The development of rhythm regularity, neuromuscular strategies, and movement smoothness during repetitive reaching in typically developing children

Journal of Electromyography and Kinesiology 22 (2012) 259–265

Contents lists available at SciVerse ScienceDirect

Journal of Electromyography and Kinesiology journal homepage: www.elsevier.com/locate/jelekin

The development of rhythm regularity, neuromuscular strategies, and movement smoothness during repetitive reaching in typically developing children Robyn Traynor a, Victoria Galea a,b,⇑, Michael R. Pierrynowski b a b

McMaster Institute of Neuroscience Discovery and Study (MiNDS), McMaster University, Hamilton, Canada School of Rehabilitation Science, McMaster University, Hamilton, Canada

a r t i c l e

i n f o

Article history: Received 30 June 2011 Received in revised form 25 November 2011 Accepted 27 November 2011

Keywords: Neuromuscular strategy Coordination EMG Timing Reaching Development

a b s t r a c t Introduction: This study examined the development of paced coordinated reaching characterized by the successful entrainment of the movement to an external pacer, synchronous muscle activations and movement smoothness. Methods: Thirty children, 5–10 years of age, and ten adults were instructed to repeatedly reach for and move an object from a lower shelf to an upper shelf in time to a metronome. Surface electromyography data were recorded. Amplitude and cross-correlations were calculated on three muscle pairs crossing the shoulder and elbow. A motion capture system captured the space curve accelerations of hand, forearm and upper arm segments to quantify movement smoothness. Results: The 5–6 year old children showed the greatest amount of temporal variability, followed by 7– 10 year olds and then the adults. Correlations between muscle pairs stabilizing the shoulder girdle were higher in each group as compared to the other two muscle pairs but the correlations for all pairs were consistently higher for adults. Movement smoothness for children 9–10 years of age was closer to an adult-like pattern with respect to control of the upper arm, but the hand segment had the greatest variability across groups. Conclusions: The increased temporal variability and decreased movement smoothness of the hand and forearm segments suggest that control of more distal musculature may be more difficult in children. The neuromuscular strategies adopted by adults were more optimal than those adopted by children as reflected by smoother and more consistent reaching. Ó 2011 Elsevier Ltd. All rights reserved.

1. Introduction Reaching for objects is an indispensable life skill, the complexity of which is concealed by its simplistic, seemingly routine execution. Fluent, coordinated movements – like goal-directed reaching – are the result of well-organized neuromuscular activity and the appropriate combinations of muscles acting with optimal force and sequencing (Williams et al., 1983; Karst and Hasan, 1991). The 27th edition of Stedman’s Medical Dictionary defines basic coordination as, ‘‘the harmonious working together, especially of several muscles or muscle groups in the execution of complicated movements’’ (Stedman, 2000, p. 407). How the neuromuscular system achieves ‘‘harmonious’’ movement remains a point of uncertainty, particularly during the development of coordinated movements. Some components of the reach action are thought to be innate, as evident very early in infancy. Recognizable reaching is evident ⇑ Corresponding author at: Health Sciences Centre, HSC 1R1C, McMaster University, 1280 Main Street West, Hamilton, Ontario, Canada L8S 4K1. Tel.: +1 905 525 9140x22620. E-mail address: [email protected] (V. Galea). 1050-6411/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.jelekin.2011.11.015

within the first four months of life, characterized by substantial trajectory variability (von Hofsten, 1991). Between 9 months and 12 years of age, the reach trajectories become progressively straighter and smoother, with fewer aberrant acceleration and deceleration episodes (von Hofsten, 1991; Kuhtz-Buschbeck et al., 1998; Schneiberg et al., 2002). Many of the studies of neuromuscular development in reaching have focussed on muscles involved in postural control. Thelen and Spencer, (1998) followed four infants over the course of their first year to understand the role of postural control in reaching. As their subjects aged, they noted modified strategies of upper trapezius and deltoid activation, critical for further stabilization of the head. They concluded, as others had, that reaching onset is highly dependent on sufficient postural stability, especially with regards to the control of gaze, head balance and supported sitting (Bertenthal and Von Hofsten, 1998; Thelen and Spencer, 1998). Johnston et al. (2002) concluded that difficulty in proximal stability ultimately affects control of the arm in reaching in children with Developmental Coordination Disorder. Zaino and McCoy (2008) also found greater variability in the control of lower limb postural muscles in children with cerebral palsy during a standing reach task. In this same study, the investigators noted a

260

R. Traynor et al. / Journal of Electromyography and Kinesiology 22 (2012) 259–265

decrease in variability in the pattern of muscle activation adopted by typically developing children, suggesting an age-related modification of postural strategies. Less emphasis has been devoted to study of the muscles of the effector: in this case, the arm. The focus of this paper is to characterize the development of coordination as evident in participants from a range of ages completing a goal-directed, externally paced reaching task. The authors propose that harmonious, or coordinated movement is achieved by rhythm regularity, which can be thought of as the successful timing of segment motion, or in this case the successful entrainment of the movement to an external pacer, synchronous muscle activations (i.e. neural success) and movement smoothness (i.e. mechanical success). In this light, a model integrating neuromuscular behaviour with kinematic assessments of segmental behaviour is a viable model to quantify coordinated action with a measure of rhythm regularity that is necessary to the successful completion of the reaching task analysed in this study. Further validity of this model can then be tested in the reach behaviour of typically developing children where, if they are successful in performing the task, will presumably show differences in coordination as their motor/ sensory systems develop. It was hypothesized that adult participants would be better able to match their reaching movements to the given cadence and maintain greater regularity in the rhythm of their reaching for the duration of the task. In order to achieve this rhythm regularity, adult participants would exhibit more defined neuromuscular strategies, particularly for the muscles controlling the effector (i.e. the arm). It was also hypothesized that the youngest participants would exhibit defined strategies for muscles controlling the shoulder girdle, representing a proximal-to-distal development, but greater variation more distally. Furthermore, it was hypothesized that adult participants would exhibit greater movement smoothness throughout the task, as a result of their neuromuscular strategies. The oldest children – ages 9 to 10 – were expected to be most similar to adults in their reaching ability, based on the measures defined here, while the youngest groups of children – ages 5 to 6, 7 to 8 – would be significantly different. 2. Materials and methods 2.1. Sample and recruitment A combined University-Hospital Research Ethics Board approved this study and study participants were recruited from a University community. All participants reported being free from developmental/neurological conditions, such as motor control and coordination disorders, either independently (adults) or by a parent (children). All adult participants and the parents/guardians of the child participants gave informed written consent. The children signed an assent form following an explanation of the study. Participants from 5–10 years of age as well as adult participants, 18+ years old, were recruited. Adult participants were necessary to determine optimal performance of this novel task, thereby serving as a comparator for developmental trajectory in the children. Participants completed the study task using their dominant hands. Most participants (e.g. adults and older children) were able to self-identify the dominant hand. The youngest children were asked to catch and bounce a tennis ball with one hand. They were all also able to print their names on the study assent forms. Dominance was determined by observing which hand the child favoured to complete the ball bouncing and writing tasks. 2.2. Instrumentation and procedures Each participant stood in front of a two-tiered adjustable shelving apparatus (Fig. 1). The lower shelf was set at the height of the

participant’s navel; the upper shelf was positioned 250 mm above the lower shelf to ensure standardization in the distance a 50 g object would be moved between the shelves. The ‘targets’ on the shelves – 13  18 cm rectangles – were large enough such that each participant was able to move the object at a distance that was most comfortable, relative to their body size. The same object was used for all participants. This particular weight was chosen because the youngest children were able to comfortably transport the object without fatigue. The object was placed within a rectangular target on the lower shelf. An identical rectangular target was located on the upper shelf, directly above the lower shelf target. The participants began with both arms at their sides. Listening to a metronome set at 60 beats per minute (1 Hz), participants were instructed to extend their dominant arms to grasp the object on the lower shelf (beat 1), transport the object and place it within the rectangle on the upper shelf (beat 2), and transport the object back to the rectangle on the lower shelf (beat 3). These three beats defined one reach cycle (beat 1 to 3), which was composed of two beat intervals (beat 1 to 2; beat 2 to 3). The cyclical reaching pattern was continued for 30 s. Participants held onto the object throughout the trial, maintaining their grasp while the object was placed on the shelves for each beat. Each participant was requested to complete three of these 30 s reaching trials. They were all given a chance to practice the task prior to beginning data collection. The motion of 19 reflective markers, attached to the participant’s head, torso, dominant arm and hand were collected at 100 Hz using an eight-camera motion capture system (Vicon MX 40+, Denver, Colorado). Commercial software (Vicon Nexus, V1.3) enabled synchronized acquisition of the kinematic and EMG data as well as digital video footage of each trial. Software from the same vendor (Vicon Nexus, V1.3) was used to interpolate and smooth (8 Hz) the marker data, and then define the motion of the participant’s trunk, upper arm, lower arm and hand segments. Trunk motion was not taken into account due to the omission of neuromuscular data from that segment. Therefore in this paper, the focus is on the arm given the inclusion of EMG data from the shoulder girdle, shoulder and elbow joints (see below). The repetitive parasagittal reaching task the participants performed requires shoulder flexion and elbow extension followed by shoulder extension and elbow flexion. Based on these dominant movements, muscle activation data were collected from six muscles using a multi-channel surface EMG system. (Pre-Amplifiers are dual differential with input impedance >100,000 MO; CMRR >100 dB min at 40 Hz and equivalent input noise of <2 lV RMS nominal; Subject-mounted amplifier had an input impedance of 31 KO; and Signal to Nose ratio of >50 dB; Gain was set at 2000; Motion Lab Systems; Baton Rouge, Louisiana.) The specific muscles were: upper trapezius, serratus anterior, anterior deltoid, posterior deltoid, biceps brachii and triceps brachii. Gelled bipolar surface electrodes were collar-taped to the skin. Acceptable electrode placement was assessed by asking each participant to perform various movements and observing expected activation patterns for each muscle. A ground electrode was taped to the participant’s acromion process on the non-dominant side.

3. Data processing and analysis 3.1. Data processing To quantify rhythm regularity, the time interval between two beat events (as described above) was measured. This was documented by observing the displacement of the dominant lateral wrist marker in the vertical Z direction and the digital video record of the reaching trial. Ideally, pairs of beat events were separated by 1 s or 100 video frames, synchronized to the metronome beats. The

R. Traynor et al. / Journal of Electromyography and Kinesiology 22 (2012) 259–265

261

Fig. 1. A participant completes one reach cycle. A participant is shown here in the initial rest position (A) in front of the shelving apparatus. To the beat of the metronome, she grasps the object on the lower shelf (B), transports the object to the upper shelf (C–D), and transports the object back to the lower shelf (E–F), completing one reach cycle. This cycle was repeated for three 30-s trials.

final 20 beat intervals of each trial were used for analysis. Rhythm regularity was quantified using the standard deviation of the 20 beat intervals. The EMG signals were amplified (Motion Lab Systems), 10– 1000 Hz band-pass filtered, then sampled at 2 kHz using resident systems and Vicon Nexus software (V1.3). Using custom software (National Instruments: Diadem), the EMG signals were full-wave rectified then low-pass filtered (dual-pass; 6 Hz Butterworth). The resultant linear envelope EMG values were scaled to a percentage of each participant’s resting EMG signal by taking the average amplitude of the ‘‘resting’’ EMG activity for each muscle and scaling the trial values to that value through simple division. Resting EMG consisted of five seconds of EMG values acquired prior to the timed trials in which the participants remained quiet and did not engage in any movement.

Synchronous muscle activations were quantified by conducting two correlation analyses for three muscle pairs: upper trapezius/ serratus anterior, anterior/posterior deltoid and biceps brachii/triceps brachii. Amplitude correlations were quantified by summing over the entire trial of normalized linear envelopes from each muscle within the pair and then using Pearson Product Moment correlation statistics to quantify amplitude similarity. Cross-correlation analyses were quantified by once again using the normalized linear envelopes but this time using cross-correlation procedures that identified temporal spatial correlations within the trial (Wren et al., 2006). The correlation values were normalized, using the Fisher ‘‘r-to-z’’ transformation (Woolson and Clarke, 2002). To quantify movement smoothness, the motion of the distal end of the hand, forearm and upper arm segments was examined using an analysis of their space curve accelerations over a reach cycle

262

R. Traynor et al. / Journal of Electromyography and Kinesiology 22 (2012) 259–265

(Pierrynowski, 2009). Within a reach cycle, the ratio of the sum of the squared resultant accelerations and the sum of the squared tangential accelerations for each segment’s movement defined an acceleration ratio (ALR). The ALR penalizes motion along curves with large resultant accelerations and it has been mathematically proved to minimize total energy cost (Barr et al., 1992). All ALR values were calculated using custom software. The median value of the 20 reach cycle ALRs was used to improve the reliability of this movement smoothness estimate (Pierrynowski, 2009).

performed on the resulting Fisher transform scores. The within factors were muscle pair and trial. Finally, for movement smoothness, a repeated-measures analysis of variance was first performed for each segment with a within factor of trial. To compare segments, data were collapsed across trials (no significant effect of trial had been found in the first analysis) and another analysis of variance was conducted with a within factor of segment (hand, forearm and upper arm). Instances of group differences for all analyses were explored using Tukey’s post hoc analyses. Statistical significance was set at 0.05. Bonferoni’s correction was used to adjust for multiple comparisons.

3.2. Data analysis IBM SPSS Statistics 19 and R Version 2.3.1 (The R Development Core Team, 2006) were used to statistically analyse the data. Repeated-measures analyses of variance were calculated to analyze rhythm regularity with within factors of trial and beat interval. For the EMG data, repeated-measures analyses of variance were

Table 1 Demographic characteristics of participants. Group 1

Group 2

Group 3

Group 4

N Total Male Female

10 5 5

10 5 5

10 5 5

10 5 5

Age Year range Mean (st dev.) Male Female

5–6 6.20 (0.68) 6.38 (0.42) 6.02 (0.88)

7–8 7.89 (0.55) 7.97 (0.40) 7.80 (0.72)

9–10 9.62 (0.61) 9.49 (0.25) 9.74 (0.86)

18–31 25.32 (4.56) 24.38 (5.98) 26.25 (2.96)

Handedness Right-dominant Left-dominant

8 2

10 0

8 2

9 1

4. Results Forty participants were recruited for this study and their demographic characteristics are reported in Table 1. The participants were divided into four age groups: (1) ages 5–6; (2) ages 7–8; (3) ages 9–10; and (4) ages 18–31 (or ‘‘adult’’). Rhythm regularity varied between the four age groups (F3,36 = 15.35, p < 0.001). The 5–6 age group (mean = 1.033 s; st dev = 0.193 s) was different from the other three groups (p = 0.008, p = 0.018 and p = 0.001, respectively). The 7–8 age (mean = 1.011 s; st dev = 0.0901 s) and 9–10 age (mean = 1.019 s; st dev = 0.138 s) groups were not different from each other, but the 18–31 age group (mean = 1.001; st dev = 0.0525) was significantly different from them (p = 0.010 and p = 0.004, respectively). The 5–6 year old children showed the greatest amount of variability and the 18–31 year olds showed the least. Children 7–10 years old were similar in their level of variability, but significantly different from the two outlying groups. The amplitude correlations indicated a main effect of muscle pair (F2,72 = 29.11, p < 0.001) (Fig. 2). Amplitude correlations between upper trapezius and serratus anterior were higher in each group as compared to the other two muscle pairs. There was also a significant difference between groups (F3,36 = 7.67, p < 0.001), particularly between the 18–31 age group and the 5–6 age

Group 1 (ages 5-6)

Amplitude Correlations

Group 2 (ages 7-8) Group 3 (ages 9-10)

0.80

Mean Amplitude Correlation (Fisher Z)

0.70

Group 4 (ages 18-31)

*

**

0.60

0.50

0.40

0.30

0.20

0.10

0.00

Upper Trapezius & Serratus Anterior

Anterior Deltoid & Posterior Deltoid

Biceps & Triceps

Muscle Pair Fig. 2. Mean EMG amplitude correlations between selected muscle pairs. This figure depicts mean EMG amplitude correlations between selected muscle pairs during the reaching task. The error bars indicate the standard error of the mean. Upper trapezius and serratus anterior were more highly correlated in all groups (⁄). Adults exhibited significantly stronger correlations in the remaining two muscle pairs, as compared to all three groups of children (⁄⁄). A large amount of variability was also exhibited, particularly in the correlation between anterior deltoid and posterior deltoid in adults.

R. Traynor et al. / Journal of Electromyography and Kinesiology 22 (2012) 259–265

(p < 0.001), 7–8 age (p = 0.017) and 9–10 age (p = 0.004) groups. The amplitude correlations were consistently higher across muscle pairs for adults, while the groups of children exhibited significantly lower correlations. When the adults were removed from the analysis of variance, an effect of group approached significance between the remaining younger groups (F2,27 = 3.22, p = 0.056), specifically between groups 1 and 2 (p = 0.057). A main effect of muscle pair (F2,72 = 13.26, p < 0.001) was revealed in an analysis of variance comparing cross-correlation values (i.e. temporal-spatial correlations) (Fig. 3). The upper trapezius and serratus anterior were more highly correlated than the other two muscle pairs for every group. There were no statistically significant differences between groups. There was a high degree of variability in each muscle pair for all groups. This variability is present in both the amplitude and temporal-spatial correlations but appears more prominently in the former correlation. Movement smoothness was different for each group and segment (hand: F3,36 = 9.48, p < 0.001; forearm: F3,36 = 7.80, p < 0.001; upper arm: F3,36 = 8.69, p < 0.001). Tukey’s post hoc analyses for the hand and forearm reveal differences between the 18– 31 age group (hand: median ALR = 1.20; forearm: median ALR = 1.15) and the 5–6 age (hand: median ALR = 1.36, p < 0.001; forearm: median ALR = 1.21, p < 0.001), 7–8 age (hand: median ALR = 1.30, p < 0.001; forearm: median ALR = 1.20, p < 0.001) and 9–10 age (hand: median ALR = 1.28, p = 0.006; forearm: median ALR = 1.17, p = 0.035) groups. For the upper arm, the 18–31 year old group (median ALR = 1.07) was only different from the 5–6 age and 7–8 age groups (both with median ALR = 1.13, p < 0.001) and not significantly different from the 9–10 age group (median ALR = 1.09, p = 0.99). These data were collapsed across trials to compare between segments. A main effect of segment was found: F2,72 = 588.57, p < 0.001. In this regard, the ALR for the upper arm (median ALR = 1.11) was closest to 1.0 for every group, followed by the forearm (median ALR = 1.18) and the hand (median ALR = 1.28), respectively.

263

5. Discussion This study sought to characterize the development of reach coordination with respect to rhythm regularity, synchronicity of muscle activity, and movement smoothness during a timed and repetitive task. This type of task is a good indicator of cerebellar function integration with visual, auditory and sensori-motor areas of brain. The establishment of developmental trajectories to such a task has not been studied in the literature to any great extent. Children as young as five were able to successfully complete the task. However, they were less regular and less smooth in their movements during reaching as compared to adults. Adult neuromuscular strategies were more highly correlated on average, particularly in amplitude, suggesting a superior ability to finely modulate motor output. The adult participants were highly proficient at maintaining the given beat and showed little variability as a group. The youngest participants, on the other hand, were able to maintain the beat on average but showed substantial variability. This observation is in agreement with the broadly assumed notion of age-related error reduction in rhythmic tapping tasks (Rosenbusch and Gardner, 1968; Smoll, 1974). This study found a tight coupling between the upper trapezius and serratus anterior muscles. This strategy is clearly necessary to maintain the correct dynamic alignment of the scapula for activation of the prime movers of the shoulder joint and elbow (Falla and Farina, 2008). Correlations between upper trapezius and serratus anterior surpassed those of anterior deltoids/posterior deltoids and biceps/triceps, muscle pairs necessary for movement about the shoulder and elbow to ensure arm transport (Tortora and Grabowski, 2003). This relationship was particularly evident in the three youngest groups with children 5–6 years of age exhibiting the lowest correlation values in all muscle groups. The adult group showed similar correlation levels in each muscle pair examined. From this observation, it is assumed that the neuromuscular strategy for scapular control, represented here with

Fig. 3. Mean EMG temporal-spatial correlations between selected muscle pairs. This figure depicts mean EMG temporal-spatial correlations between selected muscle pairs during the reaching task. The error bars indicate the standard error of the mean. As in the amplitude domain, upper trapezius and serratus anterior were again more highly correlated than the other muscle pairs in all groups (⁄). Temporal-spatial correlations were higher overall than amplitude correlations. There was no main effect of group.

264

R. Traynor et al. / Journal of Electromyography and Kinesiology 22 (2012) 259–265

serratus anterior and upper trapezius, reaches a high degree of maturation at an early age. This observation may provide support for the view of proximal-to-distal development (von Hofsten and Fazel-Zandy, 1984) in this case, referring to the development of neuromuscular strategies. The ability to stabilize the scapula for reaching may be developed at an early age while control of more distal musculature required for transport of the humerus and forearm develops later. Movement smoothness was most optimal for the arm, followed by the forearm and then the hand. The hand, as the most distal segment, has the greatest amount of physiological ‘‘freedom’’ in movement and must be tightly controlled to maintain the same repetitive cyclical motion. The upper arm follows a much more restricted and thus stereotyped path during cyclical reaching. This rather obvious trend – greater smoothness in the shoulder, which is most constrained biologically, with less smoothness in the hand, which is least constrained – was expected, and speaks to the model’s ability to reflect the reality of the system under study. What is interesting is the difference between groups within this general trend. The youngest two groups of children had very similar smoothness values for all segments. The smoothness values of children 9–10 years of age appeared to be moving closer to an adultlike pattern, as evident in the upper arm where these children were not significantly different from adults. The greatest spread of smoothness values between groups was present in the hand segment. The ability to control the hand within the constraints of the task clearly improves with age. The neuromuscular strategies of children do not match a more adult-like pattern, even by 10 years of age. Fine control of more distal musculature may be more difficult in children, as evident in this study by increased temporal variability and decreased movement smoothness of the hand and forearm segments. The neuromuscular strategies adopted by the adults were more optimal than those adopted by children as reflected by reduced ALR values and more consistent reach trajectories, resulting in greater temporal consistency. Despite the strong neuromuscular correlations reported here, particularly in adults, great neuromuscular variability existed within groups. As has been noted elsewhere, the remarkably consistent temporal-spatial kinematic parameters of reaching in adults (Jeannerod, 1984; Kalaska and Crammond, 1992; Hollerbach and Flash, 1982) are achieved by highly variable muscle activation patterns (Requin et al., 1984; Hepp-Reymond et al., 1996). This finding therefore brings to attention the notion of variability and what this implies for movement in typical, healthy populations. Clearfield et al. (2007) interpreted the kinematic similarity and electromyographic variability in the emerging patterns of reaching of infants as representing the ‘‘many-to-one’’ arrangement of available muscle patterns to a given movement. Chang et al. (2006) made similar observations in the development of walking where more stereotyped gait patterns emerged kinematically, but muscle activation patterns remained variable. This phenomenon has also been observed in the developing postural control system (Zaino and McCoy, 2008; Galea et al., 2004). The issue of variability of neuromuscular strategies within groups may exemplify the degrees-of-freedom issue in neuromuscular control. Bernstein (1957) proposed that early motor development involves fixing joints in an attempt to ‘freeze’ available degrees-of-freedom, assuming that overall control may be made easier with fewer variables. This produces rigidity in the system and the typical dyscoordination and inefficiency of early movements, but is an initial solution to an otherwise overwhelming amount of redundancy (Steenbergen et al., 1995). With experience, additional degrees-of-freedom are systematically released and the system becomes more flexible, ensuring smoother coordination (Bernstein, 1957; Turvey, 1990). A more contemporary approach

proposes that this issue of redundancy is instead a matter of abundance, enabling both stability and flexibility (Zaino and McCoy, 2008; Ranganathan and Newell, 2008; Latash et al., 2007). All degrees-of-freedom contribute to the overall performance outcome but constraint must be applied to any combinations of these elements that would otherwise undesirably alter the final outcome. Less constraint is required for the combinations of degrees-of-freedom that induce the same performance outcome (Latash et al., 2007). The development of coordination as an issue of too many degrees-of-freedom may instead be an issue of too little choice. As indicated above, effective motor control is both stable and flexible. This means flexibility in the sense of a system that is able to select different muscle combinations in order to optimize the neuromuscular strategy according to particular self, task, and environment constraints, producing relatively stable external movements. In this study, adults exhibit greater temporal consistency with less movement effort but high variability in neuromuscular strategies. Development may therefore be seen as one’s improving ability to manipulate inherent neuromuscular flexibility. Schneiberg et al. (2002) suggest a prolonged developmental timeline for movements requiring mastery of many degrees-offreedom. The results are in accordance with this statement as even children as old as 10 years of age were not similar to adults in their temporal consistency, neuromuscular strategies and overall smoothness. Stable, adult-like kinematic patterns of reaching have been demonstrated in children 8–10 years of age during self-driven prehension (Schneiberg et al., 2002). The developmental timeline of neuromuscular strategy maturation and repetitive temporal control is less understood than the kinematic components of reaching. Reaction time and movement time during eye-hand coordination tasks has been shown to decrease with age, particularly evident around 7 and 8 years of age, reaching more optimal levels by age 9 (Hay, 1990). The maturation of neuromuscular and sensory-motor control systems may be the reason for this age-related improvement (Hay et al., 2005). Pierrynowski (2009) has published the reproducibility and responsiveness of the ALR measure of effort and smoothness in walking. It requires a cyclical motion path that is inherent in gait but may be violated in more varied upper limb movements. While the authors have attempted to create a reaching task that minimizes these violations by roughly imitating the repetitive, cyclical nature of the lower limbs during gait, this analysis must still be viewed with caution. For example, the observed effects of reach interval in the movement effort analysis are likely a result of cycle violations. To the authors’ knowledge, no equivalent analyses have been attempted on the upper limb during a cyclical reaching task. 6. Conclusion The purpose of this study was to establish developmental trajectories to a timed and paced task, a good indicator of cerebellar function integration with visual, auditory and sensori-motor areas of brain. This task was investigated via neuromuscular activation strategies and a measure of movement smoothness. The authors expected to uncover a ‘‘transition’’ period around 7 years of age and more ‘‘adult-like’’ coordination by 9 and 10 years of age. Instead, children ages 5–8 exhibited similar coordination patterns while children ages 9–10 represented a potential transition group based on high levels of accuracy but also variability. The adult group was significantly different from all younger groups in all measures. Inclusion of a wider age range (e.g. 4–12 years of age) would undoubtedly provide a more detailed examination of the emergence of mature reach coordination patterns. Nonetheless, this work was instrumental in establishing an age appropriate baseline upon which the mechanism of action in children with

R. Traynor et al. / Journal of Electromyography and Kinesiology 22 (2012) 259–265

developmental disorders, such as Developmental Coordination Disorder (DCD) and cerebral palsy, may be further studied. Acknowledgements Funding for this study was provided in part by a Canadian Institutes of Health Research studentship award. The authors sincerely thank all of the participants (and their parents) for devoting their time to this research. References Barr AH, Currin B, Gabriel S, Hughes JF. Smooth interpolation of orientations with angular velocity constraints using quaternions. ACM SIGGRAPH Computer Graphics 1992;26(2):313–20. Bernstein N. Some emergent problems in the regulation of motor acts (1957). In: Whiting HTA, editor. Human motor actions: Bernstein reassessed. Amsterdam: Elsevier Science Publishers B. V; 1984. p. 343–72. Bertenthal B, von Hofsten C. Eye, head and trunk control: the foundation for manual development. Neurosci Biobehav Rev 1998;22(4):515–20. Chang CL, Kubo M, Buzzi U, Ulrich B. Early changes in muscle activation patterns of toddlers during walking. Infant Behav Dev 2006;29(2):175–88. Clearfield MW, Feng J, Thelen E. The development of reaching across the first year in twins of known placental type. Motor Control 2007;11(1):29–53. Falla D, Farina D. Neuromuscular adaptation in experimental and clinical neck pain. J Electromyogr Kinesiol 2008;18(2):255–61. Galea V, Wright MJ, Barr RD. Measurement of balance in survivors of acute lymphoblastic leukemia in childhood. Gait & Posture 2004;19(1):1–10. Hay L. Developmental changes in eye-hand coordination behaviors: preprogramming versus feedback control. In: Bard C, Fleury M, Hay L, editors. Development of eye-hand coordination across the life span. Columbia: University of South Carolina Press; 1990. p. 217–44. Hay L, Bard C, Ferrel C, Olivier I, Fleury M. Role of proprioceptive information in movement programming and control in 5 to 11-year old children. Hum Mov Sci 2005;24(2):139–54. Hepp-Reymond M, Huesler EJ, Maier MA. Precision grip in humans: temporal and spatial synergies. In: Wing AM, Haggard P, Flanagan JR, editors. Hand and Brain: the neurophysiology and psychology of hand movements. San Diego: Academic Press; 1996. p. 37–68. Hollerbach MJ, Flash T. Dynamic interactions between limb segments during planar arm movement. Biol Cybern 1982;44(1):67–77. Jeannerod M. The timing of natural prehension movements. J Motor Behav 1984;16(3):235–54. Johnston LM, Burns YR, Brauer SG, Richardson CA. Differences in postural control and movement performance during goal directed reaching in children with developmental coordination disorder. Hum Mov Sci 2002;21:583–601. Kalaska JF, Crammond DJ. Cerebral cortical mechanisms of reaching movements. Science 1992;255(5051):1517–23. Karst GM, Hasan Z. Timing and magnitude of electromyographic activity for twojoint arm movements in different directions. J Neurophysiol 1991;66(5):1594–604. Kuhtz-Buschbeck JP, Stolze H, Johnk K, et al. Development of prehension movements in children: a kinematic study. Exp Brain Res 1998;122(4):424–32. Latash ML, Scholz JP, Schoner G. Toward a new theory of motor synergies. Motor Control 2007;11:276–308. Pierrynowski MR. Measurement properties of simple biomechanical measures of walking effort. Comput Math Methods Med 2009;10(3):219–28. Ranganathan R, Newell KM. Online feedback and the regulation of degrees of freedom in motor control. Hum Mov Sci 2008;27(4):577–89. Requin J, Semjen A, Bonnet M. Bernstein’s purposeful brain. In: Whiting HTA, editor. Human motor actions: Bernstein reassessed. Amsterdam: Elsevier Science Publishers B. V; 1984. p. 467–504. Rosenbusch MH, Gardner DB. Reproduction of visual and auditory rhythm patterns by children. Percept Mot Skills 1968;26(3):1271–6. Schneiberg S, Sveistrup H, McFadyen B, McKinley P, Levin MF. The development of coordination for reach-to-grasp movements in children. Exp Brain Res 2002;146(2):142–54. Smoll FL. Development of rhythmic ability in response to selected tempos. Percept Mot Skills 1974;39:767–72. Stedman TS. Stedman’s Medical Dictionary. 27th ed. Baltimore: Lippincott Williams and Wilkins; 2000. Steenbergen B, Marteniuk RG, Kalbfleisch LE. Achieving coordination in prehension: joint freezing and postural contributions. J Motor Behav 1995;27(4):333–48.

265

Thelen E, Spencer JP. Postural control during reaching in young infants: a dynamic systems approach. Neurosci Biobehav Rev 1998;22(4):507–14. Tortora GJ, Grabowski SR. The muscular system. In: Tortora GJ, editor. Principles of Anatomy & Physiology. tenth ed. New York: John Wiley & Sons Inc.; 2003. p. 308–80. Turvey MT. Coordination. Am Psychol 1990;45(8):938–53. von Hofsten C, Fazel-Zandy S. Development of visually guided hand orientation in reaching. J Exp Child Psychol 1984;38(2):208–19. von Hofsten C. Structuring of early reaching movements: a longitudinal study. J Motor Behav 1991;23(4):280–92. Williams HG, Fisher JM, Tritschler KA. Descriptive analysis of static postural control in 4, 6, and 8 year old normal and motorically awkward children. Am J Phys Med 1983;62(1):12–26. Woolson RF, Clarke WR. Statistical methods for the analysis of biomedical data. second ed. Toronto: J Wiley & Sons Inc.; 2002. Wren TA, Do KP, Rethlefsen SA, Healy B. Cross-correlation as a method for comparing dynamic electromyography signals during gait. J Biomech 2006;39(14):2714–8. Zaino CA, McCoy SW. Reliability and comparison of electromyographic and kinetic measurements during a standing reach task in children with and without cerebral palsy. Gait & Posture 2008;27(1):128–37.

Robyn Traynor completed her Masters of Science in the McMaster Integrative Neuroscience Discovery & Study (MiNDS) graduate program at McMaster University (Hamilton, Ontario). The research presented here was a component of this graduate work. Ms. Traynor’s general research interests include child development, both typical and pathological, and knowledge translation. She currently works for a research-service organization focused on knowledge transfer and exchange in public health.

Dr. Galea is an Associate Professor with the School of Rehabilitation Science at McMaster University (Hamilton, Ontario). She is also on the graduate faculty for the Rehabilitation Science, Human Biodynamics, and Graduate Degree in Neuroscience programs. She is the co-Director of the Human Movement Laboratory, the location for this research. Dr. Galea’s research interests involve the study of motor behaviour using neurophysiological assessments as a window into typical and altered motor control of the upper limb. One focus of her interests has been on the development of motor control in typically developing infants and children and children with coordination disorders. Dr. Galea was Ms. Traynor’s graduate supervisor.

Dr. Pierrynowski is an Associate Professor with the School of Rehabilitation Science and the Department of Kinesiology at McMaster University (Hamilton, Ontario). He is also co-Director of the Human Movement Laboratory, the location of this research. Dr. Pierrynowski’s research interests lie in threedimensional analysis of human movement, both in pathological and typical populations, and Directional Statistics. Specifically, his research is focused on (1) evaluating innovative modelling and statistical tools to understand the motion of the ankle, knee, spine, and neck; and (2) the application of these models to enable health care professionals to better treat their patients.