In recent years evidence has accumulated on the association between age-related decline in multisensory temporal processing and postural control. This proposed project aims at examining GABA-mediated inhibition as the potential mechanism behind this link. Our overarching hypothesis is that changes in the excitatory-inhibitory balance with a reduction of GABAergic inhibition is a common mechanism underlying age-related multisensory and postural deficits potentially mediated by age-related reduction in network segregation. To test this hypothesis, we will assess the relationship between age-related multisensory/postural deficits and age-related reduction in GABA concentration in related brain areas and probe the role of GABA-mediated inhibition using cognitive training that specifically targets inhibitory functions. Our multimodal approach is innovative, and findings from this study has the potential to lead to the development of safe and effective rehabilitation protocols for older adults with impaired multisensory temporal integration and postural control.

The investigators will examine the effects of cognitive training in the elderly on multisensory temporal processing and postural control. The investigators will include 25 young participants and use their performance to establish age-related differences and target performance for training in the elderly. 50 elderly participants will be randomly assigned to one of the two training groups and will be asked to undergo 20 days (~30 mins/day) of inhibitory control training or knowledge-based training. Multisensory temporal processing abilities, postural stability and gait performance, brain GABA concentration, and resting-state based network segregation will be assessed before and immediately after training. Follow-up assessments will be carried out 30 days after training to examine the persistence of training effects.

Stimulus presentation. Visual and auditory stimuli will be generated using MATLAB and the psychophysics toolbox. Visual and auditory stimuli will be delivered via Display++ system and an AudioFile stimulus processor, respectively (Cambridge Research Systems). The use of these two devices allows for precise control of stimulus timing.

Inhibitory control training. Participants will be trained with tasks that involve both response withholding and rule switching. For example, participants will be asked to sort polygons based on features (e.g., shape or color) and will be asked to either provide or withhold sorting responses when presented with different cues. Rule-switching is introduced when participants must inhibit the "rule" used in the previous trial set and change the focus of their attention to a new sorting and/or inhibitory rule.

Knowledge-based training. This condition serves as the control. Participants will be presented with questions from different categories such as vocabulary, science, or geography and asked to select the correct answer from 4 alternatives within a certain time limit.

In both training conditions, task difficulty will progressively increase over the course of training based on participants' abilities. Pre- and post-training assessment of multisensory temporal processing abilities, postural stability and gait performance, brain GABA concentration, and resting-state based network segregation will be conducted.

Inclusion Criteria:

• Age between 65 and 80 (elderly group) and between 20 and 35 (young group)

Exclusion Criteria:

• History of epilepsy, stroke, Parkinson's disease, Alzheimer's disease, attention deficit-hyperactivity disorder
• History of head injury with loss of consciousness, brain surgery, or psychiatric disorders
• Early dementia or cognitive impairment
• Serious vision problems (including best-corrected binocular visual acuity worse than 20/25)
• Uncorrected hearing problems (pure tone threshold > 40 dB for 1000-2000 Hz)

Results from proposed experiments will be reported either as group findings that do not identify individual participants, or by labeling the data of each participant with a unique code that does not include the participant's name or initials. De-identified individual participant data will be shared upon request and for research purpose only starting 1 year after publication.

Starting 1 year after publication

Data will be shared upon request and for research purpose only.

50 elderly participants will be randomly assigned to one of the two training groups but will not be told whether they are in the inhibitory control (experiment) or knowledge-based (control) training group.