Related Experiment Video
Updated: May 28, 2026

A Prediction Error-driven Retrieval Procedure for Destabilizing and Rewriting Maladaptive Reward Memories in Hazardous Drinkers
Published on: January 5, 2018
Risk dishabituation: in repeated gambling, risk is reduced following low-probability "surprising" events (wins or
Heath A Demaree1, Kevin J Burns, Michael A Dedonno
1Department of Psychological Sciences, Case Western Reserve University, Cleveland, OH, USA. Heath@Case.edu
Abstract:
In path-dependent risk taking, like playing a slot machine, the wager on one trial may be affected by the outcome of the preceding trial. Previous studies have shown that a person's risk-taking preferences may change as a result of the preceding trial (win or loss). For example, the "house money effect" suggests that risk taking may increase after a win, whereas the "break even effect" posits that risk taking increases after a loss. Independent of those findings, a person's emotional state has been found to influence risk taking. For example, the "mood maintenance hypothesis" supports the notion that positive affect decreases risk taking, and related research finds that increased negative affect increases risk taking. Because winning and losing may influence one's emotional state, we sought to investigate how both previous outcomes, as well as a person's emotional responses to those outcomes, independently influence subsequent risk taking. To do this, data were collected using three simplified slot machines where the chance of winning each trial was set to 13%, 50%, and 87%, respectively. Evidence for the break even and house money effects were found on the 13% and 87% games, respectively. Likewise, emotional valence was found to predict risk taking on these two tasks, with emotional valence fully explaining the break even effect observed on the 13% game. In addition to these results, the present research revealed that risk taking is reduced following low-probability ("surprising") events (i.e., a win in the 13% condition or loss in the 87% condition). Dubbed "risk dishabituation," this phenomenon is discussed, along with its likely corresponding emotional experience--surprise.
Related Concept Videos
Instinctive Drift
Unusual Results
According to the range rule of thumb, any value above or below two standard deviations, 2σ from the mean, μ is considered unusual.
Maximum unusual value = μ + 2σ
Minimum unusual value...
Bias
In statistics, a sampling bias is created when a sample is collected from a population, and some members of the population are not as likely to be chosen as others (remember, each member...
Hazard Rate
Hindsight Biases
Regression Toward the Mean

