Empatia | Empathy and sentiment in strategic games

Info: 

Economic theory is used to analyze interactions between consumers, producers, participants of the market, but also ordinary men in the street using so-called game theory. Mathematical approach led economists to describe repetitive mechanisms of the behavior of people participating in the games. And here, by games we can understand almost all possible situations, where there is an interaction between at least two people – game theory allows to provide some insights into both: multi billion dollars business transactions as well as strategies of man and women on a first date. But the standard games theory assumes, that people benefit only from material profits, and, as psychology suggests this is not always the truth. In order to use mathematical language of game theory, often we have to deal with the solutions, that do not correspond with the decisions made in practice.

In this project, we are still taking game theory very seriously, and treat it as a useful tool, but we also allow players for altruistic behavior – which means that it is rational for them to make decisions that are not necessarily beneficial for their profits, but can provide satisfactions from wellbeing of other players. As many studies shows, this approach is much closer to a natural behavior of humans.

This game theory extension, that will be provided by us in this project, even though seems to be so simple and obvious, was never before implemented in the mathematical description of the interactions between people. Allowing players for empathetic decisions is a complex problem when it comes to a game theory. It provides more randomness to the games, and makes solutions no longer straightforward, as it was when the players was simply selfish.

Our project is mostly theoretical, but we will look at the aspect of empathy from several angles. First, we will analyze optimal strategies of players that are responsive to wellbeing of others. Second, we will look at the problem from the perspective of game designer who wants to provide and effective game for empathetic players. Third,  we will analyze concrete relationship – between workers and managers, and how allowing for empathy correlate with business cycle and wages. Finally, we will test theoretical solutions in practice – by set of experiments we will provide evidence whether games theory shows the same mechanisms that we can observe between real people.

Budget: 

Źródło finansowania | Financing: Narodowe Centrum Nauki, OPUS 17

Projekt realizowany | Timeline: 01/2020 -- 01/2024

Kierownik | Principal Investigator: Marek Weretka

Budżet łączny | Total budget: 380 846 zł

  • wynagrodzenia dla podstawowych wykonawców | compensation to researchers: 123 000 zł
  • stypendia dla młodych badaczy | scholarships for young scholars: 108 000 zł
  • komputery i oprogramowanie | hardware and software: 6 500 zł
  • konferencje i inne wyjazdy | conference travels: 44 400 zł
  • eksperymenty | experiments: 24 00 zł
  • koszty pośrednie dla FAME | overheads for FAME: 90 000 zł
Purpose: 

>>Research project objectives/ research hypothesis

The goal of the project is to identify the principles governing strategic interactions among players who directly derive positive or negative utility from the welfare of others. In particular, we will investigate the strategic incentives among empathetic players and contrast them with those among the egoistic players (including egoistic agents with altruistic motive and with preference for reciprocity). We will derive normative implications, useful in designing economic mechanisms that involve empathetic agents as well as provide testable implications of empathy.

>>Research project methodology

The traditional game theoretic framework does not take into account more complex (and in many contexts: more natural) human preferences such as empathy or altruism. We extend simple framework to allow players to care for utilities of others. We provide a solution concept and characterize an equilibrium set. Testable implications of empathy will be derived for particular games, and contrasted with those from the model having selfish and paternalistic agents. Theoretical solutions will be tested in laboratory experiments.

>>Expected impact of the research project on the development of science, civilization and society

We plan three research papers. In the first paper a non-cooperative framework of the game with interdependent preferences of agents will be introduced. The payoffs within the game will depend not only on strategy profile (as in traditional framework), but also on preferences and welfare of other players. The dependency will be either positive (which will represent sympathy of a player) or negative (which will represent antipathy). In this paper we will provide a novel enrichment of Nash equilibrium structure by endowing players with deterministic beliefs on other players’ utilities. In the second paper, the problem will be considered from the perspective of a game designer who wants to provide an efficient outcome of an empathetic game. We will look at the relationship between business cycle and workers’ empathy towards employers. We will demonstrate that such setup provides micro foundations for a puzzling result in macroeconomics that changes in workers situation (wages and employment) vary less than productivity. Finally, the third paper will consist of set of laboratory experiments which will test the prevalence and characteristics of empathic games when compared to the conventional alternatives. The framework of empathetic games is built on a standard game theory, but assumption regarding preferences are derived from social science such as psychology and behavioral economics, which makes the model interdisciplinary tool for analyzing interactions between agents. Last paper will provide a proof for the usefulness of the new model and comparison between a novel approach proposed by us and traditional game theory approach.

Opublikowane | Published

  • Co-worker altruism and unemployment | Games and Economic Behavior

    It is well-known that social relationships and altruism among workers foster cooperation in the workplace and, therefore, may have beneficial effects for firms. Yet it is unclear how and to what extent co-worker altruism impacts labor market outcomes. In this paper, we find that, although co-worker altruism may be seamless in good times, it may impact the functioning of labor markets during bad times. Specifically, co-worker altruism may potentially lead to wage rigidity and involuntary unemployment in economic downturns. These results seem to be consistent with recent empirical findings.

    Marek
    Weretka
    Jorge
    Vasquez
  • Affective empathy in non-cooperative games | Games and Economic Behavior

    According to psychology, affective empathy is one of the key processes governing human interactions. It refers to the automatic transmission and diffusion of emotions in response to others' emotions, which gives rise to emotional contagion. Contrary to other forms of empathy, affective empathy has received little attention in economics. In this paper, we augment the standard game-theoretic framework by allowing players to affectively empathize. Players' utility functions depend not only on the strategy prole being played, but also on the realized utilities of other players. Thus, players' realized utilities are interdependent, capturing emotional contagion. We offer a solution concept for these empathetic games and show that the set of equilibria is non-empty and, generically, finite. Motivated by psychological evidence, we analyze sympathetic and antipathetic games. In the former, players' utilities increase in others' realized utilities, capturing unconditional friendship; whereas in the latter the opposite holds, resembling hostility.

    Marek
    Weretka
    Jorge
    Vasquez

W toku | Work in progress

  • Empathy in risky choices regarding others

    One of the assumptions of the affective empathy theory is that individuals have consistent preferences over outcomes. However, there is empirical evidence showing violation of such assumption. In particular, in Asian disease experiment subjects are more likely to choose risky over safe rescue plan (with the same expected outcome) under loss than under gain framing. In this study, we induce empathy in the Asian disease study to test whether providing sympathetic or antipathetic relationship between decision-maker and the others affect the size of the framing gap (GvL gap). We find that inducing affection leads to reduction of the gap.

    Magdalena
    Smyk-Szymańska
    Erita
    Narhetali
    Marek
    Weretka
  • Contracts with interdependent preferences

    This paper studies contracting between a principal and multiple agents, as in Lazear and Rosen (1981) and Green and Stokey (1983). The setup is classical except for the assumption that agents have interdependent preferences. We characterize cost effective contracts, and relate the direction of co-movement in rewards --  "joint liability" (positive) or "tournaments" (negative) --  to the assumed structure of preference interdependence. We also study the implications of preference interdependence for the principal's payoffs. We identify two asymmetries. First, the optimal contract leans towards joint liability rather than tournaments, especially in larger teams, in a sense made precise in the paper. Second, when the mechanism-design problem is augmented by robustness constraints designed to eliminate multiple equilibria, for small teams the principal may prefer teams linked via adversarial rather than altruistic preferences.

    Marek
    Weretka