People with chronic stroke (PwCS) often exhibit balance deficits linked to a fear of falling and an increased fall risk while walking (Weerdesteyn et al., 2008). This problem has motivated development of numerous biomechanical metrics of walking balance, one of which quantifies step-by-step modulation of mediolateral foot placement based on pelvis dynamics (Stimpson et al., 2018). Essentially, the foot tends to be placed more laterally when the pelvis has a larger displacement or velocity away from the stance foot (Wang and Srinivasan, 2014). This “foot placement strategy” is believed to be important for ensuring walking balance (Bruijn and van Dieën, 2018), but can be disrupted among PwCS (Dean and Kautz, 2015, Stimpson et al., 2019). In this population, deficits in paretic foot placement modulation (i.e., weaker positive relationship between pelvis displacement and foot placement) are linked with poorer clinical measures of balance, albeit moderately or weakly (Howard et al., 2023).

One factor that may contribute to post-stroke deficits in step-by-step foot modulation is altered sensory processing. This strategy likely requires individuals to form a perception of the state of their pelvis and stance leg, and control swing leg position to land in an appropriate location to prevent a lateral loss of balance (Rankin et al., 2014). Many PwCS have a reduced ability to perceive or interpret somatosensory signals (e.g., proprioception, cutaneous sense) (Borstad and Nichols-Larsen, 2014, Connell et al., 2008), which may disrupt this strategy.

Musculotendon vibration can provide insight into the role of somatosensory feedback in controlling human movement. Applying vibration to the hip abductors during standing tends to cause sway away from the vibration (Bonan et al., 2017, Popov et al., 1999, Roden-Reynolds et al., 2015). This behavior is consistent with participants perceiving a stretch of the vibrated muscle that would normally accompany sway toward the vibration and responding with a motor command to counteract this unintended sway. During walking in neurologically-intact adults, vibration of the swing leg hip abductors tends to cause more lateral foot placement, while vibration of the stance leg hip abductors tends to cause more medial foot placement of the contralateral swing leg (Arvin et al., 2018, Roden-Reynolds et al., 2015). Both behaviors are consistent with participants counteracting the perceived motion that would normally accompany stretch of the vibrated muscle.

Building upon prior work, we have extended the use of hip abductor vibration to augment somatosensory feedback during walking. In neurologically-intact adults, foot placement modulation was strengthened by applying vibration linked to real-time motion (Knapp et al., 2021), demonstrating the potential of this artificial sensory information. However, it is presently unclear whether such methods would have similar effects in PwCS. The development of a non-invasive method that strengthens the positive relationship between pelvis motion and foot placement in this population could benefit their walking balance.

Beyond potential clinical uses of hip abductor vibration, the mechanism underlying strengthened foot placement modulation is not entirely clear. The effects of musculotendon vibration are typically attributed to ascending proprioceptive signals (e.g., from muscle spindles) (Proske and Gandevia, 2012) altering the perceived mechanical state of the body and causing foot placement adjustments (Arvin et al., 2018, Roden-Reynolds et al., 2015). Alternatively, vibration may provide a tactile cue, as in other sensory augmentation applications (Sienko et al., 2018). Vibration delivered over the skin can evoke cutaneous feedback as would occur when contacting the environment (Shull and Damian, 2015), which has been used as cueing (Lee et al., 2013) to reduce sway during standing posture and walking (Haggerty et al., 2012, Sienko et al., 2013, Wall, 2010). Therefore, in the present study we investigated whether augmentative vibration delivered over the hip abductors (likely eliciting proprioceptive feedback from this muscle group) had unique effects on foot placement modulation when compared with vibration over the lateral trunk (similar to prior work using vibration as a cue).

The primary purpose of this study was to investigate for the first time whether hip abductor vibration can increase foot placement modulation among PwCS. We hypothesized that delivering hip abductor vibration scaled with real-time pelvis motion would strengthen mediolateral foot placement modulation for both paretic and non-paretic steps. We anticipated that this increase in foot placement modulation would be observed in comparison to both a condition with no vibration and a condition with a corresponding cutaneous cue delivered to the trunk, thus providing insight into the underlying mechanism. Secondarily, we explored whether the effects of hip vibration during quiet standing predict the effects of augmentative hip vibration during walking.