Understanding Intelligence and Brain Function

Understanding Intelligence and Brain Function

Linas Juozenas
Skip to the article

Intelligence Unleashed · Foundations

Intelligence & the brain a modern guide to how minds learn, reason and adapt

Intelligence is not a substance hidden in one brain region, and an IQ score is not a complete measure of a person. Human cognitive ability emerges from many interacting systems: perception, memory, attention, knowledge, motivation, emotion, bodily state, culture and opportunity. This guide builds a careful map of that complexity without losing sight of what researchers can measure reliably.

A construct, not an objectResearchers infer intellectual abilities from patterns of performance. There is no single physical unit called intelligence.
Networks, not isolated centresSpecialised brain areas matter, but complex thought depends on communication across distributed cortical and subcortical systems.
Capacity meets contextDevelopment, health, education, language, experience and testing conditions all shape what a person can express at a given moment.

The field in one sentence

Intelligence is measurable, multidimensional and never context-free

People who perform well on one demanding cognitive task tend, on average, to perform well on others. That positive pattern is one of psychology’s most replicated findings. Yet it does not make every ability identical, reveal one biological cause, or turn a standardised score into a verdict on human potential.

A good introduction therefore holds two truths together: general cognitive differences are real and useful to study, and every measurement samples only part of a developing person’s abilities under particular conditions.

i

Educational context

This guide explains research concepts; it is not a clinical assessment or a substitute for educational, psychological or medical advice. For cognitive concerns, developmental questions, disability determinations and high-stakes testing, consult appropriately qualified professionals who use validated methods and consider the person’s wider history.

01 · Set the compass

Three distinctions prevent most confusion

Everyday language bundles together aptitude, learning, judgement, creativity and character. Science has to separate them before it can ask how they relate.

Knowledge

What has been learned

Facts, concepts, vocabulary, procedures and culturally organised expertise. Knowledge changes what problems a person can recognise and which solutions are available.

Intellectual ability

How information is handled

Reasoning, detecting relationships, learning, maintaining goals, solving unfamiliar problems and using prior knowledge. These capacities overlap but are not interchangeable.

Wisdom

How judgement serves a life

Balancing evidence, values, uncertainty, other people’s needs and long-term consequences. Intelligence can support wisdom, but clever reasoning alone does not guarantee ethical judgement.

A fourth distinction is equally important: maximum performance is not typical behaviour. A timed reasoning test asks what someone can do under standard instructions. Daily decisions also depend on interest, effort, emotion, habits, health, resources, goals and cooperation. A person can know the better course and still not take it; another may solve a practical problem without resembling a conventional test-taker.

≠

Intelligence is not human worth

An IQ or general-ability score does not establish a person’s kindness, dignity, rights, loyalty, courage, imagination, moral standing or value. These lie outside the score’s intended interpretation.

02 · Define the question

There is no single universally accepted essence of intelligence

Definitions usually converge on learning, reasoning, problem-solving and adaptation, while disagreeing about boundaries, structure and the role of culture.

In scientific work, intelligence is best treated as a family of related capacities that help an organism acquire and use information, pursue goals and adapt when familiar routines are insufficient. A classic multidisciplinary task-force report emphasised both the importance of measured cognitive abilities and the major questions that remained unsettled.1 Later reviews make the same basic point: intelligence research has strong empirical regularities, but no definition captures every legitimate use of the word.2

Scroll horizontally to compare →
One word, several research questions
Level Typical question What the answer can establish What it cannot establish alone
Task performance How accurately or quickly can a person solve this kind of problem? Observable performance under defined conditions. A complete, context-free capacity or permanent limit.
Psychometric structure Which tasks correlate, and which underlying statistical dimensions summarise those correlations? Patterns of individual differences across a sample. That a statistical factor is a single biological mechanism.
Cognitive process Which operations—attention, retrieval, updating, strategy—produce success? A functional account that can be tested experimentally. That one process explains every form of intelligent behaviour.
Brain mechanism Which cells, circuits and networks support the operations? Biological constraints and candidate mechanisms. Meaning, value or causation from an activation map alone.
Development and ecology How do abilities emerge in particular bodies, families, schools and cultures? How opportunity and adaptation unfold over time. A universal ranking independent of lived conditions.

A definition is a research tool. It should clarify what is being measured—not quietly turn one useful measure into the whole meaning of a mind.

Orientation principle

03 · Read the pattern

Why psychologists find both general and specific abilities

Different cognitive tests usually correlate positively, yet clusters of similar tasks also share something more specific.

If a large, varied group takes tests of vocabulary, spatial reasoning, memory, processing speed and novel problem-solving, their results are not independent of one another. Better performance in one domain tends to accompany better performance in others. Researchers call this the positive manifold. A general factor, g, summarises the variance shared across the battery. It is a statistical dimension—not an organ, a mental fuel or proof that one cause drives every task.

Even the positive manifold can have more than one theoretical explanation. One proposal is that mutually beneficial cognitive processes strengthen one another during development, allowing a general pattern to emerge without a single master mechanism.3

General

Shared performance

g captures what diverse cognitive tests have in common within a data set. Its exact interpretation depends on the model and measures.

Broad

Families of ability

Reasoning, acquired knowledge, visual and auditory processing, working memory, retrieval fluency and processing speed can form distinguishable broad domains.

Narrow

Task-relevant skills

Specific relations, word knowledge, spatial scanning, phonetic coding, memory span and many other narrower abilities explain additional differences.

The Cattell–Horn–Carroll family of models is an influential hierarchical taxonomy that organises many narrow and broad abilities, often beneath a general factor.4 It should not be reduced to only two boxes, but its best-known distinction is useful:

Scroll horizontally to compare →
Selected broad abilities—not separate kinds of people
Ability Plain-language description Example demand Important caution
Fluid reasoning Finding relations, inferring rules and solving unfamiliar problems. Discovering the missing element in a novel pattern. It is not “purely innate”; strategy, schooling and cultural familiarity still matter.
Comprehension–knowledge Breadth and depth of acquired language and culturally organised knowledge. Explaining a word or applying a learned concept. Knowledge is not mere storage; it changes reasoning and what can be noticed.
Working memory capacity Maintaining and updating information while pursuing a task. Following multi-step instructions while transforming intermediate results. Scores depend on attention, strategy and task format, not one mental container.
Processing speed Efficient performance of simple or overlearned cognitive operations. Rapidly matching symbols under a time limit. Slow speed does not imply shallow understanding, and motor or visual demands can affect scores.
Visual processing Analysing, transforming and remembering visual–spatial patterns. Mentally rotating an object. A visual task may still require language, working memory and learned strategies.
Long-term storage and retrieval Learning information and later retrieving it fluently through associations. Generating examples from a category or recalling paired items. Retrieval failure is not the same as knowledge never having been learned.
i

Factors describe patterns of correlation

Factor analysis helps organise correlations among test results. Choosing among hierarchical, bifactor, network and process models requires further theory and evidence. A factor’s name does not explain how the brain produces the observed pattern.

04 · Broaden without blurring

Multiple intelligences, successful intelligence and emotional skill

Several influential theories illuminate abilities that ordinary IQ batteries sample incompletely. They do not all carry the same kind of evidence.

Gardner

Multiple intelligences

The theory highlights linguistic, logical–mathematical, spatial, musical, bodily–kinaesthetic, interpersonal, intrapersonal and naturalistic capacities. It encouraged educators to notice varied strengths. Direct testing, however, found substantial correlations among proposed domains and a large role for general ability.5

Sternberg

Successful intelligence

The theory frames analytical, creative and practical abilities around adapting to, shaping and selecting environments.7 The lens is broader than conventional test content, but evidence that its three abilities are psychometrically independent is contested.8

Emotion

Emotional intelligence

Ability measures ask people to perceive, understand and manage emotional information; trait and mixed questionnaires ask how people see their own tendencies and competencies. These approaches overlap only partly and should not be treated as one universal “EQ”.9

Multiple intelligences are not learning styles. Gardner has explicitly rejected the common idea that every learner should be labelled “visual”, “auditory” or another type and taught only through that channel.6 Good teaching may use several representations because the subject benefits from them—not because a questionnaire has discovered a fixed neural style.

Emotional abilities matter for relationships, teamwork and learning. A meta-analysis found a modest association between emotional-intelligence measures and academic performance, with ability-based and mixed measures differing in what they captured.10 That is far from the popular claim that “EQ determines 80 or 90 per cent of success”. There is no defensible universal percentage, and outcomes reflect opportunity, personality, knowledge, health, values, support and chance as well as cognitive skills.

✓

Use theories as lenses, not identity boxes

A child is not “a musical intelligence”, an adult is not “left-brained”, and a questionnaire does not reveal a fixed learning type. Profiles can guide questions and support; they should not narrow what someone is allowed to learn.

05 · Think in networks

The brain combines specialisation with integration

Complex cognition is neither located in one tiny centre nor produced by every region doing the same work.

The nervous system is organised across scales: molecules and synapses, cells and local circuits, long-range fibre tracts, subcortical loops and large-scale functional networks. Some regions are relatively specialised; complex behaviour emerges when specialised operations are coordinated. Network neuroscience studies how activity separates into specialised systems and integrates across them.1112

Scroll horizontally to compare →
Useful landmarks without “one region, one function” mythology
System Selected contributions Oversimplification to avoid
Cerebral cortex Specialised sensory and motor territories plus distributed association systems supporting language, perception, memory and flexible behaviour. “The sole seat of higher thought.” Cortex depends on subcortical and cerebellar interactions.
White matter White matter contains many myelinated axons linking nearby and distant regions; its organisation constrains network communication. “Passive wiring.” Axons, glia and myelin can change with development and experience.62
Thalamus Multiple nuclei route and regulate signals and help coordinate cortical dynamics, attention and arousal. “A simple sensory relay.”
Hippocampal formation Relational and contextual learning, episodic recollection and navigation, supported through interactions with distributed cortical systems. “Converts short-term memory into one permanent long-term store.”
Amygdala Heterogeneous circuits involved in learning about threat, reward, valence and behavioural relevance. “The fear centre” or a universal emotional stamp.
Basal ganglia Interconnected loops contributing to action selection, reinforcement learning, habits, movement and cognition. “A drawer that stores procedural memories.”
Cerebellum Timing, prediction, calibration and learning across motor, cognitive and affective domains. “Only a balance and movement structure.”
Brainstem and hypothalamus Arousal, sleep–wake regulation, autonomic control and homeostasis that make cognition possible. “Background machinery unrelated to intelligence.”

Functional brain images are powerful but indirect. fMRI usually tracks changes in blood oxygenation associated with neural activity, and an activated region may participate in many processes. Inferring a specific thought merely because an area is active is a probabilistic “reverse inference”, not an automatic conclusion.1314

Research on intelligence therefore increasingly examines efficiency, flexibility and reconfiguration across networks rather than searching for a single intelligence centre.15 In 831 healthy young adults from the Human Connectome Project, with a conceptual replication in 145 participants, a 2026 study associated a modelled general factor with distributed structural and resting-state network features. Replication across ages and populations is still needed; this is not an individual prognostic scan.16

06 · Assemble the toolkit

Cognitive functions cooperate, compete and compensate

Researchers separate functions to study them, but real tasks recruit several at once.

Attention

Selecting what gains priority

Alerting, orienting, sustained attention, selective attention and executive control are partly distinguishable. Cortical and subcortical networks bias processing towards current goals and behaviourally relevant events.21

Working memory

Keeping useful information available

Working memory supports temporary maintenance and transformation for an ongoing task. It relies on distributed, dynamic activity and connectivity—not a single box for short-term memory.18

Long-term memory

Many systems, not one archive

Episodic events, semantic knowledge, skills, habits, priming and conditioning depend on partly distinct, interacting systems. The hippocampal formation is especially important for rapid relational and episodic learning. Long-term memories depend on distributed systems.17

Executive control

Holding goals and regulating action

Inhibition, updating and shifting show both unity and diversity.19 Prefrontal regions are important hubs, but control depends on interactions among frontoparietal and cingulo-opercular networks and subcortical regions—not a solitary mental chief executive.20

Perception

Interpreting constrained evidence

Perception is active and recurrent: sensory evidence is combined with context, expectations and task goals. Expectations can facilitate or bias processing, but perception is not unconstrained imagination.22

Language and metacognition

Representing, communicating and monitoring

Language lets people compress, share and transform knowledge. Metacognition—monitoring what one knows and regulating strategy—can improve learning even when raw task capacity is unchanged.

!

Every cognitive task is “impure”

A working-memory task also requires seeing or hearing, understanding instructions, sustaining attention and producing a response. A low result can arise through different pathways. This is why professional assessment uses converging evidence rather than inferring an underlying ability from one score.

07 · Understand change

Neuroplasticity is lifelong, constrained and value-neutral

The nervous system can reorganise, but “the brain changes” does not tell us whether the change is large, durable, transferable or beneficial.

Plasticity includes changes in synaptic strength, inhibition, intrinsic excitability, circuit structure, myelination, glial regulation and network recruitment. Modern accounts therefore go far beyond the metaphor of merely turning synaptic “weights” up or down.23242562

Experience-dependent

Repeated demands shape systems

Practice can make representations more efficient, refine coordination and alter which strategies are recruited. Change is usually strongest for the practised skill and related tasks.

Constrained

Biology and history matter

Age, prior learning, injury, sleep, stress, health, task difficulty and available support influence how much adaptation is possible and how long it takes.26

Not automatically good

Brains also learn unhelpful patterns

Persistent pain, addiction, fear generalisation and maladaptive compensation also involve plasticity. The word describes the capacity for change, not improvement by definition.25

Claims that a few weeks of music, gaming or meditation “thicken the cortex” need careful reading. Structural MRI detects changes in image-derived measures; it does not by itself identify which cellular process caused them. Small samples, flexible analyses and short follow-up periods can exaggerate certainty.27 One proposed account is expansion during skill acquisition followed by partial renormalisation as representations and circuits are refined—so “bigger” is not a universal synonym for “better”.28

Plasticity means that experience matters. It does not mean that every limit disappears, every intervention works, or every neural change transfers to unrelated abilities.

Learning principle

08 · Follow trajectories

Cognition develops on overlapping timelines

There is no single birthday that marks full cognitive maturity, and no single age at which every ability peaks.

Stage theories such as Piaget’s remain historically important because they drew attention to qualitative changes in children’s reasoning. Modern evidence shows more variability by task, knowledge, culture, instruction and testing method than a rigid four-stage timetable suggests. Development is better pictured as interacting trajectories with periods of rapid change, stability, refinement and compensation.

Scroll horizontally to compare →
Broad developmental patterns—not deadlines or diagnoses
Period Common changes Context that matters Myth to leave behind
Infancy and early childhood Rapid sensory, motor, language, memory and social learning; growing ability to represent absent objects, intentions and rules. Responsive interaction, nutrition, sleep, safety, language exposure, play and access to care. That competence appears only when a classic task first shows it.
Middle childhood Expanding knowledge, strategy use, metacognition, attention control and academic skills. Quality of instruction, practice, language, expectations, disability support and belonging. That one early score fixes the adult outcome.
Adolescence Continued refinement of learning, reward, social and control systems; greater capacity for abstract and self-directed thought. Peers, autonomy, sleep timing, stress, opportunity and meaningful challenge. That adolescents simply have an “unfinished prefrontal cortex”. Adolescence can be a period of opportunity as well as vulnerability.29
Adulthood Expertise and knowledge can deepen while some measures of processing speed or novel problem-solving change gradually. Education, occupation, health, caregiving, stress, continued learning and social participation. That the brain becomes fixed after childhood.
Later life Highly individual mixtures of stability and change across speed, memory, reasoning, vocabulary and expertise. Cardiovascular and sensory health, disease, activity, support, education and accumulated knowledge. That ageing is either uniform decline or guaranteed wisdom.

The age of 25 is not a universal scientific milestone at which “the brain finishes developing”. Different structural and functional measures follow different courses, samples vary, and no single imaging measure defines adult judgement.30 Large pooled lifespan MRI datasets model multiple non-linear age-associated trajectories rather than one finish line.31

Likewise, there is no one cognitive peak. In largely cross-sectional datasets, processing-speed measures tended to peak relatively early, vocabulary later, and performance on one emotion-perception task later still. These group-average age patterns are not fixed within-person timetables. Even within memory and reasoning, different tasks peak at different ages.32

Cognitive reserve is a theoretical construct referring to the adaptability—efficiency, capacity or flexibility—of cognitive processes that may help explain why cognition differs despite similar age-related brain change or pathology. Education and occupational complexity are imperfect proxies, not reserve itself, and reserve is not immunity from disease.33

09 · Replace the false contest

Genes and environments shape development together

“Nature versus nurture” is the wrong contest: inherited differences can influence exposure to and responses to environments, and environments influence which possibilities develop.

Heritability is a population statistic: it estimates how much variation in a measured trait, in a specified population and environment, is associated with genetic differences. It does not say what percentage of one person’s intelligence is genetic, whether a trait can change, or why average differences between groups exist.34

Age-dependent

Estimates can change across development

A large twin study in high-income settings estimated that the heritability of general cognitive ability increased from childhood to young adulthood.35 This may partly reflect people selecting and shaping environments in ways that correlate with their dispositions.

Highly polygenic

No single “intelligence gene”

Cognitive-test performance is highly polygenic: very many variants show tiny average statistical associations, while most causal variants and mechanisms remain unresolved. In a 2018 study of 269,867 participants, a polygenic score explained up to 5.2% of variation across four independent samples.36

Context-bound

Associations include social pathways

Population genomic estimates can mix direct biological effects with ancestry, family and demographic structure. In the cited study, within-sibship estimates were attenuated for cognitive ability and several other phenotypes, showing that population-based genomic associations can include indirect and demographic components.37

Scroll horizontally to compare →
What common findings mean—and do not mean
Finding Responsible interpretation Incorrect leap
A trait is substantially heritable Genetic differences account for part of observed variation in that population and context. “The trait is fixed” or “support cannot help”.
A polygenic score predicts an outcome In a studied population, a weighted set of variants carries limited statistical information. “DNA reveals an individual’s destiny” or “the score transfers equally across ancestries and environments”.
Socioeconomic conditions moderate estimates Opportunity may alter how individual differences are expressed; results vary by country and design. “The same interaction is universal.” A meta-analysis found different patterns in the United States and Western Europe/Australia.38
Schooling changes test performance Education is one of the best-supported environmental influences on measured cognitive ability. “Every extra year guarantees the same gain.” Across three quasi-experimental designs, a meta-analysis reported estimates of about 1–5 IQ points per extra year of education (combined estimate about 3.4), with estimates varying by study design and outcome. This is an average estimate, not a guaranteed individual gain.39
Exposure predicts poorer cognition Some biological and social conditions can constrain development; dose, timing and confounding must be examined. “Every correlation proves an irreversible brain effect.” Lead is a notable preventable neurotoxicant with prospective evidence even at comparatively low exposure.40

Epigenetic mechanisms regulate gene activity and can respond to development and exposure. In human studies, however, methylation measured in blood may be a marker, consequence or correlate rather than a cause in the brain. Tissue specificity, confounding and reverse causation make “experience switches intelligence genes on or off” far too simple.41

↔

Genes participate in pathways; they do not write a finished score

Development is transactional. Children influence, select and experience environments partly in relation to their characteristics; families, schools and societies distribute opportunities unequally; biological systems respond. The result is neither gene-only nor environment-only.

10 · Measure with purpose

An IQ score is an estimate, not an essence

Good assessment begins by identifying the decision to be informed, then asks whether the scores support that specific interpretation and use.

Modern IQ scores are usually age-normed standard scores, often centred on 100 with a standard deviation of 15. They are no longer a literal “mental age divided by chronological age” ratio. A score locates performance relative to the test’s norm group and comes with measurement uncertainty.42

Standardisation

Same procedure

Administration and scoring follow defined rules so results can be compared meaningfully.

Reliability

Consistent enough

Scores should not fluctuate so much that the intended interpretation becomes unstable.

Validity

Evidence for the use

Evidence must support the interpretation and decision in the relevant population—not merely a claim that “the test is valid”.

Fairness

Comparable access and meaning

Language, disability, culture, opportunity and consequences must be examined, not assumed away.

The joint testing standards of the American Educational Research Association, American Psychological Association and National Council on Measurement in Education treat validity as evidence supporting intended score interpretations and uses, with fairness integrated throughout testing practice.42

Scroll horizontally to compare →
Read a score as one line of evidence
A test may help estimate It does not directly measure Questions before use
Performance in reasoning and acquired knowledge Every practical, creative, social or emotional capability. Does the battery adequately sample the abilities relevant to this decision?
Relative strengths and weaknesses A perfectly stable neural profile. Are index differences reliable, unusual and supported by history and observation?
Current performance against age norms A fixed ceiling on future learning. Are the norms current and appropriate for age, language and population?
A probabilistic estimate of later performance or risk, calibrated from group data An individual’s guaranteed educational, work or health outcome. How well calibrated is the estimate for the relevant population, and what other information changes the decision?
Possible need for further assessment or support A diagnosis from IQ alone. Have adaptive functioning, development, schooling, sensory or motor access and clinical context been assessed?60

Why the same person can receive different results

  • Every observed score contains measurement uncertainty; reports should use an interval estimate appropriate to the test model and intended decision.
  • Practice and familiarity can raise retest performance, especially over shorter intervals.43
  • Sleep loss, pain, illness, anxiety, medication, distraction, effort and sensory or motor barriers can suppress expression of ability.
  • Translation is not enough: items and norms must function appropriately across languages and cultures.
  • Rank-order stability is low in preschool years, rises rapidly through childhood and is generally high from late adolescence to late adulthood; it also declines as the retest interval grows. Stable ranking does not imply an immutable score.44

Scores from appropriately validated cognitive tests show probabilistic associations with some educational and occupational outcomes; they do not determine an individual outcome. A meta-analysis found a substantial relation between intelligence measures and school grades, but much variation remained unexplained.45 In employee selection, updated analyses have revised older, often repeated validity estimates downward and reinforced the need to combine appropriate methods rather than crown one universal predictor.46

Average test performance also shifts across generations—the Flynn effect—and its size and direction vary across countries, periods and ability domains.47 Regular renorming is therefore essential. A score of 100 means average relative to a particular reference group at a particular time; it is not a timeless quantity of brain power.

11 · Read the electrical weather

Brain rhythms can coordinate activity; bands are not mental labels

Delta, theta, alpha, beta and gamma are convenient frequency ranges. None corresponds to one exclusive emotion, ability or level of consciousness.

Neural populations fluctuate at many timescales. Oscillatory timing can help organise communication, excitability and plasticity within and between circuits.48 Scalp EEG records mixtures of these dynamics through the skull, together with signals from the eyes and muscles, movement artefacts and electrical noise. Published band definitions vary by laboratory; individual peak frequencies vary with age, region, state and task.63

Scroll horizontally to compare →
Conventional bands are descriptive ranges, not five modes of mind
Common label Approximate range Examples of observed associations What not to conclude
Delta About 0.5–4 Hz Prominent slow activity during deep non-REM sleep; slower task-related and pathological activity also exists. That all delta is restorative or releases a particular hormone.
Theta About 4–8 Hz Scalp theta often appears with drowsiness; frontal-midline theta is associated with control demands in some tasks. Hippocampal theta-like rhythms are linked to memory and navigation, especially in intracranial and animal recordings.6465 That theta universally means creativity, meditation or subconscious access.
Alpha About 8–13 Hz Often strong over posterior scalp with relaxed eyes closed; also implicated in selective inhibition and attention. That alpha is simply “relaxation” or a direct measure of calmness.
Beta About 13–30 Hz Sensorimotor maintenance and change, prediction, cognition and transient bursts in several systems. That more beta always means better focus.
Gamma About 30–80 Hz (definitions vary) Local sensory and cognitive processing and synchronisation; activity above this range is often called high gamma and may include broadband population activity rather than a narrow oscillation.66 That gamma proves peak performance, unity of consciousness or superior intelligence.

Most associations between a frequency band and a mental operation are correlational; experimental manipulation, converging methods and precise timing are needed to argue for a causal role.49 Power spectra also contain a non-rhythmic (aperiodic) background component. Separating rhythmic peaks from this background helps prevent a change in the spectrum’s overall slope from being mistaken for a change in a named frequency band.50

High-frequency scalp activity is especially vulnerable to muscle contamination. Electrical activity from facial and neck muscles can resemble beta or gamma changes, so an impressive colour map is not automatically a clean neural measurement.51

≈

Neurofeedback is a learning protocol, not a frequency dial

Neurofeedback gives a person information linked to a measured signal with the aim of helping them learn to alter it. Effects specific to the feedback must be distinguished from expectations, coaching, engagement and repeated practice. Consensus reporting standards call for preregistration, adequate controls, blinding where possible and evidence that participants actually learned to regulate the targeted signal.52

12 · Turn evidence into practice

Support the system instead of chasing a “smarter brain” switch

The most dependable gains usually come from learning meaningful knowledge and skills under conditions that make attention, practice and recovery possible.

  1. Practise the skill you want to develop

    Transfer is usually strongest near the trained task. Commercial brain games can improve performance on practised or similar exercises, but evidence for broad, durable gains in general intelligence or everyday competence is weak.53

  2. Retrieve; do not just reread

    Trying to recall, explain or solve before looking at the answer strengthens later access to knowledge and exposes gaps. Retrieval practice has robust classroom evidence when aligned with the material and used without punitive pressure.55

  3. Space encounters over time

    Distributing study time usually supports longer retention than concentrating the same amount of study in one session. The useful interval depends on when the knowledge will be needed.5467

  4. Build knowledge and vary application

    Well-organised background knowledge can reduce working-memory demands in familiar domains and gives new ideas a framework to connect with. Once a method is understood, comparing examples and practising in varied contexts helps reveal when it applies.

  5. Protect sleep and recovery

    Sleep supports attention, learning and memory consolidation across development; insufficient or mistimed sleep can impair performance. It is not accurate to assign one memory type or “intelligence frequency” to one sleep stage.56

  6. Support whole-body health and access

    Physical activity has broad health benefits and is associated with cognitive and brain outcomes, though effect sizes and causal estimates vary by age, intervention and analysis.57 Addressing hearing, vision, pain, nutrition, illness and environmental barriers may remove constraints on performance more directly than a cognitive app.

  7. Use reflection and feedback

    Set a concrete goal, attempt the task, obtain informative feedback, identify the error, adjust the strategy and try again. Feedback works best when it guides the next action rather than merely labelling the learner.

Meditation practices aim to train attention and emotion regulation and may benefit some people, but studies vary widely in the practices examined, outcomes measured and methodological quality. Structural-imaging meta-analyses report associations, not a guaranteed anatomical upgrade.58 A major methodological review cautioned against clinical and neuroscientific hype while encouraging better-powered, more precise research.59

+

The practical target is capability, not a number

A person may improve by knowing more, choosing a better representation, automating a component skill, using an external aid, collaborating, reducing a barrier or redesigning the environment. All can increase real-world capability without pretending that intelligence is one muscle.

13 · Keep the person larger than the model

Measurement creates responsibilities

A cognitive classification can open access to support—or become a lasting barrier after its uncertainty has been forgotten.

Purpose

Test only when the result can help

Define the question first. Curiosity does not justify unnecessary data collection, and a score should not be used for decisions it was not validated to inform.

Proportionality

Match evidence to consequence

The higher the stakes, the stronger the need for current norms, multiple sources, qualified interpretation, accommodations and a route to review or appeal.

Agency

Explain and involve the person

People should know why they are being assessed, what information will be collected, who can access it, how uncertainty is represented and how results can support—not replace—their voice.

Neurodiversity changes the question

Variation in attention, language, sensory processing, social communication, memory or speed is not captured well by a single “more versus less intelligent” axis. A useful profile asks what a person can do, where barriers arise, which supports work and what the person values. It avoids romanticising disability while also avoiding the assumption that difference is only a deficit.

Group comparisons require exceptional care

Mean score differences do not identify their causes. Many factors may matter: whether a test measures the same construct across groups, sampling, unequal opportunity, health, discrimination, language, schooling, migration, stereotype threat and historical conditions. Heritability within groups cannot explain differences between group means. Treating a group mean as though it were an individual’s score is an invalid group-to-individual inference.

Prediction is not permission

A model that predicts educational difficulty may justify offering support; it does not justify withholding ambitious teaching. A hiring test with predictive validity still raises questions about job relevance, alternatives, adverse impact, disability access and whether the organisation is treating applicants fairly. Accuracy is only one ethical consideration.

The responsible purpose of intelligence science is not to decide whose mind matters. It is to understand variation well enough to improve learning, access, autonomy and participation.

Human-centred principle

14 · Inspect the claim

A ten-question check for claims about intelligence

Use this before trusting a headline, product, brain scan, genetic score or online quiz.

  • Definition: What exactly does “intelligence”, “memory” or “focus” mean here?
  • Measure: Which task or instrument produced the number, and is it validated for this use?
  • Comparison: With which group, under which conditions and against which current norm was performance compared?
  • Sample: How many participants were studied, and what were their ages, languages, cultures and health backgrounds?
  • Design: Was the study correlational, longitudinal, genetically informative or randomised—and does its design support causal inference?
  • Magnitude: Is the effect practically meaningful, not merely statistically detectable?
  • Uncertainty: Are confidence intervals, measurement error, missing data and alternative explanations visible?
  • Transfer: Did improvement extend beyond the practised test to durable everyday capability?
  • Replication: Has an independent team found a similar result in a relevant population?
  • Consequence: Who benefits, who may be excluded, and can the person understand, challenge or revisit the decision?
?

Beware the compressed headline

“Region X controls intelligence”, “gene Y determines IQ”, “gamma creates genius” and “this game makes you smarter” each collapse a chain of assumptions. Good science names the task, the people, the method, the effect size and the boundary of the conclusion.

15 · Frequently asked questions

Short answers to difficult questions

Each answer is a starting point; later articles in this series examine the domains in depth.

Is intelligence one ability or many?

Both descriptions capture part of the evidence. Diverse cognitive tasks correlate, supporting a general factor, while broad and narrow abilities explain meaningful additional variation. Theories disagree about the causal structure behind that pattern.

Does IQ measure intelligence?

A well-constructed IQ battery estimates several important cognitive abilities and summarises some of them. It does not measure every form of practical, creative, emotional or moral functioning, and its meaning depends on norms, conditions and intended use.

Can IQ change?

Yes. People often remain in a similar rank order relative to one another, especially over shorter intervals in adulthood, but scores can change through development, education, health, major environmental change, practice and measurement noise. A single change should be interpreted with its confidence interval and context.

Are intelligence differences genetic?

Genetic differences contribute to variation within studied populations, but the contribution is highly polygenic and develops through interaction with environments. Heritability is not an individual percentage, a fixed ceiling or an explanation of group differences.

Do people use only 10 per cent of their brains?

No. Different systems vary in activity by moment and task, and not every neuron fires simultaneously—nor should it—but the healthy brain does not contain a vast, functionless 90 per cent waiting to be unlocked.

Are people left-brained or right-brained?

Some functions are lateralised, and the hemispheres are not identical. Ordinary reasoning, creativity, language and emotion nevertheless depend on bilateral networks communicating through major pathways. A two-type personality label does not follow from hemispheric specialisation.

Can brain waves reveal thoughts or intelligence?

EEG can reveal time-varying electrical patterns related to arousal, sensory processing, movement preparation, sleep and tasks. Broad bands do not translate directly into specific thoughts, personality or an intelligence score. Claims about decoding thoughts or intelligence require task-specific validation and rigorous artefact control.

Does a larger brain mean greater intelligence?

Total brain volume shows a small-to-moderate positive population-level association with cognitive-test scores (about r = .24 in a large synthesis), explaining only a small share of individual differences.61 Body size, development, tissue properties, connectivity, network dynamics and measurement choices matter, and a scan cannot rank minds reliably.

What is the best way to become more capable?

Build relevant knowledge, practise the real skill, use retrieval practice and space learning over time, seek feedback, protect sleep and health, use tools and collaborate. These strategies improve capability without promising a universal increase in one hidden quantity.

When should someone seek a professional assessment?

A professional assessment may be useful when cognitive, learning, developmental or memory concerns interfere with daily life, education or work, or when formal support depends on documentation. Sudden confusion or trouble speaking can be a sign of stroke; seek emergency medical help immediately.68

Evidence map

Sources and further reading

The bibliography prioritises peer-reviewed research, professional standards and first-party academic material. It supports the claims above; it is not a ranking of every theory or a substitute for a systematic review of each subfield.

Link behaviour: External source links open in a new tab.

  1. Neisser, U. et al. (1996), “Intelligence: Knowns and Unknowns,” American Psychologist, 51(2), 77–101. DOI record.
  2. Deary, I. J. (2012), “Intelligence,” Annual Review of Psychology, 63, 453–482. DOI record.
  3. van der Maas, H. L. J. et al. (2006), “A Dynamical Model of General Intelligence: The Positive Manifold of Intelligence by Mutualism,” Psychological Review, 113(4), 842–861. DOI record.
  4. McGrew, K. S. (2009), “CHC Theory and the Human Cognitive Abilities Project: Standing on the Shoulders of the Giants of Psychometric Intelligence Research,” Intelligence, 37(1), 1–10. DOI record.
  5. Visser, B. A., Ashton, M. C. & Vernon, P. A. (2006), “Beyond g: Putting Multiple Intelligences Theory to the Test,” Intelligence, 34(5), 487–502. DOI record.
  6. Gardner, H. (2013), “Multiple Intelligences Are Not Learning Styles,” Harvard Graduate School of Education. Author’s explanation.
  7. Sternberg, R. J. (1999), “The Theory of Successful Intelligence,” Review of General Psychology, 3(4), 292–316. DOI record.
  8. Chooi, W.-T., Long, H. E. & Thompson, L. A. (2014), “The Sternberg Triarchic Abilities Test (Level-H) Is a Measure of g,” Journal of Intelligence, 2(3), 56–67. DOI record.
  9. Robinson, M. D. (2024), “Ability-Related Emotional Intelligence: An Introduction,” Journal of Intelligence, 12(5), 51. DOI record.
  10. MacCann, C. et al. (2020), “Emotional Intelligence Predicts Academic Performance: A Meta-Analysis,” Psychological Bulletin, 146(2), 150–186. DOI record.
  11. Bassett, D. S. & Sporns, O. (2017), “Network Neuroscience,” Nature Neuroscience, 20, 353–364. DOI record.
  12. Pessoa, L. (2014), “Understanding Brain Networks and Brain Organization,” Physics of Life Reviews, 11(3), 400–435. DOI record.
  13. Poldrack, R. A. (2006), “Can Cognitive Processes Be Inferred from Neuroimaging Data?” Trends in Cognitive Sciences, 10(2), 59–63. DOI record.
  14. Logothetis, N. K. (2008), “What We Can Do and What We Cannot Do with fMRI,” Nature, 453, 869–878. DOI record.
  15. Barbey, A. K. (2018), “Network Neuroscience Theory of Human Intelligence,” Trends in Cognitive Sciences, 22(1), 8–20. DOI record.
  16. Wilcox, R. R. et al. (2026), “The Network Architecture of General Intelligence in the Human Connectome,” Nature Communications, 17, 2027. DOI record.
  17. Squire, L. R. & Dede, A. J. O. (2015), “Conscious and Unconscious Memory Systems,” Cold Spring Harbor Perspectives in Biology, 7(3), a021667. DOI record.
  18. D’Esposito, M. & Postle, B. R. (2015), “The Cognitive Neuroscience of Working Memory,” Annual Review of Psychology, 66, 115–142. DOI record.
  19. Friedman, N. P. & Miyake, A. (2017), “Unity and Diversity of Executive Functions: Individual Differences as a Window on Cognitive Structure,” Cortex, 86, 186–204. DOI record.
  20. Friedman, N. P. & Robbins, T. W. (2022), “The Role of Prefrontal Cortex in Cognitive Control and Executive Function,” Neuropsychopharmacology, 47, 72–89. DOI record.
  21. Petersen, S. E. & Posner, M. I. (2012), “The Attention System of the Human Brain: 20 Years After,” Annual Review of Neuroscience, 35, 73–89. DOI record.
  22. de Lange, F. P., Heilbron, M. & Kok, P. (2018), “How Do Expectations Shape Perception?” Trends in Cognitive Sciences, 22(9), 764–779. DOI record.
  23. Magee, J. C. & Grienberger, C. (2020), “Synaptic Plasticity Forms and Functions,” Annual Review of Neuroscience, 43, 95–117. DOI record.
  24. Sancho, L., Contreras, M. & Allen, N. J. (2021), “Glia as Sculptors of Synaptic Plasticity,” Neuroscience Research, 167, 17–29. DOI record.
  25. Cramer, S. C. et al. (2011), “Harnessing Neuroplasticity for Clinical Applications,” Brain, 134(6), 1591–1609. DOI record.
  26. Lövdén, M. et al. (2010), “A Theoretical Framework for the Study of Adult Cognitive Plasticity,” Psychological Bulletin, 136(4), 659–676. DOI record.
  27. Thomas, C. & Baker, C. I. (2013), “Teaching an Adult Brain New Tricks: A Critical Review of Evidence for Training-Dependent Structural Plasticity in Humans,” NeuroImage, 73, 225–236. DOI record.
  28. Wenger, E. et al. (2017), “Expansion and Renormalization of Human Brain Structure During Skill Acquisition,” Trends in Cognitive Sciences, 21(12), 930–939. DOI record.
  29. Fuhrmann, D., Knoll, L. J. & Blakemore, S.-J. (2015), “Adolescence as a Sensitive Period of Brain Development,” Trends in Cognitive Sciences, 19(10), 558–566. DOI record.
  30. Somerville, L. H. (2016), “Searching for Signatures of Brain Maturity: What Are We Searching For?” Neuron, 92(6), 1164–1167. DOI record.
  31. Bethlehem, R. A. I. et al. (2022), “Brain Charts for the Human Lifespan,” Nature, 604, 525–533. DOI record.
  32. Hartshorne, J. K. & Germine, L. T. (2015), “When Does Cognitive Functioning Peak? The Asynchronous Rise and Fall of Different Cognitive Abilities Across the Life Span,” Psychological Science, 26(4), 433–443. DOI record.
  33. Stern, Y. et al. (2020), “Whitepaper: Defining and Investigating Cognitive Reserve, Brain Reserve, and Brain Maintenance,” Alzheimer’s & Dementia, 16(9), 1305–1311. DOI record.
  34. Visscher, P. M., Hill, W. G. & Wray, N. R. (2008), “Heritability in the Genomics Era—Concepts and Misconceptions,” Nature Reviews Genetics, 9, 255–266. DOI record.
  35. Haworth, C. M. A. et al. (2010), “The Heritability of General Cognitive Ability Increases Linearly from Childhood to Young Adulthood,” Molecular Psychiatry, 15, 1112–1120. DOI record.
  36. Savage, J. E. et al. (2018), “Genome-wide Association Meta-analysis in 269,867 Individuals Identifies New Genetic and Functional Links to Intelligence,” Nature Genetics, 50, 912–919. DOI record.
  37. Howe, L. J. et al. (2022), “Within-Sibship Genome-Wide Association Analyses Decrease Bias in Estimates of Direct Genetic Effects,” Nature Genetics, 54, 581–592. DOI record.
  38. Tucker-Drob, E. M. & Bates, T. C. (2016), “Large Cross-National Differences in Gene × Socioeconomic Status Interaction on Intelligence,” Psychological Science, 27(2), 138–149. DOI record.
  39. Ritchie, S. J. & Tucker-Drob, E. M. (2018), “How Much Does Education Improve Intelligence? A Meta-Analysis,” Psychological Science, 29(8), 1358–1369. DOI record.
  40. Lanphear, B. P. et al. (2005), “Low-Level Environmental Lead Exposure and Children’s Intellectual Function: An International Pooled Analysis,” Environmental Health Perspectives, 113(7), 894–899. DOI record.
  41. Mill, J. & Heijmans, B. T. (2013), “From Promises to Practical Strategies in Epigenetic Epidemiology,” Nature Reviews Genetics, 14, 585–594. DOI record.
  42. AERA, APA & NCME (2014), Standards for Educational and Psychological Testing. Official open-access edition.
  43. Scharfen, J., Peters, J. M. & Holling, H. (2018), “Retest Effects in Cognitive Ability Tests: A Meta-Analysis,” Intelligence, 67, 44–66. DOI record.
  44. Breit, M., Scherrer, V., Tucker-Drob, E. M. & Preckel, F. (2024), “The Stability of Cognitive Abilities: A Meta-Analytic Review of Longitudinal Studies,” Psychological Bulletin, 150(4), 399–439. DOI record.
  45. Roth, B. et al. (2015), “Intelligence and School Grades: A Meta-Analysis,” Intelligence, 53, 118–137. DOI record.
  46. Sackett, P. R. et al. (2022), “Revisiting Meta-Analytic Estimates of Validity in Personnel Selection: Addressing Systematic Overcorrection for Restriction of Range,” Journal of Applied Psychology, 107(11), 2040–2068. DOI record.
  47. Pietschnig, J. & Voracek, M. (2015), “One Century of Global IQ Gains: A Formal Meta-Analysis of the Flynn Effect (1909–2013),” Perspectives on Psychological Science, 10(3), 282–306. DOI record.
  48. Buzsáki, G. & Draguhn, A. (2004), “Neuronal Oscillations in Cortical Networks,” Science, 304(5679), 1926–1929. DOI record.
  49. Herrmann, C. S. et al. (2016), “EEG Oscillations: From Correlation to Causality,” International Journal of Psychophysiology, 103, 12–21. DOI record.
  50. Donoghue, T. et al. (2020), “Parameterizing Neural Power Spectra into Periodic and Aperiodic Components,” Nature Neuroscience, 23, 1655–1665. DOI record.
  51. Whitham, E. M. et al. (2007), “Scalp Electrical Recording During Paralysis: Quantitative Evidence that EEG Frequencies Above 20 Hz Are Contaminated by EMG,” Clinical Neurophysiology, 118(8), 1877–1888. DOI record.
  52. Ros, T. et al. (2020), “Consensus on the Reporting and Experimental Design of Clinical and Cognitive-Behavioural Neurofeedback Studies (CRED-nf checklist),” Brain, 143(6), 1674–1685. DOI record.
  53. Simons, D. J. et al. (2016), “Do ‘Brain-Training’ Programs Work?” Psychological Science in the Public Interest, 17(3), 103–186. DOI record.
  54. Cepeda, N. J. et al. (2006), “Distributed Practice in Verbal Recall Tasks: A Review and Quantitative Synthesis,” Psychological Bulletin, 132(3), 354–380. DOI record.
  55. Agarwal, P. K., Nunes, L. D. & Blunt, J. R. (2021), “Retrieval Practice Consistently Benefits Student Learning: A Systematic Review of Applied Research in Schools and Classrooms,” Educational Psychology Review, 33, 1409–1453. DOI record.
  56. Mason, G. M. et al. (2021), “Sleep and Human Cognitive Development,” Sleep Medicine Reviews, 57, 101472. DOI record.
  57. Erickson, K. I. et al. (2019), “Physical Activity, Cognition, and Brain Outcomes: A Review of the 2018 Physical Activity Guidelines,” Medicine & Science in Sports & Exercise, 51(6), 1242–1251. DOI record.
  58. Fox, K. C. R. et al. (2014), “Is Meditation Associated with Altered Brain Structure? A Systematic Review and Meta-Analysis of Morphometric Neuroimaging in Meditation Practitioners,” Neuroscience & Biobehavioral Reviews, 43, 48–73. DOI record.
  59. Van Dam, N. T. et al. (2018), “Mind the Hype: A Critical Evaluation and Prescriptive Agenda for Research on Mindfulness and Meditation,” Perspectives on Psychological Science, 13(1), 36–61. DOI record.
  60. American Association on Intellectual and Developmental Disabilities (accessed 4 September 2026), “Defining Criteria for Intellectual Disability.” AAIDD criteria.
  61. Pietschnig, J., Gerdesmann, D., Zeiler, M. & Voracek, M. (2022), “Of Differing Methods, Disputed Estimates and Discordant Interpretations: The Meta-Analytical Multiverse of Brain Volume and IQ Associations,” Royal Society Open Science, 9(5), 211621. DOI record.
  62. Sampaio-Baptista, C. & Johansen-Berg, H. (2017), “White Matter Plasticity in the Adult Brain,” Neuron, 96(6), 1239–1251. DOI record.
  63. Kane, N. et al. (2017), “A Revised Glossary of Terms Most Commonly Used by Clinical Electroencephalographers and Updated Proposal for the Report Format of the EEG Findings: Revision 2017,” Clinical Neurophysiology Practice, 2, 170–185. DOI record.
  64. Cavanagh, J. F. & Frank, M. J. (2014), “Frontal Theta as a Mechanism for Cognitive Control,” Trends in Cognitive Sciences, 18(8), 414–421. DOI record.
  65. Buzsáki, G. & Moser, E. I. (2013), “Memory, Navigation and Theta Rhythm in the Hippocampal–Entorhinal System,” Nature Neuroscience, 16(2), 130–138. DOI record.
  66. Ray, S. & Maunsell, J. H. R. (2011), “Different Origins of Gamma Rhythm and High-Gamma Activity in Macaque Visual Cortex,” PLoS Biology, 9(4), e1000610. DOI record.
  67. Carpenter, S. K. et al. (2012), “Using Spacing to Enhance Diverse Forms of Learning: Review of Recent Research and Implications for Instruction,” Educational Psychology Review, 24(3), 369–378. DOI record.
  68. Centers for Disease Control and Prevention (updated 15 May 2024; accessed 4 September 2026), “Signs and Symptoms of Stroke.” CDC guidance.

Continue the Intelligence & Brain Function sequence

Explore each foundation in depth

  1. 1 · Definitions and Perspectives on Intelligence
  2. 2 · Brain Anatomy and Function
  3. 3 · Types of Intelligence
  4. 4 · Theories of Intelligence
  5. 5 · Neuroplasticity and Lifelong Learning
  6. 6 · Cognitive Development Across the Lifespan
  7. 7 · Genetics and Environment in Intelligence
  8. 8 · Measuring Intelligence
  9. 9 · Brain Waves and States of Consciousness
  10. 10 · Cognitive Functions

Evidence reviewed through 4 September 2026 · Educational overview; not a clinical assessment

Back to blog