Effects of Night Illumination on Behavior, Body Mass and Learning in Male Zebra Finches

Round 1

Reviewer 1 Report

//

The manuscript examines the response of Zebra Finches to extended natural light exposure.

Adult male zebra finches (n-16) are singly housed in an aluminum mesh cage (46x40x46 cm) placed inside soundproof box (81x68x66 cm).  Half of the birds were subjected to dim light at night (12 h).  A gregarious, flocking bird such as finches are also typically in communal nests.  The authors need to provide such information on zebra finches indicating why confining single male zebra finches in a limited volume subjected to unusually prolonged lighting does not result in responses other than to light prevalence.

Conclusions drawn from the experiment as conducted rely upon the responses recorded (Fig. 1 and Table 1) and ascribed to the changes in light under unusual conditions.  Can the authors demonstrate that the experiment constraints may not be responsible for the observed results such as gain in weight, daily behavior, sleep behavior and exploratory response.  The statistical analysis performed may be subjected to the induced change in lightning but only if the effects of the above factors are eliminated or accounted for.

The color learning and memory retrieval component of the experiment may be less suspect to excess lightning than the other conclusions.

Minor Issues:

Abstract:         reads as a Summary and contains undefined acronyms, superfluous phrases and words, such as “We observed”, “Overall, we found” and “Interestingly” (misspelled)

Keywords:      LAN is not a word.  Artificial light as the central issue of the manuscript is not included.

Title:    Cumbersome – A more compact title might be:  “Effects of night illumination and the objective learning of male zebra finches”

Introduction:   Needs re-writing

The Introduction borders on redundancy and contradiction (lines 5 and 75).  Grammatical errors should be corrected, e.g., “both” (line 42 vs line 45).  

Line 46 – causes “a” mismatch

Line 50 – across “the” animal kingdom

Line 58 – use of the word “even” – eliminate

Line 63 – “Evidence shows”

Lines 66-67 – rats plural

Line 78 – not singular – need to be consistent

Contradiction – lines 54-67 vs lines 75-80

Use of words “Still” (line 75), “Also” (line 77) and “Therefore” (line 78) all superfluous

Recommendation:  Contingent upon the authors response to the above, the manuscript deserves positive consideration of publication.

Author Response

Comment #1: The manuscript examines the response of Zebra Finches to extended natural light exposure. Adult male zebra finches (n-16) are singly housed in an aluminum mesh cage (46x40x46 cm) placed inside soundproof box (81x68x66 cm).  Half of the birds were subjected to dim light at night (12 h).  A gregarious, flocking bird such as finches are also typically in communal nests.  The authors need to provide such information on zebra finches indicating why confining single male zebra finches in a limited volume subjected to unusually prolonged lighting does not result in responses other than to light prevalence.

Response #1: Thank you. We have mentioned this in materials and methods section. It is a common practice in these types of experiments when we take multiple observations from a single individual (e.g., rhythm analysis and individual food intake data). We take daily observations of bird’s behaviour (activity, sleep and feeding) and measure body mass weekly to monitor the effect of confinement, if any, however we didn’t see any effect of confinement. Similar protocols on Zebra Finch have been published in many publications (Prabhat et al., 2020, Hormones and Behavior; Prabhat et al., 2020, Journal of Photochemistry and Photobiology B; Batra et al., 2019, Environment Pollution).

Comment #2: Conclusions drawn from the experiment as conducted rely upon the responses recorded (Fig. 1 and Table 1) and ascribed to the changes in light under unusual conditions. Can the authors demonstrate that the experiment constraints may not be responsible for the observed results such as gain in weight, daily behavior, sleep behavior and exploratory response. The statistical analysis performed may be subjected to the induced change in lightning but only if the effects of the above factors are eliminated or accounted for.

Response #2: Thank you. We have mentioned this in discussion. We did not discount the effect of other factors than light, however, since light at night (not the amount of light overall) was the only variable for birds, we speculated that it was a cumulative effect of light at night in daily behavior, sleep behavior and exploratory response. However, further molecular experiments will be able to address the direct effect of light, if any.

Comment #3: The color learning and memory retrieval component of the experiment may be less suspect to excess lightning than the other conclusions.

Response #3: Thank you. We argue that effect of learning and memory due to changes in brain areas responsible for learning and memory rather than direct effect of light as mentioned in Taufique et al., 2018, European Journal of Neuroscience and Taufique et al., 2018, Neurobiology of Learning and Memory.

Comment #4: Minor Issues:

Abstract: Reads as a Summary and contains undefined acronyms, superfluous phrases and words, such as “We observed”, “Overall, we found” and “Interestingly” (misspelled)

Response: Thank you. We have revised the abstract. The acronyms have been defined.

Keywords: LAN is not a word.  Artificial light as the central issue of the manuscript is not included.

Response: Thank you. We have added “Artificial light” and removed “LAN” from keywords.

Title: Cumbersome – A more compact title might be: “Effects of night illumination and the objective learning of male zebra finches”

Response: Thank you. We have revised the title.

 Introduction:   Needs re-writing

Response: Introduction has been revised as per the reviewer’s comment.

The Introduction borders on redundancy and contradiction (lines 5 and 75).  Grammatical errors should be corrected, e.g., “both” (line 42 vs line 45). 

Response: We have removed “both” and other grammatical mistakes have been corrected. Also, the contradictions have been removed and revised.

Line 46 – causes “a” mismatch

Response: Thank you. Added.

Line 50 – across “the” animal kingdom

Response: Thank you. Added.

Line 58 – use of the word “even” – eliminate

Response: Thank you. Eliminated.

Line 63 – “Evidence shows”

Response: Thank you. Corrected.

Lines 66-67 – rats plural

Response: Thank you. Corrected.

Line 78 – not singular – need to be consistent

Response: Thank you. Revised.

Contradiction – lines 54-67 vs lines 75-80

Response: Thank you. Revised.

Use of words “Still” (line 75), “Also” (line 77) and “Therefore” (line 78) all superfluous

Response: Thank you. We have reworded those lines accordingly.

Reviewer 2 Report

This study explores the effects of dim (5 lux) light at night on the physiology and behaviour of captive zebra finches. The results of the study are consistent with previous studies in birds and laboratory rodents, which indicate that light at night can affect timing of behaviour, sleep behaviour, food intake, body mass, and performance. I find the observed link between decreased sleep behaviour and increased body mass (despite similar food intake) particularly interesting. While I believe these findings are valuable to our growing understanding of the effects of light at night, some results need to be interpreted more carefully and there are limitations of the study that need to be more clearly acknowledged. Some aspects of the study and results are also currently unclear, but I believe these issues can be easily rectified. Please find my detailed comments below.

General comments:

1. In a natural setting, songbirds like zebra finches would rarely, if ever, experience 0 lux at night. There would usually be some source of astronomical light that would allow birds to see some of their environment. Were the birds in this study truly in complete darkness at night? If so, the control conditions in this study might have produced unrealisitically low levels of activity at night. Consequently, the study might overestimate the effects of dLAN on behaviour and physiology. This is an important limitation of the study design that needs to be acknowledged when interpreting the results.
2. Although the manuscript is generally clear and well-structured, there are errors in grammar and syntax that sometimes make the authors’ logic difficult to follow. The manuscript will need to be edited more thoroughly before publication, either by the authors or by an external editor.
3. Throughout the manuscript, the justification for the study and the significance of the results are not particularly clear or compelling. It would be helpful to more clearly and explicitly highlight which effects you are testing are novel, which are replicated (bearing in mind that replication studies are also valuable), what insights your study has to offer, and why zebra finches are a useful or relevant study system for this research.
4. I am not completely convinced by the interpretation of the birds’ response to the novel object, although this may just be because I am less familiar with this test. Are there other studies that have used novel object exploration as a measure of an animal’s “mood”? How much does this behaviour usually vary? Why would you expect such behaviour to be influenced by exposure to light at night? How are these responses important in an ecological context and what are the implications of your findings? Addressing these questions would help the reader understand the relevance and importance of this test.
5. In the methods section, you state that body mass, fat score and food intake were measured at the start and end of the experiment (line 104). The results section also refers to an “increase” in body mass and fat under dLAN (lines 268-274). However, Table 1 only seems to present mass and fat deposition at a single timepoint, and compare these between the two treatments. Is this correct? If so, you also need to analyse and clearly present the change in body mass and fat from the start to the end of the experiment. Otherwise, there is no way for the reader to know whether the birds in the dLAN treatment just happened to be heavier and fatter from the start.
6. Throughout the manuscript, any difference between the control and treatment group tends to be described as a “negative” effect of dLAN (e.g. line 368). However, this is not necessarily true. More guarded responses to novel objects could benefit birds in dangerous environments. The ability to utilise light at night for foraging might also be beneficial when food is scarce. While I agree that at least some of the observed effects could have negative consequences, I suggest being more cautious with these assumptions.
7. Sleep has multiple states, can vary in intensity, and is not always associated with sleeping postures (e.g. see review by Aulsebrook et al. 2016 Trends in Ecology and Evolution; also Aulsebrook, Connelly et al. 2020 Current Biology). For example, some birds can sleep while sitting upright with their eyes open, and may also sit in a resting posture while awake. Consequently, it would be more appropriate to refer to ‘sleep behaviour’ and ‘decreased sleep behaviour’ throughout this study, rather than ‘sleep’ and ‘sleep loss’. ‘Sleep quality’ is also an ambiguous term, which can mean different things for different contexts and species. It would therefore be better to refer to the ‘continuity’ or ‘fragmentation’ of sleep, rather than the ‘quality’ of sleep.

Introduction:

8. Lines 77-78: I do not believe it is true to say that most studies of the effects of light at night have been limited to nocturnal mammalian model systems. There have now been many studies of diurnal birds, many of which you cite in your manuscript.

Methods:

9. Please provide additional information on the spectral properties of the light (for both L and dLAN), either in the main text or the supplementary materials. Such information is important for being able to replicate and compare studies, particularly as the spectral properties of light can influence behavioural and physiological responses.
10. What time of year was the study conducted? Seasonal timing could potentially interact with effects of light at night on behaviour and body mass.
11. I understand that 200 lux is a standard daytime light intensity to use in studies of captive animals. However, this is also not ecologically realistic; daylight from sunlight would usually exceed 1000 lux. I appreciate that you cannot resolve this for the current study, but it is worth bearing in mind for future research, particularly since daytime light exposure can also influence circadian rhythms and sleep.
12. Perhaps I have missed these details, but I am having some trouble understanding the timeline of the experiment. When, within the 6-week experiment, was finch behaviour was recorded for two consecutive days? When was the novel object exploration test? It might be helpful to include a schematic diagram of the study timeline in the supplementary materials.
13. Lines 119-120: It is a little unclear to me how having two people scan the videographs manually would remove all experimenter bias. Were these people unaware of the study purpose and treatments?
14. Lines 140-141: Here, it seems like either the wrong study has been cited, or the sentence needs to be rephrased. There is no mention of the response to novel objects in Taufique et al. 2018.
15. Line 170: I found the new subheading here a bit confusing – it gave the initial impression that there were two different sets of tasks. Perhaps consider removing the subheading or renaming the two subheadings to make this easier to follow.
16. Line 171: For how many days “at the end of the experiment”?
17. Lines 194-195: I found this explanation (“Unpaired parametric Student’s t-tests and non-parametric 194 Mann-Whitney’s U test compared…”) a little confusing; it sounded like both tests were used for each comparison. A short statement describing which tests were used for which comparison, and why, would be helpful.

Results:

18. Lines 276-277: I am not aware of evidence that response to a novel object is a good measure of ‘depression’ in animals. Please provide a source or correct this sentence.
19. Perhaps I have missed something, but why were only 10 birds used for the colour learning trials? Did some birds refuse to participate, or were they excluded for other reasons? Please include this information in the methods or results section.

Tables and figures:

20. All Tables: It would be easier to read and compare values in the tables if they were all presented with the same number of decimal places and were left- or right-aligned.
21. Table 1A: I would also like to see some information on the range and/or variation in fat scores for each group, rather than just the median. Was there much overlap between the two groups?
22. Figure 1: I would find it quicker and easier to interpret this plot if there were lines or shading to indicate the timing of day (L) and night (LO/dLAN).
23. Figure 2: Some of the images in this figure are very small. It might be better to either enlarge the figure in the final version of the manuscript or to move these images to the supplementary material.
24. Line 339 (caption of Figure 3): “trails” should be “trials”

Discussion:

25. Line 421: I would remove the statement “akin to humans”. You are implying here that your results can be generalised to humans, which I think is unnecessary and distracting. There are already many published studies investigating the effects of light at night on humans, and the light treatment used for your study is not particularly relevant for humans.
26. Lines 425-427: As mentioned in my earlier comment (3), the justification and significance of the study here is very vague. What exactly are the new insights provided by this study? How might these help us understand the behaviour, physiology and brain functions of animals in lit environments?

Author Response

This study explores the effects of dim (5 lux) light at night on the physiology and behaviour of captive zebra finches. The results of the study are consistent with previous studies in birds and laboratory rodents, which indicate that light at night can affect timing of behaviour, sleep behaviour, food intake, body mass, and performance. I find the observed link between decreased sleep behaviour and increased body mass (despite similar food intake) particularly interesting. While I believe these findings are valuable to our growing understanding of the effects of light at night, some results need to be interpreted more carefully and there are limitations of the study that need to be more clearly acknowledged. Some aspects of the study and results are also currently unclear, but I believe these issues can be easily rectified. Please find my detailed comments below.

Response: Thank you for your interest in the study. We have revised the article according to the comments.

General comments:

Comment #1: In a natural setting, songbirds like zebra finches would rarely, if ever, experience 0 lux at night. There would usually be some source of astronomical light that would allow birds to see some of their environment. Were the birds in this study truly in complete darkness at night? If so, the control conditions in this study might have produced unrealisitically low levels of activity at night. Consequently, the study might overestimate the effects of dLAN on behaviour and physiology. This is an important limitation of the study design that needs to be acknowledged when interpreting the results.

Response #1: Thank you. We have added the limitations of the present study as pointed out by other reviewers and editor at the end of the discussion. It is true that there is presence of astronomical light that may allow birds to see some of their environment. However, we measured night time illumination near the forest and university campus area (around 2 Kms) which shows 0 lux. Also, studies on songbirds have shown the dose-dependent response (0.05 lux – 5 lux) of light at night on behavior and melatonin hormonal level (DeJong et al., 2016, Physiology and Behavior). Therefore, we used only one intensity of light and used complete darkness for the controls according to the similar protocols in Taufique et al., 2018, European Journal of Neuroscience and Batra et al., 2020, Proceedings of Royal Society of Biology:B.

Comment #2: Although the manuscript is generally clear and well-structured, there are errors in grammar and syntax that sometimes make the authors’ logic difficult to follow. The manuscript will need to be edited more thoroughly before publication, either by the authors or by an external editor.

Throughout the manuscript, the justification for the study and the significance of the results are not particularly clear or compelling. It would be helpful to more clearly and explicitly highlight which effects you are testing are novel, which are replicated (bearing in mind that replication studies are also valuable), what insights your study has to offer, and why zebra finches are a useful or relevant study system for this research.

Response #2: Thank you. We have revised the introduction and clearly mentioned the novelty and replication in the present experimental design. We have also added the relevance of Zebra Finch for this research.

Comment #3: I am not completely convinced by the interpretation of the birds’ response to the novel object, although this may just be because I am less familiar with this test. Are there other studies that have used novel object exploration as a measure of an animal’s “mood”? How much does this behaviour usually vary? Why would you expect such behaviour to be influenced by exposure to light at night? How are these responses important in an ecological context and what are the implications of your findings? Addressing these questions would help the reader understand the relevance and importance of this test.

Response #3: Thank you. Yes. Novel object exploration test is commonly used for mood and depression state of an animal in both mammals and birds. A similar test was done in Jha and Kumar, 2017, Animal Behavior and Taufique et al., 2018, European Journal of Neuroscience, to address effect of constant light and dim light at night on mood, depression and personality in Zebra Finches and Indian House Crows. Since light at night have shown to disturb the daily rhythms in mammal and songbirds, both and all abovementioned behaviours are hippocampus-dependent (object location and recognition; Crusio, 2001, Behavioral Brain Research), we tested if light at night can induce loss-of interest in performance via multiple tests. We have also answered each of the mentioned question in the revised text.

Comment #4: In the methods section, you state that body mass, fat score and food intake were measured at the start and end of the experiment (line 104). The results section also refers to an “increase” in body mass and fat under dLAN (lines 268-274).

However, Table 1 only seems to present mass and fat deposition at a single timepoint, and compare these between the two treatments. Is this correct? If so, you also need to analyse and clearly present the change in body mass and fat from the start to the end of the experiment. Otherwise, there is no way for the reader to know whether the birds in the dLAN treatment just happened to be heavier and fatter from the start.

Response #4: Thank you. We have measured body mass, fat score and food intake at the start and end of the experiment. We have now included the changes in the body mass and fat in the Table 1 and amended the results accordingly.

Comment #5: Throughout the manuscript, any difference between the control and treatment group tends to be described as a “negative” effect of dLAN (e.g. line 368). However, this is not necessarily true. More guarded responses to novel objects could benefit birds in dangerous environments. The ability to utilise light at night for foraging might also be beneficial when food is scarce. While I agree that at least some of the observed effects could have negative consequences, I suggest being more cautious with these assumptions.

Response #5: Thank you. We have revised the discussion and included the point that guarded response to the novel object could benefit birds in dangerous environment and also that light at night can be utilized for the foraging in some conditions.

Comment #6: Sleep has multiple states, can vary in intensity, and is not always associated with sleeping postures (e.g. see review by Aulsebrook et al. 2016 Trends in Ecology and Evolution; also Aulsebrook, Connelly et al. 2020 Current Biology). For example, some birds can sleep while sitting upright with their eyes open, and may also sit in a resting posture while awake. Consequently, it would be more appropriate to refer to ‘sleep behaviour’ and ‘decreased sleep behaviour’ throughout this study, rather than ‘sleep’ and ‘sleep loss’. ‘Sleep quality’ is also an ambiguous term, which can mean different things for different contexts and species. It would therefore be better to refer to the ‘continuity’ or ‘fragmentation’ of sleep, rather than the ‘quality’ of sleep.

Response #6: Thank you. We have revised the word “sleep, sleep loss and sleep quality”. The heading read “Behavioural assay of sleep” in the materials and methods section. We have only used sleep states based on the Taufique et al., 2018, European Journal of Neuroscience and Batra et al., 2020, Proceedings of Royal Society of Biology:B.

Introduction:

Comment #7: Lines 77-78: I do not believe it is true to say that most studies of the effects of light at night have been limited to nocturnal mammalian model systems. There have now been many studies of diurnal birds, many of which you cite in your manuscript.

Response #7: Thank you. We have removed the line and revised the sentence.

Methods:

Comment #8: Please provide additional information on the spectral properties of the light (for both L and dLAN), either in the main text or the supplementary materials. Such information is important for being able to replicate and compare studies, particularly as the spectral properties of light can influence behavioural and physiological responses.

Response #8: Thank you. We have explained the spectral properties of light used in the methods section.

Comment #9: What time of year was the study conducted? Seasonal timing could potentially interact with effects of light at night on behaviour and body mass.

Response #9: This study was conducted in the month of January and February. We have kept birds on similar light:dark cycle (12L: 12D) throughout the year in our breeding facility at same temperature. During the experiment, only light at night was the changed variable. We have mentioned this in methods.

Comment #10: I understand that 200 lux is a standard daytime light intensity to use in studies of captive animals. However, this is also not ecologically realistic; daylight from sunlight would usually exceed 1000 lux. I appreciate that you cannot resolve this for the current study, but it is worth bearing in mind for future research, particularly since daytime light exposure can also influence circadian rhythms and sleep.

Response #10: Thank you for the valuable suggestion. The sunlight intensity exceeds 1000 lux; however, shadows show 100-400 lux during day time. We have also standardized 150-200 lux intensities during daytime in multiple experiments on songbirds (Prabhat et al., 2020, Hormones and Behavior; Taufique et al., 2018, European Journal of Neuroscience and Batra et al., 2020, Proceedings of Royal Society of Biology:B).

Comment #11: Perhaps I have missed these details, but I am having some trouble understanding the timeline of the experiment. When, within the 6-week experiment, was finch behaviour was recorded for two consecutive days? When was the novel object exploration test? It might be helpful to include a schematic diagram of the study timeline in the supplementary materials.

Response #11: Thank you. We have now added the study timeline in Figure 1.

Comment #12: Lines 119-120: It is a little unclear to me how having two people scan the videographs manually would remove all experimenter bias. Were these people unaware of the study purpose and treatments?

Response #12: Thank you. We have revised the statement. Two people unaware of the study purpose and treatments scanned the videographs.

Comment #13: Lines 140-141: Here, it seems like either the wrong study has been cited, or the sentence needs to be rephrased. There is no mention of the response to novel objects in Taufique et al. 2018.

Response #13: We apologized for the inadvertent mistake. Taufique et al., 2018, European Journal of Neuroscience (Illuminated night alters hippocampal gene expressions and induces depressive-like responses in diurnal corvids) tested the novel object exploration in Indian house crows in dim light at night. We have corrected the reference.

Comment #14: Line 170: I found the new subheading here a bit confusing – it gave the initial impression that there were two different sets of tasks. Perhaps consider removing the subheading or renaming the two subheadings to make this easier to follow.

Response #14: Thank you. We have removed the different subheadings and included only 1 subheading for colour learning task.

Comment #15: Line 171: For how many days “at the end of the experiment”?

Response #15: Thank you. We have tested the novel object exploration in 6^th week. We have mentioned this in the timeline (Figure 1).

Comment #16: Lines 194-195: I found this explanation (“Unpaired parametric Student’s t-tests and non-parametric 194 Mann-Whitney’s U test compared…”) a little confusing; it sounded like both tests were used for each comparison. A short statement describing which tests were used for which comparison, and why, would be helpful.

Response #16: Thank you. We have revised the statistics according to the test and explained separately.

Comment #17: Results:

Lines 276-277: I am not aware of evidence that response to a novel object is a good measure of ‘depression’ in animals. Please provide a source or correct this sentence.

Response #17: Thank you. We have added the reference and novel object test is commonly used to address the ‘depressive’ like state in mammals (Fonken et al., 2012, Dim nighttime light impairs cognition and provokes depressive-like responses in a diurnal rodent. Journal of Biological Rhythms, 27, 319–327).

Comment #18: Perhaps I have missed something, but why were only 10 birds used for the colour learning trials? Did some birds refuse to participate, or were they excluded for other reasons? Please include this information in the methods or results section.

Response #18: Thank you. As mentioned in the Figure 4 and Table 2, 6 birds under LD and 5 birds under dLAN performed the task. We have mentioned this in the results section.

Comment #19: Tables and figures:

All Tables: It would be easier to read and compare values in the tables if they were all presented with the same number of decimal places and were left- or right-aligned.

Response #19: Thank you. We have left aligned all the values and presented in the same number of decimals (up to two decimals).

Comment #20: Table 1A: I would also like to see some information on the range and/or variation in fat scores for each group, rather than just the median. Was there much overlap between the two groups?

Response #20: Thank you. We have included the range of fat score of each group in the results section. The range was 0-1 for LD and 2-5 of dLAN. There was almost no overlap between two groups after 6 weeks of light treatment.

Comment #21: Figure 1: I would find it quicker and easier to interpret this plot if there were lines or shading to indicate the timing of day (L) and night (LO/dLAN).

Response #21: Thank you. We have included the white and black bar denoting the LD and dLAN in the Figure 2.

Comment #22: Figure 2: Some of the images in this figure are very small. It might be better to either enlarge the figure in the final version of the manuscript or to move these images to the supplementary material.

Response #22: Thank you. We have enlarged the images and now explained in Figure 1.

Comment #23: Line 339 (caption of Figure 3): “trails” should be “trials”

Response #23: Thank you. Corrected.

Comment #24: Discussion:

Line 421: I would remove the statement “akin to humans”. You are implying here that your results can be generalised to humans, which I think is unnecessary and distracting. There are already many published studies investigating the effects of light at night on humans, and the light treatment used for your study is not particularly relevant for humans.

Response #24: Thank you. We have removed and revised the statements.

Comment #25: Lines 425-427: As mentioned in my earlier comment (3), the justification and significance of the study here is very vague. What exactly are the new insights provided by this study? How might these help us understand the behaviour, physiology and brain functions of animals in lit environments?

Response #25: Thank you for your inputs. We have revised the discussion and addressed abovementioned points.

Reviewer 3 Report

This is an interesting study building on previous studies on the same topic. It is generally well-written, but some work needs to be done to improve the flow and clarity. Further work on the methods may need to be done, particularly regarding inter-rater reliability. More details are provided below.

Simple summary:

Line 12 - 13: Change "a dim light at night protocol" to "dim light at night".

Line 14: Remove "The".

Line 15: Remove "and" before "colour".

Line 20: "dLAN" abbreviation not previously used. Include in parenthesis at first mention of "dim light at night" above.

Abstract:

Line 30: "LD" abbreviation not previously explained.

Introduction:

Line 43: Remove "both".

Line 44: Is "clock" referring to the biological/endogenous clock? Please elaborate.

Line 47: Please provide references supporting this statement. Is "reduced" supposed to be "deduced"?

Line 80: Why were male zebra finches chosen rather than female as in the previous study cited? Please explain and justify.

Methods:

Line 107 - 111: Please provide a figure demonstrating each of these scores.

Line 112 - 113: It may also be useful to provide a figure demonstrating this. Elaboration of the "different times" may also be important.

Line 119: If 2 people assessed the videos, was a Cohen's kappa coefficient used to assess inter-rater reliability? If so, it should be included in the "statistics" section, otherwise it should be done and reported.

Line 177: Remove "ca.".

Line 191: Similar to the comment for line 119, was there any assessment done for inter-rater reliability?

Results:

Figure 2: Can the front/black sleep pictures be coloured? The black and white ones included are hard to see. The novel object exploration photos are also too small to see. Maybe these pictures should be made larger in a separate figure and included in the methods section instead.

Figure 3: Similar to the comments for Figure 2, maybe the pictures demonstrating the training protocol should be included as a separate figure in the methods section.

Discussion:

If part of the objective of the study was the compare male zebra finches with the findings from the cited studies on female zebra finches, then there needs to be more discussion regarding any effects sex may have on the changes observed in this study.

Author Response

This is an interesting study building on previous studies on the same topic. It is generally well-written, but some work needs to be done to improve the flow and clarity. Further work on the methods may need to be done, particularly regarding inter-rater reliability. More details are provided below.

Response: Thank you for your interest and thorough review of the manuscript. We have tried to address all the comments.

Comment #1: Simple summary:

Line 12 - 13: Change "a dim light at night protocol" to "dim light at night".

Response: Thank you. Changed.

Line 14: Remove "The".

Response: Removed.

Line 15: Remove "and" before "colour".

Response: Removed.

Line 20: "dLAN" abbreviation not previously used. Include in parenthesis at first mention of "dim light at night" above.

Response: Thank you. We have revised the line.

Comment #2: Abstract:

Line 30: "LD" abbreviation not previously explained.

Response: Thank you. We have explained the LD abbreviation.

Introduction:

Line 43: Remove "both".

Response: Removed.

Line 44: Is "clock" referring to the biological/endogenous clock? Please elaborate.

Response: Thank you. Yes, clock referred to the endogenous clock. We have mentioned this in introduction.

Line 47: Please provide references supporting this statement. Is "reduced" supposed to be "deduced"?

Response: Thank you. We have provided the reference and revised the sentence.

Line 80: Why were male zebra finches chosen rather than female as in the previous study cited? Please explain and justify.

Response: Thank you. We have used only one sex to eliminate the sex effects and also for the novelty as the physiological responses to the dim light at night on females has already been discussed in Batra et al., 2020, Environmental Pollution. We have discussed this now in the manuscript.

Comment #3: Methods:

Line 107 - 111: Please provide a figure demonstrating each of these scores.

Response: Thank you. We have given the figure demonstrating scores in the Figure 1.

Line 112 - 113: It may also be useful to provide a figure demonstrating this. Elaboration of the "different times" may also be important.

Response: Thank you. We have given the experimental protocol in detail in Figure 1 and mentioned in the methods section.

Line 119: If 2 people assessed the videos, was a Cohen's kappa coefficient used to assess inter-rater reliability? If so, it should be included in the "statistics" section, otherwise it should be done and reported.

Response: Thank you. It was wrongly stated as we have not used 2 people to assess each video. We have revised the statement. We have used researchers who were unaware of the light parameters and overall experimental condition to remove the experimenter’s bias, if any.

Line 177: Remove "ca.".

Response: Removed.

Line 191: Similar to the comment for line 119, was there any assessment done for inter-rater reliability?

Response: Thank you. Similar to behavioural videos, all the test videos are scored by persons unaware of the light conditions, similar to Taufique et al., 2016, Behavioral Brain Research and Taufique et al., 2018, European Journal of Neuroscience.

Comment #4: Results:

Figure 2: Can the front/black sleep pictures be coloured? The black and white ones included are hard to see. The novel object exploration photos are also too small to see. Maybe these pictures should be made larger in a separate figure and included in the methods section instead.

Response: Thank you. We have now made a separate Figure 1, specially dedicated for the methods. Since, all the sleep pictures are taken from night-vision IR camera, they are black and white and we don’t think artificially colouring them would be a good idea. We have now added better pictures with higher resolutions.

Figure 3: Similar to the comments for Figure 2, maybe the pictures demonstrating the training protocol should be included as a separate figure in the methods section.

Response: Thank you. This has been included in the Figure 1 separately.

Comment #5: Discussion:

If part of the objective of the study was the compare male zebra finches with the findings from the cited studies on female zebra finches, then there needs to be more discussion regarding any effects sex may have on the changes observed in this study.

Response #5: Thank you. We have discussed the changes observed in the study with females.

Round 2

Reviewer 2 Report

Thank you to the authors for your responses to my comments and for your revisions to the manuscript. Most of my comments and concerns have been addressed and I find the manuscript much improved. However, there are still some outstanding issues that I believe need to be resolved before the manuscript is ready for publication.

General comments:

1. Thank you for your response to my concerns about the light intensity used for the control conditions. I also want to highlight that from a physical perspective, “0 lux” does not exist; it just means that the light intensity is below the threshold that can be detected by your light meter. Consequently, rather than stating that the control conditions (and measured conditions) were 0 lux, you should provide a threshold based on the sensitivity of the radiometer (< X lux). In addition, it would better to use the term “darkness” rather than “complete darkness” throughout the manuscript.
2. I also strongly suggest checking and reporting the accuracy of the radiometer at low light intensities. Based on your response, it seems like the Macam Q203 Quantum radiometer may not be sensitive enough to detect light intensities below around 0.05 lux or so (which you might expect to observe around a university). Lux meters also tend to be less accurate towards their detection limits. For example, some lux meters can have an uncertainty of even 3 lux at lower light intensities, which would be critically important for your experiment.
3. As you have mentioned, even dim light at night can influence animal behaviour and physiology in a dose-dependent manner. If the light intensity used for your control condition is unrealistically low (which it could be, if the radiometer cannot distinguish between low light and almost no light), then you may be overestimating effects of dLAN on behaviour. Please also include this point in the ‘limitations’ paragraph of your discussion.
4. The added information regarding the novel object exploration test is helpful. Still, I think the link between the test and ‘mood’ or ‘depression’ is overstated and some citations seem misleading. Of the studies cited in line 312, only one (Taufique et al, 2018) actually seems to use a novel object exploration test as a measure of ‘mood’ or ‘depressive-like symptoms’. This study also did not use the novel exploration test as the sole measure of mood, but as just one response in a suite of responses that also included self-mutilation, decreased eating, and decreased self-grooming. Consequently, it would be more appropriate to remove mentions of ‘depression’ in the manuscript and to change the text to more accurately represent the existing literature. For example, lines 310-312 could instead read: “To look at the effect on mood, we performed the novel object exploration test, which has been previously used to assay mood in other songbirds (34)”.

Title:

5. The new title is clearer and more concise but does not fully encapsulate the study. Something like: “Effects of night illumination on behaviour, body mass and learning in male Zebra finches” would perhaps be more appropriate.

Methods:

6. Lines 101-102 and 116: When I requested more information on the spectral composition of the lighting, I meant the spectral power at different wavelengths (e.g. Fig S1 in de Jong et al. 2017, Journal of Biological Rhythms: https://doi.org/10.1177/0748730417719168). This is important because even white fluorescent lights vary in their emission of different wavelengths, including blue light, which can affect animal behaviour and physiology. Please also include this information for the dLAN treatment (Line 116).
7. There is some inconsistency regarding the number of people who analysed the videos, which needs to be clarified in the manuscript. In your response to my review, you say there were two researchers, and the manuscript refers to “researchers” in the plural. But in your response to another reviewer (report 1), you seem to dismiss the need for an analysis of inter-scorer reliability because “we have not used 2 people to assess each video”. If there was more than one person analysing videos, then there should be some analysis of inter-scorer reliability.

Discussion:

8. In your response you state that you removed the statement “akin to humans” but it is still there.

Author Response

Thank you to the authors for your responses to my comments and for your revisions to the manuscript. Most of my comments and concerns have been addressed and I find the manuscript much improved. However, there are still some outstanding issues that I believe need to be resolved before the manuscript is ready for publication.

Response: Thank you for your excellent inputs. We have tried to address each question sincerely.

General comments:

Comment #1: Thank you for your response to my concerns about the light intensity used for the control conditions. I also want to highlight that from a physical perspective, “0 lux” does not exist; it just means that the light intensity is below the threshold that can be detected by your light meter. Consequently, rather than stating that the control conditions (and measured conditions) were 0 lux, you should provide a threshold based on the sensitivity of the radiometer (< X lux). In addition, it would better to use the term “darkness” rather than “complete darkness” throughout the manuscript.

Response #1: Thank you. We have added the threshold of the radiometer (<0.01 lux). We have now used “darkness” in place of complete darkness.

Comment #2: I also strongly suggest checking and reporting the accuracy of the radiometer at low light intensities. Based on your response, it seems like the Macam Q203 Quantum radiometer may not be sensitive enough to detect light intensities below around 0.05 lux or so (which you might expect to observe around a university). Lux meters also tend to be less accurate towards their detection limits. For example, some lux meters can have an uncertainty of even 3 lux at lower light intensities, which would be critically important for your experiment.

Response #2: Thank you. We have checked the radiometer. In the response #1, we have mentioned the lowest threshold of light measured accurately. It is possible that we missed the very low light intensity or fluctuations due to the limits of radiometer. However, as an experimenter we tried to measure as accurately as possible. We have also added this in methods section and discussion.

Comment #3: As you have mentioned, even dim light at night can influence animal behaviour and physiology in a dose-dependent manner. If the light intensity used for your control condition is unrealistically low (which it could be, if the radiometer cannot distinguish between low light and almost no light), then you may be overestimating effects of dLAN on behaviour. Please also include this point in the ‘limitations’ paragraph of your discussion.

Response #3: Thank you. We have mentioned this in limitations of the study.

Comment #4: The added information regarding the novel object exploration test is helpful. Still, I think the link between the test and ‘mood’ or ‘depression’ is overstated and some citations seem misleading. Of the studies cited in line 312, only one (Taufique et al, 2018) actually seems to use a novel object exploration test as a measure of ‘mood’ or ‘depressive-like symptoms. This study also did not use the novel exploration test as the sole measure of mood, but as just one response in a suite of responses that also included self-mutilation, decreased eating, and decreased self-grooming. Consequently, it would be more appropriate to remove mentions of ‘depression’ in the manuscript and to change the text to more accurately represent the existing literature. For example, lines 310-312 could instead read: “To look at the effect on mood, we performed the novel object exploration test, which has been previously used to assay mood in other songbirds (34)”.

Response #4: Thank you. We have revised the discussion and removed “depression” with appropriate word according to the literature. We have revised the line 310-312.

Title:

Comment #5: The new title is clearer and more concise but does not fully encapsulate the study. Something like: “Effects of night illumination on behaviour, body mass and learning in male Zebra finches” would perhaps be more appropriate.

Response #5: Thank you for suggesting a better title. We highly appreciate it.

Methods:

Comment #6: Lines 101-102 and 116: When I requested more information on the spectral composition of the lighting, I meant the spectral power at different wavelengths (e.g. Fig S1 in de Jong et al. 2017, Journal of Biological Rhythms: https://doi.org/10.1177/0748730417719168). This is important because even white fluorescent lights vary in their emission of different wavelengths, including blue light, which can affect animal behaviour and physiology. Please also include this information for the dLAN treatment (Line 116).

Response #6: Thank you. We have used white light (cool daylight) having colour temperature of 6500K. Currently, we do not have the facility to measure spectral power at different wavelengths like de Jong et al. 2017 (Journal of Biological Rhythms: https://doi.org/10.1177/0748730417719168). As correctly mentioned by you, white fluorescent lights vary in their emission of different wavelengths and in the present study (cool daylight at 6500K) high amount of blue intensity could be more prevalent as per reports (Chellappa et al. 2011, Non-visual effects of light on melatonin, alertness and cognitive performance: can blue-enriched light keep us alert?. PloS one, 6(1), e16429.). Since we did not measure the spectral property, we will keep this in mind for future experiments. Thank you for enlightening us. We have mentioned this in methods and added as a limitation in the discussion.

Comment #7: There is some inconsistency regarding the number of people who analysed the videos, which needs to be clarified in the manuscript. In your response to my review, you say there were two researchers, and the manuscript refers to “researchers” in the plural. But in your response to another reviewer (report 1), you seem to dismiss the need for an analysis of inter-scorer reliability because “we have not used 2 people to assess each video”. If there was more than one person analysing videos, then there should be some analysis of inter-scorer reliability.

Response #7: Thank you. We apologize for the misunderstanding from our part. Every behaviour was assessed by single researcher unaware of the experimental protocol. We have clarified this in manuscript. If it had been multiple researchers for the same, we would have provided the inter-scorer reliability.

Discussion:

Comment #8: In your response you state that you removed the statement “akin to humans” but it is still there.

Response #8: We apologize for the inadvertent mistake. We have revised the statement.

Reviewer 3 Report

The improvements made to the manuscript have increased the overall scientific value. There is just one more correction below to be made before full acceptance.

Line 308: "fragmentation" was misspelt. Please correct.

Author Response

The improvements made to the manuscript have increased the overall scientific value. There is just one more correction below to be made before full acceptance.

Response: Thank you for your kind review.

Comment #1: Line 308: "fragmentation" was misspelt. Please correct.

Response #1: Corrected.