Learning outcomes
By the end of this lesson you should be able to: define intelligence and describe the major psychometric, cognitive and multiple factor theories of it; calculate and interpret a deviation IQ and name the properties that make a test usable; explain what heritability does and does not mean, and what the Flynn effect demonstrates; describe the principal trait theories of personality, including the Big Five, alongside the psychodynamic, humanistic and social-cognitive accounts; and apply this to the assessment of intellectual disability and to the recognition of personality traits that shape vulnerability to disorder.
Prerequisites
Lesson 3 (Memory and thinking). Working memory and processing speed are components of any measured intelligence test, and the psychometric ideas of reliability and validity rely on the memory and test-retest logic from that lesson.
The big idea
Both constructs describe differences between people rather than deficits within a person, and both are used to predict, never to rank worth. Every measurement in this lesson carries the same warning: a standardised score describes performance under particular conditions with particular instruments, and the conditions are never neutral. A patient educated in a second language, with a hearing impairment, distracted by voices, in a ward with an open door, or disabled by depression cannot be assessed as though those conditions were absent.
The structure of intelligence
| Theory | Author and era | Core claim | Strength and limitation |
|---|---|---|---|
| General intelligence, g | Charles Spearman, 1904 | All cognitive tasks correlate positively because they share one general factor, supplemented by task-specific s factors | The positive manifold between tasks is the most replicated finding in psychology; g does little to explain which processes differ |
| Primary mental abilities | Louis Thurstone, 1938 | Several correlated group factors (verbal comprehension, number, reasoning, spatial, memory, perceptual speed, word fluency) rather than one g | Explains specific profiles; the factors remain correlated, so g reappears at a higher level |
| Fluid and crystallised | Raymond Cattell and John Horn | Gf is novel reasoning; Gc is accumulated knowledge. Gf peaks in early adulthood and declines; Gc rises into late life | Explains why older patients remain excellent at vocabulary and poor at novel matrices — a clinically essential distinction |
| Multiple intelligences | Howard Gardner, 1983 | Seven or more independent intelligences including musical, bodily-kinaesthetic, interpersonal and intrapersonal | Persuasive in education; poorly supported by factor analysis, since most of these correlate and load on g |
| Triarchic | Robert Sternberg | Analytical, creative and practical intelligence | Useful critique of a purely academic construct; hard to measure psychometrically |
| Three-stratum / CHC | Carroll, later McGrew and Flanagan | g at the apex, broad abilities (Gf, Gc, Gv, Gsm, Gs, Ga) in the middle, narrow abilities below | The framework on which modern IQ batteries, including Wechsler, are built and reported |
Practical summary: g and its broad abilities describe real and predictive differences, but the specific profile matters clinically. A patient with a spatial ability of the 5th percentile and a verbal ability of the 75th has an average IQ and a specific, remediable problem that the average conceals — and that profile will not be seen at all if you never break the score down.
Measuring intelligence
The Wechsler scales (WAIS in adults, WISC in children) report a Full Scale IQ that is a deviation IQ: the test population is normed so that the mean is 100 with a standard deviation of 15, and the score is a statement about position in that distribution, not a count of correct answers. Because the distribution is normal, roughly two thirds of people score between 85 and 115, and roughly 2.3 percent deviate more than 30 points from the mean in either direction.
| Property | What it means | Why the examiner cares | How it fails in practice |
|---|---|---|---|
| Standardisation | Uniform instructions, materials and conditions; a reference group provides the norms | Any departure changes the score's meaning | Testing in a foreign language, with a deaf patient unaided, or with a mimicking respondent |
| Reliability | Consistency: test-retest over weeks, internal consistency, and inter-rater agreement | A score that changes between two unchanged administrations cannot be interpreted | Practice effects, retrieval practice, retesting in a different mood state |
| Validity | Whether the test measures what it claims: content, criterion and construct validity | Without it reliability is measuring something irrelevant, reliably | Culture-loaded vocabulary items; tests of 'achievement' presented as tests of ability |
| Sensitivity and specificity | Detection of cases and exclusion of non-cases, which trade against each other | Screening scales are false-positive machines by design and require confirmation | The PHQ-9 administered and treated as a diagnosis |
| Floor and ceiling effects | Whether the test can detect very low and very high ability | A normal-range patient scoring at the floor is a measurement problem, not a patient problem | Using an adult scale on a child or a dementia patient without a lower-bound form |
NoteRaven's Progressive Matrices was designed as a culture-reduced test of fluid reasoning: non-verbal, minimal instruction, no vocabulary requirement. It is the best available approximation to a culture-fair test, and it remains an approximation — motivation, schooling and familiarity with test formats still affect scores, so 'culture-reduced' is a statement about reducing, not eliminating, bias.
Heritability is the proportion of variance in a trait attributable to genetic differences within a specified population in a specified environment. It is a population statistic, not a statement about any individual, and it rises with environmental equality: as schooling becomes universal and nutrition improves, the remaining differences become more genetic. Heritability of measured intelligence is moderate to high in adulthood and increases with age, which puzzles people who expect genes to matter less as experience accumulates — the mechanism is that people increasingly select environments that match their genes.
The Flynn effect is the steady rise of raw scores on IQ tests through the twentieth century in many countries, roughly three points per decade, with some countries reporting declines since the 1990s. Its significance is that the rate is far too fast for genetic change: it must reflect environmental change — nutrition, infectious disease, schooling, test familiarity, and the cognitive demands of modern life. Flynn himself argued that this demonstrates that IQ is not a measure of inborn fixed capacity.
Intellectual disability: diagnosis is functional, not a number
Current definitions (DSM-5, AAIDD) require deficits in both intellectual functioning and in adaptive functioning — conceptual, social and practical skills of everyday life — with onset in the developmental period. Severity is specified from the adaptive profile, and a score of 70 or below on a well-standardised instrument is a necessary but not sufficient part of the judgement.
| Severity | Approximate IQ range | Adaptive picture | Support implications |
|---|---|---|---|
| Mild | 50-69 | Academics to about primary level; independent living possible with support; social judgement immature | Support with finances, decision-making and transitions; this is the most common group and the most under-recognised |
| Moderate | 35-49 | Academics to early primary; supported work; independent living with substantial support | Supervised work settings, structured daily living, support with safety |
| Severe | 20-34 | Limited communication and conceptual skills; continuous supervision | High levels of hands-on support; movement and communication aids assessed early |
| Profound | Under 20 | Very limited communication; constant support for all daily care | Specialist residential care; sensory impairment is often coexisting and easily overlooked |
Watch outNever use an IQ score as a label. Say 'a score of 65 on the WAIS-IV at age 9' — a measurement with a date and an instrument — rather than 'a mild retardate'. 'Retardation' and 'borderline' as nouns are obsolete, and 'low functioning' is vague while carrying stigma. Diagnose intellectual disability on adaptive function, and remember that depression, ADHD, autism, hearing impairment and a poor testing environment can all drag a single score down.
Personality: trait theories
| Approach | Proponents | Structure | Contribution |
|---|---|---|---|
| Lexical and trait | Gordon Allport; Raymond Cattell (16PF) | Personality described by the trait words of ordinary language | The catalogue of traits; trait terms are how clinicians describe patients in handovers |
| Biological trait | Hans Eysenck | Three dimensions — extraversion (cortical arousal), neuroticism (limbic lability), psychoticism — with psychophysiological grounding | Predicts vulnerability to disorder: high neuroticism to anxiety and depression, low extraversion to withdrawal, high psychoticism to antisocial conduct |
| Five-factor model | Costa and McCrae | Five broad factors: Openness, Conscientiousness, Extraversion, Agreeableness, Neuroticism — with six facets each | The dominant empirical model; robust across cultures and languages, moderately heritable, and stable in adulthood |
| Type theory | Myers-Briggs and similar | Four dichotomies producing 16 types | Popular in workplaces; test-retest reliability and predictive validity are poor, and it should not be used for clinical or occupational decisions |
The Big Five is the best-evidenced description of normal personality variation, and its clinical value is predictive: high neuroticism is the strongest trait predictor of anxiety and depressive disorders, low agreeableness and low conscientiousness strongly predict substance misuse and antisocial outcomes, and conscientiousness predicts adherence and longevity. Two qualifications keep it honest. Personality is not destiny — traits are one of several interacting influences — and the model describes dimensions on which everyone sits, so high neuroticism is a risk factor, not a diagnosis.
Personality: psychodynamic, humanistic and social-cognitive accounts
| Approach | Central claim | Key structures or stages | Enduring clinical use |
|---|---|---|---|
| Psychoanalytic (Freud) | Behaviour is driven by unconscious conflict between impulse and defence, formed in childhood | Id (instinct), ego (reality-testing and defence), superego (conscience and ideal); psychosexual stages oral, anal, phallic, latency, genital | The concept of defence mechanisms remains indispensable: repression, projection, displacement, sublimation, reaction formation, rationalisation, denial, regression |
| Psychosocial (Erikson) | Development proceeds through eight psychosocial crises across the whole lifespan | Trust versus mistrust, autonomy versus shame, initiative versus guilt, industry versus inferiority, identity versus role confusion, intimacy versus isolation, generativity versus stagnation, integrity versus despair | The framework for thinking about adolescent identity, midlife crisis and the tasks of ageing |
| Humanistic (Rogers, Maslow) | People have an innate actualising tendency and become disordered when conditions of worth override their own valuing | Self-concept, incongruence, unconditional positive regard, the fully functioning person | The relational conditions of effective therapy — empathy, warmth and genuineness — predict outcome across modalities |
| Social-cognitive (Bandura, Rotter, Mischel) | Behaviour arises from person, environment and behaviour interacting; cognition mediates both | Self-efficacy, reciprocal determinism, locus of control, if-then behavioural signatures | The basis of self-management, relapse-prevention and CBT's emphasis on efficacy expectations |
Temperament completes the picture. Thomas and Chess's New York Longitudinal Study identified easy, difficult and slow-to-warm patterns, and the dimension most relevant to psychiatry is behavioural inhibition — the tendency to withdraw and be vigilant toward novelty — which predicts later social anxiety with moderate strength. Inhibition is observable in infancy, is heritable, and is one of the clearest examples of a personality-relevant disposition present before any learning could have occurred.
Common mistakes
Watch outSaying 'heritability is 60 percent' and meaning that 60 percent of a person's intelligence came from their parents. It means 60 percent of the variance observed in that population in that environment is associated with genetic differences between individuals. It says nothing about any one person, and the figure changes when the environment changes.
Watch outTreating a full-scale IQ as an explanation rather than a measurement. The IQ does not cause anything; it summarises correlations between tasks. If you find yourself explaining a patient's poverty of speech by their IQ, you have inverted the direction of inference.
Watch outUsing the Myers-Briggs type inventory as if it were a clinical instrument. It has poor test-retest reliability and no demonstrated predictive validity for occupational or clinical outcomes; it is a culture-fair-sounding, psychometrically weak questionnaire, and clinical decisions about capacity or risk must not rest on it.
Check your understanding
A 30-year-old with a lifelong history of reading difficulty scores at the 20th percentile on block design and vocabulary but at the 70th on matrix reasoning. How should this be interpreted?
How does a high score on the neuroticism dimension differ from borderline personality organisation, given that both predict affective instability?
Why is adaptive behaviour required before a diagnosis of intellectual disability can be made from an IQ score alone?
Summary and next
Intelligence is a construct summarising real, predictive correlations between cognitive tasks, with a general factor, broad fluid and crystallised abilities, and narrow task-specific skills; deviation IQs locate a person in a normal distribution with mean 100 and standard deviation 15, and are only as meaningful as the standardisation, reliability, validity and fairness of the instrument. Personality is described best by trait dimensions such as the Big Five, enriched by psychodynamic, psychosocial and social-cognitive accounts. Both constructs describe variation between people, not a hierarchy of worth.
Next: the foundations of psychiatry. The behavioural sciences end here and clinical psychiatry begins — with classification, the mental status examination, and the interview that produces the diagnosis.
Multiple choice
If a Wechsler Full Scale IQ is a deviation IQ with a mean of 100 and a standard deviation of 15, approximately what proportion of the population scores between 85 and 115?
Which statement about crystallised (Gc) and fluid (Gf) intelligence is correct?
Which finding most directly supports the conclusion that environment contributes substantially to measured intelligence scores?
Which Big Five trait shows the strongest and most replicated association with vulnerability to anxiety and depressive disorders?
A patient scores 67 on an intelligence test but manages her own medication, works in a supermarket, and maintains a bank account. What is the correct conclusion?