From good to bad intentions: Origins and misappropriation of intelligence testing

The foundation for viewing intelligence as a fixed and heritable trait associated with direct Mendelian inheritance can be traced to Francis Galton, who coined the term “eugenics” in the late nineteenth century and argued that mental ability was largely inherited, drawing on the evolutionary principles developed by his half-cousin, Charles Darwin (Galton, Reference Galton1883; Paul & Moore, Reference Paul, Moore, Bashford and Levine2012). Galton generalized the concept of evolution to include social policy, promoting the selective breeding of individuals believed to possess superior traits (i.e., positive eugenics). In Hereditary Genius, Galton defined eugenics as “the study of all agencies under human control which can improve or impair the racial quality of future generations” (Galton, Reference Galton1883). Galton sought to operationalize these ideas by promoting anthropometric measurement practices aimed at ranking individuals and populations, which he argued provided a scientific basis for social stratification and reproductive control.

Alfred Binet and Théodore Simon designed the first formal cognitive assessment with the explicitly designed aim to assess functional performance. Commissioned by the French Ministry of Education, the Binet-Simon scale was designed to identify school aged children who were performing poorly academically and who could benefit from targeted educational support. Binet and Simon emphasized that their test was “to determine, in a practical manner, the degree to which a child is able to profit from instruction. … It is not a question of measuring innate intelligence, but of evaluating the child’s performance” (Binet & Simon, Reference Binet and Simon1908/Reference Binet and Simon1916).

This distinction was abandoned with the scale’s introduction to the United States, where Henry Goddard’s 1910 translation and adaptation reframed it as a “scale of intelligence” (Goddard, Reference Goddard1910, Reference Goddard1913). Through his work at the Vineland Training School and later at the Ohio Committee on the Sterilization of the Feeble Minded, Goddard used test results to classify individuals into formal categories using now offensive labels (e.g., “moron,” “imbecile”). While Binet developed the test to identify students who would benefit from additional educational support, Terman applied it to identify high performing individuals and to characterize “superior intelligence” (Terman, Reference Terman1925). As a result, the focus on high performing individuals reinforced the belief that IQ scores provided a definitive and comprehensive representation of intellectual capacity at all levels of ability.

The seduction of quantification

The emergence of IQ testing in the early 20th century coincided with major advances in statistical methods, including correlation and factor analysis. Figures such as Karl Pearson, Charles Spearman, and Ronald Fisher developed techniques that later became foundational to modern data science. Applied to intelligence scores, these methods conveyed an appearance of scientific rigor while, within the social context of the time, reinforcing existing patterns of social stratification.

Karl Pearson, the first Galton Chair of Eugenics at the University of London, applied correlation to study heredity and human characteristics, asserting that intelligence and other biological attributes were inherited and statistically predictable (Pearson, Reference Pearson1907). Charles Spearman observed the “positive manifold,” the pattern whereby individuals’ scores across different cognitive tests tend to correlate, which prompted him to propose a general intelligence factor, or “g,” as part of his Two-Factor Theory of intelligence (Spearman, Reference Spearman1927).

Ronald A. Fisher succeeded Karl Pearson’s as Chair of Eugenics at the University of London, where he also served as editor of the Annals of Eugenics and developed statistical techniques that became central to eugenic research (Fisher, Reference Fisher1930). Recognized primarily for creating analysis of variance, with the F being his initial, Fisher actively championed eugenic causes throughout his career. As a member of the British Eugenics Society’s executive committee, he advocated for policies designed to encourage reproduction among individuals he deemed genetically “superior” (he had eight children) and opposed efforts to separate the movement from its discriminatory foundations.

Eugenics: The scientific rationalization of social hierarchy

During the early 1900s, the eugenics movement expanded throughout the United States with the stated goal of “enhancing” the nation’s genetic quality through controlled reproduction and the systematic exclusion of individuals labeled as biologically deficient. Proponents drew parallels to agricultural practices, contending that just as selective breeding enhanced crops and livestock, similar methods could be applied to human populations. Affiliations with universities and emerging disciplines such as statistics, psychology, and cognitive assessment provided the movement with the appearance of scientific legitimacy, extending its influence to education, medical practice, and governmental policy (Figure 1).

Figure 1.

Program from the Second International Congress of Eugenics (1921) proclaiming that “eugenics is the self-direction of human evolution.” The tree’s roots depict disciplines deemed foundational to the eugenics movement including psychology, mental testing, and statistics highlighting the perceived scientific legitimacy drawn from these emerging fields to justify eugenic ideology and policy “which bear upon the improvement of racial qualities in man.” Public domain.

The Eugenics Record Office (ERO) at Cold Spring Harbor, New York, established by biologist Charles Davenport and directed by Harry H. Laughlin, was at the center of the American eugenics movement (Allen, Reference Allen1986). The ERO gathered and examined family genealogies (Figure 2), institutional records, and intelligence test results to develop accounts of inherited “deficiency,” thereby reinforcing hierarchies based on race, class, and disability status. Its findings influenced public health and educational policies in ways now recognized as discriminatory.

Figure 2.

Heredity chart from early 20th century categorizing individuals by sex, ancestry, and alleged hereditary traits. Conditions marked as inheritable include “feeble-mindedness,” “criminalistic” behavior, “insanity,” and “epilepsy,” which was red symbols labeled “E.” These notations were used in the construction of family pedigrees. Reproduced from Davenport, C. B. et al. The Study of Human Heredity: Methods of Collecting, Charting and Analyzing Data. Eugenics Record Office, Cold Spring Harbor, NY (1911). Public domain.

The Second International Congress of Eugenics, held in 1921, symbolized the movement’s mainstream acceptance and international ambitions (DeSalle, Reference DeSalle2021; Kevles, Reference Kevles1985). Henry Fairfield Osborn, president of the American Museum of Natural History, opened the Congress remarking “the right of the state to safeguard the character and integrity of the race or races on which the future depends is, to my mind, as incontestable as the right of the state to safeguard the health and morals of the people” (Osborn, Reference Osborn1921). He argued that science must guide government in preventing the “multiplication of worthless members of society.”

As eugenic principles gained wider implementation, compulsory sterilization became a central policy. Beginning with Indiana’s 1907 law, more than 30 states enacted statutes targeting those deemed unfit, guided by the ERO’s Model Eugenical Sterilization Law (Laughlin, Reference Laughlin1922). These measures were widely endorsed by medical and public health authorities and later influenced Nazi Germany’s 1933 sterilization law (Sofair & Kaldjian, Reference Sofair and Kaldjian2000). American eugenics thus shaped both domestic policy and international practice, with the belief in sterilization as a public good persisting until the last U.S. statute was repealed in 2013 (West Virginia Legislature, 2013).

From army testing to classification bias

Intelligence testing extended beyond institutional settings to the military, where Robert Yerkes designed the Army Alpha test for literate recruits and the Army Beta test, a nonverbal, picture-based measure, for illiterate or non–English-speaking recruits (Yerkes, Reference Yerkes1921). Developed for rapid wartime assessment, these tests were administered to 1,726,966 soldiers between 1917 and 1919. Although the Army Beta acknowledged the impact of education and sociocultural background, scores from both tests were treated as equivalent and combined to generate overall intelligence rankings.

Princeton psychologist Carl Brigham used test results to rank immigrant groups by IQ and promote hierarchical views of intelligence by nationality (Brigham, Reference Brigham1923; see Figure 3). In his original analysis, native-born white Americans of Northern and Western European descent scored highest, whereas immigrants from Southern and Eastern Europe scored lower and were labeled intellectually inferior. Brigham later retracted these conclusions, acknowledging the cultural biases inherent in the tests that were not previously considered, explicitly stating “test scores may not represent unitary things” (Brigham, Reference Brigham1930).

Figure 3.

Ranking of immigrant groups by average IQ scores based on Army Alpha and Beta test results. These results were widely circulated to support eugenic arguments and immigration restrictions despite being heavily biased by language, education, and cultural familiarity. Reproduced from Carl C. Brigham, A Study of American Intelligence (Princeton University Press, Reference Brigham1923). Public domain.

Intelligence testing in the U.S. played a key role in shaping immigration policy, reinforcing discriminatory ideas about race, intelligence, and national identity. Laughlin’s influence helped embed these views in the Johnson-Reed Immigration Act of 1924, which barred entry to groups labeled “intellectually and racially inferior” (Lombardo, Reference Lombardo2008). The eugenics movement’s promotion of IQ testing provided supposed scientific justification for a heredity-focused, meritocratic vision, mirroring the arguments of Madison Grant’s The Passing of the Great Race (Grant, Reference Grant1916).

Intelligence testing in the early twentieth century served as a mechanism of immigration control and exclusion at Ellis Island and other U.S. ports of entry. Under Howard A. Knox, cognitive assessments, including nonverbal, performance-based measures, were implemented to manage immigrant intake and reduce apparent cultural and linguistic bias. Nevertheless, these approaches remained embedded in hereditarian assumptions, and, as Richardson (Reference Richardson2011) documented, routinely misclassified individuals in ways that reinforced existing social hierarchies.

Intelligence, eugenics, and the epileptic other

Religious and charitable organizations initially established epilepsy colonies to provide residents with specialized care, education, and vocational training. These communities were designed as supportive, self-contained environments where individuals with epilepsy could live, work, and receive treatment. Over time, however, several colonies were co-opted by eugenicists, who framed epilepsy as a hereditary defect threatening the “germ plasm” and used these institutions to justify segregation and restrict reproduction among those deemed genetically “unfit.” Notable examples include the Monson State Hospital for Epileptics in Massachusetts and the New Jersey State Village for Epileptics in Skillman, which embodied both the therapeutic and coercive dimensions of these early institutional approaches (Loring & Hermann, Reference Loring, Hermann, Barr and Bieliesksas2017).

Wallin (Reference Wallin1912) used the Binet-Simon intelligence test to assess 333 patients in the New Jersey State Village for Epileptics between 1910 and 1911, providing early empirical data and helping integrate psychological testing into neurological settings. Subsequently, Fox (Reference Fox1924) showed that academic underachievement among children with epilepsy excluded from public education was not fully explained by Binet-Simon IQ scores, underscoring both the value and limitations of early intelligence testing in epilepsy (Fox, Reference Fox1924; Hermann, Reference Hermann2010).

As eugenics gained scientific legitimacy, epilepsy colonies increasingly incorporated intelligence testing. At that time, epilepsy was frequently conflated with cognitive disability and mental illness, and IQ scores became a central tool in labeling residents as permanently “deficient.” Eugenic field workers created detailed family pedigrees to trace perceived hereditary “taint,” regularly identifying individuals as socially unfit based on seizure history, low test scores, and assumptions of intellectual inferiority. Intelligence testing was interpreted without considering seizure activity, medication effects, or broader social disadvantages, and low IQ scores were presented as objective proof of intrinsic deficiency. Consequently, some residents of epilepsy colonies were subjected to forced sterilization including surgical castration under the belief that they posed a threat to the genetic health of future generations (Kelley, Reference Kelley2006).

Eugenic ideology represented what was regarded at the time as progressive medical and scientific thinking. William J. Mayo, founder of the Mayo Clinic, used a 1923 address published in the Boston Medical and Surgical Journal to endorse sterilization and immigration restriction as means of “public health improvement.” Mayo argued that poverty and illness reflected “hereditary inferiority” and advocated reducing the number of people cared for “at taxpayer expense” through eugenic measures (Lombardo, Reference Lombardo2024).

Endorsements of eugenic policies by medical authorities supplied the scientific and moral justification for subsequent legal measures. The most consequential of these was Buck v. Bell (1927), a landmark U.S. Supreme Court case that upheld the constitutionality of Virginia’s eugenics-based sterilization law. The case involved Carrie Buck, institutionalized at the Virginia Colony for the Epileptic and Feebleminded and labeled “feebleminded,” a designation also applied to her mother and daughter. Buck v. Bell served as the test case for whether the state could involuntarily sterilize individuals deemed unfit to reproduce. In an 8–1 decision, Justice Oliver Wendell Holmes Jr. famously declared, “Three generations of imbeciles are enough.” This ruling legitimized compulsory sterilization under the guise of public health and the purported goal of genetic improvement.

Early IQ studies unfortunately suffered from ascertainment bias. Because intelligence test data were largely derived from residents of institutional epilepsy colonies, early reports frequently showed low median IQs around 70 (Hermann, Reference Hermann2019), reflecting not only epilepsy severity and underlying neurologic etiologies but also treatment effects such as bromides and the cognitive and social consequences of prolonged exclusion from formal education. In the 1940s, however, epileptologist William Lennox, despite his support of eugenics (Offen, Reference Offen2003), and psychologist A. Louise Collins studied patients in private practice and demonstrated that average IQ scores were common, revealing bias in earlier institutional samples and challenging prevailing assumptions about cognition in epilepsy (Collins & Lennox, Reference Collins and Lennox1947; Collins, Reference Collins1951). By then, however, misuse of IQ testing had already contributed to enduring stigma shaping clinical and societal attitudes.

Early neuropsychology in epilepsy heavily relied on IQ-based patient classification. Donald Hebb (Reference Hebb1939a, Reference Hebb1939b) observed that patients undergoing extensive epilepsy surgeries, including unilateral and bilateral frontal or unilateral temporal resections, often showed stable, unimpaired, or even improved postoperative IQ scores. These findings led him to emphasize the assessment of specific cognitive function. Subsequent work by Meyer and Yates (Reference Meyer and Yates1955) and Scoville and Milner (Reference Scoville and Milner1957) reinforced this perspective, laying the groundwork for the structure–functions approach in epilepsy neuropsychology.

Cognitive functioning in epilepsy exhibits substantial heterogeneity both between and within syndromes, rendering global IQ scores insufficient for accurately characterizing domain-specific impairments. In many cases, particularly for self-limited epilepsies, overall IQ is typically average, yet individuals may exhibit selective deficits in specific cognitive domains. Individuals with temporal lobe epilepsy, for example, often experience selective memory and language impairments that can be obscured by average or above-average overall IQ, without consistent verbal versus performance IQ differences corresponding to seizure onset laterality (Hermann et al., Reference Hermann, Gold, Pusakulich, Wyler, Randolph, Rankin and Hoy1995; Loring et al., Reference Loring, Strauss, Hermann, Barr, Perrine, Trenerry, Chelune, Westerveld, Lee, Meador and Bowden2008). By integrating neuropsychology early into epilepsy surgery outcomes research, and aligning cognitive findings with EEG and neuroimaging data, domain specific impairments rather than global IQ scores emerged as essential for determining surgical candidacy and evaluating treatment outcomes (Chelune et al., Reference Chelune, Naugle, Hermann, Barr, Trenerry, Loring, Perrine, Strauss and Westerveld1998; Loring & Hermann, Reference Loring, Hermann, Barr and Bieliesksas2017; Loring, Reference Loring2010). These findings underscore the importance of moving beyond IQ as a primary measure of cognitive ability and highlight the need for more domain-specific assessments to accurately identify clinically meaningful cognitive deficits associated with epilepsy (IC Code) (Norman et al., Reference Norman, Wilson, Baxendale, Barr, Block, Busch, Fernandez, Hessen, Loring, McDonald and Hermann2021).

Contemporary neuropsychology research in epilepsy increasingly emphasizes cognitive phenotypes as a framework for characterizing both seizure syndromes and variability within syndromes, moving beyond reliance on global IQ indices. Studies in adults with frontal lobe epilepsy demonstrate discrete cognitive phenotypes defined by differential involvement of language, attention, executive functioning, processing speed, and learning, highlighting substantial heterogeneity that is obscured by summary measures of overall ability (Arrotta et al., Reference Arrotta, Reyes, Kaestner, McDonald, Hermann, Barr, Sarmey, Sundar, Kondylis, Najm, Bingaman and Busch2022). Parallel work examining subjective and objective cognition shows domain specific correspondence between patient reported difficulties and measured performance in areas such as attention, memory, and executive functioning, supporting the clinical utility of standardized multi domain assessment approaches (Hohmann et al., Reference Hohmann, Berger, Kastell and Holtkamp2023). In pediatric epilepsy, comparisons between localization related and absence syndromes reveal syndrome specific cognitive vulnerabilities, including focal verbal memory weaknesses, despite broadly preserved intellectual functioning (Kernan et al., Reference Kernan, Asarnow, Siddarth, Gurbani, Lanphier, Sankar and Caplan2012). Similarly, studies comparing childhood epilepsy syndromes identify both shared and distinctive patterns of impairment, such as differential effects on visual attention versus spatial abilities, further reinforcing the value of domain based cognitive characterization (Cheng et al., Reference Cheng, Yan, Gao, Xu, Zhou and Chen2017). Collectively, this literature reflects a shift toward cognitive phenotyping as a means of refining syndrome classification, capturing within syndrome variability, and identifying clinically meaningful cognitive risk profiles in epilepsy.

From fixed trait to dimensional constructs: Contemporary characterization

Binet recognized a fundamental limitation of intelligence tests, noting that “the scale does not permit the measure of intelligence, because intellectual qualities are not superposable, and therefore cannot be measured as linear surfaces are measured” (1909, as cited in Zenderland, Reference Zenderland1998). Such tests capture performance on specific tasks rather than a fixed underlying intelligence. Sternberg (Reference Sternberg1999) extended this view by defining “successful intelligence” as the ability to balance the needs to adapt to, shape and select environments in order to attain success, emphasizing the capacity to navigate real-world, context-dependent, and ambiguous problems and illustrating intelligence as a multifaceted, contextually shaped ability rather than a linear trait.

Modern approaches to cognitive assessment, exemplified by Process Overlap Theory (POT; Kovacs & Conway, Reference Kovacs and Conway2016), conceptualize intelligence as an emergent outcome of interacting executive and domain general processes rather than a single causal factor. By treating “g” as a statistical product of overlapping processes instead of a unitary construct, POT provides a contemporary framework that better accounts for cultural and contextual influences on cognitive performance.

Cognitive abilities and traits show substantial correlation heritability and longitudinal stability (Breit et al., Reference Breit, Scherrer, Tucker-Drob and Preckel2024; Polderman et al., Reference Polderman, Benyamin, de Leeuw, Sullivan, van Bochoven, Visscher and Posthuma2015); however, this does not demonstrate that cognitive test scores provide a direct measure of intelligence. Cattell and Horn explicitly rejected Spearman’s notion of a unitary general intelligence factor “g,” arguing that it reflects a statistical artifact of correlations among heterogeneous cognitive tasks rather than a single underlying mental capacity (Cattell, Reference Cattell1963; Horn, Reference Horn, McGrew, Werder and Woodcock1991). The common practice of interpreting “g” as evidence of global intelligence, particularly in the absence of a universally accepted definition of intelligence, reflects the subjective assumptions and biases of the researchers choosing that specific descriptive label.

The Cattell–Horn–Carroll (CHC) model is a hierarchical, empirically derived framework that organizes cognitive abilities across multiple levels of specificity. At its broadest level, the model includes a general factor that may emerge statistically from correlations among cognitive tasks but is not assumed to represent a unitary or fixed biological construct. Beneath this level, the CHC model specifies multiple broad cognitive abilities that are relatively distinct yet interrelated, including fluid reasoning, crystallized knowledge, processing speed, visual and auditory processing, short term and working memory, long term storage and retrieval, and quantitative knowledge. Each broad ability is further decomposed into narrower, more specific skills that correspond closely to performance on individual neuropsychological and psychometric tests. The CHC framework provides a structured approach for understanding cognitive strengths and weaknesses across domains rather than reducing performance to a single global score. Despite its prominence in cognitive and educational assessment, and its frequent use in test development and research, neuropsychological test interpretation in routine clinical practice rarely adopts the CHC framework as an organizing model (Hermann et al., Reference Hermann, Loring, Bowden and Sarkis2023; Jewsbury et al., Reference Jewsbury, Bowden and Duff2017).

Contemporary implications and ongoing inequities

Despite advances in theory and measurement, the legacy of intelligence testing, including its constructs, terminology, and interpretive frameworks, continues to shape neuropsychological practice and policy, sometimes reinforcing inequities. Within large-scale datasets such as the Adolescent Brain Cognitive Development study, broad cognitive composites derived from factor-analytic or predictive modeling approaches are described as indexing General Cognitive Ability (GCA) (Sripada et al., Reference Sripada, Angstadt, Taxali, Clark, Greathouse, Rutherford, Dickens, Shedden, Gard, Hyde, Weigard and Heitzeg2021). GCA does not denote a single, unitary ability, but rather an empirically derived dimension capturing shared variance across diverse cognitive tasks, reflecting the well-documented tendency for performance across cognitive domains to covary. Research with this approach indicates that these general factors are associated with a range of educational, occupational, and health-related outcomes and are robustly linked to socioeconomic conditions, while also reflecting highly distributed, small-effect contributions across neural systems rather than localized or modular substrates.

Consistent with this empirical approach, terms such as Total Cognition Composite, introduced with the NIH Toolbox (Heaton et al., Reference Heaton, Akshoomoff, Tulsky, Mungas, Weintraub, Dikmen, Beaumont, Casaletto, Conway, Slotkin and Gershon2014), provide descriptive, data-driven summaries of broad cognitive performance by averaging standardized scores across neuropsychological batteries to capture overall efficiency relative to normative expectations. Characterizing group performance averages offers a practical and transparent summary across multiple measures without implying a fixed or essentialized construct of intelligence. Such composites, widely used in clinical trials to track cognitive status, disease progression, and treatment effects in conditions including HIV (Carey et al., Reference Carey, Woods, Gonzalez, Conover, Marcotte, Grant and Heaton2004), multiple sclerosis (Erlanger et al., Reference Erlanger, Kaushik, Caruso, Benedict, Foley, Wilken, Cadavid and Deluca2014), Alzheimer’s disease (Donohue et al., Reference Donohue, Sperling, Salmon, Rentz, Raman, Thomas, Weiner and Aisen2014), traumatic brain injury (Schneider et al., Reference Schneider, Pike, Elser, Coresh, Mosley, Diaz-Arrastia and Gottesman2024), and epilepsy (Meador et al., Reference Meador, Seliger, Le, Li, Razavi, Falco-Walter, King, Graham, Cunningham, Leeman-Markowski, Boyd, Loring and Gerard2025), represent efficient tools for summarizing and monitoring cognitive functioning over time rather than as direct measures of intelligence or formal models of discrete cognitive abilities.

Clarifying terminology is necessary but insufficient for meaningful reform. The continued use of intelligence based language reflects entrenched practices in how cognitive performance is measured, normed, interpreted, and applied across educational, medical, and legal contexts. Promoting equity in neuropsychological evaluation requires pairing linguistic revision with procedural modernization, including diversified normative samples, socioculturally informed interpretation, standardized language access practices, and reduced reliance on rigid cutoff scores that obscure individual variability. Aligning conceptual precision with systemic reform strengthens both scientific validity and ethical integrity.

Equitable assessment also depends on structural change, including multi-method, context sensitive evaluation, elimination of fixed IQ thresholds in legal settings, and training in the historical and sociocultural foundations of testing. Larry P. v. Riles (1979) exposed racial bias in school placement; Atkins v. Virginia (2002) and Hall v. Florida (2014) demonstrated the harms of inflexible IQ cutoffs in capital cases; and the NFL Players’ Concussion Injury Litigation (2015) showed that race-based baseline assumptions produced systematically lower compensation for Black players. Together, these cases underscore that meaningful reform in neuropsychology requires coordinated procedural and structural change to ensure fairness and transparency. Despite well-documented methodological and sociocultural limitations, some secondary analyses of large datasets continue to misinterpret summary cognitive scores as indices of intelligence to rank racial and ethnic groups (Human Varieties, Reference Varieties2023). Additional examples of the continued misuse of IQ scores are provided in the Supplementary File.

Beyond terminology: Toward systemic reform and equity

Although traditionally referred to as “intelligence scales,” instruments such as the Wechsler Scales are more accurately understood as assessing a range of distinct cognitive processes that differ in their relevance to everyday functioning. Continued use of terms like “IQ” and “intelligence” in research, or the inclusion of “Intelligence” as a section header in clinical reports, mischaracterizes what these measures actually capture and risks reinforcing reductive and stigmatizing views of cognition. Reporting “intelligence levels” is particularly problematic given the absence of a universally accepted definition of intelligence and the documented potential for such labels to carry pejorative implications. Adopting terminology that specifies discrete cognitive functions rather than invoking an assumed global construct promotes more accurate interpretation and aligns with principles of nonmaleficence by minimizing harm and distancing clinical practice from the cultural and historical baggage associated with outdated conceptions of intelligence.

Psychological science has repeatedly revised terminology to reflect current conceptual accuracy, as seen in the transition from “mental retardation” to “intellectual disability” in both the DSM and ICD frameworks (Harris, Reference Harris2013). Similarly, research by Fletcher et al. (Reference Fletcher, Lyon, Fuchs and Barnes1994) and Siegel (Reference Siegel1989) demonstrated how empirical studies of cognitive and developmental functioning prompted revisions to learning disorder and childhood psychopathology classifications. These examples illustrate the shift from outdated constructs to evidence-based, data-driven diagnostic frameworks.

Epilepsy provides a clear precedent for reevaluating terminology in neuropsychology. In South Korea, the replacement of the term for “epilepsy,” which carried pejorative connotations, with a neutral, biologically grounded alternative (i.e., cerebroelectric disorder) led to measurable improvements in public awareness, reduced stigma, and more patient-centered framing (Koh et al., Reference Koh, Pyo, Jang, Kim, Won Seo, Huh, Kwon and Young Choi2024; Shorvon, Reference Shorvon2024). This demonstrates how diagnostic language shapes social meaning and clinical expectations.

The continued use of “intelligence” in neuropsychology mirrors this problem. Like the historical term for epilepsy, intelligence is an overutilized construct increasingly at odds with contemporary scientific models and, as evidence shows, directly linked to stigma and prejudice. Persisting in its use reinforces reductive interpretations and reflects an unwarranted reluctance to adopt modern conceptual frameworks. Epilepsia, the official journal of the International League Against Epilepsy, now discourages the term “intelligence” in its research reports, highlighting the field’s growing awareness of the potentially stigmatizing effects of specific language.

Abandoning historically burdened labels, as seen with the epilepsy name change in South Korea, can shift public and patient perceptions, strengthen clinical care, and reduce misinterpretation. Revising “intelligence” and “IQ” aligns with this scientific imperative. Over 35 years ago, Muriel Lezak declared IQ “dead” in her Reference Lezak1988 INS presidential address, calling for a move beyond reductive measures. Advancing neuropsychology requires both multidimensional cognitive models and precise language to uphold scientific accuracy and mitigate stigmatizing implications.

Language change alone is insufficient; meaningful reform requires systemic action. Transitioning away from the term intelligence will be more challenging in pediatrics, where it is embedded in formal diagnostic labels such as intellectual disability and in educational systems that use it for student placement into gifted programs. Even in these contexts, framing performance in terms of abilities rather than intelligence remains appropriate and preserves diagnostic specificity. Importantly, there are growing efforts to integrate historical and cultural literacy into training, such as those advanced through the Minnesota Conference (Stringer et al., Reference Stringer, Fuchs, Bordes Edgar, Bristow, Correia, Penna, Reyes, Whiteside, Block, Bodin, Butt, Calamia, Didehbani, Dodzik, Dotson, Fernandes, Freece, Fuller, Getz and Schmitter-Edgecombe2025), with relevance to the interpretation of cognitive performance across all assessment approaches. Updating terminology reflects the natural evolution of scientific language toward greater precision and contextual validity. As neuropsychology advances, its vocabulary should reflect, rather than constrain, our understanding of cognitive diversity.