Universität zu Köln

Mensching, Johannes Ole (2011). Can we trust trust explanations?An experimental illustration of how outcome based accounts of trust struggle to explain a basic phenomenon of human life. PhD thesis, Universität zu Köln.

Preview

PDF KUPPDFOleMensching.pdf Bereitstellung unter der CC-Lizenz: Creative Commons Attribution Non-commercial No Derivatives. Download (1MB)

Abstract

In Chapter 4, I will provide an overview of the findings and insights gained in experimental trust research. I will then discuss common consequential explanations and potential methodological reasons for high trust rates. Finally, I will present an empirical study in which all these explanations were tested using a between-subjects design. In this study, my coauthors and I could not find out why people trust, but we showed which explanations fail to explain trust. First, we showed that trust rates did not decrease when participants had to hand over their own money, compared to past studies in which participants played with a show-up fee (house money effects). For the first time, we conducted a trust game in which trustors could only hand over their own money; however, similar to what occurred in past studies, 57.1% of trustors did so. Second, anticipating high rates of trustworthiness was not the reason for participants to trust. In fact, they underestimated this share by almost 20 percentage points. Third, high trust rates were not explainable by high risk tolerance. Only 25.3% of participants were willing to bet money, making a risky decision identical in gains and similar in risks to the trust game but void of trust. However, trust games entail second players who can benefit from the money trustors hand over. Therefore, trustors might hand over money because they have preferences for equality or want to enlarge the pie. Contradicting these arguments, only 28.3% of participants bet money in a risky decision, which was not only similar to the trust game regarding gains and risks but also those involving a second player. Thus, we could not find evidence that trust can be explained by one of the explanations we tested. Chapter 5 deals with the question of how people behave in trust situations in which they cannot be better off if their trust is reciprocated by the partner with whom they interact (trustee). Former studies have examined only trust decisions in which people who trust (trustors) were rewarded if the person with whom they interacted proved to be trustworthy. Thus, it is not clear to what extent the strategic motive to be better off plays a role in trust decisions in general. In Chapter 5, I present a study on that issue. Using a full between-subjects design, this study examined non-strategic trust decisions in which participants cannot be better off or even only worse off by trusting others for the first time. To do this, my coauthors and I systematically manipulated the potential gains in trust games and compared them with lotteries void of any trust but equal in risks and gains. Our results show that both trust behavior and risky behavior unrelated to trust were dependent on potential gains and losses. However, whereas the number of risky decisions decreased to almost zero in lotteries entailing no or negative gains, trust behavior was comparably stable. Chapter 5 provides evidence that people do not trust strategically and that high trust rates are sustainable in different kinds of trust situations. While I considered only explanations for trust behavior that are basically consequential in Chapters 4 and 5, I go one step further in Chapter 6. Here, I examine whether trust behavior in trust games is driven by the curiosity trustors feel in the moment they make their trust decision. Thereby, I illuminate the question whether the paradigm of the trust game itself causes the phenomenon of high trust rates. In addition, I examine the influence of regret aversion (the tendency of people to avoid future regret) on trust. In order to give an answer to this question, I present a study in Chapter 6 in which my coauthors and I compared a trust game with conditional feedback to a trust game with unconditional feedback (between-subjects). Trustors in ordinary trust games receive conditional feedback only. That means they learn the trustworthiness of their trustee on the condition that they hand over their money to him or her. Thus, trustors might hand over money in trust games because they want to know whether their trustee is reliable. To find out, we compared an ordinary trust game with conditional feedback to a trust game with unconditional feedback in which trustors always learned the trustworthiness of their interaction partner. In this trust game, trustors were always informed about the decision of their assigned trustee, no matter whether they kept or handed over their money. Whereas the curiosity hypothesis predicts that more trustors hand over money when they receive conditional feedback, regret aversion would predict that fewer people will do so. In the trust game with unconditional feedback, trustors cannot avoid potential regret triggered by their decision to keep or to hand over money. Trustors, who keep the money, run the risk of learning that their trustee was reliable and that they would have doubled their money had they handed it over. Trustors, who hand over their money, run the risk of learning that their trustee was untrustworthy and that they would not have lost their money had they kept it. Thus, regret aversion should not influence trustors to make a particular decision in the trust game with unconditional feedback. However, in the trust game with conditional feedback, trustors can avoid potential regret by keeping the money because then they will never learn whether their assigned trustee was reliable and that they would have doubled their money, if they had handed it over. Therefore, regret aversion should influence trustors in the trust game with conditional feedback to keep their money. However, we did not find any difference between the trust rates in the trust game with conditional feedback and the trust game with unconditional feedback. Hence, neither the curiosity nor the regret hypothesis could be supported. At the end of Chapter 6, I discuss which implications these findings could have for further research.

Item Type:	Thesis (PhD thesis)
Creators:	Creators Email ORCID ORCID Put Code Mensching, Johannes Ole ole.mensching@uni-koeln.de UNSPECIFIED UNSPECIFIED
URN:	urn:nbn:de:hbz:38-42896
Date:	5 July 2011
Language:	English
Faculty:	Faculty of Management, Economy and Social Sciences
Divisions:	Faculty of Management, Economics and Social Sciences > Social Sciences > Sociology and Social Psychology > Department of Economic and Social Psychology
Subjects:	Psychology Social sciences Economics
Uncontrolled Keywords:	Keywords Language Trust explanations, homo oeconomicus, house money, rationality, curiosity, risk preferences UNSPECIFIED
Date of oral exam:	21 June 2011
Referee:	Name Academic Title Fetchenhauer, Detlef Prof. Dr. Willer, Robb Prof. Ph. D.
Refereed:	Yes
URI:	http://kups.ub.uni-koeln.de/id/eprint/4289