The effects of dual tasking on gait synchronization during over-ground side-by-side walking

Human Movement Science 59 (2018) 20–29

Contents lists available at ScienceDirect

Human Movement Science journal homepage: www.elsevier.com/locate/humov

Full Length Article

The effects of dual tasking on gait synchronization during overground side-by-side walking

T

⁎

Ari Z. Zivotofskya,1, , Hagar Bernad-Elazarib,1, Pnina Grossmana,f, Jeffrey M. Hausdorffb,c,d,e a

Gonda Brain Research Center, Bar Ilan University, Ramat Gan, Israel Center for the Study of Movement, Cognition, and Mobility, Neurological Institute, Tel Aviv Sourasky Medical Center, Tel Aviv, Israel c Department of Physical Therapy, Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel d Sagol School of Neuroscience, Tel Aviv University, Tel Aviv, Israel e Rush Alzheimer’s Disease Center and Department of Orthopaedic Surgery, Rush University Medical Center, Chicago, IL, United States f City College, New York, United States b

A R T IC LE I N F O

ABS TRA CT

Keywords: Stride length Stride time Synchronization Human Gait Dual tasking

Recent studies have shown that gait synchronization during natural walking is not merely anecdotal, but it is a repeatable phenomenon that is quantifiable and is apparently related to available sensory feedback modalities. However, the mechanisms underlying this phase-locking of gait have only recently begun to be investigated. For example, it is not known what role, if any, attention plays. We employed a dual tasking paradigm in order to investigate the role attention plays in gait synchronization. Sixteen pairs of subjects walked under six conditions that manipulated the available sensory feedback and the degree of difficulty of the dual task, i.e., the attention. Movement was quantified using a trunk-mounted tri-axial accelerometer. A gait synchronization index (GSI) was calculated in order to quantify the degree of synchronization of the gait pattern. A simple dual task resulted in an increased level of synchronization, whereas a more complex dual task lead to a reduction in synchronization. Handholding increased synchronization, compared to the same attention condition without handholding. These results indicate that in order for two walkers to synchronize, some level of attention is apparently required, such that a relatively complex dual task utilizes enough attentional resources to reduce the occurrence of synchronization.

1. Introduction Two people who walk together from one point to another must match their gait speeds if they wish to stay together. To achieve this goal, each person can select from an array of cadence and stride length combinations. Surprisingly, casual observation suggests and recent studies confirm that a significant fraction of people who ambulate together do this in synchrony, with near identical cadence and stride lengths. Step length and step time can be varied in numerous ways to ensure that walking partners move forward at the same rate and arrive at the desired destination together. However, instead of achieving this goal through a random, timevarying, combination of these parameters, remarkably, couples often seem to March to a single drummer. The first paper to examine this phenomenon (Zivotofsky & Hausdorff, 2007) analyzed it in a qualitative manner. More recently, this finding was quantitatively

⁎

Corresponding author at: Gonda Brain Research Center, Bar Ilan University, Ramat Gan 52900, Israel. E-mail address: [email protected] (A.Z. Zivotofsky). 1 These authors contributed equally. https://doi.org/10.1016/j.humov.2018.03.009 Received 9 November 2017; Received in revised form 19 March 2018; Accepted 20 March 2018 0167-9457/ © 2018 Elsevier B.V. All rights reserved.

Human Movement Science 59 (2018) 20–29

A.Z. Zivotofsky et al.

characterized under natural, over-ground walking conditions (Zivotofsky, Gruendlinger, & Hausdorff, 2012) and during treadmill walking (Nessler & Gilliland, 2009; van Ulzen, Lamoth, Daffertshofer, Semin, & Beek, 2008). Gait is not the only activity that humans synchronize. Intentional synchronization between an organism and an external cue or between two organisms is a common occurrence in nature. Humans do this during clapping, singing, dancing, and other social activities. A meta-analysis of 42 studies found that synchrony affects social interactions (Mogan, Fischer, & Bulbuliab, 2017). It has positive effects on pro-social behaviours, perceived social bonding, social cognition, and positive affect. Several studies showed that people were able to create improvised arm movements while staying synchronized together (Gueugnon et al., 2016). Such an improvisation situation involves the lower and upper part of the body with the creation of synchronized movements of the arms and legs, and it also requires the maintenance of balance with delicate coordination between the lower and upper parts of the body (Bardy, Marin, Stoffregen, & Bootsma, 1999; Lee, 1989; Stoffregen & Riccio, 1988). Thus, deliberate synchronization of human motor tasks is not uncommon. Synchronization is so fundamental that it often also occurs spontaneously. While we focus here on over-ground walking, a similar phenomenon exists for other tasks. In one of the first studies to examine this question, Schmidt and O'Brien (1997) asked ten pairs of participants to swing pendulums. In trials in which the subjects had visual information about the other participant, dynamical organizing principles were involved in natural, interpersonal synchrony. Schmidt, Bienvenu, Fitzpatrick, and Amazeen (1998) examined the strength of intentional inter-subject coordination of two people oscillating their arms. Pairs of subjects were asked to coordinate pendulum swings to an auditory pulse while also looking at their partner’s pendulum and visually coordinating the oscillation of their pendulum with that of their partner’s. Although there was indeed interpersonal coupling, it was weaker than intrapersonal interlimb coupling. In a test of the effect of somatosensory contact, Sofianidis, Hatzitaki, Grouios, Johannsen, and Wing (2012) observed that even light fingertip contact can increase the spontaneous coordination dynamics of two persons performing rhythmic sway together, and that this synchronization is more pronounced in expert dancers. In addition, “pro-social” individuals were more likely to spontaneously synchronize their repetitive arm curls with another person (Lumsden, Miles, Richardson, Smith, & Macrae, 2012). Demos, Chaffin, Begosh, Daniels, and Marsh (2012) reported that pairs of participants seated side by side in rocking chairs spontaneously coordinated their rocking when they saw or heard the other person, another example of synchronization that occurs during the performance of a motor task even in the absence of deliberate decisions to couple the movements. Several studies examined the role of attention on synchronization on motor tasks other than gait. Richardson, Marsh, and Schmidt (2005) used a form of a dual task paradigm in a task that manipulated attention, when they asked pairs of participants to swing handheld pendulums while engaging in a task of identifying the differences between two cartoon pictures. Visually coupled pairs showed spontaneous synchrony, while those who had only verbal interaction, i.e., intermittent attention, did not. Richardson, Marsh, Isenhower, Goodman, and Schmidt (2007) found that individuals rocking in side-by-side chairs unintentionally entrain, and that the strength of that entrainment is influenced by attention. In other words, attention apparently can influence the synchronization of certain motor tasks. In order to manipulate both the type of information transferred and the knowledge of what perceptual modality the other person has access to, Gipson, Gorman, and Hessler (2016) cleverly used visual occlusion goggles and noise cancelling headphones in a study in which twelve dyads performed a finger oscillation task. They found that there is a complex relationship between prior knowledge and the specific perceptual modality combination between the two participants. The mechanisms whereby two people synchronize their gait, seemingly without any conscious awareness or effort, are not fully understood and there remain many unanswered questions about interpersonal gait synchronization. The aspect that we address in the present study is the role of attention in gait synchronization. Previous investigations sought to determine the degree of synchronization that occurs under varying visual, auditory, and tactile conditions during natural walking (Nessler & Gilliland, 2009; van Ulzen et al., 2008; Zivotofsky, Eldror, Mandel, & Rosenbloom, 2012). The results of these studies verified the existence of this repeatable, heretofore unrecognized, phenomenon that occurs in a subset of walking pairs and quantified the presence and degree of synchronization of gait during side by side over-ground and treadmill stepping under varying visual, auditory, and tactile conditions. That synchronization occurs is not a total surprise; entrainment of the gait rhythm to external cues has been studied in healthy adults and various patient groups, such as patients with Parkinson’s disease who are known to react well to rhythmic sensory cues (Dotov et al., 2017) and, as described above, other interpersonal movement patterns like arm swing may be synchronized. Here we focus on the degree that attention is involved. Today, it is well-recognized that gait is a highly complex, hierarchical process that is regulated by multiple internal brain networks and feedback mechanisms (Maidan et al., 2016; Mirelman et al., 2014, 2015; Nieuwhof et al., 2017; Pelosin et al., 2016) and that gait depends on attention (Al-Yahya et al., 2011; Yogev-Seligmann, Hausdorff, & Giladi, 2008). Thus, dual tasking, a condition that demands the sharing of attention among two tasks, alters the gait pattern of older adults and patients with a wide array of diseases (Amboni, Barone, & Hausdorff, 2013; Belghali, Chastan, Davenne, & Decker, 2017; Montero-Odasso, Verghese, Beauchet, & Hausdorff, 2012; Montero-Odasso et al., 2017; Yogev-Seligmann, Giladi, Gruendlinger, & Hausdorff, 2013). Even healthy young adults change certain aspects of their gait when they walk while performing another task. For example, texting while walking decreases gait speed and alters gait dynamics. The simultaneous performance of another task not only leads to a reduced gait speed in healthy young adults, it can also negatively impact the quality of the “secondary” task (Agostini, Lo, Massazza, & Knaflitz, 2015; Agostini et al., 2015; Lim, Amado, Sheehan, & Van Emmerik, 2015; Srygley, Mirelman, Herman, Giladi, & Hausdorff, 2009). These results highlight the idea that walking utilizes attention and it is not a rote task, even in healthy young adults. In the present study, we set out to investigate the role of attention in gait synchronization when healthy, young adults walk sideby-side over-ground. Specifically, we addressed three questions: 1) Does dual tasking impact gait synchronization? 2) Does the level of cognitive load make a difference, i.e., is the effect on synchronization similar for a relatively simple and a more cognitively demanding dual task? 3) Is the effect of dual tasking on synchronization dependent on the presence or absence of tactile feedback? 21

Human Movement Science 59 (2018) 20–29

A.Z. Zivotofsky et al.

While it is not clear what sort of communication is utilized when a couple walks together in synchrony, previous work (Zivotofsky et al., 2012) has shown that tactile communication, i.e. handholding, is a strong facilitator of synchronized walking. Therefore, we studied the effects of a simple and more complex dual task on synchronization, with and without hand holding. Given the impact of attention on other aspects of gait, even in healthy young adults, we hypothesized that attention, as evaluated using a dual task paradigm, would negatively impact synchronization. 2. Methods 2.1. Subjects Sixteen pairs of healthy young adults (mean age: 26 ± 6 years) participated in the study. Subjects were included if they were between the ages of 20 and 40, and had no known musculo-skeletal, orthopedic, neurologic (e.g., Parkinson’s disease, multiple sclerosis, ADHD, ADD), or other conditions that might affect their gait. Within each pair, subjects were matched for height (less than 10 cm difference; mean height: 168 ± 8 cm) and gender (12 pairs of females and 4 pairs of males). They were recruited from the local student population at Bar Ilan University and from the staff at Tel Aviv Sourasky Medical Center. Subjects provided written informed consent prior to participation, and the study was approved by the human studies committee of the Tel Aviv Sourasky Medical Center. 2.2. Protocol Testing took place along a 70 m long, obstacle free, well-lit path. There was no pre-experiment practice. Subjects were told beforehand to wear comfortable shoes, not high heels or flip-flops. The instructions were to “walk side by side” along the path, with no further elaboration. Subjects were instructed simultaneously to start walking using a gentle tap on their backs, delivered by each hand of the experimenter. Subjects were requested not to speak to each other or to communicate in any other form during the walks. The study participants walked under three conditions that modified the cognitive load and attention; these were performed with and without handholding, i.e., tactile feedback. The conditions were: (1) Baseline (usual walking): Regular walking with auditory and visual feedback (i.e. subject’s hearing and vision were not obstructed), no handholding, and no concurrent task; (2) Simple Dual Task: Regular walking with a simple cognitive load. In this condition, the subjects listened to a section from a story through headphones during the walk. They were told to pay attention and listen for two phonemes and to the content of the story and that they would be asked questions about the story after the walk. (3) Complex Dual Task: Regular walking with a complex cognitive load. The additional task in these trials required the subjects to listen to a section from a story during the walk, and they were told to pay attention and listen for four phonemes as well as the content of the story, and that they would be asked questions about the story after the walk. These conditions were repeated with tactile feedback (i.e., handholding), more specifically, the three additional conditions were: (4) Baseline (usual walking) with Tactile Feedback: Walking with auditory, visual, and tactile feedback. For tactile feedback, the subjects held hands for the duration of the walk. (5) Tactile Feedback and a Simple Dual Task: Walking with Tactile Feedback and a Simple Cognitive Load (as in condition 2); (6) Tactile feedback and a Complex Dual Task: Walking with Tactile Feedback and a Complex Cognitive Load (as in condition 3). Note that in all of the dual task trials, auditory communication between subjects was limited due to the presence of headphones that blocked sound from the environment and the other walker. The content of the story was different in each dual task condition. Visual feedback, which previous studies have shown to be of little significance to synchronization of gait during straight-line walking (Zivotofsky et al., 2012), was not modified. The various conditions were presented in a fixed sequence, as listed above. 2.3. Measurements and data analysis The movements of each subject during all of the conditions were recorded using identical trunk-mounted tri-axial accelerometers (DynaPort Hybrid, McRoberts, the Hague, Netherlands; 87 × 45 × 14 mm, 74 g) that were time-locked and then secured with a belt on the lower back of each subject at the height of L5. The sensor includes a tri-axial accelerometer (sensor range and resolution: ± 6 g and ± 3 mg, respectively) and a tri-axial gyroscope (sensor range and resolution: ± 100°/sec and ± 0.0069°/sec, respectively). The data were sampled at 100 Hz and were downloaded offline, where the vertical acceleration component was low-pass filtered at 3 Hz (using the finite impulse response, fircls1, function in Matlab by MathWorks) and used to calculate mean stride time, asymmetry of the cadences of the two walkers, and a gait synchronization index (GSI), as previously detailed (Zivotofsky et al., 2012). The GSI quantified synchronization on a scale of 0–1 using the phase-synchronization method, a previously described measure of synchronization between two periodic processes (Tass et al., 1998). The GSI is a dimensionless number that measures the ratio between the observed degree of synchronization and the maximum possible level of synchronization. A large GSI suggests that there is a relatively constant phase-shift between the stepping patterns of the two walkers, i.e., they are synchronized, but gives no information about the magnitude of the phase shift. A trial was classified as “synchronized” based on the GSI score. More specifically, if the GSI was greater than 0.20, the trial was considered to be synchronized (Zivotofsky et al., 2012). This threshold was previously identified as being above chance levels by comparing conditions that were not synchronized. To quantify the phase shift, the average absolute phase difference was calculated using circular statistics (Zar, 2007), as previously described (Zivotofsky et al., 2012). In addition to the phase-based measures, we calculated an indicator of gait similarity of each pair of walkers based on the asymmetry of the cadences of the two walkers: 22

Human Movement Science 59 (2018) 20–29

A.Z. Zivotofsky et al.

Fig. 1. Example of synchronized and unsynchronized pair. (a) Raw Signal of vertical acceleration. (b) Filter acceleration signal. (c) Phase obtained from Hilbert transform (d) Wavelet analysis of the synchrony between two signals.

Cadence asymmetry = |cadence1−cadence2|/((cadence1 + cadence2)) where cadence (steps per minute) is defined as stride rate × 120 (2 steps × 60 s/min) or 120/(average stride time). We also calculated the coefficient of variation (CV) for the series of average stride times of each walker as:

CV (%) = 100 x (standard deviation of stride time )/(mean of stride time )

2.4. Statistical analyses Statistical analysis for comparing different conditions was performed using non-parametric pairwise tests (i.e., the Friedman test, a non-parametric test similar to a repeated measure ANOVA, and Wilcoxon signed-rank test, a non-parametric test parallel to the paired t-test). Pairwise comparisons studied the effects of attention, comparing the value obtain during single task, usual walking to those obtained during the simple and complex dual task. We repeated this analysis to evaluate the effects of attention during the handholding conditions and to compare handholding to walking without handholding. In exploratory post hoc analyses, we also used non-parametric tests to compare the gait pattern of the two groups of walkers that emerged, i.e., those that walked in synchrony and those that did not, in order to identify possible differences that may explain why some pairs displayed synchronization and others did not. Measures are reported as means ± SEM (Standard Error of the Mean). SPSS was used for the statistical analyses. A pvalue < 0.05 was considered as statistically significant. 3. Results Examples of the raw and processed data for a synchronized and non-synchronized walk are shown in Fig. 1. As can been seen, the Hilbert transformed signals were clearly overlapping during this example of a synchronized walk, while this overlap was absent 23

Human Movement Science 59 (2018) 20–29

A.Z. Zivotofsky et al.

Table 1 GSI, stride time, and cadence asymmetry in the different walking conditions. Results from all subjects. Walking Condition

GSI

Stride time (sec)

Cadence Asymmetry

Gait speed (m/s)

GSI ∗ 100 % Phase difference

Without Handholding

Baseline (Usual walking) Simple Dual Task Complex Dual Task

0.08 ± 0.10** 0.11 ± 0.09** 0.09 ± 0.11**

1.04 ± 0.05 1.07 ± 0.06*,** 1.09 ± 0.08*,**

0.017 ± 0.013 0.015 ± 0.011 0.013 ± 0.011

1.36 ± 0.06 1.32 ± 0.07*,** 1.31 ± 0.09*,**

0.15 ± 0.18** 0.14 ± 0.13 0.20 ± 0.24

With Handholding

Baseline (Usual walking) Simple Dual Task Complex Dual Task

0.21 ± 0.18* 0.29 ± 0.18* 0.19 ± 0.15*

1.02 ± 0.05 1.06 ± 0.08** 1.08 ± 0.09*,**

0.012 ± 0.018 0.008 ± 0.009* 0.011 ± 0.009*

1.39 ± 0.07 1.33 ± 0.09*,** 1.32 ± 0.09*,**

0.31 ± 0.41* 0.41 ± 0.39 0.46 ± 0.52*

0.83 0.04

0.01 0.002

0.30 0.20

0.009 0.005

0.04 0.64

Dual Task Effect p-value without HH*** Dual Task Effect p-value with HH***

* indicates a significant difference between the marked condition as compared to usual walking (without handholding). ** indicates a significant difference between the marked condition as compared to usual walking while handholding. *** P-value was calculated from Friedman test.

during the walk that was classified as unsynchronized. The wavelet transform provides localized phase-frequency information for two signals. By employing the wavelet transform, time-series movement data is plotted onto a time–frequency plane (Fig. 1). In the time–frequency plane, the frequency components are illustrated on the y-axis while the x-axis represents the time line, and the power spectrum illustrates the power in the frequency domain, which changes throughout the time line. As with the Fourier transform, cross-spectrum analysis can be conducted using the wavelet transform, and cross-wavelet coherence represents the similarity between the two time series at each component frequency throughout the time line (Fujiwara & Daibo, 2016). Thus, the wavelet relative phase pattern evaluated the synchronization and non-synchronization between the couples. This example is consistent with the general findings. Subjects walked using a comfortable gait such that each walk took approximately 90 s, yielding an average of 115.3 ± 7.1 strides per subject per walk. The different walking conditions influenced synchronization (GSI), stride time, cadence asymmetry and gait speed (see Table 1). When walking without handholding, gait speed significantly decreased during the Simple Dual Task and the Complex Dual Task, compared to the baseline value, as expected. The GSI score tended to increase during the Simple Dual Task, compared to the baseline value (see Table 1 & Fig. 2). When evaluated by the rate of synchronized walks as a function of task type, more synchronized walks tended to occur in the Simple Dual Task (see Fig. 3). On the other hand, the mean GSI values and the rate of synchronization tended to decrease in the Complex Dual Task condition, compared to baseline. When walking with handholding, the GSI score tended to increase during the Simple Dual Task, compared to the baseline value (see Table 1 & Fig. 2). In contrast, the GSI significantly decreased (p = 0.003) during the Complex Dual Task. When evaluated by the rate of synchronized walks as a function of task type, a similar pattern occurs, with the greatest percentage of synchronized walks occurring in the Simple Dual Task, and the lowest percentage during the Complex Dual task (see Fig. 3). When comparing the effects of handholding on synchronization, the GSI and the rate of synchronization significantly increased during handholding during all conditions (see Figs. 2 and 3). Handholding had a similar effect on cadence asymmetry, gait speed and stride time. i.e., walking with handholding, improved these measures (see Table 1). As previously observed, we noted that some couples were more likely to walk in synchrony than others. Of the 16 pairs in this study, 50% (8 pairs) walked in synchrony fewer than three of the six walks, while 50% (8) synchronized on four or more of the 6 trials. To better understand and explore the phenomenon under investigation, in post hoc analyses, we divided the subject pairs into two groups: “synchronizers” and “non-synchronizers”. As expected, the GSI value was higher in the synchronized group compared to the non-synchronized group under all conditions (see Table 2); the differences in the GSI values were significant (p < 0.01) in all but two conditions. Among the synchronizers, handholding increased the GSI value in baseline walking and during both dual task

Fig. 2. Handholding promotes synchronization, regardless of the attentional task. HH: handholding. DT: dual task; SDT: simple dual task; CDT: complex dual task.

24

Human Movement Science 59 (2018) 20–29

A.Z. Zivotofsky et al.

Fig. 3. Association between numbers of synchronized couple and the different conditions. HH: handholding; DT: dual task; SDT: simple dual task; CDT: complex dual task.

conditions (see Table 2). Furthermore, in the synchronized subgroup, there was a dual task difficulty effect on the GSI score. During handholding, the Simple Dual task increased the GSI score compared to baseline, while the Complex Dual task decreased the GSI score (p = 0.012). When evaluating the effects of handholding, the GSI values higher (p = 0.02) during Simple Dual task with handholding, compared to without handholding, and the GSI value was significantly higher during the Complex Dual task with handholding, compared to without handholding (p = 0.05). In contrast, among the non-synchronizers there was no significant difference in the level of GSI as a function of task (see Table 2). Synchronized walkers had a higher stride time, i.e., lower cadence (p < 0.002) and lower variability (p < 0.001), compared to unsynchronized pairs in all of the walking conditions. The synchronized subgroup also had significantly smaller stride time variability compared to the unsynchronized couples in the handholding baseline task, the complex dual-task and handholding complex dual task, suggesting that reduced gait variability may be a contributing factor for couples to synchronize (see Fig. 4). To better understand the relation between time domains to synchronization, we calculated average GSI/ absolute value of phase differences over all time. Phase difference decreased as GSI increased (Pearson correlation = −0.24; p = 0.004), reflecting phase differences closer to 0 in the more synchronized walks. As expected, the ratio between GSI to absolute phase difference was significantly higher in the synchronized group compared to the unsynchronized group in each of the tasks: handholding walking (p = 0.04), complex task without handholding (p = 0.02) and simple dual task with handholding and complex dual task with handholding (p = 0.003, p = 0.007, respectively).

4. Discussion The main findings of this study support some of the basic findings previously reported, i.e., that gait synchronization is a repeatable and real phenomenon, that there are couples that are prone to synchronization and others that will rarely or never synchronize, and that handholding promotes gait synchronization. In addition, this work revealed a new finding: a mild cognitive load apparently can increase synchronization, while a more demanding cognitive load decreases synchronization. In other words, not only does dual tasking influence the gait parameters of a lone walker, the performance of secondary task also affects the coupling between two individuals who walk together. With respect to gait, dual tasking, that is the concurrent performance of a motor and cognitive task, can result in the facilitation of the motor task, interference, or have no effect (Woollacott & Shumway-Cook, 2002). The underlying mechanisms of such interference are not fully clear. Leone et al. (2017) summarize three theories: 1) the central capacity sharing model that postulates that in order to accomplish two tasks at once, resources must be re-distributed; 2) the bottleneck model assumes that the two tasks utilize some of the same networks and that certain critical tasks must be carried out sequentially, resulting in a bottleneck; and 3) the cross-talk model proposes that the processing of each of the two tasks will disturb the processing of the other task. In the current study, the parameter of interest in the motor task was synchronized walking. While it is generally accepted that walking involves cognitive processing and is thus affected by dual tasking, it is also well accepted that walking is also controlled by a spinal central pattern generator that is capable of producing basic locomotor movements independently. On the other hand, very little is known about the automaticity/ cognitive control over the synchronization between walkers. The results revealed that a simple dual task tended to increase synchrony while a complex dual task reduced the level of synchrony. A parsimonious explanation for this finding is that gait synchrony, like gait more generally, has an automatic component with cognitive oversight and control. Thus, in the simple dual task condition, the small level of distraction returns the task to a more automatic mode, which in this case results in more synchronization. The complex dual task, which involves a higher level of 25

26

0.07

0.14 0.18 0.15 0.34 0.40 0.31 0.88

± ± ± ± ± ±

0.12** 0.14** 0.17** 0.17* 0.08* 0.08

0.32

1.06 1.09 1.10 1.05 1.09 1.10 0.04

± ± ± ± ± ±

0.05 0.07** 0.09** 0.06 0.09 0.117*

0.16

0.012 0.012 0.008 0.004 0.005 0.006 0.41

± ± ± ± ± ±

0.01** 0.01** 0.01 0.01 0.01 0.01*

0.39

1.35 1.30 1.29 1.35 1.29 1.29 0.02

± ± ± ± ± ±

0.056 0.08** 0.09*,** 0.07 0.10** 0.11

0.45

0.24 0.18 0.36 0.55 0.69 0.81 0.05

± ± ± ± ± ±

0.21** 0.11** 0.25 0.52* 0.32* 0.47*

GSI ∗ 100 % Phase difference

0.36

0.03 0.05 0.02 0.09 0.17 0.05 0.88

± ± ± ± ± ±

0.02** 0.08 0.03** 0.11* 0.20 0.06

* indicates a significant difference between the marked condition, as compared to usual walking (without handholding). ** indicates a significant difference between the marked condition, as compared to usual walking while handholding. HH^ = Handholding. *** P-value was calculated from Friedman test.

***

Usual walking Simple Dual Task Complex Dual Task Usual walking while HH^ Simple Dual Task while HH^ Complex Dual Task while HH^ Dual Task Effect p-value without HH *** Dual Task Effect p-value with HH

Gait speed (m/s)

± ± ± ± ± ±

0.002

1.02 1.06 1.08 0.99 1.03 1.05 0.19

0.05 0.06** 0.08** 0.04 0.04** 0.04*,**

Stride time (sec)

GSI

Cadence Asymmetry

GSI

Stride time (sec)

Unsynchronized pairs (N = 8)

Synchronized pairs (N = 8)

Table 2 GSI, stride time and cadence asymmetry in the different walking conditions. Results from the Synchronized and Unsynchronized pairs (mean ± SD).

0.36

0.022 0.018 0.017 0.021 0.012 0.015 0.60

± ± ± ± ± ±

0.01 0.01 0.01 0.02 0.01 0.01

Cadence Asymmetry

± ± ± ± ± ±

0.005

1.38 1.34 1.32 1.42 1.37 1.35 0.30

0.06 0.07** 0.08** 0.04 0.06** 0.07**

Gait speed (m/s)

0.37

0.03 0.09 0.03 0.10 0.08 0.05 0.47

± ± ± ± ± ±

0.02 0.15 0.02 0.09 0.07 0.03

GSI ∗ 100 % Phase difference

A.Z. Zivotofsky et al.

Human Movement Science 59 (2018) 20–29

Human Movement Science 59 (2018) 20–29

A.Z. Zivotofsky et al.

Fig. 4. Association between stride time variability (CV%) and the different conditions for synchronized and unsynchronized pairs. The significant differences are noted.

distraction, leads to poorer synchronization based on one of the three explanations for dual task interference. As noted above, dual tasking is known to change an individual’s gait parameters. One can suggest, therefore, that the secondary task affected the gait of each participant and that those changes led to the observed changed in synchrony. Alternatively, perhaps the dual task and attention demanding task directly affects the coupling between two subjects. In other words, it is interesting to speculate: does the secondary task affect gait by itself or does the secondary task affect the coupling between two actors? Our study does not enable us to successfully tease apart these interesting alternatives. Future work is needed to further address this interesting question. In the present study, the simple and complex tasks were similar in terms of the nature of the cognitive challenge and attentional resources that were called into play, and differed only in terms of the level of complexity. Different types of cognitive tasks may have different effects on gait (Patel, Lamar, & Bhatt, 2014) and, perhaps, on synchronization. While the present study demonstrates the role of attentional demands in synchronization among couples, in the future, it would be interesting to examine the impact that various types of cognitive tasks have on the level of synchronization. As noted in the Introduction, previous research has revealed that interpersonal synchrony can have affiliative and pro-social consequences, and a more recent study (Lumsden, Miles, & Macrae, 2014) showed that individuals felt better about themselves following a period of synchronous, compared to asynchronous movement, while they also perceived a greater self-other overlap with their partner, a concept used by psychologists to describe how much a person views themselves and their partner as one unit. Moving in time with others may result in individuals feeling better about themselves compared to when they move to their own rhythm. In follow-up work, it will be interesting to examine both the role of attention and possible interacting influences of pro-social consequences on gait synchronization. A possible limitation of the study is that during the dual task trials auditory communication between subjects was minimized, while in typical walking auditory communication is not blocked. Furthermore, auditory communication is known to facilitate synchronization (Zivotofsky et al., 2012). However, owing to the increase in synchronization that was found during the simple dual task, it can be assumed that the task methodology was not a contributing factor in the findings. The sample size should also be noted as a potential limitation of the study, as it may have led to false negative results. Future work with a larger number of subjects is needed to further evaluate the role of attention. Intentional synchronization between organisms is a common occurrence that seems to have a beneficial role. Because of this, unintentional synchrony often occurs in various motor activities (e.g., clapping, arm movement). The present findings have again demonstrated that gait spontaneously synchronizes in a dyad of walkers, and that dual tasking can negatively influence this synchronization. Dual task training which is similar to a targeted task can yield reliable, positive training outcomes (Worden & Vallis, 2014). We speculate that this could suggest that training to synchronize gait while engaged in a simultaneous cognitive task may be transferable to other gait duals tasks and this may have significance for elderly walkers. The negative effects that dual tasking has on gait have been associated with an increased risk of falls (Amboni et al., 2013; Dorfman et al., 2014; Hausdorff, Schweiger, Herman, Yogev-Seligmann, & Giladi, 2008; Montero-Odasso et al., 2012; Springer et al., 2006; Yogev et al., 2005; Yogev-Seligmann et al., 2013) and older people are disproportionally represented in motor vehicle fatalities (Zivotofsky et al., 2012). This may be because they have trouble with dual tasking while crossing the street or navigating obstacles (Maidan et al., 2017; Neider et al., 2011; Simieli et al., 2015). Thus, we speculate that training of synchronized walking while engaged in a dual task may, perhaps, help older adults to learn how to successfully overcome some of the risks of everyday activity that involve dual tasking during walking.

27

Human Movement Science 59 (2018) 20–29

A.Z. Zivotofsky et al.

Acknowledgement We thank the participants for their time. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.humov.2018. 03.009. References Agostini, V., Lo, F. F., Massazza, G., & Knaflitz, M. (2015). Does texting while walking really affect gait in young adults? Journal of NeuroEngineering and Rehabilitation, 12, 86. Al-Yahya, E., Dawes, H., Smith, L., Dennis, A., Howells, K., & Cockburn, J. (2011). Cognitive motor interference while walking: A systematic review and meta-analysis. Neuroscience & Biobehavioral Reviews, 35, 715–728. Amboni, M., Barone, P., & Hausdorff, J. M. (2013). Cognitive contributions to gait and falls: Evidence and implications. Movement Disorders, 28, 1520–1533. Bardy, B. G., Marin, L., Stoffregen, T. A., & Bootsma, R. J. (1999). Postural coordination modes considered as emergent phenomena. Journal of Experimental Psychology Human Perception & Performance, 25, 1284–1301. Belghali, M., Chastan, N., Davenne, D., & Decker, L. M. (2017). Improving dual task walking paradigms to detect prodromal Parkinson’s and Alzheimer’s diseases. Frontiers in Neurology, 8, 207. Demos, A. P., Chaffin, R., Begosh, K. T., Daniels, J. R., & Marsh, K. L. (2012). Rocking to the beat: effects of music and partner's movements on spontaneous interpersonal coordination. Journal of Experimental Psychology: General, 141, 49–53. Dorfman, M., Herman, T., Brozgol, M., Shema, S., Weiss, A., Hausdorff, J. M., et al. (2014). Dual task training on a treadmill to improve gait and cognitive function in elderly idiopathic fallers. Journal of Neurologic Physical Therapy, 38, 246–253. Dotov, D. G., Bayard, S., Cochen, D. C. V., Geny, C., Driss, V., Garrigue, G., et al. (2017). Biologically-variable rhythmic auditory cues are superior to isochronous cues in fostering natural gait variability in Parkinson's disease. Gait & Posture, 51, 64–69. Fujiwara, K., & Daibo, I. (2016). Evaluating interpersonal synchrony: Wavelet transform toward an unstructured conversation. Frontiers in Psychology, 7, 516. Gipson, C. L., Gorman, J. C., & Hessler, E. E. (2016). Top-down (prior knowledge) and bottom-up (perceptual modality) influences on spontaneous interpersonal synchronization. Nonlinear Dynamics, Psychology, and Life Sciences, 20, 193–222. Gueugnon, M., Salesse, R. N., Coste, A., Zhao, Z., Bardy, B. G., & Marin, L. (2016). Postural coordination during socio-motor improvisation. Frontiers in Psychology, 7, 1168. Hausdorff, J. M., Schweiger, A., Herman, T., Yogev-Seligmann, G., & Giladi, N. (2008). Dual task decrements in gait: Contributing factors among healthy older adults. The Journals of Gerontology. Series A, Biological Sciences and Medical Sciences, 63, 1335–1343. Lee, W. A. (1989). A control systems framework for understanding normal and abnormal posture. The American Journal of Occupational Therapy, 43, 291–301. Leone, C., Feys, P., Moumdjian, L., D'Amico, E., Zappia, M., & Patti, F. (2017). Cognitive-motor dual-task interference: A systematic review of neural correlates. Neuroscience & Biobehavioral Reviews, 75, 348–360. Lim, J., Amado, A., Sheehan, L., & Van Emmerik, R. E. (2015). Dual task interference during walking: The effects of texting on situational awareness and gait stability. Gait & Posture, 42, 466–471. Lumsden, J., Miles, L. K., & Macrae, C. N. (2014). Sync or sink? Interpersonal synchrony impacts self-esteem. Frontiers in Psychology, 5, 1064. Lumsden, J., Miles, L. K., Richardson, M. J., Smith, C. A., & Macrae, C. N. (2012). Who syncs? Social motives and interpersonal coordination. Journal of Experimental Social Psychology, 48, 746–751. Maidan, I., Eyal, S., Kurz, I., Geffen, N., Gazit, E., Ravid, L., et al. (2017). Age-associated changes in obstacle negotiation strategies: Does size and timing matter? Gait & Posture, 59, 242–247. Maidan, I., Nieuwhof, F., Bernad-Elazari, H., Reelick, M. F., Bloem, B. R., Giladi, N., et al. (2016). The role of the frontal lobe in complex walking among patients with Parkinson's disease and healthy older adults: An fNIRS study. Neurorehabilitation and Neural Repair, 30, 963–971. Mirelman, A., Bernad-Elazari, H., Nobel, T., Thaler, A., Peruzzi, A., Plotnik, M., et al. (2015). Effects of aging on arm swing during gait: The role of gait speed and dual tasking. PLoS One, 10, e0136043. Mirelman, A., Maidan, I., Bernad-Elazari, H., Nieuwhof, F., Reelick, M., Giladi, N., et al. (2014). Increased frontal brain activation during walking while dual tasking: An fNIRS study in healthy young adults. Journal of NeuroEngineering and Rehabilitation, 11, 85. Mogan, R., Fischer, R., & Bulbuliab, J. A. (2017). To be in synchrony or not? A meta-analysis of synchrony's effects on behavior, perception, cognition and affect. Journal of Experimental Social Psychology, 72, 13–20. Montero-Odasso, M. M., Sarquis-Adamson, Y., Speechley, M., Borrie, M. J., Hachinski, V. C., Wells, J., et al. (2017). Association of dual task gait with incident dementia in mild cognitive impairment: Results from the gait and brain study. JAMA Neurology, 74, 857–865. Montero-Odasso, M., Verghese, J., Beauchet, O., & Hausdorff, J. M. (2012). Gait and cognition: A complementary approach to understanding brain function and the risk of falling. Journal of the American Geriatrics Society, 60, 2127–2136. Neider, M. B., Gaspar, J. G., McCarley, J. S., Crowell, J. A., Kaczmarski, H., & Kramer, A. F. (2011). Walking and talking: Dual task effects on street crossing behavior in older adults. Psychology and Aging, 26, 260–268. Nessler, J. A., & Gilliland, S. J. (2009). Interpersonal synchronization during side by side treadmill walking is influenced by leg length differential and altered sensory feedback. Human Movement Science, 28, 772–785. Nieuwhof, F., Bloem, B. R., Reelick, M. F., Aarts, E., Maidan, I., Mirelman, A., et al. (2017). Impaired dual tasking in Parkinson's disease is associated with reduced focusing of cortico-striatal activity. Brain, 140, 1384–1398. Patel, P., Lamar, M., & Bhatt, T. (2014). Effect of type of cognitive task and walking speed on cognitive-motor interference during dual task walking. Neuroscience, 260, 140–148. Pelosin, E., Ogliastro, C., Lagravinese, G., Bonassi, G., Mirelman, A., Hausdorff, J. M., et al. (2016). Attentional control of gait and falls: Is cholinergic dysfunction a common substrate in the elderly and Parkinson's disease? Frontiers in Aging Neuroscience, 8, 104. Richardson, M. J., Marsh, K. L., Isenhower, R. W., Goodman, J. R., & Schmidt, R. C. (2007). Rocking together: Dynamics of intentional and unintentional interpersonal coordination. Human Movement Science, 26, 867–891. Richardson, M. J., Marsh, K. L., & Schmidt, R. C. (2005). Effects of visual and verbal interaction on unintentional interpersonal coordination. Journal of Experimental Psychology Human Perception & Performance, 31, 62–79. Schmidt, R. C., Bienvenu, M., Fitzpatrick, P. A., & Amazeen, P. G. (1998). A comparison of intra- and interpersonal interlimb coordination: Coordination breakdowns and coupling strength. Journal of Experimental Psychology Human Perception & Performance, 24, 884–900. Schmidt, R. C., & O'Brien, B. (1997). Evaluating the dynamics of unintended interpersonal coordination. Ecological Psychology, 9, 189–206. Simieli, L., Barbieri, F. A., Orcioli-Silva, D., Lirani-Silva, E., Stella, F., & Gobbi, L. T. (2015). Obstacle crossing with dual tasking is a danger for individuals with Alzheimer's disease and for healthy older people. Journal of Alzheimer's Disease, 43, 435–441. Sofianidis, G., Hatzitaki, V., Grouios, G., Johannsen, L., & Wing, A. (2012). Somatosensory driven interpersonal synchrony during rhythmic sway. Human Movement Science, 31, 553–566.

28

Human Movement Science 59 (2018) 20–29

A.Z. Zivotofsky et al.

Springer, S., Giladi, N., Peretz, C., Yogev, G., Simon, E. S., & Hausdorff, J. M. (2006). Dual tasking effects on gait variability: The role of aging, falls, and executive function. Movement Disorders, 21, 950–957. Srygley, J. M., Mirelman, A., Herman, T., Giladi, N., & Hausdorff, J. M. (2009). When does walking alter thinking? Age and task associated findings. Brain Research, 1253, 92–99. Stoffregen, T. A., & Riccio, G. E. (1988). An ecological theory of orientation and the vestibular system. Psychological Review, 95, 3–14. Tass, P., Rosenblum, M. G., Weule, J., Kurths, J., Pikovsky, A. J., Volkmann, J., et al. (1998). Detection of n:m phase locking from noisy data: Application to magnetoencephalography. Physical Review Letters, 81, 3291–3294. van Ulzen, N. R., Lamoth, C. J., Daffertshofer, A., Semin, G. R., & Beek, P. J. (2008). Characteristics of instructed and uninstructed interpersonal coordination while walking side-by-side. Neuroscience Letters, 432, 88–93. Woollacott, M., & Shumway-Cook, A. (2002). Attention and the control of posture and gait: A review of an emerging area of research. Gait & Posture, 16, 1–14. Worden, T. A., & Vallis, L. A. (2014). Concurrent performance of a cognitive and dynamic obstacle avoidance task: Influence of dual-task training. Journal of Motor Behavior, 46, 357–368. Yogev, G., Giladi, N., Peretz, C., Springer, S., Simon, E. S., & Hausdorff, J. M. (2005). Dual tasking, gait rhythmicity, and Parkinson's disease: Which aspects of gait are attention demanding? European Journal of Neuroscience, 22, 1248–1256. Yogev-Seligmann, G., Giladi, N., Gruendlinger, L., & Hausdorff, J. M. (2013). The contribution of postural control and bilateral coordination to the impact of dual tasking on gait. Experimental Brain Research, 226, 81–93. Yogev-Seligmann, G., Hausdorff, J. M., & Giladi, N. (2008). The role of executive function and attention in gait. Movement Disorders, 23, 329–342. Zar, J. H. (2007). Biostatistical analysis (5th ed.). Prentice Hall. Zivotofsky, A. Z., Eldror, E., Mandel, R., & Rosenbloom, T. (2012). Misjudging their own steps: Why elderly people have trouble crossing the road. Human Factors, 54, 600–607. Zivotofsky, A. Z., Gruendlinger, L., & Hausdorff, J. M. (2012). Modality-specific communication enabling gait synchronization during over-ground side-by-side walking. Human Movement Science, 31, 1268–1285. Zivotofsky, A. Z., & Hausdorff, J. M. (2007). The sensory feedback mechanisms enabling couples to walk synchronously: An initial investigation. Journal of NeuroEngineering and Rehabilitation, 4, 28.

29