Feature

Clumsy Children

Why do some children have a harder time moving gracefully than others?

Dr Greg Anson

About 12% of New Zealanders are clumsy! That observation is a prediction based on numerous studies of clumsy behavior. Most of the studies have involved children and were completed in other countries.

Clumsiness is a phenomenon much easier to observe than it is to pin down with a scientific description. Some individuals appear to fall more frequently, collide with furniture more often, have greater difficulty tying shoelaces or organising books on a shelf, and often give the impression through their movements that some parts of their bodies are not necessarily aware of what other parts are doing. In school playgrounds, children who are clumsy are often left out of team games and activities, either because they choose not to participate or because they are not selected by their peers to join in the game.

But what causes clumsiness? Many specific conditions may result in clumsy behaviour, such as specific pathological conditions like Down's syndrome and Parkinson's disease; temporary disablement such as a broken limb or injured body part, or other transitory situations like puberty. In these examples the clumsy behaviour can be explained. In contrast, clumsiness associated with performance of motor activities in which no specific condition has been identified, has no apparent cause.

This puzzling observation intrigued Justine Higgott and me. In 1995, for her final year honours research project Justine carried out an experimental investigation to test one possible explanation of clumsiness in children. Specifically, we define clumsiness as "movement difficulties in the absence of neurological, physical and intellectual problems". The clumsy movement is certainly obvious, but the cause remains elusive.

The project had two main parts:

• assessment and testing of children to establish the presence or absence of clumsiness
• a reaction-time experiment to determine whether clumsiness was associated with problems of movement preparation and organisation

Movement preparation describes the work that the brain does before the actual movement begins -- for example, "thinking" about accelerating the body out of the blocks before the starter's gun actually sounds.

The children who participated in the study were drawn from two age-groups, 6-9-year-olds ("young") and 11-13-year-olds ("old"). Having two age groups allowed us to consider differences in performance that might be due to different stages of a child's development. Each age group contained two subgroups, one comprising clumsy children, the second consisting of children who were not clumsy. All children in each age group had similar reading ages. What has reading age got to do with this you may ask? We used reading age as an estimate of intellectual function -- to meet the criteria of "clumsy", children should not have intellectual problems.

Our experiment was conducted in the Neuromotor Control Project Laboratory in the beautiful new Physical Education building at the University of Otago in Dunedin. The laboratory is a large open room with computers and electronic equipment to accurately measure characteristics of human movement control such as accuracy, speed, timing and decision making.

Each child attended the project laboratory several times. During the first visit, Justine explained the testing procedure and asked the participant if they felt OK about continuing with the tests. Children were able to withdraw their participation at any time if they wished. The first test, the Basic Motor Abilities Test (BMAT), involved a series of activities that evaluated different dimensions of movement control. These included:

• manipulation (using the hands and fingers quickly to perform sequential movements)
• whole body control (balance and jumping, for example)
• large limb movement (throwing a bean bag at a target)

The results from the BMAT would tell us whether the original description of the child as "clumsy" was correct.

In this instance, the BMAT results confirmed the referral status of each child -- children in the clumsy group scored much lower (young 30%  and old 34.8%) than did those in the control group (54.4% and 59.8% for young and old respectively).

These results set the stage for the next part of the study, in which we tried to answer the following question: Why are clumsy children "clumsy"? A review of the scientific literature revealed that one often-cited view suggested that clumsiness results from difficulty in organising or preparing movements prior to their initiation. In the Neuromotor Control Laboratory we can examine preparation by measuring reaction time, or RT as it often called. RT is the time needed to make a decision about when to start a movement. Again, if we think about the sprinter in the blocks, the time between the sound of the starter's gun and the beginning of the movement is called RT.

Snap!

What happens during the RT interval? Imagine you are playing the card game Snap with a friend. Your friend places a seven of spades on the pile, then you place a seven of hearts! RT is the time between seeing the seven of hearts, recognising it as a "seven", perceiving that it forms a pair, selecting the correct response (slapping your hand on the pile and shouting "Snap!" at the same time) and initiating the movement of your hand toward the pile.

Even simple games like Snap involve decision making. The decision making requires the brain to perform many complex tasks. And just think, all this work happens in less than half a second, though for the brain 500 milliseconds is quite a long time to play with.

Deciding what movement to make involves choosing the correct parameters for the movement, such as its direction, distance, speed, and end-point. This kind of information processing is work that the brain does in order to provide humans with a phenomenal repertoire of movement capabilities, whether they be utilised in sport, recreation, video games or playing Snap with a friend. One feature of games like Snap is that success depends on how fast the information processing can be completed and the speed with which the movement can be initiated.

Justine and I wondered if clumsiness in children occurred when the processes used in movement preparation were altered or impaired in some way.

We used a special kind of RT task called a parameter pre-cuing task. It allows us to present participants with selected information about the movement that they will perform when the "go signal" occurs. For example, the pre-cue might indicate what direction to move but not how far to move -- thus direction is specified but distance is not; the participant won't know the exact distance until the go signal occurs. Therefore they can use the direction pre-cue to help prepare the movement but will have to organise the distance after the go signal. We used combinations of direction and distance information to provide different parameter pre-cue conditions in the experiment.

So, what did the children do in this experiment?

The child sat in a dental chair (for maximum comfort and postural support) and placed their arm on a padded arm rest. In their hand they grasped a handle much like a video game controller, attached to a pointer that could be moved to one of four targets. Two of the targets were to the left and two to the right of a centrally placed green light emitting diode (LED). The centre of each target was marked by a red LED and the borders by red adhesive tape. To begin a trial, the pointer had to be located at the central (green) LED and this signalled to the computer that the participant was ready.

Each trial began with the display (a red LED in the centre of each target) of the pre-cue parameters. When the pre-cue consisted of a single red LED, it provided the participant with all the necessary information (direction and distance) except when to start. The go signal was a loud tone that occurred, on average, 500 ms following the 5-second pre-cue display. From an earlier study with adult participants we knew that RT would be shortest (fastest) when the pre-cue contained both direction and distance information.

In addition to the single red LED pre-cue condition, there were four other conditions in which the pre-cue provided incomplete information. This meant that the participant couldn't organize and prepare the whole response until the go signal actually occurred. For example, if the pre-cue was two red LEDs left of centre, then direction was specified but distance remained unknown (could be short or long). Likewise, if the red LEDs on the far left and far right of centre formed the pre-cue, distance would be specified but not direction. When all four LEDs were lit, the participant was unable to select specific direction and distance information until the "GO" signal indicating the exact target occurred. Use of the parameter pre-cue method allowed us to assess the effect of different kinds and amounts of information on the speed of movement initiation as measured by RT.

Each child took part in 600 trials; each of the five pre-cue conditions was presented 120 times (30 at each target). Trials were grouped in blocks of 100. Each block contained an equal number of trials for each pre-cue condition. The distribution of conditions within a block was pseudo random so that participants could not reliably anticipate the next condition or target location. Each child completed two blocks of 100 trials in one experimental session and returned at approximately weekly intervals to complete the remaining blocks. The instructions required that each trial be performed as quickly and accurately as possible.

After all this hard work by the children, we were keen to look at the results. We wanted to get a picture of the effect of age (young and old) and movement ability (not clumsy, clumsy) on RT for each of the five parameter pre-cue conditions. To do this we calculated the average RT for each group and each pre-cue condition.

Several important findings were observed. First, the pattern of results for each group was similar. Second, the older, non-clumsy group had the fastest RTs and the young, clumsy children had the slowest RTs. We can see that RT is influenced by both age and movement ability. When both direction and distance information is specified, RT is the fastest for each group; and knowing direction but not distance leads to faster RT than knowing distance but not direction. As predicted, when all targets are lit for the pre-cue (the direction and distance is not specified by the pre-cue) RT is slowest for all groups.

The experiment that Justine and I completed provided results indicating that clumsy children (both young and old) were slower than non-clumsy children. However the results also indicate that clumsy children are able to organise and prepare movement information in the same way that non-clumsy children do, but they require more time to process the information.

At this time it is not known whether this effect is persistent beyond the age groups we looked at, or whether it might be altered, for example, by having clumsy children participate in tasks that facilitate speeded information processing. This would be an interesting question to investigate and the results may tell us more about the mysteries that underpin the phenomenon known as clumsiness.

Dr Greg Anson is a Senior Lecturer in Kinesiology in the School of Physical Education and the Neuroscience Research Centre, University of Otago.