Motor learning of a bimanual task in children with unilateral cerebral palsy

https://doi.org/10.1016/j.ridd.2013.03.008Get rights and content

Highlights

  • Children with unilateral CP showed reduced and slow learning of a bimanual task.

  • The learning curves fit to an exponential function for children with unilateral CP.

  • The importance of sufficient practice of motor tasks for rehabilitation schedules.

Abstract

Children with unilateral cerebral palsy (CP) have been shown to improve their motor performance with sufficient practice. However, little is known about how they learn goal-oriented tasks. In the current study, 21 children with unilateral CP (age 4–10 years old) and 21 age-matched typically developed children (TDC) practiced a simple bimanual speed stack task over 15 days of practice. Both groups demonstrated their ability to learn the current bimanual task, but their rate of improvement and learning pattern differed. Children with unilateral CP overall were slower and improved ~10% less than TDC. Most of the improvement occurred during the first 3 days for the TDC, whereas performance did not plateau until 6–8 days for the children with unilateral CP. This initial slower learning rate for children with unilateral CP was also confirmed by better fitting of the curve to an exponential function than the power law function (p < 0.05). Therefore, when working with children with unilateral CP, sufficient practice is important (two to three times more than for TDC), and delayed improvement is expected.

Introduction

Children with unilateral cerebral palsy (CP) have early non-progressive lesions of the brain that result in a number of impairments predominantly on one side of their body (e.g., Coluccini et al., 2007, Domellof et al., 2009). In addition, they were found to have bimanual coordination problems with less synchronized and more sequential movements of their two hands (e.g., Charles and Gordon, 2006, Hung et al., 2004, Hung et al., 2010, Steenbergen et al., 2008, Utley et al., 2004), which affect their functional independence and quality of life.

Recently several promising approaches to improve hand function for children with unilateral CP have been developed, including “Constraint-Induced Movement Therapy (CIMT)” (e.g., Charles et al., 2006, Gordon et al., 2005, Sakzewski et al., 2009, Stearns et al., 2009), intensive bimanual training (e.g., Charles and Gordon, 2006, Gordon et al., 2008, Gordon et al., 2007, Sakzewski et al., 2011a, Sakzewski et al., 2011b), and a combination of both unimanual and bimanual intensive training (e.g., Aarts et al., 2011). These approaches usually consist of 10–15 days of intensive treatment with very few breaks in between. Significant improvements in unimanual or bimanual hand function were found before and after these intensive treatments.

These intensive treatments are based on motor learning approaches to rehabilitation (Carr & Shepherd, 2010), however, we still don’t know much about how children with CP learn goal-oriented tasks. Children with CP have shown improvement in their performance after practice during goal-oriented tasks (e.g., Gordon and Duff, 1999, Valvano and Newell, 1998). The typical motor learning concept (based on healthy young adults) that random practice (practice multiple tasks in a random order) facilitates better learning than block practice (consecutively repeating the same task before moving on to the next task) has been reported to not occur for children with CP, i.e., learning rates were similar for both types of practice schedules (Duff & Gordon, 2003). Another motor learning concept whereby reduced feedback frequency benefits learning was found to be true for adults with CP (Hemayattalab & Rostami, 2010), but not for typically developed children (TDC, Goh et al., 2012, Sidaway et al., 2012, Sullivan et al., 2008). TDC seem to benefit more from higher feedback frequency than adults, especially for difficult tasks. Therefore, general motor learning principles that have been established based on the research of healthy young adults may not always apply to children with CP. Most studies of motor learning in children with CP only investigated in the changes of task performance before and after practice. It is unknown if most of the improvements occur during early, middle, or late practice. Better understanding of the learning pattern for children with CP can help us attend to their learning needs to maximize their learning efficiency (i.e., it could determine the optimal amount of practice).

There are two proposed learning patterns for healthy adults. The first pattern is the “power law of practice” (Crossman, 1959, Newell and Rosenbloom, 1981, Snoddy, 1926): a linear log–log relationship between performance measures and practice trials. The second pattern is the “multiple time scales of nonlinear dynamic systems” (Liu et al., 2010, Newell et al., 2001): persistent and transitory changes in task performance evolve from different subsystem time scales. Since the current study only had 45 practice trials and no manipulating control parameters, it may fit a single exponential function, i.e., small changes in the beginning of practice and a more rapid improvement subsequently (Liu et al., 2003, Stratton et al., 2007) instead of the multiple exponential functions. It is not known which learning pattern children with unilateral CP and TDC will follow.

In the current study, we investigate the learning pattern for children with unilateral CP and compare it with TDC using a novel and simple bimanual task. Given the hand function limitation of the more affected side, we hypothesize a different, and delayed learning rate, for children with unilateral CP.

Section snippets

Participants

Twenty-one children with unilateral CP and 21 age-matched, typically-developing right-handed children participated in this study (Table 1). Handedness of TDC was defined by the Edinburgh Handedness Inventory (mean absolute L.Q. = 79, Oldfield, 1971). Inclusion criteria for participants with unilateral CP included unilateral CP with a difference of at least 50% between the two limbs on the Jebsen-Taylor Test of Hand Function (without the writing subtest, Jebsen, Taylor, Trieschmann, Trotter, &

Results

Fig. 2 shows the averaged times over 15 days of practice for both groups. There were significant improvements over practice and significant differences between groups (practice: F(14, 546) = 24.87, p < 0.001, η2 = 0.39, group: F(1, 39) = 58.52, p < 0.001, η2 = 0.60). In addition, the practice effect had different impact on the two groups of children's task performance (practice × group: F(14, 546) = 10.05, p < 0.001, η2 = 0.21). Based on Post hoc analysis (Table 2), the participants with CP showed no significant

Discussion

Children with unilateral CP demonstrated potential to learn the current bimanual speed stack task in a relatively short period of time (15 practice sessions). However their improvement rate was not as large as the TDC (about 10% less improvement than TDC), and the learning pattern was different from TDC. For the current simple bimanual speed stacks task, TDC children took 1–3 days to achieve most of their improvement while children with unilateral CP took around 6–8 days before a plateau in

Conclusion

For therapy, sufficient practice is important for children with unilateral CP (two to three times more time than TDC) and delayed improvement is expected. No change or little improvement is expected early in practice for children with unilateral CP. More rapid improvement will follow subsequently. This finding can help set reasonable expectations from parents and children who want immediate improvement. In addition, the current findings of delayed improvement for children with unilateral CP may

Acknowledgement

We thank the children and their families for participating in this study.

References (35)

  • E. Domellof et al.

    Impairment severity selectively affects the control of proximal and distal components of reaching movements in children with hemiplegic cerebral palsy

    Developmental Medicine and Child Neurology

    (2009)
  • S.V. Duff et al.

    Learning of grasp control in children with hemiplegic cerebral palsy

    Developmental Medicine and Child Neurology

    (2003)
  • H.T. Goh et al.

    Movement pattern and parameter learning in children: Effects of feedback frequency

    Research Quarterly for Exercise and Sport

    (2012)
  • A.M. Gordon et al.

    Both constraint-induced movement therapy and bimanual training lead to improved performance of upper extremity function in children with hemiplegia

    Developmental Medicine and Child Neurology

    (2008)
  • A.M. Gordon et al.

    Fingertip forces during object manipulation in children with hemiplegic cerebral palsy I: Anticipatory scaling

    Developmental Medicine and Child Neurology

    (1999)
  • A.M. Gordon et al.

    Bimanual training and constraint-induced movement therapy in children with hemiplegic cerebral palsy: A randomized trial

    Neurorehabilitation and Neural Repair

    (2011)
  • A.M. Gordon et al.

    Efficacy of a hand-arm bimanual intensive therapy (HABIT) in children with hemiplegic cerebral palsy: A randomized control trial

    Developmental Medicine and Child Neurology

    (2007)
  • Cited by (15)

    • Motor learning in unilateral cerebral palsy and the influence of corticospinal tract reorganization

      2020, European Journal of Paediatric Neurology
      Citation Excerpt :

      By using different paradigms, previous literature has provided some evidence of impaired motor learning in CP, but without taking into consideration the role of corticospinal circuit plasticity. For instance, Hung and Gordon (2013) showed that children with unilateral CP are able to improve their performance in a bimanual speed task, but their rate of learning is lower than that of TD children [52]. Gagliardi and colleagues (2011) showed impaired sequence learning skills in children with CP (i.e. they required more trials to complete the task, making more spatial errors), as compared to TD controls; in line with our findings, such learning difficulties were unrelated to the severity of the hand motor disorder, measured with the Gross Motor Function Classification System [53].

    • Motor learning paradigm and contextual interference in manual computer tasks in individuals with cerebral palsy

      2017, Research in Developmental Disabilities
      Citation Excerpt :

      Better understanding of the specifics of motor learning of individuals with CP can help in attending to their learning needs and in maximizing learning efficiency (Hung & Gordon, 2013). Considering motor learning in CP, Hung and Gordon (2013) investigated the learning of 21 children with unilateral CP (4–10 years) and compared them with 21 age-matched typically developing (TD) children using a simple bimanual task. They verified that both groups demonstrated the ability to learn the bimanual task, however their rate of improvement and learning pattern differed.

    • Very prolonged practice in block of trials: Scaling of fitness, universality and persistence

      2015, Physica A: Statistical Mechanics and its Applications
      Citation Excerpt :

      Scientists have done many experiments throughout the years in order to investigate the learning process; the focus has been tasks on blocks with short time of execution divided into short blocks, punctuated with rest breaks. A non-extensive but representative list of examples is pressing buttons, ski-simulator [8,9], tracing geometric figures in a mirror [10], 1023 choice reaction time task [11], alphabet arithmetic task in the study of learning curve [12], memory search task in study of age-related learning differences [13], easy 4 choice reaction time task in the investigation of very prolonged practice [14], movement of a finger or pointer from one place to another in the study of Fitts’ Law [15], human stick balancing [16], speed stack bimanually with pyramid of cups in the study of motor learning in children with unilateral cerebral palsy [17], and reading text in the study of text reading fluency in adult readers [18]. Scientists have often performed trials in blocks with the objective of removing some presumed transient randomlike changes from trial to trial while emphasizing the persistent changes or the global trend of learning over the trials [19].

    • Transfer of motor learning from virtual to natural environments in individuals with cerebral palsy

      2014, Research in Developmental Disabilities
      Citation Excerpt :

      However, at present this is mostly speculation; the issue of transfer to different speeds clearly deserves additional experimentation. In summary, despite the motor difficulties in individuals with CP, it is clear that they can still improve motor performance—at least on the relatively simple tasks of the present study (see also Gofer-Levi, Silberg, Brezner, & Vakil, 2013; Hung & Gordon, 2013; Krebs et al., 2012; Robert, Guberec, Sveistrup, & Levin, 2013). This of course highlights the relevance of rehabilitation programs and research to find ways to optimize these programs as much as possible.

    View all citing articles on Scopus
    View full text