All reviews of published articles are made public. This includes manuscript files, peer review comments, author rebuttals and revised materials. Note: This was optional for articles submitted before 13 February 2023.
Peer reviewers are encouraged (but not required) to provide their names to the authors when submitting their peer review. If they agree to provide their name, then their personal profile page will reflect a public acknowledgment that they performed a review (even if the article is rejected). If the article is accepted, then reviewers who provided their name will be associated with the article itself.
Thank you for addressing all comments so diligently and for making the respective changes in the manuscript. I agree that it was the better decision to take the more cautious approach and focus on what this study can specifically contribute. My only remaining observation is that the figure legends are quite short making it difficult to fully grasp what is being presented without referring back to the text. If possible, you might want to consider adding brief explanations what you would like readers to pay attention to, but that is not necessary.
[# PeerJ Staff Note - this decision was reviewed and approved by Michael Wink, a PeerJ Section Editor covering this Section #]
The authors revised their MS based on the reviewers' comments.
They answered all my concerns and further justified their approach when necessary.
They substantially modified the MS to address each reviewers' suggestions or comments.
Regarding a stronger comparison between their findings and vocal communication and tool use, whilst the authors integrated some of my suggestions, they mainly adopt the position of Reviewer 2 asking for more caution, and extensively modify the manuscript to answer Reviewer's 2 concerns. I understand this, and whilst I personally think it could be beneficial to be more explicit about a potential link between the two, I appreciate the need to be careful when making such comparison and I support this approach.
Once again, I find this paper a very valuable contribution to the field and recommend its publication in Peer J.
no comment
no comment
no comment
The authors addressed all my comments. They did a really good job and I think that the manuscript has improved a lot.
No comments
No comments
Your article entitled “Chimpanzee play sequences are structured hierarchically as games” has now been been seen by two reviewers and the reviewers’ comments are appended below. I share the reviewers' view that this is manuscript addresses an important topic and that the attempt to link play behavior to communication offers a potential way to gain new insights into both. However, I also agree with the concerns they are raising, which means that I do not think that this article is at this stage ready for publication. Both reviewers have questions about the link you aim to establish between play and communication. In particular, I share both their concerns about your decisions about when play becomes a game with rules, and both reviewers have helpful suggestions to consider around rule-breaking (reviewer 1) and about the sufficiency of the sequentiality criterion (reviewer 2). From my reading, I was also unsure to what extent external constraints, rather than the cognitive processes you elude to, structure play: for example, I would assume that play in a tree is already limited by the postures individuals can take in a tree. All these comments are about providing clearer definitions and explanations. I am looking forward to a revised manuscript.
[# PeerJ Staff Note: Please ensure that all review and editorial comments are addressed in a response letter and any edits or clarifications mentioned in the letter are also inserted into the revised manuscript where appropriate. #]
This study investigates the sequential and, potentially, hierarchical structure of play behaviour in chimpanzees. The authors demonstrate that play in chimpanzees revolves around six games. They show that whilst within a game, play retain its unpredictable nature (transition probabilities remain low, even though above chance level), added information about antecedent play elements increase predictability, confirming play is hierarchically structured in games. The paper is well-written and well-structured despite a very complex methodology. The figures presented are relevant and the data is provided. This study is an important contribution to the field and has the potential to provide further insights into the cognitive abilities of chimpanzees, with potential parallels in combinatorial communication and tool use. This could, in turn, shed light into the evolutionary progression of these key features in our species.
This brings me to the major issue I have with this paper: I think it is not ambitious enough, focused mostly on play behaviour. Whilst it is always interesting to study a species behaviour for its own sake, I think the striking element here is the comparison between sequential/hierarchical structuring in play and what is observed in language/tool use. The authors shy away from making the comparison and considering a continuity between the different kind of structuring (see L71-74; L490-495; L559-560). I would advise the authors to start the introduction with their second paragraph (L46-74) and use their first paragraph (L31-45) as a transition between communication/tool use and play. I recommend they compare play and communication/tool-use sequences more explicitly, with more references to the work that has been done in the latter. There is extensive literature linking cognitive abilities underlying tool use for instance and the emergence of language, I believe this paper could very well tap into that and discuss the potential for a continuity between play, tool-use and language. I would advise the author to emphasise in the discussion how play could help in the acquisition of either tool use or combinatorial communication. We know play is extremely relevant for ontogeny in any species. One could argue that learning the sequential/hierarchical rules of play could help the development of cognitive capacities allowing the processing of sequences and hierarchies in other domains, i.e. tool use or communication. Although I understand one should be cautious when making such comparison/connection, I believe the authors can help the reader navigate these potentially very interesting implications.
Although the methods are complex, they are extremely transparent and well-detailed, allowing replicability. The authors acknowledged the choices they made and the limitations associated and discuss this extensively in the discussion. The study represents the first exploration into the sequential nature of play in chimpanzees and the results presented here as well as the methodology are clearly suited for the exercise. For this reason, the next comment is not necessarily targeted at this paper specifically. As the authors mentioned, play behaviour was investigated on an individual level, not taking into account the response from the playing partner(s), and they highlighted six games with underlying rules (L481-482; here typo as well: wHere). A critical next step is to investigate how/when rules are violated and what happen then. Is a rule violated when the player uses play elements form other games? exhibit different kind of behaviour (e.g. aggression)? Does the play bout end then? I believe investigating not only what are the rules but also what happen when they are violated can be very informative cognitively and be transposed potentially to other sequential/hierarchical behaviours. I understand this may be out of the scope of the current study, but it might be worth discussing.
The findings are robust, statistically sound and controlled. Again, this is a very first investigation in the hierarchical structure of play in chimpanzees, and whilst the authors make some choices that can potentially bias the results, the rationale behind it is well argued and its limitations extensively discussed. This is a very promising contribution to the field, and as mentioned above, I believe the authors need to make this much more explicit.
I think that the literature references are not sufficient to understand what play behaviour and, more important, games are.
Moreover, I have some concerns about the interpretation of the results made by the authors.
Line 159-161: Why authors choose this 5 seconds criteria to separate the play bouts and not the 10 seconds criteria already used in the literature? I am not saying that this is wrong, but I would like to understand the reason at the base of authors’ choices.
Line 183-185: Authors considered one player at time in order to obtain a string of play elements. Being play the results of a continuous back and forth, could it be misleading consider only one player at time to obtain such strings? Maybe it is necessary to look at the continuous exchange of patterns and mood between players? Or at the asymmetry on the bout?
Line 194-195: I think that an inter- and also intra-observer reliability is needed, precisely because of this detailed coding approach. Without such reliability scores, authors cannot be sure of their data and analyses. This is not a little limitation, this is (I think) the kind of limitation that prevents a manuscript from being published.
Line 239: Did authors consider the software “Behatrix” (Friard and Gamba, 2021) to perform sequential analyses for the play bouts? Maybe it would be useful for them.
In the introduction, authors define “game” as follows: “In human play, there are specific, socially learned arbitrary rule systems that govern what we call ’games’ (Leisterer-Peoples et al., 2021): in a game, certain actions and sequences are allowed or not, but their order can be flexible. For example, in hide-and-seek, hiding is allowed, but laughing loudly is counterproductive, where and how to hide is up to the player. There is evidence that apes have standardised games and play them with each other and human partners (Costa et al., 2019; Pika & Zuberbühler, 2008; Tanner & Byrne, 2010) – however, these examples focus on special contexts (e.g., playing in water, playing socially with objects), and we do not have a method to determine how widespread predictable behavioural rules are.”
In the discussion, at lines 463-473, authors discuss their result interpreting the connection between several patterns as game. I am not convinced about it. What I am wondering is: is the link (and consequentiality) of some play patterns sufficient to claim that this chimpanzees have their games? I don’t think authors addressed this questions in their analyses, and this is a problem because they included the word “games” also in the title of the manuscript.
I would have expected a more complete description of what a game is and how to recognize it in the introduction, with more objective criteria than the only one gave by authors. This is why the sophisticated analyses performed by the authors does not address the key point of what a game is and how to recognize it, and this is reflected into the discussion.
Line 476-477 and line 554: Authors’ data cannot be generalized (“we show that, at least for chimpanzees…”, “we show that chimpanzee play behaviour is…”) due to the their limitation (limited sample size, few young subjects).
Line 481-483: Are authors sure that their analyses and criteria are sufficient to indicate a “rule-based system”? They checked for consequentiality and link between patterns. Moreover, they considered only one player at time, not taking into account the response of the playmate.
Line 497: “Rough-and-tumble” is often used to code a play-fight behaviour that is impossible to codified more in detailed due to the high distance between the researcher (and so the video camera) and the animals or to the intertwining of the players' bodies. All situations in which coding could not be objective.
Line 506-509: Evidence for role-reversal and self-handicapping strategies were already found in apes, including chimpanzees.
I found the introduction too vague, especially when authors explained the form of sequences in communication and tool use. Play behaviour is a completely different sphere. Communication, and tool use in particular, have clear immediate goals, while play does not have them. Another criteria (following Burghardt, 2005) is that play involve incomplete and exaggerated pattern borrowed from other “serious context” such as the aggressive one, and that those patterns are carried out in an disordered way. For this reason, I would like to read more about play and less about the organization of other spheres that have little to do with the focus of the manuscript.
Moreover, the communication during play is completely neglected. Authors refer to “communication” as it occur only outside the play behavioural sphere, but it is not true.
Line 38-41: "In many animal species, social play involves rapid exchanges of actions between several participants that often appear random to observers, making it one of the most complex social contexts individuals are involved in daily." This sentence is not so correct because the occurence of play behaviour can be rare (not daily).
Line 44-45: This is not true. There are few methods that try to give an overview of the complexity and level of competition of a play session: the Play Asymmetry Index used for primates and dogs, the Play Variability Index used for interspecific play between dog and horses... Maybe they are not what the authors are looking for, but saying that there are no methods to understand how flexible play is, is incorrect.
Line 100-101: There are several species besides chimpanzees that play also in adulthood (and in which play can serves different functions): several monkeys, dogs, wolves, dolphins, elephants, hyenas, lions, meerkats, sealions….
All text and materials provided via this peer-review history page are made available under a Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.