Do we prefer preventing mistakes or fixing them?

Our everyday lives are filled with situations in which good behaviour is rewarded and bad behaviour punished. There is always scope for mistakes, though — in every criminal justice system, some innocent people are unjustly punished (these are termed 'false positives'), while some genuine criminals walk free ('false negatives').

A similar situation holds for rewards: a business with a bonus scheme might give bonuses to some employees who don't deserve one (false positives), or fail to give bonuses to employees who do (false negatives).

So, which type of mistake do we find less tolerable — creating false negatives, or false positives? A series of studies in Psychological Science by Eitan D. Rude and Franklin Shaddy at the UCLA on almost 3,500 US-based adults finds some consistent biases in attitudes towards this question. The work also suggests that these biases could influence our opinions about a wide range of real-world situations.

In the pair's initial studies, participants were presented with a punishment or reward scenario, and asked how relatively important they thought it was prevent false positives, false negatives, or to fix either type of mistake after it had been made. One scenario involved pay schemes that either docked the pay of poorly performing salespeople or gave bonuses to high performers. Another described an insurance company that either reduced the monthly premium payments of the safest drivers or increased premiums for those deemed to be the most reckless. The third focused on fines for water users who failed to stick to a water allotment, and credits for those who did.

Some of the participants were told that these rewards or punishments were planned — so they were being asked about mistakes that could be prevented. Others were told that they had already happened, meaning that any false negatives or positives would need to be fixed.

The team's analysis of the participants' preferences revealed two consistent patterns across all three scenarios. For punishments, participants believed it was more important to avoid false negatives (such as failing to dock the pay of lazy workers) than to fix them, but more important to fix false positives (docking the pay of workers who didn't deserve it, for example) than to prevent them.

For rewards, however, the opposite was true. Participants felt it was more important to prevent false positives (to avoid giving bonuses to poorly performing workers, for example) than to fix them, and more important to fix false negatives (failing to give a bonus to workers who deserved one, for example) than to prevent them.

This quite complex pattern of results can be summed up in this way, the researchers write: "… people maintain stronger preferences for fixing false-positive punishments than for preventing them, and stronger preferences for fixing false-negative rewards than for preventing them.

"The overall effect is the endorsement of harsher policies in principle (before errors occur, when people focus more on preventing mistakes that help bad actors) than in practice (after errors occur, when people focus relatively more on fixing mistakes that harm good actors)."

In a further study, they found some evidence for why this might be: when the punishments or rewards in the scenario were described as having already happened, rather than pending, participants felt that the good salespeople who had undeservingly been punished seemed more vivid and real to them than poor salespeople who had escaped punishment. The researchers believe this could be because the participants more readily put themselves in the shoes of the 'good but unfairly treated' individual, and therefore punishments or rewards that had come to pass and required fixing seemed more concrete and real.

This work "offers a novel theoretical backdrop for understanding many real-world policy debates," the team writes.

In the real world, it seems likely, of course, that our biases may shift a little, depending on the scenario. For example, we may feel that it's relatively more important to prevent rather than fix wrongful murder convictions than to prevent rather than fix the mistaken docking of the pay of a well-performing employee.

But the direction of the biases found in this all-American sample were consistent. These results may give us hints as to why particular policies find higher public support. As the researchers note, "Our findings…offer critical insights for policymakers, managers, and marketers (among others)."

Read the paper in full:
Rude, E. D., & Shaddy, F. (2024). People Endorse Harsher Policies in Principle Than in Practice: Asymmetric Beliefs About Which Errors to Prevent Versus Fix. Psychological Science, 35(5), 529-542. https://doi.org/10.1177/09567976241228504