Language-based game theory in the age of artificial intelligence.

Understanding human behaviour in decision problems and strategic interactions has wide-ranging applications in economics, psychology and artificial intelligence. Game theory offers a robust foundation for this understanding, based on the idea that individuals aim to maximize a utility function. However, the exact factors influencing strategy choices remain elusive. While traditional models try to explain human behaviour as a function of the outcomes of available actions, recent experimental research reveals that linguistic content significantly impacts decision-making, thus prompting a paradigm shift from outcome-based to language-based utility functions. This shift is more urgent than ever, given the advancement of generative AI, which has the potential to support humans in making critical decisions through language-based interactions. We propose sentiment analysis as a fundamental tool for this shift and take an initial step by analysing 61 experimental instructions from the dictator game, an economic game capturing the balance between self-interest and the interest of others, which is at the core of many social interactions. Our meta-analysis shows that sentiment analysis can explain human behaviour beyond economic outcomes. We discuss future research directions. We hope this work sets the stage for a novel game-theoretical approach that emphasizes the importance of language in human decisions.

The dual-process approach to human sociality: Meta-analytic evidence for a theory of internalized heuristics for self-preservation.

Which social decisions are influenced by intuitive processes? Which by deliberative processes? The dual-process approach to human sociality has emerged in the last decades as a vibrant and exciting area of research. Yet a perspective that integrates empirical and theoretical work is lacking. This review and meta-analysis synthesizes the existing literature on the cognitive basis of cooperation, altruism, truth telling, positive and negative reciprocity, and deontology and develops a framework that organizes the experimental regularities. The meta-analytic results suggest that intuition favors a set of heuristics that are related to the instinct for self-preservation: people avoid being harmed, avoid harming others (especially when there is a risk of harm to themselves), and are averse to disadvantageous inequalities. Finally, this article highlights some key research questions to further advance our understanding of the cognitive foundations of human sociality. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

A synthesis of evidence for policy from behavioural science during COVID-19.

Scientific evidence regularly guides policy decisions1, with behavioural science increasingly part of this process2. In April 2020, an influential paper3 proposed 19 policy recommendations ('claims') detailing how evidence from behavioural science could contribute to efforts to reduce impacts and end the COVID-19 pandemic. Here we assess 747 pandemic-related research articles that empirically investigated those claims. We report the scale of evidence and whether evidence supports them to indicate applicability for policymaking. Two independent teams, involving 72 reviewers, found evidence for 18 of 19 claims, with both teams finding evidence supporting 16 (89%) of those 18 claims. The strongest evidence supported claims that anticipated culture, polarization and misinformation would be associated with policy effectiveness. Claims suggesting trusted leaders and positive social norms increased adherence to behavioural interventions also had strong empirical support, as did appealing to social consensus or bipartisan agreement. Targeted language in messaging yielded mixed effects and there were no effects for highlighting individual benefits or protecting others. No available evidence existed to assess any distinct differences in effects between using the terms 'physical distancing' and 'social distancing'. Analysis of 463 papers containing data showed generally large samples; 418 involved human participants with a mean of 16,848 (median of 1,699). That statistical power underscored improved suitability of behavioural science research for informing policy decisions. Furthermore, by implementing a standardized approach to evidence selection and synthesis, we amplify broader implications for advancing scientific evidence in policy formulation and prioritization.

Social and moral psychology of COVID-19 across 69 countries.

The COVID-19 pandemic has affected all domains of human life, including the economic and social fabric of societies. One of the central strategies for managing public health throughout the pandemic has been through persuasive messaging and collective behaviour change. To help scholars better understand the social and moral psychology behind public health behaviour, we present a dataset comprising of 51,404 individuals from 69 countries. This dataset was collected for the International Collaboration on Social & Moral Psychology of COVID-19 project (ICSMP COVID-19). This social science survey invited participants around the world to complete a series of moral and psychological measures and public health attitudes about COVID-19 during an early phase of the COVID-19 pandemic (between April and June 2020). The survey included seven broad categories of questions: COVID-19 beliefs and compliance behaviours; identity and social attitudes; ideology; health and well-being; moral beliefs and motivation; personality traits; and demographic variables. We report both raw and cleaned data, along with all survey materials, data visualisations, and psychometric evaluations of key variables.

Political ideology and generosity around the globe.

In a world severely put under stress by COVID-19, generosity becomes increasingly essential both when able to transcend local boundaries, building upon universalistic values, and when directed toward more local contexts, such as the native country. This study aims to investigate an underresearched determinant of generosity at these two levels, a factor that captures one's beliefs, values, and opinions about society: political ideology. We study the donation decisions of more than 46,000 participants from 68 countries in a task with the possibility of donating to a national charity and an international one. We test whether more left-leaning individuals display higher generosity in general (H1) and toward international charities (H2). We also examine the association between political ideology and national generosity without hypothesizing any direction. We find that more left-leaning individuals are more likely to donate in general and more likely to be generous internationally. We also observe that more right-leaning individuals are more likely to donate nationally. These results are robust to the inclusion of several controls. In addition, we address a relevant source of cross-country variation, the quality of governance, which is found to have significant informative power in explaining the relationship between political ideology and the different types of generosity. Potential mechanisms underlying the resulting behaviors are discussed.

"I Think This News Is Accurate": Endorsing Accuracy Decreases the Sharing of Fake News and Increases the Sharing of Real News.

Accuracy prompts, nudges that make accuracy salient, typically decrease the sharing of fake news, while having little effect on real news. Here, we introduce a new accuracy prompt that is more effective than previous prompts, because it does not only reduce fake news sharing, but it also increases real news sharing. We report four preregistered studies showing that an "endorsing accuracy" prompt ("I think this news is accurate"), placed into the sharing button, decreases fake news sharing, increases real news sharing, and keeps overall engagement constant. We also explore the mechanism through which the intervention works. The key results are specific to endorsing accuracy, rather than accuracy salience, and endorsing accuracy does not simply make participants apply a "source heuristic." Finally, we use Pennycook et al.'s limited-attention model to argue that endorsing accuracy may work by making people more carefully consider their sharing decisions.

Predicting attitudinal and behavioral responses to COVID-19 pandemic using machine learning.

At the beginning of 2020, COVID-19 became a global problem. Despite all the efforts to emphasize the relevance of preventive measures, not everyone adhered to them. Thus, learning more about the characteristics determining attitudinal and behavioral responses to the pandemic is crucial to improving future interventions. In this study, we applied machine learning on the multinational data collected by the International Collaboration on the Social and Moral Psychology of COVID-19 (N = 51,404) to test the predictive efficacy of constructs from social, moral, cognitive, and personality psychology, as well as socio-demographic factors, in the attitudinal and behavioral responses to the pandemic. The results point to several valuable insights. Internalized moral identity provided the most consistent predictive contribution-individuals perceiving moral traits as central to their self-concept reported higher adherence to preventive measures. Similar results were found for morality as cooperation, symbolized moral identity, self-control, open-mindedness, and collective narcissism, while the inverse relationship was evident for the endorsement of conspiracy theories. However, we also found a non-neglible variability in the explained variance and predictive contributions with respect to macro-level factors such as the pandemic stage or cultural region. Overall, the results underscore the importance of morality-related and contextual factors in understanding adherence to public health recommendations during the pandemic.

National identity predicts public health support during a global pandemic.

Changing collective behaviour and supporting non-pharmaceutical interventions is an important component in mitigating virus transmission during a pandemic. In a large international collaboration (Study 1, N = 49,968 across 67 countries), we investigated self-reported factors associated with public health behaviours (e.g., spatial distancing and stricter hygiene) and endorsed public policy interventions (e.g., closing bars and restaurants) during the early stage of the COVID-19 pandemic (April-May 2020). Respondents who reported identifying more strongly with their nation consistently reported greater engagement in public health behaviours and support for public health policies. Results were similar for representative and non-representative national samples. Study 2 (N = 42 countries) conceptually replicated the central finding using aggregate indices of national identity (obtained using the World Values Survey) and a measure of actual behaviour change during the pandemic (obtained from Google mobility reports). Higher levels of national identification prior to the pandemic predicted lower mobility during the early stage of the pandemic (r = -0.40). We discuss the potential implications of links between national identity, leadership, and public health for managing COVID-19 and future pandemics.

Evolution of honesty in higher-order social networks.

Sender-receiver games are simple models of information transmission that provide a formalism to study the evolution of honest signaling and deception between a sender and a receiver. In many practical scenarios, lies often affect groups of receivers, which inevitably entangles the payoffs of individuals to the payoffs of other agents in their group, and this makes the formalism of pairwise sender-receiver games inapt for where it might be useful the most. We therefore introduce group interactions among receivers and study how their interconnectedness in higher-order social networks affects the evolution of lying. We observe a number of counterintuitive results that are rooted in the complexity of the underlying evolutionary dynamics, which has thus far remained hidden in the realm of pairwise interactions. We find conditions for honesty to persist even when there is a temptation to lie, and we observe the prevalence of moral strategy profiles even when lies favor the receiver at a cost to the sender. We confirm the robustness of our results by further performing simulations on hypergraphs created from real-world data using the SocioPatterns database. Altogether, our results provide persuasive evidence that moral behavior may evolve on higher-order social networks, at least as long as individuals interact in groups that are small compared to the size of the network.

Telling people to "rely on their reasoning" increases intentions to wear a face covering to slow down COVID-19 transmission.

Finding messaging to promote the use of face masks is fundamental during a pandemic. Study 1 (N = 399) shows that telling people to "rely on their reasoning" increases intentions to wear a face mask, compared with telling them to "rely on their emotions." In Study 2 (N = 591) we add a baseline. However, the results show only a non-significant trend. Study 3 reports a well-powered replication of Study 2 (N = 930). In line with Study 1, this study shows that telling people to "rely on their reasoning" increases intentions to wear a face mask, compared to telling them to "rely on their emotions." Two internal meta-analyses show that telling people to "rely on their reasoning" increases intentions to wear a face mask compared (1) to telling them to "rely on their emotions" and (2) to the baseline. These findings suggest interventions to promote intentions to wear a face mask.

Mathematical foundations of moral preferences.

One-shot anonymous unselfishness in economic games is commonly explained by social preferences, which assume that people care about the monetary pay-offs of others. However, during the last 10 years, research has shown that different types of unselfish behaviour, including cooperation, altruism, truth-telling, altruistic punishment and trustworthiness are in fact better explained by preferences for following one's own personal norms-internal standards about what is right or wrong in a given situation. Beyond better organizing various forms of unselfish behaviour, this moral preference hypothesis has recently also been used to increase charitable donations, simply by means of interventions that make the morality of an action salient. Here we review experimental and theoretical work dedicated to this rapidly growing field of research, and in doing so we outline mathematical foundations for moral preferences that can be used in future models to better understand selfless human actions and to adjust policies accordingly. These foundations can also be used by artificial intelligence to better navigate the complex landscape of human morality.

The evolution of trust and trustworthiness.

Trust and trustworthiness form the basis for continued social and economic interactions, and they are also fundamental for cooperation, fairness, honesty, and indeed for many other forms of prosocial and moral behaviour. However, trust entails risks, and building a trustworthy reputation requires effort. So how did trust and trustworthiness evolve, and under which conditions do they thrive? To find answers, we operationalize trust and trustworthiness using the trust game with the trustor's investment and the trustee's return of the investment as the two key parameters. We study this game on different networks, including the complete network, random and scale-free networks, and in the well-mixed limit. We show that in all but one case, the network structure has little effect on the evolution of trust and trustworthiness. Specifically, for well-mixed populations, lattices, random and scale-free networks, we find that trust never evolves, while trustworthiness evolves with some probability depending on the game parameters and the updating dynamics. Only for the scale-free network with degree non-normalized dynamics, we find parameter values for which trust evolves but trustworthiness does not, as well as values for which both trust and trustworthiness evolve. We conclude with a discussion about mechanisms that could lead to the evolution of trust and outline directions for future work.

Using social and behavioural science to support COVID-19 pandemic response.

The COVID-19 pandemic represents a massive global health crisis. Because the crisis requires large-scale behaviour change and places significant psychological burdens on individuals, insights from the social and behavioural sciences can be used to help align human behaviour with the recommendations of epidemiologists and public health experts. Here we discuss evidence from a selection of research topics relevant to pandemics, including work on navigating threats, social and cultural influences on behaviour, science communication, moral decision-making, leadership, and stress and coping. In each section, we note the nature and quality of prior research, including uncertainty and unsettled issues. We identify several insights for effective response to the COVID-19 pandemic and highlight important gaps researchers should move quickly to fill in the coming weeks and months.

Lying on networks: The role of structure and topology in promoting honesty.

Lies can have a negating impact on governments, companies, and the society as a whole. Understanding the dynamics of lying is therefore of crucial importance across different fields of research. While lying has been studied before in well-mixed populations, it is a fact that real interactions are rarely well-mixed. Indeed, they are usually structured and thus best described by networks. Here we therefore use the Monte Carlo method to study the evolution of lying in the sender-receiver game in a one-parameter family of networks, systematically covering complete networks, small-world networks, and one-dimensional rings. We show that lies that benefit the sender at a cost to the receiver, the so-called black lies, are less likely to proliferate on networks than they do in well-mixed populations. Honesty is thus more likely to evolve, but only when the benefit for the sender is smaller than the cost for the receiver. Moreover, this effect is particularly strong in small-world networks, but less so in the one-dimensional ring. For lies that favor the receiver at a cost to the sender, the so-called altruistic white lies, we show that honesty is also more likely to evolve than it is in well-mixed populations. But contrary to black lies, this effect is more expressed in the one-dimensional ring, whereas in small-world networks it is present only when the cost to the sender is greater than the benefit for the receiver. Last, for lies that benefit both the sender and the receiver, the so-called Pareto white lies, we show that the network structure actually favors the evolution of lying, but this only occurs when the benefit for the sender is slightly greater than the benefit for the receiver. In this case again the small-world topology acts as an amplifier of the effect, while other network topologies fail to do the same. In addition to these main results we discuss several other findings, which together show clearly that the structure of interactions and the overall topology of the network critically determine the dynamics of lying.

Correction: People making deontological judgments in the Trapdoor dilemma are perceived to be more prosocial in economic games than they actually are.

[This corrects the article DOI: 10.1371/journal.pone.0205066.].

Author Correction: Group size effects and critical mass in public goods games.

An amendment to this paper has been published and can be accessed via a link at the top of the paper.

Increasing altruistic and cooperative behaviour with simple moral nudges.

The conflict between pro-self and pro-social behaviour is at the core of many key problems of our time, as, for example, the reduction of air pollution and the redistribution of scarce resources. For the well-being of our societies, it is thus crucial to find mechanisms to promote pro-social choices over egoistic ones. Particularly important, because cheap and easy to implement, are those mechanisms that can change people's behaviour without forbidding any options or significantly changing their economic incentives, the so-called "nudges". Previous research has found that moral nudges (e.g., making norms salient) can promote pro-social behaviour. However, little is known about whether their effect persists over time and spills across context. This question is key in light of research showing that pro-social actions are often followed by selfish actions, thus suggesting that some moral manipulations may backfire. Here we present a class of simple moral nudges that have a great positive impact on pro-sociality. In Studies 1-4 (total N = 1,400), we use economic games to demonstrate that asking subjects to self-report "what they think is the morally right thing to do" does not only increase pro-sociality in the choice immediately after, but also in subsequent choices, and even when the social context changes. In Study 5, we explore whether moral nudges promote charity donations to humanitarian organisations in a large (N = 1,800) crowdfunding campaign. We find that, in this context, moral nudges increase donations by about 44 percent.

The evolution of lying in well-mixed populations.

Lies can have profoundly negative consequences for individuals, groups and even for societies. Understanding how lying evolves and when it proliferates is therefore of significant importance for our personal and societal well-being. To that effect, we here study the sender-receiver game in well-mixed populations with methods of statistical physics. We use the Monte Carlo method to determine the stationary frequencies of liars and believers for four different lie types. We consider altruistic white lies that favour the receiver at a cost to the sender, black lies that favour the sender at a cost to the receiver, spiteful lies that harm both the sender and the receiver, and Pareto white lies that favour both the sender and the receiver. We find that spiteful lies give rise to trivial behaviour, where senders quickly learn that their best strategy is to send a truthful message, while receivers likewise quickly learn that their best strategy is to believe the sender's message. For altruistic white lies and black lies, we find that most senders lie while most receivers do not believe the sender's message, but the exact frequencies of liars and non-believers depend significantly on the payoffs, and they also evolve non-monotonically before reaching the stationary state. Lastly, for Pareto white lies we observe the most complex dynamics, with the possibility of both lying and believing evolving with all frequencies between 0 and 1 in dependence on the payoffs. We discuss the implications of these results for moral behaviour in human experiments.

Group size effects and critical mass in public goods games.

Understanding whether the size of the interacting group has an effect on cooperative behavior has been a major topic of debate since the seminal works on cooperation in the 1960s. Half a century later, scholars have yet to reach a consensus, with some arguing that cooperation is harder in larger groups, while others that cooperation is easier in larger groups, and yet others that cooperation attains its maximum in intermediate size groups. Here we add to this field of work by reporting a two-treatment empirical study where subjects play a Public Goods Game with a Critical Mass, such that the return for full cooperation increases linearly for early contributions and then stabilizes after a critical mass is reached (the two treatments differ only on the critical mass). We choose this game for two reasons: it has been argued that it approximates real-life social dilemmas; previous work suggests that, in this case, group size might have an inverted-U effect on cooperation, where the pick of cooperation is reached around the critical mass. Our main innovation with respect to previous experiments is that we implement a within-subject design, such that the same subject plays in groups of different size (from 5 to 40 subjects). Groups are formed at random at every round and there is no feedback. This allows us to explore if and how subjects change their choice as a function of the size of the group. We report three main results, which partially contrast what has been suggested by previous work: in our setting (i) the critical mass has no effect on cooperation; (ii) group size has a positive effect on cooperation; (iii) the most chosen option (played by about 50% of the subjects) is All Defection, followed by All Cooperation (about 10% of the subjects), whereas the rest have a slight trend to switch preferentially from defection to cooperation as the group size increases.