Twenty healthy young adults (mean age: 25 ± 3 years) were enrolled in this study, including ten emmetropes (5 females, mean spherical equivalent (SE) OD: 0.1 ± 0.3 D, OS: 0.1 ± 0.4 D), and ten myopes (8 females, mean SE OD: −3.7 ± 1.6 D, OS: −3.4 ± 1.9 D). All participants had normal ocular health and no history of ocular surgery. Prior to the study, participants were screened for significant sleep disturbances or stress that could affect circadian rhythms; none reported relevant issues in the weeks preceding the study. Refractive errors were assessed binocularly under non-cycloplegic conditions using an infrared photorefractor (PlusOptix a12R, www.plusoptix.com), which is commonly used for screening refractive status in children and young adults, and were converted into spherical equivalents. Written informed consent was obtained from all participants. The study protocol adhered to the Declaration of Helsinki and was approved by the Swiss Research Ethics Commission (EKNZ, reference 2023-01503).

Based on their spectacle prescriptions, participants were first corrected with trial lenses placed in a trial frame to achieve clear distance vision. In the treated eye (right eye, OD), myopic defocus was imposed by adding a + 3.5 D trial lens in front of the participant’s habitual correction. At 2 m viewing distance, as used in the experiment, this resulted in approximately +3.0 D of myopic defocus in the right eye, assuming relaxed accommodation. Participants viewed the movie binocularly. Lens power was selected based on previous literature,12 since robust short-term changes in axial length in response to imposed myopic defocus were found. No contact lenses were used.

All experimental sessions were conducted in a room with an illuminance of approximately 200 lx. Participants watched a movie on a large 65″ screen, positioned at a fixed distance of 2 m. It subtended a visual angle of about 40 deg and had a luminance between 100 and 300 cd/m². The experimental protocol consisted of five measurement sessions, conducted at approximately 10:00 am, 11:00 am, 1:00 pm, 2:00 pm, and 3:00 pm. Before each session, a 15 min washout period was implemented in which the subjects watched the movie with normal binocular vision. The wash-out period reduced or excluded potential carry-over effects from preceding defocus exposure, or of previous visual exposure outside the lab. At each time point, participants watched a movie for 30 min under monocular defocus conditions, with axial length measured immediately before and immediately after each viewing session. Thus, the protocol followed a strictly timed sequential order: a 15- minute baseline washout period (e.g., 9.45–10:00 am), followed immediately by a 30- minute movie viewing session under monocular defocus (10:00–10:30 am) with optical biometry measurements performed precisely at the beginning and end of each session. This exact 45- minute cycle was repeated identically for all subsequent sessions at 11:00 am, 1:00 pm, 2:00 pm, and 3:00 pm. A one-hour lunch break was scheduled between the morning and afternoon sessions but since the afternoon session started again with a 15 min washout period, we did not expect any impact of the visual exposure during lunch break.

Measurements were always obtained in the same sequence (right eye first, followed by left eye) and were completed within approximately two minutes. Six consecutive axial length measurements were recorded per eye at each time point. We verified that axial length values did not show a systematic decline across repeated measurements with the Lenstar device, confirming that differences between treated and fellow eyes were not attributable to the brief time delay between measurements. To minimize metabolic influences, participants were instructed to refrain from caffeine and tobacco consumption for at least two hours prior to the start of the experiment.

Axial length was measured using a low-coherence optical biometer (Lenstar LS 900, Haag-Streit, products.haag-streit-usa.com) and defined as the distance from the corneal apex to the retinal pigment epithelium (RPE). Changes in axial length are commonly interpreted as indirect indicators of underlying choroidal thickness changes1–4,8 but choroidal thickness was not directly measured in the present study. Six consecutive measurements per eye were taken before and immediately after each session. Standard deviations across repeated measures were consistently below 10 μm.

Data were analyzed using “R” (version R 4.2.3, R-project.org) and GraphPad Prism (version 10.0.2 for macOS, graphpad.com). Normality of axial length changes was assessed using the Shapiro-Wilk test. Primary outcomes included axial length changes (in μm) in the treated eye (OD) and the stability of axial length in the control eye (OS) before and after each session. Paired t-tests were used to assess intra- and inter-ocular differences across time points. The p-values reported in the Results section are uncorrected. A post hoc Bonferroni correction was applied as a sensitivity analysis and confirmed that axial shortening remained significant in lens-treated eyes at 10:00 am, 11:00 am, and 3:00 pm. To formally assess the interaction between time of day and defocus condition, a two-way repeated-measures ANOVA was conducted with time (5 levels: 10:00 am, 11:00 am, 1:00 pm, 2:00 pm, 3:00 pm) and eye (OD defocused vs. OS control) as within-subject factors, applying Greenhouse-Geisser correction for sphericity.

Consistent with previous work,6 myopic eyes did not show measurable axial length changes in response to 30 min of imposed myopic defocus at any of the tested time points (Fig. 1). This lack of response under short exposure conditions suggests that myopic eyes may exhibit reduced sensitivity to brief myopic defocus stimuli. Longer exposure may be necessary to elicit measurable structural changes. A previous study has shown that two hours of computer work per day for 12 days, with the screen filtered with simulated longitudinal chromatic aberration, could elicit choroidal thickening also in myopes.13 Taken together, these observations suggest that myopic eyes display reduced responsiveness to myopic defocus, compared to emmetropes.

A comparison between normal diurnal axial length fluctuations in emmetropic eyes with the impact of positive lenses on axial eye length shows reduced responsiveness at the time of the day where axial length is maximal and choroidal thickness at a minimum (Fig. 2; reference data are from 4 published studies,14,27,31,32 all showing nice agreement). Our findings are broadly consistent with those of Chakraborty et al. (2012) ,14 who also reported no axial shortening around midday, potentially reflecting reduced efficacy of myopic defocus at that time of day. The afternoon axial elongation observed by Chakraborty et al. under continuous positive lens wear cannot be directly compared with our results, given the fundamental difference in exposure paradigms – continuous lens wear throughout the day versus repeated intermittent 30- minute exposures. Lens power also differed between studies (+1.5 vs +3.5 D). Taken together these differences suggest that positive lenses interact with diurnal cycles in a complex manner, a conclusion also reached by Chakraborty et al. (2012) .14 Reappearance of significant axial shortening at 3:00 pm (Fig. 2, red bar) suggests that defocus sensitivity returns on the same day, making cumulative fatigue and adaptation unlikely causes for the diurnal variability.

Fig. 2.

Axial eye shortening after 30 min of positive lens wear in the emmetropic group at 5 time points, denoted by red bars. Gray bars denote changes in the untreated fellow eyes. Colored lines show relative diurnal axial length changes in untreated subjects, extracted from 4 published studies.14,27,31,32 Note that responses to positive defocus were lacking around noon when axial length is at its daytime maximum.

Moderiano et al. (2019)33 have previously examined how ocular responses to short-term defocus depend on the time of day. In their design, young adult participants were exposed for two hours to either +3 D (myopic defocus) or −3 D (hyperopic defocus) at two different phases: In the morning (10:00 am – 12:00 pm) and in the evening (5:00 – 7:00 pm). It was found that ocular responses to imposed defocus are strongly modulated by the time of day, with myopic defocus producing greater choroidal thickening and axial shortening in the evening than in the morning, while hyperopic defocus increased axial length only in the morning. Choroidal thinning with hyperopic defocus occurred regardless of time of day, indicating that axial length rather than choroidal thickness was the parameter that was most sensitive to diurnal modulation. These findings may show that defocus signals interact with the eye’s intrinsic diurnal rhythms, making the ocular system more responsive to “stop signals” (myopic defocus) later in the day and to “go signals” (hyperopic defocus) earlier in the day.

Huang et al. (2025)34 found that myopic defocus inhibited choroid thinning more effectively in the morning. Nickla et al. (2017)25 studied the effects of diurnal treatment protocols in chickens and found that “Myopic defocus in the evening is more effective at inhibiting eye growth than defocus in the morning”. Based on work with negative lenses or diffusers in the chicken model, they concluded that “… reading activities early in the day may be contraindicated in school children at risk of becoming myopic”.

They also compared the effects of blue light (460 nm) exposure in the morning or in the evening on refractive development in chickens.35 The pattern of results was much dependent on the illuminances. Evening exposure to blue light stimulated ocular growth rates relative to controls for all except for the brightest condition (600 lx). Morning exposures only resulted in growth stimulation for the 200 lx condition. A simple conclusion as to whether morning stimulation with blue light may be more effective than evening could not be drawn. Earlier studies on the impact of diffusers on the development of deprivation myopia at different times of the day by Ohngemach et al. (1997)36 showed that deprivation of sharp vision in the evening caused more myopia than in the morning, but this work was, again, in the chicken. In conclusion, there is little evidence that any treatment may have a larger effect on eye growth in the morning versus in the evening. However, our current data suggest that there is least effect around noon, at least in emmetropes.

A limitation of the present study is the relatively small sample size of 10 subjects in each group and the single-day design that prevents a comparison of individual responses across multiple days. Additionally, no measurement was obtained at 12:00 pm – a time point at which several human studies have reported the greatest daytime axial elongation. Scheduling the lunch break at 1:00 pm rather than 12:00 pm would have been preferable, as it would have allowed assessment of defocus responsiveness at this critical time point. However, differences between measurements one hour apart should not be too large. Furthermore, we have no data in the evening and night where the choroid reaches its greatest thickness. However, our statistical analysis confirms time-of-day-dependent variability. The experimental design was not intended to characterize full diurnal or circadian ocular rhythms. Instead, by applying brief, repeated defocus exposures and measuring axial length immediately before and after each exposure, the study assessed time-of-day-dependent changes in responsiveness under otherwise comparable viewing conditions.

The lack of responsiveness at midday could have been attributed to altered visual experience during lunch break. However, this seems unlikely, since prominent responses were measured in the morning when participants also came from outside the lab. In addition, washout periods preceded all movie watching periods. They should have erased the effects of previous visual exposure on choroid and axial length.