Decision making under stress: The “hiding behind a small cake” effect

Description

These data report the results of experimental research that investigated the "hide behind the small cake effect". In this study, we implemented a laboratory experiment using the ultimatum game with asymmetric information, in which only the first mover was fully informed about the size of the endowment. Our experiment compared the decisions of subjects who received a stress stimulus from an adapted Trier Social Stress Test for Groups (TSST-G) protocol with a control group. We used the Tobit model in the regression analysis because we observed a sizable number of transfers equal to the smallest possible transfer (R$0) and the largest possible transfer (R$30). A Logit model was used when the dependent variable was a dummy for the proposed categories. We found that under stress, proposers transferred more to responders. In contrast, non-stressed players were more likely to take advantage of information asymmetry by choosing lower strategic offers, which is consistent with the so-called “hide behind a small cake” effect. Regression analysis also indicated that larger offers are not necessarily associated with increased prosocial motives since stressed proposers became more confident about the responder´s ability to guess the true endowment, which decreased their incentive to take advantage of the aforementioned effect.

Steps to reproduce

1. Collect salivary cortisol after stress stimulus: Salivary cortisol can be easily assessed using the Salivette® Cortisol device, a 10 cm long polypropylene tube with a cotton swab. To collect the sample, the cotton swab is placed in the individual’s mouth for about 2 to 3 minutes until it is soaked with saliva, and it is then put back in the tube and closed. The electrochemiluminescence method was used to determine the salivary cortisol level of each participant. 2. Conduct the experimental game with assimetric information between players in order to evaluate the hide behind the small cake effect. The procedures will happen with the control and the treatment group; 3. With the endowments obtained in the game plus the cortisol salivary results and control variables gathered with participants (age, gender, and income), reproduce models using Tobit and Logit regressions. 4. Interaction variables were also added in the regressions: a) Strategic-selfish offers: A value of 1 when offers were characterized by splits where offers were lower than 50% of the endowment in the game (when the proposer retained more than the responder) and zero otherwise. b) Social preference offers: A value of 1 when transfers were higher than 50% of the true endowment and zero otherwise. c) Fair offers: A value of 1 when transfers were equal to 50% of the true endowment in the game and zero otherwise.

Institutions

Categories

Funding

Licence