ABSTRACT
-
Objective
Spontaneous motor tempo (SMT), observed in walking, tapping and clapping, tends to occur around 2 Hz. Initiating and controlling movement can be difficult for people with Parkinson’s (PWP), but studies have not identified whether PWP differ from controls in SMT. For community-based interventions, e.g. dancing, it would be helpful to know a baseline SMT to optimize the tempi of cued activities. Therefore, this study compared finger tapping (FT), toe tapping (TT) and stepping ‘on the spot’ (SS) in PWP and two groups of healthy controls [age-matched controls (AMC) and young healthy controls (YHC)], as SMT is known to change with age.
-
Methods
Participants (PWP; n = 30, AMC; n = 23, YHC; n = 35) were asked to tap or step on the spot at a natural pace for two trials lasting 40 seconds. The central 30 seconds were averaged for analyses using mean inter-onset intervals (IOI) and coefficient of variation (CoV) to measure rate and variability respectively.
-
Results
PWP had faster SMT than both control groups, depending on the movement modality: FT, F(2, 87) = 7.92, p < 0.01 (PWP faster than YHC); TT, F(2, 87) = 4.89, p = 0.01 (PWP faster than AMC); and SS, F(2, 77) = 3.26, p = 0.04 (PWP faster than AMC). PWP had higher CoV (more variable tapping) than AMC in FT only, F(2, 87) = 4.10, p = 0.02.
-
Conclusion
This study provides the first direct comparison of SMT between PWP and two control groups for different types of movements. Results suggest SMT is generally faster in PWP than control groups, and more variable when measured with finger tapping compared to stepping on the spot.
-
Keywords: Age; Finger tapping; Movement; Parkinson’s disease; Spontaneous motor tempo; Stepping; Toe tapping
Tempo refers to the rate at which something repeats with regularity over time [
1], and in music, tempo commonly corresponds to the percept of a rhythmic beat. Thus, musical tempo is often described quantitatively in terms of beats per minute (bpm). Typically, humans show a preference for music that has an interbeat interval of 500–600 ms (i.e., is performed between 100–120 bpm), so it is perhaps unsurprising that the rate of preferred repetitive movements (such as walking, tapping and/or clapping) is also within this tempo range [
2]. van Noorden and Moelants [
3] referred to this concept in general as the 2 Hz human resonance theory (i.e., two cycles per second or 500 ms interbeat intervals). Early studies have suggested that the most common spontaneous motor tempo (SMT) was around 600 ms [
4]. However, large individual differences in SMT have been observed, ranging from 300 ms to 800 ms intervals [
5]. Although SMT is very reliable (correlations of measures taken across time are between 0.75–0.95), it changes over one’s lifespan [
6]. Studies show that children between four and seven years old have a fast SMT (300–400 ms, or 200 bpm), adults’ SMT is slower (500–600 ms), and the SMT in older adults is even slower (approximately 700 ms or approximately 86 bpm) [
7-
9]. The SMT is also affected by factors other than age, such as genetics and musical training. Twin studies suggest that although identical twins have very similar SMTs, those of nonidentical siblings vary.4 Musical training increases the trajectory of change (i.e., slows the SMT) from childhood to adulthood [
8]. However, there is little evidence to suggest that an SMT is linked to an individual’s sex, handedness, body size, or heart rate [
7].
The notion of an SMT, also referred to as ‘preferred’, ‘optimal’ or ‘natural’ timing, is suggestive of a type of motor agency (in terms of linking feelings of control to volitional actions) that are problematic for people with Parkinson’s disease (PWP) [
10,
11]. This is because Parkinson’s disease (PD) affects both the perception of time and the production of timed motor activities (including initiating and regulating movements) due to the loss of dopamine-producing neurons in the substantia nigra pars compacta [
12]. The symptoms vary across individuals but can include tremor, postural instability, rigidity, akinesia and bradykinesia, resulting in functional difficulties when walking, such as freezing of gait, and/or festination (a hastening of steps difficult to stop), both of which can lead to high incidences of falling and associated complications [
13-
15]. Symptoms of PD also include nonmotor difficulties, such as depression, anxiety, apathy, disturbances in sleep and digestive cycles and cognitive decline [
16]. Although cognitive decline is often considered a problem related to later stages of PD, it is likely that the perception of time is disturbed in earlier stages of PD [
17].
Experimental studies of timing in PD have generally relied on a finger tapping paradigm used in synchronization-continuation tasks to assess timed motor production [
12]. In the task paradigm, synchronization occurs during a paced condition (i.e., tapping is guided by an external cue), whereas continuation occurs during an unpaced condition (tapping continues after the pacing cue ceases). Jones and Jahanshahi [
12] compiled these synchronization-continuation studies and found mixed results in terms of finger tapping rates during the continuation condition, which may suggest differences in SMT: PWP were either faster, slower or did not differ from controls in terms of unpaced finger tapping. However, the term unpaced is somewhat of a misnomer in synchronization-continuation task paradigms, as the tempo of the motor action has essentially been primed by the cue in the synchronization condition, which immediately precedes the unpaced (continuation) condition. To ascertain whether the SMT in PWP differs from that in controls, it is essential to measure timed motor movements in the absence of any cueing. Only two studies [
18,
19] have included a measure of explicitly spontaneous rather than self-paced movement prior to the presentation of stimuli to PWP, and both of these studies focused on finger tapping as the sole movement modality. Yahalom and colleagues [
18] reported no significant difference between PWP (
n = 51) who tapped their fingers at a rate of 680 ms (88 bpm) in comparison to controls (
n = 36) who tapped their fingers at a rate of 581 ms (103 bpm). Benoit and colleagues [
19] also found no difference in the SMT between PWP (
n = 15) and controls (
n = 10) in terms of the rate [PWP: mean = 580 ms, standard error of the mean (SEM) = 78.5 ms; Controls: mean = 600 ms, SEM = 63.9 ms] and variability [measured using the coefficient of variation (
CoV)] (PWP: mean
CoV = 0.05, SEM = 0.08; Controls: mean
CoV = 0.05, SEM = 0.04).
Although self-initiated activity can be challenging for PWP, interventions such as rhythmic auditory stimulation have shown that synchronization to external rhythmic sounds (such as metronomes or music) can improve cyclic movements, such as walking [
20]. This type of therapeutic approach includes identifying a clinical aim (such as increasing step length or reducing cadence) and then training a specific motor response to the sound cue, usually at either 10% above or below the individuals’ SMT [
21,
22]. Consequently, the tempi of the cueing stimuli should be considered relative to the type of movement because an important aim for treatments for PWP (e.g., Parkinson’s UK23) that are developed as individualized adjunct therapies is to improve the ability to perform functional movements in everyday life. Therefore, we compared the SMTs in three types of movements, toe tapping, stepping ‘on the spot’ (as a proxy for dancing), and finger tapping, in PWP, age-matched controls and young healthy controls to provide information on rehabilitation interventions for researchers, clinicians and practitioners. The first two types of movements are both types of movements that are typically related to music, and finger tapping is typically used in SMT and timed movement research.
MATERIALS & METHODS
This study was approved by the Health, Sciences, Engineering & Technology ECDA (Ethics Committee with Delegated Authority; Protocol Reference aLMS/SF/UH/02547) at the University of Hertfordshire. All participants provided written informed consent prior to the beginning of the study in accordance with the recommendations of the Helsinki Declaration.
- Participants
The sample was split into three groups: younger healthy controls [YHC; n = 36, 29 females, mean age 20.75 [standard deviation (SD) 3.18] years, age range 18–32 years]; age-matched (to the PWP group) controls [AMC; n = 26, 12 females, mean age 64.35 (SD 13.02) years, age range 32–78 years] and PWP [n = 30, 20 females, mean age 62.23 (SD 10.48) years, age range 34–77 years]. All participants underwent cognitive impairment assessments using the Mini Mental State Examination. The exclusion criterion was a score on this assessment of <24 [
24], and no participants were excluded on this basis.
The Parkinson’s group was tested during the ‘ON’ state of their stabilized medication. The average time since diagnosis was 67.27 months (just over 5.6 years, SD = 59.19 months). The time since diagnosis ranged from 5 months to 272 months (21 years). The Unified Parkinson’s Disease Rating Scale (UPDRS) [
25] was used to evaluate their current status. For the overall score of the UPDRS (max = 176), the Group mean was 25.57, and the SD = 10.15. The scores for the three factors were as follows: mentation, behavior and mood (max = 16), mea
n = 3.5, SD = 1.68; activities of daily living (max = 52), mea
n = 10.43, SD = 4.68; and motor examination (max = 108), mea
n = 11.63, SD = 5.64. The Schwab and England Activities of Daily Living Scale [
26] score for this sample ranged between 50 and 100% (mea
n = 82.33%, SD = 11.94%). The Hoehn and Yahr Scale [
27] mean score was 1.78 (SD = 0.83), ranging from 1–4 in this sample (0 = min, 5 = max). Current medications were also recorded.
Table 1 provides data for the PWP and relates the ascribed PD subtypes according to the established guidelines [
28] (further details provided in
Supplementary Table 1 in the online-only Data Supplement).
- Equipment
Finger and toe tapping data were collected using a stomp box [Acoustim8, Series 100, UK used by musicians (generally in acoustic music) to provide a bass drum sound. Full technical details are reported in Rose et al. [
29], 2019]. Heel strike data for stepping on the spot were gathered using BioPac (Biopac Systems Inc., Goleta, CA, USA) heel and toe strike transducers (Model RX111) attached to BioNomadix ankle sensors (Model BN-TX STRK2-T). The MP150 unit communicated with a UIM100C unit (for tapping) and two BioNomadix STRK2-R units (for stepping). A metal thimble provided auditory feedback for participants during the finger tapping condition. During the toe tapping and stepping conditions, the participants could hear the sounds of the transducers striking the stomp box or the floor.
- Procedure
The participants first provided demographic information and completed the screening tests, and the PWP completed the UPDRS. The participants were then asked to tap (with their finger, then with their toe, and finally step on the spot) at their “most comfortable, natural rate that was neither too fast nor too slow, but felt ‘just right’”, as established by McAuley et al. [
7] (p. 353). These data were collected in two trials lasting 30 seconds each. As the instructions focused on spontaneous repetitive movements, the participants chose whichever hand or foot they felt most comfortable to use for this specific task. Therefore, the participants were also asked which hand or foot was preferred for tapping to music to compare potential differences between the (hypothetical) tasks. These data are presented in
Table 2, and data relating to the laterality of PD are presented in the
Supplementary Table 1 in the online-only Data Supplement.
- Data preparation and analyses
The inter-onset interval (
IOI) refers to the time interval between the onsets of two successive strikes produced by a participant (i.e., finger or toe tap or a step). The mean
IOI indicates the rate of the SMT. A second dependent variable, the
CoV, measured the within-subject performance variability and was calculated as the
IOI standard deviation/
IOI mean×100) [
30,
31].
Equipment failure resulted in the loss of data from 25 out of a potential 368 trials (6.79%) across the 92 participants. Following distribution analyses, one outlier was removed from the FT
CoV data (FT
CoV = 93.42) to reach the criterion for Levene’s statistic (i.e., not significant). This adjustment did not change the nature or outcome of the analyses. Effect sizes are reported as partial eta squared (interpreted as small = 0.01, medium = 0.06, and large = 0.14) [
32,
33]. Tukey’s honestly significant difference (HSD) post hoc analyses were used to explore significant findings. Analyses were conducted using SPSS Software (ver. 23 and ver. 25, IBM Corp., Armonk, NY, USA)
RESULTS
- Descriptive
Table 3 presents the mean
IOI for the SMTs for each group in each movement modality, and
Table 4 presents the mean
CoV for the SMTs for each group in each movement modality.
- Group analyses
Analyses of variance by group was conducted for
IOI and
CoV, and the results are presented by movement modality. Post hoc Tukey’s HSD analyses were performed to illustrate the nature of the differences (
Figure 1 and
2).
Correlations
Overall, according to the two-tailed Pearson product-moment coefficients, the movement modalities were highly correlated with each other for both
IOI and
CoV in the whole sample (
Table 5), but the correlations for TT and SS (right heel) failed to reach the significance level. However, as shown in
Table 5, in contrast to the whole sample data, the PWP data showed a disruption in the relationship between effector movements (i.e., finger and toe tapping) and whole/body movement.
There were strong correlations between the right and left foot stepping conditions for the whole sample (
Table 5) and for PWP; the correlation in the
IOI mean was r(29) = 0.93,
p < 0.01, and that for
CoV was r(29) = 0.69,
p < 0.01. Thus, the two SMT rates for stepping on the spot for right heel and left heel were averaged to make a new dependent variable Stepping
IOI and Stepping
CoV for further analyses.
Finger tapping
Significant differences between groups were revealed for FT IOI F(2, 87) = 7.92, p < 0.01, ηρ2 = 0.15. The PWP had faster finger tapping than the YHC [p <0.01, mean difference ± 112.21 ms, standard error (SE) = 28.78 ms]. However, the PWP and AMC did not differ (p = 0.45). Although the AMC tended to be faster than the YHC, this difference between control groups was not significant (p = 0.05).
Significant differences between groups were also revealed for FT CoV F(2, 87) = 4.10, p = 0.02, ηρ2 = 0.09. The PWP were more variable than the AMC (p = 0.01, mean difference ± 10.47, SE = 3.75) but not more variable than the YHC (p = 0.37). The YHC were also more variable than the AMC (p = 0.04, mean difference ± 7.42, SE = 3.55).
Toe tapping
Significant differences between groups were revealed for TT IOI F(2, 87) = 4.89, p = 0.01, ηρ2 = 0.10. The PWP were faster than the AMC (p < 0.01, mean difference ± 89.67, SE = 29.57 ms) and YHC (p = 0.04, mean difference ± 54.87, SE = 26.55 ms). The AMC and YHC did not differ (p = 0.23). No significant differences in the CoV were revealed between groups for TT (p = 0.62).
Stepping on the spot, right heel
A significant difference between groups was revealed for SS right heel IOI F(2, 80) = 3.49, p = 0.04, ηρ2 = 0.08. The PWP were faster than the YHC (p = 0.03, mean difference ± 44.99 ms, SE = 17.41 ms) but were not different from the AMC (p = 0.21). The YHC and AMC did not differ (p = 0.79). No significant differences in the CoV were revealed between groups for stepping on the spot (right heel) (p = 0.13).
Stepping on the spot, left heel
A significant difference between groups was revealed for SS left heel IOI F(2, 81) = 4.10, p = 0.02, ηρ2 = 0.09. The PWP were faster than the YHC (p < 0.01, mean difference ± 50.77 ms, SE = 18.11 ms) but were not different from the AMC (p = 0.07). The YHC and AMC did not differ (p = 0.44). No significant differences in the CoV were revealed between groups for stepping on the spot (left heel) (p = 0.59).
Stepping
The mean of the data for both feet were used to generate two dependent variables: Stepping IOI and stepping CoV. A significant difference between groups was revealed for stepping IOI F(2, 77) = 3.26, p = 0.04, ηρ2 = 0.08. The PWP were faster than the YHC (p < 0.05, mean difference ± 43.50 ms, SE = 17.26 ms), but were not different from the AMC (p = 0.30). The YHC and AMC did not differ (p = 0.65). No significant differences in the CoV were revealed between groups for stepping (p = 0.91).
Additional exploratory analyses
As analyses revealed significant differences between the PWP and controls, additional analyses were performed to understand which (if any) specific aspects of PD might predict SMT performance. A series of linear regressions were conducted on all the SMT dependent variables using the UPDRS total scores and the subscale scores (I, II, III, and IV) as predictor variables. Although significant results are reported below, once alpha p was adjusted for multiple comparisons, these findings did not remain significant. Therefore, these findings are provided for clinical interest only.
The UPDRS II (activities of daily living) predicted variability in finger tapping (FT CoV) - F(1, 26) = 7.77, p = 0.01, R2 = 0.23. The model predicts that for every 1.76 increase in the score on the UPDRS II, the FT SMT CoV (i.e., variability) will increase by 18.03 ms. The score on the UPDRS II explained 23% of the variance in finger tapping. Higher scores (i.e., more difficulties in activities of daily living) were associated with more variability in finger tapping. Similarly, the UPDRS II score predicted toe tapping variability (TT CoV): UPDRS II F(1, 28) = 4.56, p = 0.04, R2 = 0.14 (therefore explaining 14% of the variance in toe tapping). This model suggests that for every 0.89 increase in the score on the UPDRS II, the TT SMT CoV (i.e., variability) will increase by 24.69 ms.
These findings led to the evaluation of whether the hand or foot used in the SMT task, in comparison to the hand or foot affected by PD, impacted the results. Although no significant effect was found in relation to the hand used in the PWP, the effect of PD on the foot used was significant. By comparing the foot used with the side affected by PD, a foot match issue was confirmed in relation to the SMT rates for toe tapping: F(1, 26) = 5.60, p = 0.02, accounting for 18.7% of the variance (R2 = 0.19). The foot used significantly predicted the SMT variability (CoV) for toe tapping: F(1, 28) = 12.23, p < 0.01, R2 = 0.30. In this group, 22 PWP used their right foot (mean CoV, 31.47, SD = 8.15, range 21.6–44.57), five used their left foot (mean CoV, 32.71, SD = 7.64, range 24.27–41.58) and three used one foot for each of the two trials (mean CoV, 54.52, SD = 14.32, range 42.84–70.49). This result suggests that for PWP whose feet were affected by PD, using either side did not overcome the problem of tapping consistency. In contrast, PWP were able to compensate with their hands, for which there was no apparent significant effect.
These analyses were also conducted using PD duration and severity (according to the Hoehn and Yahr stages, and the Schwab and England percentiles), but these factors as independent variables did not predict SMT rate or stability. Furthermore, as performed in a previous study [
18], the PWP participants were grouped according to PD subtypes [
28] (
Table 1). As Yahalom et al. [
18] found a difference in SMT between the unclassified (UC) and freeze predominant subtypes, it was important to compare the tremor dominant (TD,
n = 10), postural instability/gait difficulty (PIGD,
n = 6), and UC (
n = 14) subtypes in these data. However, no PD subtype differences were revealed, and none of the PD subtypes predicted SMT performance in this sample of PWP.
Due to the differences in the number of males and females in the groups, analyses by sex were also conducted. For the whole sample, a significant difference between males and females was revealed for the stepping on the spot only, mean IOI F(1, 75) = 5.112, p = 0.027, ηρ2 = 0.064. The males (mean = 539.09 ms, SD = 73.38 ms) stepped on the spot more slowly than the females (mean = 504.49 ms, SD = 58.54 ms). There was no interaction with Group (p > 0.17), and no significant differences according to sex for any of the SMT dependent variables were revealed within the PWP group only.
DISCUSSION
This study compares SMTs in different types of movement in people with and without PD. Age-matched and younger controls were included to provide information on rehabilitation interventions for researchers, clinicians and practitioners. There are two main findings. First, the SMT rates in the PWP were faster than those in both control groups during toe tapping but were faster than that in the younger control group only during finger tapping and stepping on the spot. The PWP were also more variable than both control groups for finger tapping, but no group differences in variability were observed for toe tapping and stepping on the spot, for which the least amount of variance was observed. Second, although the whole group analyses suggested that the three types of movements (finger and toe tapping and stepping on the spot) were correlated with the SMT rates, this result did not hold for the PWP. These findings are now discussed in relation to those reported in previous studies and the literature relating to timed motor behaviors in individuals with PD.
Finger tapping is commonly used in timing studies and therefore can be directly compared. In the present study, the difference between the PWP and the younger control group amounted to a difference of 24 bpm for finger tapping (with the PWP performing the task faster than the YHC group), and this finding had a large effect size. As the PWP did not differ significantly from the AMC in the SMT rate, the most parsimonious interpretation of the finding would be that the younger controls were slower than the PWP and AMC. A similar finding was reported previously in a large-scale finger tapping study, whereby McAuley et al. [
7] noted what they described as a “potential blip in SMTs” (p. 354) in the 18- to 38-year-old group of individuals included in their study. However, two studies [
18,
19] have previously reported slower SMTs for PWP when finger tapping, although different methods were used. Yahalom et al. [
18] collected SMT data for 16 seconds using the least affected limb for the PD participants to limit the effects of motor deficits on the timing tasks. Furthermore, for 75% of these 51 participants with PD, their least affected hand corresponded to their nondominant hand. Benoit et al. [
19] collected SMTs for both hands for 60 seconds each and seemingly reported the mean of these data (though this is not explicitly stated). There are therefore two important points to consider: 1) the use of the hand (and in our study, foot) and 2) the faster SMT rate reported in this study compared with previous studies.
To address the first of these points, as previously mentioned, there are several ways to measure spontaneous movements, and there are difficulties associated with all of them, at least in the context of PD research. For example, finger tapping is used in event-based or predictive timing studies because it is thought to enable the parsing of variance caused by motor ‘noise’ and to identify an individual’s motor intent, whereas the continuous movement of stepping has been associated with emergent timing [
34-
37]. However, finger tapping is not naturally associated with spontaneous movements. Therefore, toe tapping was included in this study as a comparable effector type movement associated with spontaneous responses to music. Similarly, gait is often used as a measure of spontaneous (bipedal) timing, but it is not directly comparable to tapping due to the forward motion associated with gait. Therefore, we included stepping on the spot as a whole-body spontaneous motion specifically because it has been shown to be associated with emergent rather than predictive timing due to the continuous nature of the movement.
In this study, the participants chose whichever hand or foot they felt most comfortable using for the task, but we also gathered information regarding which hand they would use if they were completing a goal-orientated task (in this case, we suggested tapping to music, in comparison to a SMT, which is simply tapping at one’s most comfortable speed) and which hand and foot were most affected by PD. We analyzed these data and found that for PWP, although there was a hand match issue (that is, the first choice was compromised by PD) in 2/3rds of the participants, it did not seem to affect performance regarding the rate of tapping; it only seemed to affect the variance. However, with toe tapping, although the same ratio was recorded in terms of the match issue, the use of the foot did significantly affect performance. Although no group differences in the variance were observed for toe tapping, it was noticeable that all groups performed with more variance in toe tapping than in finger tapping for this task. Furthermore, the least amount of variance was apparent for stepping on the spot. This finding suggests that in general, stepping was the easiest task, that the PWP found finger tapping the hardest task to sustain, and that toe tapping was the most difficult task. Why then, would PWP perform this task in particular faster than both control groups?
There are two possible explanations suggested by the literature: hastening and kinesia paradoxica. Hastening is a phenomenon whereby tapping is executed at a higher rate than required (in comparison to a target tempo), and it is reported to occur in older people and people with PD, for whom it may be related to freezing and/or festination [
18]. As research has suggested that differing clinical phenotypes in PD may be related to risk factors for motor symptoms (i.e., TD or PIGD), we classified the PWP in this study into these subtypes [
28] to ascertain whether such differences in presentation manifested in a SMT for the movements observed. No associations between the PD subtypes and SMT variables were established, but this result may be because the sample was not sufficiently large to detect differences. Future studies, preferably longitudinal studies, that include measures for both hands, both feet, stepping on the spot and gait, as well as sex matching groups, should be conducted to determine how SMT might change over time. Although sex differences have not been reported in SMT studies per se, gait studies have suggested that the differences reported herein for the whole sample for stepping may be related to wholebody kinematics, such as hip movement and arm swing [
38]. This comprehensive approach to future research will also help elucidate whether the posited slowing of the internal clock theory is linked to cognitive decline in PWP [
7,
17]. The second possible explanation is kinesia paradoxica, which is the idea that a motor response is partially dependent on a person’s emotional state, and this may or may not be associated with bradykinesia in PD [
39]. Although kinesia paradoxica is also usually associated with external triggers, heightened emotional arousal can affect performance in spontaneous motor tasks. Hypothetically, this theory can be tested using a measure of momentary affective states. However, this theory is highly speculative, and although we know of no study that has considered this theory, it is another possible avenue for future SMT research.
Finally, by comparing the correlations between SMT for the three different movement types, for the whole group and for PWP only, this study presents evidence of a disconnect between effector movements (finger and toe tapping) and the whole-body movement of stepping on the spot. These data also provided evidence that PWP (and YHC) were significantly more variable than age-matched controls (AMC) in finger tapping. This finding had a medium to large effect size, and it was most noticeable that no differences between groups were observed in the other movement modalities for this measure of variability. Medication is known to impact timing performance [
12], and future studies should consider testing PWP during both the ON and OFF medication regimes. However, from this study, the results suggest that different types of movement should be considered for different applications. For example, finger tapping may provide evidence of PD impairment for research, but stepping on the spot appears to be a relatively preserved form of spontaneous movement for PWP. This result may be because emergent timing is thought to be relatively unaffected due to the compensatory support from the cerebellum, whereas predictive timing (as associated with finger tapping) relies on the basal ganglia [
34-
37]. As reduced performance in bimanual tasks compared to unimanual tapping tasks has been observed [
40], this result strengthens the suggestions that it is the nature of stepping on the spot in particular that may be useful for therapeutic application. This activity can be performed safely (by holding the back of a chair, for example) in the patient’s own home to increase activity levels and fitness, and further research has also shown that this type of movement is particularly good for sensorimotor synchronization to music (rather than metronomes) [
29]. If practitioners and clinicians can assess the patients’ SMT for this movement, they can match the timing to preferred music (therefore bpm measures are included in
Tables 3 and
4) to optimize therapeutic goals and increase the enjoyment of the activity.
However, the potential claims of this study are limited because the order in which the SMT data was collected was not counterbalanced, the groups were not sex matched, and the PWP were not tested in both the ON and OFF states of medication. Overall, the main finding is that practitioners and clinicians should not assume that their age and/or PD slows patients’ spontaneous movement tempo for all types of motor actions. There are large individual differences in SMT [
21], so it is important to establish an individual’s baseline SMT to personalize treatment to achieve therapeutic goals.
Supplementary Materials
Supplementary Table 1.
Extended version of the Parkinson’s disease participant information
jmd-19043-suppl1.pdf
Notes
-
Conflict of Interest
This study was not funded, but the research was supported by the authors’ institutions, including the University of Hertfordshire (UK), Lucerne University of Applied Sciences and Arts (Switzerland), McMaster University (Canada), and Western University (Canada). This study was conducted at the University of Hertfordshire as part of the first author’s postdoctoral research fellowship (2016-2018), during which a research visit to Canada resulted in the collaboration. The study was compiled at Lucerne University of Applied Sciences and Arts in 2019 as part of the first authors’ new role as a Senior Research Associate. The authors declare no conflicts of interest.
-
Author Contributions
Conceptualization: Dawn Rose, Daniel J. Cameron, Jessica A. Grahn, and Lucy E. Annett. Data curation: Dawn Rose and Daniel J. Cameron. Formal analysis: Dawn Rose. Investigation: Dawn Rose, Daniel J. Cameron, and Lucy E. Annett. Methodology: Dawn Rose, Daniel J. Cameron, and Lucy E. Annett. Project administration: Dawn Rose and Daniel J. Cameron. Resources: All authors. Software: Dawn Rose. Supervision: Lucy E. Annett. Validation: Daniel J. Cameron and Lucy E. Annett. Visualization: Dawn Rose. Writing—original draft: Dawn Rose. Writing—review & editing: All authors.
Acknowledgments
The authors acknowledge Professor Yvonne Delevoye-Turrell and Dr. Laurent Ott from SCALab at University of Lille, France for extracting the data.
Figure 1.The group differences in spontaneous motor tempo (SMT) inter-onset intervals (IOI). The error bars display the standard deviation. An asterisk (*) identifies significant differences between groups (p < 0.05). FT: finger tapping, TT: toe tapping, SR: ss right, SL: ss left, PWP: people with Parkinson’s disease, AMC: age-matched controls, YHC: young healthy controls.
Figure 2.The group differences in spontaneous motor tempo (SMT) coefficient of variation (CoV). The error bars display the standard deviation. An asterisk (*) identifies significant differences between groups (p < 0.05). FT: finger tapping, TT: toe tapping, SR: ss right, SL: ss left, PWP: people with Parkinson’s disease, AMC: age-matched controls, YHC: young healthy controls.
Table 1.Parkinson’s disease participant information
Age*
|
Sex |
PD duration†
|
PD sub-type‡
|
UPDRS total |
H&Y§
|
LEDD (mg) |
66 |
F |
42 |
UC |
3 |
0 |
290 |
44 |
F |
48 |
TD |
36 |
3 |
710 |
48 |
M |
48 |
TD |
31 |
3 |
240 |
76 |
F |
43 |
UC |
34 |
2 |
- |
75 |
F |
252 |
PIGD |
29 |
5 |
925 |
65 |
F |
228 |
TD |
39 |
3 |
550 |
70 |
F |
48 |
TD |
25 |
1 |
280 |
63 |
M |
108 |
TD |
46 |
3 |
- |
71 |
M |
60 |
TD |
27 |
3 |
1,056 |
69 |
M |
192 |
PIGD |
25 |
3 |
1175 |
65 |
F |
36 |
TD |
25 |
2 |
- |
56 |
F |
144 |
PIGD |
33 |
2 |
2,356 |
68 |
F |
108 |
UC |
12 |
1 |
540 |
77 |
F |
36 |
UC |
34 |
2 |
- |
59 |
M |
180 |
TD |
44 |
3 |
- |
49 |
M |
11 |
TD |
29 |
2 |
80 |
65 |
F |
24 |
UC |
41 |
2 |
- |
73 |
F |
6 |
UC |
20 |
2 |
375 |
59 |
F |
69 |
PIGD |
45 |
2 |
328 |
54 |
F |
72 |
TD |
34 |
2 |
720 |
58 |
F |
20 |
PIGD |
36 |
2 |
500 |
60 |
M |
90 |
UC |
25 |
1 |
- |
34 |
M |
43 |
PIGD |
62 |
2 |
1,880 |
67 |
F |
72 |
UC |
32 |
2 |
663 |
48 |
F |
120 |
UC |
19 |
2 |
1274 |
70 |
M |
24 |
UC |
43 |
3 |
- |
68 |
F |
20 |
UC |
12 |
1 |
340 |
52 |
M |
5 |
UC |
26 |
2 |
100 |
63 |
F |
30 |
UC |
23 |
1 |
100 |
Table 2.Hand and foot that was used spontaneously and was preferred for tapping to music by group
Group |
Used during SMT task
|
General preference
|
Hand |
n
|
Foot |
n
|
Hand†
|
n
|
Foot†
|
n
|
PWP |
Right |
25 |
Right |
22 |
Right |
17 |
Right |
19 |
|
Left |
4 |
Left |
5 |
Left |
2 |
Left |
3 |
|
Both*
|
1 |
Both*
|
3 |
Alternating |
1 |
Alternating |
2 |
|
|
|
|
|
Either |
2 |
Either |
4 |
|
|
|
|
|
Prefers foot |
8 |
Prefers hand |
2 |
AMC |
Right |
24 |
Right |
24 |
Right |
10 |
Right |
17 |
|
Left |
2 |
Left |
2 |
Left |
3 |
Left |
3 |
|
Both*
|
0 |
Both*
|
0 |
Alternating |
1 |
Alternating |
1 |
|
|
|
|
|
Either |
3 |
Either |
4 |
|
|
|
|
|
Prefers foot |
9 |
Prefers hand |
0 |
YHC |
Right |
34 |
Right |
34 |
Right |
24 |
Right |
22 |
|
Left |
2 |
Left |
2 |
Left |
2 |
Left |
3 |
|
Both*
|
0 |
Both*
|
0 |
Alternating |
6 |
Alternating |
10 |
|
|
|
|
|
Either |
2 |
Either |
1 |
|
|
|
|
|
Prefers foot |
2 |
Prefers hand |
0 |
Table 3.Mean inter-onset interval (IOI) for spontaneous movement tempo by group and movement modality
IOI
|
Group |
n
|
Mean (ms) |
SD (ms) |
Minimum (ms) |
Maximum (ms) |
Bpm conversion |
Finger tapping |
Whole sample |
90 |
531.77 |
125.12 |
223.09 |
916.31 |
112.83 |
PWP |
30 |
476.57 |
127.40 |
226.02 |
852.34 |
125.90 |
AMC |
24 |
515.39 |
78.74 |
223.09 |
636.65 |
116.42 |
YHC |
36 |
588.69 |
127.11 |
412.25 |
916.31 |
101.92 |
Toe tapping |
Whole sample |
88 |
509.03 |
111.30 |
281.14 |
878.63 |
117.87 |
PWP |
30 |
463.78 |
101.40 |
281.14 |
739.37 |
129.37 |
AMC |
23 |
553.40 |
86.11 |
436.45 |
878.63 |
108.42 |
YHC |
35 |
518.65 |
122.04 |
288.34 |
870.14 |
115.68 |
Stepping on the spot, right heel |
Whole sample |
83 |
519.34 |
69.92 |
326.79 |
749.48 |
115.53 |
PWP |
29 |
493.31 |
62.29 |
361.77 |
621.43 |
121.63 |
AMC |
22 |
526.07 |
44.87 |
445.71 |
628.56 |
114.05 |
YHC |
32 |
538.30 |
83.82 |
326.79 |
749.48 |
111.46 |
Stepping on the spot, left heel |
Whole sample |
82 |
523.69 |
72.16 |
349.25 |
807.90 |
114.57 |
PWP |
29 |
495.09 |
64.09 |
349.25 |
627.26 |
121.19 |
AMC |
23 |
530.83 |
50.45 |
423.24 |
629.81 |
113.03 |
YHC |
30 |
545.86 |
85.32 |
421.15 |
807.90 |
109.92 |
Table 4.Mean coefficient of variation (CoV) for spontaneous movement tempo by group and movement modality
CoV
|
Group |
n
|
Mean (ms) |
SD (ms) |
Minimum (ms) |
Maximum (ms) |
Finger tapping |
Whole sample |
89 |
32.77 |
15.30 |
13.16 |
93.42 |
PWP |
29 |
38.16 |
19.81 |
19.97 |
93.42 |
AMC |
24 |
25.71 |
9.36 |
13.16 |
48.75 |
YHC |
36 |
33.14 |
12.61 |
19.71 |
69.77 |
Toe tapping |
Whole sample |
88 |
33.11 |
10.96 |
14.65 |
70.49 |
PWP |
30 |
33.98 |
10.93 |
21.64 |
70.49 |
AMC |
23 |
31.19 |
9.07 |
15.44 |
48.64 |
YHC |
35 |
33.64 |
12.19 |
14.65 |
63.84 |
Stepping on the spot, right heel |
Whole sample |
83 |
26.45 |
12.04 |
8.87 |
64.43 |
PWP |
29 |
24.97 |
10.67 |
8.87 |
50.09 |
AMC |
22 |
23.62 |
9.89 |
9.20 |
42.48 |
YHC |
32 |
29.75 |
13.98 |
9.61 |
64.43 |
Stepping on the spot, left heel |
Whole sample |
82 |
27.34 |
17.66 |
9.02 |
112.16 |
PWP |
29 |
26.93 |
16.85 |
9.02 |
74.50 |
AMC |
23 |
30.42 |
23.07 |
9.54 |
112.16 |
YHC |
30 |
25.37 |
13.46 |
9.39 |
52.66 |
Table 5.Pearson correlation results for the whole sample and for the Parkinson’s disease group for the inter-onset intervals (IOI) and coefficients of variation (CoV) between movement modalities
Movement modalities |
SMT IOI
|
SMT CoV
|
Whole sample |
|
|
Finger tapping—toe tapping |
r(86) = 0.71, p < 0.01 |
r(84) = 0.61, p < 0.01 |
Finger tapping—stepping on the spot (right heel) |
r(82) = 0.31, p < 0.01 |
r(80) = 0.24, p = 0.03 |
Finger tapping—stepping on the spot (left heel) |
r(81) = 0.463, p < 0.01 |
r(79) = 0.30, p < 0.01 |
Toe tapping—stepping on the spot (right heel) |
r(80) = 0.22, p = 0.05 |
r(80) = 0.22, p = 0.05 |
Toe tapping—stepping on the spot (left heel) |
r(78) = 0.38, p < 0.01 |
r(78) = 0.33, p < 0.01 |
Stepping on the spot (right heel–left heel) |
r(77) = 0.96, p < 0.01 |
r(77) = 0.33, p < 0.01 |
Finger tapping—stepping on the spot*
|
r(76) = 0.36, p < 0.01 |
r(74) = 0.34, p < 0.01 |
Toe tapping—stepping on the spot*
|
r(74) = 0.24, p = 0.04 |
r(74) = 0.29, p = 0.01 |
Parkinson’s disease group |
|
|
Finger tapping—toe tapping |
r(30) = 0.77, p < 0.01 |
r(28) = 0.64, p < 0.01 |
Finger tapping—stepping on the spot (right heel) |
ns (p = 0.11) |
ns (p = 0.09) |
Finger tapping—stepping on the spot (left heel) |
r(29) = 0.48, p < 0.01 |
r(27) = 0.68, p < 0.01 |
Toe tapping—stepping on the spot (right heel) |
ns (p = 0.55) |
ns (p = 0.12) |
Toe tapping—stepping on the spot (left heel) |
ns (p = 0.13) |
ns (p = 0.083) |
Stepping on the spot (right heel–left heel) |
r(29) = 0.93, p < 0.01 |
r(29) = 0.69, p < 0.01 |
Finger tapping—stepping on the spot*
|
r(29) = 0.40, p = 0.03 |
r(27) = 0.58, p < 0.01 |
Toe tapping—stepping on the spot*
|
ns (p = 0.28) |
ns (p = 0.07) |
REFERENCES
- 1. Moelants D. Preferred tempo reconsidered. In: Stevens C, Burnham D, McPherson G, Schubert E, Renwick J, editors. Proceedings of the 7th international conference on music perception and cognition. Sydnney: Adelaide: Causal Productions; 2002:580–583.
- 2. Moelants D. Dance music, movement and tempo preferences. In: Kopiez R, editor. Proceedings of the 5th Triennual ESCOM conference. Hanover Institute for Research in Music Education; 2003:649–652.
- 3. van Noorden L, Moelants D. Resonance in the perception of musical pulse. J New Music Res 1999;28:43–66.Article
- 4. Fraisse P. Rhythm and tempo. In: Deutsch D, editor. The psychology of music. 1st ed. New York: Academic Press; 1982:149–180.
- 5. McAuley JD. Tempo and rhythm. In: Jones MR, Fay RR, Popper AN, editors. Music perception. New York: Springer-Verlag; 2010:165–199.
- 6. Rimoldi HJ. Personal tempo. J Abnorm Psychol 1951;46:283–303.ArticlePubMedPDF
- 7. McAuley JD, Jones MR, Holub S, Johnston HM, Miller NS. The time of our lives: life span development of timing and event tracking. J Exp Psychol Gen 2006;135:348–367.ArticlePubMedPDF
- 8. Drake C, Jones MR, Baruch C. The development of rhythmic attending in auditory sequences: attunement, referent period, focal attending. Cognition 2000;77:251–288.ArticlePubMed
- 9. Vanneste S, Pouthas V, Wearden JH. Temporal control of rhythmic performance: a comparison between young and old adults. Exp Aging Res 2001;27:83–102.ArticlePubMed
- 10. Gallese V. The inner sense of action. Agency and motor representations. J Conscious Stud 2000;7:23–40.
- 11. Moore JW, Ruge D, Wenke D, Rothwell J, Haggard P. Disrupting the experience of control in the human brain: pre-supplementary motor area contributes to the sense of agency. Proc Biol Sci 2010;277:2503–2509.ArticlePubMedPMC
- 12. Jones CR, Jahanshahi M. Motor and perceptual timing in Parkinson’s disease. In: Merchant H, de Lafuente V, editors. Neurobiology of Interval Timing. New York: Springer; 2014:265–290.
- 13. Allen NE, Schwarzel AK, Canning CG. Recurrent falls in Parkinson’s disease: a systematic review. Parkinsons Dis 2013;2013:906274.ArticlePubMedPMCPDF
- 14. Findley LJ. The economic impact of Parkinson’s disease. Parkinsonism Relat Disord 2007;13 Suppl:S8–S12.ArticlePubMed
- 15. Giladi N, Treves TA, Simon ES, Shabtai H, Orlov Y, Kandinov B, et al. Freezing of gait in patients with advanced Parkinson’s disease. J Neural Transm (Vienna) 2001;108:53–61.ArticlePubMedPDF
- 16. Jankovic J. Parkinson’s disease: clinical features and diagnosis. J Neurol Neurosurg Psychiatry 2008;79:368–376.ArticlePubMed
- 17. Allman MJ, Meck WH. Pathophysiological distortions in time perception and timed performance. Brain 2012;135(Pt 3):656–677.ArticlePubMedPDF
- 18. Yahalom G, Simon ES, Thorne R, Peretz C, Giladi N. Hand rhythmic tapping and timing in Parkinson’s disease. Parkinsonism Relat Disord 2004;10:143–148.ArticlePubMed
- 19. Benoit CE, Dalla Bella S, Farrugia N, Obrig H, Mainka S, Kotz SA. Musically cued gait-training improves both perceptual and motor timing in Parkinson’s disease. Front Hum Neurosci 2014;8:494.ArticlePubMedPMC
- 20. de Dreu MJ, van der Wilk AS, Poppe E, Kwakkel G, van Wegen EE. Rehabilitation, exercise therapy and music in patients with Parkinson’s disease: a meta-analysis of the effects of music-based movement therapy on walking ability, balance and quality of life. Parkinsonism Relat Disord 2012;18 Suppl 1:S114–S119.ArticlePubMed
- 21. Thaut MH, McIntosh GC, Hoemberg V. Neurobiological foundations of neurologic music therapy: rhythmic entrainment and the motor system. Front Psychol 2015;5:1185.ArticlePubMedPMC
- 22. Dalla Bella S, Dotov D, Bardy B, de Cock VC. Individualization of musicbased rhythmic auditory cueing in Parkinson’s disease. Ann N Y Acad Sci 2018;1423:308–317.Article
- 23. Parkinson’s UK. Improving life through research [Internet]. London: Parkinson’s UK; c2019 [accessed on 2018 Nov 15]. Available at: https://www.parkinsons.org.uk/research/improving-life-through-research.
- 24. Folstein MF, Folstein SE, McHugh PR. “Mini-mental state”. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975;12:189–198.ArticlePubMed
- 25. Fahn S, Elton RL; UPDRS Program Members. Unified Parkinson’s disease rating scale. In: Fahn S, Marsden CD, Goldstein M, Calne DB, editors. Recent developments in Parkinson’s disease. Vol 2. Florham Park: Macmillan Healthcare Information; 1987:513–163.293–304.
- 26. Schwab RS, England AC. Projection technique for evaluating surgery in Parkinson’s disease. In: Billingham FH, Donaldson MC, editors. Third symposium on Parkinson’s disease. Edinburgh: Livingstone; 1968:152–157.
- 27. Hoehn MM, Yahr MD. Parkinsonism: onset, progression, and mortality. Neurology 1967;17:427–442.ArticlePubMed
- 28. Stebbins GT, Goetz CG, Burn DJ, Jankovic J, Khoo TK, Tilley BC. How to identify tremor dominant and postural instability/gait difficulty groups with the movement disorder society unified Parkinson’s disease rating scale: comparison with the unified Parkinson’s disease rating scale. Mov Disord 2013;28:668–670.ArticlePubMed
- 29. Rose D, Delevoye-Turrell Y, Ott L, Annett LE, Lovatt PJ. Music and metronomes differentially impact motor timing in people with and without Parkinson’s disease: effects of slow, medium and fast tempi on entrainment and synchronization performances in finger tapping, toe tapping, and stepping on the spot tasks. Parkinsons Dis 2019;2019:6530838.ArticlePubMedPMCPDF
- 30. Repp BH. Sensorimotor synchronization: a review of the tapping literature. Psychon Bull Rev 2005;12:969–992.ArticlePubMedPDF
- 31. Wilquin H, Delevoye-Turrell Y, Dione M, Giersch A. Motor synchronization in patients with schizophrenia: preserved time representation with abnormalities in predictive timing. Front Hum Neurosci 2018;12:193.ArticlePubMedPMC
- 32. Bakeman R. Recommended effect size statistics for repeated measures designs. Behav Res Methods 2005;37:379–384.ArticlePubMedPDF
- 33. Cohen J. Statistical power analysis for the behavioral sciences. 2nd ed. Hillsdale: Erlbaum; 1988:285–287.
- 34. Wing AM, Kristofferson AB. The timing of interresponse intervals. Percept Psychophys 1973;13:455–460.ArticlePDF
- 35. Grahn JA, Brett M. Impairment of beat-based rhythm discrimination in Parkinson’s disease. Cortex 2009;45:54–61.ArticlePubMed
- 36. Spencer RMC, Ivry RB. Comparison of patients with Parkinson’s disease or cerebellar lesions in the production of periodic movements involving event-based or emergent timing. Brain Cogn 2005;58:84–93.ArticlePubMed
- 37. Dione M, Delevoye-Turrell Y. Testing the co-existence of two timing strategies for motor control in a unique task: the synchronisation spatial-tapping task. Hum Mov Sci 2015;43:45–60.ArticlePubMed
- 38. Bruening DA, Frimenko RE, Goodyear CD, Bowden DR, Fullenkamp AM. Sex differences in while body gait kinematics at preferred speeds. Gait Posture 2015;41:540–545.ArticlePubMed
- 39. Ballanger B, Thobois S, Baraduc P, Turner RS, Broussolle E, Desmurget M. “Paradoxical kinesis” is not a hallmark of Parkinson’s disease but a gneral proeprty of the motor system. Mov Disord 2006;21:490–495.Article
- 40. Brown RG, Jahanshahi M, Marsden CD. The execution of bimanual movements in patients with Parkinson’s, Huntington’s and cerebellar disease. J Neurol Neurosurg Psychiatry 1993;56:295–297.ArticlePubMedPMC