THE EFFECT OF BLOCKED VERSUS RANDOM TRAINING SCHEDULES ON BOCCIA SKILLS PERFORMANCE IN EXPERIENCED ATHLETES WITH CEREBRAL PALSY

Efficiently maximizing sporting performances through structuring of training sessions has become a very important focus over the past twenty years. Very little study directly related to boccia exists for athletes with cerebral palsy (CP) and consequently information on performance measures is scarce. The aim of study was to measure the effect of blocked versus random training schedules on boccia skills performance. Fourteen experienced national-level athletes with CP (M = 23.6 years) participated in this study. Athletes were divided in two intervention groups and followed prescribed blocked (n = seven athletes) and random (n = seven athletes) training schedules for 10 weeks. Data collection included preand post-tests original testing procedures. Two paired-samples Wilcoxon Signed Rank tests and a Mann-Whitney U test only generated statistically significant evidence to support blocked training (z = 2.29, p = 0.02, CI.90 = 9.50, 22.50). Despite this, findings and individual cases of improvements plus athletes and coaches remarks indicated a practical significance towards blocked training schedules over random training schedules in terms of improving boccia skills performance. Greater gains across all three boccia skills measured and in total mean difference between preand post-test were shown in favor of athletes following blocked training schedules.


INTRODUCTION
Organized sports for persons with cerebral palsy (CP) were slower to both gain attention and develop when compared to other disability sports.However, individuals with CP have shown to be able to compete and the number of athletes with CP worldwide grows and continues to grow (Jones, 1988;Čurdová, 2005).Furthermore, over the past twenty years, substantial increases in the levels of government and private support have seen numerous, consistent and improved performances of many athletes and teams (Doll-Tepper, 1999;DePauw & Gavron, 2005).Consequently, maximizing sporting performances in the most efficient way has become a very important focus for many athletes and coaches.
A large amount of study has been done on how motor skill acquisition and development is affected by the organization of practice schedules (Magill & Hall, 1990;Shea & Morgan, 1979in Hynes-Dusel, 2002).In particular, focus has been on how practice or training sessions should be structured to result in the greatest and most efficient improvement.It is generally accepted that the total amount of practice is the most important factor to consider (Schmidt & Lee, 1999in Choi, Qi, Gordon & Schweighofer, 2008).In addition, literature suggests practice should be randomized, which degrades performance during practice but improves performance over time in retention tests when compared to blocked practice (Schmidt, 1991;Hanlon, 1996).
However, the question remains if these findings for training and motor skill development also apply for athletes with disabilities, in particular individuals with CP.Currently, very few studies directly relating to CP and boccia exists (Marta, 1998) and as a result, information relating to measures of performance is limited.Boccia is a game designed for people for CP and is a cross between petanque and indoor bowls that can be played as individuals, and in pairs or teams.The aim of boccia is to propel (by throwing, kicking or using a ramp device with use of an assistant) your colored boccia balls closer to the white jack than your opponents.For competition purposes, players are classified into one of the boccia sport-specific classes BC1, BC2, BC3 or BC4 (CPISRA, 2010).
The aim of the current study was to identify the effect of blocked versus random training schedules on boccia skills performance in persons with CP by measuring (1) jack and first ball placement, (2) drawing shots, and (3) driving shots skill performances.The jack and first ball refers to the first shot of the game, while drawing and driving refer to throwing to score points and throwing to hit and move boccia balls respectively.The current study will, in practice, provide boccia players and coaches with further explanations relating to optimal training schedules and implications for practice, improvement and performance.
Consequently, the following research questions were derived: (1) whether blocked training schedules improve boccia skills performance, (2) whether random training schedules improve boccia skills performance, and (3) whether there is a difference between blocked and random training schedules in relation to improving boccia skills performance.

Participants
Fourteen experienced national-level athletes with CP, comprised of nine males and five females, participated in this study, ages ranging from 17 to 40 years (M = 23.6 years; SD ± 6.9 years).The athletes were based in two major sports clubs in the Czech Republic: Kociánka Brno, o.s (SKK Brno) and Tukani APA VČAS Olomouc, as well as one participant from the ParaFed Canterbury boccia club in Christchurch, New Zealand.
The inclusion criteria included the provision of informed consent coupled with the mostrecent classification as determined by the CPISRA (2010) classification rules for classes BC1, BC2, and BC3.

Measurements and study design
To determine the effect of blocked and random training schedules on boccia performance, original testing and intervention procedures were used as there has been little previous research into the area of CP and boccia performance measures to date.This testing procedure was developed and trailed within the New Zealand boccia team over the past seven years and has been beneficial in skill improvement, tracking progress and evaluating training.The testing procedure measures three components: (1) jack and first ball placement, (2) drawing shots, and (3) driving shots, and requires participants to propel 84 boccia balls for a combined total possible maximum score of 252.The reason for propelling 84 boccia balls are so that scores aren't based on chance and fatigue is accounted for as consistency is needed to obtain a high skill performance score.A twogroup (blocked and random) pre-test and post-test designed study (Portney & Watkins, 2009) was used that follows CPISRA (2010) international boccia rules at all times during both testing and intervention.The two boccia skills performance assessments (pre-and post-tests) and intervention training sessions were done over a period of 10 weeks.The reason for using the pre-and post-tests was to measure if there is an actual difference, if any, between blocked and random training schedules both before and after the training intervention programs in relation to boccia skills performance.An initial pre-test was done to determine a starting boccia skills test score for each athlete.Half of the total number of athletes, both within a classification grade and overall, were then randomly assigned to one of the experimental training schedule groups to follow the respective intervention training programs.This was followed by a post-test using the same matched testing protocol to determine a final performance score and data for analysis purposes.

Testing protocol
Athletes were given an outline of the testing procedure prior to commencing testing.During testing athletes were to sit inside one of the two central playing boxes.The authors were present and conducted the (pre-and post-) testing procedure with assistance of the club coaches (all with international and Paralympic Games boccia experience) at their respective clubs.The same boccia balls, equipment (assistive devices), venues and testing protocol were used in both the pre-and post-test for each athlete.
(1) The first part of the test measured jack ball placement and the first ball shot.To score, athletes were required to propel the jack anywhere within the scoring dimensions of the court.If the jack failed to meet this criterion then no points could be scored on this attempt.Once the jack ball had been achieved, athletes were to follow this shot up by propelling one colored boccia ball as close to the jack as possible.If the colored boccia ball ended up within one ball width (9.0cm)three points were scored; within two ball widths (18.0cm) -two points were scored; within three ball widths (27.0cm) -one point was scored; and anything outside of this range scored zero.After these two shots had been thrown, the boccia balls were retrieved and returned to the athlete to be repeated twelve times with the chance to score a maximum 36 points.
(2) The second part of the test measured the ability to draw onto the jack ball at three different set lengths of three, five and nine meters.The jack was initially placed directly in front of the used playing box at three meters, while each athlete had one colored boccia ball with which they were to propel as close as possible to the jack.The same threepoint scoring system was used.This was repeated twelve times with a possible maximum score of thirty six, and then the jack was moved back to five meters, followed by nine meters respectively, and the same procedure repeated.The total maximum score possible for the drawing section of the test was 108.
(3) The third and final part of the test measured the ability to drive the jack away at three different set lengths of three, five and nine meters.The jack was again placed directly in front of the used playing box at three meters, while the athlete had one colored boccia ball with which they were to propel and hit the jack off the original mark as far as possible.This time, if the jack is moved three or more ball widths (27.0cm) from the original position -three points were scored; moved two ball widths (18.0cm) -two points were scored; moved one ball width (9.0cm)one point was scored; and no contact or moving the ball less than 9.0cm scored zero.This was repeated twelve times with a possible maximum score of thirty six, and then the jack was moved back to five meters, followed by nine meters respectively, and the same procedure repeated.The total maximum score possible for the driving section of the test is 108.
A modified ruler with three distance markings of 9.0cm, 18.0cm and 27.0cm was used to determine the score for each shot measured during testing.Measurements were always taken from the closest outermost edge of the target jack boccia ball to the furthest outermost edge of the propelled colour boccia ball.

Intervention
All athletes received a prescribed blocked or random training intervention schedule to follow for 10-weeks with training sessions lasting 90 minutes each twice a week.Every training session consisted of: (1) jack and first ball practice, (2) drawing practice, and (3) driving practice activities.All were explained in simple text with pictures to follow.Athletes following the prescribed blocked training schedule repeatedly performed each individual skill: (1) jack and first ball placement, (2) drawing shots, or (3) driving shots, for 10 minutes at each length of three meters, five meters and nine meters respectively before changing skill.Conversely, athletes following the random training schedule performed each individual skill: (1) jack and first ball placement, (2) drawing shots, or (3) driving shots, for the same time of 30 minutes per skill, but instead varied the length for each attempt before changing skill (Schmidt & Wrisberg, 2008).
The training schedules and activities were designed and modified based on the authors' international and Paralympic Games boccia coaching experience.Prior to the first training intervention sessions, the authors held oneon-one workshop sessions with the respective club coaches explaining the aims and reasons of the study and why the training programs have been designed.The session also included specific step-by-step instructions and practical guides on how to run the respective training activities.In addition, the participating clubs were each given personal printed and electronic copies of the blocked and random training schedules along with the athletes' names and their respective schedules to follow.The first training sessions with the athletes were jointly-run by the authors and coaches to demonstrate and explain the training procedures to the athletes.The authors continued to coach and monitor eight athletes who were based in Olomouc, whereas the club coaches coached the Brno and Canterbury based athletes respectively.Weekly email and phone contact was maintained with the Brno and Canterbury club coaches to monitor the prescribed twicea-week training sessions for athletes.Furthermore, the authors visited the Brno club every third week (a total of five visits including pre-and post-tests) to personally check the prescribed training schedules.

Data analysis
Two separate paired-samples Wilcoxon Signed Rank Tests were used (one for blocked and one for random training) to determine whether blocked and random training schedules respectively improved boccia skills performance between pre-and post-tests.A two-sample Mann-Whitney U test was used to determine whether there was a difference between blocked and random training schedules in terms of an increase in individual and combined boccia skills performance.The non-parametric tests were used due to the small sample sizes.

RESULTS
Table 1 presents the details of classification grade, age, gender, training schedule, sum results for pre-and post-test scores for each of the athletes, and the mean scores and SDs for blocked and random training schedules respectively.It can be seen that all athletes, except three (athletes 7, 11 and 12), improved in overall boccia skills performance following intervention.The most relevant statistics for analysis purposes are the means reflecting the pre-and post-test scores for blocked and random training schedules.Examination between these scores show the difference in mean sum results of the blocked training schedule group (M= 15.43, SD = 4.77) was greater than the difference in mean sum results of the random training schedule group (M= 4.43, SD = 1.93).With 90% confidence, the median for total scores following blocked training is somewhere between 9.50 and 22.50 points higher than the median total score prior to blocked training.A 90% confidence interval was used as the authors were unable to assume that the score differences came from a normal distribution and that the sample size was too small to construct a 95% confidence interval.
The sum breakdown of boccia skills results for pre-and post-test scores for each of the athletes relative to training schedule are shown in Table 2. Mean differences between pre-and post-test scores for the blocked training schedule group compared to the random training schedule group were: improvements of 5.57 versus 1.71 for jack and first ball; improvements of 5.43 versus 3.86 for drawing skills; and improvement of 3.86 versus a decline of 1.14 for driving skills.
Table 3 presents the results of two pairedsamples Wilcoxon Signed Rank tests, comparing the differences for blocked and random training schedules pre-and post-tests scores respectively.Participants undergoing blocked training improved significantly overall (z = 2.29, p = 0.02, CI .90= 9.50, 22.50).
The results of a 2-sample Mann-Whitney U test (p < 0.05) comparing the differences in the three boccia skills tested: (1) jack and first ball, (2) drawing, and (3) driving, as well as the total score between pre-and post-tests between blocked and random training schedules are shown in Table 4.No statistical differences were demonstrated between blocked and random training schedules for any of the three boccia skills tested, nor for the overall score.

DISCUSSION
Although differences between the effects of blocked versus random training schedules from the current study were not statistically significant, the greater gain in mean sum results in boccia skills performance were shown by athletes using the blocked training schedule versus athletes using the random training schedule indicating a practical significance.Furthermore, the three athletes (7, 11 and 12) who did not improve on their pre-test scores each followed the random training schedule; while the greatest and worst improvements (athletes 5 and 12) followed blocked and random training schedules respectively (see Table 1).
It is evident that the blocked training schedule group, on average, showed greater gains across all three boccia skills measured when compared to the random training schedule group (see Table 2).It is also important to note that the random training group on average declined in the driving category indicating that random training is detrimental to the performance of this boccia skill.This suggests that athletes following the random training schedule found it difficult to adjust to the changing conditions (length) of each shot, as opposed to the repetition of the blocked training schedule group.These findings challenge the beliefs (Schmidt, 1991;Hanlon, 1996) that random training is more beneficial to improvement and learning than blocked training.However current findings are consistent with Giuffrida, Shea & Fairbrother (2002) who showed that blocked training was most beneficial when repeatedly learning a task in the same way it is performed.Results from the current study reinforce the idea that individuals with specific neurologic impairments and disabilities, in this case CP, require blocked training to learn more effectively as they are unable to process too much information at once (Lin, Sullivan, Wu, Kantak & Winstein, 2007), such as choosing the parameters for the type of shot needed between a blocked and random training schedule.
The current study used two training schedule interventions that proved somewhat effective in improving boccia skills performance in athletes with CP.While the p value for the blocked training schedule total between pre-and post-tests for the intervention was statistically significant (see Table 3); no statistical significance was observed for the random training schedule (see Table 3), nor the difference between blocked and random training schedules (see Table 4) in relation to improving boccia skills performance.The comparison between preand post-test scores for athletes following the blocked training schedule shows evidence to suggest that the median total score after blocked training is higher than the median total score before training.
Despite findings that the blocked training schedule improves the overall boccia skills performance score, the authors were unable to specify which area of the athletes' game improved, as the authors were unable to find significant differences for any of the three individual skill areas between the pre-and post-tests.A possible explanation for this is that two or three of the three skills: (1) jack and first ball, (2) drawing and (3) driving improved, but not enough for a significant difference between pre-and post-test scores for these skills to be found.However, when the differences were added together for a total score, the accumulative improvement in the three skill areas leads to a difference between the pre-and post-tests that was large enough to find a significant effect for blocked training.With a larger sample size, it is likely that the authors would be able to establish which particular area(s) improved following blocked training.However, there was no evidence to suggest a significant difference between median total scores before and after random training, nor was there any significant evidence that the median total score after random training was the same as the median total score after blocked training.Lack of statistical significance may be due to a small sample size.
This study only assessed three skills: (1) jack and first ball, (2) drawing, and (3) driving in boccia performance in experienced national level athletes with CP.Other findings can also be counted in from the current research in terms of boccia skills: it is important to emphasize training of boccia skills played at longer distances and playing in the farthest part of the boccia court as the measurements taken for longer distances were lower compared to shorter distances; males and females are capable of obtaining similar scores in boccia skills performances; better measurement results (test scores) were obtained in the pre-and post-tests for BC3 players using assistive devices when compared to BC2 throwing players.These are additional areas that require more attention and should be further examined in the future.
The first and most important limitation was that this study only involved fourteen athletes.Consequently, a small sample size makes it more difficult to produce findings of statistical significance as statistical tests normally require a larger sample size to justify that the effect did not just happened by chance alone.Provided that the current findings were replicated with a larger sample size, results may have been found statistically significant.Generalization of the current findings should therefore be undertaken with caution because of the relatively small sample size and restricted number of retention tests.Consequently, more well-controlled studies using longitudinal data, larger sample sizes, more tests and an additional experimental control group would be helpful to track the changes in boccia performance over time relative to both training schedules.This would also help remove the variability of results due to the nature and degree of CP.
Secondly, this study only involved experienced athletes.Findings from Painter, Inman and Vincent (1994) showed that it is more difficult to detect changes in learning and performance in subjects with a mild intellectual disability who were already somewhat competent and skilled with a set motor task.Therefore, this test should be repeated with a larger population of differing skill levels and competition, including a sample size of equal numbers for males and females, and classification groups, to ensure suitable data for statistical analysis between beginner, advanced and expert players and types of training.
Thirdly, due to practicality issues such as travel and costs within the limited amount of time available to execute this study, the intervention training periods were not able to be run for long enough to yield more reliable data, hence the statistical insignificance in the results.In addition, this research was only conducted in two sports clubs in the Czech Republic.
Furthermore, as boccia is a minority sport, there are a limited number of coaches and support assistants which affects the amount of time that athletes can spend practicing.Therefore, in order to achieve a stronger overall understanding of the effect of blocked versus random training schedules on boccia performance in persons with CP, it would be necessary to complete similar research in a number of different clubs and athletes across the country.It would also be worthwhile to conduct a cross-cultural research by testing players from different countries to gain a better understanding on boccia skills performances worldwide.
Previous literature and findings have suggested that blocked training provide better learning results, especially during the skill acquisition phase (Magill & Hall, 1990;Shea & Morgan, 1979in Hynes-Dusel, 2002).Traditionally, teachers and coaches begin teaching motor skill performance with an aim on improving technique through the use of a blocked training schedule so the beginning learner can gain proficiency in the skill itself before they are forced to apply this into a more challenging game-like situation.This is generally followed by a more random training schedule once the learner has become competent and confident in their performance of the motor skills.Therefore, the application of advanced skills and strategy may become more of a focus, suggesting that there is a place for both blocked and random training schedules in sport (Schmidt & Wrisberg, 2008) Maximizing sporting performances in the most efficient way has become an increasingly important focus throughout the world.It is therefore necessary to have knowledge on how to structure training sessions to improve the consistency of skill levels and performances in athletes.Although there has been research into the areas of training schedules and types of practice, there has been little previous research directly related to skill techniques or performance measures in boccia in athletes with CP.To conclude, findings from the current study provide an insight into training and motor skill development and how these apply to athletes with CP in terms of skill performances in boccia.It also addresses some of the gaps in the literature concerning disability sports.

Table 1
Athlete Characteristics, Intervention Training Schedule and Results from Pre-and Post-Tests

Table 2
Athlete Intervention Schedule and Boccia Skills Results from Pre-and Post-Tests

Table 3
Results of Paired-Samples Wilcoxon Signed Rank Test for Differences between Respective Blocked and Random Training Schedules Pre-and Post-Tests Scores

Table 4
Results of 2-Samples Mann-Whitney U Test for Differences between Blocked and Random Training Schedule Pre-and Post-Tests Scores