Kosfeld et al (2005)

Procedure

- 128 healthy male students (mean age 22) randomly allocated into either the oxytocin or placebo group - Substances were administered via an intra-nasal spray - The researchers designed a trust game with real monetary stakes -- Subjects were paired anonymously and played the role of either an investor or a trustee - Step 1: experimenter gives both the investor and the trustee an endowment of 12 monetary units - Step 2: the investor needs to decide how much of that to send to the trustee (there are 4 options: 0, 4, 8, 12). The experimenter triples whatever is sent to the trustee. The trustees have their own endowment, which is added to their total. - Step 3: the trustee decides how much of the now available money to send back to the investor. The idea is that if the investor completely trusts the trustee, the investor would send 12 that would turn into 26, and the trustee would send back at least 18. The pair only interacts once during the experiment, so there is the temptation to keep the whole sum. - Participants played the game 4 times in the same role, each time paired randomly with a new partner. - At the end of the experiment the total earned monetary units were exchanged for real money.

Conclusions

- Oxytocin specifically affects trust in interpersonal interactions.

Method

- Stratified sampling - Random allocation of participants

Results

- The level of trust in those participants who received a dose of oxytocin was higher than in the control group. The median transfer of investors was 10 in the oxytocin group and 8 in the control group. 45% of the subjects in the oxytocin group showed the maximum trust level (12 monetary units) while 21% in the control group showed the maximum trust level. 2 alternative explanations: -- Oxytocin reduces risk aversion in general -- Oxytocin increases people's trust in other humans - Follow-up study to clarify -- An independent group of subjects played a similar trust game, but instead against a random mechanism (software). The algorithm in the software was modeled after decisions of real people (trustees) in previous experiments → no difference was observed between the oxytocin and placebo groups.

Year of study

2005

Researchers

Kosfeld et al (2005)

Evaluate (method and ethics)

Method: - Ecological validity - can an all male study be generalized to other genders/sexes? - Demand characteristics - were there any expectations/biases due to the all-male sample? Ethics: ?

Title/subject of research

Role of oxytocin in interpersonal trust

Aim

To find evidence to support that oxytocin increases trust in humans