It's good and it's bad: Sampling processes can generate decision-framing effects.

Abstract

Framing a choice in terms of gains versus losses can have a dramatic impact on peoples' decisions, sometimes completely reversing their choices. This decision-framing effect is often assumed to stem from individuals' inherent motivational biases to react more strongly to negative information. However, more recent work suggests these decision biases can also stem from biases in the information samples based on which people make their decisions. Here, we test how biases in the frequency of information people have about each decision option can produce decision-framing effects via a sampling process. Specifically, we hypothesize that a gain versus loss framing determines whether people probe their memories for positive or negative information about each decision option. This can lead to inaccuracy if there are biases in the amount of information people have about each option. That is, when people have more information about one option, it is possible that they retrieve both more positive and more negative information about it, creating a bias to select it as being both more and less likely to result in success, depending on decision framing. Three experiments show that people's decisions are more accurate with a gain (vs. loss) framing when a high (vs. low) frequency option has a higher proportion of success; but decisions are less accurate with a gain (vs. loss) framing when the high frequency option has a lower proportion of success. The current results suggest that decision-framing effects do not necessarily indicate a motivational bias. Instead, sampling processes may underlie decision-framing effects. (PsycInfo Database Record (c) 2022 APA, all rights reserved).