Table 1.
Studies analyzing the effects of rhythmic auditory cueing on gait.
| Author | Sample description, age: (M ± S.D years) | PEDro score | Assessment tools | Research design | Auditory feedback elements | Conclusion |
|---|---|---|---|---|---|---|
| Dotov, et al. [100] | 7F, 12M (60) | 6 | Coefficient of variation of inter-stride interval, cadence, gait velocity, stride length, DFA of short-long term series of inter-response-interval correlations, circular statistics for synchronization of footfall & beat | Pre-test, gait performance with/without RAC (no variability, biological variability, non-biological variability; randomized), post-test | RAC with no variability, biological variability & non-biological variability at +10% of preferred cadence Magnitude of biological & non-biological variability: 2% of inter-beat-interval Metronome sequence: triangle timbre Musical excerpts Amplitude modulated noise: Modulated on musical excerpt with drum ensemble, discarding tonal information |
Significant enhancement in coefficient of variation for inter-stride interval after RAC in all conditions. Significant effect of RAC that was amplitude modulated for biological variability as compared to IC on short-long term correlation for term series of inter-response-interval correlations. Enhanced synchronization, cadence but reduced short-long term correlation for term series of inter-response-interval correlations during metronome based IC as compared to feedback with amplitude modulated for biological variability. |
| Maculewicz, et al. [141] | 5F, 15M (24.4±3.2) | 4 | Mean square error for the asynchrony between target & performed measure & trend of tempo change obtained from slope of line fitted to measured tempo, questionnaire | Gait performance with/without real-time auditory feedback (adaptive), RAC (constant) &/or haptic feedback, with instructions to perform gait at preferred cadence or the tempo of the sound | Real-time auditory feedback (adaptive), RAC (constant) by sine, wood & gravel sounds | Significantly enhanced step wise interaction with real time auditory feedback with (sinusoid >wood>gravel). Significant reduction in asynchrony with audio-haptic feedback & real-time auditory feedback as compared to no feedback. Significant enhancement in comfort for perceiving haptic & audio-haptic feedback as compared to haptic only or no feedback in self-reported questionnaire. |
| Schreiber, et al. [97] | 5F, 12M (37.4±15.7) | 4 | Cadence, gait speed, rhythmicity, stance time, double support time, gait symmetry, step length, stride length, step width, EMG activity of (tibialis anterior, soleus, gastrocnemius medialis, vastus medialis, rectus femoris, semitendinosus, gluteus medius & gluteus maximus), kinematics for pelvis, hips, knees & ankle joint (sagittal, frontal, transverse plane) | Gait performance with/without RAC cueing at preferred, reduced cadence (instructions & cueing randomized) | RAC at preferred & reduced cadence | Significantly reduced gait speed with RAC at preferred cadence as compared to preferred speed gait without cueing. No effect of RAC on cadence, rhythmicity, stance time, double support time, gait symmetry for RAC at preferred or reduced cadence as compared to no cueing. Significantly reduced step width with RAC at reduced cadence as compared to reduced speed gait without cueing. Significantly enhanced step length with RAC at reduced cadence as compared to reduced speed gait without cueing. Significant differences for ankle dorsiflexion, hip flexion & hip abduction of the gait cycle with RAC at reduced cadence as compared to reduced speed gait without cueing. |
| Hamacher, et al. [104] | Young: 8F, 12M (24.9±4.1) Old: 11F, 9M (67.4±5.3) |
5 | Stride length, minimum foot clearance, stride time, stride to stride analysis (mean & coefficient of variation) | Gait performance with/without dual-task (arithmetic subtraction in 3’s task) &/RAC (randomized) | RAC at preferred cadence | Significant enhancement in stride length, stride time with RAC (with/without dual-task) in both younger & older adults. Significantly enhanced coefficient of variation of stride time in older participants under dual-task condition & with RAC Enhancement in coefficient of variation of stride to stride in older participants under dual-task condition & with RAC |
| Terrier [96] | 22F, 14M (33±10) | 4 | DFA of coefficient of variability for stride time, stride length, stride speed, stride length, stride speed & stride time | Gait performance on treadmill with/without visual (stepping stones), RAC | RAC at preferred cadence | Significant reduction in stride time & stride speed with RAC as compared to no cueing. No effect on coefficient of variation for stride length, stride time & stride speed (mean & coefficient of variation) with RAC |
| Roerdink, et al. [162] | 5F, 7M (28±6) | 5 | Stride-to-stride DFA for persistence of stride time, stride length, stride speed & anterior-posterior center of pressure sway | Treadmill gait performed with/without RAC with isochronous metronome & non-isochronous metronome containing inter-beat interval sequences with distinct scaling exponents (randomized) | RAC with (IC) containing equidistant inter-beat interval & 4 (non-isochronous) metronome containing inter-beat interval sequences with distinct scaling exponents Frequency: 600Hz RAC with mean inter-beat intervals being equal to mean stride time of preferred cadence. |
Significant effect of IC cueing for changing the stride-to-stride fluctuations of stride length & stride time to anti-persistent & vice versa for the non-IC. Significant effect of isochronous & non-isochronous metronome cueing for changing the stride-to-stride fluctuations of stride speed to anti-persistent for both the cueing. |
| Wright, Spurgeon and Elliott [163] | 8F, 2M (20-33) | 5 | Mean asynchrony, step time variability & mean percentage step correction | Gait performance with/without RAC &/or visual cueing | RAC, 500 ms (cue duration 30 ms), 800Hz | Significant enhancement in & mean percentage step correction with audio & audio-visual cueing as compared to only visual cueing Significant reduction in mean synchrony of step with RAC with audio-visual cueing as compared to only audio or visual cueing. Significant reduction in step time variability with audio & audio-visual cueing as compared to only visual cueing |
| Young, et al. [138] | 6F, 4M (63.9±4) II: same as I III: same as I |
5 | I: Mean step length, % change stride length, mean step duration, % change in variability of stride length, duration II: same as I III: same as I |
I: Gait performance with/without verbal instruction, verbal instruction-metronome cueing, stepping sound, stepping sound-verbal instructions, for small and wide stride length (randomized) II: Gait performance with/without stepping sound, verbal instruction-stepping sound feedback, synthesized gravel sound, synthesized gravel sound-verbal instructions, for small and wide stride length (randomized) III: Gait performance with/without motor imagery, motor imagery-stepping sound feedback, synthesized gravel sound, synthesized gravel sound-motor imagery, for small and wide stride length (randomized) |
I: RAC (Ct: 550-649ms, Exp: 600-700ms), foot step feedback on gravel (500, 600, 700ms) II: RAC (Ct: 550-649ms, Exp: 600-700ms), foot step feedback on gravel (500, 600, 700ms), synthesized gravel step sound corresponding to plantar force (developed by using ground reaction forces vector to modulate both intensity envelop, and central frequency of bandpass filter applied to stochastic noise impulse signal) III: same as II |
Significant enhancement in stride length for healthy Ct in all cueing conditions. No effect of auditory cueing or instructions on mean step duration. Significant reduction in stride length variability with synthesized feedback as compared to footstep feedback-verbal instruction, synthesized feedback-verbal instructions. Significant reduction in stride length variability with stepping, synthesized feedback, stepping-verbal instructions. Significant enhancements in stride length with rhythmic auditory cueing (synthesized) and motor imagery together. No effect on stride duration parameters. |
| Leow, et al. [105] | 24F, 19M (18-20) | 5 | Stride velocity, step length, step time, stride width, double support, & coefficient of variability for stride length | Gait performance with/without rhythmic music, RAC (low/high groove) at 0% & +22.5% of preferred cadence | RAC (low/high groove music) at 0% & +22.5% of preferred cadence (50ms 1kHz sine tones) | Significant enhancement in stride velocity with rhythmic music cueing (high groove) & metronome at +22.5% of preferred cadence as compared to no cueing. Significant reduction in double support with metronome cueing at 0% & +25% of preferred cadence as compared to no cueing. Significant reduction in step length with high groove music at +25% of preferred cadence. Significant reduction in step time in low (0% also), high groove music cueing & RAC at +25% of preferred cadence cueing as compared to no cueing. Significant enhancement in coefficient of variability for stride length with low & high groove RAC at 0% & +25% of preferred cadence. |
| Sejdić, et al. [164] | 8F, 7M (23.9±4.7) | 5 | Gait speed, mean stride interval, stride interval variability, stride interval dynamics, dynamic stability of gait in anterior-posterior, vertical & medio-lateral dimension (short: between 0th & 1st stride & long: between 4th & 10th stride, term Lyapunov exponent) | Gait performance with rhythmic auditory, visual & haptic cueing (randomly spate or together) at preferred cadence during 2 sessions | RAC at preferred cadence | Significantly reduced stride interval variability with RAC (alone & combined with visual & haptic cueing) as compared to no cueing condition. Significantly reduced stride interval dynamics (long term Lyapunov exponent) with RAC (alone & combined with visual & haptic cueing) as compared to no cueing condition. Significant enhancement in dynamic stability of gait with RAC (alone & combined with visual & haptic cueing) as compared to no cueing condition. |
| Terrier and Dériaz [165] | 10F, 10M (36±11) | 4 | DFA on time series of stride time, stride length & stride speed Short & long-term local dynamic stability in anterior-posterior & medial-lateral direction |
Gait performance on treadmill at slow (0.7 times preferred cadence), fast (1.3 times preferred cadence) & at preferred cadence with/without RAC (randomly) | RAC at slow (0.7 times slower than preferred cadence), fast (1.3 times faster than preferred cadence) cadence | Significant enhancement in long term local dynamic stability with RAC Significant reduction of stride time & stride length variability at slow speed with RAC No effect on short term local dynamic stability with RAC |
| Roerdink, et al. [166] | 10F, 10M (63.2±3.6) | 5 | Cadence, mean relative timing between footfalls & auditory stimuli, variability of mean relative timing (by circular statistics) | Participants performed gait at preferred cadence followed by 7 random trials with adjusted RAC i.e. 77.5%, 85%, 92.5%, 100%, 107.5%, 115% or 122.5% | Auditory input from drum RAC at 77.5%, 85%, 92.5%, 100%, 107.5%, 115% or 122.5% of preferred cadence Different pitch to pace for RAC i.e. for step left: 440Hz, right: 1000Hz |
Significant effect of RAC on cadence, mean relative timing & variability of mean relative timing between footfalls & auditory inputs. Significantly fewer steps required to reach synchronization |
| Lohnes and Earhart [67] | Young: 7F, 4M (24±0.8) Old: 7F, 4M (70.8±10.4) |
5 | Gait velocity, cadence & stride length | Patients performed gait with/without RAC at -10%, +10% of preferred cadence or with additional cueing strategy “think about larger strides” with/without -10% & +10% of auditory inputs tone, with/without dual-task “word generation task” | RAC at ±10% of preferred cadence. | Significant effect on gait velocity stride length, cadence for both groups with ±10% of RAC under both single and dual-task conditions. Larger effects noted in young participants as compared to older counterparts. Verbal instructions had no influence on cadence among both groups under both single and dual-task conditions. |
| Trombetti, et al. [167] | Exp: 64F, 2M (75±8) Ct: 65F, 3M (76±6) |
8 | Gait velocity, stride length, cadence, double, single support phase, stride time/length variability, TUG test, trunk angular displacement, Tinetti tests & assessment of falls | Exp: Pre-test, gait & exercise training with auditory input performed for 1-hour session/week for 12 months, 6-month test, post-test, with/without dual-task (counting backward aloud task) Ct: started 6-month delayed intervention, with/without dual-task (counting backward aloud task) |
RAC as piano music | Single task: Significant enhancement in gait velocity, stride length & stride time variability for the Exp as compared to Ct. Dual-task: Significant enhancement in stride length, decrease in stride length variability in Exp as compared to Ct Significant enhancement in 1 legged stance, Tinetti tests, TUG & decreased mediolateral angular velocity. Significantly reduced incidences of falls in Exp as compared to Ct. |
| Wittwer, et al. [136] | 12F, 7M (79±7.8) | 4 | Swing time, stride time, velocity, stride length, double support %, stride width, stride length & time variability | Participants performed gait with/without auditory feedback “randomly” i.e. music or RAC | Music or metronome or RAC at participants preferred cadence | Significant enhancement in velocity, stride length with music as compared to no sound. Significant reduction in stride time, double limb support & enhancement in cadence with both music & RAC input, as compared to no auditory input. No effect on mean step width, mean temporal or spatial gat variability. |
| Yu, et al. [93] | 13F (21.8±0.4) | 5 | Stride length, cadence & gait speed | Gait performance with/without RAC at 0% & ±10% of preferred cadence | RAC at 0% & ±10% of preferred cadence | Significant enhancement in stride length, cadence & gait speed with +10% RAC as compared to all conditions. Significant reduction in cadence & gait speed with -10% of RAC as compared to 0% & no cueing. |
| Almeida, et al. [92] | Exp I: 9 (42.7±6.6) Exp II: 10 (42.4±4.5) Ct: 9 (41.7±5) |
4 | Gait speed, heart rate, maximal oxygen consumption, rating of perceived exertion | Gait performance with/without (Ct) RAC at 90 bpm (Exp II) & 140 bpm (Exp I) for 30 minutes with re-tests at every 5-minute interval | RAC at 90 & 140 bpm | Significant enhancement in gait performance in Exp I as compared to Exp II & Ct. No effect on heart rate & maximal oxygen consumption in Exp or Ct. |
| Hunt, McGrath and Stergiou [168] | 4F, 6M (28.1±5.3) | 4 | Stride time, sample entropy of stride time interval for individualized fractal RAC, DFA for auditory signals scaling exponent & stride time scaling exponent | Gait performance with/without individualized fractal RAC for white, pink & brown noise (randomized) | Individualized fractal RAC (embedding white, pink & brown noise variables into inter-beat interval of music) Inter-beat interval: stretched or compressed based on dynamics of pink, white or brown noise time series Amplitude: standard deviation of inter-beat intervals matched standard deviation of step time Tempo: at preferred cadence |
Significant effect of RAC on sample entropy of stride interval time series (brown>pink>white>no sound) Significant enhancement of fractal scaling exponent with auditory feedback of stride interval time series (brown>pink>white>no sound) |
| Marmelat, et al. [169] | 7F (28±6) | 5 | DFA of inter slide interval variability, inter-beat interval variability & asynchrony with metronome between two successive right heel strikes | Gait performed on treadmill with/without RAC with either IC or fractal feedback | RAC with either IC or fractal feedback Inter-beat intervals contained fractal Gaussian noise with corresponding scaling exponent (600 Hz) |
Significant effects of pacing rhythmic metronome feedback on global exponents of inter-beat & slide intervals (persistent correlations) No effect on inter slide interval, asynchrony with RAC Participants anticipated the metronome & adapted with pacing stimuli No significant correlations between inter-beat intervals & inter-slide intervals (increased correlation with increased variability) |
| 5F, 7M (28±6) | 5 | DFA of inter slide interval variability, inter-beat interval variability & asynchrony with metronome between two successive right heel strikes | Gait performed on treadmill with/without RAC with either IC or fractal feedback | RAC with non-IC (different scaling exponents) | Significant effects of pacing rhythmic metronome feedback on global exponents of inter-slide intervals (anti-persistent correlations) No significant correlations between inter-beat intervals & interslide intervals (increased correlation with increased variability) |
|
| Franěk, et al. [68] | 30F, 42M (20.2±1.2) | 4 | Gait speed, synchronization (inter step times) | Gait performed with/without rhythmic music feedback at 114, 124, 133 bpm | RAC at 114, 124, 133 bpm | Significant enhancement in gait speed with faster tempo music feedback as compared to slower tempo RAC & no feedback. No effect on synchronization with rhythmic music feedback. |
| 60F, 61M (20.6±1.5) | 4 | Gait speed, synchronization (inter step times) | Gait performed with/without] RAC (music motivational/non-motivational) | RAC (music motivational: 131-200 bpm, non-motivational: 52-96 bpm) | Significant enhancement in gait speed with motivational rhythmic music feedback as compared to non-motivational RAC & no feedback. | |
| Leman, et al. [142] | 11F, 7M (22-51) | 4 | Gait speed, gait tempo, synchronization of steps to tempo | Gait performance with 52 rhythmic music excerpts (activating & relaxing) | RAC (relaxing or activating effects) at 130 beats per minute, short fade in of 50 ms & fade out of 100 ms applied to each musical excerpt RAC superimposed at position 1, 12, 23, 34, 45, & 58 |
Significant effect of activating (increased gait speed), relaxing (reduced gait speed) in gait speed with RAC with same tempo. Significant enhancement in synchronization of steps with RAC |
| Peper, et al. [170] | Young: 4F, 8M (22-28) Old: 5F, 7M (55-69) |
5 | Mean reaction time, gait speed, step length, step width | Gait performed with/without RAC & visual feedback (stepping stones), dual-task (probe reaction task generating vibrating stimuli) | RAC Left (440Hz), right (1000Hz) Temporal shift of ±1/6th of interval between consecutive ipsilateral beeps, causing ±60º phase delay/advance |
Significantly enhanced step length & step width RAC No effect on gait speed in young & older adults with RAC Significantly enhanced reaction times with RAC as compared to no cueing. Significantly reduced reaction time with RAC as compared to visual cueing. |
| Bank, Roerdink and Peper [171] | 10F, 10 M (63.2±3.6) | 5 | Mean normalized step time, step length, relative phase shift between gait & cues | Gait performance with RAC ±22.5% (introduced in steps of ±7.5% randomly) of preferred cadence &/or stepping stone visual feedback | RAC at ±22.5% of preferred cadence Temporal shift of ±1/6th of interval between consecutive ipsilateral beeps, causing ±60º phase delay/advance |
Significant effect of phase delay on increasing/decreasing step length, step time with auditory & visual feedback. However visual cueing > RAC Significantly enhanced phase shift from auditory to visual cueing condition. Significant reduction in coordination of RAC with gait as compared to visual cueing |
| Wellner, et al. [91] | 17 (28±8) | 4 | Obstacle hit %, average obstacle clearance & individually chosen gait speed | Gait performance on robot assisted device with/without Rhythmic auditory feedback (distance to obstacle &/or foot clearance feedback) | Rhythmic real-time feedback for distance to obstacle & foot clearance Obstacle distance: Rhythm (repeating sound with shorter pause interval as distance decreases), continuous/discrete pitch (continuous sound with higher pitch as distance increases/decreases), dynamics (increase in volume as distance decreases) Absolute foot clearance: harmony (dissonant/consonant chords below/above obstacle), pitch with 2 & 3 levels, noise (Gaussian noise below, no sound above obstacle) |
Significantly enhanced self-chosen gait speed with auditory feedback as compared to only visual feedback. Significant enhancement in gait speed with rhythmic feedback for distance to obstacle &/or foot clearance as compared to no feedback |
| Arias and Cudeiro [102] | 6F, 5M (65.7±7.6) | 5 | Cadence, gait velocity, step amplitude, coefficient of variation for step amplitude & stride time | Patients performed gait with/without rhythmic cueing from auditory, visual & audio-visual condition, with frequency ranging from 70-110% increment/decrement at ±10% of preferred cadence | RAC with wave frequency of 4625 Hz delivered at frequency ranging from 70-110% increment/decrement at ±10% of preferred cadence | Significant enhancement in cadence, step amplitude in Ct with RAC No effects on gait velocity, coefficient of variability for stride time & stride amplitude. |
| Baker, et al. [172] | 7F, 5M (71.5±2.5) | 7 | Gait speed, coefficient of velocity for (step time, double limb support time) | Pre-test, functional gait performance with/without RAC -10% of preferred cadence, attentional cue instructions "try to take big steps", together "take a big step with the beat", & with/without a dual-task (a tray with 2 cups of water on top), post-test | RAC at -10% of preferred cadence | Significant effect of RAC back and verbal instructions on enhancing stride length, gait velocity. Significantly reduced cadence with RAC and verbal instructions. Reduced gait speed, cadence with -10% RAC No effect on stride length. |
| Hausdorff, et al. [117] | 14F, 12M (64.6±6.8) | 5 | Stride time, gait speed, stride length, swing time, stride time variability & swing time variability | Pre-test, gait performance with/without RAC at preferred cadence, +10%, Post-test 2 & 15 min short term retention test | RAC at 0% & +10% of preferred cadence | Significant enhancement in gait speed with +10% RAC Significant reduction in stride time with +10% RAC No effect on stride length, swing time, stride time variability, swing time variability with RAC |
| Willems, et al. [103] | 9 (68.1±7.3) | 5 | Steps (number, time, height, width, length), step length, step width, step duration, coefficient of variation of step duration | Gait performance while turning with/without RAC | RAC at preferred cadence | Enhancement in step length. No effects on steps (number, time, height, width), step length, step width, step duration, coefficient of variation of step duration with RAC |
| Baram and Miller [99] | 6F, 5M (25.4±1.9) | 4 | Gait speed, stride length, 10 meters walking test | Pre-test, followed by rhythmic auditory feedback & 10 min follow-up short term residual performance test | Rhythmic auditory feedback generated with gait step in real-time | No effects on stride length and gait velocity with rhythmic feedback generated in real-time |
| Willems, et al. [173] | 10 (67.2±9.1) | 4 | Step frequency, gait speed, stride length & double support (%) phase | Pre-test, gait performance at 0%, -20%, -10%, +10%, +20% of RAC (randomized), post-test | RAC at 0%, -20%, -10%, +10%, +20% preferred cadence | Significant effect of RAC on cadence, gait speed, with 0%, -10%, +10%, +20% pacing of RAC No significant effects on double limb support, stride length |
| Baker, et al. [101] | 7F, 4M (71.5±2.5) | 6 | Gait speed, step amplitude & step frequency | Pre-test, functional gait performance with/without RAC -10% of preferred cadence, attentional cue instructions "try to take big steps", together "take a big step with the beat", & with/without a dual-task (a tray with 2 cups of water on top), post-test | RAC at -10% of preferred cadence | Significant effect of RAC & attentional cue "big steps with beat" on step frequency in gait speed (single-task only), step amplitude, step frequency in Ct in both single & dual-task conditions Non-significant effects on gait speed, step amplitude & step frequency with RAC only. Effects not evitable once the RAC was removed, in post-test |
| Rochester, et al. [94] | 4F, 6M (63.5±7) | 6 | Step length, step frequency, walking speed, time duration & cadence | Complex functional walking & sitting task under single & dual-motor task (carrying a tray) condition with/without RAC | RAC generated per preferred speed of patients. | No effects of RAC on gait speed, step length & cadence under single/dual-task conditions. However, reduction in cadence under dual-task conditions with RAC |
| Thaut, et al. [174] | 10F, 6M (25-40) | 4 | Stride symmetry, stride duration & EMG amplitude variability (Gastrocnemius) | Gait performance tested with/without RAC 3 times for 5 weeks | RAC at 4/4-time signature (1st & 3rd beat accentuated by tambourine beat, 70dB) at preferred cadence, at slower, faster than preferred cadence | Significant enhancement in stride rhythmicity between right & left limb with RAC Significantly delayed & shortened onset of gastrocnemius EMG activity with RAC Significant reduction in EMG variability of gastrocnemius muscle with RAC Significantly enhanced integrated amplitude ratios for gastrocnemius EMG activity |
| McIntosh, et al. [175] | 6F, 4M (72±5) | 4 | Gait velocity, stride length, cadence & cadence-auditory stimulus synchronization | Gait performance by participants with pre-test, with & without RAC at +10% of preferred cadence, post-test | RAC at 0%, +10% of preferred cadence | Significant enhancement in gait velocity and cadence with RAC Enhancement in stride length. No effect on gait symmetry |
F: Female, M: Male, Exp: Experimental group, Ct: Control group, RAC: Rhythmic auditory cueing, DFA: Detrended Fluctual Analysis, PD: Parkinson’s disease, EMG: Electromyography, IC: Isosynchronous cueing, bpm: beats per minute.