Since ancient times, cooperation has played an important role in the struggle for human survival. The solutions for common problems such as climate change, resource reduction and environmental pollution require cooperation behaviors (Dietz et al., 2003; Fehr and Fischbacher, 2003; Milinski et al., 2006, 2008; Hauser et al., 2014). Cooperation, such as public education and public transportation, provides individual needs in a way that can maximize resource utilization. Cooperation exists at many levels and takes place between individuals and organizations as well as between states and countries. Cooperation allows participants to exchange valuable information that helps both sides improve their knowledge bases and work in a time- and resource-efficient manner (White, 2005).

In order to maintain and facilitate cooperation, abundant research has been conducted to explore influential factors (Cronk, 1999; White, 2005; Tabellini, 2008; Balliet, 2010; Espín et al., 2012; Gross et al., 2016; Molenmaker et al., 2016; Yilmaz and Bahçekapili, 2016) and underpinning mechanisms (Gardner and West, 2004; Fowler, 2005; Boyd et al., 2010; Wang et al., 2017; Fu et al., 2019) of cooperation. Specifically, there is by now considerable and convincing evidence that group members’ behavior within a social dilemma is influenced by both expectations and observations of others’ behavior (Braver and Barnett, 1974; Dawes et al., 1977; Messick et al., 1983; Schroeder et al., 1983; Yamagishi and Sato, 1986; Komorita et al., 1992; Bornstein and Ben-Yossef, 1994). There has also been much research conducted on the voluntary contributions mechanism (Brandts and Schram, 2001; Hauert et al., 2002; Juberías et al., 2004; Croson et al., 2005; Chaudhuri, 2011; García and Traulsen, 2012; Inglis et al., 2016; Iwamura et al., 2020), in which participants can choose between a private account and a public account to invest in; participants can obtain more profits from the public account if the collective investment of all the group members to the public account is high enough. Jennifer Zelmer (2003) performed a meta-analysis and concluded that average contributions, friendship among subjects, group size and cash reward are the main factors that have a significant effect on people’s tendencies to contribute to the public account. In the voluntary contributions game, subjects can benefit from both the private account and the public account.

However, few researchers have noted that society provides individual solutions as a complementary choice to shared problems. For example, a person can drive a private car to replace public transportation, or a house owner can employ private security instead of depending on a publicly funded police force. Theoretically, the availability of individual solutions leads to stronger tension than does the voluntary contributions game because people can only choose one of the two solutions (i.e., the individual solution or the collective solution) (Juberías et al., 2004; Gross and De Dreu, 2019). People try to obtain more independence through individual solutions, even if doing so consumes more resources (Fehr et al., 2002; Juberías et al., 2004). However, from existing theories, we know little about how individual solutions to shared problems influence the human inclination for cooperation and coordination.

Gross and De Dreu (2019) were the first to explore how the cost of an individual solution influences decision-making when the collective solution and the individual solution exist simultaneously; in their study, the researchers set different levels of cost ratios of an individual solution versus a collective solution and observed the subjects’ decision-making. The results showed that the participants’ cooperation tendencies increased with the increase in the cost ratio of the individual solution versus the collective solution. When the cost of the individual solution was over 1.7 times that of the collective solution, then 80% of participants would steadily choose the collective solution. Across different cost-benefit ratios of individually versus collectively solving the shared problem, individuals display a remarkable tendency toward group-independent, individual solutions. This “individualism” leads to inefficient resource allocations and coordination failure. In the presence of individual solutions to shared problems, groups struggle to balance self-reliance and collective efficiency, leading to a “modern tragedy of the commons.” In order to facilitate cooperation in this case, more research need to be done to know the underpinning mechanisms. We hold that partners’ cooperation tendency (i.e., participants’ judgment of the credibility of their partners in a cooperation activity) also plays an equal or more important role than that of cost ratio on participants’ cooperation tendencies in the presence of the individual solution.

Our research frame is divided into two parts. In the first part, we aimed to prove that team members’ credit is a significant factor which influences people’s cooperation tendencies when the individual solution is accessible. Then in the second part, we intended to investigate the exact nature of the relationship between others’ and our own behavior in a detailed way. A number of scholars have proposed uncooperative group members have greater impact on our behavior than equally-extreme cooperative ones. It has been reported (Messick et al., 1983; Samuelson et al., 1984) that providing a relatively wide distribution of false harvesting feedback in a resource-conservation dilemma leads to faster depletion of the shared resource than feedback with a narrow distribution and the same mean, just as one would expect if the extremely low cooperator had greater relative impact on others’ behavior. Furthermore, work by Kerr et al. (2009) has shown that the influence of a few non-cooperative group members is larger than the influence of a few cooperative group members (termed the “bad apple” effect). However, there has been no research exploring the influence of distribution of the self-interest or altruism tendencies on people’s decision making. Our current focus is on allocating the same self-interest or altruism tendencies to 2–5 group members, to explore how the distribution of self-interest or altruism tendencies influences people’s decisions.

To explore the question above, we designed the following four experiments to explore the effect of the quantity and distribution of teammates’ tendencies of self-interest and altruism on individual decision-making when an individual solution is available. First, the definition of relative items of this paper is given herein. The tendency of self-interest means that the group member contributes less than his or her fair share to the collective solution, while the tendency of altruism means that the group member contributes more than his or her fair share. In Study 1a and Study 1b, we explored whether one group member’s behavior can affect subjects’ decision-making. If one group member can influence subjects’ decision-making, we can infer that the credit of one’s partners is a basic factor in people’s individual or collective solution decision-making; furthermore, with more group members behaving in a similar way, the effect would be more significant. The hypothesis of Study 1a and Study 1b was that participants’ decision-making would be affected by one group member’s tendencies toward self-interest and altruism (i.e., that it would increase with an increase in the altruism tendencies and decrease with an increase in the self-interest tendencies). After we prove one group member’s behavior can influence participants’ individual or collective strategy decision-making in Study 1a and Study 1b, then, in Study 2a and Study 2b, we allocate the same self-interest or altruism tendencies to 2 to 5 group members. The hypothesis of Study 2a was that, with the total altruistic tendencies remaining unchanged, the more altruistic group members the altruistic tendencies were allocated to, the higher a participant’s level of trust in the team would be, while the hypothesis of Study 2b was that, with the total self-interest tendencies remaining unchanged, the fewer self-interest-based group members the self-interest tendencies were allocated to, the higher a participant’s level of trust in the team would be.

According to previous research (Gross and De Dreu, 2019), when the cost ratio (i.e., cost ratio = the cost of the individual solution/the cost of the collective solution) increases to 1.75, people tend to choose the collective solution. In Study 1a, the cost ratio of the individual versus the collective solution was set to 1.75 as a “cooperation-oriented” situation. Because the cost of the collective solution is usually much lower than that of the individual solution in real life, we set the cooperation-oriented situation to better simulate reality.

Study 1a aimed to explore how the altruism and self-interest of one of the partners would influence people’s cooperation tendencies under the “cooperation-oriented” situation when an individual solution was available.

In Study 1b, the cost ratio of individually versus collectively solving the shared problem was set as 1.35, which was the “individual-oriented” situation.

Study 1b aimed to explore how the altruism of one partner influences people’s cooperative tendency under an individual-oriented situation when an individual solution is available.

After we prove one group member’s behavior can influence participants’ individual or collective strategy decision-making in Study 1a and Study 1b, we intended to investigate the exact nature of the relationship between others’ and our own behavior in a detailed way.

Study 2a aimed to explore whether people pay more attention to the number of altruists or are more likely to be influenced by a few obvious altruists. In this study, we distributed a generous amount of 10 RPs (i.e., see Table 1 for the definition of “a generous amount”) equally among 2 to 5 virtual subjects in the group to explore whether there was a difference in people’s trust in the a collective solution under the convergent (i.e., a generous amount of 10 RPs was equally allocated to 2 virtual subjects) or divergent (i.e., a generous amount of 10 RPs was equally allocated to 5 virtual subjects) distribution of an equivalent generous amount.

Study 2b aimed to explore whether people pay more attention to the number of teammates who are self-interested or are more likely to be influenced by a few obvious self-interested teammates. Study 2b distributed a self-interest-based amount of 5 RPs equally among 2–5 virtual subjects in the team to explore whether there was a difference in people’s trust degree under the convergent (i.e., the self-interest amount of 5 RPs was equally allocated to two virtual subjects) or divergent (i.e., the self-interest amount of 5 RPs was equally allocated to five virtual subjects) distribution of an equivalent self-interest-based amount.

The results are consistent with our hypothesis. The results of Study 1a and Study 1b not only showed that participants’ decision-making was affected by their teammates’ tendency toward self-interest and altruism but also presented an upper limit of the influence of their teammates’ credibility. The reason for this outcome may be that once the teammates’ credibility exceeded a certain amount, the cost ratio of the individual solution versus the collective solution mattered more in participants’ decision-making. The results of Study 2a and Study 2b showed that, with the total altruistic/self-interest tendency unchanged, the participants attached more importance to the number of team members who presented a slightly altruistic or self-interest-based tendency rather than the generously altruistic team members or the extremely self-interested team members. The participants might have thought that the proportion of their teammates who were willing to cooperate in a team could better predict the possibility of cooperation success in the second trial.

Our study found that the divergent distribution of generosity can improve the cooperative tendency of the subjects, which has some implications for people’s preference for fairness. Chou et al. (2016) demonstrated that the monomorphic society of fair preference would be the only evolutionarily stable equilibrium by establishing evolutionary game models and random evolution simulation models. Humans’ preference for fairness may result from the adaptive advantage of instinctive fairness behavior in early human evolution (Chen et al., 2020; Iwamura et al., 2020). Cooperation is a necessary and more effective means for human society to increase individual welfare; however, people who are in the search for cooperation need to face the instability of cooperation opportunities and benefits. Although people can obtain the maximum profit during one-shot cooperation when focusing on the benefit of cooperation, it will become more difficult for these individuals to find such partners in the future. Thus, their opportunities for cooperation will decrease in the long term. In contrast, although it will increase their chance of cooperation when people contribute much more than their fair share to the collective solution, the profit attached to one-shot cooperation decreases. Regardless of whether cooperation opportunities decrease or the cooperation benefits decline, this approach is not conducive to an individual’s survival. As a result, a fair state of cooperation is also preferred from the individual’s point of view. The group in which every member both contributes and benefits equally tends to convince participants that the group can sustain effective cooperation in the long run.

We must acknowledge that the sensitivity difference between the generous factor and the self-interest factor in Study 1a can also be explained by people’s preference for fairness. Participants had a higher tendency to cooperate in the control round, which may be because the subjects preferred a fair state in which everyone contributed equally to the collective target from the overall perspective. From a local point of view (i.e., the participants paid independent attention to the generous factor and the self-interest factor), however, this may be because the subjects were indeed more sensitive to the self-interest factor when generous and self-interest factors coexisted. In either way, this outcome reflects the fact that when individual solutions are available, people’s propensity to collaborate increases when all partners tend to cooperate to a similar extent. Further research needs to be done to explore the sensitivity difference.

Overall, the results indicated that the participants showed largely rational responses. Except for a few participants who did not understand the experimental rules thoroughly, there was no one to invest RPs in either the public pool or private pool. Because only one of the individual targets and the collective target needed to be reached, if the participants invested RPs in both pools, then the remaining RPs would decrease dramatically. Even if there were several participants investing in both pools during the first or second round, after observing his or her teammates’ behavior, he or she would quickly switch to the one-pool strategy (i.e., considering that there were only several rounds of two-pool investing, in the analysis process, the larger number of the two invested values was regarded as the participants’ real decision). Moreover, the variables we manipulated reasonably affected the participants’ probability of contributing to the collective solution. Any sort of variability in the contributions of others would suggest that their contributions were somewhat unpredictable, thus creating a risk that one’s own contribution would be wasted and thereby changing one’s own decision-making. Thus, we must ask the following question: if everyone was simply rational and insisted on the collective solution regardless of any sort of variability, would the Nash equilibrium (i.e., everyone can obtain the maximal profit) (Holler and Klose-Ullmann, 2020) be achieved? Obviously, the “if” is impossible to determine. It should be emphasized that since one participant must know the others’ strategies to know his or her own feasible strategy set, the others cannot determine their feasible strategies without knowing the one participant’s strategy (Ye and Hu, 2017; Yu et al., 2017; Rubinstein, 2018; Holler and Klose-Ullmann, 2020). In a real interactive game that used the same paradigm as that used in our study, Gross and De Dreu (2019) observed that few people attempted to achieve the Nash equilibrium by insisting on the collective solution. More specifically, there is literature (Tyran, 2004) on step-level games such as ours in which it is argued that the rational strategy is to contribute more when the probability of being “pivotal” is higher. Interestingly, there seems to be a “bandwagon” effect beyond this point, such that people contribute even when their contribution is not needed. These results all indicate that people cannot be completely rational, i.e., similar to machines, and that the Nash equilibrium is impossible to achieve in these circumstances.

As material resources become increasingly abundant, human individualism is becoming more abundant (Santos et al., 2017). An increasing number of individual solutions to shared problems have emerged (Izquierdo et al., 2010). As a result, it is worthwhile for us to conduct more research to understand how individual solutions to shared problems influence the human inclination for cooperation and coordination. In future research, we plan to design more delicate experiments to explore the sensitivity difference between the generous factor and the self-interest-based factor. The explicit reason why the divergent distribution of generous amounts and the convergent distribution of self-interest-based amounts are preferred is also worthwhile to determine. Moreover, the interaction between the cost ratio and the teammates’ credibility may be a practical tissue worth examining. If the strong Nash equilibrium (i.e., solutions for a non-cooperative environment where players are allowed to discuss their strategies and to form coalitions without any binding commitment) (Petrosyan et al., 2017) was utilized when the individual solution was available, would the Nash equilibrium be achieved or would cooperation at least be facilitated?

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author.

The studies involving human participants were reviewed and approved by South China Normal University. The patients/participants provided their written informed consent to participate in this study.

MZ developed the study concept and drafted the manuscript. JF performed testing and data collection. YY, JJL, JBL, and LM did the proofreading. NQ, JBL, and JDD provided critical revisions and participated in the final revision of the manuscript. All authors approved the final version of the manuscript for submission.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.