NOTE TO USERS This reproduction is the best copy available. UMI Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. AN INVESTIGATION INTO THE ATTENTION PROFILE OF BOYS WITH AND WITHOUT DEVELOPMENTAL COORDINATION DISORDER A Thesis presented to the School of Kinesiology Lakehead University Submitted in partial fulfillment of the degree of Masters of Science in Kinesiology By; Laura Anne Sheehan Supervisor: Dr. Jane Taylor April 2009 ©2009 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Library and Bibliothèque et Archives Canada Archives Canada Published Heritage Direction du Branch Patrimoine de l'édition 395 Wellington Street 395, rue Wellington Ottawa ON K1A0N4 Ottawa ON K1A0N4 Canada Canada Your file Votre référence ISBN: 978-0-494-49958-0 Our file Notre référence ISBN: 978-0-494-49958-0 NOTICE: AVIS: The author has granted a non­ L'auteur a accordé une licence non exclusive exclusive license allowing Library permettant à la Bibliothèque et Archives and Archives Canada to reproduce, Canada de reproduire, publier, archiver, publish, archive, preserve, conserve, sauvegarder, conserver, transmettre au public communicate to the public by par télécommunication ou par l'Internet, prêter, telecommunication or on the Internet, distribuer et vendre des thèses partout dans loan, distribute and sell theses le monde, à des fins commerciales ou autres, worldwide, for commercial or non­ sur support microforme, papier, électronique commercial purposes, in microform, et/ou autres formats. paper, electronic and/or any other formats. The author retains copyright L'auteur conserve la propriété du droit d'auteur ownership and moral rights in et des droits moraux qui protège cette thèse. this thesis. Neither the thesis Ni la thèse ni des extraits substantiels de nor substantial extracts from it celle-ci ne doivent être imprimés ou autrement may be printed or otherwise reproduits sans son autorisation. reproduced without the author's permission. In compliance with the Canadian Conformément à la loi canadienne Privacy Act some supporting sur la protection de la vie privée, forms may have been removed quelques formulaires secondaires from this thesis. ont été enlevés de cette thèse. While these forms may be included Bien que ces formulaires in the document page count, aient inclus dans la pagination, their removal does not represent il n'y aura aucun contenu manquant, any loss of content from the thesis. 1^1 Canada Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Abstract Attention problems have been identified as an associated problem in children with Developmental Coordination Disorder (DCD), using the Child Behaviour Checklist, and the Covert Orienting of Visuospatial Attention Task. Their patterns indicate a deficit in the voluntary disengagement of attention, while reflexive orienting seems to be unaffected. Recently, attention has been investigated using the Attention Network Test (Fan, McCandliss, Sommer, Raz, & Posner, 2002) which measures the efficiency of orienting, alerting and executive control networks. As no research presently exists, the goal was to examine attention networks in boys with and without DCD using the Attention Network Test for Children (ANT-C; Rueda et al., 2004). Twenty-five boys between 7 and 10 years were recruited to participate in the study. Each participant was screened using the MABC, and then tested on the ANT-C. Fourteen boys with a mean age of 9 years comprised the DCD group (MABC percentile mean = 5.5), while eleven boys with a mean age of 8.6 years comprised the comparison group (MABC percentile mean = 51.1). A series of independent sample t-tests revealed the boys with DCD were not significantly different from the comparison group on the alerting (t(23)=-0.44, p=0.61, d=-0.18); orienting (f(23)= -1.39, p = 0.18, d= -0.55); or executive control (f(Z3)=-0.68, p=0.51, d=0.28) networks. In addition, the two groups were similar on error rates (t(2S)=0.94, p=0.36) and overall reaction time (t(23)=0.61,p=0.55). In contrast, using bivariate correlations, relationships were found between the alerting network and both the orienting (r = 0.70, p = 0.02) and executive control (r = 0.64, p = 0.04) networks in the group without DCD. In the group with DCD, these relationships were not observed. The presence or lack of relationship between networks suggests the two groups use differing strategies to achieve similar efficiency scores. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Based on the previous literature, which described attention difficulties in children with DCD, the results of the study were unexpected. High variability within and between each group, demonstrated by individual profiles and standard deviation, may have had a strong effect on the outcome of the inferential statistics. In addition, the validity of the ANT-C in regard to previous studies has also been questioned. It appears the ANT-C is not measuring the same aspects of attention that have been determined problematic in previous studies. Therefore, the attention profiles of boys with and without DCD remain similar on network efficiency, error rate and median reaction time, but may differ on the strategies used to achieve them. 11 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Acknowledgements I can still remember my first day of Graduate Orientation like it was yesterday. I was sitting in Jane’s office balling my eyes out, so overwhelmed with the choice I had made to return to Thunder Bay to complete this Master’s. All I saw ahead of me was a mountain ten times the elevation of the Sleeping Giant and I had no idea where to start climbing. Well here I am two and a half years later, standing on the top! I have persevered around marshes, over boulders and across rivers, taking two steps forward and one step back along this journey for knowledge and unexpected self growth. I would like to start by thanking my participants and their parents. I appreciate the time you gave up from your busy schedules to take part in my study. Without you this project could not have been completed. To those who helped with the recruitment process, we weren’t always successful, but in the end we triumphed. I appreciate the time you put forth filling out questionnaires, contacting family and friends, handing out letters, and hanging posters. Next, the entire kinesiology department including faculty, staff and students deserves a big thank you for not giving up on me and sticking with me right until the end. Thank you to those who were there to teach, to those who were there for guidance, and to those who were there to have fun. A special thank you must go to my committee members, Jim and Eryk. Jim, you had a great idea that helped put this project in motion. Eryk, as a student yourself, you always understood and provided advice. And to the both of you, thank you for always listening to my questions and directing me to my answers. To Jane, my advisor, I do not think I have enough words to thank you for everything you have done for me. You offered me this great challenge, held my hand when I needed comfort, but iii Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. gave me a kick in the butt when I needed to grow up. I will forever be in your debt for all the lessons I have learned. Finally and most importantly, I would like to thank my family and friends who initially pushed me to accept this challenge; they were always there with their unconditional love and support; and they will be with me to celebrate this great achievement and all those to come. IV Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table of Contents Introduction 1 Statement of Problem 2 Definition 3 Review of Literature 5 Developmental Coordination Disorder 5 Assessment of Developmental Coordination Disorder 11 Attention 14 Assessment of Attention 18 Developmental Coordination Disorder and Attention 20 Method 24 Procedure 24 Recruitment 24 Screening 25 Participants 25 Testing 26 Preliminary Evaluation 26 Analysis 27 Interpretation of the ANT-C Scores 29 Hypothesis 29 Results 31 Group Characteristics 31 Alerting Network 33 Orienting Network 35 Executive Control Network 37 Error Rates 39 Median Reaction Time 42 Correlations 42 Discussion 44 Alerting Network 44 Orienting Network 46 Executive Control Network 49 Overall Error Rates 51 Median Reaction Time 51 Correlations 52 Summary 54 Limitations and Recommendations 56 References 58 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendices 64 Appendix A - Principal Cover Letter and Consent Form 64 Appendix B - Teacher Cover Letter and Consent Form 69 Appendix C - Parent and Participant Cover Letter and Consent Form 73 Appendix D - Motor Behaviour Checklist 76 Appendix E - Poster 79 Appendix F - Newsletter 81 Appendix G - Handout 83 Appendix H - Movement Assessment Battery for Children Protocol 85 Appendix I - Adapted Movement Assessment Battery for Children Checklist 88 Appendix J - Development of the Attention Network Test for Children 91 Appendix K - Attention Network Test for Children Protocol 93 Appendix L - North American Federation of Adapted Physical Activity 2008 Poster 96 Appendix M - Participants Performance on the Movement Abilities Battery for Children 98 Appendix N - Individual Scores on the Attention Network Test for Children 100 VI Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. List of Figures Figure: 1. Warning conditions of the Attention Network Test for Children. 18 2. Target conditions of the Attention Network Test for Children. 19 3. Sequence of events of each trial of the Attention Network Test for Children 20 4. Network efficiency calculations. 28 5. Comparison of group means on the Total Impairment Score (TIS) on the Movement Assessment Battery for Children. 32 6. Comparison of group mean percentile scores on the Movement Assessment Battery for Children. 32 7. Individual alerting network efficiency scores 34 8. Comparison of group means on median reaction time in the no cue and double cue conditions. 34 9. Comparison of Group means and standard deviations on alerting scores. 35 10. Individual orienting network efficiency scores 36 11. Comparison of group means on median reaction time in the center cue and spatial cue conditions. 36 12. Comparison of group means and standard deviations on orienting scores. 37 13. Individual executive control efficiency network scores 38 14. Comparison of group means on median reaction time in the incongruent and congruent conditions. 38 15. Comparison of group means and standard deviations on executive control scores. 39 16. Comparison of error rates for target type illustrated by group mean. 40 17. Comparison of group mean error rates for warning cue conditions. 41 18. Overall error rates for boys with and without DCD. 41 19. Overall median reaction times for boys with and without DCD. 42 20. Individual alerting, orienting and executive control scores. 102 21. Individual error rate by target type. 102 22. Individual error rate by warning cue condition. 103 23. Individual total error rate. 103 24. Individual overall median reaction time (msec). 104 Vll Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. List of Tables Table: 1. Estimates of network efficiency (msec) by age 17 2. Individual characteristics including age and Movement Assessment Battery for Children Scores 99 3. Individual alerting, orienting and executive control scores 101 Vlll Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Introduction In 1994, the term Developmental Coordination Disorder (DCD) was adopted at the International Consensus Meeting on Children and Clumsiness to describe children with motor difficulties (Polatajko, Fox & Missiuna, 1995). DCD is a motor-based performance problem that limits a child’s ability to fully participate in the everyday activities of childhood. It is estimated that 6% of children between the ages of 5 and 11 years have DCD. The major characteristics of DCD are difficulties mastering either gross or fine motor tasks or both, generalizing learned movements to other tasks, and organizing and coordinating movements to accomplish a specific task, in comparison to children of the same age (Polatajko & Cantin, 2005; American Psychiatric Association, 1994, p.54). The current diagnostic criteria for DCD requires a child’s performance in daily motor coordination activities to be substantially below that expected given the child’s chronological age and measured intelligence. The disturbance must significantly interfere with academic achievement or activities of daily living, and is not due to another medical condition. If mental retardation is present, the motor difficulties must be in excess of those usually associated with it (American Psychiatric Association, 1994, pp.53-54). There are a variety of factors that have been identified in an attempt to determine the cause of DCD, however it is still largely unknown (Cermak & Larkin, 2002, p. 16). One contributing factor that has been identified is a deficit in visual-spatial processing (Wilson & McKenzie, 1998). The results of this meta-analysis have led to an investigation into the orienting attention network of children with DCD with attention cueing tasks. The orienting network is responsible for directing attention from an unattended location to a target location or object (Posner & Badgaiyan, 1998, pg. 62). Orienting may be overt, with eye movements, or covert, without any eye movement (Posner & Rothbart, 2007). It can also be reflexive or voluntary Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (Posner, 1980). Children with DCD have been identified with a deficit in the voluntary disengagement of attention, while reflexive orienting is not implicated in the disorder (Wilson & Maruff, 1999). In addition to the orienting network, there are two other attention networks, the alerting and the executive control. The alerting network is responsible for achieving and maintaining a vigilant state to incoming stimuli (Fan, McCandliss, Sommer, Raz & Posner, 2002). The executive control network is responsible for the more complex mental operations engaged during monitoring and resolving conflict among stimuli (Fan et ah). Together these three networks form the organ system of attention with its own anatomy and circuitry (Fan & Posner, 2004). Attention problems have been identified as an associated problem in children with DCD, and these children have been identified as a group at risk for attention problems (Sugden & Chambers, 2005, pg. 14; Dewey, Kaplan, Crawford & Wilson, 2002). Other than the research completed on the orienting network, to date, there is no research that specifically explores all three attention networks. Therefore, it is of interest that the efficiency of the alerting, orienting and executive control networks be examined in children with DCD. Statement o f the Problem The purpose of the present study was to use the Attention Network Test for Children to investigate the attention profile of boys with and without DCD and determine if there was a difference between groups. The aspects explored within the profile included, the efficiency of the alerting, orienting and executive control networks, error rates, overall median reaction time, and the relationships between the networks (Rueda et al., 2004). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Definitions Attention Network Test for Children (ANT-C) is a measurement tool developed to examine individual differences in the efficiency of the attention networks of alerting, orienting and executive control using reaction times to various conditions of the test (Fan et ah, 2002; Posner & Rothbart, 2006). There are 12 conditions based on four warning cues, and three target types. The four warning cues are no cue, double cue, center cue and spatial cue, while the three target types include congruent, incongruent and neutral (Rueda et ah, 2004). Attention can be described as the mental ability to select stimuli, responses, memories and thoughts that are behaviourally relevant among a host of others in our environment (Raz, 2004; Sugden & Chambers, 2005). Alerting is the function characterized by the process of achieving and maintaining a state of high sensitivity to incoming stimuli (Raz, 2004). In general, a larger alerting score indicates difficulty in maintaining attentiveness without a warning cue (Fan & Posner, 2004). Developmental Coordination Disorder (DCD) is a neurodevelopmental (motor skill) disorder characterized by a marked impairment in the development of motor coordination abilities that significantly interferes with performance of daily activities and/or academic achievement. The difficulties observed are not consistent with the child’s intellectual abilities and are not caused by a pervasive developmental disorder or general medical conditions that could explain the coordination deficits (Polatajko & Cantin, 2005). Executive Control is the function which involves more complex mental operations engaged during monitoring and resolving conflict among stimuli. A greater executive control score generally indicates difficulty in resolving conflict (Fan et al., 2002; Fan & Posner, 2004). Movement Assessment Battery for Children (M-ABC) is a clinical and educational tool designed to identify and describe the strengths and weaknesses of motor function impairments in Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. children aged 4 to 12. The test provides objective quantitative data on three performance areas (manual dexterity, ball skills and balance) for a diagnosis. Children with Total Impairment Scores below the 5‘̂ percentile are considered to have DCD, while children between the 5th and 15th percentile are considered at risk (Henderson & Sugden, 1992, p. 108). Orienting is the function characterized by the process of aligning attention and selection of information from a source of sensory stimuli (Posner & Rothbart, 2007; Raz, 2004). The network directs attention from an attended location to a target location or object (Posner & Badgaiyan, 1998, p.62). Usually, a larger orienting score indicates difficulty disengaging from the center cue, where the target does not appear (Fan & Posner, 2004). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Review of Literature Developmental Coordination Disorder The concept o f Developmental Coordination Disorder is not new (Cermak & Larkin, 2002, p. 2). Over the past 100 years, DCD has been described as clumsiness, a motor/learning disability, a perceptuomotor dysfunction, and developmental dyspraxia (Polatajko & Cantin, 2005). In October 1994 at the International Consensus Meeting on Children and Clumsiness, the term ‘clumsy’ was rejected as a label for children with motor difficulties, and the term DCD, endorsed by the American Psychiatric Association, was adopted (Polatajko et al.,1995). DCD is a motor-based performance problem that limits a child’s ability to fully participate in the everyday activities of childhood, in comparison to children of the same age. A child with DCD may have difficulties mastering either gross or fine motor coordination tasks or both which may be apparent in locomotion, agility, manual dexterity, complex skills and/or balance (Sugden, 2006). In addition, children with DCD may have difficulty learning new movements, generalizing learned movements to other tasks, and organizing and coordinating their movements to accomplish a specific task. The motor performance of children with DCD is consistently slower, less accurate, less precise and more variable than that of their peers (Polatajko & Cantin, 2005). The American Psychiatric Association and the World Health Organization recognize DCD as a disorder, and provide varying, but similar diagnostic criteria. The main consensus for a diagnosis of DCD is that a performance in daily activities that requires motor coordination is substantially below what is expected given the child’s age and intelligence that significantly interferes with academic achievement or activities of daily living, but is not due to a general medical condition. However, if mental retardation is present the motor difficulties must be in excess of those usually associated with it (American Psychiatric Association, 1994, pg. 55). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 6 Children with DCD form a heterogeneous group and there are no typical cases (Cermak & Larkin, 2002, p.42). Numerous research studies have been completed using cluster or factor analysis to confirm the heterogeneity between cases (Visser, 2003; Sugden & Chambers, 2005). Henderson and Sugden (1992) suggest there are clear subgroups of children with DCD that exist with different patterns of performance (p. 121). The first group of children demonstrates poor gross motor skills in comparison to their fine motor skills, while the second group is often competent in tasks requiring large body movements but has difficulty with fine motor tasks. The final group of children shows an equal impairment in both gross and fine motor skills. In 1994, Hoare investigated the possibility that the movement difficulties associated with DCD might be divisible into subtypes. She tested her participants on six variables, which consisted of kinesthetic acuity, motor free visual perception, visual-motor integration, static balance, running and the Purdue Pegboard. The results of the cluster analysis demonstrated clear heterogeneity into five clusters. The first cluster was below average on dynamic balance and kinesthetic acuity. Clusters two and three were identified by visual perception competencies with poor kinesthetic acuity, and visual motor deficits, respectively. The fourth cluster had poor static balance and visual-motor functions, while cluster five had poor static and dynamic balance and manual dexterity. McNabb, Miller, and Polatajko (2001) later repeated Hoare’s study, and successfully replicated the five clusters. Dewey and Kaplan (1994), Wright and Sugden (1996), and Wilson, Kaplan, Crawford, Campbell and Dewey (2000) also investigated whether subtypes of developmental motor deficits could be identified. All three research groups chose different variables from Hoare and consequently found different subtypes. The variables used by Dewey and Kaplan were balance, bilateral coordination, upper limb coordination, transitive gestures, and motor sequencing. Again after a cluster analysis, they found four subtypes. Subtype number one demonstrated deficits in Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 7 motor sequencing, while subtype two had deficits in balance, coordination and gestural performance. The third subtype had severe deficits in all motor skills areas, and the fourth subtype showed no motor deficits when compared to the other groups. In contrast, Wright and Sugden, and Wilson et al. used a single test to divide their participants. The test of choice for Wright and Sugden was the Movement Abilities Test for Children (MABC). The cluster analysis produced four subtypes from the participant’s scores. Cluster one showed generalized low scores, but not severe in any. The second cluster had poor performance in dynamic environments, while the third cluster demonstrated generalized poor scores across motor tasks, particularly in dynamic environments. Finally, cluster four was identified by poor fine motor control, speed and dynamic balance. Wilson et al. (2000), on the other hand, chose to use a factor analysis of the scores on the Developmental Coordination Disorder Questionnaire (DCD-Q) to group their participants. Similar to Wright and Sugden (1996), Wilson et al. found four groups, which were separated by a deficit in fine motor skills, a deficit in gross motor skills, a deficit in ball skills and control during movement, or complex general motor problems. Although each study produced differing clusters, a sensorimotor subtype appeared in all of the studies regardless of the specific sensorimotor variables used in the study, whereas the presence of other subtypes depended on the inclusion and combination of the particular measures (Visser, 2003). The lack of consistency in the results may be attributed to the excess variation resulting from different source populations, differences in the choice of variables, and differences in statistical methods (MacNabb et al., 2001). Therefore, since subgroups exist, it is not surprising that a child with DCD would score differently on different tests or exhibit a different profile on subtests within a test (Wilson et al., 2000). Even with differing profiles, it is important to remember each child with DCD has the ability to learn motor skills, however he or she usually Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 8 requires more practice than children without DCD, and the quality of movement may be compromised (Cermak & Larkin, 2002, p.7). As previously mentioned by Henderson and Sugden (1992), children with DCD may or may not have difficulty with gross or fine motor skills. Gross motor skills are tasks that involve the larger muscle groups of the body, such as crawling, walking, jumping, catching, and kicking, while fine motor skills are tasks that involve the use of smaller muscle groups that produce more intricate movements such as, writing, cutting, tying shoe laces, buttoning clothing, and moving pieces in a board game (American Psychiatric Association, 1994). The gross motor skills of walking and catching have been of interest to researchers studying children with DCD. Woodruff, Bothwell-Myers, Tingley and Albert (2002) developed an index of walking performance to investigate the differences in gait patterns between children with DCD and controls. They were unable to find significant differences between the means of the toe off, single stance, toe off and stride length all expressed as a percentage of the gait cycle. However, they found children with DCD had a much larger variance around the means than the controls. These characteristics suggest an abnormality in the gait pattern of children with DCD. More specifically, an abnormality in the time and distance patterns of children with DCD. The catching skills of children with DCD are also significantly different from their peers. VanWaelvelde, DeWeerdt, DeCock and Smits-Engelsman (2004) examined children with DCD and matched younger typically developing children who were able to catch a similar number of balls, on a two handed catching task. The qualitative ball catching performance was significantly poorer in the children with DCD than their match. In the preparation phase of catching, the children with DCD showed less elbow flexion and held their hands in front of the body. In preparation for contact with the ball, there was less arm extension, and less flexion of the elbows to absorb force upon contact. They also had a greater number of grasping errors. On a one- Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. handed catching task completed by Deconick et al. (2006), boys with DCD were not found to have differences in timing of the grasp, but failed to achieve maximal hand opening and peak closing velocity as high as their age matched peers. Research examining the fine motor skills of children with DCD is focused on their drawing and handwriting skills. The drawing skills of children with DCD, as determined by Barnett and Henderson (2002) using the “Draw a Man” test, are characterized by irregular, poorly controlled lines. Shapes drawn were often incomplete, lines did not meet at junctions clearly, and shading was often inaccurate and variable. These characteristics suggest poor control of pencil pressure or force in children with DCD. The group was also poor at representing proportions, depicting features, and providing details in those features. In addition, they had a tendency to draw smaller objects on the page than their well coordinated age matched peers. Smits-Engelsman, Niemeijer and van Galen (2001) used the Flower Trail component of the MABC to compare children with DCD to published norms and matched controls. Similar to Barnett and Henderson’s research (1992), they found higher drawing errors. In contrast, there was no significant difference between groups with respect to pen pressure. Other characteristics of the poor writers identified were, less time to complete the task and a higher velocity with fewer velocity peaks. There was also no significance between the variables of average trajectory length, and number of times the pen was lifted from the page. In addition, there was a tendency for good writers to spend more time pausing above the paper prior to writing. Figures of the handwriting skills of children with DCD, by Sugden and Chambers (2005) illustrate the difficulties with letter formation, spacing and alignment (pp. 170-171). In addition, children with DCD write slower than their peers (Sugden & Chambers, pp. 170-171). Approximately, 6% of children between the ages of 5 and 11 years are estimated to have DCD (American Psychiatric Association, 1994, p.54). The onset of DCD is typically in the early Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 10 years of life when the child first attempts running, holding a knife and fork, buttoning clothes, or playing ball games (American Psychiatric Association, p.54), but is rarely diagnosed before the age of 5 years when the child first attends school. This may be because the child’s lack of coordination only becomes a problem when it results in failure to satisfy his or her particular environmental demands, and it is in school where the child’s inability to meet requirements becomes problematic (Sugden, 2006; Cermak & Larkin, 2002, p. 15). The disorder has varying, but significant impact throughout the life span. A small portion of children do appear to improve, but more often than not a child’s motor difficulties continue into adolescence and adulthood (Sugden). The long-term outcomes of DCD were examined by Geuze and Borger (1993). They retested children between the ages of 11 and 17 who were diagnosed as clumsy, 5 years prior. At least 50% of the participants showed persistent motor difficulties into adolescence. Similarly, Losse et al. (1991) performed a ten year follow up with children who were determined to have motor coordination difficulties when they were 6 years old. At 16 yeas of age, the majority of the young adults continued to have difficulties with motor coordination. In 2003, Cousins and Smyth gathered adults diagnosed or self-identified with a history of motor impairments and tested them on manual dexterity, handwriting, construction, obstacle avoidance, dynamic balance, static balance, dual task performance, ball skills, reaction time, movement time and sequencing. The gross motor skills provided the participants with greater difficulty than the fine motor skills, but on a whole the adults were found to retain their motor difficulties. Fitzpatrick and Watkinson (2003) explored the retrospective views of adults experiencing life with physical awkwardness by conducting an interview. All participants had similar experiences growing up, which began with a breakdown in the execution of a sport skill followed Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 11 by self-evaluation and humiliation from reactions of others, expression of the consequences of the failure, and attempts to evade any further exposure to awkwardness in the future. On a positive note, these participants were able to come to terms with their physical awkwardness and each participant currently participates in physical activity to some extent. Assessment o f Developmental Coordination Disorder. In the process of identifying children with DCD, there are two distinct phases, including screening and evaluation (Taylor, 2006). There are a number of tests designed to identify and others to assess motor impairment or dysfunction in children (Dewey & Wilson, 2001). The Motor Behaviour Checklist (MBC) is a quick checklist devised to aid parents, teachers, and other professionals in screening children with motor performance difficulties (Taylor). Initially, the teacher answers, yes or no to the statement, “I am concerned about the motor development of this child”. If concern is indicated, then the teacher completes 10 additional questions. The questions describe general motor abilities, performance of simple everyday activities, and behavioural patterns of each student. The questions are answered using a 4-point likert scale of descriptions ranging from well-coordinated behaviour to those associated with a performance below the expected level of proficiency (Przysucha & Taylor, 2004). The MBC has been used reliably to screen children for DCD in a catching study by Lefebvre and Reid (1998), and a balance study by Przysucha and Taylor. Other screening tools exist, such as the Movement Assessment Battery for Children Checklist and the Developmental Coordination Disorder Questionnaire (Taylor), however they are longer in length, and therefore require more time to complete (Henderson & Sugden, 1992, p.2; Wilson et al., 2000). The tests developed to assess motor impairment or dysfunction in children, include the Movement Assessment Battery for Children Performance Test (MABC), the Bruininks-Oseretsky Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 12 Test for Motor Proficiency (BOTMP), and the Developmental Coordination Disorder Questionnaire (Dewey & Wilson, 2001). None of these tests is considered the gold standard when it comes to the identification of DCD, however the MABC does appear to be emerging as the most frequently used motor assessment tool used by researchers and clinicians internationally (Polatajko & Cantin, 2005). The MABC was created to yield an estimate of movement competence in children aged 4 to 12 years (Henderson & Sugden, 1992, pp.2-3). It is not designed for identification of children who have above average motor profiles, as it is not sufficiently discriminating at that end of the normal distribution (Sugden & Chambers, 2005, p. 143). There are four age bands (4-6, 7-8, 9-10 and 11-12) of eight tasks divided into three performance areas, manual dexterity, ball skills, and balance (Henderson & Sugden, p.2). The test is scored in four steps. First, the raw score for each task is recorded. The raw score for each task is then converted into a scale score, ranging from zero to five, with lower scores indicating a better performance. The individual task scores are then summed to produce the Total Impairment Score (TIS), which is next converted into percentile form (Burton & Miller, 1998, p. 173). A TIS that is at or below the 5th percentile is indicative of a definite motor problem, while a TIS between the 5th and 15th percentile suggests the child is considered at risk of DCD or has a borderline motor dysfunction (Henderson & Sugden, p. 107). The MABC is widely used in research all around the world to classify children into groups designated as clumsy, motor impaired, DCD or typically developing (Sugden & Chambers, 2005, p. 143; Smits-Engelsman, Henderson & Michels, 1998). Researchers in China, Japan, Scotland, Australia, Sweden, the Netherlands, and Singapore have found the MABC to be a useful tool to identify children with DCD (Chow, Henderson & Barnett, 2001; Miyahara et al., 1998; Mon-Williams, Pascal & Wann, 1994; Pick & Edwards, 1997; Rosblad & Gard, 1998; Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 13 Smits-Engelsman et al. 1998; Wright, Sugden, Ng & Tan, 1994). Smits-Engelsman et al., Rosblad et al. and Wright et al. determined the MABC to be a suitable tool in the Netherlands, Sweden, and Singapore in differentiating children with DCD from their peers. The norms provided in the MABC manual are satisfactory, but may require some alteration. Chow et al. and Miyahara et al. found the test content was suitable for children in China and Japan, however cross-cultural differences were found in the scores. Therefore, norms for these countries may need adjustment. Many of the studies were however, limited to one age band and may yield different conclusions if expanded to other age categories. Although the MABC is widely used and has unique features that enhance its usefulness for screening, intervention planning, and clinical exploration, Burton and Miller (1998) suggest there is insufficient evidence to adequately establish the reliability and validity of the test (pp.176-177). In contrast, Croce, Horvat, and McCarthy (2001) determined the test-retest reliability to be high. They believe the high test-retest reliability supports the use of the MABC and allows teachers, clinicians, and researchers to be confident of the initial assessment of a child and identification of motor deficiencies. Most recently. Van Waelvelde, Peersman, Lenoir and Smits-Engelsman (2007) identified the reliability of most individual scores and the sub-scores for ball skills were poor, while the manual dexterity and balance scores showed good to moderate reliability. On a whole, they believe the total impairment score of the MABC is a reliable measure for identification of mild to moderate motor impairment in young children. The validity of the MABC has been determined by comparing scores with the BOTMP. There are a number of studies that examined whether children identified with DCD on one test were consistently identified on the other. Dewey and Wilson (2001) discovered it was not uncommon for children to score within the average range on one test, but to be impaired on the other. Crawford, Wilson and Dewey (2001) also found low levels of agreement between the two Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 14 tests. Over one third of the children identified with DCD by the BOTMP were not identified by the MABC, whereas one quarter of those not identified with DCD by the BOTMP were identified with DCD on the MABC. Croce et al. (2001) highlighted a statement by Henderson and Sugden (1992) that acknowledged the two tests are similar in some respects, but different in others. Each test was designed with a different goal in mind. The MABC focuses on the identification of impairment, while the BOTMP measures motor ability across both gross and fine motor function (Henderson & Sugden, p.206). With that in mind, Croce et al. believe the concurrent validity between the MABC and the BOTMP is good. Other aspects that make the MABC an appealing assessment tool are the ease of administering the items of the MABC to participants, reported by Croce et al. (2001). Their participants also commented that the testing was not tedious, difficult, or discouraging, which was not the case for the BOTMP. In addition, Chow and Henderson (2005) found a relatively inexperienced tester can be trained to use the MABC quite reliably by studying the manual and testing children of widely differing ability. Attention Attention has most recently been viewed through a neurological approach (Fan et al., 2002). This involves perceiving attention as an organ system with its own anatomy and circuitry. An organ system is defined as differentiated structures made up of various cells and tissues that are adapted for the performance of some specific function and grouped with other structures into a system, in this case attention. The specific functions of attention are broken down into three networks, alerting, orienting, and executive control (Fan & Posner, 2004). The alerting network is responsible for achieving and maintaining a vigilant state to incoming stimuli (Fan et al., 2002). The main function of the network is to reduce background Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 15 noise and maintain an adequate amplification for the task at hand (Posner & Badgaiyan, 1998, p.69). It is critical for optimal performance in tasks involving higher cognitive functions, such as tasks involving reaction time and the appearance of infrequent stimuli (Raz, 2004; Posner & Badgaiyan, p.68). The structures of the brain associated with the alerting network are the frontal and parietal regions of the right hemisphere, and the locus ceruleus (Fan et al.; Raz). The locus ceruleus is the originating location of norepinephrine, which is the neurotransmitter linked to the network (Posner & Badgaiyan, p.68). The orienting network is responsible for selecting information and aligning attention with a source of sensory signals (Raz, 2004; Posner & Rothbart, 2007). The network directs attention from an unattended location to a target location or object (Posner & Badgaiyan, 1998, p.62). This process may be overt, when eye movements accompany movements of attention, or covert, without any eye movement (Posner & Rothbart). It can also be reflexive, a shift of attention to a location due to a sudden event, or voluntary, a conscious search for information in the visual field (Posner, 1980). Orienting can be manipulated by presenting a cue indicating where a target is likely to occur, thereby directing attention to the cued location (Posner). The frontal eye fields and posterior structures of the brain, including the parietal lobe, pulvinar of the thalamus and superior colliculus are the brain structures involved in the orienting network (Raz; Posner & Badgaiyan, p.62). Acetylcholine (cholinergic systems) arising in the basal forebrain is the neurotransmitter involved in orienting (Raz). The executive network involves more complex mental operations engaged during monitoring and resolving conflict (Fan & Posner, 2004). The network becomes more active and is most needed during tasks that involve complex discrimination in processes such as conflict resolution, error correction, inhibitory control, and planning and resource allocation (Posner & Badgaiyan, 1998, p. 65; Raz, 2004). These processes are involved in variations of the flanker task Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 16 and the Stroop task (Fan et ah, 2002). These tasks activate anterior structures of the brain, such as the anterior cingulate cortex, lateral ventral prefrontal cortex and basal ganglia (Raz). The basal ganglia supply dopamine, the neurotransmitter of the executive control network, to the frontal lobe (Posner & Badgaiyan, p.65). In 2002, Fan et al. set out to develop a behavioural task to measure the efficiency of each of the three networks in adults. By combining the cued reaction time test developed by Posner in 1980, and the flanker test created by Friksen and Friksen In 1974, the group developed the Attention Network Test (ANT; Fan et al., 2002). Their results indicate that the ANT produces reliable estimates of alerting (M = 47 msec, SD = 18), orienting (M = 51 msec, SD = 21), and executive control (M = 84 msec, SD = 25). Fan et al. also determined that the efficiencies of the three networks were uncorrelated and therefore assumed they work independently. In a later study Callejas, Lupianez, and Tudela (2004), modified the ANT to introduce a short duration high frequency tone that would enable them to independently measure the three networks and the effect of each one on the other two networks. They were able to find interactions between all three networks. First, the executive control network is inhibited by the alerting network, whereas the orienting network raises the efficiency of the executive control network. Second, they found that the alerting network influences the orienting network by accelerating rather than enhancing its effect. Callejas et al. were able to conclude that, although the three attention networks are independent, and are subtended by different neural networks, the three attention networks act under the constant influence of one another in order to produce an efficient and adaptive behaviour. The distinctions and overlaps between attention networks were replicated by Callejas, Lupianez, Funes and Tudela (2005) and Fan, McCandliss, Fossella, Flombaum and Posner (2005). Callejas et al. used a similar ANT to Callejas et al. (2004) but with a lengthened Stimulus Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 17 Onset Asynchrony (SOA), while Fan et al. used the original ANT and functional magnetic resonance imaging (fMRI). The development of the attention networks has also been examined. Rueda et al. (2004) used both the ANT and the Attention Network Test for Children (ANT-C) on children and adults. The estimates of efficiency and standard deviation for the three attention networks for the age range studied are presented in Table 1. Rueda et al. determined that reaction time (RT) and accuracy improve at each age interval, and increases in efficiency were found for two of the three attention networks. The alerting network was found to develop up to and beyond the age of 10 years into adolescence and adulthood. The executive control network showed strong development from 4 to 7 years of age and stabilized after the age of 7 years. Finally, the orienting network seems to be formed by the age of 4, and therefore did not change in the age range studied (Fan & Posner, 2004; Raz, 2004; Rueda et al.). Table 1 Estimates o f network efficiency (msec) by age (Rueda et al., 2004) Age (years) Alerting Orienting Fxecutive Control 7 100(75) 62(67) 63 (83) 8 73(67) 63(66) 71(77) 9 79(47) 42(48) 67(38) 10 41(47) 46(44) 69(44) Group Mean 7T25 5T25 5Œ25 * Standard deviations for the RT data are presented between parentheses. Konrad et al. (2005) were also interested in the development of the attention networks. They modified the ANT to present the stimuli peripherally, rather than centrally and also used fMRI with their adult and child participants. Their data agreed with the findings of Rueda et al. (2004), suggesting that there is a transition from functional yet immature systems supporting attention functions in children to more definitive, mature networks in adults. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 18 Assessment o f Attention. The Attention Network Test for Children (ANT-C) was adapted from the adult version, the ANT, by Rueda et al. (2004). The ANT-C, attempts to quantify the processing efficiency of the alerting, orienting and executive control networks of children (Fan et al., 2002; Raz, 2004). It was adapted with the intent of making the task more appealing for children by replacing the target arrows with swimming goldfish (Rueda et al.). The ANT-C uses the reaction time (RT) between conditions to measure efficiency of each network. Depending on the cue condition, each trial may or may not begin with a cue that informs the participant that a target will soon appear and of the potential location of the target (Posner, Sheese, Odludas & Tang, 2006). There are four warning cue conditions (no cue, centre cue, double cue, and spatial cue). Fach is illustrated in Figure 1 (Fan et al.). In the no cue trials, the participant is not presented with a centre or spatial cue. They continue to see only the fixation point. In the centre cue trials, the participant is presented with an asterisk where the fixation dot is located. In the double cue trials, the participant is presented with two asterisks, one above and one below the fixation point. Finally, in the spatial cue trials, only one asterisk is presented to the participant, either above or below the fixation point. The spatial cues are always valid. In other words, the targets are always presented in the same location as the warning cue (Fan et al.). * * + * + 4- + * No Cue Centre Cue Double Cue Spatial Cue Spatial Cue (Up) (Down) Figure i. Warning conditions of the Attention Network Test for Children (Rueda et al., 2004). The target is then presented to the participant. The participant must determine which direction the fish is swimming, left or right. The target appears above or below the fixation point, and is potentially accompanied by four flankers. The flankers surrounding the target either match Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 19 the targets direction (congruent condition), are in the opposite direction (incongruent condition), or do not appear (neutral condition), as seen in Figure 2. When the researchers present the test instructions to their young participants, they ask for help with feeding the fish. The participant helps feed the fish by pressing the button corresponding to the direction in which the middle fish or single fish is swimming (Rueda et ah, 2004). or or or Congruent Incongruent Neutral Figure 2. Target conditions of the Attention Network Test for Children (Rueda et al., 2004). Each trial consists of five events, which are illustrated in Figure 3. The events include an initial fixation period of random duration varying from 400 msec to 1600 msec, followed by a warning cue for 150 msec. A second fixation period then appears for 450 msec, followed by the target and flankers, which stay on the screen until the participant responds or a time span of 1700 msec is reached. Finally, there is a feedback screen for 2000 msec if the correct response is given, followed by a constant post-target fixation period of 1000 msec. If the incorrect response is given, the feedback screen is skipped and the sequence continues with the post-target fixation period (Rueda et al., 2004). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 20 1 + + + + + + Initial Warning Second Target and Feedback Post-Target Fixation Cue Fixation Flankers (200 msec) Fixation Period (150 msec) Period (<1700 msec) Point (400-1600 (450 msec) (1000 msec) msec) Figure 3. Sequence of events of each trial of the Attention Network Test for Children (Rueda et al., 2006). The reliability and validity of the ANT-C have not been formally examined. However in the adult ANT, the test-retest correlations for the cognitive subtractions used to provide three numbers that describe the efficiency of each of the three attention networks are somewhat less reliable than that of the raw RT (r = 0.87). The alerting network appears to be the least reliable (r = 0.52), whereas the executive control network is the most reliable (r = 0.77) and the orienting network falls between the two (r = 0.61; Fan et al., 2002). To add to the reliability of the ANT-C, Rueda et al. (2004) tested a group of adults and a group of 10 year old children on both the adult ANT and the ANT-C. Neither version showed a significant difference between children and adults. However, the adult ANT provided conflict scores nearly twice as high as the ANT-C. This suggests the adult ANT is considerably more difficult. Developmental Coordination Disorder and Attention The etiology of DCD is unknown (Cermak & Larkin, 2002, p. 16). To date, a single factor has not been identified to be the cause of DCD. However, various factors have been investigated. Factors considered over the years include brain damage or dysfunction, genetic predisposition, impairment in information processing, or an impoverished environment (Cermak & Larkin, 2002, Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 21 p. 16). In a meta-analysis conducted with some of the literature aimed at identifying the mechanisms responsible for DCD, Wilson and McKenzie (1998) suggested that relative to matched controls, children with DCD most consistently demonstrate a deficiency with respect to the processing of visuospatial information. This conclusion is not all that surprising, since one of the associated behaviours of DCD is difficulties in attention (Sugden & Chambers, 2005, p. 14). In a study by Dewey et al. (2002), the research team investigated the problems of attention experienced by children with DCD. The parents of the participants were asked to complete the Attention Problems Subscale of the Child Behaviour Checklist and the Hyperactivity Index from the Abbreviated Symptom Questionnaire (Achenbach, 1991). The results revealed that both children with DCD and children suspected of having DCD scored poorer on the measures of attention than the comparison children. Based on the findings of the meta-analysis by Wilson and McKenzie (1998), Wilson and Maruff (1999) investigated children with DCD on their movement of attention through visual space to designated target locations using the Covert Orienting of Visuospatial Attention Task (COY AT). The CO VAT provides a measure of an individual’s ability to direct visuospatial attention to areas of the visual field without accompanying eye movements. In the voluntary mode, the RTs for both Children with DCD and the control children were faster when the stimulus appeared at the cued rather than the uncued location (spatial precue effect). More importantly, the magnitude of the effect was significantly greater for the children with DCD than for the control participants. Therefore, only the children with DCD demonstrated results consistent with a deficit in the disengagement operation of directing covert attention. These results confirmed the earlier findings of Wilson, Maruff and McKenzie in 1997. It is also important to acknowledge that the attention deficit was not evident in all of the children with DCD. However, the RTs of most children with DCD were slower than those of the Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 22 controls, which may be explained by the motor deficits that occur as a part of DCD, although it is unlikely that the motor disabilities led to the abnormal orienting response (Wilson & Maruff, 1999). Mandich, Buckolz and Polatajko (2003) attempted to replicate Wilson and M aruff s (1999) findings, and continue the examination of the inhibitory function of children with DCD with respect to the movement of attention. Their intention was to clear up the uncertainty of whether the disengagement operation was volitional or whether it was affected by automatic factors. To do this they used a spatial precue task with both informative and uninformative precue conditions. Mandich and colleagues were successfully able to demonstrate that children with DCD exhibit a number of inhibitory deficits with respect to the intentional movement of attention through visual space. More specifically, the children with DCD took longer to disengage attention from a voluntary cued location, so that it could be moved to the target position. The patterns of attention deficits exhibited by the children with DCD parallel those that have been observed in children with attention deficit hyperactivity disorder (ADHD). Both groups have been shown to perform within normal limits on COVAT in the reflexive orienting mode, but display abnormalities within the voluntary orienting mode (Wilson & Maruff, 1999). In addition to the orienting network, children with ADHD have been tested using a modified ANT to investigate the efficiency of the attention networks, alerting, orienting and executive control. Konrad, Neurfang, Hanisch, Fink and Herpertz-Dahlmann (2006) modified the ANT by having the arrow targets appear in a vertical row to either side of the fixation point, rather than in a horizontal row above or below the fixation point. The children with ADHD were boys between the age of 8 and 12 years. They demonstrated efficiency scores of 55 msec, 138 msec and 122 msec for the alerting, orienting and executive control, respectively. These results suggest a significant deficit in the executive control network when compared to their peers without ADHD. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 23 The ANT has also been a convenient and effective tool in the evaluation of attention abnormalities associated with strokes and other brain injuries. It has also been used successfully to test a variety of clinical populations, including individuals with borderline personality disorder and schizophrenia. The findings from these studies may be useful in designing better interventions and determining the effectiveness of pharmacological and behavioural interventions (Fan & Posner, 2004). It would therefore be interesting to examine the results of children with DCD on the ANT-C. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Method Procedure Recruitment. After receiving ethical approval from the Ethics Review Board at Lakehead University, participants were recruited through the Thunder Bay Catholic District School Board, the Motor Development Clinic at Lakehead University, summer camps offered through the Athletic and Science departments at Lakehead University, local community groups, health care providers and day care centres, and word of mouth. A proposal similar to that submitted to the Ethics Review Board at Lakehead University was submitted to the Thunder Bay Catholic District School Board. Approval was granted and a number of schools were identified to be contacted. Information packages, including cover letters and consent forms for the principal, teachers and participants along with a sample Motor Behaviour Checklist (MBC), were delivered to each school (refer to Appendices A through D). Upon confirmation of participation in the study. Teacher Cover Letters, Consent Forms and MBCs with instructions were delivered. Each teacher was asked to complete a checklist for each boy in his or her class. Once the checklists were completed by the teachers, participant information packages including a cover letter and consent form were dropped off for teachers to send home with students. At this time, the completed teacher consent forms and checklists were collected. It was then left to the parents of the children to contact the researcher to set-up a time for testing. As well, past and present participants of the Motor Development Clinic were contacted by the researcher if they were the correct age for the study. Contact information was supplied by Dr. Jane Taylor, the coordinator of the program. Also, participants were recruited through referrals of an Occupational Therapist in Thunder Bay to the upcoming Motor Development Clinic. 24 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 25 In addition, the organizers of the Thunderwolves Basketball Camp and Superior Science Camp sent home participant information packages with their participants that met the age requirements. Again, it was left up to the parent of the child to contact the researcher if they were interested in participating. In addition, posters were hung at local community centres and a chiropractic office, an advertisement was published in a local skating club newsletter, and flyers were distributed at a local soccer complex and day care centre to children that fell into the identified age range (refer to Appendices E through G). Finally, awareness of the study was passed around by word of mouth through the Kinesiology and Athletic Departments at Lakehead University. A number of participants who were recruited were children, relatives, or friends of faculty, staff and students. Screening. The MBC was used as an initial screening tool for potential participants. The information provided by the checklist offered an early indication of group association. Children recruited from the Thunder Bay District Catholic School Board were the only group to be screened using the MBC. Past and present participants of the Motor Development Clinic and those referred for the upcoming clinic were screened using other assessment techniques by the occupational therapist that referred them. All participants were assigned to a group based on their percentile ranking on the MABC. Participants. A sample of twenty-five males between the age of 7 and 10 years were recruited to participate in the study. All participants with a Total Impairment Score (TIS) on the MABC at or below the 15"’ percentile were identified as the group with DCD. The remaining participants had a TIS at or above the 20'" percentile and formed the group without DCD. Both groups had normal or corrected to normal vision, an added requirement of the ANT-C. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 26 Testing. Testing was completed in either one or two sessions. Most participants completed both the MABC and the ANT-C in a single session, approximately 1 hour in length. However, boys recruited through referrals for the upcoming Motor Development Clinic completed the testing in two sessions. The first session, roughly 1 hour in length, included the MABC test as well as other tests included in the initial assessment for the clinic. A second visit was requested to complete the ANT-C. This session was completed in approximately 30 minutes. All testing sessions took place in room 1028A in the C.J. Sanders Field House at Lakehead University. Assessment of the MABC was completed according to the manual by Henderson and Sugden (1992). Refer to Appendix H for the protocol. The participant’s parent or guardian was asked to complete an MABC Checklist to provide descriptive information on the child’s behavioural problems related to motor difficulties. The MABC Checklist was adapted to include a question about medication related to Attention Deficit Disorder with or without Hyperactivity (see Appendix I). This information was collected to assist in the interpretation of the results and was not used as inclusion or exclusion criteria. The ANT-C was programmed on the computer by the researcher. The appearance of the test and the procedure replicated the program guidelines used by Rueda et al. (2004). Refer to Appendices J and K for the developmental procedure and protocol. Preliminary Evaluation. A preliminary examination of data using sixteen participants was completed and presented as a poster at the North American Federation of Adapted Physical Activity (NAFAPA) conference in Indianapolis Indiana, on September 6'", 2008 (see Appendix L). The review provided an opportunity to become familiar with the organization, analysis and interpretation of data, offered a good indication of what to expect when the entire sample was included, and identified factors to examine for interpretation of the results. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 27 Analysis The independent variable of interest is the two groups, boys with and with DCD, while the dependent variables are median reaction time (RT) on accurate trials and error rate. RT is recorded as the time from onset of the target to the time when the participant pushes the button on the keyboard. RTs from correct trials were trimmed to exclude outlying responses. The lower cut­ off was set at 100 msec to exclude no responses and anticipatory responses, and the upper cut-off was set at 1500 msec (Callejas et al., 2005; Roberts et al., 2006). Incorrect trials, which included trials where the child responded by pushing the incorrect button on the keyboard were also excluded, but used to calculate the error rate and the overall error percentage (Van Donklear et al., 2005). The error rate is calculated as the percentage of incorrect trials within a condition (congruent, incongruent or neutral), or a percentage of all the trials (Van Donklear et al., 2005). To determine the efficiency of each attention network, first the RTs are organized by their warning cue and target type. The median of each condition was then used to determine the median RTs for the no cue, centre cue, double cue, spatial cue, congruent and incongruent conditions (Rueda et al., 2004). Next, three simple calculations were computed to produce the network efficiency scores. First, subtracting the median RT obtained in the double-cue conditions from the median RT in the no cue conditions measures the alerting network due to the presence of a warning signal (Fan et al., 2002; Posner et al., 2006). Second, the orienting network is measured by subtracting the median RT of the spatial cues from the median RT of the central cues, since the spatial cue, but not the central cue, provides valid information on where the target will occur (Fan et al., 2002; Posner et al., 2006). Lastly, the executive control network is responsible for conflict resolution. It is measured by subtracting the median RT of the congruent target trials from the median RT of the incongruent trials (refer to Figure 4; Fan et al., 2002; Posner et al., 2006). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 28 Alerting = median RT no cue condition - median RT double cue condition Orienting = median RT centre cue condition - median RT spatial cue condition Executive Control = median RT incongruent conditions - median RT congruent conditions Figure 4. Network efficiency calculations (Fan et al., 2002). Next, independent sample t-tests were performed on the no cue, double cue, centre cue, and spatial cue, as well as the alerting, orienting and executive control network scores to determine if there was a significant difference between the efficiency of the networks in boys with and without DCD. The percentage of error between target type (congruent, incongruent and neutral) and warning cue type (no cue, double cue, centre cue and spatial cue) were also examined using independent sample t-tests. Bonferroni corrections for inflated Type I error were applied to both analyses in order to control for the number of analyses conducted (Konrad et al., 2006). Due to the small sample size Cohen’s d effect size calculations were also completed. Independent sample t-tests were also used to examine the difference between the overall median RT and error rate between groups (Van Donklear et al., 2005). Additional, pairwise t-tests were used to examine the difference between the no cue and the double cue, the centre cue and the spatial cue, and the incongruent and congruent target conditions within each group. Finally, bivariate correlations were computed to examine two types of relationships. The first correlation was used to investigate the relationship between the alerting, orienting and executive control networks within each group. The second correlation explored the relationship between the group with DCD and the group without DCD on each of the attention networks. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 29 Interpretation o f the ANT-C Scores In general, higher network scores, or a greater difference between the warning cues or targets used in the network score calculations represent less efficient networks, while lower network scores, or a lesser difference between the conditions represent more efficient networks. In the alerting network, a higher network score would describe difficulty maintaining alertness without a cue. A high score in the orienting network indicates a difficulty disengaging from the centre cue where no target appears, whereas a high score in the executive control network suggests difficulty resolving conflict (Fan & Posner, 2004). If this interpretation were true, smaller network scores or a lesser difference between conditions would be considered more efficient. Unfortunately, it is not always that straight forward. High network scores may also be considered efficient when a participant is making proper use of the cues or even expending increased effort. This explanation is more probable when the overall RTs are relatively quick. Similarly, the smaller network score described in the previous paragraph would only be efficient if associated with quicker overall RTs. Consequently, if the small network score or lesser difference was calculated using slow RTs, the network would be considered less efficient. Hypothesis Prior to performing all statistical analyses, it was expected that the overall reaction time of the boys with DCD would be slower than that of the boys without DCD, similar to the results of Wilson and Maruff (1999). In regard to the individual attention networks, it was expected that the children with DCD would demonstrate deficits in the orienting and executive control networks. Previously in the orienting network, both boys with DCD and those with ADHD were found to have a deficit on the voluntary condition of the COVAT (Wilson & Maruff, 1999). While in the Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 30 executive control network, a deficit was found in children with ADHD using a modified ANT (Konrad et al., 2006). Similar results are expected for the boys with DCD based on the similarities on the COVAT and the additional attention difficulties associated with DCD (Sugden & Chambers, 2005, pg. 14). On the final attention network, alerting, it has been demonstrated that the network does not completely develop until the age of 10 years or beyond, therefore it could be speculated that 10-year-old boys with DCD might produce scores associated with the development of the alerting network (Fan & Posner, 2004). In addition, as previously mentioned, attention problems are an associated behaviour of some children with DCD (Sugden & Chambers). Consequently, some boys with DCD may also demonstrate a deficit in the development of the alerting network compared to boys without DCD of the same age, and similar to the deficit in performance of motor tasks (American Psychiatric Association, 1994, pg. 55). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Results Group Characteristics Initially, twenty-seven participants were tested on the MABC, however one did not complete the ANT-C test and a second was subsequently found to have Autism. Therefore, twenty-five participants were included in the data analysis. All of the participants were boys between the age of 7 and 10 years. Fourteen boys were identified as, at risk or with DCD. Together, these boys formed the first group, boys with DCD. They had a mean age of 9 years (SD = 10.97). The group Total Impairment Scores (TIS) ranged from 10 to 27.5, which correspond with the 15'" and below the C' percentile. The group TIS mean was 16.3 (SD = 5.75), while the rank of the group mean was at the 6'" percentile. The remaining eleven boys were not identified with DCD, and therefore formed the second or control group, boys without DCD. The mean age of this group was 8.6 years (SD = 0.96). The individual TIS scores ranged from 1 to 8.5, or the 89'" to the 20'" percentile. The mean TIS was 4.6 (SD = 2.26), which is equivalent to the 51®' percentile. An independent sample t-test revealed that the boys with DCD were not significantly older than the boys without DCD when age was examined (t(23) = 1.17, p > 0.05). Therefore, the two groups were assumed to be the same chronological age. In contrast, analysis of the MABC scores, also by independent sample t-tests, showed that on both the TIS and the percentile ranking, the two groups were significantly different. The boys without DCD scored significantly higher on the TIS (t(23) = 6.96, p < 0.05), while scoring significantly lower on the percentile ranking (t(23) = -6.812, p < 0.05; see Figures 5 and 6). Based on these two analyses, it was concluded that the two groups were of similar age, and the boys assigned to the group without DCD demonstrated significantly better overall motor abilities than 31 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 32 the boys from the group with DCD. Complete individual scores on the MABC are available in Appendix M. Non DCD G ro u p Figure 5. Comparison of Group means on the Total Impairment Score (TIS) on the Movement Assessment Battery for Children. 0) 30 Non DCD G ro u p Figure 6. Comparison of group mean percentile scores on the Movement Assessment Battery for Children. Finally, the results of the MABC Checklist identified only one participant taking medication for ADD or ADHD. From parent interviews for the Motor Development Clinic, eight Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 33 of eleven boys with DCD who participated in the clinic were diagnosed or in the process of identification for ADD or ADHD. Alerting Network The efficiency of the alerting network is calculated by subtracting the median RT obtained in the double-cue conditions from the median RT in the no cue conditions (Fan et al., 2002). The median RT of no cue and the double cue conditions for boys with DCD were 833 msec and 789 msec, respectively (see Figure 8). The individual alerting scores for the group ranged from -65 msec to 141 msec, while the groups’ mean alerting score was 44 msec with a standard deviation of 69 (see Figure 7 and 9). In comparison, the boys without DCD had a no cue median of 844 msec and a double cue median RT of 788 msec (see Figure 8). The individual alerting scores of the group ranged from -64 msec to 212 msec, while the groups’ mean alerting score was 56 msec with a standard deviation of 72 (see Figure 7 and 9). The difference between the alerting score of the boys with and without DCD was not found to be significant when an independent sample t-test with a Bonferroni correction was utilized (t(23) = -0.43, p = 0.51). Further confirmation of this result is supported by the small effect size (d = -0.18). The two groups were also not significantly different on their no cue and double cue RT scores (No Cue, t(23) = -0.28, p - 0.78; Double Cue, t(23) = 03, p = 0.98). In contrast, a pairwise sample t-test determined a significant difference between the no cue and double cue conditions within each group (DCD, t(i3) = 2.37, p - 0.03; Non DCD, t(io) = 2.59, p = 0.03). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 34 1 2 3 4 5 10 11 12 13 15 16 17 i 19 20 21 22 23 24 25 Participant Figure 7. Individual alerting network efficiency scores. * Participants 1-14 are boys with DCD, participants 15-25 are boys without DCD ?> 830 i r 810 - o 800 I DCD 789 788 I Non DCD No C ue Double C ue Warning Cue Figure 8. Comparison of group means on median reaction time in the no-cue and double-cue conditions. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 35 140 120 H 100 2 I 80 I Standard Deviation ■£ i 60 ( I Network Score « ^ 40 - 56 20 - 0 DCD N onD C D Group Figure 9. Comparison of group means and standard deviations on alerting scores. Orienting Network The boys with DCD had individual orienting scores that ranged from -89 msec to 154 msec and a group mean score of -2 msec with a standard deviation of 72 in the orienting network (see Figure 10 and 11). The boys without DCD had had individual orienting scores that ranged from -141 msec to 207 msec and a group mean score of 46 msec with a standard deviation of 100 (see Figure 10 and 11). The orienting score was calculated by subtracting the median RT of the spatial cue from the median RT of the centre cue conditions (Fan et al., 2002). The median centre cue values for the boys with and without DCD were 833 msec and 818 msec, respectively (see Figure 12). The boys with DCD had a spatial cue median RT of 835 msec and the boys without DCD had a median RT of 773 msec (see Figure 12). Similar to the alerting network, the independent sample t-test also found no significant difference between the orienting scores of the boys with and without DCD {t(23) = -1.387, p = 0.18). Although no difference was found between the groups, a medium effect size was revealed (d= -0.55). In addition, no significant difference was found on the RTs of the spatial cue and Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 36 centre cue conditions between the two groups (Centre Cue, t(23) = 0.36, p = 0.72; Spatial Cue, t(23) = 1.48, p = 0.15), or within each group on the centre cue and spatial cue (DCD, t(i3) = O.-0.11, p = 092; Non DCD, = 1.51, p = 0.16). * 0 - 19 20 21 22 23 24 25 Participants Figure 10. Individual orienting network efficiency scores. * Participants 1-14 are boys with DCD, participants 15-25 are boys without DCD 840 1 833 835 830 " 150 ■ S tandard Deviation g B Networl< S core100 - 135 50 - 117 DCD N onD C D Group Figure 15. Comparison of group means and standard deviations on executive scores. Error Rates The error rates were examined by both target and warning cue condition. The error rates were determined by dividing the number of incorrect trials by the total number of trials in each specific target or warning cue type. As a review, the target conditions included congruent, incongruent and neutral. The group with DCD had an error percentage of five, eleven and seven, respectively. Where as, the group without DCD had equal or lower error percentages of five, seven and five for the congruent, incongruent and neutral warning cue types. Using an independent sample t-test with a Bonferroni correction, the two groups were not found to be significantly different in error rates on any of the target types (Congruent; = -0.1 l ,p = 0.91; Incongruent, t(23) = 1.07, p = 0.29; Neutral, t(23) = 0.69, p = 0.5; see Figure 13). Individual target type error rates are available in Appendix N. In addition, a one-way ANOVA was used to determine if there was a difference in responses to the target types themselves in each group. They too, were not significantly different in the DCD group (F^g) = 1.80, p = 0.18), or the non DCD group (F(43) = 0.32, p = 0.81). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 4 V 12 11 JO 10 p o DCD is a motor-based performance problem that limits a child's > To investigate the attention profile of boys with DCD by @ominlng > Mean Alerting, Orienting, and Executive Control Function ability to fully participate in the eve^day aithfkies of childhood the efficiency of the three attention networks, alerting, orienting, and. (Polaiajko & Cantin, 2005: American Psychiatric Association. 1994). executive control, using the Attention Network Test for Children (ANT- C; Reuda et 3l.. 2004)______________________________________ > There are a variety of factors that have been identified in an attempt to determine the cause of DCD. however It is still unknown (American Psychiatric /teociation. 1994). One factor that has been M ethod MUM* identified Is a defic) in visual spatial processing (Wbon & kkKentie. > Participant? 1998). This theory has led to àn irrvestigation into the orienting • The sample included 16 boys. 8 with DCD and 8 without, attentiori network o f children with DCD, The orienting network Is between the ages of 7 and 10 years (mean = 8.8 years). responsible for directing atention from an unattended location to a : > ivfean o f hifedian RTs (msec)• 'tAifthi DCD idemhled by an MABC Composite Score at or target location or object (Posner & Badgaryan, 1998. pg. 62). Children below the 3™ percentile (mean * M'pementile) and a with DCD have been identified with a défbit inthe voluntary Nitinuai Dexterity Score at or below.the IS^^percentile KFCtF BMWCkF CtFWChF 4m l C»F HN> disengagement of amentiort, while refladve orienting is not implicated (mean = below 5^ percentile).. fU T fjt/ iK.TTf-.j «feTc r i r ; i / v . t fit.1 in the disorder (Wilson atvtiru ff. 1999). "VWthout DCD identified by an MABC Composite Score at i««nwi i»:.r t::.t c j j cw. 0 | > or above the 20 ̂ percentile (mean = 36*'percentile)and a In addition to the orienting, there are two other attention networks, MUbybnuai Dexterity feore at or above the )5*^percentile ■ .rjl TT;,-»tT.T 01 alerting and executive control. Alerting is responsible for achieving (mean = above 15*'percentile): and maintaining a vigilant stateto incoming stimuli (Fan. tvfcCsndliss. >• fifc-il/.'. Sommer. Raa & Posner. 2002), while executive control is responsible > Recruitment for more complex mental operations engaged during monitoring and resolving conflict among stimuli (Fan et al.). Most recently,attention ■ Participants with DCD were recruited through the Motor 0 0 0 0 Mh* | Fkxlt* has been viewed through a neurological approach (Fan et al.) in Development Clinic à lakehead University, while participants without DCD were recruited through local schools and suirmer which attention is perceived as an organ ^ e m with its o m anatomy and circuitry (Fan S Posner. 2904) Together the three net works form the organ system of attention; D isc u ss io n > Data Coileaion > Using an Independent Sample T-Test with Bonferronl correction, > Attention problems have been identified as an associated problem • Parerits of the partcipants completed an Adapted MABC there was no significant difference found between boys with and In children with DCD (Sugden & Chambers, 2005. pg; 14; Dewey, Checklist. without DCD in their scores on network efficiency for alerting • • Participants completed the MABC and the ANT-C Kaplan. Crawford SVWIson. 2002). As no research presently exists, it 0.298, p =. 0.772). orienting (f (i*) = 0 828, p = 0.826) and executive Is of interest that the efficiency of the alerting, orienting and executive control ( f ( i* )= -0.885, j»s 0.391). control networks be examined using the ANT-G in children with DCD (Reuda et al.. 2094). The findings fromthis study may be useful in A tten tio n N e tw o rk T e s t fo rC h ild ren >-The effect sizes for alerting (d = 0.14), orienting (fi/ » -0.41) and designing better behaviourai interventions (Fan & Posner, 2004). executive control (d = 0.44) were also small. > Although it appears that there are differences in the orienting and executive control networks.the variation within the network scores R e fe re n c e s may contribute to the lack of significant difference between groups. . . . . . . . . . . - . .. .. . . . The alerting, orienting and executive control standard deviaions for ti, «M il. the DCD and control groups are 48.9 and 43.6.51 and 81.1. and 53.2 and .3. Inspect ively. ) & I"WI«WW. kf.iiw ; «M» > I «OOtfWWIII > In addition,the overall number of errors committed by participants >.iiMX.Vu’M)M>rM>Mi«>nuNi<.k»i,lik: h>jF*#«.Kimi.iA«6l.Çi’«vw-'flr.#we«#weim AckBowlcdfCBCits 6Winr.F.ii* l |.ll» tkk F*«***,«#*«(«*m l i * k * k irbw .# > Or. Jim NfcAulirfe. Nipissing Unrvei^ity > Therefore, the hypothesis of an orienting defick in boys with DCD. > Professor Byk PfTysucha, Lakehead University which was previously suggested by WIson and ivbruff (1999) using )lMl«f|F)',tHlWWI».kP.II*F|»,kiFuf»t.W fi«i*v mtKut fijnnitifw1iFsw B̂y.FÂfettfjW.F»U lCM_NF̂_lVr_ the COVAT. was not confirmed.> Professor Carlos Zerpa, Lakehead Unwersmr Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix M - Participant Performance on the Movement Assessment Battery for Children 98 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 2 Individual characteristics including age and Movement Assessment Battery fo r Children Scores Participant Group Age Manual Ball Balance Total MD TMIS Number Dexterity Skills Score Motor Percentile Percentile Score Score Impairment (MD) Score (TMIS) 6 1 7.17 9 1.5 0 10.5 <1 13 1 1 8 2.5 4 4 10.5 >15 13 4 1 8.67 3 2.5 7.5 13 >15 6 7 1 8.25 8.5 4 8 20.5 <1 0 9 1 8.42 8.5 1 0.5 10 <1 15 17 1 8.17 10 2.5 4 16.5 <1 2 10 1 9.58 9.5 0 7 16.5 <1 2 13 1 9.67 7.5 0 5.5 13 <1 6 14 1 9.5 9.5 3 2 14.5 <1 4 16 1 9.33 14.5 7 3 24.5 <1 <1 18 1 9.33 8 0 2 10 <1 15 22 1 9.33 10.5 9 8 27.5 <1 <1 19 1 10.58 14 1 3 18 <1 1 21 1 10.58 9.5 5 8.5 23 <1 <1 2 2 7.75 3 0 0 3 >15 65 8 2 7.5 1 0 0 1 >15 89 27 2 7.75 4 1.5 0 5.5 >15 40 3 2 8.17 1.5 0 1.5 3 >15 65 5 2 8.42 4 0 0 4 >15 54 11 2 8.92 1 1.5 2 4.5 >15 49 12 2 8 2 0 0 2 >15 79 24 2 8.08 2 1 3 6 >15 36 25 2 9.75 3 0 3 6 >15 36 26 2 9.92 4 0 3 7 >15 29 20 2 10.17 5 1 2.5 8.5 15 20 23 3 8 4.5 4 8.5 17 >15 2 15 3 9.3 13 8 10.5 21.5 <1 <1 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix N - Individual Scores on the Attention Network Test for Children 100 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 3 Participant # Participant Group Alerting Orienting Executive Number on Control Figures 6 1 1 92 -79 276 1 2 1 99 81 92 4 3 1 58 -31 118 7 4 1 7 7 54 9 5 1 137 29 252 17 6 1 -39 154 208 10 7 1 141 -49 98 13 8 1 -65 -89 7 14 9 1 2 -70 150 16 10 1 4 -12 138 18 11 1 36 10 103 22 12 1 77 -85 80 19 13 1 116 42 201 21 14 1 -50 67 114 2 15 2 48 106 51 8 16 2 10 -141 117 27 17 2 17 -67 166 3 18 2 -64 -68 22 5 19 2 123 207 172 11 20 2 85 77 82 12 21 2 47 89 143 24 22 2 212 116 200 25 23 2 94 92 95 26 24 2 29 36 108 20 25 2 18 58 129 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 300 1 250 - 200 - 150 - A « A g g| # ♦ Alerfng 50 * ■ OrienSng I A Executive Control 0 1 2 ^ 4 5 6 7 8 9 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 -50 -1 0 0 - -150 -200 J Participant Figure 20. Individual alerting, orienting and executive control scores. * Participants 1-14 are boys with DCD, participants 15-25 are boys without DCD 45 -, 40 - 0 C o ngruent B incongruent ■ Neutral 1 2 3 4 5 6 7 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 P artic ip an t Figure 21. Individual error rate by target type. * Participants 1-14 are boys with DCD, participants 15-25 are boys without DCD Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 103 35 30 26 S I 20 I E No C u e B D ouble C ue 15 B C en tre C ue ■ Spatial C u e 10 5 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Participant Figure 22. Individual error rate by warning cue condition. * Participants 1-14 are boys with DCD, participants 15-25 are boys without DCD 30 25 - 20 2 15 10 5 - 1 2 3 4 5 6 7 P a r t ic ip a n t Figure 23. Individual total error rate. * Participants 1-14 are boys with DCD, participants 15-25 are boys without DCD Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1200 1 1019 1003 1000 907 819 828 796 801 791800 753 751 751 763 678 600 - 400 200 1 2 3 4 5 6 7 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Participant Figure 24. Individual overall median reaction time (msec). * Participants 1-14 are boys with DCD, participants 15-25 are boys without DCD Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.