ADAPTED PHYSICAL ACTIVITY QUARTERLY, 1995.12.250-261 O 1995 Human Kinetics Publishers, Inc.

Conditioned Patellar Tendon Reflex Function in Children With and Without Developmental Coordination Disorders Harriet G. Williams and Jeanmarie R. Burke University of South Carolina

A conditioned patellar tendon reflex paradigm was used to study the contributions of crossed spinal and supraspinal inputs to the output of the alpha motoneuron pool in children with and without developmental coordination disorders. The basic patellar tendon reflex response was exaggerated in children with developmental coordination disorders. Crossed spinal and supraspinal influences on the excitability of the alpha motoneuron pool were similar in both groups of children. However, there was evidence of exaggerated crossed spinal and supraspinal inputs onto the alpha motoneuron pool in individual children with developmental coordination disorder.

There has been increasing interest in the so-called clumsy child in recent years (cf. Gubbay, 1978; Lundy-Ekman, Ivry, Keele, & Woollacott, 1991; Williams, Woollacott, & Ivry, 1992). The term refers to a child who has a developmental coordination disorder but is otherwise nondisabled. Although the etiology of clumsiness has not been established, there is general consensus that the basis for developmental coordination disorders is not general intellectual or gross sensory, motor, or neurological impairments (Dare & Gordon, 1970; Smyth & Glencross, 1986; Williams et al., 1992). The motor control deficits exhibited by these children include poor balance and postural control (Williams & Woollacott, in press), inappropriate or inconsistent timing of interlimb coordination actions (Geuze & Kalverboer, 1987; Williams et al., 1992), and inability to produce appropriate levels of force in carrying out a variety of tasks (Lundy-Ekman et al., 1991). Distractibility, decreased ability to attend to tasks, and slower, less precise processing of visual and proprioceptive information have also been reported as characteristic of some children with developmental coordination disorders (Denckla & Rudel, 1978; Hulme, Smart, Morgan, & McKinlay, 1984; Zental, 1985). The authors are with the Department of Exercise Science, University of South Carolina School of Public Health, Columbia, SC 29208.

Patellar Tendon Reflex

251

At the behavioral level, it is quite apparent that motor skill performances of children with developmental coordination disorders are poorer than those of other children of similar chronological ages. At the present time, however, little is known about the potential differences in the central nervous system (CNS) that may be related to the impaired motor performances observed in these children. Although the planning and execution of skillful motor acts is a complex process, one important aspect of such performance involves the precise regulation of the input-output characteristics of the alpha motoneuron pool. The basic patellar tendon reflex response provides researchers with a safe, noninvasive method to study the basic input-output characteristics of the alpha motoneuron pool in humans. The output of the alpha motoneuron pool is influenced by a variety of inputs: crossed spinal inputs, descending inputs from various brain centers, and inputs from the periphery. The inhibitory and excitatory effects of crossed spinal pathways and supraspinal pathways on the output characteristics of the alpha motoneuron pool may be assessed with a conditioned patellar tendon reflex paradigm. In this paradigm, a tap to the left patellar tendon (conditioning stimulus) precedes a tap to the right patellar tendon (test stimulus) at varying time or conditioning intervals. The conditioning interval is the time that elapses between the conditioning stimulus (i.e., left tap) and the test stimulus (i.e., right tap). The purposL of the conditioning stimulus is to activate cross spinal and supraspinal pathways that may influence the input-output characteristicsof alpha motoneuron pool. Whether crossed spinal or supraspinal pathways are activated depends on the duration of the conditioning interval. By comparing the force of the test (right leg) patellar tendon reflex response at each conditioning interval to the basic patellar tendon reflex response (the conditioned patellar tendon reflex recovery profile), the tester may assess contributions of crossed spinal pathways (conditioning intervals I 75 ms) and supraspinal pathways (conditioning intervals 2 90 ms) (Burke, Kamen, & Koceja, 1993). We hypothesize that there are developmental differences in the input-output properties of the alpha motoneuron pool between children with developmental coordination disorders and their nondisabled peers. Specifically, changes in sensory feedback andlor motoneuron excitability of the peripheral reflex loop, crossed-spinal pathways, and supraspinal pathways may be impaired in children who are clumsy. This may account for (a) their inability to precisely regulate muscle force as shown by Lundy-Ekman et al. (1991) and (b) their reduced precision in processing proprioceptive feedback as shown by various authors (Denckla & Rudel, 1978; Hulme et al., 1984; Zental, 1985). By comparing the force of the basic patellar tendon reflex response in nondisabled children and children with developmental coordination disorders, we can address the issue of developmental changes in the sensitivity of the peripheral reflex loop. The force of the conditioned patellar reflex response, then, can be used to identify changes in spinal and supraspinal influences that affect the excitability of the alpha motoneuron pool in children who are clumsy. An understanding of these spinal and supraspinal influences is important because the successful performance of voluntary movements depends upon the ability of the CNS to precisely regulate the excitability (i.e., output) of the alpha tnotoneuron pool, the final common pathway. Thus, an increased excitabilityof the alpha motoneuron pool from spinal and supraspinal influences or peripheral feedback, as measured by increases in reflex force,

Williams and Burke

252

may contribute to the motor control problems of children with developmental coordination disorders. The purposes of this study were (a) to compare patellar tendon reflex function of children with developmental coordination disorders with that of children with nonimpaired motor development and (b) to assess the potential influences of crossed spinal and supraspinal inputs on these reflex responses. The overall goal was to identify potential nervous system mechanisms underlying motor control difficulties in children with developmental coordination disorders.

Method Five children with developmental coordination disorders (all males, age = 7.95 1.18 years) were recruited from a clinical population of children enrolled in sensory-motor enrichment programs at The University of South Carolina. Eight children with nonimpaired motor development (7 males, 1 female, age 8.25 i 1.69 years) were recruited from a local elementary school and equated for age. Children were screened on gross and fine motor tasks. Gross motor control tasks (Williams's Gross Motor Control Test Battery, 1992) included one-foot balance on the preferred foot with eyes open and closed (reliability = 32, 50-ft (15.24-111) hop on both right and left feet (reliability = .83), and an agility task (reliability = .79). The agility task required the child to complete 10 continuous trips between two tape marks 10 ft (3.1 m) apart. Fine motor tasks included bimanual and unimanual object manipulation tasks (Frey, 1979) and a pencilusage task (Hammill, Pearson, & Voress, 1993). The bimanual object manipulation task involved using both hands to pick up, turn, and place 12 circular discs (1 in. [2.56 cm] in diameter) in a peg board. The unimanual task required the child to use the preferred hand to pick up and place 12 small pegs (114 in [0.64 cm] in diameter) in a peg board. The pencil-usage task was the Eye-Motor Coordination subtest from the Frostig battery. The conditioned patellar tendon reflex paradigm, in which a tap to the left patellar tendon (conditioning stimulus) preceded a tap to the right patellar tendon (test stimulus), was used to assess reflex function. The conditioning intervals were 15, 30, 45, 60, 75, 90, 105, 120, 135, 150, 300, and 600 ms. Three reflex trials were randomly elicited at each of the 12 conditioning intervals. Three control reflex trials (test reflex stimulus only) were randomly interspersed among the conditioned reflex trials. The children were seated comfortably on a custom-designed reflex bench. Rubber-tipped electromagnetic solenoids were aligned to deliver a patellar tendon tap to the right and left legs independently. Each solenoid was in series with a piezoelectric force transducer to monitor the force of each tendon tap. The outputs of the right and the left piezoelectric force transducers were measured on each trial to ensure that the conditioning and test reflex stimuli were supramaximal. A load cell transducer was secured against the right ankle to measure the force of the patellar tendon reflex response. A BASIC computer program was used to control the timing of the tendon taps, to monitor the forces of the tendon taps, and to measure the force of the right patellar tendon reflex. The tendon tap forces and the reflex force were sampled at a rate of 2 kHz per channel. Peak forces of the control patellar tendon reflex responses were compared

+

Patellar Tendon Reflex

253

between the two groups of children using a t test for independent means. A group-by-conditioning-interval ANOVA model was used to reveal differences in the conditioned patellar tendon reflex recovery curve between children with developmental coordination disorders and their nondisabled peers. The Dunnett's post hoc test was used to determine which conditioned reflex forces were significantly different from the control patellar tendon reflex force. All statistical tests were conducted at .05 level of significance. For each subject, percentage of control values were calculated at all conditioning intervals.' These percentages of control values were also analyzed using the group-by-conditioning-interval ANOVA model described in the preceding paragraph. The purpose of this analysis was to verify that differences in baseline reflex force did not bias the analysis of the conditioned patellar tendon reflex responses. The conditioned patellar tendon reflex percentages repolted in the text reflect these percentage of control values. Individual conditioned patellar tendon reflex recovery curves were also qualitatively examined for children who were clumsy to identify potential intersubject variability.

Results Motor performance data are given in Table 1. Children with developmental coordination disorders balanced with eyes open for 6.4 s; the average for children with nonimpaired motor development was 21 s. Time in balance with eyes closed was 3.8 s for clumsy children and 13.5 s for children with nonimpaired motor development. The average time for the 50-ft (15.24-m) hop for children with developmental coordination disorders was 12.7 s; nondisabled children were faster and completed the distance in 7.5 s. Two children who were clumsy could not hop. Time in the agility task was 18.1 s for clumsy children and 17.0 s for children with nonimpaired motor development. Children with coordination disorders were slower in both bimanual (21.1 s vs. 11.0 s) and unimanual (32.4 s vs. 17.0 s) fine motor object manipulation tasks. Children with developmental coordination disorders were more than 2 years behind usual age expectations in performing pencil usage tasks (age equivalent scores were 6 years 0 months vs. 8 years 3 months, respectively). In general, children with coordination disorders, as a group, were more variable in performance than were the other children. The force of the basic patellar tendon reflex was greater by 105% in children with coordination difficulties than in comparison children (Figure 1; p i.05). This result suggests that one characteristic of the children with coordination difficulties was that a greater proportion of the alpha motoneuron pool was recruited by the peripheral reflex loop. The main effect of conditioning interval was significant O, < .05). There was no significant Group x Conditioning Interval interaction @ > .05). Dunnett's post hoc test indicated that reflex forces were facilitated only between the 75

'Percentage of control = {[(CI - Control/Control) . 1001 + loo), where CI refers to reflex force at the specific conditioning interval and Control refers to force of the control patellar tendon reflex response.

Williams and Burke

254

Table 1 Motor Performance of Children With and Without Developmental Coordination Disorders (DCD) DCD children (n = 5)

Non-DCD children (n = 8)

Task

M

SD

M

SD

Balance, eyes open Balance, eyes closed 50-ft hop Agility Bimanual coordination Unimanual coordination Pencil usage

6.4 3.8 12.7 18.1

4.2 2.5 4.3 3.9

21.0 13.5 7.5 17.0

3.5 1.8 2.1 3.0

24.1

5.2

11.0

3.5

32.4 610

6.3 212

17.0 813

3.9 115

Note. All values except pencil usage are in seconds. Pencil usage values are in years/ months.

PATELLAR TENDON REFLEX BASIC RESPONSE

t

I ICLUMSY

0 CONTROLS

LL

MOTOR DEVELOPMENT LEVELS Figure 1 - Comparison of the basic patellar tendon reflex responses of children with developmental coordination disorders and age-equated controls. The error bars denote the standard errors of the means.

Patellar Tendon Reflex

255

ms and 150 ms conditioning intervals and were facilitated similarly in both groups of children (Figure 2; p < .05). Analysis of the percentage of control values indicated that there was again a significant conditioning interval main effect only between 75 ms and 150 ms conditioning intervals @ < .05) and no significant Group x Conditioning Interval interaction. This supports the analysis of the absolute force data which showed that the amount of reflex facilitation was similar for both groups (Table 2).

7

%

50 T

0

=-. 0-0

CLUMSY CONTROLS

I I

I

I I

,, #,

I

I

I

I I

I d

I,

RIGHT 15 45 75 105 135 300 600 ONLY INTERVAL BETWEEN LEFT HAMMER TAP & RIGHT HAMMER TAP (MSEC)

Figure 2 - Comparison of the conditioned patellar tendon reflex recovery curves of children with developmental coordination disorders (DCD) and age-equated controls. The error bars denote the standard errors of the means (up, DCD children; Down, age-equated controls). The data curves were fitted with a cubic spline function. Table 2 Percentage of Control Values Between 75 ms and 150 ms Conditioning Intervals in Children With and Without Developmental Coordination Disorders (DCD) Conditioning intervals (ms) Groups Non-DCD DCD

M SD M SD

5

90

105

120

135

150

340.8 213.88 267.9 176.27

374.0 196.17 271.1 238.06

357.6 238.04 248.7 219.03

288.4 117.47 223.9 182.39

336.8 230.27 240.4 235.50

323.7 156.49 225.3 161.80

256

Wi//iarnsand Burke

A similar amount of reflex facilitation at the 75 ms conditioning interval suggests that the contribution of crossed spinal pathways to motoneuron excitability was not different for either group. However, short-latency reflex effects (those occurring between the 15 ms and 60 ms conditioning intervals) were observed in some children with developmental coordination disorders but in none of the comparison children (Figure 3). Supraspinal influences on the excitability of the alpha motoneuron pool were also not different between the two groups of children as indicated by the similar amount of reflex facilitation between the 90 ms and 150 ms conditioning intervals. It is important to note that qualitative analyses of case studies showed some potentially interesting differences in the supraspinal contribution to reflex excitability in individual children who were clumsy (Figure 3). The percentages reported in the Case Studies section reflect percentage of control values for each subject, with 100% being the baseline control value.

Case Studies Case A. This child showed increased activity in crossed-spinal inhibitory and excitatory pathways as indicated by the pronounced reflex inhibition at the 15 ms (14%), 30 ms (18%), and 45 ms (71%) conditioning intervals and the subsequent reflex facilitation at the 60 ms (209%) conditioning interval. There was also a strong supraspinal inhibitory effect at the 600 ms (1 1%) conditioning interval. Reflex facilitation between the 75 ms and 150 ms conditioning intervals was similar to age-equated controls. Case B. This child also showed increased excitability of crossed-spinal inhibitory and excitatory pathways; the nature of the effects at early conditioning intervals was different from Case A (Figure 3B). There was an initial reflex inhibition (39%) at the 15 ms conditioning interval and a subsequent reflex facilitation (170%) at the 30 ms conditioning interval. This was followed by a return to baseline force value at the 45 ms conditioning interval. This early cyclic conditioned reflex response has been observed previously in young adults (Burke, Kamen, & Koceja, 1989). However, the initial inhibition and subsequent facilitation were only 80% and 120% of control values. Unlike Case A and the age-equated controls, the conditioned reflex responses at the 300 ms (173%) and 600 ms (122%) intervals were facilitated. The cyclic nature of the short-latency effects and the long-latency facilitation effects implies that influences from crossed-spinal and supraspinal pathways may contribute to the coordination difficulties in this child. Case C . The most noteworthy characteristic in this child was the extremely exaggerated baseline patellar tendon reflex response (31.8 16.00 N). This subject also showed a short-latency inhibition at the 30 ms (57%) conditioning interval and a strong long-latency inhibition at the 300 ms (36%) and 600 ms (26%) conditioning intervals. Although there appeared to be a strong reflex facilitation at the 75 ms conditioning interval, the amount of facilitation was only 143%. As a group, the children who were clumsy showed a reflex facilitation of 268% at the 75 ms conditioning interval (Table 1). Case D. This child demonstrated a strong short-latencyreflex inhibition at the 30 ms (40%) and 45 ms (49%) conditioning intervals and a strong supraspinal inhibitory effect at the 600 ms (47%) conditioning interval; these results were similar to those for Case itated conditioned reflex responses (75 ms

+

.

Patellar Tendon Reflex

257

to 150 ms) were similar to the age-equated controls in terms of percent of control values; this was due to the vigorous baseline reflex response (26.4 f 3.67 N). Case E. This child had strong excitatory inputs from crossed-spinal pathways as evidenced by the short-latency reflex facilitation at the 15 ms (428%), 30 ms (444%), and 45 ms (611%) conditioning intervals. There was also a strong supraspinal excitatory effect at the 300 ms (471%) conditioning interval. All other conditioned reflex responses were similar to age-equated controls.

Discussion The most noteworthy neuromuscular characteristic identified in the children with developmental coordination disorders was the exaggerated patellar tendon reflex response. This finding implies that the sensitivity of the peripheral reflex loop is increased in these children. This suggests that stronger peripheral input from the Ia afferents for a given stimulus intensity recruits a greater proportion of the alpha motoneuron pool to participate in the reflex response. Consequently, reflex force output is greater in children who are clumsy than in children without coordination difficulties. Two potential mechanisms to account for enhanced tendon reflexes are (a) an increase in muscle spindle gain via the gamma motor system, a central mechanism, or (b) an increase in muscle tendon stiffness, a peripheral mechanism. There is no evidence to date suggesting that increases in muscle tendon stiffness are characteristic of developmental coordination disorders. Moreover, the children in our study did not experience flexibility problems. An increase in muscle spindle gain in the resting muscle could explain reduced precision in processing of proprioceptive feedback during movement control as shown by various authors (Denckla & Rudel, 1978; Hulme et al., 1984; Zental, 1985). The precision of proprioceptive feedback is maintained during the performance of voluntary movements by alpha-gamma coactivation (cf. Hagbaxth, 1993). A high background activation of the gamma motoneuron pool would adversely affect the ability of the CNS to properly set muscle spindle gain during the performance of voluntary movements (Hulliger, 1993; Llewellyn, Yang, & Prochazka, 1990). Thus, the ability to maintain the precision of proprioceptive feedback during movement control would be compromised in developmental coordination disorders. In support of the central mechanism proposed above, timing control problems observed in clumsy children have been attributed to deficits in a central timing mechanism and not to peripheral mechanisms involved with response outcome (Williams et al., 1992). Cerebellardysfunction may underlie these central timing mechanisms (Ivry & Keele, 1989). Related to our results, the cerebellum influences the background activity of the gamma motoneuron pool (cf. Gorassina, Prochazka, & Taylor, 1993). Moreover, exaggerated tendon reflexes have been observed in patients between the ages of 2 and 14 years with a cerebellar disorder and no concomitant peripheral neuropathy (Ozeren, Arac, & Ulku, 1989). Regardless of the specific CNS structures, it appears that developmental coordination disorders may be related to CNS dysfunction and not to developmental delay. Our data suggest that spinal and supraspinal influences affecting the excitability of the alpha rnotoneuron pool are similar in children with and without

Williams and Burke

INlDWK BEFWW MW M E R TAP & MDTT W E R TAP (YSEC)

a W 0

5

.. 1s

:

L

:

h

ONLY INlDWKBEFWWMHAYU

1L ,h

:

:%Em

TAP&RK;HTWMERTAP

(ma

nz

lk

O'Rft3.k : : 7i5 ONLY INlDWKBEFWWMWM

:

1k

(a

: lAi : T m

TAP&RK;HTWERTAP

Figure 3 - Conditioned patellar tendon reflex recovery profiles for individual children with developmental coordination disorders compared to age-equated controls. For individual children who were clumsy, the error bars denote the standard errors of means of the three trials at each conditioning interval. The error bars denote the standard errors of the means for the age-equated controls. The data curves were fitted with a cubic spline function. Continued on next page.

Patellar Tendon Reflex

=- .....m mDl s 0-.

40--

=-I

S "I: "oo W

P

..

O'm&

ONLY

1:s

: 4 i : 7il : 1&

:

1$

: ~ J M ) m

BOWW M HAYUW TAP & B(;M WUER TAP

(-1

-

Ei =:

..--. --•

mmls

540

P Figure 3

'

A

: : 7'5 : 1& 11$ ' T m ONLY I N l D W A ~ l E T H A Y U ~ T A P & R l M ~ T A P

01m4r i5

- continued.

(=I

developmental coordination disorders. An increase in the excitability of the alpha motoneuron pool in children with coordination problems could easily explain the inability to precisely regulate muscle force as shown by Lundy-Ekman et al. (1991). However, for individual children who were clumsy, we did observe potential differences in short-latency crossed-spinal effects (160 ms conditioning intervals) and long-latency supraspinal effects (300 ms and 600 ms conditioning intervals) on the excitability of the alpha motoneuron pool. The sources of these supraspinal effects may involve transcortical, transcerebellar, or reticulospinal pathways (cf. Burke et al., 1989, 1993). Although the conditioned reflex data were not conclusive, the ability of the CNS to precisely regulate the excitability (i.e., output) of the alpha motoneuron pool during performance of voluntary movements may be impaired in children with developmental coordination disorders. It is also plausible that the increased sensitivity of the peripheral reflex loop in children who are clumsy could explain their inability to precisely regulate muscle force. It is well known that Ia afferent input can assist with the recruitment of alpha motoneurons for the production of voluntary movements, such as walking (Capaday & Stein, 1986, 1987), running (Capaday &

Williams and Burke

260

Stein, 1987; Dietz, Schmidtbleicher, & Noth, 1979), and isolated muscle contractions (Butchart, Farquhar, Part, & Roberts, 1993; Hultborn, Meunier, Pierrot-Deseilligny, & Shindo, 1987). Thus, enhanced proprioceptive feedback from the peripheral musculature during voluntary movements may impair the ability of the CNS to regulate precisely the recruitment of alpha motoneurons and consequently muscle force gradation. In conclusion, developmentalcoordinationdisorders may be associated with increased sensitivity of the peripheral reflex loop. This neuromuscular finding in children with coordination difficulties may contribute to their inability to precisely regulate muscle force and their reduced precision in processing proprioceptive feedback. There is also some evidence to suggest that influences of crossedspinal and supraspinal pathways on the excitability of the alpha motoneuron pool are affected in some children with coordination disorders.

References Burke, J.R., Kamen, G., & Koceja, D.M. (1989). Long-latency enhancement of quadriceps excitability from stimulation of skin afferents in young and old adults. Journal of Gerontology, 44, 158-163. Burke, J.R., Kamen, G., & Koceja, D.M. (1993). The relationship between body height extremes and the conditioned patellar tendon reflex response. International Journal of Neuroscience, 72, 45-58. Butchart, P., Farquhar, R., Part, N.J., & Roberts, R.C. (1993). The effect of age and voluntary contractionon presynaptic inhibition of soleus muscle Ia afferent terminals in man. Experimental Physiology, 78, 235-242. Capaday, C., & Stein, R.B. (1986). Amplitude modulation of the soleus h-reflex in the human during walking and standing. Journal of Neuroscience, 6, 1308-13 13. Capaday, C., & Stein, R.B. (1987). Difference in the amplitude of the human soleus h reflex during walking and running. Journal of Physiology, 392, 513-522. Dare, M.T., & Gordon, N. (1970). Clumsy children: A disorder of perception and motor development. Developmental and Child Neurology, 12, 178-185. Denckla, M., & Rudel, R. (1978). Anomalies of motor development in hyperactive boys. Annals of Neurology, 3, 23 1-233. Dietz, V., Schmidtbleicher,D., & Noth, J. (1979). Neuronal mechanisms of human locomotion. Journal of Neurophysiology, 42, 1212-1222. Frey, C. (1979). Frey fine motor control tests. Unpublished master's thesis, University of Toledo, Toledo, OH. Geuze, R., & Kalverboer, A. (1987). Inconsistency and adaptation in timing of clumsy children. Journal of Human Movement Studies, 13,421-432. Gorassina, M., Prochazka, A., & Taylor, J.L. (1993). Cerebellar ataxia and muscle spindle sensitivity. Journal of Neurophysiology, 70, 1853-1862. Gubbay, S. (1978). The management of developmental apraxia. Developmental Medicine and Child Neurology, 20, 643-646. Hagbarth, K. (1993). Microneurography and applications to issues of motor control: Fifth annual Stuart Reiner memorial lecture. Muscle and Nerve, 16, 693-705. Hammill, D., Pearson, N., & Voress, J. (1993). Developmental Test of Visual Perception. Austin, TX:Pro-Ed. Hulliger, M. (1993). Fusimotor control of proprioceptive feedback during locomotion and

Patellar Tendon Reflex

261

balancing: Can simple lessons be learned for artificial control of gait? Progress in Brain Research, 97, 173-180. Hulme, C., Smart, A., Morgan, G., & McKinlay, I. (1984). Visual, kinesthetic and crossmodal judgments of length by clumsy children: A comparison with young normal children. Child Care, Health, and Development, 10, 117-125. Hultbom, H., Meunier, S., Pierrot-Deseilligny, E., & Shindo, M. (1987). Changes in presynaptic inhibition of Ia fibres at the onset of voluntary contraction in man. Journal of Physiology, 389, 757-772. Llewellyn, M., Yang, J.F., & Prochazka, A. (1990). Human h-reflexes are smaller in difficult beam walking than in normal treadmill walking. Experimental Brain Research, 83, 22-28. Lundy-Ekman, L., Ivry, R., Keele, S., & Woollacott, M. (1991). Timing and force control deficits in clumsy children. Journal of Cognitive Neuroscience, 3, 367-376. Ozeren, A., Arac, N., & Ulku, A. (1989). Early-onset cerebellar ataxia with retained tendon reflexes. Acta Neurologica Scandinavica, 80, 593-597. Smyth, T., & Glencross, D. (1986). Information processing deficits in clumsy children. Australian Journal of Psychology, 38, 13-22. Williams, H., & Woollacott, M. (in press). Neuromuscular characteristics underlying postural control in clumsy children. Advances in Motor Development Research. Williams, H., Woollacott, M., & Ivry, R. (1992). Timing control in clumsy children. Journal of Motor Behavior, 24, 165-172. Zental, S. (1985). Structured tasks: Effects on activity and performance of hyperactive and comparison children. Journal of Educational Research, 79, 91-95.

Acknowledgments This research was supported by the University of South Carolina Research and Productive Scholarship Faculty Grant Program (1 1050-E130).