Perceptions of Declining Cooperation: Insights from a Cross-Cultural Survey
Recent findings reveal that a majority of both American and Chinese respondents believe that people have become less considerate, helpful, or cooperative over the past sixty years—a period extending a few years into the future. Specifically, 65% of Americans and 62.1% of Chinese participants expressed this view, while 23% of Americans and 28.5% of Chinese believed the opposite, and the remainder anticipated no significant change. Statistical analysis showed no significant difference between the two countries’ distributions (chi-square(2) = 5.36, P = 0.069), highlighting a shared societal perception despite diverse cultural and historical backgrounds.
This perception taps into a common sentiment about social decline, now quantitatively measured across two vastly different societies. The study offers a rare comparative glimpse into cross-cultural beliefs about human cooperation.
The Incentivized Guesswork Behind the Study
Yi Liu, Giuliana Spadaro, Sümeyye Ergün, Yingling Li, and Paul Van Lange conducted a preregistered survey involving 628 American and 449 Chinese adults, published in Communications Psychology in March 2026. Unlike typical opinion polls, this survey required participants to engage with a classic prisoner’s dilemma game scenario. Respondents read the rules—where defecting against a cooperator yields a payoff of 60 points, mutual defection yields 20 points each, and mutual cooperation yields 40 points each—and then estimated the percentage of people in various years who would have chosen to cooperate.
Importantly, these estimates were incentivized. Respondents who guessed within ±5% of benchmark cooperation rates received a small bonus. These benchmarks were derived from archived experimental data spanning past decades and included a forecast for 2030 based on historical trends. This design encouraged thoughtful, data-informed responses rather than casual opinions.
Each participant evaluated cooperation rates for four specific years, divided into two groups: one assessed 1960, 1980, 2000, and 2020; the other 1970, 1990, 2010, and 2030. Both groups indicated a downward trend in cooperation over time, with older participants perceiving a steeper decline. Although the samples originated from professional online panels—which may over-represent urban, wealthier, and better-educated individuals—weighting the data by age, education, and income did not alter the overall pattern of perceived decline. The Chinese sample fell short of its recruitment target, and some responses were excluded due to incompleteness (12% of Chinese and 4% of American responses).
Contrasting Beliefs with Empirical Data
What makes this study particularly intriguing is the paradox it presents. The authors contrast their survey findings with a substantial body of prior empirical research housed in the Cooperation Databank, which compiles 2,636 studies conducted between 1958 and 2017 across 78 societies. Yuan and colleagues, in two meta-analyses, found evidence of increased cooperation over time: a 21% rise in the United States from 1956 to 2017, and a 36% increase in China from 1999 to 2019. These figures encompass various cooperation games such as the prisoner’s dilemma, public goods, and resource dilemma games.
While the current survey reports these historical increases as “small,” it does not clarify whether these percentages represent absolute percentage points or relative changes, which is critical for interpretation. The archival data served as benchmarks for incentivizing participant guesses but were never used to test hypotheses in the present study. Thus, the comparison between perceived and actual cooperation is indirect: it juxtaposes people’s estimates about strangers’ behavior in laboratory games with actual recorded behavior from similar games decades earlier.
It is essential to note that this data does not capture everyday cooperative behaviors outside experimental settings, such as holding doors open or helping neighbors, limiting the scope of conclusions about real-world social conduct.
Distinct Cultural Perspectives on Cooperation Decline
Despite similar overall pessimism, the survey revealed nuanced differences in how Americans and Chinese perceive the decline in cooperation. The Chinese sample expressed a steeper belief in declining cooperation within the game context (β = −0.31, 95% CI [−0.34, −0.28]) compared to the American sample (β = −0.22, 95% CI [−0.25, −0.20]). Both trends were clearly negative.
Respondents evaluated people from each era on four attributes: warmth, morality, assertiveness, and competence. Chinese participants perceived sharper declines in warmth and morality, while American respondents noted more pronounced declines in assertiveness and competence. Notably, Chinese respondents’ ratings of competence slightly increased over time (β = 0.04, 95% CI [0.01, 0.07], P = 0.050), suggesting a perception of stable competence across decades.
Participants who believed cooperation had changed—either positively or negatively—were asked to select and rank up to three reasons from a list of twelve derived from prior literature. Both groups prioritized “trust” as the leading cause of change (weighted rank score of 1.26), followed by stress (0.80), wealth (0.77), and individualism (0.77). Beyond this shared hierarchy, Americans emphasized social media, face-to-face contact, technology, and religiosity, while Chinese respondents attributed more weight to trust, stress, wealth, and education.
The authors caution that these self-reported reasons likely represent folk explanations, potentially reflecting post-hoc rationalizations or prevailing cultural narratives rather than definitive causal mechanisms. Additionally, participants may have conflated trust with trustworthiness, subtly complicating interpretation.
Signs of Stabilization in Future Cooperation Perceptions
The study found no significant evidence of continued decline in cooperation when projecting from 2020 to 2030. The difference in game-based cooperation estimates between these years was statistically nonsignificant (β = −0.10, 95% CI [−0.28, 0.09], P = 0.104), and warmth ratings showed virtually no change (β < 0.01, 95% CI [−0.11, 0.12], P = 0.930). Moreover, for morality, assertiveness, and competence, participants actually rated future decades more positively than the present.
However, the confidence intervals are broad enough to allow for a possible decline that the study was unable to detect, and different participant groups rated 2020 and 2030, limiting direct longitudinal comparisons. The authors suggest this pattern may reflect a tension between a generally functional pessimism about others’ behavior and the inherent optimism people tend to hold about their own futures.
Implications and the Path Forward
It is crucial to emphasize that this research captures beliefs about cooperation rather than directly measuring cooperative behavior in real life. The authors note that distrust can have real consequences, such as discouraging people from seeking needed help—a phenomenon that can reinforce social withdrawal and isolation.
The authors highlight a significant gap in the literature: the absence of longitudinal data tracking cooperation as it naturally occurs outside laboratory settings. Establishing such a record would provide a more definitive understanding of how cooperative behavior has evolved over decades.
Until then, the most defensible conclusion is that the belief in a six-decade decline in cooperation is well documented across two culturally distinct populations. In contrast, the extensive game data cited show an opposing trend of increasing cooperation in formal experimental contexts. This paradox invites further research and reflection on how perceptions and realities of cooperation interplay in shaping social trust.
For those interested, the full study and discussion can be accessed here.
