Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Jul 1.
Published in final edited form as: J Neurol Phys Ther. 2025 Feb 27;49(3):162–170. doi: 10.1097/NPT.0000000000000508

Personalized auditory rhythmic cues to optimize gait in older adults and people with Parkinson disease

Elinor C Harrison 1,4, Lauren E Tueth 1, Allison M Haussler 1, Kerri S Rawson 1,2, Gammon M Earhart 1,2,3,*
PMCID: PMC12188473  NIHMSID: NIHMS2069424  PMID: 39913311

Abstract

Background and Purpose:

Gait dysfunction in people with Parkinson disease (PwPD) is common and debilitating. Internal cues (i.e., mental singing) may provide a useful alternative to external cueing (i.e., listening to music) to improve walking for PwPD. In this study, we individually tailored cues to each participant by allowing them to choose their own cue song and by selecting the cue rate that elicited the longest strides.

Methods:

Fifty-six participants (28 Controls, 28 PwPD) performed walking trials in a single-session, cross-sectional trial. Initial, uncued trials were compared to cued conditions in which participants walked to music or while mentally singing at four randomized cue rates (90%, 100%, 110%, and 120% of uncued walking cadence). Participants chose their song from a curated catalogue. Cues were optimized according to longest stride lengths as measured by APDM wearable sensors. Repeated-measures ANOVAs assessed differences between gait characteristics (velocity, cadence, and stride length) and gait variabilities (coefficients of variations) across conditions and groups.

Results:

Both cue types improved velocity, cadence, and stride length as compared to uncued walking. Internal cues were most beneficial to gait variability. The majority of participants exhibited the longest strides and fastest velocities at the highest cue rate of 120%, though a cue rate of 110% was most beneficial to gait variability.

Discussion and Conclusions:

This study shows how internal and external cues personalized by song choice and cue rate may be tailored to individuals. The information gained may help tailor cues to optimize gait performance for older adults and PwPD.

Introduction

People with Parkinson disease (PwPD) experience difficulty with rhythmic movements including gait.(Studenski et al. 2011) Gait dysfunction in Parkinson disease (PD) is characterized by reduced speed, reduced stride length, and increased temporal and spatial variability.(Meg Morris et al. 2005) Multiple studies suggest uneven timing of footfalls underlies walking difficulties in PD and show a positive correlation between increased stride-to-stride variability and fall risk.(Ebersbach et al. 1999; Skodda, Flasskamp, and Schlegel 2010; Thaut et al. 2001) The basal ganglia, subcortical structures affected in PD, are linked to impaired rhythm perception and disordered gait timing.(Grahn and Brett 2009) Gait dysfunction may manifest early in PD and is among the most debilitating motor symptoms, impacting everyday function and quality of life.(Nisenzon et al. 2011) Furthermore, gait impairments are often resistant to pharmacological or surgical approaches, thus, alternative treatments are needed.

One of the most studied approaches to treating gait timing disturbance in PD is use of external rhythmic auditory cueing. Decades of research show that auditory stimulation, mostly in the form of music or metronome beats, can improve temporal (i.e., gait speed, cadence) and spatial (i.e., stride length) gait features.(McIntosh 1997; Rochester et al. 2005; Thaut et al. 1996; 2001; Willems et al. 2006a) These methods may take advantage of biological linkages between auditory and motor systems, allowing synchronization between movement and externally produced rhythms.(Rossignol and Jones 1976) External auditory cues may facilitate sensorimotor network activity by engaging cerebello-thalamo-cortical pathways; little is known about the neurophysiological mechanism underlying internal cue use.(Vitorio et al. 2018)

External rhythmic auditory cues are an imperfect tool. External cues require cueing devices, are not adaptable in the moment, and elicit effects that fade quickly after cue removal.(Hausdorff et al. 2007) Furthermore, temporal variability (i.e., cadence CV) may worsen in the presence of external cues for both older adults(Shashank Ghai and Shashank Ghai 2017) and PwPD(Harrison and Earhart 2023), potentially reflecting difficulty synchronizing to an outside source. Other factors that may influence responsiveness to rhythmic cues include past musical training(Dalla Bella et al. 2017), synchronization ability(Leow, Parrott, and Grahn 2014), disease severity(Erra et al. 2019; Pablo Arias and Cudeiro 2010), baseline gait impairment(Cochen De Cock et al. 2018), and familiarity with cue song.(Leow, Rinchon, and Grahn 2015)

To address these shortcomings, we previously tested internal cueing as an alternative approach. We showed internal, self-generated rhythmic cues, such as singing aloud or mentally singing in one’s head, were not different from external cues with respect to improving gait speed and stride length.(Harrison, McNeely, and Earhart 2017; Harrison, Horin, and Earhart 2018) Furthermore, mental singing, or singing in one’s own head, was associated with significant reductions in both spatial and temporal variability measures.(Harrison, Horin, and Earhart 2019)

In the present study, we further examined the benefits of external versus internal cueing by personalizing cues to individuals. Past research suggests not everyone responds the same to external rhythmic cues and personalized cues might improve responsiveness.(Horin et al. 2020; Harrison et al. 2020) However, no study systematically personalized either external or internal cues to maximize effectiveness for each individual. We set out to elucidate the benefits of personalized cues, in terms of song choice, cue rate, and cue type. Our study design allowed participants to select their own cue song and tailored each individual’s cue rate to maximize strides using both external and internal cue types.

While in prior work we used only one song, here we provided a range of songs from which participants could choose. We previously showed internal cues of up to 110%PW were most effective for improving gait(Harrison, Horin, and Earhart 2019) and herein we expanded our cue offerings up to 120%PW. This higher tempo of 120%PW has been effective during external cueing for older adults(Shashank Ghai and Shashank Ghai 2017) and sometimes for PwPD.(Ghai et al. 2018) We then systematically tested each individual to determine which tempo elicited their longest strides, which we refer to as their “optimal cue rate”. We chose stride length as the parameter for optimization because movement scaling difficulty in PwPD is thought to most negatively impact stride length regulation.(M. Morris et al. 1998)

The first aim of this study was to compare the immediate effects of personalized external cues (music) and personalized internal cues (mental singing) with uncued walking on specific gait spatiotemporal measures among older adults and PwPD. We hypothesized that personalized cues would allow both groups, older adults and PwPD, to improve gait characteristics (velocity, stride length, cadence) as compared to uncued walking. This hypothesis is based on robust past work showing that PwPD are able to effectively utilize internal cues despite beat impairment.(Harrison, McNeely, and Earhart 2017; Harrison, Horin, and Earhart 2018; 2019; Horin et al. 2020; Harrison et al. 2020) The second aim was to compare the immediate effects of personalized external cues (music) and personalized internal cues (mental singing) with uncued walking on specific gait variability measures among older adults and PwPD. We hypothesized that internal cueing, in the form of mental singing, would reduce variability more than music for both groups as compared to uncued. The third aim was to explore the relation between optimal cue rate and individual characteristics (i.e., age, disease severity, musical training, and uncued gait characteristics) that could explain differences in individual responsiveness to cueing in healthy older adults and PwPD. We hypothesized optimal cue rate would correlate to individual characteristics.

The systematic personalization of cues for each individual presents a clinically meaningful template for individualized approaches in rehabilitation settings. By expanding song choices and optimizing cue rates, this study may provide clinicians with valuable information for tailoring cues in future interventions.

Methods

This study was approved by the Institutional Review Board (IRB) of Washington University in St. Louis School of Medicine and registered at ClinicalTrials.gov (NCT04246476). A single center, cross-sectional, interventional study design was utilized.

Participants

Participants were recruited for a larger study that included MRI brain scans. PwPD were recruited through the Movement Disorders Clinic at Washington University in St. Louis School of Medicine and via the American Parkinson Disease Association. Age- and sex-matched controls were recruited through the Washington University Older Adults Participant Pool and the Research Participant Registry. Additional recruitment occurred through advertisements, flyers, and social media.

Inclusion criteria were: 1) at least 30 years of age; 2) right-handed (criterion for MRI); 3) normal hearing; 4) able to walk for 10 continuous minutes independently; and 5) willing and able to provide informed consent. Additional inclusion criteria for PwPD were: 1) diagnosis of idiopathic, typical PD according to the UK Brain Bank Criteria; 2) Hoehn & Yahr stages 2–3(Hoehn and Yahr 1998); 3) score ≥ 1 on the Movement Disorders Society Unified Parkinson Disease Rating Scale – Part III – Motor Aspects (MDS-UPDRS-III)(Goetz, Tilley, Shaftman, Stebbins, Fahn, Martinez-Martin, Poewe, Sampaio, Stern, Dodel, Dubois, Holloway, Jankovic, Kulisevsky, Lang, Lees, Leurgans, LeWitt, Nyenhuis, Olanow, Rascol, Schrag, Teresi, van Hilten, and LaPelle 2008) item #10 indicating observable gait impairment; 4) score of ≤1 on the New Freezing of Gait Questionnaire (nFOG-Q)(Alice Nieuwboer et al. 2009) item #7 indicating freezing episodes are not moderately or significantly disturbing; and 5) stable PD medications for at least two months prior. Potential participants were excluded if they had a diagnosis of any other neurological condition, other medical conditions that might interfere with safe participation or cognitive impairment (i.e., Mini-Mental Status Examination (MMSE) score <24).

Protocol

All participants came to the Movement Science Research Center for a single session lasting 2–3 hours. PwPD were tested while on their normal PD medications.

Upon providing written informed consent, eligibility was determined during a pre-assessment that included a demographics questionnaire, pure-tone electronic hearing test, MMSE, MDS-UPDRS-III(Goetz, Tilley, Shaftman, Stebbins, Fahn, Martinez-Martin, Poewe, Sampaio, Stern, Dodel, Dubois, Holloway, Jankovic, Kulisevsky, Lang, Lees, Leurgans, LeWitt, Nyenhuis, Olanow, Rascol, Schrag, Teresi, van Hilten, LaPelle, et al. 2008), and nFOG-Q(A. Nieuwboer et al. 2009).

Gait Assessments

Gait data were recorded using the APDM Mobility Lab System gait and posture analysis system (APDM, Inc., Portland, OR).(R. Morris et al. 2019) This system is equipped with six wearable, inertial Opal sensors placed on the wrists, feet, sternum, and lumbar spine. It is lightweight and wireless, allowing full freedom of movement and continuous monitoring of ambulation.

All walking trials were performed in a 30m hallway where participants walked continuously in an ellipse. The ends of the ellipse were marked by cones that participants walked around. A tone produced by the APDM software indicated the start and end of each trial. In each condition, participants performed three 30-second trials from which turns were automatically excluded from analysis. Rest breaks were offered as needed. For safety, participants were followed by a trained physical therapist. No falls or adverse events occurred during data collection.

Uncued walking

During uncued walking trials, participants were asked to walk in silence at their preferred walking (PW) speed (Fig 1A). These trials were used to represent ‘normal’ walking and provide a point of comparison for other conditions. Uncued walking trials were averaged to determine preferred walking cadence which was used to calculate four individualized cue rates: 90%, 100%, 110%, and 120% of PW cadence. The range of cue tempos was chosen to reflect those most likely to improve gait for PwPD. While the majority of studies of external cueing show PwPD improve most with cues set to tempos higher than baseline walking cadence(Howe et al. 2003; Picelli et al. 2010; P. Arias and Cudeiro 2008), some PwPD lengthen strides more when presented with slower cues.(Willems et al. 2006b)

Figure 1.

Figure 1.

Open in a new tab

Schematic of motor assessment tasks.

Songs were prerecorded using single notes on a digital piano keyboard. Cues were rounded to the nearest five beats per minute increment, so songs were ready for any cue rate necessary ranging from 60 to 160bpm.

Song selection

After three uncued walking trials, participants selected a cue song from a curated catalogue of ten options compiled by a music therapist (Fig 1B). These commonly known songs were selected for having characteristics appropriate for gait training such as 4/4 timing, beat salience, familiarity, and simple lyrics.(Leman et al. 2013) Participants were asked to select the song most familiar to them as improvements in velocity and stride length have been seen in PwPD when synchronizing to a highly familiar song(de Bruin et al. 2010) and lyrics sung on a life-long familiar melody result in better consolidation and higher retention.(Moussard et al. 2014) Only the chorus was played to keep cues simple and repetitive to facilitate motor synchronization.(Repp and Su 2013)

The songs available were as follows: “I’ve Been Working on the Railroad”, “You Are My Sunshine”, “This Land is Your Land”, “Skip to My Lou”, “She’ll Be Coming Around the Mountain”, “Don’t Sit Under the Apple Tree”, “When the Saints Go Marching In”, “You Can’t Hurry Love”, “When Johnny Comes Marching Home Again”, and “Seventy-Six Trombones.”

Once participants selected a song, they were given the lyrics to review. We then removed the lyrics and recorded participants singing the song aloud to ensure memorization. Participants performed three practice trials singing the lyrics aloud while walking to familiarize them with performing the two tasks simultaneously.

Cued walking

Cues were delivered from a laptop with external speakers. Cue rates were presented in randomized blocks and two cue types (Music and Mental) were randomized within each tempo (Fig 1C). Three trials of 30 seconds each were performed in each condition at each tempo.

Music:

This condition represented traditional cueing techniques where music played and participants were asked to walk to the beat. Each song played once through before the starting tone sounded and participants began walking to the music. Participants were asked to not sing aloud or in their heads and were monitored to ensure compliance. After every music trial, participants were asked if they were accidentally singing in their heads. If they reported they were, the trial was excluded and repeated.

Mental:

This condition represented our experimental cueing paradigm in which participants were asked to mentally sing while walking without music playing. Each song played once through, then turned off before the starting tone sounded and participants began walking and mentally singing. To ensure participants were correctly doing the task, participants were asked to confirm they were mentally singing after each trial. If they reported they forgot to sing or forgot the lyrics, the trial was excluded and repeated.

Optimal Cue Rate:

Upon completion of all walking trials, each participant’s optimal cue rate (i.e., tempo with longest strides) was determined by averaging stride length across the music and mental trials within each cue rate (Fig 1D).

Statistical Analysis

Statistical analyses were conducted using IBM SPSS 27. Independent samples t-tests (p< .05) were used to compare uncued walking between groups (control, PwPD). Repeated measures (RM) ANOVAs (2×3) included main effect of group (control, PwPD), main effect of condition (uncued, music, mental), and the interaction of group and condition for each gait characteristic (velocity, cadence, stride length, and their coefficients of variation). For cued conditions, optimal cue rate (i.e., tempo that elicited longest strides) for each participant was used for analysis. For significant main effects of condition, Bonferonni adjusted pairwise comparisons are reported. Greenhouse-Geisser estimates of sphericity are reported when Mauchley’s test indicated assumption of sphericity had been violated.

Standardized residuals were examined for outliers (± 1.96 SD) by condition and winsorized to the next highest value within ± 1.96 SD for the particular outlier’s group.

Correlation analyses examined participant characteristics (age, disease severity, musical training, and uncued gait velocity, cadence, stride length) related to cue rate optimization or responsiveness.

Using preliminary data on gait variability measures (cadence CV) from our lab, the power analysis indicated a sample size of 50 would be needed to detect a moderate effect size (Cohen’s d = .5) in a RM ANOVA with two groups (control, PwPD) and three conditions (uncued, music, mental) at 80% power. To account for attrition, we aimed to recruit 60 participants.

Results

Twenty-nine older adult controls and 29 PwPD, matched by age and sex, were recruited (Table 1). One participant with PD who was an outlier for 89% of data points was removed from analyses along with their matched control. The final analysis included 28 participants per group. Comparisons below are between uncued walking and cued conditions (music and mental) at optimized tempo unless otherwise stated.

Table 1.

Participant Demographics

Controls (n=28) PwPD (n=28)
Sex, % male 71.4 71.4
Age, years 67.75 (6.14) 66.76 (5.84)
LEDD, mg … 581 (742)
MDS-UPDRS-III … 35.86 (13.19)
MMSE, median (range) 29 (26,30) 29 (27,30)
Fallers (%) 25.0 35.7
Open in a new tab

Abbreviations: LEDD, levodopa equivalent daily dose; MDS-UPDRS-III: Movement Disorders Society – Unified Parkinson Disease Rating Scale Part III; MMSE, Mini-Mental Status Examination; Fallers, % of participants who reported 1 or more falls in the past year; PwPD, People with Parkinson disease.

*

Values represent mean (SD), except where noted.

Gait

For baseline, uncued walking, t-tests confirmed the groups were similar in velocity (p=.307), cadence (p=.73), and stride length (p=.183). Complete results for all cue rates are in Table 2.

Table 2.

Gait characteristics

90% 100% 110% 120%
Uncued Music Mental Music Mental Music Mental Music Mental
Velocity (m/s) Control 1.17 (.13) 1.10 (.15 ) 1.12 (.13) 1.18 (.14) 1.20 (.14) 1.24 (.15) 1.24 (.15) 1.32 (.16) 1.29(.15)
PD 1.12 (.14) 1.06 (.13) 1.07 (.14) 1.15 (.14) 1.15 (.15) 1.24 (.16) 1.23 (.16) 1.31 (.17) 1.28 (.16)
Cadence (steps/minute) Control 112.39 (7.18) 106.22 (10.53) 108.62 (8.17) 112.44 (10.73) 113.70 (10.82) 116.87 (12.00) 116.17 (11.11) 121.32 (14.20) 119.30 (11.69)
PD 112.65 (7.30) 105.64 (6.96) 107.25 (6.92) 111.67 (7.05) 112.27 (6.95) 118.53 (7.79) 117.12 (6.79) 123.21 (8.16) 120.05 (7.73)
Stride Length (m) Control 1.25 (.14) 1.24 (.11) 1.24 (.11) 1.26 (.12) 1.27 (.11) 1.28 (.11) 1.28 (.11) 1.31 (.12) 1.30 (.12)
PD 1.20 (.13) 1.20 (.11) 1.19 (.12) 1.23 (.13) 1.22 (.13) 1.26 (.13) 1.26 (.14) 1.27 (.14) 1.28 (.13)
Velocity CV Control .017 (.004) .017 (.003) .016 (.004) .016 (.005) .014 (.003) .016 (.005) .015 (.004) .017 (.006) .015 (.003)
PD .019 (.004) .019 (.005) .018 (.005) .015(.004) .016 (.005) .015 (.004) .015 (.004) .018 (.006) .016 (.006)
Cadence CV Control .010 (.002) .012 (.003) .009 (.002) .011 (.004) .008 (.002) .011 (.004) .009 (.002) .012 (.004) .009 (.002)
PD .010 (.003) .013 (.004) .010 (.003) .010 (.002) .010 (.002) .010 (.003) .008 (.002) .012 (.005) .010 (.003)
Stride Length CV Control .012 (.003) .015 (.004) .012 (.002) .015 (.004) .011 (.002) .014 (.004) .012 (.003) .014 (.006) .011 (.002)
PD .014 (.003) .014 (.004) .013 (.004) .012 (.003) .012 (.004) .013(.004) .011 (.004) .015 (.007) .011 (.004)
Open in a new tab

Gait characteristics for both conditions, Music and Mental, for both Controls and people with Parkinson disease, shown at all tempos, 90%, 100%, 110%, and 120% of preferred walking cadence. CV, coefficient of variation; PD, Parkinson disease

For velocity, there was a within-subject main effect of condition (F1.166,62.981 = 70.378, p<.001) with pairwise comparisons indicating both cue types (music and mental) significantly increased velocity compared to uncued (p<.001) and cue types were not different from each other (p=.332) (Fig. 2A). The between-subject main effect of group was not significant (F1.000, 54.000 = .296, p=.589). The condition*group interaction was not significant (F1.166,62.981 =1.270, p=.271) indicating controls and PwPD responded similarly.

Figure 2.

Figure 2.

Open in a new tab

Gait characteristics and coefficients of variation (CV) compared between three conditions. *p<.05 within-subject main effect of condition. There were no differences between groups.

For cadence, there was a within-subject main effect of condition (F1.22,65.896 = 40.483; p<.001) with pairwise comparisons indicating both cue types (music and mental) significantly increased cadence compared to uncued (p<.001) and cue types were not different from each other (p=.143) (Fig. 2B). The between-subject main effect of group was not significant (F1.000, 54.000 = .357, p=.553). The condition*group interaction was not significant (F1.22, 65.896=1.333; p=.26) indicating controls and PwPD responded similarly.

For stride length, there was a within-subject main effect of condition (F1.194,64.453=58.835; p<.001) with pairwise comparisons indicating both cue types (music and mental) significantly increased stride length (p<.001) and cue types were not different from each other (p=.999) (Fig. 2C). The between-subject main effect of group was not significant (F1.000, 54.000 =1.154, p=.288). The condition*group interaction was not significant (F1.194,64.453=1.02; p=.33) indicating controls and PwPD responded similarly.

Gait variability

For velocity CV, there was a within-subject main effect of condition (F1.594,86.09=7.417; p=.002). Pairwise comparisons indicated mental singing significantly decreased velocity variability compared to uncued (p=.004) (Fig. 2D). Music was not different from uncued (p=.144) or mental singing (p=.075). The between-subject main effect of group was not significant (F1.000, 54.000 =.843, p=.363). The condition*group interaction was not significant (F1.594,86.09=.125; p=.836) indicating controls and PwPD responded similarly.

For cadence CV, there was a within-subject main effect of condition (F1.823,98.418=15.880; p<.001). Pairwise comparisons indicated cadence variability was higher for music compared to uncued (p=.002) and mental singing (p<.001) (Fig. 2E). Mental singing was not different from uncued (p=.138). The between-subject main effect of group was not significant (F1.000, 54.000 =.121, p=.729). The condition*group interaction was also not significant (F1.823,98.418=2.508; p=.092) indicating controls and PwPD responded similarly.

For stride length CV, there was a within-subject main effect of condition (F2,53=17.528; p<.001) (Fig. 2F). Pairwise comparisons indicated stride length variability was higher for uncued (p=.015) and music (p<.001) compared to mental singing. The between-subject main effect of group was not significant (F1.000, 54.000 =.071, p=.791). The condition*group interaction was not significant (F2.000,53.000 =2.336; p=.107) indicating controls and PwPD responded similarly.

Personalized song choice

The songs chosen most frequently by controls were “You are my sunshine” (44.8%) and “I’ve been workin’ on the railroad” (20.7%). The songs chosen most frequently by people with PD were “Skip to my Lou” (34.5%) and “You are my sunshine” (31%) (Fig. 3A). Three songs were not chosen by any participant.

Figure 3.

Figure 3.

Open in a new tab

A) Percentage of controls and PwPD who were optimized to each: A) song (as selected by participants; songs that were not chosen by anyone are not shown); B) beats per minute (as determined by which rate elicited the longest stride lengths during cueing); C) stride length; D) velocity; E) cadence variability (CV).The y-axis shows percent of participants who performed optimally at each tempo (i.e., longest stride lengths, highest velocity, lowest stride time variability).

Optimized cue rate

Both groups overall exhibited the longest strides lengths at the highest cue rate of 120% of baseline cadence, with t-tests revealing no differences between groups (p=.649) (Fig. 3C). The median cue rate was 130 beats per minute for both controls (range: 100–155) and PwPD (range: 100–150), with t-tests revealing no differences between groups (p=.732) (Fig. 3B).

No significant correlations were noted between demographics and optimized cue rate (See Supplemental Table 1.).

Cue optimization based on other variables

Although we analyzed all data according to optimization of stride length, we also explored optimization based on other factors such as velocity and variability. We did this to determine if other methods of optimization might elicit different results and therefore might be important to consider when prescribing individualized cues. When considering which cue rates increased velocity the most, we noted the highest cue rate of 120% PW was still best for most participants (Fig. 3D). However, when considering cadence variability, we found the majority of participants performed best at 110%PW or less (Fig. 3E).

Discussion

The goal of this study was to explore personalized internal and external rhythmic cues for older adults and PwPD to offer insight into how to tailor cues individually. Our results indicate personalized cues are effective for improving gait for older adults and PwPD, and that cue administration should not be one-size-fits-all but rather tailored to individual needs.

In terms of our aims, our results show that 1) personalized external cues (music) and personalized internal cues (mental singing) convey immediate improvements in spatiotemporal gait measures among older adults and PwPD; 2) personalized internal cues (mental singing) are more effective than personalize external cues (music) at improving specific gait variability measures among older adults and PwPD; 3) none of the factors we explored explain differences in individual responsiveness to cueing in healthy older adults and PwPD.

Participants selected a range of songs from the curated list. Songs most frequently selected had particularly simple lyrics that were easy to remember. All were songs commonly sung to children, indicating they might be life-long melodies that participants remembered from childhood.(Halpern and Zatorre 1999) Offering options of song choices for gait rehabilitation could give patients a sense of agency.

The majority of participants in both groups were optimized to the highest cue rate of 120%PW, indicating they took the longest strides during the fastest-paced stimuli. This aligns with past work showing older adults can modulate gait parameters using external cues of up to ±20%PW. (Peper, Oorthuizen, and Roerdink 2012; Roerdink et al. 2011; Shashank Ghai and Shashank Ghai 2017) For PwPD, however, past research into external cueing offers conflicting reports. Some studies indicate PwPD can modulate gait at tempos of 120%PW(Ghai et al. 2018), whereas others suggest such tempos are too high for PwPD, perhaps because such fast tempos exceed their motor synchronization capabilities.(Willems et al. 2006b) In our study, we showed most PwPD were able to successfully modulate gait to high tempos, even when mentally singing. As ours is the first study to explore tempos of 120%PW during internal cueing, this confirms mental singing is both possible and effective at even higher tempos than previously explored.

While most participants elongated strides the most at the highest cue rate, this was not true for everyone. Some participants from each group were optimized at slower tempos (90%–110%PW). Past research suggests slower tempos can be especially useful for increasing stride length for people at more advanced stages of disease.(Willems et al. 2006b) As the disease progresses, physical limitations or motor inflexibility may restrict ability to modulate gait to higher tempos. Offering slower tempos than PW may potentially re-balance the stride length-cadence trade-off by allowing more time to execute steps and reducing shuffling.(Willems et al. 2006b)

When exploring optimization via different variables, we noted that 120%PW not only elongated strides the most, it also elicited the highest velocities. This indicates that, for most people, synchronizing cadence to higher cue rates led to concurrent increases in stride length and velocity. Though similar results have been shown with external cues(Ghai et al. 2018), this is the first report of people achieving such high cadence increases during mental singing.

We were also interested in seeing which cue rate most improved gait variability. Regarding optimization of cadence variability, 120%PW was not the optimal tempo for most. Rather, lower tempos were often better for improving gait variability. This suggests that, for some participants, cue of 120%PW may be possible but may sacrifice gait stability. For someone taking short, shuffling steps or exhibiting festination, 120%PW may help lengthen strides the most. However, for someone who is struggling with gait stability or prone to falling, a lower tempo may be most beneficial. These results highlight the importance of taking each individual’s needs into consideration and tailoring cues accordingly.

Mental singing was not significantly different from external musical cues with respect to improving gait, corroborating past work showing mental singing may be useful for gait rehabilitation in PD.(Harrison, McNeely, and Earhart 2017; Harrison, Horin, and Earhart 2018; 2019; Satoh and Kuzuhara 2008) These results also show internal cues improve gait variability more than external cues, potentially due to ease of synchronization during vocal-motor coupling tasks.(Harrison, Horin, and Earhart 2019) Gait variability results, which are more reliable over longer distances(Rennie et al. 2018), are notable as we utilized wearable sensors that allowed analyzing more steps per trial than we had done previously.

Although correlation analyses did not reveal significant relationships between demographics and optimal cue rates, participants who were optimized at slower cue rates were generally older, had higher disease severity, and had more impaired baseline gait. We also noted the optimized beats per minute was highly variable, with a range spanning from 100–155 bpm among our sample. The median of 130 bpm closely aligns with the preferred musical tempo for humans (120–130bpm)(Hine et al. 2022) and is close to the middle of the range for average gait cadence for older adults (100–150).(Tudor-Locke et al. 2021) However, this cue rate is relatively high compared to other rhythmic cueing studies(Ghai et al. 2018) and should be considered in light of a high-performing sample in this study. Previous work suggests a cadence of 130 steps/min is metabolically inaccessible over long durations for most older adults.(Tudor-Locke et al. 2021) As our gait optimization protocol only measured gait in short, 30-second bouts, it is unclear if these cue rates could be maintained over longer durations.

Conclusions

The results of this study indicate personalized cues can effectively improve gait for older adults and PwPD. Both groups utilized internal and external cues to benefit gait, though internal cues reduced gait variability more than external cues. The range of responsiveness to cueing suggests cue rate optimization for gait rehabilitation is not one-size-fits-all. Future studies could: 1) explore a broader range of personal characteristics to see what else may contribute to responsiveness and 2) train participants in how to utilize internal cueing techniques to maximize responsiveness.

Supplementary Material

Supplementary Table 1

Supplemental Table 1: Correlations between optimal cue rate and participant demographics.

Acknowledgements

This work was supported by the National Center for Complementary and Integrative Health of the National Institutes of Health (NIH) under Award Number R61/R33AT010753 and the National Center for Advancing Translational Sciences of the National Institutes of Health under Award Number TL1TR002344. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.

Sources of Funding:

Research reported in this publication was supported by the National Center for Complementary and Integrative Health of the National Institutes of Health under Award Number R61AT010753 and the National Center for Advancing Translational Sciences of the National Institutes of Health under Award Number TL1TR002344. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Footnotes

Conflicts of Interest:

The authors declare no conflicts of interest.

References

  1. Arias P, and Cudeiro J. 2008. “Effects of Rhythmic Sensory Stimulation (Auditory, Visual) on Gait in Parkinson’s Disease Patients.” Exp Brain Res 186 (April):589–601. 10.1007/s00221-007-1263-y. [DOI] [PubMed] [Google Scholar]
  2. Arias Pablo, and Cudeiro Javier. 2010. “Effect of Rhythmic Auditory Stimulation on Gait in Parkinsonian Patients with and without Freezing of Gait.” PLOS ONE 5 (3): e9675. 10.1371/journal.pone.0009675. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bruin Natalie de, Doan Jon B., Turnbull George, Suchowersky Oksana, Bonfield Stephan, Hu Bin, and Brown Lesley A.. 2010. “Walking with Music Is a Safe and Viable Tool for Gait Training in Parkinson’s Disease: The Effect of a 13-Week Feasibility Study on Single and Dual Task Walking.” Parkinson’s Disease 2010 (July):483530. 10.4061/2010/483530. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Cochen De Cock V, Dotov DG, Ihalainen P, Bégel V, Galtier F, Lebrun C, Picot MC, et al. 2018. “Rhythmic Abilities and Musical Training in Parkinson’s Disease: Do They Help?” Npj Parkinson’s Disease 4 (March):8. 10.1038/s41531-018-0043-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Dalla Bella S, Benoit CE, Farrugia N, Keller PE, Obrig H, Mainka S, and Kotz SA. 2017. “Gait Improvement via Rhythmic Stimulation in Parkinson’s Disease Is Linked to Rhythmic Skills.” Sci Rep 7 (February). [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Ebersbach G, Heijmenberg M, Kindermann L, Trottenberg T, Wissel J, and Poewe W. 1999. “Interference of Rhythmic Constraint on Gait in Healthy Subjects and Patients with Early Parkinson’s Disease: Evidence for Impaired Locomotor Pattern Generation in Early Parkinson’s Disease.” Movement Disorders: Official Journal of the Movement Disorder Society 14 (4): 619–25. [DOI] [PubMed] [Google Scholar]
  7. Erra Carmen, Mileti Ilaria, Germanotta Marco, Petracca Martina, Imbimbo Isabella, Biase Alessandro De, Rossi Stefano, et al. 2019. “Immediate Effects of Rhythmic Auditory Stimulation on Gait Kinematics in Parkinson’s Disease ON/OFF Medication.” Clinical Neurophysiology 130 (10): 1789–97. 10.1016/j.clinph.2019.07.013. [DOI] [PubMed] [Google Scholar]
  8. Ghai S, Ghai I, Schmitz G, and Effenberg AO. 2018. “Effect of Rhythmic Auditory Cueing on Parkinsonian Gait: A Systematic Review and Meta-Analysis.” Sci Rep 8 (January):506. 10.1038/s41598-017-16232-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Goetz Christopher G., Tilley Barbara C., Shaftman Stephanie R., Stebbins Glenn T., Fahn Stanley, Pablo Martinez-Martin Werner Poewe, Sampaio Cristina, Stern Matthew B., Dodel Richard, Dubois Bruno, Holloway Robert, Jankovic Joseph, Kulisevsky Jaime, Lang Anthony E., Lees Andrew, Leurgans Sue, LeWitt Peter A., Nyenhuis David, Olanow C. Warren, Rascol Olivier, Schrag Anette, Teresi Jeanne A., van Hilten Jacobus J., and LaPelle Nancy. 2008. “Movement Disorder Society-Sponsored Revision of the Unified Parkinson’s Disease Rating Scale (MDS-UPDRS): Scale Presentation and Clinimetric Testing Results.” Movement Disorders : Official Journal of the Movement Disorder Society 23 (15): 2129–70. 10.1002/mds.22340. [DOI] [PubMed] [Google Scholar]
  10. ———. 2008. “Movement Disorder Society-Sponsored Revision of the Unified Parkinson’s Disease Rating Scale (MDS-UPDRS): Scale Presentation and Clinimetric Testing Results.” Movement Disorders: Official Journal of the Movement Disorder Society 23 (15): 2129–70. 10.1002/mds.22340. [DOI] [PubMed] [Google Scholar]
  11. Grahn JA, and Brett M. 2009. “Impairment of Beat-Based Rhythm Discrimination in Parkinson’s Disease.” Cortex 45 (January):54–61. 10.1016/j.cortex.2008.01.005. [DOI] [PubMed] [Google Scholar]
  12. Halpern AR, and Zatorre RJ. 1999. “When That Tune Runs through Your Head: A PET Investigation of Auditory Imagery for Familiar Melodies.” Cereb Cortex 9 (November):697–704. [DOI] [PubMed] [Google Scholar]
  13. Harrison Elinor C, and Earhart Gammon M. 2023. “The Effect of Auditory Cues on Gait Variability in People with Parkinson’s Disease and Older Adults: A Systematic Review.” Neurodegenerative Disease Management 13 (2): 113–28. 10.2217/nmt-2021-0050. [DOI] [PubMed] [Google Scholar]
  14. Harrison Elinor C., Horin Adam P., and Earhart Gammon M.. 2018. “Internal Cueing Improves Gait More than External Cueing in Healthy Adults and People with Parkinson Disease.” Scientific Reports 8 (1): 15525. 10.1038/s41598-018-33942-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. ———. 2019. “Mental Singing Reduces Gait Variability More Than Music Listening for Healthy Older Adults and People With Parkinson Disease.” Journal of Neurologic Physical Therapy 43 (4): 204. 10.1097/NPT.0000000000000288. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Harrison Elinor C., Horin Adam P., Myers Peter S., Rawson Kerri S., and Earhart Gammon M.. 2020. “Changes in Parkinsonian Gait Kinematics with Self-Generated and Externally-Generated Cues: A Comparison of Responders and Non-Responders.” Somatosensory & Motor Research, January, 1–8. 10.1080/08990220.2020.1713740. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Harrison Elinor C., McNeely Marie E., and Earhart Gammon M.. 2017. “The Feasibility of Singing to Improve Gait in Parkinson Disease.” Gait Posture 53 (March):224–29. 10.1016/j.gaitpost.2017.02.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Hausdorff JM, Lowenthal J, Herman T, Gruendlinger L, Peretz C, and Giladi N. 2007. “Rhythmic Auditory Stimulation Modulates Gait Variability in Parkinson’s Disease.” Eur J Neurosci 26 (October):2369–75. 10.1111/j.1460-9568.2007.05810.x. [DOI] [PubMed] [Google Scholar]
  19. Hine Kyoko, Abe Koki, Kinzuka Yuya, Shehata Mohammad, Hatano Katsunobu, Matsui Toshie, and Nakauchi Shigeki. 2022. “Spontaneous Motor Tempo Contributes to Preferred Music Tempo Regardless of Music Familiarity.” Frontiers in Psychology 13 (November):952488. 10.3389/fpsyg.2022.952488. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Hoehn MM, and Yahr MD. 1998. “Parkinsonism: Onset, Progression, and Mortality. 1967.” Neurology 50 (2): 318 and 16 pages following. 10.1212/wnl.50.2.318. [DOI] [PubMed] [Google Scholar]
  21. Horin Adam P., Harrison Elinor C., Rawson Kerri S., and Earhart Gammon M.. 2020. “People with Parkinson Disease with and without Freezing of Gait Respond Similarly to External and Self-Generated Cues.” Gait & Posture 82 (October):161–66. 10.1016/j.gaitpost.2020.09.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Howe TE, Lovgreen B, Cody FW, Ashton VJ, and Oldham JA. 2003. “Auditory Cues Can Modify the Gait of Persons with Early-Stage Parkinson’s Disease: A Method for Enhancing Parkinsonian Walking Performance?” Clin Rehabil 17 (July):363–67. [DOI] [PubMed] [Google Scholar]
  23. Leman Marc, Moelants Dirk, Varewyck Matthias, Styns Frederik, van Noorden Leon, and Martens Jean-Pierre. 2013. “Activating and Relaxing Music Entrains the Speed of Beat Synchronized Walking.” PloS One 8 (7): e67932. 10.1371/journal.pone.0067932. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Leow LA, Parrott T, and Grahn JA. 2014. “Individual Differences in Beat Perception Affect Gait Responses to Low- and High-Groove Music.” Front Hum Neurosci 8:811. 10.3389/fnhum.2014.00811. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Leow LA, Rinchon C, and Grahn J. 2015. “Familiarity with Music Increases Walking Speed in Rhythmic Auditory Cuing.” Ann N Y Acad Sci 1337 (March):53–61. 10.1111/nyas.12658. [DOI] [PubMed] [Google Scholar]
  26. McIntosh GC 1997. “Rhythmic Auditory-Motor Facilitation of Gait Patterns in Patients with Parkinson’s Disease.” J Neurol 62:22–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Morris M, Iansek R, Matyas T, and Summers J. 1998. “Abnormalities in the Stride Length-Cadence Relation in Parkinsonian Gait.” Mov Disord 13 (January):61–69. 10.1002/mds.870130115. [DOI] [PubMed] [Google Scholar]
  28. Morris Meg, Iansek Robert, McGinley Jennifer, Matyas Thomas, and Huxham Frances. 2005. “Three-Dimensional Gait Biomechanics in Parkinson’s Disease: Evidence for a Centrally Mediated Amplitude Regulation Disorder.” Movement Disorders 20 (1): 40–50. 10.1002/mds.20278. [DOI] [PubMed] [Google Scholar]
  29. Morris Rosie, Stuart Samuel, Grace McBarron Peter C. Fino, Mancini Martina, and Curtze Carolin. 2019. “Validity of Mobility Lab (Version 2) for Gait Assessment in Young Adults, Older Adults and Parkinson’s Disease.” Physiological Measurement 40 (9): 095003. 10.1088/1361-6579/ab4023. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Moussard Aline, Bigand Emmanuel, Belleville Sylvie, and Peretz Isabelle. 2014. “Learning Sung Lyrics Aids Retention in Normal Ageing and Alzheimer’s Disease.” Neuropsychological Rehabilitation 24 (6): 894–917. 10.1080/09602011.2014.917982. [DOI] [PubMed] [Google Scholar]
  31. Nieuwboer A, Rochester L, Herman T, Vandenberghe W, Emil GE, Thomaes T, and Giladi N. 2009. “Reliability of the New Freezing of Gait Questionnaire: Agreement between Patients with Parkinson’s Disease and Their Carers.” Gait & Posture 30 (4): 459–63. 10.1016/j.gaitpost.2009.07.108. [DOI] [PubMed] [Google Scholar]
  32. Nieuwboer Alice, Rochester Lynn, Herman Talia, Vandenberghe Wim, Emil George Ehab, Thomaes Tom, and Giladi Nir. 2009. “Reliability of the New Freezing of Gait Questionnaire: Agreement between Patients with Parkinson’s Disease and Their Carers.” Gait & Posture 30 (4): 459–63. 10.1016/j.gaitpost.2009.07.108. [DOI] [PubMed] [Google Scholar]
  33. Nisenzon AN, Robinson ME, Bowers D, Banou E, Malaty I, and Okun MS. 2011. “Measurement of Patient-Centered Outcomes in Parkinson’s Disease: What Do Patients Really Want from Their Treatment?” Parkinsonism Relat Disord 17 (February):89–94. 10.1016/j.parkreldis.2010.09.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Peper CL, Oorthuizen JK, and Roerdink M. 2012. “Attentional Demands of Cued Walking in Healthy Young and Elderly Adults.” Gait & Posture 36 (3): 378–82. 10.1016/j.gaitpost.2012.03.032. [DOI] [PubMed] [Google Scholar]
  35. Picelli A, Camin M, Tinazzi M, Vangelista A, Cosentino A, Fiaschi A, and Smania N. 2010. “Three-Dimensional Motion Analysis of the Effects of Auditory Cueing on Gait Pattern in Patients with Parkinson’s Disease: A Preliminary Investigation.” Neurol Sci 31 (August):423–30. 10.1007/s10072-010-0228-2. [DOI] [PubMed] [Google Scholar]
  36. Rennie Linda, Löfgren Niklas, Moe-Nilssen Rolf, Opheim Arve, Dietrichs Espen, and Franzén Erika. 2018. “The Reliability of Gait Variability Measures for Individuals with Parkinson’s Disease and Healthy Older Adults - The Effect of Gait Speed.” Gait & Posture 62:505–9. 10.1016/j.gaitpost.2018.04.011. [DOI] [PubMed] [Google Scholar]
  37. Repp BH, and Su YH. 2013. “Sensorimotor Synchronization: A Review of Recent Research (2006–2012).” Psychon Bull Rev 20 (June):403–52. 10.3758/s13423-012-0371-2. [DOI] [PubMed] [Google Scholar]
  38. Rochester Lynn, Hetherington Victoria, Jones Diana, Nieuwboer Alice, Willems Anne-Marie, Kwakkel Gert, and Van Wegen Erwin. 2005. “The Effect of External Rhythmic Cues (Auditory and Visual) on Walking during a Functional Task in Homes of People with Parkinson’s Disease.” Archives of Physical Medicine and Rehabilitation 86 (5): 999–1006. 10.1016/j.apmr.2004.10.040. [DOI] [PubMed] [Google Scholar]
  39. Roerdink M, Bank PJ, Peper CL, and Beek PJ. 2011. “Walking to the Beat of Different Drums: Practical Implications for the Use of Acoustic Rhythms in Gait Rehabilitation.” Gait Posture 33 (April):690–94. 10.1016/j.gaitpost.2011.03.001. [DOI] [PubMed] [Google Scholar]
  40. Rossignol S, and Melvill Jones G. 1976. “Audio-Spinal Influence in Man Studied by the H-Reflex and Its Possible Role on Rhythmic Movements Synchronized to Sound.” Electroencephalography and Clinical Neurophysiology 41 (1): 83–92. 10.1016/0013-4694(76)90217-0. [DOI] [PubMed] [Google Scholar]
  41. Satoh M, and Kuzuhara S. 2008. “Training in Mental Singing While Walking Improves Gait Disturbance in Parkinson’s Disease Patients.” Eur Neurol 60:237–43. 10.1159/000151699. [DOI] [PubMed] [Google Scholar]
  42. Ghai Shashank, Ghai Ishan, and Ghai Ishan Ghai Shashank. 2017. “Effect of Rhythmic Auditory Cueing on Aging Gait: A Systematic Review and Meta-Analysis.” Aging and Disease 9 (5): 901–23. 10.14336/AD.2017.1031. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Skodda Sabine, Flasskamp Andrea, and Schlegel Uwe. 2010. “Instability of Syllable Repetition as a Model for Impaired Motor Processing: Is Parkinson’s Disease a ‘Rhythm Disorder’?” Journal of Neural Transmission (Vienna, Austria: 1996) 117 (5): 605–12. 10.1007/s00702-010-0390-y. [DOI] [PubMed] [Google Scholar]
  44. Studenski S, Perera S, Patel K, Rosano C, Faulkner K, Inzitari M, Brach J, et al. 2011. “Gait Speed and Survival in Older Adults.” Jama 305 (1): 50–58. 10.1001/jama.2010.1923. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Thaut MH, McIntosh GC, Rice RR, Miller RA, Rathbun J, and Brault JM. 1996. “Rhythmic Auditory Stimulation in Gait Training for Parkinson’s Disease Patients.” Movement Disorders: Official Journal of the Movement Disorder Society 11 (2): 193–200. 10.1002/mds.870110213. [DOI] [PubMed] [Google Scholar]
  46. Thaut MH, McIntosh KW, McIntosh GC, and Hoemberg V. 2001. “Auditory Rhythmicity Enhances Movement and Speech Motor Control in Patients with Parkinson’s Disease.” Functional Neurology 16 (2): 163–72. [PubMed] [Google Scholar]
  47. Tudor-Locke Catrine, Jose Mora-Gonzalez Scott W. Ducharme, Aguiar Elroy J., Schuna John M., Barreira Tiago V., Moore Christopher C., et al. 2021. “Walking Cadence (Steps/Min) and Intensity in 61–85-Year-Old Adults: The CADENCE-Adults Study.” The International Journal of Behavioral Nutrition and Physical Activity 18 (September):129. 10.1186/s12966-021-01199-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Vitorio Rodrigo, Stuart Samuel, Gobbi Lilian T. B., Rochester Lynn, Alcock Lisa, and Pantall Annette. 2018. “Reduced Gait Variability and Enhanced Brain Activity in Older Adults With Auditory Cues: A Functional Near-Infrared Spectroscopy Study.” Neurorehabilitation and Neural Repair 32 (11): 976–87. 10.1177/1545968318805159. [DOI] [PubMed] [Google Scholar]
  49. Willems AM, Nieuwboer A, Chavret F, Desloovere K, Dom R, Rochester L, Jones D, Kwakkel G, and Van Wegen E. 2006a. “The Use of Rhythmic Auditory Cues to Influence Gait in Patients with Parkinson’s Disease, the Differential Effect for Freezers and Non-Freezers, an Explorative Study.” Disability and Rehabilitation 28 (11): 721–28. 10.1080/09638280500386569. [DOI] [PubMed] [Google Scholar]
  50. ———. 2006b. “The Use of Rhythmic Auditory Cues to Influence Gait in Patients with Parkinson’s Disease, the Differential Effect for Freezers and Non-Freezers, an Explorative Study.” Disabil Rehabil 28 (June):721–28. 10.1080/09638280500386569. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Table 1

Supplemental Table 1: Correlations between optimal cue rate and participant demographics.

NIHMS2069424-supplement-Supplementary_Table_1.docx (15KB, docx)

RESOURCES