Review Reports

Round 1

Reviewer 1 Report (Previous Reviewer 1)

Comments and Suggestions for Authors

This version is considerably improved.   I appreciate the careful attention given to the previous concerns. In particular, replacing the original photopic–scotopic crossover analysis with CIE S 026 melanopic weighting is much better. The addition of corneal-plane calculations, indoor exposure modeling, and explicit testable predictions also strengthens the manuscript considerably.

The revised paper now reads, more appropriately, as a computational hypothesis rather than a demonstration of a physiological mechanism.  A few issues, however, warrant additional clarification.

First, although the modeled changes in melanopic daylight efficacy ratio are substantial, changes in this ratio do not necessarily correspond to proportional changes in physiological melanopsin signaling. Absolute retinal irradiance, adaptation, temporal exposure, and contributions from other photoreceptor classes remain relevant.  I would further emphasize the distinction between spectral composition and physiologically effective melanopic stimulation.

Second, I would clarify whether the proposed effects of spectral reddening can be distinguished from those associated with declining overall light exposure. The manuscript would benefit from identifying an experimental approach capable of separating these influences.

Third, I see that you acknowledged the limitations of the analytic daylight model.  nonetheless, the accuracy of the predicted spectral ratios under low-solar-elevation conditions remains unclear. I would make these numerical predictions with appropriate caution.

Finally, the proposed daylight-saving-time test is interesting, but changes in circadian timing and behavioral routines across these transitions are complicated. It should be described as a useful test of the hypothesis rather than a definitive discriminator between spectral and nonspectral explanations.

Overall, the revisions have adequately addressed the principal methodological problems. I believe the manuscript is potentially suitable for publication as a hypothesis paper following these relatively minor clarifications.

Author Response

Dear Referee,

please find our reply to your constructive and positive report.

Sincerely,

Afif

Author Response File: Author Response.pdf

Reviewer 2 Report (Previous Reviewer 2)

Comments and Suggestions for Authors

I thought the revised manuscript was more structured and clearer than the previous version, and I appreciated these improvements. After reviewing it, I have a few further suggestions that may be helpful to consider. I hope the authors find them useful. They are mainly intended to clarify points and provide better context.

1. Lines 51 to 53. I appreciate the inclusion of the scope. However, it would be better placed near the research aim or integrated with the hypothesis toward the end of the introduction.
2. Lines 104 to 108. Please reference relevant models (e.g., CIE).
3. To help readers understand the sites, dates, and climate conditions, the authors could add monthly average charts showing daylight availability throughout the year.
4. Line 133 refers to Table 1, which does not appear until page 12, much later in the manuscript. I recommend moving it closer to where it is first introduced.
5. I valued the inclusion of S2.5. However, the overall size of the home setting and the internal reflectances of its surfaces are not provided. The text later notes that the window is small; indicating the window-to-wall ratio may help readers better understand performance in this context. Please also remove “idealised” on line 224.
6. S3.4. This section would benefit from information about the overall depth of the floor plan away from the glazed perimeter. The authors may also want to highlight that illuminance is likely to vary considerably throughout the year. Although the authors selected milestone periods (e.g., solar noon at the equinox), these do not always represent annual daylight distributions. I believe this can be highlighted on page 15, line 428. The authors could also connect this to existing literature on the topic: “A dimensionality reduction method to select the most representative daylight illuminance distributions”.
7. A brief discussion of the time series selection could be added. Although it was intended to characterize annual performance, the selected dates and times may not always represent annual daylight performance.

Author Response

Dear Referee,

Please find the attached report as a reply to your constructive and positive report.

Sincerely,

Afif

Author Response File: Author Response.pdf

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This paper presents an interesting hypothesis.  The major problem with the paper is that it is completely speculative.  No actual patients were included, no manipulations/interventions, no even correlations between say symptoms and light changes.  This is a paper that should be published in more of a scientific hypotheses type journal where speculation is welcomed.  Here are just a few big additional points to consider.

1. The big crossover point is pretty arbitrary and not physiologically meaningful (their computable time window does not make a lot of sense).  
2. Some of the modeling is pretty off (especially for the photoreceptors).  Rods and melanopsin are not interchangable for example.  Scotopic and photopic integrals cannot be interpreted as short and long-wave light.  The modeled exposure is not similar to what an actual patient would experience.  
3. Even the atmospheric modeling is off.  One value for turbidity? a chlose sky, one viewing direction, etc.  None of this is particularly realistic
4.   Even the motivation is pretty weak.  You noted a clinical study that cites a 2.4% prevalence with no consistent seasonal pattern.

I could go on for a while but, leave it to say, the paper is not ready for prime time.

Author Response

Please refer to the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Although I found the topic interesting and worth exploring, I believe the manuscript needs many revisions to improve clarity. The section order is unclear and lacks logical flow, making it hard to understand the method details and main results. The content in both sections also feels mixed, and a more organized presentation of the details is necessary. I have provided detailed suggestions below that I hope the authors find constructive and helpful to the manuscript.

1. The introduction is well-written overall. I suggest slightly restructuring the context to start with circadian and circannual rhythms. This would provide more background on daylight hours and clarify whether polar day and night are relevant to the study. Mentioning the place and date is also important. Afterwards, the authors can specify details about wavelength and how non-image forming effects impact observers with dementia.
2. Figure 1. Although I value the detailed information in both charts, their placement here appears somewhat arbitrary. The authors might consider relocating this to Section 2 or to the methods.
3. An important aspect of the manuscript that requires attention is the absence of a methods section. The manuscript goes from the introduction to the results, and then to the methods, which is quite confusing. I would suggest restructuring these to go from the introduction to the methods and then to the results. The authors may want to separate the results and discussion sections.
4. Would it be possible to include a dedicated methods section that describes the case study selection, including climatic details, as well as the simulation parameters? Additionally, consider consolidating all method-related information into Section 2. If sundown hours differ by location based on the time of day, the authors could specify the average sunset time and its standard deviation to illustrate the variation throughout the year.
5. Figure 3 could be improved for clarity by standardizing the x-axis marks, which are inconsistent across the three plots. The font style varies and should be uniform. The details within the top plot are obscured by vertical lines, and these lines can be cropped to prevent them from extending beyond the plot area.
6. Sections 3.4 and 3.5 can be somewhat confusing to follow, as 3.4 discusses photoreceptor sensitivity for visual pathways, while 3.5, although on a related topic, introduces the non-image forming pathways. I believe these are separate systems, but there are some overlaps between them that could be worth mentioning. The authors may also want to highlight some common metrics used to measure circadian stimulation, such as Equivalent Melanopic Lux.
7. I suggest renaming Section 4 to ‘Discussion,’ highlighting testable predictions as a key discussion topic. The authors could also add limitations as a subsection alongside the future research directions. Section 2.5 could also be added to this section. Currently, it appears that limitations are a significant part of the manuscript, which might need further consideration in how this is communicated.

Author Response

Please refer to the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

the late afternoon agitation, called sundowning syndrome can be found in people with dementia showing disorientation and anxiety. The present manuscript tries to extrapolate these symptoms from the change of the spectral components of the daylight spectrum in the afternoon. At this time the shortwave component of the spectrum declines steadily, whereas the long wavelength component remains nearly constant. 

The authors focus on the crossover of these spectral components which is dependent on the latitude, geometry and turbidity. They claim that this is the physical clue to the diminished short wavelength drive to the melanopsin expressing retinal ganglion cells (RGCs) which have a peak in the depleted spectral region. 

In sum this would make sense because these RGCs are drivers of the day / night biorhythm. 

I suggest that the authors reduce the part Astronomy because most of this should be common knowledge...

More important is that the authors should take exactly the same y - axis in Fig. 2! Here the picture for 17:30 jeopardizes their statement that the blue component diminishes because they magnify the y axis from 145 to 100 RSRP!

Please take the same y axis also here.  

Author Response

Please refer to the attachment.

Author Response File: Author Response.pdf