Abstract: BACKGROUND: Individuals with osteoarthritis fall at a greater rate than the general population, likely as a result of weakness, pain, movement limitations, and a decline in balance. Due to the high prevalence of osteoarthritis in the population, understanding the mechanisms leading to greater fall risk is an important issue to better understand. RESEARCH QUESTION: What is the influence of unilateral knee osteoarthritis on the characteristics of performing a voluntary step (i.e., similar to that performed to avoid a fall after a perturbation), compared to healthy age-matched controls? METHODS: Case-control study performed in a Health maintenance organization physical therapy clinic. The research group consisted of a referred sample of 21 patients with unilateral knee osteoarthritis. The control group consisted of 22 age-matched healthy individuals. All participants were over 65 years of age. Participants were excluded if they had a surgical procedure to back or lower limb within one year before testing, oncological or neurological disease or a deficit in tactile sense. Movements were performed with and without dual tasking. MEASUREMENTS: Duration of the initiation phase (cue to step initiation), preparatory phase (step initiation to foot off) and swing phase (foot off to foot contact). RESULTS: In the preparatory phase and swing phase, the osteoarthritis group moved more slowly than the control group, and these differences were larger for forward compared to backward movements. Dual-tasking slowed responses in the pre-movement initiation stage across groups. SIGNIFICANCE: The differences in basic parameters, and the slower movements in the osteoarthritis group, are consistent with known features of osteoarthritis, being a disease commonly regarded as primarily “mechanical”, and are likely to increase fall risk. These response deficits suggest we should take advantage of advanced rehabilitation techniques, including cognitive loading, to help prevent falls in older adults with osteoarthritis.

Abstract: Natural human behavior arises from continuous interactions between the cognitive and motor domains. However, assessments of cognitive abilities are typically conducted using pen and paper tests, i.e., in isolation from “real life” cognitive-motor behavior and in artificial contexts. In the current study, we aimed to assess cognitive-motor task performance in a more naturalistic setting while recording multiple motor and eye tracking signals. Specifically, we aimed to (i) delineate the contribution of cognitive and motor components to overall task performance and (ii) probe for a link between cognitive-motor performance and pupil size. To that end, we used a virtual reality (VR) adaptation of a well-established neurocognitive test for executive functions, the 'Color Trails Test' (CTT). The VR-CTT involves performing 3D reaching movements to follow a trail of numbered targets. To tease apart the cognitive and motor components of task performance, we included two additional conditions: a condition where participants only used their eyes to perform the CTT task (using an eye tracking device), incurring reduced motor demands, and a condition where participants manually tracked visually-cued targets without numbers on them, incurring reduced cognitive demands. Our results from a group of 30 older adults (>65) showed that reducing cognitive demands shortened completion times more extensively than reducing motor demands. Conditions with higher cognitive demands had longer target search time, as well as decreased movement execution velocity and head-hand coordination. We found larger pupil sizes in the more cognitively demanding conditions, and an inverse correlation between pupil size and completion times across individuals in all task conditions. Lastly, we found a possible link between VR-CTT performance measures and clinical signatures of participants (fallers versus non-fallers). In summary, performance and pupil parameters were mainly dependent on task cognitive load, while maintaining systematic interindividual differences. We suggest that this paradigm opens the possibility for more detailed profiling of individual cognitive-motor performance capabilities in older adults and other at-risk populations.