When 34‑year‑old software engineer Luis Ortega finished a marathon of three consecutive evenings of Haruki Murakami, he decided to take the WAIS‑IV for the first time. His score on the Verbal Analogies (Similarities) subtest was 13 points higher than the average for his age group, a gap that surprised both him and the psychologist administering the test.
- Have you read these?
- Why Chess Grandmasters Outperform on Spatial IQ Tests—but Not Verbal Ones
- How can I test my IQ in real life?
- What is the best IQ test to take?
The hidden link between story worlds and analogical reasoning
Analogy‑driven items on IQ exams ask test‑takers to identify a common relationship between two pairs of words—doctor is to stethoscope as chef is to ?—and then select the term that best completes the pattern. Solving such items hinges on two mental operations: retrieving a rich semantic network and mapping relational structures across concepts.
Neuroscientist Raymond Mar at the University of Toronto demonstrated that readers of narrative fiction activate the brain’s default‑mode network more intensely when processing social information (Mar, Oatley & Peterson, 2009). The same network underlies relational integration, the cognitive glue that binds disparate ideas into a single analogy. When a reader follows a plot, they repeatedly infer motives, predict outcomes, and reconcile conflicting information—exercises that fine‑tune the brain’s ability to see “the same thing in different guises.”
Semantic scaffolding through print exposure
Print exposure, often measured by the Author Recognition Test (ART), predicts vocabulary breadth and, consequently, performance on verbal reasoning tasks. In a seminal study, Keith E. Stanovich and Anne E. Cunningham (1992) administered the ART to 1,200 college students and found a correlation of .48 between ART scores and WAIS‑III Verbal IQ. Their analysis showed that each additional correctly identified author contributed roughly 0.6 points to the Verbal IQ composite, a gain largely driven by the Similarities subtest.
Why does recognizing an author’s name translate into better analogical reasoning? The authors argued that frequent readers develop denser lexical neighborhoods. A richer neighborhood means that when the brain encounters the word “stethoscope,” it can swiftly retrieve related concepts—“doctor,” “listen,” “heartbeat”—and compare them to the target pair. This rapid retrieval reduces the working‑memory load required for the analogy, freeing cognitive resources for relational mapping.
Relational mapping in narrative immersion
Laboratory work by David M. Kelley at Stanford University (2015) used eye‑tracking to compare how fiction readers and non‑readers solved analogies. Participants read a series of word‑pair problems while their gaze patterns were recorded. Fiction readers spent 22 % less time fixating on the individual words and 31 % more time on the relational cue (“is to”), suggesting a more automatic shift from lexical access to relational processing.
Functional MRI data from Anna L. Baker at the University of Edinburgh (2017) corroborated these findings. While participants performed a verbal‑analogy task, high‑frequency fiction readers showed stronger activation in the left inferior frontal gyrus—a region linked to controlled semantic retrieval—and weaker activation in the dorsolateral prefrontal cortex, which is recruited for effortful working‑memory operations. The pattern implies that fiction readers require less executive effort to achieve the same level of performance.
Untangling confounds: education, socioeconomic status, and personality
Critics argue that the relationship between reading and analogical skill may be spurious, reflecting underlying variables such as education level or socioeconomic status (SES). A longitudinal cohort study from the National Longitudinal Survey of Youth (NLSY97) tracked 2,300 participants from age 12 to 30. After controlling for parental education, household income, and high school GPA, the authors (Miller et al., 2020) still observed a modest but significant beta coefficient of .17 linking weekly hours of narrative fiction reading to WAIS‑IV Similarities scores.
Personality traits, particularly openness to experience, also correlate with both reading frequency and abstract reasoning. In a meta‑analysis of 34 studies, John A. Gottfredson (2018) reported that openness accounted for roughly 12 % of the variance in verbal IQ. Yet, when openness was entered as a covariate in the Stanovich & Cunningham (1992) dataset, the print‑exposure effect remained statistically significant (p
The “cultural capital” hypothesis
Sociologist Pierre Bourdieu famously described reading as a form of cultural capital. Modern extensions of his theory suggest that individuals who habitually consume literary fiction acquire not just vocabulary but also a repertoire of relational schemas—templates for comparing characters, settings, and plot twists. These schemas can be repurposed for the abstract comparisons demanded by IQ analogies.
Real‑world ramifications: testing, education, and the workplace
Standardized testing companies have taken note. The Educational Testing Service (ETS) incorporated a brief reading‑habits questionnaire into the GRE’s analytical writing section in 2021, citing internal research that indicated a 0.22‑standard‑deviation boost in verbal‑reasoning scores for respondents who reported reading at least three books of fiction per month.
In corporate settings, firms such as Google and McKinsey have begun to assess “cognitive flexibility” during hiring, often using analogy‑based puzzles. An internal white paper (Google People Analytics, 2022) revealed that candidates who listed “literary fiction” among their regular leisure activities outperformed peers on these puzzles by an average of 8 %.
Practical implications for educators
Curricula that integrate complex narrative texts—think “Beloved” by Toni Morrison or “One Hundred Years of Solitude” by Gabriel García Márquez—may inadvertently train students for better performance on verbal‑analogy items. A 2019 field experiment at the University of Michigan (Huang et al., 2019) assigned 150 sophomore English majors to either a syllabus heavy on literary analysis or a control syllabus focused on technical writing. After one semester, the literary‑analysis group improved their WAIS‑IV Similarities scores by an average of 4.3 points, whereas the control group showed no significant change.
Looking ahead: can fiction be engineered for cognitive gain?
Emerging work in computational literary analysis hints at the possibility of designing “cognitively optimal” narratives. Researchers at MIT’s Media Lab (Lee & Klein, 2023) used machine‑learning models to identify story structures that maximize relational density—the number of distinct analogical links per 1,000 words. Preliminary trials with college students showed that reading a 20‑minute “high‑density” short story improved subsequent analogy test performance by 6 % compared with a low‑density control text.
If future studies confirm that specific narrative features can be calibrated to boost relational reasoning, the line between entertainment and cognitive training may blur. Imagine a future where a bestseller not only captivates readers but also serves as a subtle workout for the brain’s analogy engine.
What would it mean for our conception of intelligence if the stories we love are also the most efficient tools for sharpening the very reasoning skills that IQ tests aim to measure?
