Sixty Years of Tinbergen’s Four Questions and Their Continued Relevance to Applied Behaviour and Welfare Research in Zoo Animals: A Commentary

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This is a well-structured and well-written manuscript in which the authors present a critical review of the literature on Tinbergen’s approach to behavioral research (through the seminal 4 questions paradigm), and propose an insightful application of this theoretical framework for zoo animal welfare.

The topic was well introduced. The background literature was appropriately reviewed (mixing seminal/historical and recent/up-to-date key articles) and critically addressed. The objectives were clearly laid out and ultimately reached.

Overall, the layout of the manuscript is logical, its tone is light, and its style is clear; these qualities result in an easy and pleasant read. Findings from previous research were clearly summarized and put into perspective in light of this review manuscript. The authors did an excellent job at making a relatively complex conceptual analysis simple, which makes the manuscript accessible to a non-expert audience (while maintaining the interest of experts in this field!). I have no major issues with the outline/structure and conclusion of the manuscript. The figures are informative and relevant (but please see below for a critical comment). This is an interesting manuscript that also fits an information gap.

Before I can recommend this manuscript for publication, I just have a few comments that I would like the authors to acknowledge and possibly address in a revised version of the manuscript.

If the authors agree with my points, I anticipate they will be able to handle these comments easily. If the authors are willing to consider my suggestions below, then I am happy to recommend this manuscript for publication.

Figure 1. I am not convinced that the example provided under Phylogeny really matches the typical requirements that should apply to this Tinbergen’s question, which is about exploring the ancestral origins and evolutionary pathways leading to various forms of a trait. Please explain more clearly and possible revise.

I think your paper could benefit from including elements more recent and even more integrative proposition about TFQs. These authors’ main argument is that the dichotomy between Proximate and Ultimate causes (as emphasized in your manuscript, following Tinbergen’s tradition approach) is false. This approach should be updated by taking into account new findings supporting the view that the main distinction between research questions should be between causes and consequences of behavior (with the possible addition of behavioral structure between them).

Please find below two examples of such relevant references:

Bergman TJ, Beehner JC. Leveling with Tinbergen: Four levels simplified to causes and consequences. Evol Anthropol. 2022 Jan;31(1):12-19. doi: 10.1002/evan.21931.

Leca JB. Towards a three-level neo-Tinbergenian approach to object play: Structure, causes and consequences of a behavioral puzzle. Neurosci Biobehav Rev. 2023 Sep;152:105290. doi: 10.1016/j.neubiorev.2023.105290.

Author Response

This is a well-structured and well-written manuscript in which the authors present a critical review of the literature on Tinbergen’s approach to behavioral research (through the seminal 4 questions paradigm), and propose an insightful application of this theoretical framework for zoo animal welfare.

The topic was well introduced. The background literature was appropriately reviewed (mixing seminal/historical and recent/up-to-date key articles) and critically addressed. The objectives were clearly laid out and ultimately reached.

Overall, the layout of the manuscript is logical, its tone is light, and its style is clear; these qualities result in an easy and pleasant read. Findings from previous research were clearly summarized and put into perspective in light of this review manuscript. The authors did an excellent job at making a relatively complex conceptual analysis simple, which makes the manuscript accessible to a non-expert audience (while maintaining the interest of experts in this field!). I have no major issues with the outline/structure and conclusion of the manuscript. The figures are informative and relevant (but please see below for a critical comment). This is an interesting manuscript that also fits an information gap.

Before I can recommend this manuscript for publication, I just have a few comments that I would like the authors to acknowledge and possibly address in a revised version of the manuscript.

If the authors agree with my points, I anticipate they will be able to handle these comments easily. If the authors are willing to consider my suggestions below, then I am happy to recommend this manuscript for publication.

 

Thank you for the useful and development feedback and comments. We have actioned all comments accordingly and edited the manuscript in line with your revisions.

Figure 1. I am not convinced that the example provided under Phylogeny really matches the typical requirements that should apply to this Tinbergen’s question, which is about exploring the ancestral origins and evolutionary pathways leading to various forms of a trait. Please explain more clearly and possible revise.

Thank you for the comment. We have revised the phylogeny section accordingly to make our meaning and explanation clearer. The figure caption has also be edited.

I think your paper could benefit from including elements more recent and even more integrative proposition about TFQs. These authors’ main argument is that the dichotomy between Proximate and Ultimate causes (as emphasized in your manuscript, following Tinbergen’s tradition approach) is false. This approach should be updated by taking into account new findings supporting the view that the main distinction between research questions should be between causes and consequences of behavior (with the possible addition of behavioral structure between them).

Thank you for the useful information, we have included these new concepts within the manuscript, towards the end of the discussion.

Please find below two examples of such relevant references:

Bergman TJ, Beehner JC. Leveling with Tinbergen: Four levels simplified to causes and consequences. Evol Anthropol. 2022 Jan;31(1):12-19. doi: 10.1002/evan.21931.

Leca JB. Towards a three-level neo-Tinbergenian approach to object play: Structure, causes and consequences of a behavioral puzzle. Neurosci Biobehav Rev. 2023 Sep;152:105290. doi: 10.1016/j.neubiorev.2023.105290.

Thank you for the useful and development references. We have included these sources in the paper (at the end of section 5.3 before the Conclusion).

Reviewer 2 Report

Comments and Suggestions for Authors

 

This manuscript argues for applying Tinbergen’s seminal “Four Questions” to zoo animal behaviour. I wholeheartedly endorse this view – the four questions remain relevant.

 

However, I think this manuscript could do better at explaining the specifics of how the four qs apply to zoo animals. This is typically couched in generalities. The link between the four qs and zoo animal behaviour currently (at least to me) amounts to essentially the “TFQs” being shorthand for “a scientific approach to studying animal behaviour” (e.g., in “Using TFQs to understand the individual and species-level importance of different behaviour can assist with planning and implementation of welfare-focused husbandry”, L450-453). This raises questions over the ms’s novelty. The manuscript could also be clearer on what considering the four questions adds, as pretty much all the issues it raises are already areas of concern in zoo research. In short, the structure and central argument need clarification.

 

Likewise, within sections, the central point could be clarified. A good start would be opening each section with a mini-intro and ending them with mini-summaries to emphasise the key points. Examples could also be clearer. Currently, the manuscript very rarely gives examples where each of the four questions are applied to an animal welfare issue (essentially only one figure). Usually, examples are raised and then a vague comment made about the four qs’ applicability stated but not explained. I’d like to see lots of clear examples of where applying the Four Qs would be beneficial, why, and what the outcome would be. Indeed, some of the Four qs are possibly not welfare-linked (e.g., phylogeny). Right from the outset, I’d like a clear rationale for why each of the four questions is relevant to welfare, and why they form a useful framework for welfare science.

 

 

Specific Points:

*L28: “continue TO advance”

*L36: I’m not really sure what this means (see also L244).

*L48: As written, this definition of causation could apply to ontogeny.

*L58-60: This feels like the wrong place for this sentence, which does not consider all four of the questions.

*L62: I’m not sure this acronym’s necessary but, if it is, it should be defined on first use.

*L97: In my view, animal welfare science is too interdisciplinary to be classified as a “behavioural science”.

*L137-139: I’m unclear, based on these explanations, why this must be developmental plasticity.

*L230-233: I don’t understand the point that’s being made with this example. It doesn’t seem to link to the provision of enrichment, and captive meerkats in outdoor enclosures are at risk from aerial predators.

*L245-246: Is it subjective? Always and totally? This feels like an over-generalisation.

*L248: For technical terms like QBA, I’d always define on first usage.

*L253-256: This “example” doesn’t really apply the four questions. Specifically, how does addressing each of the four questions (or selecting one or more of the four) improve welfare assessment?

*L257: Why?

*L336: Grammar

*L365: At least some of the causative links in this table (Phylogeny > Challenges in the Zoo) are somewhat speculative. Perhaps it could be framed in terms of phylogenetic hypotheses generated from challenges that emerge in zoos?

*L436-438: This point could be unpacked. Specifically, how has this knowledge affected, say, enclosure design?

*L446-448: As above, vague.

*L469: Presumably not if the natural behaviour involves hiding.

*L475: Often the “Relevant outcomes” in this table feel unrelated to several conside points in the “Considering Tinbergen” column, leading to the possible appearance of superficiality. Can each “Applied concept” have four rows linking to the Four qs, so the table explains the relevant outcome for each of the qs?

*L584: This feels very late to be introducing new references and ideas.

Comments on the Quality of English Language

Fine

Author Response

This manuscript argues for applying Tinbergen’s seminal “Four Questions” to zoo animal behaviour. I wholeheartedly endorse this view – the four questions remain relevant.

However, I think this manuscript could do better at explaining the specifics of how the four qs apply to zoo animals. This is typically couched in generalities. The link between the four qs and zoo animal behaviour currently (at least to me) amounts to essentially the “TFQs” being shorthand for “a scientific approach to studying animal behaviour” (e.g., in “Using TFQs to understand the individual and species-level importance of different behaviour can assist with planning and implementation of welfare-focused husbandry”, L450-453). This raises questions over the ms’s novelty. The manuscript could also be clearer on what considering the four questions adds, as pretty much all the issues it raises are already areas of concern in zoo research. In short, the structure and central argument need clarification.

 

Thank you. We have attempted to clarify and examine our examples more clearly. We have provided key links to the zoo, for example our penguin swimming / welfare example, and big cat enrichment example, that link specifically back to the FQs. We have added further clarity and key points that link back to the FQs throughout the paper.

 

Likewise, within sections, the central point could be clarified. A good start would be opening each section with a mini-intro and ending them with mini-summaries to emphasise the key points. Examples could also be clearer. Currently, the manuscript very rarely gives examples where each of the four questions are applied to an animal welfare issue (essentially only one figure). Usually, examples are raised and then a vague comment made about the four qs’ applicability stated but not explained. I’d like to see lots of clear examples of where applying the Four Qs would be beneficial, why, and what the outcome would be. Indeed, some of the Four qs are possibly not welfare-linked (e.g., phylogeny). Right from the outset, I’d like a clear rationale for why each of the four questions is relevant to welfare, and why they form a useful framework for welfare science.

 

Thank you for the comment and feedback, we have edited and clarified the examples and context accordingly throughout the manuscript. However, other reviewers have asked for sections to be reduced so this has been tricky to manage.

 

 

Specific Points:

*L28: “continue TO advance”

We have edited this typo. Thank you.

 

*L36: I’m not really sure what this means (see also L244).

We have removed line 36 to be clearer in our explanation, and we have also edited line 244 to make it clear we are referring to behaviour and welfare.

 

 

*L48: As written, this definition of causation could apply to ontogeny.

We have edited the definition of causation to be clearer. Thank you for the comment here.

 

*L58-60: This feels like the wrong place for this sentence, which does not consider all four of the questions.

Thank you for the comment. We have moved this sentence within the paragraph and expanded on our meaning here to include application to captive care.

 

*L62: I’m not sure this acronym’s necessary but, if it is, it should be defined on first use.

Thank you for the comment. We define this abbreviation on line 42 and we have used this to save words and make sentences flow more easily. We would like to keep it for those reasons.

 

 

*L97: In my view, animal welfare science is too interdisciplinary to be classified as a “behavioural science”.

We have edited this to include the term interdisciplinary and stated that AWS has a strong grounding in the behavioural sciences.

 

 

*L137-139: I’m unclear, based on these explanations, why this must be developmental plasticity.

Thank you for the comment, we have changed this example to one showing that the flying abilities of starlings are impacted upon by competition in the nest during their time as a nestling.

 

 

*L230-233: I don’t understand the point that’s being made with this example. It doesn’t seem to link to the provision of enrichment, and captive meerkats in outdoor enclosures are at risk from aerial predators.

Thank you for the comment. We have moved this example to further up the paragraph and included further explanation of the meaning of this example.

 

 

*L245-246: Is it subjective? Always and totally? This feels like an over-generalisation.

Thank you for the comment. We have edited this sentence accordingly and included a different reference.

 

*L248: For technical terms like QBA, I’d always define on first usage.

Thank you for the comment. We do define QBA, “for inferring emotional outputs from animal body language” but we have made this clearer.

 

 

*L253-256: This “example” doesn’t really apply the four questions. Specifically, how does addressing each of the four questions (or selecting one or more of the four) improve welfare assessment?

Thank you for the comment. We have expanded and provided further explanation of this example.

 

*L257: Why?

We have edited to explain that we need more evidence for behavioural normality for many captive species.

 

 

*L336: Grammar

Edited

 

*L365: At least some of the causative links in this table (Phylogeny > Challenges in the Zoo) are somewhat speculative. Perhaps it could be framed in terms of phylogenetic hypotheses generated from challenges that emerge in zoos?

Thank you for the suggestion. We have edited this section and the table accordingly.

 

 

*L436-438: This point could be unpacked. Specifically, how has this knowledge affected, say, enclosure design?

Thank you for the comment. We have provided ideas for how such information can be used to inform enclosure design.

 

*L446-448: As above, vague.

We have edited and expanded to provide more applied concepts.

 

*L469: Presumably not if the natural behaviour involves hiding.

Thank you for the useful point! We have edited this to include further context and explanation.

 

 

*L475: Often the “Relevant outcomes” in this table feel unrelated to several conside points in the “Considering Tinbergen” column, leading to the possible appearance of superficiality. Can each “Applied concept” have four rows linking to the Four qs, so the table explains the relevant outcome for each of the qs?

We have edited and expanded the outcomes column. Thank you for the ideas and suggested improvements.

 

*L584: This feels very late to be introducing new references and ideas.

Thank you for the comment. We have moved this information to the start of 5.3 and better integrated it into the text.

Reviewer 3 Report

Comments and Suggestions for Authors

This is a well written and  thoroughly researched review of the application of the principles of the TFQ to the understanding of the motivation for food seeking, reproduction and social behaviour in captive zoo animals. It is however very long. I suggest that it would have much more impact if it were half the length. The essence of the tables could be conveyed in about 20% of the words.  All the current examples quoted from the literature can be retained  but identified in brief as topics for further reading (if desired).

Most of the review deals with food seeking, which is something that can be addressed in zoos. Other motivations, like reproduction, and interspecies interactions are difficult/impossible to meet. Some of your assumptions appear inconsistent with the TFQ; e.g. presumably penguins are primarily motivated to swim by the proximate need to seek food. Just swimming in a zoo pond might get boring in short time.

Author Response

This is a well written and  thoroughly researched review of the application of the principles of the TFQ to the understanding of the motivation for food seeking, reproduction and social behaviour in captive zoo animals. It is however very long. I suggest that it would have much more impact if it were half the length. The essence of the tables could be conveyed in about 20% of the words.  All the current examples quoted from the literature can be retained  but identified in brief as topics for further reading (if desired).

Thank you for the comment. We are not sure what to edit here as other reviewers have asked for more information. We have attempted to rationalise the manuscript where possible. We appreciate the feedback.

Most of the review deals with food seeking, which is something that can be addressed in zoos. Other motivations, like reproduction, and interspecies interactions are difficult/impossible to meet. Some of your assumptions appear inconsistent with the TFQ; e.g. presumably penguins are primarily motivated to swim by the proximate need to seek food. Just swimming in a zoo pond might get boring in short time.

Thank you for the comment. We have applied the Four Questions to a range of outputs from the zoo in Table 2 so we have covered more than just feeding behaviour.

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

Thanks for addressing my feedback - I think this will be a valuable contribution to the literature.

Author Response

Thank you for the kind remarks and helpful feedback on our paper.