Responsibility is Divisible by Two, But Not by Three or Four: Judgments of Responsibility in Dyads and Groups

Abstract

When two individuals are doing a joint task, most people seem to think that the responsibility should be divided proportionally between them in a complementary fashion, so that an increase in one actor's responsibility leads to a corresponding decrease in the responsibility of the second actor. however, with three or four actors, the individual responsibilities of the first two are not reduced. As a consequence, the sum of responsibility assessments exceeds 100%, and a change in one actor's responsibility does not have to affect the perceived responsibility of the others, indicating a singular (case-based), rather than a distributional (class-based) view. The shift from additive to nonadditive responses is demonstrated in four vignette studies, using different response formats, and for actions involving causal as well as moral responsibility. The pattern of results is compatible with the singularity principle in judgment, according to which only one hypothesis, or one comparison, can be considered at a time.1. Cross-country skiing combined with target shooting.

Acting knowingly: effects of the agent's awareness of an opportunity on causal attributions

“…Second, we considered the possibility that in the awareness condition, people will be more likely to interpret responsibility or blame in a moral sense. If so, we may expect them to apportion responsibility or blame between two causes (Teigen & Brun, 2011), and these scores will therefore sum to 100%. Consistent with this view, we found that the mean sum of proximal and distal scores for responsibility judgements (M D 119%) and blame (M D 111%) judgements was significantly closer to 100% than causal importance judgements (129%, t D 2.71, p < .01 and t D 4.56, p < .001, respectively).…”

Causal superseding

“…Then, if the norm-violation of one agent"s action increases that agent"s causality, it follows under this intuition that some other factor"s causality will have to be reduced. Though this explanation might at first seem compelling, there is already empirical evidence that causal responsibility is not generally a zero-sum judgment (Kominsky, Phillips, Gerstenberg, Lagnado, & Knobe, 2014;Lagnado, Gerstenberg & Zultan, 2013;Teigen & Brun, 2011). For example, when an outcome was brought about by a collection of causes that were each individually necessary for its coming about, then each cause was judged as fully responsible (Lagnado, Gerstenberg & Zultan, 2013;Zultan, Gerstenberg & Lagnado, 2012).…”

Additivity neglect in probability estimates: Effects of numeracy and response format

“…One explanation for additivity neglect has therefore been that people either fail to use their mathematical skills when making probabilistic judgments of uncertain events, or that they altogether lack the mathematical skills necessary to give normative responses (Robinson & Hastie, 1985;Teigen, 1983). Another explanation is that the bias occurs due to a case-based approach to the probability judgments, where people consider one alternative at the time, thus failing to take the complementarity of the alternatives into account (Sanbonmatsu, Posavac, & Stasney, 1997;Teigen & Brun, 2011).…”

“…Several studies have suggested that people can adopt two different approaches to probability judgments, namely a case-based or class-based approach (Fox & Rottenstreich, 2003;Fox & Ülkümen, 2011;Kahneman & Tversky, 1982;Reeves & Lockhart, 1993;Teigen & Brun, 2011). Within psychology, the distinction originates from Kahneman and Tversky (1982), who distinguish between singular versus distributional approaches (Kahneman & Tversky, 1982;Klar, Medding, & Sarel, 1996;Reeves & Lockhart, 1993).…”