1.11.5 Neuroimaging and its role in understanding development

What each modality measures

MRI dominates developmental research because it uses no ionising radiation, so children can be scanned repeatedly.

Study designs

• Cross-sectional studies compare age groups once; they are quick but confound age with cohort and individual differences.

• Longitudinal studies rescan the same people and show that development varies between individuals and is often non-linear.

• Longitudinal studies suffer attrition, scanner changes, practice effects and measurement error.

Shape of brain development

• Brain volume is about a quarter of adult size at birth, three quarters by age two and 90 to 95% by about six.

• Later, white matter volume increases while many cortical grey matter measures decline.

• Synaptogenesis peaks earlier in visual than in prefrontal cortex.

• Sensory and motor systems mature before association regions, which go on changing into adulthood.

• Timing varies by region, measure and person, so no scan places an individual on a universal maturity scale.

Tap to enlarge

Cortical thinning

• MRI measures thickness between tissue contrast boundaries, so it does not directly measure synaptic pruning.

• Apparent thinning can reflect changes in synapses, dendrites, glia, water, folding or analysis method.

• Increasing myelination can shift the grey-white matter boundary without any tissue loss.

White matter and networks

• White matter develops into early adulthood, including long-range frontal connections.

• Rising fractional anisotropy and falling diffusivity may reflect myelination and axonal organisation, but fibre geometry also affects both.

• Functional networks become more differentiated from one another and more integrated internally.

• Head movement can mimic the reported childhood pattern of stronger short-range and weaker long-range connectivity.

Adolescence

• Dual systems accounts propose that reward responses and cognitive control mature on different schedules, contributing to context-dependent risk-taking.

• Adolescents may reason about risk like adults when calm, yet behave differently under arousal or peer influence.

• Group findings cannot establish an individual young person's maturity, intent or capacity for control.

Tap to enlarge

Ageing

• HAROLD (hemispheric asymmetry reduction in older adults) describes more bilateral recruitment.

• The posterior-anterior shift in ageing describes greater frontal than posterior recruitment.

Both accounts may apply in different tasks.

Interpreting findings

• Group-level brain-behaviour associations do not show direction, because experience changes the brain and shared causes affect both.

• Reverse inference wrongly concludes that a process occurred because a region linked to it activated; most regions serve several functions.

• Research scans cannot determine maturity, verify symptoms or diagnose most psychiatric disorders.

• Clinical imaging answers specific questions, such as suspected lesions, malformations, hydrocephalus or neurological disease.