This commentary complements the article by De Dreu and Gross (2019) from the perspectives of behavioral economics and game theory. It aims to provide a bridge between psychology/neuroscience research and economics research in attack-and-defense by stipulating relevant literature, clarifying theoretical structures, and suggesting improvements in experimental designs and possible further investigations.

A collective action problem exists not only in offensive warfare, but also in defensive situations. The collective action problem is dealt with in the same way in offensive and defensive warfare: by strong leadership, discipline, rewards and punishments, strong group identification, strict religiosity, and intolerance of deviants. This behavior is explained in terms of evolutionary psychology.

Is there a strategic mechanism that explains role-contingent differences in conflict behavior? I sketch a theory in which differences in optimal behavior for attackers and defenders arise under initially symmetric conditions through the dynamic accumulation of differences in the distributions of traits in the subpopulations of potential opponents.

The target article focuses on how attacker-defender conflicts are fought. This commentary complements it by considering how attacker-defender conflicts may be resolved at the bargaining table. I highlight multiple linkages between asymmetric intergroup conflict as modeled with the attacker-defender game and negotiation research and illustrate how the proposed model of attacker-defender conflicts can inspire new research on intergroup negotiation.

De Dreu and Gross argue that only asymmetric games allow the motives underlying defense and attack to be disentangled. However, the Prisoner's Dilemma Game Alt matrix (PDG-Alt matrix), a modified symmetric PDG, also allows these motives to be disentangled. Studies using the PDG-Alt matrix produced findings contradicting a central claim of De Dreu and Gross.

Current psychological perspectives emphasize “attack” and “defense” as the behavioral mechanisms underlying conflict. Here, we extend this view by highlighting the relevance of pathological altruism and the neuroendocrine pathways associated with hostile behaviors. Specifically, we elucidate the modulatory role of the neuropeptide hormone oxytocin in motivating extraordinary levels of in-group commitment that can promote extreme behaviors and endure conflict with out-groups.

Here, we argue that attackers in intergroup conflicts are also likely to hold strong identity fusion, anticipate threat from the out-groups, and retaliate by signaling preemptive aggressiveness, which may not be asymmetrically exclusive to defenders. We propose that the study of the intergroup and intragroup dynamics could highlight more specific, robust markers to differentiate types of defenders from attackers.

De Dreu and Gross (D&G) seem to have disregarded some relevant experimental literature on games of conflict, most notably variations on “matching pennies” games. While in such games, “attacker” and “defender” are typically not explicitly labelled, players’ differentiated roles yield naturally to such notions. These studies partly validate some of D&G's findings and interpretations.

De Dreu and Gross offer a compelling synthesis of a growing literature on the psychology of attack and defense. I argue that human raiding ecology suggests the need to endogenize attacker-defender move order as well as opportunities for tactical mismatch available to defenders. Perhaps most significantly, I draw attention to the surprising lacunae in sex differences across attack and defense.

De Dreu and Gross's description of the proximate mechanisms conditioning success in intergroup conflict omits humans' deontological morality. Drawing on research on sacralization and moral objectivism, I show how “moral rigidity” may have evolved through partner selection mechanisms to foster coalitions’ cohesion and combativeness in intergroup conflict.

We argue that adopting a sociocultural lens to the origins of intergroup bias is important for understanding the nature of attacking and defending behavior at a group level. We specifically propose that the potential divergence in the development of in-group affiliation and out-group derogation supports De Dreu and Gross's framework but does indicate that more emphasis on early sociocultural input is required.

Gain or omission/termination of loss produces approach; while loss or omission/termination of gain produces withdrawal. Control of approach/withdrawal motivation is distinct from valuation of gain/loss and does not entail learning – making “reward” and “punishment” ambiguous. Approach-withdrawal goal conflict engages a neurally distinct Behavioural Inhibition System, which controls “anxiety” (conflict/passive avoidance) but not “fear” (withdrawal/active avoidance).

We focus on the implications of De Dreu and Gross's findings for the evolutionary perspective on out-group aggression and in-group cooperation. Although their experimental protocols are potentially useful in determining the origins of out-group aggression in humans, they so far provide inconclusive evidence only. We suggest ways of furthering our understanding of the connection between parochial cooperation and intergroup conflict.

Many modern-day groups differ from prehistoric ones regarding the proportion of members who are related to any particular individual. From an evolutionary mismatch lens, an appreciation of this disparity could help better explain the potential dilution of group cohesion during peacetime and inform novel, more effective approaches to enhancing group unity – strategies that might enhance national security around the globe.

With respect to De Dreu and Gross's article, we comment on the psychological functions for attack and defense, focusing on associations between individual differences in psychopathic personality traits and the behavioral patterns observed in attack-defense conflicts. We highlight the dimensional nature of psychopathy and formulate hypothetical associations between distinct traits, their different behavioral outcomes, and associated brain mechanisms.

We argue that how players perceive the attack-defense game might matter far more than its actual underlying structure in determining the outcomes of intergroup conflict. Leaders can use various tactics to dynamically modify these perceptions, from collective victimization to the distortion of the perceived payoffs, with some followers being more receptive than others to such leadership tactics.

Attempts to understand the fundamental forces shaping conflict between attacking and defending groups can be hampered by a narrow focus on humans and reductionist, oversimplified modelling. Further progress depends on recognising the striking parallels in between-group conflict across the animal kingdom, harnessing the power of experimental tests in nonhuman species and modelling the eco-evolutionary feedbacks that drive attack and defence.

There is a large body of research on conflict in nonhuman animal groups that measures the costs and benefits of intergroup conflict, and we suggest that much of this evidence is missing from De Dreu and Gross's interesting article. It is a shame this work has been missed, because it provides evidence for interesting ideas put forward in the article.

We argue that the roles of attacker and defender in asymmetric intergroup conflict are structurally ambiguous and their perception is likely to be subjectively biased. Although this allows for endogenous selection into each role, we argue that claiming the role of the defender likely is more advantageous for conflict participants.

The attack-and-defense game is a game in which an attacker (a group of attackers) has an incentive to revise the status quo and a defender (a group of defenders) wants to protect it. The asymmetry in objectives creates incompatible interests and results in a mixed-strategy Nash equilibrium. However, this equilibrium could be heavily impacted by behavioral considerations.