Summary: Researchers report declines in verbal reasoning, matrix reasoning and numerical series scores in a large U.S. sample from 2006–2018, while spatial reasoning (3D rotation) showed gains for the years available (2011–2018).
Source: Northwestern University
The Flynn effect historically describes steady rises in measured IQ across the 20th century—often three to five IQ points per decade. New analysis of a large online U.S. sample suggests a partial reversal of that trend between 2006 and 2018 for most cognitive domains studied.
Researchers from Northwestern University examined cognitive ability scores collected through the Synthetic Aperture Personality Assessment (SAPA) Project and found consistent downward trends in three of four cognitive domains assessed. Scores for verbal reasoning (vocabulary and logic), matrix reasoning (visual problem solving and analogies), and letter-and-number series (computational and mathematical reasoning) declined across the study period. In contrast, scores for three-dimensional rotation, a measure of spatial reasoning, generally increased for the years available (2011–2018).
Overall composite ability scores—derived from multiple test items—were also lower for more recent samples. These declines remained visible when the data were stratified by age, education level and gender, indicating the trends were not confined to a single demographic group.
Elizabeth Dworak, a research assistant professor of medical social sciences at Northwestern University Feinberg School of Medicine and a corresponding author, emphasized that lower scores do not necessarily mean people are less intelligent. “These findings reflect changes in test performance across samples, not a simple statement that mental ability has fallen,” she said. “It could reflect changes in test-taking behavior, motivation, familiarity with these formats, or other factors that influence performance on these measures.”
The study, published in the journal Intelligence, used the SAPA Project, a free online survey originally designed to provide feedback on personality and temperament traits. SAPA also includes items from the International Cognitive Ability Resource (ICAR), an open-source multiple-choice assessment. The researchers analyzed responses from 394,378 U.S. adults collected between 2006 and 2018 to assess whether cognitive ability scores shifted over that 13-year window. A focused subset of 303,540 participants, recruited between 2011 and 2018, provided the data set that included 3D rotation items, which were added later.
Why might some scores decline while spatial ability rises? The study itself did not establish causal explanations, but Dworak noted several hypotheses that warrant further investigation: changes in nutrition or overall health, evolving media environments and digital exposure, shifts in educational emphasis, and altered social values that change what skills are reinforced in daily life. For example, increased focus on STEM in education could boost spatial and technical skills while attention to other forms of abstract reasoning might decline.
Another practical explanation is motivation and test engagement. Because the SAPA Project is presented primarily as a personality survey, some test-takers may be more invested in temperament items and less engaged with the cognitive sections, producing lower measured ability in specific tasks. The research team is pursuing follow-up work, including accessing a longer-term dataset that spans approximately 40 years to better contextualize these trends and explore possible causes.
Dworak highlighted that not all cognitive domains moved in the same direction. “If every ability had risen or fallen together, it would be easier to point to a single cause,” she said. “The mixed pattern suggests complex underlying influences and underscores the need for deeper, longitudinal research.”
About this cognition and IQ research news
Author: Kristin Samuelson
Source: Northwestern University
Contact: Kristin Samuelson – Northwestern University
Image: The image is in the public domain
Original Research: Open access. “Looking for Flynn effects in a recent online U.S. adult sample: Examining shifts within the SAPA Project” by Elizabeth Dworak et al., published in Intelligence. The study analyzes trends in ICAR-derived composite and domain scores across 13 years of SAPA Project data.
Abstract (summary)
Looking for Flynn effects in a recent online U.S. adult sample: Examining shifts within the SAPA Project
Compared with much of the European literature, fewer studies have addressed whether the Flynn effect—or its reversal—is apparent among U.S. adults. Using 13 years of cross-sectional SAPA Project data (n = 394,378), the authors examined trends in standardized composite cognitive scores and domain-specific scores (matrix reasoning, letter-and-number series, verbal reasoning, and three-dimensional rotation). Test items came from the International Cognitive Ability Resource (ICAR), with overlapping item sets of 35 items (2006–2018) and 60 items (2011–2018).
Composite scores and domain scores for matrix reasoning and letter-and-number series showed patterns consistent with a reversed Flynn effect from 2006 to 2018 across age, education and gender strata. Verbal reasoning slopes were smaller and did not consistently exceed the threshold used in this analysis. A reversed Flynn effect was also present for composite scores from the 60-item set between 2011 and 2018. In contrast, three-dimensional rotation scores provided evidence of a positive Flynn effect, with the largest increases observed when data were stratified by age.