The subadditivity effect is a cognitive bias that occurs when people estimate the probability of the combination of two or more uncertain events to be less than the probability of the exclusive components.
--------
Example from Hilbert (2012) page 28.
For example, Redelmeier, Koehler, Liberman, & Tversky (1995) asked physicians to provide probabilities for the likelihood of four exclusive and exhaustive survival chances of patients: dies during hospitalization, dies within a year after release, lives between 1-10 years, lives more than 10 years. Subjects were confronted with each of those alternatives, making four separate judgments. The sum of the likelihood estimates should add up to 100 %, but was instead equal to 164 % (Redelmeier et al., 1995). The decomposed estimate is therefore said to be subadditive with 100/164 = 0.61. In a similar exercise, Witteman, Renooij, and Koele (2007) detected that the subadditivity effect increases with the level of unpacking: the decomposition into three alternatives led to a sum of 120 % on average, while the decomposition into six alternatives led to a sum of 180 % on average (see also Fiedler et al., 2009). Additionally, Fiedler et al. showed that the degree of subadditivity depends on the extremity of the input evidence.
As a cognitive bias, the subadditivity effect falls in the cognitive domain of functioning (SCOPES model).
Reference
Hilbert, M. (2012). Toward a synthesis of cognitive biases: How noisy information processing can bias human decision making. Psychological Bulletin, 138(2), 211–237. https://doi.org/10.1037/a0025940
Geoffrey W. Sutton, PhD is Emeritus Professor of Psychology. He retired from a clinical practice and was credentialed in clinical neuropsychology and psychopharmacology. His website is www.suttong.com
See Geoffrey Sutton’s books on AMAZON or GOOGLE STORE
Follow on FACEBOOK Geoff W. Sutton
TWITTER @Geoff.W.Sutton
You can read many published articles at no charge:
Academia Geoff W Sutton ResearchGate Geoffrey W Sutton
Dr. Sutton’s posts are for educational purposes only. See a licensed mental health provider for diagnoses, treatment, and consultation.
No comments:
Post a Comment