Date of Award

Winter 12-15-2018

Author's School

Graduate School of Arts and Sciences

Author's Department

Movement Science

Degree Name

Doctor of Philosophy (PhD)

Degree Type



Abstract of the Dissertation

Singing as a Therapeutic Technique to

Improve Gait for People with Parkinson Disease


Elinor Clare Harrison

Doctor of Philosophy in Movement Science


Washington University in St. Louis, 2018

Professor Gammon Earhart, Chair

Gait impairment is common in older adults and even more prevalent for people with Parkinson disease (PD). Gait dysfunction is often characterized by reductions in speed, step frequency, and step length. In addition, decreased ability to regulate step length and step frequency may contribute to increased gait variability, making walking less stable and increasing risk for falls. As gait deficits are often resistant to drug therapy, there is a need to find alternative therapies that improve mobility. Rhythmic cueing in the form of listening to music is effective at enhancing walking for people with PD, helping people lengthen strides and increase velocity. However, research on rhythmic facilitation of movement has been limited to external cues and it is unknown if self-generated rhythmic cues, such as singing, may provide the same or greater benefit. This projects described in this dissertation are among the first to examine the effects of singing on walking and may reveal a novel, low-cost, non-invasive, accessible and adaptable therapeutic technique to normalize gait in PD.

In order to study the effects of internal cues on movement patterns in PD, we conducted four experiments (chapters 2-5). In the first experiment (chapter 2), we tested the feasibility of singing as a cueing technique by comparing it to traditional external cueing and to dual-task walking. We showed that while a dual task slowed and destabilized gait, singing while walking did not have this detrimental effect. In fact, singing did not negatively affect velocity, cadence, or stride length, and it positively impacted measures of gait variability. These results indicated that singing is not only feasible for people with PD but that it may hold potential to improve gait stability.

Buoyed by the results of our pilot study, we then set out to examine how best to administer singing as a therapeutic technique to elicit the most benefit for people with PD. In experiment two (chapter 3), we assessed the differential effects of internal and external cueing techniques on basic walking as well as more challenging gait situations. We tested both forward walking, commonly considered an automatic motor pattern, and backward walking, which tends to reveal more pronounced gait impairment and is related to fall risk. We included people with PD and a healthy control group to provide additional insight into how the role of beat impairment in PD may differentially affect task performance. Our results showed that internal cueing was associated with improvements in gait velocity, cadence, and stride length in the backward direction, and reduced variability in both forward and backward walking. In contrast, external cues minimally benefitted gait characteristics and detrimentally affected gait variability. We also confirmed that people with PD may exhibit greater improvement than their healthy counterparts, particularly in more challenging gait situations such as backward walking.

In experiment three (chapter 4), we investigated how different cue rates might alter responses in healthy controls and people with PD. In order to test this, we assessed cued walking conditions at tempos above, at, and below preferred gait cadence. We also added a second internal cueing condition of mental singing, in which participants sang in their heads, to determine if it could elicit the same benefits as singing aloud. The results indicated that mental singing was more effective than overt singing at eliciting gait improvement, which renders this technique more practical for everyday use. When done at rates of 10% above preferred cadence, mental singing allowed people to increase velocity while simultaneously reducing variability and gait asymmetry.

In our final experiment (chapter 5), we sought to compare the same cued conditions using motion capture technology in order to determine if rhythmic cues can improve movement quality as well as spatiotemporal gait features. In our assessment of lower extremity sagittal plane joint angles, we showed that cues may combat downregulation of movement amplitude by increasing range of motion at all lower limb joints. These increases in movement amplitude may be associated with longer strides and reduced stride-to-stride variability. We were able to distinguish some key features that may predict likelihood of responding positively to internal cueing techniques, such as freezing status, fall history, and prior musical experience. The results indicate that internal cues may benefit a range of people with PD, even those at risk of more debilitating gait impairments such as falling or freezing of gait, and that those with prior musical experience are most likely to respond.

Taken together, these results provide compelling evidence that internal cues are a promising therapeutic technique that may transform gait rehabilitation for older adults as well as people with PD. The experiments detailed herein contribute to a burgeoning field of literature concerning rhythm processing and are among the first to examine singing as a cueing technique for people with PD.


English (en)

Chair and Committee

Gammon M. Earhart

Committee Members

Michael Harris, Catherine Lang, David Marchant, Pietro Mazzoni,


Permanent URL: https://doi.org/10.7936/8qzt-nf80