Search

We present eight experiments exploring gender biases in GPT. Initially, GPT was asked to generate demographics of a potential writer of fourty phrases ostensibly written by elementary school students, twenty containing feminine stereotypes and twenty with masculine stereotypes. Results show a strong bias, with stereotypically masculine sentences attributed to a female more often than vice versa. For example, the sentence “I love playing fotbal! Im practicing with my cosin Michael” was constantly assigned by GPT-3.5 Turbo to a female writer. This phenomenon likely reflects that while initiatives to integrate women in traditionally masculine roles have gained momentum, the reverse movement remains relatively underdeveloped. Subsequent experiments investigate the same issue in high-stakes moral dilemmas. GPT-4 finds it more appropriate to abuse a man to prevent a nuclear apocalypse than to abuse a woman. This bias extends to other forms of violence central to the gender parity debate (abuse), but not to those less central (torture). Moreover, this bias increases in cases of mixed-sex violence for the greater good: GPT-4 agrees with a woman using violence against a man to prevent a nuclear apocalypse but disagrees with a man using violence against a woman for the same purpose. Finally, these biases are implicit, as they do not emerge when GPT-4 is directly asked to rank moral violations. These results highlight the necessity of carefully managing inclusivity efforts to prevent unintended discrimination.

Abstract 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.

We report on three pre-registered studies testing whether people in the position of describing a decision problem to decision-makers exploit this opportunity for their benefit, by choosing descriptions that may be potentially beneficial for themselves. In Study 1, recipients of an extreme dictator game (where dictators can either take the whole pie for themselves or give it entirely to the receiver) are asked to choose the instructions used to introduce the game to dictators, from six different instructions known from previous research to affect dictators’ decisions. The results demonstrate that some dictator game recipients tend to choose instructions that make them more likely to receive a higher payoff. Study 2 shows that people who choose descriptions that make them more likely to receive a higher payoff indeed believe that they will receive a higher payoff. Study 3 shows that receivers are more likely than dictators to choose these self-serving descriptions. In sum, our work suggests that some people choose descriptions that are beneficial to themselves; we also found some evidence that deliberative thinking and young age are associated with this tendency.

Understanding collective behaviour is an important aspect of managing the pandemic response. Here the authors show in a large global study that participants that reported identifying more strongly with their nation reported greater engagement in public health behaviours and support for public health policies in the context of the pandemic.

Generations of social scientists have explored whether males and females act differently in domains involving competition, risk taking, cooperation, altruism, honesty, as well as many others. Yet, little is known about gender differences in the trade-off between objective equality (i.e., equality of outcomes) and efficiency. It has been suggested that females are more equal than males, but the empirical evidence is relatively weak. This gap is particularly important, because people in power of redistributing resources often face a conflict between equality and efficiency. The recently introduced Trade-Off Game (TOG) – in which a decision-maker has to unilaterally choose between being equal or being efficient – offers a unique opportunity to fill this gap. To this end, I analyse gender differences on a large dataset including N=6,955 TOG decisions. The results show that females prefer objective equality over efficiency to a greater extent than males do. The effect turns out to be particularly strong when the TOG available options are “morally” framed in such a way to suggest that choosing the equal option is the right thing to do.

In this paper we investigate the effect of moral suasion on ingroup favouritism. We report a well-powered, pre-registered, two-stage 2x2 mixed-design experiment. In the first stage, groups are formed on the basis of how participants answer a set of questions, concerning non-morally relevant issues in one treatment (assorting on non-moral preferences), and morally relevant issues in another treatment (assorting on moral preferences). In the second stage, participants choose how to split a given amount of money between participants of their own group and participants of the other group, first in the baseline setting and then in a setting where they are told to do what they believe to be morally right (moral suasion). Our main results are: (i) in the baseline, participants tend to favour their own group to a greater extent when groups are assorted according to moral preferences, compared to when they are assorted according to non-moral preferences; (ii) the net effect of moral suasion is to decrease ingroup favouritism, but there is also a non-negligible proportion of participants for whom moral suasion increases ingroup favouritism; (iii) the effect of moral suasion is substantially stable across group assorting and four pre-registered individual characteristics (gender, political orientation, religiosity, pro-life vs pro-choice ethical convictions).

Understanding whether preferences are sensitive to the frame has been a major topic of debate in the last decades. For example, several works have explored whether the dictator game in the give frame gives rise to a different rate of pro-sociality than the same game in the take frame, leading to mixed results. Here we contribute to this debate with two experiments. In Study 1 (N=567) we implement an extreme dictator game in which the dictator either gets $0.50 and the recipient gets nothing, or the opposite (i.e., the recipient gets $0.50 and the dictator gets nothing). We experimentally manipulate the words describing the available actions using six terms, from very negative (e.g., stealing) to very positive (e.g., donating) connotations. We find that the rate of pro-sociality is affected by the words used to describe the available actions. In Study 2 (N=221) we ask brand new participants to rate each of the words used in Study 1 from “extremely wrong” to “extremely right”. We find that these moral judgments can explain the framing effect in Study 1. In sum, our studies provide evidence that framing effects in an extreme Dictator game can be generated using morally loaded language.

Whether there are gender differences in lying has been largely debated in the past decade. Previous studies found mixed results. To shed light on this topic, here I report a meta-analysis of 8,728 distinct observations, collected in 65 Sender-Receiver game treatments, by 14 research groups. Following previous work and theoretical considerations, I distinguish three types of lies: black lies, which benefit the liar at a cost for another person; altruistic white lies, which benefit another person at a cost for the liar; and Pareto white lies, which benefit both the liar and another person. The results show that: males are significantly more likely than females to tell black lies (N=4,173); males are significantly more likely than females to tell altruistic white (N=2,940); and results are inconclusive in the case of Pareto white lies (N=1,615). Furthermore, gender differences in telling altruistic white lies are significantly stronger than in the other two cases.

Decades of experimental research show that some people forgo personal gains to benefit others in unilateral anonymous interactions. To explain these results, behavioral economists typically assume that people have social preferences for minimizing inequality and/or maximizing efficiency (social welfare). Here we present data that cannot be explained by these standard social preference models. We use a “Trade-Off Game” (TOG), where players unilaterally choose between an equitable option and an efficient option. We show that simply changing the labelling of the options to describe the equitable versus efficient option as morally right completely reverses the correlation between behavior in the TOG and play in a separate Dictator Game (DG) or Prisoner’s Dilemma (PD): people who take the action framed as moral in the TOG, be it equitable or efficient, are much more prosocial in the DG and PD. Rather than preferences for equity and/or efficiency per se, our results suggest that prosociality in games such as the DG and PD are driven by a generalized morality preference that motivates people to do what they think is morally right.

The appearance of a new coronavirus (Covid-19) and its rapid expansion throughout the world has forced all countries to establish regulations based on social confinement. In the early days of a pandemic, the adherence to regulations is crucial to be able to block its spread. This research aims to analyse the relationship between motivational variables associated with physical distancing and self-interested consumption behaviours in the first 10 days of confinement in Spain. A total of 1,324 people participated throughout the country (mean age 28.92 years). Participants answered an online survey about socio-demographic, motivational variables, which included a) risk information seeking, b) confidence in self- and collective efficacy in coping with the pandemic, and c) the four higher-order personal values ‒conservation (security, conformity, and tradition), self-transcendence (universalism and benevolence), openness (self-direction actions and stimulation), and self-improvement (hedonism and power) ‒ and the aforementioned behaviours in coping with Covid-19. Results showed a positive association between self- and collective efficacy and both coping behaviours analysed: a protective role of conservation values on normative behaviours; and a negative relationship between self-transcendence values and self-interested consumption. Additionally, risk information seeking was positively associated with the development of physical distancing behaviour.

New technological solutions play an important role in preventing the spread of Covid-19. Many countries have implemented tracking applications or other surveillance systems, which may raise concerns about privacy and civil rights violations but may be also perceived by citizens as a way to reduce threat and uncertainty. Our research examined whether feelings evoked by the pandemic (perceived threat and lack of control) as well as more stable ideological views predict the acceptance of such technologies. In two studies conducted in Poland, we found that perceived personal threat and lack of personal control were significantly positively related to the acceptance of surveillance technologies, but their predictive value was smaller than that of individual differences in authoritarianism and endorsement of liberty. Moreover, we found that the relationship between the acceptance of surveillance technologies and both perceived threat and lack of control was particularly strong among people high in authoritarianism. Our research shows that the negative feelings evoked by the unprecedented global crisis may inspire positive attitudes towards helpful but controversial surveillance technologies but that they do so to a lesser extent than ideological beliefs.

We report on an experimental study that was set up to reveal differences in the tendencies of men and women to cooperate in same-sex interactions. Former studies on this subject were mostly conducted in industrialized modern societies. In contrast, we tested the cooperation tendency among Buryats, a people from Southern Siberia of Mongolian origin. All subjects participated in (1) one iterated Public Goods Game in a group of four individuals of the same sex and (2) four one-shot Prisoner’s Dilemma games with different partners of the same sex. The interactions were in a face-to-face setting, but any intentional communication during the experiments was prohibited. We found that Buryat men were more cooperative than Buryat women in both types of same-sex interactions. In particular, the fraction of men employing a strategy of unconditional cooperation in the iterated Public Goods Game was much higher (36%) than the fraction of unconditional cooperators among women (21%). In general, the behavior of men was less context dependent than the behavior of women. In both sexes, individuals who were more cooperative in one type of game tended to be more cooperative in the other type of game. Although direct communication was prohibited, the interaction partners in the Prisoner’s Dilemma games employed the same strategy much more frequently than expected by chance. We conclude that, even among strangers, the exchange of subtle signals is sufficient to coordinate strategic decisions.

Interpersonal trust is an important source of social and economic development. Over decades, researchers debated the question whether and how public institutions influence interpersonal trust, making this relationship a much-discussed issue for scientific debate. However, experimental and behavioral data and insights on this relationship and the underlying psychological processes are rare and often inconsistent. The present set of studies tests a model which proposes that institutional trust indirectly affects trust among unrelated strangers by enhancing individuals’ feelings of security. Study 1 (survey on trust in a broad spectrum of state institutions), Study 2 (nationally representative data from 16 countries), and Study 3 (experimental manipulation of institutional trust) provide convergent support for this hypothesis. Also, the results show that the effect remains consistent even after controlling for individual dispositions linked to interpersonal and institutional trust (Study 1 and 3) and country level indicators of institutional performance (Study 2). Taken together, these findings inform and contribute to the debate about the relationship between institutions and interpersonal trust by showing that when institutions are trusted, they increase feelings of security, and therefore promote interpersonal trust among strangers.

Reputation plays a major role in human societies, and it has been proposed as an explanation for the evolution of cooperation. While the majority of previous studies equates reputation with a transparent and complete history of players’ past decisions, reputations in real life are often ambiguous and opaque. Using web-based experiments, we explore the extent to which opaque reputation works in isolating defectors, with and without partner selection opportunities. We found that low reputation works as a signal of untrustworthiness, whereas medium or high reputations are not taken into account by subjects for orienting their choices. Reputation without partner selection does not promote cooperative behavior; that is, defectors do not turn into cooperators only for the sake of getting a positive reputation. Finally, in a third study, when reputation is pivotal to selection, then a substantial proportion of would-be-defectors turn into cooperators. Taken together, these results provide insights about the characteristics of reputation and about the way in which humans make use of it when selecting partners, and also when knowing that they will be selected.

In the last few decades, numerous experiments have shown that humans do not always behave so as to maximize their material payoff. Cooperative behavior when non-cooperation is a dominant strategy (with respect to the material payoffs) is particularly puzzling. Here we propose a novel approach to explain cooperation, assuming what Halpern and Pass call translucent players. Typically, players are assumed to be opaque, in the sense that a deviation by one player in a normal-form game does not affect the strategies used by other players. But a player may believe that if he switches from one strategy to another, the fact that he chooses to switch may be visible to the other players. For example, if he chooses to defect in Prisoner's Dilemma, the other player may sense his guilt. We show that by assuming translucent players, we can recover many of the regularities observed in human behavior in well-studied games such as Prisoner's Dilemma, Traveler's Dilemma, Bertrand Competition, and the Public Goods game.

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

Methods of statistical physics have proven valuable for studying the evolution of cooperation in social dilemma games. However, recent empirical research shows that cooperative behavior in social dilemmas is only one kind of a more general class of behavior, namely moral behavior, which includes reciprocity, respecting others' property, honesty, equity, efficiency, as well as many others. Inspired by these experimental works, we here open up the path toward studying other forms of moral behavior with methods of statistical physics. We argue that this is a far-reaching direction for future research that can help us answer fundamental questions about human sociality. Why did our societies evolve as they did? What moral principles are more likely to emerge? What happens when different moral principles clash? Can we predict the break out of moral conflicts in advance and contribute to their solution? These are amongst the most important questions of our time, and methods of statistical physics could lead to new insights and contribute toward finding answers.

The opportunity to tell a white lie (i.e., a lie that benefits another person) generates a moral conflict between two opposite moral dictates, one pushing towards telling the truth always and the other pushing towards helping others. Here we study how people resolve this moral conflict. What does telling a white lie signal about a person’s pro-social tendencies? To answer this question, we conducted a two-stage 2x2 experiment. In the first stage, we used a Deception Game to measure aversion to telling a Pareto white lie (i.e., a lie that helps both the liar and the listener), and aversion to telling an altruistic white lie (i.e., a lie that helps the listener at the expense of the liar). In the second stage we measured altruistic tendencies using a Dictator Game and cooperative tendencies using a Prisoner’s dilemma. We found three major results: (i) both altruism and cooperation are positively correlated with aversion to telling a Pareto white lie; (ii) both altruism and cooperation are negatively correlated with aversion to telling an altruistic white lie; (iii) men are more likely than women to tell an altruistic white lie, but not to tell a Pareto white lie. Our results shed light on the moral conflict between prosociality and truth-telling. In particular, the first finding suggests that a significant proportion of people have non-distributional notions of what the right thing to do is, irrespective of the economic consequences, they tell the truth, they cooperate, they share their money.

Why do people make deontological decisions, although they often lead to overall unfavorable outcomes? One account is receiving considerable attention: deontological judgments may signal commitment to prosociality and thus may increase people's chances of being selected as social partners-which carries obvious long-term benefits. Here we test this framework by experimentally exploring whether people making deontological judgments are expected to be more prosocial than those making consequentialist judgments and whether they are actually so. In line with previous studies, we identified deontological choices using the Trapdoor dilemma. Using economic games, we take two measures of general prosociality towards strangers: trustworthiness and altruism. Our results procure converging evidence for a perception gap according to which Trapdoor-deontologists are believed to be more trustworthy and more altruistic towards strangers than Trapdoor-consequentialists, but actually they are not so. These results show that deontological judgments are not universal, reliable signals of prosociality.

We propose a notion of continuous path for locally finite metric spaces, taking inspiration from the recent development of A-theory for locally finite connected graphs. We use this notion of continuity to derive an analogue in Z2 of the Jordan curve theorem and to extend to a quite large class of locally finite metric spaces (containing all finite metric spaces) an inequality for the ℓp-distortion of a metric space that has been recently proved by Pierre-Nicolas Jolissaint and Alain Valette for finite connected graphs.

Groups make decisions on both the production and the distribution of resources. These decisions typically involve a tension between increasing the total level of group resources (i.e. social efficiency) and distributing these resources among group members (i.e. individuals' relative shares). This is the case because the redistribution process may destroy part of the resources, thus resulting in socially inefficient allocations. Here we apply a dual-process approach to understand the cognitive underpinnings of this fundamental tension. We conducted a set of experiments to examine the extent to which different allocation decisions respond to intuition or deliberation. In a newly developed approach, we assess intuition and deliberation at both the trait level (using the Cognitive Reflection Test, henceforth CRT) and the state level (through the experimental manipulation of response times). To test for robustness, experiments were conducted in two countries: the USA and India. Despite absolute-level differences across countries, in both locations we show that: (i) time pressure and low CRT scores are associated with individuals' concerns for their relative shares and (ii) time delay and high CRT scores are associated with individuals' concerns for social efficiency. These findings demonstrate that deliberation favours social efficiency by overriding individuals' intuitive tendency to focus on relative shares.

In a world in which many pressing global issues require large scale cooperation, understanding the group size effect on cooperative behavior is a topic of central importance. Yet, the nature of this effect remains largely unknown, with lab experiments insisting that it is either positive or negative or null, and field experiments suggesting that it is instead curvilinear. Here we shed light on this apparent contradiction by considering a novel class of public goods games inspired to the realistic scenario in which the natural output limits of the public good imply that the benefit of cooperation increases fast for early contributions and then decelerates. We report on a large lab experiment providing evidence that, in this case, group size has a curvilinear effect on cooperation, according to which intermediate-size groups cooperate more than smaller groups and more than larger groups. In doing so, our findings help fill the gap between lab experiments and field experiments and suggest concrete ways to promote large scale cooperation among people.

Cooperation is fundamental to the evolution of human society. We regularly observe cooperative behaviour in everyday life and in controlled experiments with anonymous people, even though standard economic models predict that they should deviate from the collective interest and act so as to maximise their own individual payoff. However, there is typically heterogeneity across subjects: some may cooperate, while others may not. Since individual factors promoting cooperation could be used by institutions to indirectly prime cooperation, this heterogeneity raises the important question of who these cooperators are. We have conducted a series of experiments to study whether benevolence, defined as a unilateral act of paying a cost to increase the welfare of someone else beyond one's own, is related to cooperation in a subsequent one-shot anonymous Prisoner's dilemma. Contrary to the predictions of the widely used inequity aversion models, we find that benevolence does exist and a large majority of people behave this way. We also find benevolence to be correlated with cooperative behaviour. Finally, we show a causal link between benevolence and cooperation: priming people to think positively about benevolent behaviour makes them significantly more cooperative than priming them to think malevolently. Thus benevolent people exist and cooperate more.

Social dilemmas are situations in which collective interests are at odds with private interests: pollution, depletion of natural resources, and intergroup conflicts, are at their core social dilemmas. Because of their multidisciplinarity and their importance, social dilemmas have been studied by economists, biologists, psychologists, sociologists, and political scientists. These studies typically explain tendency to cooperation by dividing people in proself and prosocial types, or appealing to forms of external control or, in iterated social dilemmas, to long-term strategies. But recent experiments have shown that cooperation is possible even in one-shot social dilemmas without forms of external control and the rate of cooperation typically depends on the payoffs. This makes impossible a predictive division between proself and prosocial people and proves that people have attitude to cooperation by nature. The key innovation of this article is in fact to postulate that humans have attitude to cooperation by nature and consequently they do not act a priori as single agents, as assumed by standard economic models, but they forecast how a social dilemma would evolve if they formed coalitions and then they act according to their most optimistic forecast. Formalizing this idea we propose the first predictive model of human cooperation able to organize a number of different experimental findings that are not explained by the standard model. We show also that the model makes satisfactorily accurate quantitative predictions of population average behavior in one-shot social dilemmas.