Aging is associated with increased sensory and motor noise, requiring older adults to manage everyday tasks—such as walking to the bus station—under greater uncertainty. This becomes particularly challenging when they must adapt to unexpected situations. The aim of this project is to advance our understanding of the mechanisms and strategies older adults use to deal with uncertainty and to develop novel training approaches.
Study 1 (Joint Walking in Older Adults):
Data collection for Study 1 is complete. Participants performed a tango-inspired joint walking task in which they faced a trained leader and walked forwards and backwards. The leader changed direction after an unpredictable number of steps, so participants had to coordinate with the leader in real time under uncertainty. To test how different sensory inputs support this coordination, we restricted vision, audition or haptics. In total, 48 older women took part. Half of them attend dance classes regularly, while the other half do not attend regular dance or sports classes and served as a control group. We have analysed the data, discussed it in small internal symposia, and are currently writing the manuscript.
Study 2 (Joint Walking in VR):
Building on Study 1, we next investigated whether coordination in this task improves with practice and whether training in virtual reality (VR) transfers to real-world performance. Here, we did not manipulate sensory inputs but focused on joint walking without restrictions. We developed a VR implementation of the task in which the leader is projected as a virtual avatar into a VR environment based on the motion-tracking data. Participants were 36 healthy young students from the University of Bern, 18 men and 18 women. We tested their coordination performance before and after training, once in VR and once in the real environment. For the training, participants were split into three groups. One group trained with the real leader, one group trained with the leader shown as an avatar in VR, and a third group performed the steps alone without a leader, serving as a control group. Data collection and analysis are complete, and the manuscript will be submitted soon.
In a follow-up study, we plan to build on Study 2 using the same approach. By increasing the difficulty of the joint walking task, we aim to take a deeper look at the mechanisms behind the transfer from VR training to real-world performance. In addition, we will write a review that gives an overview of joint walking, its effects on gait parameters and its possible use in rehabilitation.
Our experimental paradigm involves a joint walking task in which participants are asked to coordinate their steps with a leader (Figure 1), without knowing in advance how many steps will be taken. This creates a situation of uncertainty, requiring real-time coordination.
In the first study, we manipulate the availability of sensory information (Figure 2) to examine participants’ reliance on different sensory modalities. We also investigate whether performance and coordination strategies differ between older adults with dance experience and an age-matched control group.
In the next step, we will examine how these strategies evolve with learning, and whether improvements in the joint walking task transfer to general gait parameters in other walking tasks. In a third step, we aim to develop and validate an extended reality (XR) version of the task that can be used independently by older adults to train stable and adaptable walking behaviour.
Currently, the first data is collected—comparing older adults with dance experience to a control group. In parallel, the development of the XR-task and a scoping literature review are in progress.
