This paper explores the impact of multiple social norm types (specifically, injunctive versus descriptive norms) on the behaviour of a coupled social-ecological model of a population of harvesters. The authors find a wide variety of behaviours are possible, including states of intermediate sustainability, and the results point to the importance of factors such as the threshold for conforming to a norm, and timescale.  I found the results on how system response to policy interventions depends on group size and composition to be particularly interesting.  The paper expands the social structure considerably, relative to existing literature.  The model is adequately described, and appropriate literature is referenced.  The analysis is sufficiently thorough to illustrate the features of the model. Both injunctive and descriptive norms are modelled as a contagion process (occurring at different organizational levels) which would not have been my first choice since injunctive norms are “ought”, whereas descriptive norms are “is”, and their flavour is therefore very different, but maybe that’s just a modelling choice.  I only have some minor revisions to suggest that I hope might improve the paper:

1. The paper touches on social learning, but I don’t see mention of social learning in section 3.3, where ‘individual learning’ is described. This should be corrected if it was just an oversight, or the difference between individual learning and social learning in their framework should be clarified.
2. I was confused by the results in section 3.4. The intervention strength is the fraction of policy-influenced groups and the x-axis of figure 3.4a has a wide range with many different values for the intervention strength.  But if there are only two groups, how can the intervention strength be anything other than 0, 0.5 or 1?
3. Line 569: Is the lack of convergence for high values of Delta T interesting in some way, such as leading to oscillations or other transients, or does lack of convergence mean some unsociological behaviour?
4. There is a good amount of repetition that could be removed to make the paper shorter.
5. The Conclusion or Discussion section should include a section on model limitations. For instance, model assumptions that might impact results could be highlighted and the relative lack of sensitivity analysis could be mentioned. Similarly, the model assumes that each harvester has their own personal stash, which might be a good approximation for farmers, but even in that case, the decisions of neighbouring farmers influence one another.

This study describes a novel approach to modelling social norms in the context of resource extraction in World-Earth modelling. The authors implement a new model in the existing copan:CORE framework that consists of different layers: a biophysical layer, a layer representing individual agents and their interactions, and a novel layer that represents group processes. By adding the additional group level layer, the authors can model the effects of social norms more realistically by representing the effects of both descriptive and injunctive norms simultaneously. This advances the current representation of social norms in agent based models and World-Earth modelling.

I think this paper is well thought out and well-written, and represents a worthwhile addition to the existing literature and methodology for modelling social norms. Yet, I wonder whether the authors can make more explicit in this paper how the current approach compares to and relates to previous efforts to represent social norms in modelling efforts. I also think the authors can be more explicit about their assumptions regarding the dynamics of social norms and which processes they decided to model and how. I expand on my points below.

First, I think that the representation of both descriptive and injunctive norms using a group level layer represents a clear step forward for representing social norm effects in agent based models. As I was reading the paper, I however found it difficult to establish how this approach builds upon and differs from other approaches to modelling social norms. I believe a game theory/commons dilemma approach is the main theoretical lens through which decision-making surrounding resource extraction is analyzed and modelled, but the authors do not discuss how their approach relates to this.

Importantly, the commons dilemma framework suggests at least two additional processes related to social norm effects, which are already commonly included in agent based models but which are not represented in this paper. First, if there is a norm to cooperate (i.e., act sustainably), this increases the payoff associated with defecting (i.e., acting unsustainably), which could be a motivating force to shift behaviour to defect in order to maximize personal gains (i.e., freeriding) (e.g., (Tavoni et al., 2012; Tu et al., 2024).  Second, the power of social norms lies in their enforcement, i.e., the fact that there are social punishments associated with not acting in line with the norm (e.g., (Nhim et al., 2019; Tavoni et al., 2012)). I was wondering whether the authors could expand upon why they did not choose to incorporate such dynamics into the current representation. I realize that making models overly complex is not desirable, but I also think it is important that modelling efforts are cumulative and thus clearly connect to and build upon existing works rather than developing separate approaches that are not integrated.

Second, I think distinguishing between injunctive and descriptive norms is a great addition to the literature, but I think the current theoretical review and modelling application is still a bit agnostic and simplistic about the different mechanics through which injunctive and descriptive norms influence behaviour. People can conform to social norms because of informational (i.e., assuming that the behaviour most people do will likely be the correct/most effective approach) or normative reasons (i.e., wanting to fit in/not stand out) (e.g., (McDonald & Crandall, 2015)). The latter of these also relies on the sanctioning of norm violations by other group members, as I also identified above as a missing mechanism. Studies also indicate that there are key interactions that occur when descriptive and injunctive norms do or do not align. Specifically, if an injunctive norm is contrasted with a conflicting descriptive norm, its effect on behaviour is minimized (e.g., (Bonan et al., 2020; Smith et al., 2012; Staunton et al., 2014)).

I am not asking the authors to implement all these mechanics in the current application of the model, but I do think it this paper could present the full (or at least a fuller) picture of our understanding of the effects of social norms, and then more clearly show which elements are and are not represented in the current modelling representation and why.

Lastly, if I understand correctly, the effect of group membership only matters for injunctive norms and not descriptive norms. I think the model would be more realistic if group membership also affected how people react to the observed behaviour of others. Specifically, there is literature which shows that the effects of social norms differ based on whether this information is received from ingroup or outgroup members ((Spears, 2021). For example, people are more likely to follow the behaviour of other ingroup members compared to outgroup members. Similarly, deviation from a norm is perceived far more negatively for ingroup compared to outgroup members.

Bonan, J., Cattaneo, C., d’Adda, G., & Tavoni, M. (2020). The interaction of descriptive and injunctive social norms in promoting energy conservation. Nature Energy, 5(11), 900-909. https://doi.org/10.1038/s41560-020-00719-z

McDonald, R. I., & Crandall, C. S. (2015). Social norms and social influence. Current Opinion in Behavioral Sciences, 3, 147-151. https://doi.org/10.1016/j.cobeha.2015.04.006

Nhim, T., Richter, A., & Zhu, X. (2019). The resilience of social norms of cooperation under resource scarcity and inequality — An agent-based model on sharing water over two harvesting seasons. Ecological Complexity, 40. https://doi.org/10.1016/j.ecocom.2018.06.001

Smith, J. R., Louis, W. R., Terry, D. J., Greenaway, K. H., Clarke, M. R., & Cheng, X. (2012). Congruent or conflicted? The impact of injunctive and descriptive norms on environmental intentions. Journal of Environmental Psychology, 32(4), 353-361. https://doi.org/10.1016/j.jenvp.2012.06.001

Spears, R. (2021). Social Influence and Group Identity. Annu Rev Psychol, 72, 367-390. https://doi.org/10.1146/annurev-psych-070620-111818

Staunton, M., Louis, W. R., Smith, J. R., Terry, D. J., & McDonald, R. I. (2014). How negative descriptive norms for healthy eating undermine the effects of positive injunctive norms. Journal of Applied Social Psychology, 44(4), 319-330. https://doi.org/10.1111/jasp.12223

Tavoni, A., Schluter, M., & Levin, S. (2012). The survival of the conformist: social pressure and renewable resource management. J Theor Biol, 299, 152-161. https://doi.org/10.1016/j.jtbi.2011.07.003

Tu, C., Wu, Y., Chen, R., Fan, Y., & Yang, Y. (2024). Balancing Resource and Strategy: Coevolution for Sustainable Common-Pool Resource Management. Earth Systems and Environment. https://doi.org/10.1007/s41748-024-00489-8