**2. Influence of various rhythm shifts on reactive periodic movement**

The EMG‐RT is defined as the interval between the stimulation signal and the onset of vol‐ untary electromyographic activity of a response, and reflects the time of the central process. The presentation of periodic stimuli creates predictions and expectations. During the reac‐ tion‐time task, the EMG‐RTs are shortened in the first three stimuli when applying the peri‐ odic stimuli. On the other hand, the different intervals inserted in periodic stimuli sequences causes a delay in EMG‐RT [9]. Regarding the facilitation of reactive periodic movement dur‐ ing physical therapy, if the physical therapists provide periodic rhythm by handclap, it will be impossible to replicate the exact time interval without deviation in the absence of a met‐ ronome. The studies described in this section show the EMG‐RT data responses to various rhythm shifts. Additionally, we clarify the relationship between the degree of stimulus inter‐ val deviation and the delay of EMG‐RT in the reactive periodic movement.

#### **2.1. Apparatus for recording the EMG‐RT**

The same apparatus was used for EMG‐RT measurements in experiments 1–4, which are discussed in this section. The telemetry EMG measuring system (MQ8; KISSEICOMTEC, Matsumoto City, Japan) was loaded onto a PC (VersaPro VY20F/AG‐W; NEC). The auditory stimulus system was set up using SoundTrigger2Plus (KISSEICOMTEC). Auditory stimuli were delivered via headphones. The presented auditory stimulus and EMG signals were recorded with a data acquisition system (VitalRecorder2; KISSEICOMTEC), and recorded sig‐ nals were analyzed by an EMG signal analysis program (BIMUTAS‐Video; KISSEICOMTEC). The surface electrodes (LecTrode NP; ADVANCE, Tokyo, Japan) were placed over the right tibialis anterior muscle.

#### **2.2. EMG‐RT measurements**

impaired in patients with motor diseases such as Parkinson's or those who have experienced a stroke. It is possible that an auditory pacing event could be useful as physical therapy in treat‐ ing movement disorders. In a physical therapy study, during a target‐hitting task using flex‐ ion and extension of the elbow, variability of electromyogram patterns of the biceps brachii decreased with pacing using a regular auditory rhythm, compared with that during the no‐ pacing and irregular auditory rhythm conditions [1]. In clinical studies, it has been reported that an intervention using rhythmic auditory stimulation improved gait velocity, cadence, and stride length in patients with Parkinson's disease [2] and gait velocity, stride length, and elec‐ tromyographic activity of the medial gastrocnemius in patients with hemiparetic stroke [3].

When performing a rhythmic movement using periodic auditory stimuli as a trigger, the subjects can select a variety of movement patterns (such as reaction, synchronization, and syncopation). In this chapter, the electromyographic reaction time (EMG‐RT) data responses to various rhythm shifts [4–7] are discussed in Section 2, and the variabilities of intertap inter‐ val (ITI) in the continuation paradigm of sensorimotor synchronization [8] are discussed in Section 3. Furthermore, each section includes an explanation of the clinical consideration in

**2. Influence of various rhythm shifts on reactive periodic movement**

val deviation and the delay of EMG‐RT in the reactive periodic movement.

**2.1. Apparatus for recording the EMG‐RT**

tibialis anterior muscle.

The EMG‐RT is defined as the interval between the stimulation signal and the onset of vol‐ untary electromyographic activity of a response, and reflects the time of the central process. The presentation of periodic stimuli creates predictions and expectations. During the reac‐ tion‐time task, the EMG‐RTs are shortened in the first three stimuli when applying the peri‐ odic stimuli. On the other hand, the different intervals inserted in periodic stimuli sequences causes a delay in EMG‐RT [9]. Regarding the facilitation of reactive periodic movement dur‐ ing physical therapy, if the physical therapists provide periodic rhythm by handclap, it will be impossible to replicate the exact time interval without deviation in the absence of a met‐ ronome. The studies described in this section show the EMG‐RT data responses to various rhythm shifts. Additionally, we clarify the relationship between the degree of stimulus inter‐

The same apparatus was used for EMG‐RT measurements in experiments 1–4, which are discussed in this section. The telemetry EMG measuring system (MQ8; KISSEICOMTEC, Matsumoto City, Japan) was loaded onto a PC (VersaPro VY20F/AG‐W; NEC). The auditory stimulus system was set up using SoundTrigger2Plus (KISSEICOMTEC). Auditory stimuli were delivered via headphones. The presented auditory stimulus and EMG signals were recorded with a data acquisition system (VitalRecorder2; KISSEICOMTEC), and recorded sig‐ nals were analyzed by an EMG signal analysis program (BIMUTAS‐Video; KISSEICOMTEC). The surface electrodes (LecTrode NP; ADVANCE, Tokyo, Japan) were placed over the right

physical therapy.

58 Clinical Physical Therapy

#### *2.2.1. Subliminal rhythm shift with shortened interval*

The purpose of experiment 1 was to investigate the influence of a subliminal rhythm shift with a shortened interval on the control of reactive movement. Fourteen healthy subjects (10 men, 4 women; mean age, 25.4 years) participated in this experiment. The subjects were right‐foot dominant and kicked the balled with their right foot. Subjects had no motor function abnor‐ malities of the right ankle and no hearing abnormalities. The experiment was conducted in a quiet room. The subjects were seated with 90° of knee flexion. All subjects performed the reac‐ tion‐time tasks, raising their right ankles in response to the auditory stimuli. Their eyes were closed to exclude the influence of vision during the tasks. The following three test conditions were applied: (1) periodic auditory stimuli with an interstimulus‐onset interval (ISI) of 1500 ms and a shift in the last stimulus interval only to (2) 1425 ms (the interval shortened 5% of 1500 ms), and (3) 1200 ms (the interval shortened 20% of 1500 ms), respectively, in successive stimulus sequences with an ISI of 1500 ms. Each condition was composed of 6–10 stimuli. There were 15 trials performed during each condition for a total of 45 trials. EMG‐RT values for the last stimulus were compared among the three conditions. One‐way repeated measures analysis of variance (ANOVA) revealed a significant difference between the EMG‐RT values under different conditions (see **Figure 1**). Tukey's posthoc test showed that the EMG‐RT was

**Figure 1.** EMG‐RT values of conditions 1–3.

significantly delayed under condition 3 (184.7 ± 24.6 ms, *p* < 0.01) compared with that under conditions 1 (138.0 ± 29.7 ms) and 2 (144.3 ± 29.4 ms). A comparison of conditions 1 and 2 revealed no significant differences.

#### *2.2.2. Subliminal rhythm shift with lengthened interval*

Experiment 2 aimed to investigate the influence of a subliminal rhythm shift with a length‐ ened interval on the control of reactive movement. Thirteen healthy individuals (10 men, 3 women; mean age, 27.4 years) were included in this study. All subjects were right‐footed according to the Chapman's foot‐preference inventory (mean score, 14.8) [10]. The subjects had no motor function abnormalities of the right ankle and no hearing abnormalities. All subjects performed the same reaction‐time tasks as those in experiment 1. The following three test conditions were applied: (1) periodic auditory stimuli with an ISI of 1500 ms, and a shift in the last stimulus interval only to (2) 1575 ms (the interval lengthened 5% of 1500 ms), and (3) 1800 ms (the interval lengthened 20% of the 1500 ms), respectively, in successive stimuli sequences at an ISI of 1500 ms. Each condition was composed of 6–10 stimuli. There were 15 trials performed during each condition for a total of 45 trials. EMG‐RT values for the last stimulus were compared among the three conditions. One‐way repeated measures ANOVA revealed a significant difference between the EMG‐RT values under different conditions (see **Figure 2**). Tukey's posthoc test showed that the EMG‐RT was significantly delayed under

**Figure 2.** EMG‐RT values of conditions 1–3.

condition 3 (189.8 ± 17.3 ms, *p* < 0.01) compared with that under conditions 1 (130.9 ± 11.8 ms) and 2 (131.1 ± 11.2 ms). A comparison of conditions 1 and 2 revealed no significant differences.

#### *2.2.3. Subliminal rhythm shift with random interval*

significantly delayed under condition 3 (184.7 ± 24.6 ms, *p* < 0.01) compared with that under conditions 1 (138.0 ± 29.7 ms) and 2 (144.3 ± 29.4 ms). A comparison of conditions 1 and 2

Experiment 2 aimed to investigate the influence of a subliminal rhythm shift with a length‐ ened interval on the control of reactive movement. Thirteen healthy individuals (10 men, 3 women; mean age, 27.4 years) were included in this study. All subjects were right‐footed according to the Chapman's foot‐preference inventory (mean score, 14.8) [10]. The subjects had no motor function abnormalities of the right ankle and no hearing abnormalities. All subjects performed the same reaction‐time tasks as those in experiment 1. The following three test conditions were applied: (1) periodic auditory stimuli with an ISI of 1500 ms, and a shift in the last stimulus interval only to (2) 1575 ms (the interval lengthened 5% of 1500 ms), and (3) 1800 ms (the interval lengthened 20% of the 1500 ms), respectively, in successive stimuli sequences at an ISI of 1500 ms. Each condition was composed of 6–10 stimuli. There were 15 trials performed during each condition for a total of 45 trials. EMG‐RT values for the last stimulus were compared among the three conditions. One‐way repeated measures ANOVA revealed a significant difference between the EMG‐RT values under different conditions (see **Figure 2**). Tukey's posthoc test showed that the EMG‐RT was significantly delayed under

revealed no significant differences.

60 Clinical Physical Therapy

**Figure 2.** EMG‐RT values of conditions 1–3.

*2.2.2. Subliminal rhythm shift with lengthened interval*

Experiment 3 included 14 healthy subjects (11 men, 3 women; mean age, 28.5 years) who performed the same reaction‐time tasks as those in experiments 1 and 2. All subjects were right‐footed according to the Chapman's foot‐preference inventory (mean score, 14.6) [10]. EMG‐RT was measured under the following three test conditions: (1) periodic auditory stimuli with an ISI of 1500 ms, (2) a random ISI shift in the range of 1463–1537 ms (range of ±5% of the 1500 ms), and (3) a random ISI shift in the range of 1350–1650 ms (range of ±20% of 1500 ms). In condition 3, the time differences between consecutive ISIs were set longer than 75 ms and 10 auditory stimuli were provided per trial. There were 10 trials performed during each con‐ dition for a total of 30 trials. EMG‐RT values corresponding to the 1st–10th stimuli were com‐ pared within each condition using one‐way repeated measures ANOVA. When a significant difference was recognized, paired comparisons were performed using Tukey's posthoc test. The EMG‐RT values for conditions 1–3 are shown in **Table 1**. In conditions 1 and 2, EMG‐ RT value responses to the 2nd–10th stimuli were significantly shortened compared with the response to the 1st stimulus (*p* < 0.01), and the responses to the 3rd–10th stimuli were signifi‐ cantly shortened compared with the response to the 2nd stimulus (*p* < 0.01). In condition 3, EMG‐RT value responses to the 2nd–10th stimuli were significantly shortened compared with the response to the 1st stimulus (*p* < 0.01), and the responses to the 3rd stimulus was sig‐ nificantly shortened compared with the response to the 2nd stimulus (*p* < 0.01). On the other hand, EMG‐RT value responses to the 7th–10th stimuli were significantly delayed compared with the response to the 3rd stimulus (*p* < 0.01).

#### *2.2.4. Differences in rate of rhythm shift*

Experiment 4 aimed to investigate the influence of differences in the rate of rhythm shift on the control of reactive movement. Ten healthy individuals (8 men, 2 women; mean age, 25.5 years) performed the same reaction‐time tasks as those in the previous EMG‐RT measurements. All subjects were right‐footed according to the Chapman's foot‐preference inventory (mean score,


**Table 1.** EMG‐RT values for conditions 1–3 (unit: ms).

13.8) [10]. The following 21 test conditions were applied: periodic auditory stimuli with an ISI of 1500 ms, and only the last stimulus interval shifted to 1485, 1470, 1455, 1440, 1425, 1410, 1395, 1380, 1365, 1350, 1335, 1320, 1305, 1290, 1275, 1260, 1245, 1230, 1215, and 1200 ms (inter‐ vals shortened from 1 to 20% of 1500 ms), in successive stimuli sequences at an ISI of 1500 ms. Each condition was composed of 6–10 stimuli. There were 10 trials performed during each condition for a total of 210 trials; each subject performed the 210 trials over a period of 5 days (42 trials per day). EMG‐RT values for the last stimulus were compared among the 21 condi‐ tions. EMG‐RT value responses to stimuli with an ISI of 1500 ms (no rhythm shift) were 142.4 ± 8.9 ms. The EMG‐RT values for 20 test conditions with rhythm shift are shown in **Table 2**. The EMG‐RT values of shifts in ISI from 8 to 20% of the 1500 ms were significantly delayed compared with those during the periodic auditory stimuli with an ISI of 1500 ms with shifts in ISI from 1 to 7% of 1500 ms (*p* < 0.01). The EMG‐RT values of shifts in ISI from 15 to 20% of 1500 ms were significantly delayed compared with shifts in ISI from 8 to 11% of 1500 ms (*p* < 0.01).

#### **2.3. Practical considerations for the use of reactive periodic movement in physical therapy**

Small changes below 5% of the base interval are considered to be below the threshold of con‐ scious recognition [11]. For a base interval of 1500 ms, our three experiments revealed that the reactive movements can be performed without delay under conditions with an ISI shift below the threshold of conscious recognition. The periodic rhythmic stimulation is predicted to comprise some time duration. The prediction system for stimuli plays the same role as in the case of con‐ stant rhythm in such an ISI shift, and the readiness for movement will be maintained in the cen‐ tral nervous system. On the other hand, changes of 20% of the base interval are considered to be above the threshold of conscious recognition [11]. Our data showed that EMG‐RTs were delayed during the 20% ISI shift. The prediction system for the stimulus loses its ability, and the central nervous system will need to provide new motor commands. Furthermore, experiment 4 revealed that the periodic rhythmic stimulation with the 1500‐ms interval is predicted to comprise an approximately 100 ms duration. The reactive movements can be performed without delay with an ISI shift of 7% of 1500 ms. Finger tapping produces a series of intervals with substantial vari‐ ability, even when they are intended to be regular; the typical SD is 3–6% of an ISI within a range of 200–2000 ms [12]. When the physical therapist facilitates rhythmical reactive periodic move‐ ment using a handclap, it will be desirable to keep the rhythm shift in a range within 7% of the ISI.


**Table 2.** EMG‐RT values for 20 test conditions with rhythm shift (unit: ms).
