A more economical explanation for the results – and one that must be ruled out before a mechanistic explanation based on rituals and coordination of actions is proposed – would be that oxytocin increases participants` memory for the sequence of events (there is evidence of effects of oxytocin on memory) or that it reduces their stress levels and increases rest and lethargy (there is evidence in humans and animals) and reduced their contributions. Even in the literature on oxytocin and economic games (which should be the conceptual framework of this study), the results must be aligned with those showing individuals giving more on oxytocin. Is this a cultural phenomenon specific to Chinese culture? Although these collaborations indicate that oxytocin is a possible neurobiological mechanism underlying group coordination, three questions remain unclear. First, we lack empirical evidence of the possibility that oxytocin promotes coordination of collective action at the group level in general and during competition and group conflict in particular. Second, we misunderstand how group members coordinate collective action during competition and conflict between groups, and whether oxytocin can directly affect the coordination and/or strategy that group members use to coordinate. Third, we do not know whether oxytocin modulates tacit coordination within groups differently when collective contributions are aimed at attacking the rival rather than defending the group`s interior against possible attacks by the rival. We apologize for not being clear and for revising Figure 2 accordingly. In addition, we added a table to give a clear description of the rule for the IADC game, as well as the calculation of the payment for each person (i.e. Table 1 in the revision). In the concrete examples of Figure 2: In the simultaneous block, we showed a turn for a group of attackers who lose this round (a total of 6 compared to the pool of 7 of a group of defenders), so that each attacker ends up with what remains of his contribution: attacker 1 received 17 (as 20-3), Forward 2 won 19 (like 20-1), and forward 3 won 18 (like 20-2). In the sequential block, we represented a round for a group of defenders to win this round (a total of 10 compared to the group of 7 men in an attack group), leaving each defender with what remains of their contribution: Defender1 won 16 (like 20-4), Defender 2 won 16 (like 20-4) and Defender 3 won 18 (like 20-2). Treatment effects were also observed when we looked at the number of non-contributors.
There were more non-contributors in the attacking groups than in the defense groups (M ± SE = 20.23 ± 0.90 versus 4.25 ± 0.44; F(1, 78) = 408.489, p < 0.001; η2 = 0.840) and more non-contributors in groups receiving oxytocin as placebo (M ± SE = 13.46 ± 0.90 vs 11.01 ± 0.75; F(1.78) = 4.345, p = 0.040; η2 = 0.053). Crucially, oxytocin increased the number of non-contributors in the attacker groups, but not in the defense groups (role × treatment, F (1, 78) = 5.043, p = 0.028, η2 = 0.061, Figure 3B). This role × treatment effect applies in particular when decisions were taken simultaneously (F(1, 78) = 5.712, p = 0.019, η2 = 0.068), but less so when decisions were taken sequentially (F(1, 78) = 2.143, p = 0.147; η2 = 0.027). We then looked at the participants` decision time to decide not to contribute and showed that individuals belonging to groups of attackers made the decision not to contribute faster than to contribute (F (1, 78) = 137.679, p < 0.001, η2 = 0.641). Oxytocin increased the rate at which individuals in the attacking groups decided not to contribute (treatment × contribution: F(1, 77) = 4.857, p = 0.031; η2 = 0.059, Figure 4). As we discuss in the manuscript, if oxytocin allows groups to coordinate on a "peaceful strategy without attack", we would expect 1) smaller victories in groups of attackers receiving oxytocin as a placebo, 2) the effect of oxytocin on the number of non-contributing attackers should not differ if the attacks succeed or fail, and 3) no effect of oxytocin on tracking the rival`s defense history. The results of our subsequent analyses of the success rate (oxytocin did not reduce the success rate), the non-contributor to successful or failed attacks (oxytocin increased the non-contributing attacker only in failed attacks but not in successful attacks) and the tracking parameter (oxytocin increases the continuation of the rival`s history by the attacker, fewer attacks when rivals are strongly defended) support the second interpretation, that oxytocin increases the effective attack strategy and not the peaceful non-attack strategy. Consistent with previous studies, oxytocin may increase the non-contributing attacker through its function of facilitating group service bias (De Dreu and Kret, 2016; De Dreu, et al., 2011), Promoting social learning and cognitive flexibility (Ma et al., 2016a; Sala, et al., 2011) and the growing importance of social feedback (Shamay-Tsoory, et al., 2016).
We ask that you disagree with the reviewer that conceptions between subjects deviate from the gold standard in oxytocin studies, and believe that bias in randomization cannot explain our results. First, we have now described the randomization procedure in more detail (subsection “Procedures”) and pointed out that participants were measured not only by demographic data, but also by current mood and psychological variables. We showed that individuals in the four groups (Attacker/Defender x Oxytocin/Placebo) did not differ in terms of demographic information, mood swings and prosocial characteristics (Supplementary Dossier 1. Supplementary tables 1A, 1B). Second, several meta-analysis studies have summarized behaviour (Van IJzendoorn, Bakermans-Kranenburg, 2012; Shahrestani, et al., 2013) and the neural effects (Wang, et al., 2017) of oxytocin. For example, in the meta-analysis of the behavioural effects of oxytocin (van IJzendoorn and Bakermans-Kranenburg, 2012, Table 1), of the 31 studies that examined the effects of intranasal oxytocin on facial recognition, trust in the group, and trust in external group members, 10 studies were adopted in the subject design, while the majority (i.e. 21 studies) between subjects used plans for the administration of oxytocin and placebo. Shahrestani and colleague summarized 7 studies that examined the effects of intranasal oxytocin on emotions and facial recognition, all of which used a plan between subjects (Shahrestani, et al., 2013). Similarly, we summarized the neural effects of intranasal oxytocin in our recent meta-analysis of fMRI studies of oxytocin published prior to March 2017 (Wang et al., 2017). There were 50 fMRI studies administering oxytocin intranasally in healthy individuals, 29 of which used a plan between subjects (21 studies in the subject plan, Table S1 in Wang et al., 2017). We believe that designs are often used between subjects and provide reliable and reproducible results for oxytocin/placebo contrast, as we have also seen here.
But Zhang et al. now show that the “love hormone” also helps individuals launch more coordinated “attacks” against outside groups. In a study that included a multi-stage economic competition game between groups of “attackers” and “defenders,” oxytocin did not make attackers less aggressive. Instead, it allowed them to better coordinate their attacks. Each competitive game included three attackers who individually deposited money into a group pool to outbid the other group and earn more money, and three defenders who made similar contributions to their own group pool to defend it against rival attacks and protect themselves from losing all their money. Attackers who used an oxytocin nasal spray were better at pursuing their rivals` defensive strategies than attackers whose nasal spray contained a placebo. Under the influence of oxytocin, the attackers planned their attacks to occur when their rivals were vulnerable. Over time, oxytocin users have become more effective at coordinating their behavior with other members of their group. This has led to more returns. Inter-subject correlation – the author`s index for behavioral synchronicity – was only measured in the “attacking” group (unclear reason), but this group also significantly reduced its contribution over time on oxytocin. It is possible that the increase in correlation between group members results from this decrease and that zero donation is correlated between participants (although we accept the suggestion that it is a proxy for synchronicity, it is a synchronicity of “no action” and not action).
Another important issue is design between subjects, which is not the gold standard in research on oxytocin administration. Particularly with respect to economic games, a recent study showed different patterns of brain activation in a design within and between subjects of an OT administration study that used the prisoner`s dilemma (Chen X, Gautam P, Haroon E, Rilling JK Within vs. Inter-Subject effects of intranasal oxytocin on the neural response to cooperative and non-cooperative social interactions. Psychoneuroendocrinology. 2017 78:22-30. doi: 10.1016/j.psyneuen. 2017.01.006). At a minimum, the authors should describe how randomisation was carried out and show in-depth comparisons between groups not only in demographic terms, but also in relevant physiological variables (e.B. BMI). Anagnostou E, L Soorya, J Brian, A Dupuis, D Mankad, S Smile, S Jacob “Intranasal Oxytocin in the Treatment of Autism Spectrum Disorders: A Review of the Literature and Early Data on Safety and Efficacy in Youth.” Brain Res (2014); 1850:188-198. .
. .