Whose name is usually associated with developing the earliest intelligence test?

Measuring the Trait of Intelligence

One trait that could be expressed quantitatively was intelligence, tests for which were developed particularly in the United States. In 1912, Davenport arranged for his long-time friend, Henry H. Goddard (1856–1962), then Director of the Training School for Feebleminded Boys and Girls at Vineland, New Jersey, to administer versions of the French Binet-Simon test to immigrants arriving at Ellis Island. Although the Binet-Simon test was intended to measure only an individual's mental functioning at a given point in time, Goddard and a host of American psychometricians considered that it also measured innate, or genetically determined intelligence. Goddard coined the term feeblemindedness to refer to those people who scored below 70 on his tests and claimed that it “was a condition of the mind or brain which is transmitted as regularly and surely as color of hair or eyes.” Because Goddard was convinced that feeblemindedness was a recessive Mendelian trait, he reformulated the concept of intelligence from a continuous character to that of a discrete character. And it was Goddard who carried out the famous study demonstrating the supposed inheritance of mental deficiency in a New Jersey family known by the pseudonym Kallikak.

For psychometricians and eugenicists, the belief that their tests measured innate capacity rather than merely accumulated knowledge meant that the tests could be used as an instrument for carrying out educational and social policy, not merely as a measure of an individual's progress at a specific point in time. For eugenicists, the new mental tests, especially the Stanford-Binet test first published in 1916, were seen as a precise, quantitative tool for measuring an otherwise elusive, but fundamental human trait. The fact that much of the material, including terminology, on which the tests were based was culture-bound did not deter psychometricians or eugenicists from claiming that the tests measured only innate learning capacity. Even when results from the U.S. Army tests during World War I showed that the longer recruits from immigrant families had lived in the United States, the better they did on the tests, Carl C. Brigham (1890–1943), a Princeton psychologist who analyzed the data, argued that the trends showed a decline in the quality of immigrants over time, not their degree of familiarity with the cultural content of the tests.

10.03.8 Cross-Cultural Testing when Translation is Required

When a test is translated from one language to another, the research findings discussed thus far do not hold. It is inappropriate simply to translate a test and apply it in a different linguistic culture. A test must be redeveloped from scratch (although constructs may be retained) basically before any such application would be appropriate. New items, new normative data, and new scaling would all be required. This has been known since the early days of psychological assessment and testing. In the early 1900s, when the Binet–Simon tests were brought to the USA from France, approximately 30 different versions of the test were developed in the USA by various researchers. However, most of these were mere translations or contained minor modifications to adapt to American culture. The Stanford–Binet Intelligence Scale, in its various incarnations, however, became the standard bearer for measurement of intelligence for nearly 60 years and was even more popular in France at one time than the original French Binet–Simon scales. The reason for the domination of the Stanford–Binet series was Lewis Terman's insight and tenacity in redeveloping the test in the USA. After determining that Binet's theory of intelligence applied, new items were written, tried out, and a new scale devised for norming that was conceptually consistent with the Binet–Simon scales but in its practical application was a new and different test.

The problems in translating verbal and nonverbal concepts across linguistic cultures are difficult but in any event the redevelopment of tests in such circumstances seems required. Cronbach and Drenth (1972) provide a book length treatment of these problems and various experiences with proposed solutions to cross-cultural adaptation of psychological tests stemming from some 30 nations from throughout the world. The various contributors describe both the strengths and limitations of adapting tests cross-culturally from one country to another, providing perspectives from such diverse discipline psychometrics, cognitive development, psychology, and anthropology. More recent guidelines and reviews of the issues involved in cross-cultural adaptation of psychological and educational tests can be found in Hambleton (1994), Hambleton and Kanjee (1995), and Van de Vijver and Hambleton (1996).

DEVELOPMENT OF PSYCHOLOGY: THE NINETEENTH AND TWENTIETH CENTURIES

The core concepts and approaches of developmental-behavioral pediatrics are as solidly rooted in psychology as they are in pediatrics. The following brief summary highlights major developments in psychology that were particularly relevant to current practice and research.

Charles Darwin (1809-1882) has been credited with introducing the study of human behavioral development, which evolved into the psychology of children (Kessen, 1999). His essay, entitled, “A Biographical Sketch of an Infant” was a meticulous account of the capacities of his infant son. He carefully described developments in a variety of domains—movement, vision, emotions (anger, fear, and pleasure), reasoning, moral sense, and communication. This inventory presaged the domains of functioning further described and studied by subsequent contributors and formed the basis of how we view child development in the current era.

Francis Galton (1822-1911), Darwin's cousin, launched the study of human intelligence. He was particularly interested in the variation among individuals. Galton's legacy is developmental and intelligence testing, a foundation of current developmental-behavioral pediatric practice (Kessen, 1999). Alfred Binet (1857-1911) collaborated with Theodore Simon in designing a carefully constructed scale that could be used to differentiate children who were developing typically from children who required special education because of slow development. The Binet-Simon test was first published in 1905. Lewis Terman (1877-1956) standardized the Simon-Binet test on a large sample of U.S. children, creating the Stanford-Binet test of intelligence. Arnold Gesell (1880-1961) used a similar empirical approach to create an evaluation of the development of young children. His book, entitled An Atlas of Infant Behavior and published in 1934, described the typical developmental milestones. Although the developers of these assessments were clear about the limitations of the quantitative approach to measuring intelligence, the Eugenics Movement used the work of Galton and results of intelligence testing to support their claims about the superiority of white race and inferiority of African Americans, immigrants, and individuals with disabilities and mental health disorders. Eugenics advocated for improvements in the human race through selective breeding, prenatal testing, birth control, sterilization, and euthanasia (Kanner, 1964). This history emphasizes the ethical obligations of professionals in assessing the capacities of young children.

In a concurrent but independent tradition of psychology, Sigmond Freud (1856-1939) described the development of emotions and emotional disorders (Kessen, 1999). Freud proposed a three-part structure of the mind: the id, the ego, and the superego. He described five stages of psychosexual development: the oral, anal, phallic, latency, and genital stages. Freud also articulated the concept of the unconscious. Psychoanalysis became the method for helping patients acquire insights into the unconscious conflicts in their upbringing that caused emotional disorders. Most of these concepts have been severely criticized or reworked throughout the 20th and 21st centuries. Erik Erikson (1902-1994) later reconceptualized Freudian stages in psychosocial rather than psychosexual terms. The major tasks that children face at various points in development are still described in Erikson's terms.

James Mark Baldwin (1861-1934) was a leading figure in the area of sensation and perception. His experimental work on infant development strongly influenced Jean Piaget (1896-1980), whose intense observation of his three children formed the foundation of an integrated theory of cognitive development. In Piaget's theory, the sensorimotor stage of development preceded the preoperational, operational, and formal operational stages. Children progressed through these stages through processes of assimilation of environmental experiences and accommodations to those experiences. These concepts remain a foundation in experimental cognitive development.

Another influential tradition within psychology that emerged in the 19th century was the study of learning. Ivan Pavlov (1849-1936), a Russian physiologist, psychologist, and physician, described what he called the “conditioned reflex.” The conditioned reflex is the ability of a once neutral stimulus, such as a bell, to cause a physiologic reaction, such as salivation, in an animal or human based on pairings of the neutral stimulus with a motivating stimulus, such as food. These concepts are current in areas such as the causes and treatments of phobias. In the United States, James B. Watson (1878-1958) was an early behaviorist, who argued for cutting out consciousness and other intangibles from the dialogue of psychology. His hope was to control children's emotions through conditioning. B. F. Skinner (1904-1990) elaborated on operant conditioning, the ability of a reinforcing stimulus to change the probability of the appearance of behaviors. Operant conditioning still plays a central role in behavior management of children developing typically and children with disabilities. Following Skinner, behavioral approaches scrutinize antecedent conditions, behaviors, and consequences in the search for reinforcers. In addition, the frequency and pattern of reinforcement are still considered important to the maintenance of behavior change.

2 History and Theoretical Foundations: From Institutionalization to Inclusion

People with disabilities have always been part of society. What has changed is how society defines and explains these differences and how people with special needs are treated. Early explanations of disabilities were grounded in superstitious belief systems, and treatments included abandonment and extermination. By the 1800s, these gave way to quasi-scientific explanations and institutionalization of people with disabilities. Early efforts to educate children with disabilities were often conducted by physicians (e.g., Itard, Seguin, Howe, Gallaudet) whose work established the foundation for current practices such as individualized instruction and an emphasis on functional skills. Because societal demands were different (e.g., for literacy), early efforts to provide special education were focused on people with severe disabilities such as mental retardation and psychosis. During this period, asylums for the treatment of individuals with disabilities began to appear. Such institutions were prevalent in the United States until the relatively recent past and still exist widely in many other countries. Although asylums were originally intended to protect people with disabilities from cruel treatment in the outside world, patients were sometimes abused and neglected.

Assessment of cognitive processes and intelligence

Lower-order cognitive processes may be considered as a heterogeneous group of various cognitive functions (e.g. perceptual motor learning, visual short-term memory, selective attention etc.). Thus, there is no unitary-assessment approach. The evaluation of cognitive processes ranges from simple reaction times (RTs) measure to the query of strategies adopted during problem-solving tasks, and depends on the particular problem. For example, mirror tracing is a visual motor task to measure visual integration, hand–eye coordination and learning of new motor skills. During skill acquisition, subjects have to copy a figure, but with their hand being placed behind a barrier. Feedback from their copying performance is only visible in a mirror. Because a detailed description of cognitive assessments would go beyond the scope of this article, we will exclusively focus on the assessment of intelligence.

Intelligence can only indirectly be inferred. Basically, three main approaches may be distinguished (1) inference from an external criterion, (2) psychometric testing [e.g. the classical intelligence test, resulting in intelligence quotient (IQ) scores] and (3) inference from neurobiological correlates.

Intelligence can be inferred from apparent behaviour or linked to an external criterion. Usually, professional success is assumed to reflect the underlying intelligence. Intelligence correlates positively with professional success, income or career advancement, and seems to be the single best predictor for the later position in society (Brody, 1992). According to a meta-analysis by Fraser et al. (1987), correlation coefficients between intelligence scores and school marks vary from 0.34 to 0.51, depending on the type of school and years of education. Thus, intelligence only moderately correlates with school performance (Cohen, 1992), reflecting the fact that intelligence does not exactly mirror school performance, and limiting the significance of intelligence testing for prognostic use in the prediction of individual school careers. In fact, many aspects such as concrete learning opportunity, individual support and personality factors (e.g. achievement motivation), have an impact on external criteria such as school marks which explains the moderate nature of correlations between intelligence scores and school marks.

Psychometric testing is the standard approach to infer about intelligence. Specific tasks, which are assumed to reflect individual differences in intellectual ability, are applied. Introduced by Albert Binet and published as the Simon–Binet test in 1905 (Binet and Simon, 1905), the first tool for the assessment of intelligence was aimed to detect and support children with special needs. The items included in this test battery reflected age-appropriate intellectual ability. Individual items ranked for chronological age (CA) allowed for inference about children's mental age (MA). Accordingly, the individual mental ability score for a given child was related to its CA, thus providing a mean to evaluate its general state of intellectual development – advance or delay. With the transformation of the original difference score (MA minus CA) into a quotient score (MA/CA × 100) by William Stern (1912), the well-known IQ score was born. Two basic principles from this seminal work are still present in today's IQ tests: (1) adaptive testing, that is the principle to start with items assumed to match approximately the intellectual level of a given subject, with a subsequent gradual progression of item difficulty and (2) relation of individual test scores to age-standardized population norms which is based on the assumption that intellectual ability just as many other human characteristics should be normally distributed in the general population.

Finally, intelligence can be inferred from neurobiological correlates. Early anthropometric approaches from the 19th century used total brain weight and volume as correlates of intelligence, assuming that bigger brains are better brains. Even though, there are in fact positive correlations between total brain volume and IQ scores, correlation coefficients are only small to moderate (r ∼ 0.3) (McDaniel, 2005), and there is no causal relation. Besides, many of these approaches were biased by sexual and racial aspects, frequently suiting scientifically justified racism (Gould, 1994). Summarizing studies that are based on voxel-based morphometry, Jung and Haier (2007) elaborated and upgraded the original idea of linking grey matter volume to intelligence. The parieto-frontal integration theory (P-FIT) implies that networks of grey matter volumes in specific brain areas (local aspects) and interactions dependent on the connecting fibre tracts are the biological substrates of individual differences in intelligence.

A different approach, the so-called mental-speed approach is based on the assumption that speed of processing in elementary cognitive tasks is related to psychometric intelligence. In general, RTs are slower with increasing load of information. However, subjects with higher IQ scores need less extra time to process additional information. In fact, the increase in RT with increasing load of information correlates about −0.18 to −0.23 with IQ scores (Jensen, 1987). Techniques such as positron emission tomography (PET), EEG and functional magnetic resonance imaging (fMRI) have considerably contributed to a refinement of the mental-speed approach. A seminal study was performed by Richard Haier et al. (1988). Subjects engaged in a non-verbal IQ test were simultaneously scanned for their glucose metabolism rate (GMR). Surprisingly, those that reached the highest IQ scores displayed the lowest GMR. In other words, the higher the IQ score, the lower the cognitive effort or demand, which has become known as the principle of neural efficiency. Haier interpreted his finding as follows ‘… intelligence is not a function of how hard the brain works, but rather of how efficiently it works’. (Haier et al., 1992). Since then, many studies have replicated and further characterized neural efficiency which is modulated by the level of task complexity and a task–sex interaction (for review, see Neubauer and Fink, 2009).

In sum, over the past decades common lines emerged regarding the neurobiology of intelligence. First, general statements are formulated for specific abilities rather than global aptitude. Second, neurobiological correlates are related to the local, instead of the global anatomical level. Third, the focus is on dynamic rather than static aspects of processing. As a notion of caution, it has to be mentioned that neurobiological correlates are easily mistaken as a direct measurement of intelligence. There is, however, no causal relationship between physiological parameters and psychometrically assessed intelligence. Furthermore, there is no solid theoretical model linking physiological parameters of the central nervous system to intelligence as a behavioural trait and the reported magnitude of correlations between IQ scores and physiological parameters of the waking brain (e.g. waveforms of event-related potentials, cerebral metabolic rates) are only moderate (r ∼ 0.4) (Deary and Caryl, 1997).

Assessment of intellectual functions

Psychological tests were originally developed to identify intelligent people. Francis Galton believed that human intelligence is mostly inherited. But he wondered how could the most intelligent people be identified? A test of intelligence would have to be created. Galton took on the job and in 1883 published the first intelligence test.

Abstract

This paper examines the first moments of the emergence of “psychometrics” as a discipline, using a history of the Binet–Simon test (precursor to the Stanford–Binet) to engage the question of how intelligence became a “psychological object.” To begin to answer this, we used a previously-unexamined set of French texts to highlight the negotiations and collaborations that led Alfred Binet (1857–1911) to identify “mental testing” as a research area worth pursuing. This included a long-standing rivalry with Désiré-Magloire Bourneville (1840–1909), who argued for decades that psychiatrists ought to be the professional arbiters of which children would be removed from the standard curriculum and referred to special education classes in asylums. In contrast, Binet sought to keep children in schools and conceived of a way for psychologists to do this. Supported by the Société libre de l'étude psychologique de l'enfant [Free society for the psychological study of the child], and by a number of collaborators and friends, he thus undertook to create a “metric” scale of intelligence—and the associated testing apparatus—to legitimize the role of psychologists in a to-that-point psychiatric domain: identifying and treating “the abnormal”. The result was a change in the earlier law requiring all healthy French children to attend school, between the ages of 6 and 13, to recognize instead that otherwise normal children sometimes need special help: they are “slow” (arriéré), but not “sick.” This conceptualization of intelligence was then carried forward, through the test's influence on Lewis Terman (1877–1956) and Lightner Witmer (1867–1956), to shape virtually all subsequent thinking about intelligence testing and its role in society.

1 Early developments

ID has a longer history of research on comorbid psychopathology compared to ASD. This observation is understandable as ID has a much longer history as a field of study in the disciplines of education, health, and mental health. Binet famously developed the Binet–Simon intelligence test as a method of identifying and separating children with ID into classrooms separate from typically developing children. These developments occurred in Paris and culminated in the first version of the Binet–Simon test in 1905. Later, Terman translated and modified the test for use in the U.S. The first edition of the Stanford–Binet appeared in 1916 with revisions following in 1937 and 1960 (Sears, 1957).

Formative efforts in defining ASD came much later. Almost 40 years had passed since the development of the Binet–Simon before Kanner (1943) first described autism in a professional journal. Even at that point, major modifications and changes to the diagnosis of ASD continued, with scale development following even later. This preoccupation with defining core symptoms in our view was an impediment to the development of the field of comorbid conditions.

For some time after the development of modern definitions of ID and ASD, and accompanying tests to help identify these conditions, comorbid conditions had not been addressed. Additionally, various rationales for why these disorders could not overlap with mental health conditions, in particular, were common. Insufficient ego strength or poor insight into their own problems were reasons cited for these beliefs. It was not until the 1960s that researchers began to acknowledge the presence of co-occurring psychopathology among persons with ID (Gardner, 1967).

Despite these developments, there was considerable resistance to change in the field. One of the primary difficulties was the general separation of services into two tracks: ID and mental health. Thus, persons with ID and mental health concerns often found themselves in a proverbial health services no-man's-land. This service model also shaped how services were provided and how patients were viewed. The ID centers and outpatient programs tended to focus on psychological and educational services. Over time these services became more and more focused on methods and procedures adhering to an operant conditioning paradigm. These methods as a group are often referred to as applied behavior analysis. The mental health side, conversely, adopted a medical/biological model. The focus has been on differential diagnosis and psychotropic medication. Supportive psychological therapies were also employed in some instances. Thus, the types of services received were greatly affected depending on to which of the two types of agencies the individual was assigned. In truth, another problem with this approach was that both treatment models have merit. However, while these methods certainly could complement one another, that approach was seldom followed. Rather, many professionals tended to gravitate to one model or the other. The opposing camp was viewed as a rival treatment model rather than a potential asset to a particular program. The notion was more about sorting out if the person fell into an ID or a mental health box, with no consideration for overlap. This drastically limited the focus on comorbid mental health conditions in persons with ID.

ASD also developed as a singular disorder with respect to the delivery of services. Once the condition had been defined, early researchers were of the opinion that the disorder was rare and occurred among children with above average intelligence. As a result, most programs tended to be housed in medical schools or were administered by private providers. It was not until the 1960s and 1970s that the disorder was reframed. At this point, ASD made the move from being defined as a mental health disorder to a developmental disorder. Later, it became evident that a high overlap occurred between ASD and ID (Hill & Furniss, 2006; Matson & Shoemaker, 2009). In essence, this became the first major advance into the field of ASD and comorbidity. One could also argue that research showing as many as 70% of individuals with ASD also evince ID hastened and solidified the establishment of ASD as a form of developmental disability.