Analysis of real-world eye level light exposure through spectacle lens Authors: Emmanuel Kobia-Acquah, Camille Prince, Eléonore Pic, Des Coughlan, Joshua Hazle, Coralie Barrau Published: 2026 ABSTRACT This study presents a simulation-based methodology for evaluating lens-modulated light exposure across diverse real-world environmental conditions. A total of 987 outdoor light conditions were captured using calibrated spectroradiometers in 12 countries, recording spectral irradiance (280–780 nm, 1-nm resolution) and illuminance without lenses. Conditions varied by geographic region, weather, head inclination, and solar orientation. A lens transmittance model assigned transmittance values based on ambient illuminance and temperature to simulate the illuminance reaching the eye through clear, sun, and light-adaptive lenses. Melanopic equivalent daylight illuminance (EDI) was also computed to assess biologically relevant light exposure. Mean ambient illuminance was 42,561 ± 45,480 lux (range: 266–288,178) and 5934 ± 6738 lux (243–6738 lux) after light-adaptive lens simulation, with significant variations across environmental conditions. High-illuminance exposure (>10,000 lux) decreased from 68% (clear) to 6% (light-adaptive lens), while exposure between 1000 and 10,000 lux increased from 30% to 90%, and low-illuminance (<1000 lux) remained low (1.7% vs 3.1%). Melanopic analysis showed largely preserved melanopic EDI, with only 6% of conditions falling below 250 melanopic EDI. Light-adaptive lens reduced high-intensity light while maintaining moderate daylight levels and melanopic EDI. This methodology provides a more ecologically valid evaluation of lens performance than conventional lab-based testing. Introduction Light is a primary environmental regulator of both visual and non-visual processes in humans. Beyond its fundamental role in image formation, light exposure influences a range of physiological and behavioural outcomes, including mood1,2, sleep-wake cycles3,4, alertness5,6, academic performance7, metabolic health8,9, and ocular growth10,11. For instance, high ambient light exposure, particularly from outdoor environments, has been consistently associated with reduced incidence and progression of myopia in both humans and animal models12–16. Proposed mechanisms include the stimulation of retinal dopamine release and activation of non-visual pathways such as opsin OPN5 and intrinsically photosensitive retinal ganglion cells (ipRGCs), which are sensitive to specific bands of the visible spectrum17–19. Additionally, light is the dominant zeitgeber for circadian entrainment, synchronising the central circadian pacemaker in the suprachiasmatic nucleus through melanopsin-mediated ipRGC signalling, with downstream effects on melatonin suppression, sleep timing, and circadian alignment20–22. The biological effects of light depend not only on intensity but also on spectral composition and temporal characteristics23,24. Short-wavelength light is an important determinant of circadian effectiveness and melatonin suppression, with melanopic metrics such as melanopic equivalent daylight illuminance (melanopic EDI) providing a more physiologically relevant measure than photopic lux alone25,26. Consequently, there is growing interest in accurately quantifying the duration as well as the intensity and spectral composition of light reaching the eye. Conventional approaches to quantify light exposure often rely on self-reported questionnaires, daily diaries, or activity logs, which are prone to recall bias and lack temporal or spectral granularity27. Objective assessments using wearable light loggers mounted on the wrist or lapel have improved measurement reliability but do not accurately reflect the actual light exposure at eye level due to anatomical misalignment or shielding effects from body position28,29. Even with frame-mounted sensors and instrumented eyewear systems30,31, which provide continuous real-world measurements closer to the eye, these systems still do not account for how spectacle lenses alter the transmission and spectral profile of light reaching the eye. Spectacle lenses, particularly photochromic/light-adaptive and tinted designs, can significantly modify both the intensity (illuminance) and spectral composition of light transmitted to the eye. Different methods, including modelling and experimental, have been used to characterise ocular light exposure. For example, modelling studies, particularly those focused on ultraviolet light, have incorporated solar geometry, environmental conditions, and anatomical factors to quantify corneal irradiance under real-world conditions32,33. These have been complemented by mannequin-based measurements, which provide direct estimates of the light reaching the eye in outdoor environments34. In addition, spectral filtering methods, where spectral irradiance is combined with spectral transmittance curves, have been implemented in optical modelling and radiometry to determine the amount and spectral composition of light transmitted through optical media35–38. However, these approaches are usually applied under controlled or laboratory conditions and, while based on widely recognised optical principles, do not fully reflect the dynamic behaviour of light-adaptive lenses or the variability of light exposure in everyday environments. In this study, we present a simulation-based methodology for characterising light exposure through spectacle lenses. Our framework integrates (1) a large dataset of real-world spectral irradiance and illuminance measurements collected under diverse environmental conditions across multiple countries; (2) a dynamic lens transmission model that assigns transmittance values based on ambient illuminance and temperature; and (3) a spectral integration to compute transmitted light intensity across biologically relevant wavelength bands. This approach allows quantitative comparison of different lens designs, including light-adaptive, tinted, and static clear lenses, under variable real-world conditions, providing insights into lens performance that are not obtainable from laboratory assessments. Results Mean ambient outdoor eye-level illuminance across all conditions was 42,561 ± 45,480 lux (range: 266–288,178 lux). There were geographical variations in outdoor illuminance, with average ambient illuminance highest in the US West Coast (82,316 ± 68,088 lux) and lowest in Europe (33,475 ± 31,299 lux). Head inclination strongly influenced light exposure: zenith orientation recording the highest average illuminance (73,492 ± 50,144 lux), followed by 0° (26,219 ± 28,539 lux), and −15° (5985 ± 6641 lux). Similarly, weather conditions influenced ambient light intensity, with measurements under sunny conditions averaging 45,373 ± 44,610 lux, while cloudy conditions recorded lower average illuminance (28,179 ± 48,344 lux) (Table 1).Table 1Average eye-level illuminance (lux) recorded with different lens types without taking into account temperature dependency of photochromic dyesClear lensTv 94%(mean [SD] lux)Gen S lensgrey Tv 12% (23°C)(mean [SD] lux)Sun lensCat 3 greyTv 14% (mean [SD] lux)All conditions (N = 987)42,561 (45,480)5934 (6738)6299 (6731)Location US West Coast (N = 97)82,316 (68,088)13,208 (11,752)12,182 (10,077) US East Coast (N = 113)61,163 (70,094)9355 (9804)9052 (10,374) Pacific Asia (N = 108)43,680 (36,435)7925 (6544)6464 (5392) Europe (N = 669)33,475 (31,299)3981 (3172)4954 (4632)Weather Sunny (N = 501)45,373 (44,610)6116 (6308)6715 (6602) Cloudy (N = 220)32,472 (48,343)5087 (6709)4805 (7154) Partly cloudy (N = 158)44,344 (45,888)6365 (8742)6562 (6791) Sunny – partly cloudy (108)47,461 (40,166)7378 (5385)7024 (5944)Head inclination Zenith (N = 393)73,491 (50,143)9761 (8895)10,876 (7421) 0 (N = 473)26,219 (28,539)3784 (2805)3880 (4223) −15 (N = 121)5985 (6641)1912 (1066)885 (982)Orientation Facing sun (N = 222)45,573 (31,184)5494 (2313)6744 (4615) Away from sun (N = 195)7854 (3999)2784 (770)1162 (591) Table 1Average eye-level illuminance (lux) recorded with different lens types without taking into account temperature dependency of photochromic dyesClear lensTv 94%(mean [SD] lux)Gen S lensgrey Tv 12% (23°C)(mean [SD] lux)Sun lensCat 3 greyTv 14% (mean [SD] lux)All conditions (N = 987)42,561 (45,480)5934 (6738)6299 (6731)Location US West Coast (N = 97)82,316 (68,088)13,208 (11,752)12,182 (10,077) US East Coast (N = 113)61,163 (70,094)9355 (9804)9052 (10,374) Pacific Asia (N = 108)43,680 (36,435)7925 (6544)6464 (5392) Europe (N = 669)33,475 (31,299)3981 (3172)4954 (4632)Weather Sunny (N = 501)45,373 (44,610)6116 (6308)6715 (6602) Cloudy (N = 220)32,472 (48,343)5087 (6709)4805 (7154) Partly cloudy (N = 158)44,344 (45,888)6365 (8742)6562 (6791) Sunny – partly cloudy (108)47,461 (40,166)7378 (5385)7024 (5944)Head inclination Zenith (N = 393)73,491 (50,143)9761 (8895)10,876 (7421) 0 (N = 473)26,219 (28,539)3784 (2805)3880 (4223) −15 (N = 121)5985 (6641)1912 (1066)885 (982)Orientation Facing sun (N = 222)45,573 (31,184)5494 (2313)6744 (4615) Away from sun (N = 195)7854 (3999)2784 (770)1162 (591) Average eye-level illuminance (lux) recorded with different lens types without taking into account temperature dependency of photochromic dyes When lens effects were simulated, the average illuminance reaching the eye decreased substantially with the sun lens and the activated light-adaptive lens. Across all conditions combined, the mean light level through the sun and light-adaptive lens were 6299 ± 6731 lux and 5934 ± 6738 lux, respectively. These effects were consistently observed across geographic regions, head inclination, orientations, and weather conditions (Table 1). About 68% of all outdoor conditions exceeded 10,000 lux with clear lens, levels associated with potential discomfort or glare, while only 30% fell within moderate daylight exposure levels of 1000–10,000 lux (Fig. 1). For light-adaptive lens, the frequency of outdoor light eye level exposure within 1000–10,000 lux increased to 90%, and high-intensity exposures (>10,000 lux) decreased to 6%. Light levels below 1000 lux remained low with clear (1.7%) and light-adaptive (3.1%) lenses (Fig. 1). Subgroup analyses further showed that, under high-exposure conditions such as direct zenith head inclination, facing the sun or sunny weather, the light-adaptive lens reduced excessive brightness and maintained a narrower and more stable eye-level illuminance profile (Fig. 2). In contrast, the clear lens displayed distributions skewed toward higher intensities (Fig. 2).Fig. 1Illuminance exposure profiles and distribution.a Light exposure (illuminance) density map for different lens types. b Percentage of conditions falling within each illuminance range (<1000, 1000–3000, 3000–5000, 5000–10,000, >10,000 lux) according to the lens type.Fig. 2Illuminance distribution (log scale) across different lens types under real-world outdoor lighting conditions.a All conditions (N = 987); c Zone; d Weather; e Head inclination. Data are grouped by illuminance ranges (<1000, 1000–3000, 3000–5000, 5000–10,000, >10,000 lux). b Distribution of melanopic equivalent daylight illuminance (EDI) across lens types under morning conditions at zenith (N = 229). Less than 6% of cases (N = 13) fell below 250 melanopic EDI for both the light-adaptive and sun lenses across all temperature conditions. Illuminance exposure profiles and distribution. a Light exposure (illuminance) density map for different lens types. b Percentage of conditions falling within each illuminance range (<1000, 1000–3000, 3000–5000, 5000–10,000, >10,000 lux) according to the lens type. Illuminance distribution (log scale) across different lens types under real-world outdoor lighting conditions. a All conditions (N = 987); c Zone; d Weather; e Head inclination. Data are grouped by illuminance ranges (<1000, 1000–3000, 3000–5000, 5000–10,000, >10,000 lux). b Distribution of melanopic equivalent daylight illuminance (EDI) across lens types under morning conditions at zenith (N = 229). Less than 6% of cases (N = 13) fell below 250 melanopic EDI for both the light-adaptive and sun lenses across all temperature conditions. Melanopic EDI was strongly correlated with photopic illuminance (R² = 0.99) under ambient conditions (Supplementary Fig. S1). Within the subset of morning and zenith conditions across all temperatures (N = 229), melanopic EDI was largely maintained with the light-adaptive lens. Although the distribution shifted toward lower values relative to the clear lens, most exposures remained within moderate to high melanopic ranges, with only 6% of conditions falling below 250 lux (Fig. 2). Discussion This study presents a simulation-based framework for quantifying eye-level light exposure through spectacle lenses under real-world conditions. By combining high-resolution spectral irradiance data with a dynamic lens transmittance model, we demonstrate that a light-adaptive lens can decrease photopic illuminance and stabilise light dose to sustain vision across the day while ensuring enough melanopic light reception. Across nearly 1000 real-world conditions, the light-adaptive lens consistently reduced the frequency of high-intensity exposures (>10,000 lux), with a corresponding increase in exposures within 1000–10,000 lux range. Melanopic analysis showed that while the light-adaptive lens led to a slight reduction in values, melanopic EDI was generally well maintained, with only a small proportion of conditions falling below 250 melanopic EDI, a level often used as a reference for daytime circadian-effective light. Direct sunlight typically exceeds these illuminance levels and sustained exposure can lead to visual discomfort, glare, and increased photostress, especially in aging populations or outdoor workers39. Under high ambient illuminance conditions, such as direct overhead sun, the light-adaptive lens consistently stabilised the light dose into a range that likely minimises glare and visual discomfort, whereas in lower ambient conditions, the lens allowed adequate transmission. Moderate daylight levels have been associated with improved visual comfort and performance, circadian activation, and myopia protection10,26,40, and are generally sufficient to stimulate ipRGC activity without the discomfort associated with very high illuminance25,26. As expected, these findings suggest that lens design can influence both the magnitude and distribution of ocular light exposure in everyday environments. Light-adaptive lens may therefore offer a practical advantage over clear lenses by reducing excessive exposure while maintaining functionally relevant light levels. Photopic illuminance, while widely used, does not fully represent the biological effectiveness of light. Circadian and non-visual responses are primarily driven by melanopic-sensitive ipRGCs and depend on spectral composition and timing rather than photopic illuminance alone25,26. Because light-adaptive lenses attenuate short-wavelength light to varying degrees depending on activation state, melanopic stimulation may be reduced even when photopic illuminance remains within moderate ranges. Therefore, the illuminance categories used in this study should be interpreted as descriptive exposure bands rather than physiologically optimal thresholds. While these should not be interpreted as defining optimal exposure thresholds, they provide useful context for understanding how lens-mediated changes in light exposure may influence visual experience. A recent real-world study by Didikoglu et al. highlights the importance of considering not only the intensity of light exposure but also its temporal pattern41. In adults monitored during daily life, higher daytime melanopic exposure and less fragmented light patterns were associated with improved cognitive performance, including vigilance, working memory, and visual search, as well as reduced sleepiness and faster reaction times. These findings are relevant to the present study, which captures substantial variability in environmental light exposure across different real-world conditions. In this context, fast light-adaptive lenses may help mitigate large fluctuations in retinal light exposure by maintaining transmission within a more stable and comfortable range throughout the day. By reducing abrupt changes between low and high light levels while preserving access to meaningful daytime exposure, these lenses may contribute to a more consistent visual and non-visual light environment. Consistent with this, the melanopic analysis showed that, despite a modest downward shift with the light-adaptive lens, melanopic EDI remained largely within moderate to high ranges, with only a small proportion of values falling below 250 melanopic EDI. This aligns with emerging evidence that stable, appropriately timed daytime light exposure supports alertness and cognitive function41. However, further interventional studies are needed to determine whether light-dose stabilisation through light-adaptive lens wear translates into measurable benefits for executive function or daily cognitive efficiency. Several metrics have been proposed to quantify the amount of light reaching the human eye, including corneal illuminance, mean room surface exitance42, retinal flux43, and retinal illuminance44. Recently, the concept of perceived light fractions, defined as the proportion of emitted light received by the eye, has also been explored45. These metrics are mostly derived from static, controlled measurements and rarely incorporate the modifying effects of spectacle lenses, which can alter both the spectral composition and illuminance of light reaching the retina. More recent modelling efforts have begun to incorporate environmental variability32,33. For example, Marro and colleagues’ numerical model, to quantify individual exposure to both natural and artificial light, combined environmental radiance measurements with anthropometric and geometrical data to estimate corneal irradiance under dynamic outdoor conditions, accounting for factors such as head orientation, solar position and presence of obstructions32. However, the study does not simulate how spectacle lenses alter light transmission to the eye, nor does it integrate empirical spectral data or specific lens transmittance profiles. Jiang et al. used a mannequin-based system to directly measure eye-level light exposure through violet light-blocking or penetrating lenses, demonstrating how spectral filtering influences light reaching the eye34. They reported that violet light was mostly blocked by violet light-blocking lenses in all shapes of frames, while violet light-penetrating lenses allowed more violet light to reach the eye. Compared to our study, theirs was geographically limited, and data were obtained at only one time point. Other approaches, such as multi-sensor systems combining gaze tracking and panoramic imaging, have also been proposed to estimate luminous flux reaching the eye46. While promising, these systems have largely been applied in controlled or indoor environments and do not represent the dynamic interaction between environmental conditions (such as ambient illuminance and weather) and lens behaviour, particularly for light-adaptive lenses whose transmission properties change in real time depending on ambient light and temperature47. The present study, therefore, addresses a key methodological gap in the quantification of real-world eye-level light exposure through spectacle lenses. Unlike laboratory-based assessments of lens transmission that characterise static transmittance curves under controlled lighting, our study integrates high-resolution real-world spectral irradiance data with lens-specific transmittance modelling, enabling simulations of lens performance under a wide range of outdoor light conditions. Despite its strengths, this study had some key limitations. The model assumes a uniform eye position and does not account for individual variability in facial anatomy or gaze patterns, which can affect real‑world exposure. Mannequin-based measurements do not fully replicate the human binocular field of view or shading from facial features such as the brow and orbital structures. As shown by Marro et al., corneal irradiance can vary with head orientation and facial geometry and may differ across individuals32. Future work should include real‑time behavioural data or wearable eye tracking to capture these effects. Although the transmittance model was derived empirically, only one light-adaptive lens profile was modelled; therefore, extending to include additional lens types would broaden applicability. While illuminance bands were used as functional proxies, it does not directly link these exposures to clinical outcomes, which could potentially strengthen the physiological interpretation of exposure levels. This simulation-based model demonstrates ecologically valid evaluation of lens-modulated eye-level light exposure, offering advantages over standard lab-based testing. Light-adaptive lens significantly reduced high-intensity light exposure while maintaining adequate illuminance within moderate daylight exposure levels of 1000–10,000 lux range across diverse conditions. Melanopic analysis showed that biologically relevant light exposure was largely preserved, with only a small proportion of conditions falling below 250 melanopic EDI. By integrating spectral irradiance, dynamic lens modelling, and real‑world environmental variability, this approach advances how eye-level light exposure can be quantified through spectacle lenses. Methods Light exposure dataset Outdoor light exposure data were obtained using an in-house custom-made measurement system (LUMOS and Life 360), each incorporating calibrated spectroradiometer probes (Jaz [Ocean Optics, Orlando, USA] and Gooch & Housego [Ilminster, UK], respectively) mounted at eye level of a mannequin (Fig. 3). All spectroradiometers were calibrated according to the manufacturer’s recommendation prior to the measurements, with traceability to recognised radiometric standards. Both spectroradiometers were operated using the same optical configuration (fibre coupling and input optics) and under identical geometric and environmental conditions. Measurements captured illuminance (lux) and spectral irradiance from 280 to 780 nm at 1‑nm resolution. All recordings were conducted without spectacle lenses, representing raw ambient light exposure at eye level. A total of 987 distinct outdoor lighting conditions were collected across 12 countries in Europe, Asia, and North America. Measurements were conducted across a broad range of environmental conditions, with metadata recorded for each acquisition. These included: time of day (between 08:00 and 20:00); seasonal and weather conditions (sunny, partly cloudy, or cloudy); head inclination angles (including zenith, 0°, and −15°); sensor orientation relative to the solar position (facing the sun, away from the sun, and cardinal directions) (Table 2). Environmental parameters such as solar elevation, photoperiod, and sun-sensor geometry were not artificially standardized or normalized. Instead, they were inherently captured through the diversity of real-world acquisition conditions. This approach was intentionally adopted to reflect ecologically valid exposure scenarios, rather than controlled or simulated environments. Solar geometry (including solar elevation angle and relative azimuth) was implicitly represented through the combination of time of day, geographic location, and sensor orientation. These parameters were not controlled but were captured as part of the measurement metadata, enabling post hoc interpretation of their potential influence. The selected head orientations (zenith, 0°, and −15°) were chosen to represent common viewing conditions of human gaze behaviour in outdoor environments, capturing both direct and oblique exposure scenarios relevant to daily activities. Zenith represents an upper-bound exposure scenario, corresponding to direct upward orientation relative to the sky. A 0° inclination reflects forward gaze along the primary visual axis, which is mostly representative of typical visual behaviour. A −15° inclination approximates habitual downward gaze during walking or near tasks. Together, these positions reflect viewing angles within and around the central binocular field of view, allowing assessment of how head orientation influences eye-level light exposure.Fig. 3Optical benches for quantifying outdoor light exposure.(Top) LUMOS bench; Instrumented mannequin head with a fibered spectroradiometer in the eye. (Bottom) Life 360 bench.Table 2Characteristics of outdoor light databaseLUMOSLIFE 360N191796Measurement deviceCalibrated Jaz spectroradiometerSpectral irradiance recorded from 280 to 780 nmCalibrated Gooch & Housego spectroradiometerSpectral irradiance recorded from 280 to 780 nmHead inclinationsZenith, 0° and −15°Zenith and 0°OrientationsNorth, South, East, West, North-East; Facing /Away from the sunNorth, South, East, West, North-East; Facing /Away from the sunCountriesFranceWorldwide; 12 countries in Europe, Asia and North AmericaTimes of dayFrom morning (8AM) to evening (up to 8PM)MonthsAllWeatherRecorded through weather stations; temperature recorded for each measurement, ranging from −2 °C to 37 °C Optical benches for quantifying outdoor light exposure. (Top) LUMOS bench; Instrumented mannequin head with a fibered spectroradiometer in the eye. (Bottom) Life 360 bench. Table 2Characteristics of outdoor light databaseLUMOSLIFE 360N191796Measurement deviceCalibrated Jaz spectroradiometerSpectral irradiance recorded from 280 to 780 nmCalibrated Gooch & Housego spectroradiometerSpectral irradiance recorded from 280 to 780 nmHead inclinationsZenith, 0° and −15°Zenith and 0°OrientationsNorth, South, East, West, North-East; Facing /Away from the sunNorth, South, East, West, North-East; Facing /Away from the sunCountriesFranceWorldwide; 12 countries in Europe, Asia and North AmericaTimes of dayFrom morning (8AM) to evening (up to 8PM)MonthsAllWeatherRecorded through weather stations; temperature recorded for each measurement, ranging from −2 °C to 37 °C Characteristics of outdoor light database Simulation of light exposure through lenses To simulate the effect of spectacle lenses, a transmission model was applied to the measured ambient light data. Lens visible transmittance (Tv) values and spectral transmittance curves for the light-adaptive lens were experimentally measured under controlled D65 illumination across multiple illuminance ranges (0–1000 lux, 1000–5000 lux, 5000–10,000 lux, 10,000–20,000 lux, 20,000–35,000 lux, and >35,000 lux) and at three temperatures (10 °C, 23 °C, and 35 °C) (Fig. 4). These empirical measurements were used to derive a mathematical model describing lens activation behaviour across varying illuminance and temperature conditions. The resulting model was subsequently applied to each environmental lighting condition to simulate lens transmission under real-world exposure scenarios. For each environmental condition, ambient spectral irradiance data were multiplied by the corresponding wavelength-specific spectral transmittance profile to generate lens-modified spectral irradiance. Lens-modified photopic illuminance and Tv were subsequently derived by integrating transmitted spectral irradiance across the visible spectrum. Numerical integration was performed using the trapezoidal method (trapz, pracma package) in R (version 4.3.2; R Core Team, 2024) and applied consistently across all scenarios. The resulting values represented simulated eye-level (corneal plane) light exposure.Fig. 4Temperature- and light-dependent transmittance of light-adaptive lens.a Spectral transmittance of light-adaptive lens at various activations and temperatures. b Lens transmittance (Tv%) under controlled light conditions (D65). Temperature- and light-dependent transmittance of light-adaptive lens. a Spectral transmittance of light-adaptive lens at various activations and temperatures. b Lens transmittance (Tv%) under controlled light conditions (D65). Temperature was included as an input parameter because photochromic activation and fading are thermally dependent. Higher temperatures generally accelerate bleaching and reduce steady-state darkening, whereas lower temperatures promote greater activation and longer persistence of the coloured state. Incorporating temperature provided more realistic modelling of lens behaviour across varying outdoor environments and ensured that predicted optical outcomes reflected both illuminance and thermal effects. Light exposure metrics and classification Ambient (clear/no lens) and lens-modified photopic illuminance values were classified into the following exposure levels: <1000 lux, 1000–3000 lux, 3000–5000 lux, 5000–10,000 lux, >10,000 lux. These were further grouped into three descriptive exposure levels (<1000 lux, 1000–10,000 lux, >10,000 lux) to reflect low, moderate, and high daylight exposure ranges relevant to visual and non-visual light responses26,40. To complement the photopic illuminance analysis and provide a more physiologically relevant assessment of biologically effective light exposure, melanopic EDI was performed on a subset of environmental conditions (morning and zenith across all temperatures). Melanopic EDI was categorised into three exposure levels: <250 lux, 250–1000 lux, >1000 lux. MEDI was calculated using the CIE S026/E:028 mathematical formula:\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\rm{MEDI}}=\frac{{\int }_{380}^{780}{I}_{\lambda }(\lambda ){s}_{{mel}}(\lambda )d\lambda }{{\int }_{380}^{780}{I}_{\lambda ,D65}(\lambda ){s}_{{mel}}d\lambda }\times {\int }_{380}^{780}{I}_{\lambda ,D65}(\lambda )V(\lambda )d\lambda$$\end{document}MEDI=∫380780Iλ(λ)smel(λ)dλ∫380780Iλ,D65(λ)smeldλ×∫380780Iλ,D65(λ)V(λ)dλ Iλ–the spectral irradiance of the collected data; Iλ, D65–the spectral irradiance of the D65 illuminant; smel–the melanopic spectral sensitivity function; V as the photopic luminous efficiency function Primary outcome measures included mean photopic illuminance, distribution of exposure across defined illuminance levels, and the proportion of conditions falling within each exposure category before and after lens simulation. For the melanopic analysis, corresponding melanopic metrics were also summarised. Secondary analyses examined how these outcomes varied as a function of geographic region, head inclination, orientation, and weather condition. Test lenses Although our approach can be implemented across a range of spectacle lenses, in this study, we simulated and compared the effects of eye-level light exposure through a clear lens, a light-adaptive lens (Transitions® Gen STM), and a sun lens. All lenses were commercially available hard-coated lenses on polycarbonate substrate, with Tv values of 94% (clear lens), 14% (sun lens category 3 grey), 12% at 23 °C when fully activated and 92% at all temperatures in the bleached state (Transitions® Gen STM grey). For Transitions® Gen STM, when temperature is taken into account, we acknowledge the temperature dependency of photochromic dyes, for example, at 10 °C, full activation Tv is 4%, and at 35 °C, full activation Tv is 26% (Fig. 4). Supplementary information Supplementary Figure S1. Supplementary Figure S1.