A multimodal dataset for emotional transition analysis in virtual reality.

Dementia, particularly Alzheimer’s disease (AD), is a growing public health concern that presents profound challenges to healthcare systems, families and societies throughout the world. Midlife is a critical period for the development of AD pathology and potentially, a disease-altering window prior to the manifestation of substantial brain damage. There is an urgent need for risk reduction interventions focused on midlife. However, the indicators and brain mechanisms of AD in midlife remain poorly understood. Accumulating evidence points to functional brain network segregation—the extent to which a network is distinct from others in its functional connectivity profile—as a marker of brain health in both normal and pathological ageing. Increasing adult age is associated with reduced network segregation, and such ‘dedifferentiation’ is in turn associated with age-related decline in cognitive and motor function. Conversely, preservation of network segregation is associated with maintenance of cognition in healthy ageing and in patients with brain injury, suggesting that functional network segregation supports cognitive reserve. In asymptomatic older adults (mean age >65 years), loss of network segregation was associated with the accumulation of AD pathology, such as beta-amyloid (Aβ) and tau, or the presence of apolipoprotein () ε4 allele, the main genetic risk factor for sporadic late-onset AD in the Indo-European population. Reduction of network segregation in ε4 carriers was also associated with cognitive decline, and, conversely, its maintenance was associated with preserved cognitive performance, despite the presence of AD pathology. Decreased segregation was also observed in patients with mild cognitive decline (MCI) and AD compared to age-matched controls. An accelerated decline in network segregation with ageing has been associated with increasing dementia severity. It remains unknown, however, whether network segregation shows early modifications, from midlife, in individuals who have inherited risk of Alzheimer’s disease but are presently cognitively healthy. This question has important implications for identifying intermediate phenotypes of the earliest brain changes in the preclinical stages of AD, which can provide urgently needed early disease biomarkers for timely preventative interventions from midlife. It is also not known which brain networks are more liable to segregation loss during healthy ageing and in preclinical AD populations. Associative networks are shown to be more susceptible to age-related ‘dedifferentiation’ than the sensorimotor networks, but it is not known which individual network show early risk-related changes from midlife. The default mode network (DMN) is one such candidate network, with early changes—atrophy and metabolic abnormalities—being found in its core regions at early stages of clinical AD progression.