Assessment of Pupillometry Across Different Commercial Systems of Laying Hens to Validate Its Potential as an Objective Indicator of Welfare

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This study describes an innovative method for non-invasive detection of iris-to-pupil (IP) ratio in chickens with a view to achieving real-time and reliable monitoring of chickens' welfare status and optimizing production management practices. However, the following points were found during the review process.

1. The authors fail to illustrate the necessity of the development and application of non-invasive measurement through more significant evidence and clear figures in the opening part.
2. Some of the punctuation marks in the article are not correctly formatted, such as the lack of a period at the end of the sentence in line 49 and the absence of commas before some “and”.
3. In the introduction, the authors mention that chickens show pupil dilation in response to stimuli, suggesting that pupil changes may serve as a quantitative indicator of welfare, but this speculation has not been verified in the chicken species, and there is a big difference between qualitative and quantitative, so this study does not have a sufficient foundation for prior research.
4. The authors chose conditions in the environment that may cause stress for comparison in the experimental design, but as stated in lines 57-59 of this paper, environment conditions are indirect indicators that do not directly reflect the level of stress in chickens and should be verified by physiological or behavioral indicators.
5. Although the authors mention that breed differences may be influenced by the type of farming, the potential influence of the breed itself on IP ratios was not sufficiently excluded during the analysis for the comparison of IP ratios of two breeds under different farming systems.
6. Although the study measured ambient light intensity, in practice, there may be other environmental factors that were not measured or controlled, such as temperature, humidity, noise, etc., which may also have an effect on the physiological state and IP ratio of hens, thus causing some interference with the results.

Author Response

1. The authors fail to illustrate the necessity of the development and application of non-invasive measurement through more significant evidence and clear figures in the opening part.

Response: Thank you for this valuable comment. We acknowledge the reviewer’s observation regarding the opening section of our manuscript. Our intent was to highlight the need for non-invasive welfare assessment methods by referencing the limitations of current invasive or subjective practices (Outlined in introduction section line 50 to 68). It is important to note that, to date, there is limited prior empirical evidence supporting the application of pupillometry in poultry, which underscores the novelty and necessity of our investigation.

2. Some of the punctuation marks in the article are not correctly formatted, such as the lack of a period at the end of the sentence in line 49 and the absence of commas before some “and”.

Response: We have reviewed the manuscript and corrected the typographic errors in the revised manuscript.

3. In the introduction, the authors mention that chickens show pupil dilation in response to stimuli, suggesting that pupil changes may serve as a quantitative indicator of welfare, but this speculation has not been verified in the chicken species, and there is a big difference between qualitative and quantitative, so this study does not have a sufficient foundation for prior research.

Response: Thank you for this thoughtful and important observation. We agree with the reviewer that distinguishing between qualitative and quantitative interpretations of physiological responses is crucial. In our introduction, the reference to pupil dilation in chickens in response to stimuli was intended to highlight a well-established physiological mechanism that is, the autonomic regulation of pupil diameter in response to environmental cues. This response, mediated by the sympathetic and parasympathetic nervous systems, has been described in avian species, including chickens, as a natural part of their neurophysiological regulation. However, we fully acknowledge that while this autonomic response is known, its application as a quantitative measure of welfare in poultry has not been previously validated. The central aim of our study is to bridge this gap by demonstrating how this physiological response specifically the iris-to-pupil (IP) ratio can be used as a quantitative, non-invasive welfare indicator. We have sought to do this by correlating IP ratios across different commercial husbandry systems that are known to influence welfare outcomes in laying hens.

4. The authors chose conditions in the environment that may cause stress for comparison in the experimental design, but as stated in lines 57-59 of this paper, environment conditions are indirect indicators that do not directly reflect the level of stress in chickens and should be verified by physiological or behavioral indicators.

Response: Thank you for this thoughtful and constructive comment. We fully agree with the reviewer that environmental conditions, while relevant to welfare assessments, are indirect indicators and do not directly reflect the physiological stress levels experienced by chickens. This recognition was central to the motivation behind our study. In response to this limitation, we aimed to evaluate a more objective, quantifiable, and non-invasive physiological indicator, the iris-to-pupil (IP) ratio which reflects autonomic nervous system activity and has the potential to serve as a direct measure of stress. While physiological and behavioural indicators such as corticosterone levels or feather pecking provide valuable welfare insights, it is important to recognize that they often operate above certain thresholds of detection, may require invasive procedures, or be influenced by observer bias. In contrast, the IP ratio demonstrates high sensitivity to subtle changes in the internal state of the animal and can be measured without handling, minimizing confounding effects of the assessment process itself. Our study findings show clear, statistically significant differences in IP ratio across varying husbandry systems, underscoring its responsiveness and utility in detecting welfare differences that may not be captured through conventional means.

5. Although the authors mention that breed differences may be influenced by the type of farming, the potential influence of the breed itself on IP ratios was not sufficiently excluded during the analysis for the comparison of IP ratios of two breeds under different farming systems.

Response: Thank you for this valuable observation. We agree that breed differences could potentially influence physiological parameters such as the iris-to-pupil (IP) ratio. However, in our study, we carefully considered this possibility and analysed the data to distinguish between breed effects and housing system effects. Although initial comparisons showed that Lohmann Brown (LB) hens had significantly higher IP ratios than White Leghorn (WL) hens, it is important to note that these breeds were not uniformly distributed across all housing systems. Specifically, all WL birds were housed exclusively in the pen system, while LB birds were represented across all four housing environments (outdoor, indoor, scratch, and pen). To address this imbalance and examine breed differences more objectively, we conducted a focused comparison of IP ratios between the two breeds under similar housing conditions, i. This comparison revealed no significant difference in IP ratios between the breeds under these matched environmental conditions. This finding strongly suggests that the observed overall difference in IP ratio between LB and WL breeds was more likely a reflection of the influence of husbandry system rather than an intrinsic breed effect.

6. Although the study measured ambient light intensity, in practice, there may be other environmental factors that were not measured or controlled, such as temperature, humidity, noise, etc., which may also have an effect on the physiological state and IP ratio of hens, thus causing some interference with the results.

Response: Thank you for this thoughtful and important comment. We fully acknowledge that environmental factors such as temperature, humidity, and noise can influence the physiological state of animals, including pupil dynamics. While our study focused on measuring ambient light intensity due to its direct effect on pupillary response, we agree that unmeasured environmental variables could act as potential confounders. To address this limitation, we incorporated a within-farm comparative design in our study, specifically highlighted in Figure 3C and 3D, where birds from the same farm (Kelly and Olive) were evaluated under different housing systems (e.g., outdoor, indoor, and scratch). By comparing birds that were subject to the same macro-environmental conditions (e.g., geography, general management, and weather patterns) but housed in differing systems, we effectively minimized inter-farm environmental variability. This design strengthens the reliability of our findings by isolating the influence of housing system itself on IP ratio.

Reviewer 2 Report

Comments and Suggestions for Authors

I accepted the review with interest, because I am interested in the behavior and welfare of poultry. The topic of using pupillometry seems very interesting and has great potential, however, due to the way the work is written, I have many doubts whether the research was properly conducted and the data analyzed. For this reason, my critical comments mainly refer to the chapters: Material and methods, and Results.

Material and methods

For greater clarity, I suggest dividing this chapter into sections. The methodology itself is so poorly written that there is no way to fully understand what the authors did and how, and therefore to replicate the research by another team to verify it.

In my opinion, the explanation of why the application for the performance of animal experiments was abandoned should be included in the note "Ethical statement"

Please explain the concept of IP. What does the larger or smaller value of it mean? In the Results section, it is difficult to understand which result the authors consider better/worse and why.

Please arrange the order and expand the description of the conditions of keeping individual groups of birds. For example, the authors list: "hobby free-range (6 birds), free-range (169 birds), organic EU-compliant (94 birds), and loose caged (indoor pen-based housing) (27 birds)", but describe the organic system, free-range system, hobby farm. Please describe everything in detail about the subsequent systems. In the first system, birds in the number of ... were kept in (description of conditions: what birds, flock size, lighting program, nutrition, henhouse equipment, etc.)

„A subgroup analysis was conducted using data from birds at Kelly and Olive farms to assess differences in the IP ratio across three housing sub-environments, i.e., outdoor, indoor, and scratch areas.” – I don't understand how sub-environments were separated. Were these the same birds, from the same system, but tested in different areas of the farm? Did you test different systems within one farm? If so, what did these systems look like?

“This subgrouping allowed for a targeted comparison of IP ratios among birds exposed to different levels of environmental enrichment and access within the same farm” – this raises my big doubts.Pupils react to light, so did you control the light intensity in these environments, because I imagine that the light is completely different in outdoor and indoor environments.I understand that the authors later describe that light had no effect (correlations were studied), but I am not convinced that the way it was done is appropriate.In my opinion, they should have studied the same birds, in different lighting conditions, but in the same place.Then there would be no doubt that the light factor can be eliminated as irrelevant from the point of view of pupillometry.The way it was done in this study leaves too many independent factors that could not be taken into account and this could have falsified the results.

Since the breeds used in the study were Lohmann Brown and White Leghorn, did the authors take into account that they have different reactivity and response to stimuli? The authors could compare whether the method they describe shows differences between breeds, if not, then compare the systems without taking breed into account.

In the results, the authors noted that differences in IP between breeds were due to housing systems, there is no indication that breed was important. However, I think it would be appropriate to exclude White Leghorn from the pooled analyses and stay with one breed.  In my opinion, the authors have too many variables in their analysis whose impact they cannot realistically assess (e.g. they compare 27 Leghorn hens and almost 300 Lohman Brown hens).

„Hens were sampled randomly within each system, ensuring adequate representation of each breed where available.” – - what does adequate representation mean? How was it assessed to be an adequate representation?

„Photographs of the hens' eyes were taken using a high-resolution digital camera” – What camera was used? Please specify specifications or model. From what distance were the photographs taken? What was the procedure for taking the photograph?

Although the authors include the age of the birds in the results, there is no word about this in the methodology

What was the light intensity for each system and farm? Was the number of lx measured once for each system or with each photograph?

The methodology raises a lot of questions about what has been done and how it has been done, and further, whether mistakes have been made at the planning stage. Perhaps the authors can defend their assumptions? First, however, they need to reorganize this section to make it understandable. Only after I read it several times, also after reading the entire article, did I begin to guess what was done and why.

Results

As with the previous chapter, I suggest dividing it into sections.

The correlation coefficient for lighting and age is 0.25 and 0.05, respectively. In the next sentence, the authors write: “low correlation coefficient, especially with a light intensity…” – after all, the coefficient is higher than for age. Besides, although the coefficient is 0.25, so the correlation is weak, but it is there. In addition, I would not call a correlation at the level of 0.05 significant, and the authors suggest this when discussing it.

I wonder about the reactivity of birds' pupils to light. Are there any such studies? Maybe the authors should refer to this in the introduction or even mention the methodology.

Figure 1: I wonder what could have caused the higher IP in the group of birds with 5000 lx lighting? What birds were they? As for age, I see that there were only a few birds from the oldest age category (or were all the observations so similar?). This could be the reason for this insignificant in my opinion, and according to the authors somewhat (?) significant correlation – data distortion due to the small size of the group.

If we have 6 birds from the hobby free range system, and much more in the others, then maybe this is the reason for the differences shown by this system compared to the others. I think that comparing 6 birds with almost a hundred and 27 from another group is quite a stretch. The general rule of parametric tests is that the largest group should not be more than twice as large as the smallest. 6 birds is absolutely not a representative group!

„At Kelly Farm, which operated under a free-range system, birds were housed in three distinct environments: outdoor, indoor, and scratch…” – they were housed there permanently or they chose to be there?

I think that instead of describing so many configurations of the experience, it would be better if the authors focused and planned on one or two variants.

Discussion

… because pupil size is also responsive to changes in ambient light, dilating in low light and constricting in bright light, it is essential that IP ratio measurements are conducted under controlled or consistently measured lighting conditions to ensure that observed differences reflect physiological stress responses rather than lighting artifacts” - in the discussion the authors draw attention to this, but they did not take it into account in the planning of the experiment. In general, I consider the discussion and introduction correct (although it is difficult for me to comment on the discussion due to the comments regarding the conduct of the experiment and the analysis of the results)

Author Response

Material and methods

For greater clarity, I suggest dividing this chapter into sections. The methodology itself is so poorly written that there is no way to fully understand what the authors did and how, and therefore to replicate the research by another team to verify it.

Response: Thank you for this important observation. In the revised manuscript, we have carefully restructured the Methods section into clearly defined subsections to enhance clarity and transparency. Each step of the methodology including study design, animal cohorts, housing systems, photographic procedures, light measurement, digital analysis, and statistical methods has been described in detail. Our aim was to ensure that the procedures are sufficiently documented so that the study can be fully understood and reliably replicated by other researchers. We appreciate the reviewer’s comment, which helped us improve the scientific rigor and reproducibility of the manuscript.

In my opinion, the explanation of why the application for the performance of animal experiments was abandoned should be included in the note "Ethical statement"

Response: Thank you for this thoughtful comment. We respectfully believe that the purpose of the "Ethical Statement" is to address ethical considerations relevant to the specific type of research conducted, rather than to explain why alternative research formats such as animal experiments were not pursued. Research can take many forms, and it would not be practical or customary to include justifications for the exclusion of every possible methodology. In our case, we conducted a fully non-invasive study involving observational data collection under natural husbandry conditions, with no experimental manipulation or intervention. Accordingly, our ethical statement was focused on confirming that no procedures requiring animal experimentation approval were performed, and that verbal consent from farm owners was obtained. We hope this clarifies our approach and reasoning.

Please explain the concept of IP. What does the larger or smaller value of it mean? In the Results section, it is difficult to understand which result the authors consider better/worse and why.

Response: Thank you for your valuable comment. In the revised manuscript, we have included an additional explanatory paragraph in Methods sections to clarify the concept and interpretation of the iris-to-pupil (IP) ratio. Specifically, we explain that the IP ratio is a physiological indicator reflecting the balance between sympathetic and parasympathetic nervous system activity. A larger IP ratio corresponds to greater pupil dilation, typically associated with increased sympathetic activation and heightened arousal or stress, while a smaller IP ratio indicates greater parasympathetic activity, suggestive of a more relaxed state. This clarification helps contextualize the results across different housing systems, allowing readers to better understand which conditions were associated with more favourable welfare outcomes. We appreciate your suggestion, which helped improve the clarity of our findings.

Please arrange the order and expand the description of the conditions of keeping individual groups of birds. For example, the authors list: "hobby free-range (6 birds), free-range (169 birds), organic EU-compliant (94 birds), and loose caged (indoor pen-based housing) (27 birds)", but describe the organic system, free-range system, hobby farm. Please describe everything in detail about the subsequent systems. In the first system, birds in the number of ... were kept in (description of conditions: what birds, flock size, lighting program, nutrition, henhouse equipment, etc.)

Response: Thank you for this valuable suggestion. In response to the reviewer’s comment, we have revised the Methods section of the manuscript to provide a clearer and more structured description of each housing system evaluated in this study. We have arranged the order of presentation to match the bird groupings, and for each system, we now include detailed information regarding the number of birds sampled, type of housing, environmental and management conditions. This reorganization and expansion aim to enhance clarity, facilitate replication, and address the variability in environmental factors that may influence welfare-related physiological parameters such as the iris-to-pupil (IP) ratio. We appreciate the reviewer’s input, which has contributed to improving the manuscript.

„A subgroup analysis was conducted using data from birds at Kelly and Olive farms to assess differences in the IP ratio across three housing sub-environments, i.e., outdoor, indoor, and scratch areas.” – I don't understand how sub-environments were separated. Were these the same birds, from the same system, but tested in different areas of the farm? Did you test different systems within one farm? If so, what did these systems look like?

Response: Thank you for raising this important point. We agree that clarification was needed regarding how the housing sub-environments were defined and assessed. In response, we have revised the Methods section of the manuscript to clearly explain that the subgroup analysis at Kelly and Olive farms involved different groups of birds housed in distinct areas of the same farm, not the same birds tested in multiple locations. At Kelly Farm, which operates under a free-range system, birds were managed in three separate environments within the same farm: an outdoor area (n = 60), an indoor housing area (n = 38), and a scratch area (n = 71). These areas functioned as consistent and separate sub-environments, each hosting different sets of birds. Similarly, at Olive Farm, following an EU-compliant organic system, birds were housed either in the outdoor environment (n = 42) or in the scratch area (n = 52), with no overlap of individuals across settings. These comparisons allowed us to assess the influence of immediate environmental conditions on the iris-to-pupil (IP) ratio while minimizing confounding variables by focusing on groups reared under the same farm management practices. This clarification has been incorporated into the revised manuscript to improve transparency and interpretability of the study design.

“This subgrouping allowed for a targeted comparison of IP ratios among birds exposed to different levels of environmental enrichment and access within the same farm” – this raises my big doubts. Pupils react to light, so did you control the light intensity in these environments, because I imagine that the light is completely different in outdoor and indoor environments. I understand that the authors later describe that light had no effect (correlations were studied), but I am not convinced that the way it was done is appropriate. In my opinion, they should have studied the same birds, in different lighting conditions, but in the same place. Then there would be no doubt that the light factor can be eliminated as irrelevant from the point of view of pupillometry. The way it was done in this study leaves too many independent factors that could not be taken into account and this could have falsified the results.

Response: Thank you for this insightful comment, which highlights a valid concern regarding potential confounding factors, particularly ambient light, in our subgroup analyses. We have clarified the Methods section in the revised manuscript to address this issue more explicitly. The subgroup analysis at Kelly and Olive farms involved different groups of birds, each housed in distinct sub-environments within the same farm, and not the same birds tested under multiple conditions. While we understand the value of testing the same individuals under varying light conditions, doing so would have fundamentally altered our study’s design from an observational, field-based welfare assessment to an experimental manipulation, which was not the goal of our work. Our objective was to evaluate the real-world implications of existing husbandry systems as they are currently practiced on farms. To address this limitation, we incorporated a within-farm comparative design into the study, which is specifically highlighted in Figure 3C and 3D. In this design, birds from the same farm (Kelly and Olive) were assessed under different housing systems (outdoor, indoor, and scratch areas) within the same macro-environment (e.g., geographic location, general management practices, and climatic conditions). This approach helped to minimize inter-farm environmental variability while still capturing the influence of sub-environmental conditions, such as access to natural stimuli or level of environmental complexity, on the iris-to-pupil (IP) ratio. We also performed correlation analyses to evaluate the effect of light intensity on IP ratio (Figure 1A), and found only a very weak correlation (r = 0.25), suggesting that ambient light variations in real-world farm settings had minimal influence on this physiological parameter. While we acknowledge that a fully controlled lighting study could offer further insight, our current approach aimed to preserve ecological validity and reflect practical farm conditions in welfare monitoring.

In the revised manuscript we have added a paragraph in the discussion section to highlight this limitations of our study.  

Since the breeds used in the study were Lohmann Brown and White Leghorn, did the authors take into account that they have different reactivity and response to stimuli? The authors could compare whether the method they describe shows differences between breeds, if not, then compare the systems without taking breed into account. In the results, the authors noted that differences in IP between breeds were due to housing systems, there is no indication that breed was important. However, I think it would be appropriate to exclude White Leghorn from the pooled analyses and stay with one breed.  In my opinion, the authors have too many variables in their analysis whose impact they cannot realistically assess (e.g. they compare 27 Leghorn hens and almost 300 Lohman Brown hens).

Response: Thank you for this valuable observation. We agree that breed differences could potentially influence physiological parameters such as the iris-to-pupil (IP) ratio. However, in our study, we carefully considered this possibility and analysed the data to distinguish between breed effects and housing system effects. Although initial comparisons showed that Lohmann Brown (LB) hens had significantly higher IP ratios than White Leghorn (WL) hens, it is important to note that these breeds were not uniformly distributed across all housing systems. Specifically, all WL birds were housed exclusively in the pen system, while LB birds were represented across all four housing environments (outdoor, indoor, scratch, and pen). To address this imbalance and examine breed differences more objectively, we conducted a focused comparison of IP ratios between the two breeds under similar housing conditions, i. This comparison revealed no significant difference in IP ratios between the breeds under these matched environmental conditions. This finding strongly suggests that the observed overall difference in IP ratio between LB and WL breeds was more likely a reflection of the influence of husbandry system rather than an intrinsic breed effect.

„Hens were sampled randomly within each system, ensuring adequate representation of each breed where available.” – - what does adequate representation mean? How was it assessed to be an adequate representation?

Response: Thank you for highlighting the need for clarity regarding the representation of breeds within each housing system. By “adequate representation,” we refer to the inclusion of a sufficient number of birds from each breed present within the respective system to allow meaningful comparison and subgroup analyses. We ensured this by sampling birds proportionally to their presence in the flock, maintaining a minimum number of individuals per breed to provide reliable data while reflecting the actual population structure on each farm. We have included the following paragraph in the methods section of the revised manuscript to address this concern.

Within each housing system, hens were randomly sampled to reflect the breed composition present in the flock. Specifically, the number of birds selected from each breed was proportional to their relative abundance on the farm, ensuring that both Lohmann Brown and White Leghorn birds were adequately represented where present. This approach aimed to capture the natural variation within each system and provide sufficient sample sizes for breed-specific analyses. Randomization was performed using systematic sampling methods to minimize selection bias and to ensure the representativeness of the sampled birds relative to the broader flock demographics.

„Photographs of the hens' eyes were taken using a high-resolution digital camera” – What camera was used? Please specify specifications or model. From what distance were the photographs taken? What was the procedure for taking the photograph?

Response: Thank you for your valuable suggestion. We have added detailed information about the camera specifications, the distance from which the photographs were taken, and the procedure for capturing the images in the revised Methods section to improve clarity and reproducibility.

Although the authors include the age of the birds in the results, there is no word about this in the methodology

Response: Thank you for this insightful comment. As this study was conducted under real-world commercial farm conditions rather than as a controlled experimental trial, the age of the birds could not be predetermined or standardized across all groups. Therefore, age was not included in the methodology section, as it was not part of the initial study design. However, bird age was recorded during data collection and analysed retrospectively as part of our findings, which is why it appears in the Results section. We believe this approach aligns with the observational nature of the study and provides additional context for interpreting the outcomes.

What was the light intensity for each system and farm? Was the number of lx measured once for each system or with each photograph?

Response: Thank you for this important question. In the revised manuscript, we have clarified that light intensity was measured at the time each photograph was taken using a light meter app on the iPhone 13. The ambient lighting conditions were recorded for every individual bird sampled, rather than relying on a single measurement per system. Additionally, the distribution of light intensities across all farms and housing systems has been included in Figure 1B to provide a clear representation of the range and variability of lighting conditions during image acquisition.

The methodology raises a lot of questions about what has been done and how it has been done, and further, whether mistakes have been made at the planning stage. Perhaps the authors can defend their assumptions? First, however, they need to reorganize this section to make it understandable. Only after I read it several times, also after reading the entire article, did I begin to guess what was done and why.

Response: Thank you for your detailed feedback. We acknowledge that the initial version of the methodology may not have provided sufficient clarity and could have raised concerns about study design and execution. In response, we have thoroughly revised and reorganized the Methods section in the updated manuscript to ensure greater transparency, coherence, and replicability. We have clarified the rationale behind key design choices, described procedures in more detail, and structured the section to enhance readability. We hope that the revised version now adequately addresses your concern.

Results

As with the previous chapter, I suggest dividing it into sections.

Response: Thank you for your suggestion. In response, we have organized the Results section into distinct paragraphs presented in a sequential manner to improve clarity and logical flow. This structuring highlights each major finding individually, making it easier for readers to follow the progression of results and understand the relationship between experimental variables and outcomes.

The correlation coefficient for lighting and age is 0.25 and 0.05, respectively. In the next sentence, the authors write: “low correlation coefficient, especially with a light intensity…” – after all, the coefficient is higher than for age. Besides, although the coefficient is 0.25, so the correlation is weak, but it is there. In addition, I would not call a correlation at the level of 0.05 significant, and the authors suggest this when discussing it.

Response: Thank you for this thoughtful observation. We acknowledge that neither correlation is strong enough to be considered significant in influencing the IP ratio.

I wonder about the reactivity of birds' pupils to light. Are there any such studies? Maybe the authors should refer to this in the introduction or even mention the methodology.

Response: We are not aware of any.

Figure 1: I wonder what could have caused the higher IP in the group of birds with 5000 lx lighting? What birds were they? As for age, I see that there were only a few birds from the oldest age category (or were all the observations so similar?). This could be the reason for this insignificant in my opinion, and according to the authors somewhat (?) significant correlation – data distortion due to the small size of the group.

Response: The instances of higher IP ratios observed in birds exposed to 5000 lx lighting were indeed limited in number and may be attributed to normal biological variation within the population. Given our random sampling approach, which was intended to capture real-world variability across commercial systems, we did not control for all individual-level factors that might influence pupil dynamics. As such, we are unable to definitively attribute the elevated IP ratios in this subgroup to any specific cause. We have clarified this limitation in the revised discussion section.

We would like to clarify that the oldest age category (~104 weeks) included 27 birds, all of which were White Leghorns housed in floor pen systems. While we acknowledge that this group was smaller compared to some other age categories, the sample size was still sufficient to include in correlation analysis. However, we agree that the limited variation within this subgroup both in terms of breed and housing could have contributed to the low correlation observed between age and IP ratio.

If we have 6 birds from the hobby free range system, and much more in the others, then maybe this is the reason for the differences shown by this system compared to the others. I think that comparing 6 birds with almost a hundred and 27 from another group is quite a stretch. The general rule of parametric tests is that the largest group should not be more than twice as large as the smallest. 6 birds is absolutely not a representative group!

Response: We have clarified this limitation in the revised discussion section.

„At Kelly Farm, which operated under a free-range system, birds were housed in three distinct environments: outdoor, indoor, and scratch…” – they were housed there permanently or they chose to be there?

Response: They were housed there permanently.

I think that instead of describing so many configurations of the experience, it would be better if the authors focused and planned on one or two variants.

Response: We appreciate the reviewer’s suggestion. However, we would like to clarify that this study was observational in nature, conducted under real-world farm conditions rather than as a tightly controlled experimental design. The diversity in configurations reflects the variability inherent in commercial and backyard poultry farming systems. Our goal was to capture this natural variation and report findings as they were observed, to enhance ecological validity and better understand how husbandry practices in practical settings relate to welfare indicators such as the IP ratio. We have clarified this view in the methods section of the revised manuscript.

Discussion

… because pupil size is also responsive to changes in ambient light, dilating in low light and constricting in bright light, it is essential that IP ratio measurements are conducted under controlled or consistently measured lighting conditions to ensure that observed differences reflect physiological stress responses rather than lighting artifacts” - in the discussion the authors draw attention to this, but they did not take it into account in the planning of the experiment. In general, I consider the discussion and introduction correct (although it is difficult for me to comment on the discussion due to the comments regarding the conduct of the experiment and the analysis of the results).

Response: Thank you for your insightful comment. We would like to clarify that this study was not designed as a controlled experimental investigation but rather as an observational study conducted under real-world commercial and hobby farm conditions. While it is well recognized that pupil size responds to ambient light, our findings indicate that variations in ambient light intensity across the farms studied did not have a major influence on the IP ratio. This supports the utility of the IP ratio as a robust physiological indicator under typical farming conditions. We have addressed this point further in discussion section.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The questions I raised have all been properly revised or explained.

Reviewer 2 Report

Comments and Suggestions for Authors

I have no other comments