Abstract: This study introduces and extends recent developments in the studies of framing effects in behavioral decision making, and examines the effects of self-framing on risky choice and its underlying mechanism. Until recently, framing studies had largely focused on how externally framed (phrased) choice problems affected risk preference. Little attention was given to how decision makers mentally frame decision problems themselves. Recent works by Wang and others found significant effects of self-framing of choice outcomes on risk preference of the decision maker. In the present study, we addressed several unanswered questions. Previous studies showed that a large proportion of the participants use both positive and negative frames to encode expected choice outcomes. Would the difference in the hedonic tone of self frames for different choice options (i.e., sure thing vs. a gamble of equal expected value) affect risk preference of the decision maker? We predicted that the positive hedonic tone of self-framing would increase the attractiveness of the framed option. Thus, the option that was more positively framed would be more likely to be chosen. Second, we examined the mechanism of self-faming effects with regard to how risk perception mediates the effects. We predicted that opportunity and threat perception would mediate the effects of self-framing on risky choice.
One hundred and sixteen university students participated in the study. Adopting a within-subject design, each participant was provided with three risky choice problems presented in managerial, health, and investment contexts. After viewing a graphic display (pie chart or histogram) of a decision problem, participants discretionally described expected choice outcomes (self framing) in a sentence completion task. For each choice problem, the participants were asked to rate perceived opportunities and threats associated with each of the two options, either a sure thing or a gamble. The participants were then asked to make a choice between a sure-thing option and its gamble equivalent. Chi-square analysis was conducted to test self-framing effects; logistic regression analysis was used to examine the relationship between self-framing, risk (opportunity and threat) perception and risky choice.
The study found that the hedonic tone of self-framing affected the final choice of the participant: the more positive the hedonic tone of self framing of a choice option (either a sure option or a gamble) was, the more likely the option would be chosen. This effect of self framing was partially mediated by the decision maker’s perception of opportunities and threats.
This study enriched our understanding of framing effects. When decision makers are allowed to encode and frame the expected outcomes of a risky choice problem by themselves, the small difference in the hedonic tone of self-generated frames of alternative options may influence risk perception and risky choice. That is, the hedonic tone of self-framing as an encoding of decision information affects risk (opportunity and threat) perception, which in turn affects the risk preference and choice of the decision maker

Key words: self-framing effect, risk perception, risky choice