Reading Literary Fiction and Theory of Mind: Three Preregistered Replications and Extensions of Kidd and Castano (2013)

Abstract

Scholars from diverse disciplines have proposed that reading fiction improves intersubjective capacities. Experiments have yielded mixed evidence that reading literary fiction improves performance on the Reading the Mind in the Eyes Test, a test of Theory of Mind. Three preregistered experiments revealed mixed results. Applying the “small telescopes” method developed by Simonsohn revealed two uninformative failures to replicate and one successful replication. On a measure of the importance of intentions to moral judgments, results were more mixed, with one significant effect in the expected direction, one nonsignificant effect, and one significant effect in the unexpected direction. In addition, two experiments yielded support for the exploratory but preregistered hypothesis that characters in popular fiction are perceived as more predictable and stereotypic than those in literary fiction. These findings help clarify the sociocognitive effects of reading literary fiction and refine questions for future research.

Dutch Author Recognition Test

“…There may be two reasons why introverts are not more likely to read than extraverts. First, as we saw above, it has been argued that fiction reading helps understanding others, which may interest extraverts as much as introverts (Kidd & Castano, 2019;Mar et al, 2006Mar et al, , 2009. Second, there are different types of books.…”

“…The latter has sparked much controversy, because its replicability was called into question. As a result, there have been several preregistered, high-power replication studies, including some by the original authors themselves (Kidd & Castano, 2019;Panero et al, 2016;Samur, Tops, & Koole, 2018;van Kuijk et al, 2018). Although the results have not been highly supportive for the claim that reading a few short literary texts increases scores on ToM tests, each and every study replicated the positive correlation between ART scores and performance on ToM tests (see also Mar, Oatley, Hirsh, dela Paz, & Peterson, 2006;Mar, Oatley, & Peterson, 2009, for earlier evidence relating the reading of fiction to abilities of empathy).…”

Those who ignore the past are doomed…to be heartless: Lay historicist theory is associated with humane responses to the struggles and transgressions of others

“…Another type of information exposure that should be explored is exposure to books and to visual media. There is evidence that high-quality fiction can improve one’s ability to understand another’s thoughts and feelings [ 61 ]. What types of reading might increase development of historicist thinking?…”

Cognitive Empathy as Imagination: Evidence From Reading the Mind in the Eyes in Autism and Schizotypy

“…RMET performance enhancements provide especially useful information about this test because their causes are probably not confounded with sex-related or disorder-related cognitive deficits or reductions in ability. Such enhancements have been found in an intriguing suite of studies, with higher RMET scores being reported in: (a) women with an anxious attachment style ( 25 , 26 ), a condition that is itself female biased ( 27 ); (b) women, but not men, with higher levels of social anxiety ( 28 ); (c) women with past major depression, dysphoria, or a maternal history of depression, though not with clinical depression ( 10 , 21 , 29 – 31 ); (d) women with anorexia nervosa, for emotional RMET cues but not overall ( 32 ); (e) women with borderline personality disorder, for emotional RMET cues, or overall, and non-clinical women high in borderline traits (for negative cues only) ( 17 , 33 – 35 ); (f) typical males and females who read more literary fiction ( 36 ); (g) typical males and females who exhibit higher mindfulness or undergo mindfulness training prior to testing ( 37 – 40 ); and (h) typical males and females who have been administered oxytocin (better scores), MDMA (better scores), or testosterone [worse scores, contingent upon their 24D digit ratios ( 41 – 43 )].…”