Models and Theories of Intelligence

Image: “Bildtitel” by A Health Blog. License: CC BY-SA 2.0

What is Intelligence?

There is still no standardized scientific definition for the term ‘intelligence’. W. Stern (1911), a pioneer in the field of intelligence, defined intelligence as the ‘general capacity of an individual to consciously adjust his thinking to new requirements’. Intelligence is a hypothetical construct that can only be observed in behavior. Read here about methodological basics.

The term intelligence comprises verbal and mathematical abilities as well as logical reasoning.

Cognitive flexibility comprises 2 pillars:

• Acquiring knowledge
• Adapting to changing circumstances

Intelligence Quotient

The classical IQ by Binet and Stern

Image: Alfredo Benet Junior (July 11, 1857 – October 18, 1911). License: Public Domain

Binet was one of the first persons that attempted to make intelligence measurable. At the beginning of the 20th century, he was commissioned to develop an intelligence test to measure the intellectual capacity of school children. The goal of the test was to identify children with learning disabilities in order to provide them with special individualized help.

The level of difficulty of the test was scaled in ascending order and was supposed to assess the intellectual age of a child with tasks usually passed by 6-year-olds, 7-year-olds, etc.

Based on Binet’s work, Stern developed the classic intelligence quotient (IQ), which is, however, not applicable to adults.

IQ = (mental age / chronological age) × 100

Example: John is 9 years old and solves all the tasks for 11-year-old children → (11/9) × 100 = 122.2

The deviation IQ by Wechsler

David Wechsler later developed as an alternative—the deviation IQ, which is still the most common type of intelligence measurement today. The subject’s score on the intelligence test is compared to the average IQ of people of the same age (the average is standardized to 100 points).

The quotient reflects the subject’s relative position within the selected reference group (e.g., high school students, children in special schools of the same age). Scores obtained from different reference groups are not directly comparable. One similarity between the deviation IQ and the classic IQ is that both compare the test performance with the respective age group.

The 4 Most Important Models of Intelligence

Important terms on models of intelligence

• Factor analysis: In factor analysis, the correlation of the individual tasks of an intelligence test is calculated. Tasks with closely related topics represent their content on a higher level. Example: Tasks that demand spatial sense correlate with each other to such a degree that they can be summarized as the spatial factor.
• Extraction: Possibilities of generating factors

Two-factor theory by Spearman (1904): the g factor

Spearman observed that subjects who were able to easily solve intelligence tasks (calculating) also scored high on other tasks (picture arrangements). From this, he concluded that there must be an inherent cognitive ability.

This intelligence is active in the performance of any task, independent of the type of task. Spearman called this basic ability the general factor of intelligence—the g factor. In hierarchical terms, this g factor is placed above the s factors.

For specific intelligence performances, s factors are necessary. Spearman argued that these s factors do not correlate with each other, meaning that an individual does not necessarily have to perform well or badly in multiple specific areas.

• g factor determines: processing speed, mental capacity, or intellectual performance (‘genius’ or ‘simple character’?)
• s factors determine: spatial, numerical, verbal, and mechanical abilities for specific areas

Image: S-factors. By: Lecturio

Multiple-factor theory of intelligence by Thurstone (1938)

Thurstone opposed Spearman’s idea of a superordinate general intelligence factor. Instead, he developed a multiple-factor theory of intelligence that established 7 primary factors. These primary factors are all on the same level.

Thurstone’s 7 primary mental abilities include:

• Number facility
• Verbal comprehension
• Word fluency
• Spatial visualization
• Associative memory
• Reasoning
• Perceptual speed

Berlin Model of Intelligence Structure (BIS) by Jäger (1984)

In German-speaking countries, intelligence research was strongly influenced by Jäger’s descriptive model which can be divided into 2 modalities: operational abilities and content-related abilities. Like Spearman, Jäger assumed a superior general factor g. The Berlin Model of Intelligence Structure (BIS) is derived directly from Jäger’s model.

Model of fluid and crystallized intelligence by Cattell (1971)

Fluid intelligence is the native ability to handle new situations and to solve problems without utilizing previously acquired knowledge. Fluid intelligence is independent of culture and develops during childhood.

Crystallized intelligence describes acquired knowledge. This knowledge is culture-specific with respect to knowledge contents and possible experiences. Crystallized intelligence peaks during adolescence and usually remains constant even in old age.

Changes in Intelligence

The proportion of older people in the population is continually increasing and understanding the development of intelligence in old age is becoming a focus of medicine and research. Old people notably differ in their degree of cognitive abilities. For the most part, training, not age, is crucial.

Average values regarding the development of intelligence:

• Until the age of 75: Minor changes
• Afterward: Accelerated degradation of cognitive functions

Measuring Intelligence: Intelligence Tests

Intelligence tests are psychological methods for assessing cognitive performances or intelligence, which is characterized as IQ. These tests are always standardized to a group, meaning that the results of the tests have a normal distribution in a certain population and the result of an individual is always a comparison with other individuals within this group.

Wechsler Intelligence Scale (WIS)

To date, the Wechsler Intelligence Scale for Children (WISC) and adults (WAIS), developed by the psychologist David Wechsler, is the most commonly used intelligence test. It is based on the general factor by Spearman and is broadly divided into a verbal subtest and a performance subtest. Accordingly, a verbal IQ and a performance IQ are calculated; their mean makes the full-scale IQ.

The different subtests test for:

The test is suitable for assessing the general cognitive state of development and for the examination of age-, environment-, or disease-related performance impairments in specific areas.

WIS represents an individual test where a subject is tested by an examiner. The test takes about 60–90 minutes. In order to ensure the objectivity of the administration, the examiner follows highly standardized instructions. Furthermore, WIS expresses a culture-independent assessment of intelligence.

Analysis of WIS

The mean WIS score is 100 points, the standard deviation is 15 points, and its design precludes gender differences. Furthermore, there are versions for adults, school children, and preschool children. The IQ is taken from the averaged verbal and performance subtests.

Examples of tasks included in WIS

Verbal subtest:

• Information: What is the capital of France?
• Comprehension: What do you do if you find an injured person lying on the sidewalk?
• Arithmetic: John bought 3 books for 5 dollars each, and paid a 10% sales tax. How much did he pay altogether?
• Similarities: How are a snake and an alligator alike?
• Digit span: The subject hears a number string once (e.g. 7 4 9 7 2) and has to verbally repeat it forward or backward.
• Vocabulary: What is the meaning of the word ‘articulate’?

Performance subtest:

• Digit-symbol coding: The subject receives a code that should help to assign specific symbols to specific numbers (e.g. 1, 2, 3, 4).
• Picture completion: In a displayed picture, the subject has to name what is missing in the depiction (e.g. ox without horns)
• Block design: Specific displayed patterns must be reconstructed with cubes.
• Picture arrangement: A quantity of jumbled up pictures (e.g. comics) have to be arranged in chronological order so that they make up an intelligible story.
• Object assembly: The subject has to assemble a known object from a certain quantity of pieces of a puzzle.

Intelligence Structure Test (IST)

The Intelligence Structure Test (IST) is based on the multiple-factor theory by Thurstone and is designed to be a group test that can be taken individually. It comprises verbal, pictorial, and numerical subtests; a newer version of IST 2000 also tests for memory. From the verbal, pictorial, and numerical factors, a secondary factor—reasoning is calculated. The administration of IST lasts around 90 minutes and can be taken by pen and paper or on the computer.

The objectivity of IST is very high because the subjects work individually without the interference of an examiner.

The mean IST score is 100 points and the standard deviation is 10 points. An overall IQ cannot be determined from IST; it only yields values for the separate subtests.

Culture fair test

The culture fair test is based on Cattell’s model of crystallized and fluid intelligence. The test assesses culture-independent fluid intelligence. It consists of the 4 nonverbal subtests—series, classifications, matrices, and conditions.

Progressive matrices test by Raven (1936)

Image: Raven’s progressive matrices example. By: Lifeof Riley. License: CC BY-SA 3.0

The progressive matrices test by Raven is a nonverbal multiple-choice intelligence test. During World War II, every draftee underwent the progressive matrices test, independent of literacy. The tasks of the test are based on understanding a present pattern and selecting the missing piece from a series of several provided pieces. There are 3 different types of matrices: standard, colored, and advanced matrices.

In 2007, 2 studies concluded that people with Asperger syndrome or classic autism achieve higher scores on average on the progressive matrices test than on WIS.

Intelligence and Achievement

The reason for assessments of intelligence is associated with the prediction of achievement variables: achievement in school, at work, in academics, etc. Students are divided into 2 groups, according to their achievements and intelligence:

• Underachiever: Achievement is not as high as expected (frequent cause: motivational deficits).
• Overachiever: Achievement is higher than expected (frequent cause: high conformity, ambition, and effort).

Excursus: The Dunning-Kruger Effect

‘If you’re incompetent, you can’t know you’re incompetent … [T]he skills you need to produce a right answer are exactly the skills you need to recognize what a right answer is.’ (David Dunning)

In their studies, Dunning and Kruger were able to identify certain characteristics that are typically found in individuals who are unconsciously incompetent.

Incompetent people:

• Overestimate their own abilities and skills
• Fail to recognize their own lack of competence