Sleep deprivation. Labs A, C, and D excluded recent sleep loss, while Lab B did not. Current major medical issues. Lab A excluded active major medical conditions (e.g., cardiovascular, metabolic, neurological) that could increase vulnerability to intense stimulation 49, citing examples such as unstable hypertension, significant arrhythmias, and multiple sclerosis. Other laboratories did not. Traumatic brain injury (TBI). Lab A excluded any history of TBI, including concussion or post-concussive syndrome, given the common post-TBI photophobia, headache, and sensory overload. Unexplained loss of consciousness: Lab B uniquely screened for unexplained, repeated losses of consciousness to identify possible absence seizures. Heart disease/pacemakers: Only Lab C explicitly excluded people with heart disease or cardiac pacemakers, including those without neuroimaging components, to maintain consistency in safety screening and minimise any potential cardiovascular risk. Aphantasia: Lab C excluded aphantasia (inability to voluntarily generate visual imagery). This was a study-design choice to maintain comparable phenomenological reporting. Data on SLS-induced experiences in aphantasia remains limited 50. . CC-BY-NC 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted November 19, 2025. ; https://doi.org/10.1101/2025.11.17.25340398doi: medRxiv preprint Table 1. Comparison of screening criteria for SLS across four laboratories (Labs A–D), and two commercial providers (Provider A = app-based; Provider B = device-based). Each row lists a specific exclusion criterion, with columns indicating whether it is explicitly included in the Lab’s screening protocol. Symbols are defined as follows: ‘✓ ’ Criterion explicitly included in the screening procedure. ‘—’ Criterion not included in the screening procedure. ‘○’ Criterion included in a qualified or partial form. PSE = photosensitive epilepsy; ‘≤3y’ = within last 3 years; FHx = family history; u/wk = units per week; psych. = psychiatric. 3.4 Additional Risk Management Strategies Here we note additional risk management strategies beyond screening, identifying common strategies across laboratories, as well as areas of divergence. . CC-BY-NC 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted November 19, 2025. ; https://doi.org/10.1101/2025.11.17.25340398doi: medRxiv preprint First-aid training and on-site readiness. The majority of laboratories ensured that staff conducting studies had current first-aid training, including recognising and managing seizures. In Lab C, not all staff were first-aid trained; however, a designated first-aid warden was always on call and immediately available during data collection sessions. Taster SLS exposure. Every site implemented ‘graded exposure’, including a brief ‘taster’ session before the main experiment to confirm tolerability and identify adverse responses before full participation. EEG pre-screening. In two studies, Lab D required first-time SLS participants to undergo a 10-minute EEG with SLS to assess photosensitivity. Although this approach is a clinically accepted method for identifying photosensitive epilepsy, its effectiveness is low in asymptomatic, pre-screened adults 51. Where thorough medical and migraine/photophobia history, clear risk information, supervised ‘graded exposure’, and trained staff are in place, routine SLS-EEG is unlikely to add

benefit and increases cost and participant burden. 3.5 Adverse responses to SLS Across all surveyed laboratories, no severe adverse reactions were observed, specifically, no seizures, and no syncope (transient loss of consciousness). Among 1,070 participants, there were 20 non-serious minor side-effects reported: 0/20 in Lab A (0.0 per 1,000), 7/461 in Lab B (15.2 per 1,000), 11/480 in Lab C (22.9 per 1,000), and 2/109 in Lab D (18.3 per 1,000), yielding a pooled prevalence of ~18.7 per 1,000 (1.87%). Events clustered into five main themes: anxiety/distress (n = 5; e.g., misinterpreting bodily sensations, uncertainty about session duration), eye discomfort/photophobia (n = 5), dizziness/vertigo/motion distortion (n = 3), headache/migraine (n = 3; including one mild migraine with prior history), and non- specific unpleasantness/early withdrawal (n = 4; e.g., “unpleasant,” “did not enjoy”). Where multiple symptoms were reported, we coded the primary reason for withdrawal. All events were resolved on-site without medical transfer. For context, this pooled prevalence (~1.87%) sits at the low end of routine MRI terminations in screened adults due to anxiety or claustrophobia (~1–3%; Munn et al., 2015). . CC-BY-NC 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted November 19, 2025. ; https://doi.org/10.1101/2025.11.17.25340398doi: medRxiv preprint 4.0 Results: Commercial SLS Providers 4.1 Commercial providers screening procedures Compared with laboratory protocols, commercial SLS screening procedures overlap on the core exclusions, notably a history of epilepsy/PSE and age ≥18, as well as caution around migraine, photophobia and significant psychiatric histories (Table 1). There are, however, clear divergences from laboratory-based screening. Provider A: asked for information about migraine or photophobia, cardiac issues, sleep deprivation cut-offs, and neurodiversity-related sensory intolerance. A novel feature was the inclusion, for users, of a detailed list of medications that increase the likelihood of seizures. In contrast, while Provider B included core exclusions (e.g., epilepsy, pregnancy, recent alcohol/drugs, sleep deprivation, head injury), there was no formal photophobia or migraine screening, syncope history, cardiac or pacemaker checks, or neurodiversity accommodations. 4.2 Commercial providers adverse responses to SLS Provider A. Across approximately 3.8 million closed-eye SLS sessions delivered via a mobile platform, based on ad hoc feedback and incident reporting, nine major adverse events were recorded. Major events were defined as those requiring medical attention (e.g., a primary care doctor visit), and these highlighted events encompass all reports judged to have probable or higher causation related to the SLS experience. This equates to an incidence of roughly 2.4 per million sessions, underscoring the rarity of such outcomes. Minor side-effects (e.g., transient discomfort, unpleasantness, or session withdrawal) were not systematically recorded and therefore are challenging to quantify. Given this limitation, the reported data primarily reflect serious events where users actively sought medical attention, rather than the broader spectrum of tolerability issues. Provider B. Based on operational figures, more than 400,000 closed-eye SLS sessions have been run with this device over the past five years. This conservative total excludes additional usage from consumer (non-professional) units and therefore likely underestimates total exposures. Data availability on adverse reactions from this source was minimal: as a commercial provider, they receive limited user feedback and incident reporting is largely ad hoc. Two recent cases were directly reported: one instance of transient memory loss over several days after repeated home use, and one seizure at a public event. Historical accounts from earlier operations describe a prolonged trance-like unresponsive state (resolved on-site) and two additional seizures assessed by emergency services, including one with short- lived cognitive disorientation. This leads to a total of 5 adverse events. While low . CC-BY-NC 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted November 19, 2025. ; https://doi.org/10.1101/2025.11.17.25340398doi: medRxiv preprint (incident rate of ~12.5 per million sessions), this number should be treated with caution given the lack of systematic reporting. 5.0 Discussion Across the laboratory and commercial datasets, adverse events associated with SLS were extremely rare and generally non-serious. Laboratory studies (n = 1,070) reported 20 minor withdrawals due to discomfort, with no severe adverse events. Commercial provider data (~4.2 million sessions) reported 14 severe adverse events requiring medical attention (~3.3 per million). These convergent data indicate that the absolute risk of serious adverse outcomes under controlled, closed-eye SLS is low but non-zero, and that proportionate pre-screening effectively mitigates the majority of predictable risks. To facilitate risk management and ensure proportionate, practical safeguards for SLS, we created the Sussex Strobe Safety Screening Questionnaire (4SQ). This is a brief, plain-language tool that flags established SLS sensitivities and is paired with actionable guidance for risk management across research, clinical, commercial, and public settings. The 4SQ prioritises phenotypic risk markers (photosensitivity, light- triggered discomfort, migraine/visual aura, seizure history, sensory intolerance) over diagnostic labels. It provides a simple, implementable screening pathway for research, clinical, and public contexts. 5.1 Development of the Sussex Strobe Safety Screening Questionnaire (4SQ) To translate our findings into practical screening guidance, we synthesise evidence from laboratory protocols, operational data from two commercial SLS providers, and the clinical/safety materials, balancing participant inclusion with risk reduction using clear, operational decision rules. In the sections that follow, we examine each safety-relevant sensitivity to SLS in turn, and (a) state the rationale linking the factor to elevated risk or intolerance for SLS, (b) specify screening questions phrased for non-specialists, and (c) give a default decision (exclude/include with caution and accommodations). Where evidence is limited or heterogeneous, we adopt conservative exclusion criteria. 1) Age (minimum age & safeguarding) Rationale: Photosensitive epilepsy is up to ~6× more common in those aged 7–19 than in adults (Quirk et al., 1995), and participation by minors carries additional consent and safeguarding requirements. Although photosensitivity declines with age, large-scale epidemiological evidence shows that the likelihood of seizures becoming prolonged or non-self-limiting (i.e., status epilepticus) increases from . CC-BY-NC 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted November 19, 2025. ; https://doi.org/10.1101/2025.11.17.25340398doi: medRxiv preprint around 60 years of age 53. Accordingly, ≥60 years is recommended as a conservative upper-age threshold for unsupervised SLS, with medical clearance and enhanced safeguards advised above this age. Screening: Confirm age is ≥18 or older at booking or consent. Default decision: Exclude. 2) Neurological history Photosensitive epilepsy or seizure history (including first-degree family history) Rationale: SLS can provoke epileptic seizures in susceptible individuals. The risk is higher in younger people and in families with a history of photosensitivity. A personal history of any unprovoked seizure (not only epileptic) indicates a lowered seizure threshold and uncertainty about individual triggers, and “non- photosensitive” epilepsy may still be precipitated by interacting factors (e.g., high arousal, anxiety, hyperventilation, sleep loss, recent illness, or medicines that may reduce seizure threshold). Given this diagnostic ambiguity, a conservative approach is to treat any prior seizure history as a contraindication. Screening: Personal epilepsy/seizure history; first-degree family history; prior visually triggered events. Default decision: Exclude. Migraine (especially with photophobia or aura) Rationale: Bright or flickering light is a common migraine trigger; visual aura indicates cortical hyper-excitability, and marked photophobia predicts discomfort and early withdrawal. Screening: History of migraine or frequent headaches; photophobia/visual aura. Default decision: Exclude when aura or marked photophobia is present or migraine; otherwise, include with caution and enhanced monitoring. Traumatic brain injury (TBI), concussion, or head injury (past 12 months) Rationale: Post-concussive symptoms (photophobia, headache, sensory overload, dizziness) can be aggravated by high-intensity SLS; recent TBI also carries an uncertain seizure risk. Screening: Any head injury/TBI in the prior 12 months; persistent post-concussive symptoms. Default decision: Exclude (12-month exclusion window). Unexplained loss of consciousness/episodes of altered awareness Rationale: May indicate undiagnosed epilepsy (e.g., absence seizures) or other instability. SLS could precipitate adverse events in such cases. Screening: History of fainting/blackouts/brief unresponsiveness without a clear medical explanation. Default decision: Exclude (or defer pending medical evaluation). . CC-BY-NC 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted November 19, 2025. ; https://doi.org/10.1101/2025.11.17.25340398doi: medRxiv preprint 3) Neurodevelopmental conditions & sensory sensitivity Autism spectrum disorder (ASD) Rationale: Higher rates of photophobia and sensory intolerance mean intense SLS may be overwhelming for some autistic individuals; the relevant risk marker is sensory sensitivity, and not the diagnosis itself. Screening: Brief, direct questions on light sensitivity and sensory intolerance. Default decision: Do not blanket-exclude. Include by default within an environment designed to be tolerable for neurotypical and neurodivergent participants. Exclude only if marked photophobia or sensory intolerance persists despite these provisions, or if the participant prefers not to proceed. Attention-deficit/hyperactivity disorder (ADHD) Rationale: Photophobia/light discomfort is commonly reported in ADHD. Again, sensory sensitivity, not the diagnostic label, is the risk marker. Screening: As above, focus on photophobia and sensory intolerance. Default decision: Do not blanket-exclude. Exclude when significant photophobia or sensory intolerance is reported; otherwise, include with accommodations. 4) Psychiatric history Anxiety disorders / high anxiety Rationale: SLS can elevate arousal and alter perception, potentially precipitating panic or distress. However, evidence is limited, so we take a conservative stance. Screening: STAI-T where feasible (suggested cut-off > 50) or self-reported severe anxiety/panic history. Default decision: Exclude if STAI-T >50 or severe anxiety disorder is reported, otherwise include with caution (enhanced briefing, easy exit, close monitoring). Psychosis (current or recent) Rationale: SLS-evoked visual phenomena may interact with positive symptoms, increasing distress or confusion. However, evidence is limited, so we take a conservative stance. Screening: Current psychotic symptoms; recent episodes. Default decision: Exclude with active symptoms; consider only with clinical clearance in stable remission. 5) Other medical conditions Pregnancy Rationale: Direct evidence on SLS in pregnancy is lacking; physiological changes during pregnancy may alter seizure threshold. Apply precautionary exclusion in research/public settings. Screening: Current pregnancy. Default decision: Exclude. . CC-BY-NC 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted November 19, 2025. ; https://doi.org/10.1101/2025.11.17.25340398doi: medRxiv preprint Heart disease / implanted cardiac devices / unstable medical conditions Rationale: SLS sessions can elicit autonomic arousal, pre-existing cardiac instability or serious systemic illness increases risk. Screening: History of arrhythmias, unstable angina, heart failure, implanted devices; uncontrolled hypertension; demyelinating disease with active symptoms, etc. Default decision: Exclude. 6) Substances & sleep Alcohol or recreational drug use (recent) Rationale: Intoxication might alter neural responsiveness (depending on substance), judgment and undermine reliable self-report and timely withdrawal. Screening: Ask (i) Have you used alcohol or recreational drugs recently? and (ii) Do you feel any current or residual effects (e.g., “high”, hangover, withdrawal, unusual fatigue, poor sleep)? Default decision: Exclude or reschedule if any current or residual effects are reported or observed. Operational note. Sites should pre-specify conservative wash-out guidance by class (e.g., alcohol, cannabis, stimulants, MDMA, psychedelics). Sleep deprivation (recent) Rationale: Acute sleep loss raises seizure risk and impairs self-monitoring of distress. Screening: Hours slept in the past 24 hours; acute restriction (e.g., < 4 hours). Default decision: Exclude/reschedule; proceed when sleep normalises. 7) Medications Psychotropic/psychoactive medications Rationale: Some medications may alter neural responsiveness and lower seizure threshold, complicating risk estimation. Screening: Current use of antipsychotics, bupropion (for smoking cessation), opioid analgesics, benzodiazepines, stimulants, or other psychoactive agents (see the Supplementary Material for a full list of medications that may lower the seizure threshold). Default decision: Exclude, unless a protocol explicitly permits specific classes with clinical oversight. Drawing on these sensitivities, we developed the Sussex Strobe Safety Screening Questionnaire (4SQ; Figure 2), a plain-language tool to identify known SLS sensitivities. A ‘YES’ response to any item signals the need for further assessment and eligibility decisions based on context-specific exclusion criteria. . CC-BY-NC 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted November 19, 2025. ; https://doi.org/10.1101/2025.11.17.25340398doi: medRxiv preprint Figure 2. The 4SQ is a brief, plain-language pre-participation screener for stroboscopic light stimulation (SLS). It comprises five sections covering (1) Neurological history (epilepsy/PSE, seizures, first-degree family history, unexplained loss of awareness, neurological diagnoses, recent head injury), (2) Psychiatric and mental health (diagnoses, recent difficulties, current psychotropic medication, first-degree family history of psychosis), (3) Heart conditions (cardiac disease, implanted devices), (4) Sensory sensitivities (migraine/photophobia, strong discomfort/panic with bright or flashing lights), and (5) Lifestyle and substance use (alcohol/recreational drugs in the last 24 h, significant sleep restriction, pregnancy). Responses are recorded YES/NO. Any YES flags the need for further assessment and a context-specific eligibility decision (exclude, include with caution, or include with accommodations). The 4SQ is not diagnostic and is intended to support proportionate risk management across research, clinical, commercial, and public settings. . CC-BY-NC 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted November 19, 2025. ; https://doi.org/10.1101/2025.11.17.25340398doi: medRxiv preprint 5.2 Practical Risk Mitigation for SLS Mitigating the risks associated with SLS delivery requires trained staff, proportionate screening, and controlled exposure parameters. We recommend that staff should receive regular first-aid training (including seizure first aid, syncope, panic response, and post-event documentation), with annual refreshers and scenario-based drills to maintain readiness. In our cross-laboratory synthesis, we found that a small set of procedural safeguards provided an effective balance between safety and accessibility: proportionate screening, brief supervised ‘graded exposure’ with explicit opt-out, plain-English stop-signals, and continuous monitoring. In contexts where direct supervision is not possible, such as app-based or home-use SLS, additional safety measures are recommended. These include providing a simple and immediate way for users to stop the SLS through direct interaction (e.g., a clear “stop” control, automatic termination of SLS if sudden movement is detected where feasible and allowing brightness adjustment to enhance comfort and minimise minor side-effects. Clear guidance should be provided to users on safe usage (e.g., not using the device near water, in the bath, or at height), and incident-reporting mechanisms should be in place so that users can easily report adverse reactions. On report of major incidents, users shall be guided towards suitable medical attention, ideally under the oversight of medical professionals or designated safety officers. Clinically, photosensitivity is typically assessed using intermittent photic stimulation (IPS) during EEG, normally spanning ~1–65 Hz to elicit abnormal EEG responses. Seizure-provoking SLS frequencies tend to cluster between 15 and 25 Hz (Fisher et al., 2005). While we do not think clinical IPS assessment is needed for general SLS use, to mitigate risk we encourage that stimulation should generally not exceed 15 Hz, with conservative sequences capped at 15 Hz and minimal dwell time above 15 Hz. This avoids the highest-risk frequency range while preserving the vividness of SLS experiences. Prolonged direct exposure to high-intensity light sources can cause retinal damage. Appropriate exposure limit values (ELVs) defined in the International Commission on Non-Ionizing Radiation Protection (ICNIRP) guidelines and adopted by the European Directive 2006/25/EC sets a maximum safe exposure of ~2.4 s at 7,000 lux with eyes open. In practice, the eyelid reflex (~0.1 s; Pearce, 2008) provides rapid protection, so exposure under controlled conditions is unlikely to be harmful. We recommend verifying peak luminance at the participant position with a light meter and documenting compliance with European Directive 2006/25/EC. For commercial providers, photobiological safety must also comply with the IEC 62471 standard (“Photobiological safety of lamps and lamp systems”), which classifies devices by risk group. Only devices rated as Risk Group 1 (“low risk”) or below should be used for SLS applications, confirming that normal use poses no hazard to the eye or skin. . CC-BY-NC 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted November 19, 2025. ; https://doi.org/10.1101/2025.11.17.25340398doi: medRxiv preprint 5.3 First-Episode Risk Without Screening: Benchmarks and Context Even with a robust screening tool, a residual risk of first-episode reactions persists among individuals who screen negative, so it is essential to contextualise baseline risk in unscreened or minimally screened populations. While this paper focuses on developing a screening measure to mitigate SLS-related risk, our results relate to pre-screened samples. To quantify the baseline risk of a first-episode adverse reaction to SLS, we draw on safety-critical occupational screening (e.g., aircrew), where intermittent photic stimulation (IPS) is routinely administered, and on mass public exposures to strobe at electronic dance music events. As a proxy for base-rate first-episode seizure risk, we compared the rate of adverse events across three large cohorts where screening did not take place: U.K. aircrew candidates (N = 13,658; adverse events: 322 per 100,000; 54), electronic dance-music event attendees (N = 400,343; adverse events: 9 per 100,000, likely under-ascertained; Salet et al., 2019), and Danish air-force recruits (N = 5,893; adverse events: 1,103 per 100,000; 55. Counting either an IPS-evoked photoparoxysmal response (PPR) or an observed seizure as a first-episode adverse event, reported rates range from 9 to 1,103 per 100,000. Due to differences in outcome measures (EEG PPR vs clinically observed seizures) and contexts (military screening, public events), we report both an exposure-weighted pooled estimate and a deliberately conservative unweighted mean. The pooled rate across all exposures is ~34.5 per 100,000 (145/419,894; 95% CI ~28.9–40.1), which likely underestimates the risk of first-time seizures in public settings and overestimates risk during IPS screening (not all PPRs translate to clinical seizures during SLS). The intentionally conservative unweighted mean (~478 per 100,000) should be treated as a worst-reasonable planning figure, not a population estimate. These rates should be treated as worst-case estimates from unscreened contexts. With adequate screening, the likelihood of a first-episode photosensitive reaction is substantially lower than these worst-case estimates. In addition to screening for seizure history, migraine/photophobia, and relevant medications, asking explicitly about prior exposure to strobe or flickering lights (e.g., nightclub lights, EEG IPS, SLS sessions), as a documented history of symptom-free exposure makes a first- episode reaction unlikely. Coupled with clear information with an opt-out option, brief, closely monitored ‘graded exposure’, and trained staff, this combination significantly reduces the chance of enrolling an undiagnosed photosensitive participant. Consistent with this view, no seizures were observed in our surveyed laboratories, and operational data from our commercial partners indicate that seizure events under closed-eye SLS are extremely rare. . CC-BY-NC 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted November 19, 2025. ; https://doi.org/10.1101/2025.11.17.25340398doi: medRxiv preprint 6.0 Conclusion We present a practical, evidence-based framework to make stroboscopic light stimulation (SLS) safe, scalable, and replicable. Screening is centred on phenotypic risk markers (photophobia, migraine/aura, seizure history, sensory intolerance) rather than diagnostic labels. The Sussex Strobe Safety Screening Questionnaire (4SQ) provides a concise, non-specialist tool for pre-screening sensitivities to SLS. External benchmarks indicate low but non-zero absolute risk from SLS; operational risk is further reduced by using sub-15 Hz SLS, graded exposure, and trained on-site response. The framework is platform-agnostic and suited to research, clinical feasibility studies, commercial delivery, and public installations moving the field towards a standardised, evidence-informed safety architecture that enables responsible and scalable inclusion. Funding D.J.S. is supported by a Medical Research Council Grant UKRI083. A.K.S. is supported by the European Research Council (ERC) Advanced Investigator Grant 101019254, under the European Union’s Horizon 2020 programme. T.H. is supported by the Margaret Boden PhD scholarship from the University of Sussex. R.B. is funded by the Be.AI Leverhulme PhD scholarship, from the Leverhulme Trust.

DJS, AKS, FM, TH and JWS are grateful to Collective Act Ltd for the opportunity to collaborate on the Dreamachine project, which informed aspects of our approach to stroboscopic-light safety described here. The authors are also grateful to Devraj Joshi (Whatever Together) and Cecelia Schwartzman for helpful discussions. . CC-BY-NC 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted November 19, 2025. ; https://doi.org/10.1101/2025.11.17.25340398doi: medRxiv preprint 7.0 References 1. Amaya IA, Behrens N, Schwartzman DJ, Hewitt T, Schmidt TT. Effect of frequency and rhythmicity on flicker light-induced hallucinatory phenomena. Spitschan M, ed. PLOS ONE. 2023;18(4):e0284271. doi:10.1371/journal.pone.0284271 2. Beauté R, Schwartzman DJ, Dumas G, et al. Mapping of Subjective Accounts into Interpreted Clusters (MOSAIC): Topic Modelling and LLM applied to Stroboscopic Phenomenology. arXiv. Preprint posted online February 25, 2025. doi:10.48550/arXiv.2502.18318 3. Hewitt T, Amaya I, Beauté R, Seth AK, Schmidt TT, Schwartzman DJ. Stroboscopically induced visual hallucinations: historical, phenomenological, and neurobiological perspectives. Neurosci Conscious. 2025;2025(1):niaf020. doi:10.1093/nc/niaf020 4. Shenyan O, Lisi M, Greenwood JA, Skipper JI, Dekker TM. Visual hallucinations induced by Ganzflicker and Ganzfeld differ in frequency, complexity, and content. Sci Rep. 2024;14:null. doi:10.1038/s41598-024-52372-1 5. Amaya IA, Schmidt ME, Bartossek MT, et al. Flicker light stimulation induces thalamocortical hyperconnectivity with LGN and higher-order thalamic nuclei. Imaging Neurosci. 2023;1:1-20. doi:10.1162/imag_a_00033 6. Becker C, Gramann K, Müller H, Elliott M. Electrophysiological correlates of flicker-induced color hallucinations. Conscious Cogn. 2009;18:266-276. doi:10.1016/j.concog.2008.05.001 7. Ffytche D. The hodology of hallucinations. Cortex. 2008;44(8):1067-1083. doi:10.1016/j.cortex.2008.04.005 8. Heller NH, Patel N, Faustin VM, Tse PU, Störmer VS. Neural correlates of stroboscopic stimulation to test a model of psychedelic hallucination. PsyArXiv. Preprint posted online December 4, 2023. doi:10.31234/osf.io/dex5v 9. Mauro F, Raffone A, VanRullen R. A Bidirectional Link between Brain Oscillations and Geometric Patterns. J Neurosci. 2015;35(20):7921-7926. doi:10.1523/JNEUROSCI.0390-15.2015 10. Pütz P, Braeunig M, Wackermann J. EEG correlates of multimodal ganzfeld induced hallucinatory imagery. Int J Psychophysiol. 2006;61(2):167-178. doi:10.1016/j.ijpsycho.2005.09.002 11. Schwartzman DJ, Schartner M, Ador BB, Simonelli F, Chang AYC, Seth AK. Increased spontaneous EEG signal diversity during stroboscopically-induced . CC-BY-NC 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted November 19, 2025. ; https://doi.org/10.1101/2025.11.17.25340398doi: medRxiv preprint altered states of consciousness. Preprint posted online January 4, 2019. doi:10.1101/511766 12. Sumich A, Anderson JD, Howard CJ, et al. Reduction in lower-alpha power during Ganzfeld flicker stimulation is associated with the production of imagery and trait positive schizotypy. Neuropsychologia. 2018;121:79-87. doi:10.1016/j.neuropsychologia.2018.11.004 13. Wackermann J, Pütz P, Allefeld C. Ganzfeld-induced hallucinatory experience, its phenomenology and cerebral electrophysiology. Cortex. 2008;44:1364-1378. doi:10.1016/j.cortex.2007.05.003 14. Bartossek MT, Kemmerer J, Schmidt TT. Altered states phenomena induced by visual flicker light stimulation. Eisenbarth H, ed. PLOS ONE. 2021;16(7):e0253779. doi:10.1371/journal.pone.0253779 15. Montgomery C, Amaya IA, Schmidt TT. Flicker light stimulation enhances the emotional response to music: a comparison study to the effects of psychedelics. Front Psychol. 2024;15:1325499. doi:10.3389/fpsyg.2024.1325499 16. Pearson J, Chiou R, Rogers S, Wicken M, Heitmann S, Ermentrout B. Sensory dynamics of visual hallucinations in the normal population. eLife. 2016;5:e17072. doi:10.7554/eLife.17072 17. Rule M, Stoffregen M, Ermentrout B. A Model for the Origin and Properties of Flicker-Induced Geometric Phosphenes. Sporns O, ed. PLoS Comput Biol. 2011;7(9):e1002158. doi:10.1371/journal.pcbi.1002158 18. Agger MP, Danielsen ER, Carstensen MS, et al. Safety, Feasibility, and Potential Clinical Efficacy of 40 Hz Invisible Spectral Flicker versus Placebo in Patients with Mild-to-Moderate Alzheimer’s Disease: A Randomized, Placebo- Controlled, Double-Blinded, Pilot Study. J Alzheimers Dis. 2023;92(2):653-665. doi:10.3233/JAD-221238 19. Huang F, Huang Q, Zheng L, et al. Effect of 40 Hz light flicker on behaviors of adult C57BL/6J mice. Brain Res. 2023;1814:148441. doi:10.1016/j.brainres.2023.148441 20. Kim S, Kim S, Khalid A, et al. Rhythmical Photic Stimulation at Alpha Frequencies Produces Antidepressant-Like Effects in a Mouse Model of Depression. Svenningsson P, ed. PLOS ONE. 2016;11(1):e0145374. doi:10.1371/journal.pone.0145374 . CC-BY-NC 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted November 19, 2025. ; https://doi.org/10.1101/2025.11.17.25340398doi: medRxiv preprint 21. Martorell AJ, Paulson AL, Suk HJ, et al. Multi-sensory Gamma Stimulation Ameliorates Alzheimer’s-Associated Pathology and Improves Cognition. Cell. 2019;177(2):256-271.e22. doi:10.1016/j.cell.2019.02.014 22. Yang YL, Lai TW. Chronic Visual Stimulation with LED Light Flickering at 24, 40, or 80 Hz Failed to Reduce Amyloid β Load in the 5XFAD Alzheimer’s Disease Mouse Model. eneuro. 2023;10(8):ENEURO.0189-23.2023. doi:10.1523/ENEURO.0189-23.2023 23. Quirk JA, Fish DR, Smith SJM, Sander JWAS, Shorvon SD, Allen PJ. Incidence of photosensitive epilepsy: a prospective national study. Electroencephalogr Clin Neurophysiol. 1995;95(4):260-267. doi:10.1016/0013-4694(95)00118-I 24. De Bittencourt PRM. Photosensitivity: The Magnitude of the Problem. Epilepsia. 2004;45(s1):30-34. doi:10.1111/j.0013-9580.2004.451010.x 25. Perucca P, Bahlo M, Berkovic SF. The Genetics of Epilepsy. Annu Rev Genomics Hum Genet. 2020;21(1):205-230. doi:10.1146/annurev-genom-120219- 074937 26. Badawy RAB, Curatolo JM, Newton M, Berkovic SF, Macdonell RAL. Sleep deprivation increases cortical excitability in epilepsy: Syndrome-specific effects. Neurology. 2006;67(6):1018-1022. doi:10.1212/01.wnl.0000237392.64230.f7 27. Moore JL, Carvalho DZ, St Louis EK, Bazil C. Sleep and Epilepsy: a Focused Review of Pathophysiology, Clinical Syndromes, Co-morbidities, and Therapy. Neurotherapeutics. 2021;18(1):170-180. doi:10.1007/s13311-021-01021-w 28. Giorgi FS, Maestri M, Guida M, et al. Controversial Issues on EEG after Sleep Deprivation for the Diagnosis of Epilepsy. Epilepsy Res Treat. 2013;2013:1-5. doi:10.1155/2013/614685 29. Giorgi FS, Guida M, Caciagli L, et al. What is the role for EEG after sleep deprivation in the diagnosis of epilepsy? Issues, controversies, and future directions. Neurosci Biobehav Rev. 2014;47:533-548. doi:10.1016/j.neubiorev.2014.10.005 30. Dell’Aquila JT, Soti V. Sleep deprivation: a risk for epileptic seizures. Sleep Sci. 2022;15(02):245-249. doi:10.5935/1984-0063.20220046 31. Haut SR, Hall CB, Masur J, Lipton RB. Seizure occurrence: Precipitants and prediction. Neurology. 2007;69(20):1905-1910. doi:10.1212/01.wnl.0000278112.48285.84 . CC-BY-NC 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted November 19, 2025. ; https://doi.org/10.1101/2025.11.17.25340398doi: medRxiv preprint 32. Li Y, Meador KJ. Epilepsy and Pregnancy. Continuum. 2022;28(1):34-54. doi:10.1212/CON.0000000000001056 33. Slee A, Nazareth I, Freemantle N, Horsfall L. Trends in generalised anxiety disorders and symptoms in primary care: UK population-based cohort study. Br J Psychiatry. 2021;218(3):158-164. doi:10.1192/bjp.2020.159 34. Remes O, Brayne C, Van Der Linde R, Lafortune L. A systematic review of reviews on the prevalence of anxiety disorders in adult populations. Brain Behav. 2016;6(7):e00497. doi:10.1002/brb3.497 35. Digre KB, Brennan KC. Shedding Light on Photophobia. J Neuroophthalmol. 2012;32(1):68-81. doi:10.1097/WNO.0b013e3182474548 36. Patterson Gentile C, Aguirre GK. A neural correlate of visual discomfort from flicker. J Vis. 2020;20(7):11. doi:10.1167/jov.20.7.11 37. Steiner T, Scher A, Stewart W, Kolodner K, Liberman J, Lipton R. The Prevalence and Disability Burden of Adult Migraine in England and their Relationships to Age, Gender and Ethnicity. Cephalalgia. 2003;23(7):519-527. doi:10.1046/j.1468-2982.2003.00568.x 38. Schulte LH, Jürgens TP, May A. Photo-, osmo- and phonophobia in the premonitory phase of migraine: mistaking symptoms for triggers? J Headache Pain. 2015;16(1):14. doi:10.1186/s10194-015-0495-7 39. Prevalence | Background information | Autism in adults | CKS | NICE. Accessed October 10, 2025. https://cks.nice.org.uk/topics/autism-in- adults/background-information/prevalence/ 40. Corbett BA, Schupp CW, Levine S, Mendoza S. Comparing cortisol, stress, and sensory sensitivity in children with autism. Autism Res. 2009;2(1):39-49. doi:10.1002/aur.64 41. Corbett BA, Muscatello RA, Blain SD. Impact of Sensory Sensitivity on Physiological Stress Response and Novel Peer Interaction in Children with and without Autism Spectrum Disorder. Front Neurosci. 2016;10. doi:10.3389/fnins.2016.00278 42. Kooij JJS, Bijlenga D. High Prevalence of Self-Reported Photophobia in Adult ADHD. Front Neurol. 2014;5. doi:10.3389/fneur.2014.00256 43. Charlson FJ, Ferrari AJ, Santomauro DF, et al. Global Epidemiology and Burden of Schizophrenia: Findings From the Global Burden of Disease Study 2016. Schizophr Bull. 2018;44(6):1195-1203. doi:10.1093/schbul/sby058 . CC-BY-NC 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted November 19, 2025. ; https://doi.org/10.1101/2025.11.17.25340398doi: medRxiv preprint 44. Saha S, Chant D, Welham J, McGrath J. A Systematic Review of the Prevalence of Schizophrenia. Hyman SE, ed. PLoS Med. 2005;2(5):e141. doi:10.1371/journal.pmed.0020141 45. Ishiguro Y, Takada H, Watanabe K, Okumura A, Aso K, Ishikawa T. A Follow-up Survey on Seizures Induced by Animated Cartoon TV Program “Pocket Monster.” Epilepsia. 2004;45(4):377-383. doi:10.1111/j.0013- 9580.2004.18903.x 46. Salet N, Visser M, Stam C, Smulders YM. Stroboscopic light effects during electronic dance music festivals and photosensitive epilepsy: a cohort study and case report. BMJ Open. 2019;9(6):e023442. doi:10.1136/bmjopen-2018-023442 47. Glendinning L. Please look away ... it’s the 2012 logo. The Guardian. https://www.theguardian.com/media/2007/jun/06/marketingandpr.olympics2012 . June 6, 2007. Accessed October 13, 2025. 48. Spielberger CD. State-Trait Anxiety Inventory for Adults. Published online July 9, 2012. doi:10.1037/t06496-000 49. Cortese A, Conte A, Ferrazzano G, et al. Photophobia in multiple sclerosis. Mult Scler Relat Disord. 2018;26:55-57. doi:10.1016/j.msard.2018.09.005 50. Reeder RR. Ganzflicker Reveals the Complex Relationship Between Visual Mental Imagery and Pseudo-Hallucinatory Experiences: A Replication and Expansion. Collabra Psychol. 2022;8(1):36318. doi:10.1525/collabra.36318 51. de Pémille CV, Rekik S, Amiel H, et al. Contribution of intermittent photic stimulation to routine EEG. Neurophysiol Clin Clin Neurophysiol. 2021;51(6):549- 553. doi:10.1016/j.neucli.2021.10.003 52. Munn Z, Moola S, Lisy K, Riitano D, Murphy F. Claustrophobia in magnetic resonance imaging: A systematic review and meta-analysis. Radiography. 2015;21(2):e59-e63. doi:10.1016/j.radi.2014.12.004 53. Ong CT, Sheu SM, Tsai CF, Wong YS, Chen SCC. Age-Dependent Sex Difference of the Incidence and Mortality of Status Epilepticus: A Twelve Year Nationwide Population-Based Cohort Study in Taiwan. Biagini G, ed. PLOS ONE. 2015;10(3):e0122350. doi:10.1371/journal.pone.0122350 54. Gregory RP, Oates T, Merry RTG. Electroencephalogram epileptiform abnormalities in candidates for aircrew training. Electroencephalogr Clin Neurophysiol. 1993;86(1):75-77. doi:10.1016/0013-4694(93)90069-8 . CC-BY-NC 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted November 19, 2025. ; https://doi.org/10.1101/2025.11.17.25340398doi: medRxiv preprint 55. Trojaborg W. EEG Abnormalities in 5,893 Jet Pilot Applicants Registered in a 20-Year Period. Clin Electroencephalogr. 1992;23(2):72-78. doi:10.1177/155005949202300206 . CC-BY-NC 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted November 19, 2025. ; https://doi.org/10.1101/2025.11.17.25340398doi: medRxiv preprint 8.0 Supplemental Material 8.1 List of medications that may lower seizure threshold Reviewing the literature on medications known to reduce the seizure threshold, we arrive at the following list of medications. Antimicrobials ● β-lactams: penicillins, cephalosporins, carbapenems (dose-related risk; carbapenems higher relative risk; also rapid, marked reduction of valproate levels with carbapenems). ● Isoniazid (mechanism via pyridoxal-5-phosphate/GABA; pyridoxine reverses toxicity). ● Fluoroquinolones (e.g., ciprofloxacin, ofloxacin, levofloxacin; rare but reported; GABAA_AA antagonism). Antimalarials ● Mefloquine, chloroquine (seizures reported in people with and without epilepsy). Analgesics ● Opioids (class effect is context- and dose-dependent; tramadol, buprenorphine and pethidine/meperidine most consistently implicated; others possible under certain conditions). ● NSAIDs: aspirin, diclofenac, indometacin (dose-dependent; uncommon clinically). Mefenamic acid can be pro-convulsant at toxic doses. Ibuprofen and paracetamol are not implicated. Methylxanthines (respiratory) ● Theophylline, aminophylline (adenosine A1 antagonism; can cause difficult- to-treat seizures in toxicity). Antipsychotics ● Clozapine (strongest signal; clinical seizures ~3–6%). ● Chlorpromazine and other phenothiazines (other antipsychotics have lower but present risk). . CC-BY-NC 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted November 19, 2025. ; https://doi.org/10.1101/2025.11.17.25340398doi: medRxiv preprint ● Haloperidol, benperidol, droperidol, melperone, azaperone (butyrophenones) ● Other monoaminergic agents: Venlafaxine and mirtazapine (rare, usually dose-related pro-convulsant effects, most often reported in overdose or at higher therapeutic doses). Antiepileptics (paradoxically) ● Benzodiazepine withdrawal (withdrawal lowers threshold). ● Withdrawal of topiramate, lamotrigine, gabapentin and pregabalin (can lower seizure threshold and precipitate seizures). ● Carbamazepine (can exacerbate some generalised/myoclonic epilepsies). Immunosuppressant / Antineoplastic ● Cyclosporine (high epileptogenic potential in review). ● Chlorambucil (intermediate potential). Radiology contrast ● Iodinated contrast media (low epileptogenic potential; rare events; risk affected by CNS comorbidity). Smoking cessation ● Bupropion (clear, dose-dependent risk at therapeutic use for smoking cessation). Other ● Interferon-α (minimal/inconclusive pro-convulsant potential). ● Atomoxetine and stimulant medications for ADHD (e.g. methylphenidate, amphetamine formulations) have a generally low seizure risk once treatment is established, but caution is advised during initiation or recent dose changes; unfamiliarity with a recently prescribed stimulant should prompt conservative screening and, where appropriate, clinical review. Additional References: Hitchings, A. W. (2016). Drugs that lower the seizure threshold. Adverse drug reaction bulletin, 298(1), 1151-1154. Ruffmann, C., Bogliun, G., & Beghi, E. (2006). Epileptogenic drugs: a systematic review. Expert review of neurotherapeutics, 6(4), 575-589. . CC-BY-NC 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted November 19, 2025. ; https://doi.org/10.1101/2025.11.17.25340398doi: medRxiv preprint