Ego Depletion Increases Risk-Taking

ABSTRACT

We investigated how the availability of self-control resources affects risk-taking inclinations and behaviors. We proposed that risk-taking often occurs from suboptimal decision processes and heuristic information processing (e.g., when a smoker suppresses or neglects information about the health risks of smoking). Research revealed that depleted self-regulation resources are associated with reduced intellectual performance and reduced abilities to regulate spontaneous and automatic responses (e.g., control aggressive responses in the face of frustration). The present studies transferred these ideas to the area of risk-taking. We propose that risk-taking is increased when individuals find themselves in a state of reduced cognitive self-control resources (ego-depletion). Four studies supported these ideas. In Study 1, ego-depleted participants reported higher levels of sensation seeking than non-depleted participants. In Study 2, ego-depleted participants showed higher levels of risk-tolerance in critical road traffic situations than non-depleted participants. In Study 3, we ruled out two alternative explanations for these results: neither cognitive load nor feelings of anger mediated the effect of ego-depletion on risk-taking. Finally, Study 4 clarified the underlying psychological process: ego-depleted participants feel more cognitively exhausted than non-depleted participants and thus are more willing to take risks. Discussion focuses on the theoretical and practical implications of these findings.

Keywords:

• ego depletion
• risk-taking
• self-regulation
• self-control
• sensation seeking

Notes

1. Both low depletion control conditions (zoo: N = 9; pictures: N = 8) did not significantly differ regarding their effect on sensation seeking (zoo: M = 4.19; pictures: M = 4.31). However, they both significantly differed from the high depletion condition (M = 5.05; LSD; zoo: p = .016; pictures: p = .043). Thus, both control conditions were collapsed for the main analyses.

2. We would like to thank an anonymous reviewer of a previous version of this manuscript for this important point.

3. The mean of the whole sample was M = 3.12 (SD = 0.67); the values of the 2 outliers were 5.21 (low depletion condition; 3s from the mean; highest value in the whole participant sample) and 1.10 (high depletion condition; 3s from the mean; lowest value in the whole participant sample). If those 2 outliers were included in the analysis no significant main effect for the experimental condition factor can be observed, F(2,71) = 0.60, p = .55. Without those outliers the omnibus analysis is marginally significant (see below), F(2,69) = 2.49, p = .09.

4. For both ego-depletion manipulations, the effect of low vs. high depletion on risk-taking points into the same direction: Emotion control task: low depletion (M = 2.95, SD = 0.54); high depletion (M = 3.37, SD = 0.77); cognitive control task: low depletion (M = 2.99, SD = 0.60); high depletion (M = 3.26, SD = 0.54).

5. Following the analytic strategy recommended by Rosenthal and Rosnow (1989), follow-up a priori contrasts revealed that participants in the high depletion condition reported higher levels of risk-taking (M = 3.31, SD = 0.66; contrast weight: 2) than participants in the non-depletion (M = 2.97, SD = 0.56; contrast weight: −1) and cognitive load condition (M = 3.07, SD = 0.44; contrast weight: −1), t(69) = −2.03, p < .05.

6. We would like to thank an anonymous reviewer of a previous version of this paper for this important point.