For most of the 20th century, the biological basis of intelligence remained largely opaque. Psychometrics could tell us that intelligence was measurable and heritable, but the underlying neural mechanisms were inaccessible. The advent of structural and functional brain imaging — first CT, then MRI, then fMRI and diffusion tensor imaging — changed that. We can now observe, with unprecedented precision, how high-IQ brains differ from average-IQ brains. The picture that has emerged is nuanced, sometimes counterintuitive, and deeply informative.
The Parieto-Frontal Integration Theory
The most influential neuroscientific framework for intelligence is the Parieto-Frontal Integration Theory (P-FIT), proposed by Rex Jung and Richard Haier in 2007 following a systematic review of neuroimaging studies. P-FIT proposes that intelligence is not localised to any single brain region but emerges from the efficient integration of information across a distributed network spanning the parietal and frontal lobes — areas involved in sensory processing, working memory, abstract reasoning, and executive control.
According to P-FIT, the key driver of intelligence is not raw brain size or activity level, but the efficiency and integrity of the structural and functional connections linking parietal and frontal regions. This model has received support from studies showing that white matter tract integrity — measured using diffusion tensor imaging, which maps the myelin-sheathed axons that carry signals between brain regions — correlates significantly with IQ scores (typically r = 0.3 to 0.5 in adult samples). The better insulated and more coherently organised the neural "wiring," the faster and more reliably information is transmitted, and the higher the measured intelligence.
Brain Efficiency, Not Just Size
Early brain-intelligence research focused on total brain volume, and a modest positive relationship exists (meta-analytic r ≈ 0.24 between brain volume and IQ). But volume alone is a poor proxy for intelligence. More revealing is the concept of neural efficiency — the idea that higher-IQ brains accomplish cognitive tasks using less glucose metabolism and less overall neural activation.
This counterintuitive finding, first reported by Richard Haier and colleagues in 1988 using PET scanning, has been replicated multiple times with modern fMRI. When solving the same complex problems, higher-IQ individuals show less diffuse brain activation — they recruit fewer neurons, but recruit them more precisely and effectively. Think of it as the difference between a skilled surgeon making a precise incision and an untrained person flailing with a scalpel: the expert uses less energy, less collateral activation, and achieves a better outcome.
Interestingly, this efficiency pattern reverses under extremely novel or challenging conditions. When tasks push even high-IQ individuals to the edges of their capacity, their brains show increased activation — suggesting that efficiency is not simply a global property of high-IQ brains, but a dynamic characteristic of well-practised or manageable cognitive demands.
Working Memory, the Prefrontal Cortex, and g
Among the cognitive functions most tightly linked to general intelligence is working memory capacity — the ability to hold and manipulate information in conscious awareness. Neuroimaging studies consistently show that working memory tasks heavily recruit the dorsolateral prefrontal cortex (DLPFC) and anterior cingulate cortex, and that activation strength in these regions during demanding working memory tasks correlates with IQ scores.
The genetic component of intelligence is also largely mediated through brain structure. Twin studies and GWAS (genome-wide association studies) have identified many common genetic variants associated with both IQ and brain volume, cortical thickness, and white matter integrity. A 2017 study (Davies et al.) identified 40 genetic loci associated with cognitive function, most of which were also associated with structural brain differences — confirming that the genetic influences on intelligence operate substantially through their effects on brain development.
Key Takeaway
The neuroscience of intelligence has moved decisively beyond the simplistic question of "which brain region is responsible?" The picture that emerges from contemporary imaging research is of intelligence as an emergent property of a well-integrated, efficiently operating neural network — one characterised by strong white matter connectivity, a capacity for precise and economical information processing, and robust working memory infrastructure. These are biological characteristics shaped by both genetic endowment and developmental environment, and they underline that what IQ tests measure is anchored in real, observable differences in how brains are structured and how they function.