How human aging disrupts the head direction network: evidence from VR experiments and mechanistic models

Abstract

Navigational deficits during aging can severely limit mobility and reduce quality of life. While research on the underlying neural mechanisms has primarily focused on medial temporal lobe dysfunction, the head-direction (HD) system--a core component of the mammalian navigation circuit--remains largely unexplored in the context of aging. We established an immersive virtual reality paradigm that provides direct behavioral read-outs of HD signals. In addition, we developed a biologically inspired HD model, which accommodates noise sources that simulate age-related neural changes. Compared to younger adults, older participants exhibited larger angular errors, and a brief delay increased their heading uncertainty. In addition, our novel ring-attractor architecture shows that synaptic noise and small-scale neuronal loss replicate the magnitude and dynamics of the age-related deficits observed behaviorally. Together, these behavioral and computational findings provide the first evidence that aging compromises the fidelity and stability of the HD system. By pinpointing noise accumulation and neuron attrition as mechanistic contributors, our study significantly advances the understanding of spatial navigation deficits in old age, and it highlights novel targets for interventions aimed at preserving navigational abilities and quality of life.