Misprediction of affective outcomes due to different evaluation modes: Replication and extension of two distinction bias experiments by Hsee and Zhang (2004)

Abstract Hsee and Zhang (2004) argued that when people face a decision and must predict future affective states, they are often in a joint evaluation (JE) mode where direct comparisons between different choice options are relatively easy. When actually experiencing (or predicting affective states for) only one of these options, people are usually in single evaluation (SE) mode where direct comparisons are more difficult. This situational difference in evaluation mode was observed to lead to overpredictions of positive affect (happiness) when people were in JE in comparison to SE for options that were quantitatively different, but there were no overpredictions for options that were qualitatively different. This effect was coined distinction bias. In the present paper, we replicated (and extended on) Studies 1 and 2 from Hsee and Zhang with 824 MTurk participants. In Study 1 we replicated the original findings: Relative to people in SE, people in JE overpredicted the happiness derived from quantitatively different hypothetical scenarios (i.e., selling 80 vs 160 books, or 160 vs 240 books), but did not overpredict the difference between qualitatively different hypothetical scenarios (i.e., selling 0 books vs 80 books; 0 being the implicit reference point that makes the two scenarios qualitatively different). Study 2 failed to find support for the original findings: People in JE did not consistently overpredict happiness for quantitatively different scenarios (i.e., copy-pasting 25 negative words vs 10 negative words, or 10 positive words vs 25 positive words). Taken together, the present paper provides mixed support for distinction bias.