In Japan, official reports claim that 6.3% (or about 679,000) of elementary and middle school students who are enrolled in normal classes experience learning difficulties (MEXT Japan, 2002). Children with developmental disabilities often experience serious psychological and behavioral problems, such as aggression, anxiety, depression. and attention deficit hyperactivity disorder (AD/HD). It is reported that excellent agreement between attention problems and AD/HD (Biederman et al. 1995). It was also studied that a population sample of 6 to 17 year old children and adolescents and compared the mean Child Behavior Checklist to attention problem scores of control children, and children with AD/HD. Children diagnosed with AD/HD scored higher on attention problems than did control goups (Steinhausen, Winkler, Meier & Kannenberg. 1997). Numerous attempts have been made to study and explain the problem behaviors of AD/HD. However, until now, very few attempts have been made to study the problem behaviors of children with learning difficulties.

We are concerned with problem behaviors of children with learning difficulties. Their difficulty with reading in particular affects all learning domains negatively, hindering academic performance in all subjects. A person’s inability to read well can also generate an inferiority complex that results in the loss of his or her motivation to learn, which in turn may be linked to symptoms leading to juvenile delinquency (Kimberly, Morris & Richard, 2006; Siponmaa, Kristiansson, Jonson, Nyden, & Gillberg, 2001). The inability to read also influences friendships outside of the classroom (Stanovich, 1986) and children’s ability to process feelings of anger (Kazdin, Rodgers, Colbus, & Siegel, 1987; Moffitt & Henry, 1989). All of these factors suggest that addressing reading difficulties should be one of the priorities in helping children with learning difficulties.

Emotional problems are often due to low self-esteem, one of the causes of which is a lack of confidence in learning. For such children, it may be necessary to provide support so that they remain motivated to learn and can smoothly progress through life as a student; their reading ability may improve as a result of appropriate support (Ehri et al., 2001; Temple et al., 2003). Another potential cause of low self-esteem is the strict attitude of the parents or other people close to the child who manifest a lack of understanding of the child’s needs.

In the present study, we focused on two points: academic support for children, and the support of parents in understanding their children. For the academic support, we developed and used more than thirty tasks. Every task was focused on reading, writing, or understanding language. Prior assessments were conducted to choose the appropriate tasks for each participant.

There was time for the parents to talk to the experimenter at length. Parents received information about their child, and as a result, were better able to understand the learning characteristics of their children. The results were reflected in considerable changes in both the parents and the children. Including parents in the intervention influenced not only the learning abilities of the children, but their emotional and behavioral growth as well.

An evaluation checklist was used to inspect what kind of changes arose in children according to the long-term academic support. The Child Behavior Checklist (CBCL), developed as a comprehensive checklist by Achenbach, evaluates emotional and behavioral problems of children (Achenbach, 1978; Achenbach, 1999). We used the parent form of the CBCL for the present study. This has been translated into 58 languages and is used by study organizations and clinics worldwide. It is consistent across different countries and cultures, and its reliability and validity have already been examined in several countries, including Japan (Bilenberg, 1999; Itani et al., 2001). It has been used in Japan for children, ranging in age from infancy to adolescence, who were experiencing emotional or behavioral problems. Comparisons between the children who had been abused and those who had not showed that the former had significantly higher scores for social problems, attention problems, aggressive behavior, externalizing, and total problems (Tsuboi, 2005). The CBCL has also been used with children with developmental disabilities, such as those with autism. The usefulness of the CBCL has been examined and a large body of research has demonstrated its reliability and validity in both clinical and non-clinical populations. It has been reported that score does not change with age (Derks et al., 2006; Itani et al., 2001). Based on these reports, we chose the CBCL to evaluate any changes that occurred in the children’s daily lives outside of the learning intervention.

In summary, we attempted to reduce the emotional and behavioral problems of children with learning difficulties by providing long-term academic support and counseling to both the child and the parent, and we evaluated the changes in the children’s behavior using the CBCL.

Ten children with developmental disabilities, ranging from first grade to third grade (mean age 7.8, SD = 0.55), and their parents participated in this study. Of the ten children, one child had been diagnosed with attention deficit disorder, five had pervasive developmental disorders, and four had learning disorders. Their intellectual ability was equalized (mean IQ = 92.6, range = 82-104). The second grade girl who had pervasive developmental disorder participated as control (age 8.0, IQ = 112). She did not receive learning intervention. The participating children were chosen by a doctor under contract to the Nagoya City Child Welfare Center. Children diagnosed with a developmental disability that caused delays in their learning, but for whom medicine did not need to be administered, were eligible for selection.

The study consisted of three parts. First, the parents completed the CBCL prior to the intervention. Next, children and parents participated in the intervention program (Fukushima et al., 2009) for one year. Lastly, the parents completed the CBCL again.

Children participated in the learning intervention program once a week for one year. This program was carried out in a room called the Cosmos Club in the Nagoya City Child Welfare Center. A play area and the learning area were separated. The learning desk faced a wall and only those things that were necessary for the tasks were placed on the desk. Each intervention consisted of four to five kinds of tasks according to the results of the assessment tasks performed by each child before the intervention. There were more than thirty tasks related to reading, writing and the understanding of language. There was one experimenter and one or more assistants in each case. During the intervention, the experimenter and assistants tried to praise the child frequently, both verbally and by clapping their hands. Each visit consisted of 40 minutes of learning intervention for the children, followed by 10 minutes of counseling for the parents. Children were allowed to play with assistants in the same room during the counseling of the parents. The experimenter discussed the child’s task and performance of that day, and parents talked about the child’s problems in and out of school during the counseling period. The experimenter explored potential ways to deal with the problems.

More than 30 tasks were used for the learning intervention program. Eight tasks focused on reading, 15 tasks focused on writing, and 8 tasks focused on comprehension. In one task, popular words were displayed on the monitor and participants were asked to input what was shown on the screen. An appropriate learning plan was set for each child and the learning intervention was conducted over a period of one year. Each plan was based on the results of the Screening Test of Reading and Writing for Japanese Primary School Children (STRAW) and Picture Vocabulary Test - Revised (PVT-R). The first test examined how hiragana, katakana, and kanji characters could be written in STRAW. The degree of difficulty of the stimuli was divided by school year. The PVT-R measures children’s vocabulary in comparison with normally developed children. Both tests are standardized assessments tool. The tests were administered to children prior to and following the intervention. The present study employed a paired t-test design to evaluate the differences in scores before and after the intervention.

Each parent completed the CBCL both before and after the year-long intervention. The CBCL is a well-standardized checklist for parents to report the frequency and intensity of behavioral and emotional problems, containing 118 items (Achenbach, 1978; Achenbach, 1999; Itani et al., 2001). There are eight syndrome scales: withdrawn, somatic complaints, anxious / depressed, social problems, thought problems, attention problems, aggressive behavior, and delinquent behavior. These are grouped into two categories of behavior: Internalizing (withdrawn, somatic complaints, anxious/depressed) and Externalizing (delinquent behavior, aggressive behavior). This structure allows the child’s specific characteristics to be evaluated, and it can be used for treatment and education (Nakata et al., 1999). Parents rated items such as “Cruel to animals” and “Nightmares” on a three-step scale, where 0 = “Not true (as far as you know),” 1 = “Somewhat or sometimes true,” and 2 = “Very true or often true.” For each child, the total score, and the Internalizing and Externalizing sub-scores were calculated. Higher scores denote more behavioral problems, and the borderline clinical range is defined as 25.15 (Itani et al., 2001).

The present study employed a two-way repeated-measures analysis of variance (ANOVA) design to evaluate the differences in scores before and after the intervention.

Each child received the learning intervention 44 times in the course of a year. Children were praised enthusiastically by the experimenter and assistants following every task. They were at no point scolded by the teacher during the intervention. The children’s mothers attended every counseling session and the fathers attended less than twice during the year. In all cases, the mother filled out the checklist.

Each participant progressed through the learning tasks according to characteristics and levels. The results of all participating children were averaged in STRAW and PVT-R. According to the results of STRAW, a hiragana character became almost possible to be written. Especially in the word condition, the average was almost a perfect score. The results of the kanji and katakana condition, which were low before support, had improved greatly. The average correct answer rate was less than 50% before support, but it exceeded 60% after the support (Fig. 1). As a result of the evaluation of the vocabulary by PVT-R, evaluation points improved by an average of one point, which is very significant. The result of the mean evaluation point was 11.5, and it was a result to exceed 10 points of normal development (Fig. 2). There was no significance between pre and post-intervention in STRAW scores at control (Table 1).

In counseling to the parents, the experimenter told parents about their child’s cognitive abilities related to reading and writing and explained the mechanisms of any reading and writing disabilities. In one case, for example, a boy had difficulties in reading and phonological awareness. The experimenter in this case explained to the parents the relation between phonological awareness and reading in order to help them understand their child’s difficulties.

The CBCL total scores of almost all participants decreased after intervention (Table 2). The mean total score before the intervention was 40.00 (SD = 21.90), and the mean score after the intervention was 32.50 (SD = 21.62). The two-way repeated-measures analysis of variance revealed a significant difference in total mean scores before and after intervention, p < .005.

The mean internalizing scores of the participating children was 10.10 (SD = 10.39) before the intervention and 9.20 (SD = 10.42) after the intervention. A two-way repeated-measures analysis of variance test revealed that this was not a significant difference. On the other hand,

children’s externalizing scores dropped significantly (average before: 11.00, SD = 8.32, average after: 7.80, SD = 6.43, a two-way repeated-measures analysis of variance, p < .005) .

Regarding the terms of the syndrome scales, aggressive behavior scores decreased significantly, p < .001 (Fig. 3). The scores of four of those children decreased by more than five points. Scores for social problems (p < .005) and attention problems (p < .01) also decreased significantly. The scores for both syndrome scales decreased in seven participating children. Although a decrease in score was seen in seven participating children for withdrawn and thought problems, the difference was not significant. Finally,

there was no significant difference between pre and post-intervention scores in the control group (Table 3).

Problem behaviors in children were reduced by the long-term academic support, as is shown by the syndrome scales. All the children enjoyed learning and the parents achieved a better understanding of their own children. The score decreased in each syndrome scale, especially in “aggressive behavior”, “social problems,” and “attention problems”. All of the children had learning difficulties, which represent one type of developmental disability.

Scores for aggressive behavior decreased most markedly.. It may be helpful to consider some important factors regarding aggressive behavior. Such behavior is not determined by only a single factor; rather, it is intertwined with related factors in a complex manner. It is thought that the children may have had high aggression scores before the intervention due to stress caused by comparing their own performance to that of their classmates at school. Aggressive behavior is strongly related to juvenile delinquency (Dodge, Price, Bachorowski, & Newman, 1990; Nasby, Hayden, & Depaulo, 1980; Steinberg & Dodge, 1983). It is thought that through the present intervention, stress related to learning may have been reduced, and aggressive behavior decreased as a result. It is also possible that the increased understanding between child and parent also contributed to the decrease. Because the parents came to understand the child’s characteristics and needs, children were less likely to get angry at their parents. Children’s performance was also repeatedly praised by the parents and other adults, reducing the child’s stress and aggression levels and increasing their motivation.

Furthermore, aggressiveness may easily lead to direct-attack behavior if impulsiveness is strong (Atkins, Stoff, Osborne, & Brown, 1993; Halperin, Matier, Bedi, Sharma, & Newcorn, 1992). Through this intervention, the risk of a secondary occurrence of juvenile delinquency may have been reduced because attention problems, including impulsiveness and aggressiveness, decreased.

Scores also decreased in secondary “social problems.” Various components, such as “the behavior was childish” and “unable to be friendly,” were evaluated. Through this study, the children gained confidence in learning, and their motivation to learn also increased. Before the intervention, children were not confidant in their own learning abilities, and this may have influenced their behavior and friendships. After the intervention, however, with their increased confidence, children were on a more equal footing with their friends.

The third greatest decrease was in attention problems. In this syndrome scale, various components, such as “restless,” “compulsive,” and “attention does not continue,” were evaluated. Various measures were adopted to encourage children to concentrate during their session at the Cosmos Club. First, the play area and the learning area were separated. Second, only the things that were necessary for the tasks were placed on the desk. Third, the desk faced a wall such that children could not be easily distracted by other things in the room. After a year, children who could not sit still in a chair for 10 minutes before the intervention were successfully able to concentrate for 30 minutes. Although this may not easily transfer to other environments, including their schools, they could continue increasing their attention span at the Cosmos Club. Through their weekly training, they might have learned to concentrate in other environments as well.

Externalizing scores decreased more markedly than internalizing scores. From these results, it may be said that the learning intervention of this study was especially effective against externalizing. It is also possible that improvements in internalizing behavior would be seen if the intervention continued for more than a year.

The average total score was 40.00 before the support and 32.50 afterwards. Thus, total problem behaviors significantly decreased after the support in comparison with before the support. However, this score is much higher than those of children with unimpaired development. According to a study by Itani et al. (2001), the average score of normally developing boys was 16.10. It may be said that this score of 32.50 was still a relatively low value. However, the average found by consultants from psychiatric medical offices and by the child psychiatry wards of general hospitals was 47.19. It may therefore be said that this score of 32.50 was a relatively low value compared to those children. Relative to the developmental courses in CBCL reported by Deruiter (2007), the scores in present study decreased significantly.

Several studies have been carried out on changes of the CBCL scores that occur with age in children with developmental disabilities. It is reported that the total average score was 47.19 in boys aged four to eleven and 48.69 for those aged twelve to fifteen (Itani et al., 2001). The scores in girls were 41.37 for those aged four to eleven and 43.81 for those aged twelve to fifteen. Thus, the scores did not change with age in children with developmental disabilities. There was one control in the present study, and that score did not change either.

This means that long-term academic support, including the 10-minute sessions with the parents, positively affected the children with learning difficulties. This positive influence was supported by two major factors. First, long-term support was provided over the course of a one year periodr. Children could continue the learning in a comfortable environment through appropriate learning tasks. Through this support, children could obtain high motivation for learning. It is thought that these changes may have resulted in the observed reduction in problem behaviors. Second, the children’s parents were given an opportunity to better understand their children, and through the reduction of misunderstandings a more comfortable study and living environment for the children was established in the home.