Kosfeld et al (2005)

Uses: Hormone

Aim: to investigate the effect of oxytocin on decision making.

 

Method: True experiment, double blind, placebo, control

 

Procedure:

194 healthy male students participate in the study with 128 participated in the trust experiment and 66 in the risk condition (computer). They are asked to play the trust game where they are given 12 MU and asked to give either 0, 4, 8, or 12 to the trustee. The amount they gave is tripled by the experimenter and the trustee can decide how much money they will give back to the participant. The participants are split into experimental (oxytocin nasal spray) and placebo groups and they do this entire process 4 times with different trustees. Risk condition is the same game but against a computer.

They had informed consent and were told that it is an investigation of hormone on decision making. The MU could also be directly redeemed as real life cash once the experiment was over, so the more MU they had, the more they get paid.

 

Results:

45% of the participants in the oxytocin group showed maximum trust level (12MU) and only 21% in the placebo group showed maximum trust. Only 21% of the oxytocin group showed low trust (below 8MU) but 45% of the control group showed low trust. In the risk experiment there were little to no difference in each of the group, the mean transfer being 7.5 MU.

 

Conclusion:

Oxytocin increases the trusting behaviour of investors. In the risk condition, the investors behaviour did not change between oxytocin and placebo. Kosfeld et al (2005) showed that oxytocin plays a major role in trust behaviour and decision making.

Evaluation:

  • All young male, no diversity (sampling bias) (oxytocin may affect men and women differently) (oxytocin released during childbirth)

  • Lacks ecological validity, complicated, does not reflect true investment, a poor attempt at replicate game theory.

  • Can not operationalised complex social human behaviour.

  • Reductionist

  • Trust is a hard concept to test

  • Double blind, placebo, control (highly controlled, lab study, can be replicated)

  • Risk condition knowing it's a computer

  • Large sample size

Kosfeld, M., Heinrichs, M., Zak, P. et al. Oxytocin increases trust in humans. Nature 435, 673–676 (2005). https://doi.org/10.1038/nature03701