Bayesian optimal adaptation explains age-related human sensorimotor changes.

Area of Science:

• Neuroscience
• Gerontology
• Systems Biology

Background:

• Aging is linked to sensory decline and motor control deterioration.
• Sensorimotor systems show correlations between age-related anatomical changes and behavioral deficits.
• A Bayesian framework is proposed to explain these sensorimotor relationships in aging.

Purpose of the Study:

• To investigate age-dependent changes in the human vestibulo-ocular reflex (VOR).
• To determine if Bayesian optimal adaptation explains VOR changes in response to sensory hair cell loss.
• To demonstrate longitudinal Bayesian processing in the aging human brain.

Main Methods:

• Utilized a Kalman filter model to analyze VOR temporal dynamics across age groups.
• Assessed how sensory signal-to-noise ratio degradation affects reflex output.
• Correlated model predictions with observed age-dependent changes in the VOR.

Main Results:

• Age-dependent VOR changes were accurately explained by a Kalman filter model (r = 0.93, P < 0.001).
• The model demonstrated optimal behavioral adaptation to degraded sensory information due to hair cell death.
• Findings confirm the brain's capacity for statistically optimal responses despite sensory system deterioration.

Conclusions:

• The aging brain employs Bayesian optimal adaptation to maintain motor behavior amidst sensory decline.
• This study provides evidence for longitudinal Bayesian processing throughout the human lifespan.
• Results offer insights into vestibular disorders, physical therapy, and fall prevention in the elderly.