Conflict resolution abilities in children with Specific Language Impairment

Purpose: To investigate the conflict resolution abilities of children with Specific Language Impairment, and to verify whether the time of speech-language therapy correlates to the performance on the conflict resolution task. Methods: Participants included 20 children with Specific Language Impairment (Research Group) and 40 children with normal language development (Control Group), with ages ranging from 7 years to 8 years and 11 months. To assess the conflict resolution abilities, five hypothetical contexts of conflict were presented. The strategies used by the children were classified and scored by the following levels: level 0 (solutions that do not match the other levels), level 1 (physical solutions), level 2 (unilateral solutions), level 3 (cooperative solutions), and level 4 (mutual solutions). Results: Statistical analysis showed group effect for the variable total score. There was a difference between the groups for modal development level, with higher level of modal development observed in the Control Group. There was no correlation between the period of speech-language therapy attendance and the total score. Conclusion: Children with Specific Language Impairment present difficulties in solving problems, in view of the fact that they mainly use physical and unilateral strategies. There was no correlation between the time of speech-language therapy and performance in the task.

The strategies used by the children were classified and scored by the following levels: level 0 (solutions that do not match the other levels), level 1 (physical solutions), level 2 (unilateral solutions), level 3 (cooperative solutions), and level 4 (mutual solutions).Results: Statistical analysis showed group effect for the variable total score.There was a difference between the groups for modal development level, with higher level of modal development observed in the Control Group.There was no correlation between the period of speech-language therapy attendance and the total score.Conclusion: Children with Specific Language Impairment present difficulties in solving problems, in view of the fact that they mainly use physical and unilateral strategies.There was no correlation between the time of speech-language therapy and performance in the task.
The RG was made up of 20 children (9 girls and 11 boys) with a diagnosis of SLI and ten children by age group (7 and 8 years).All the children of the RG were present in speech attendance, on weekly basis, treated in the speech research laboratory in language development and its amendments ( LIF-DLA) of Speech-Language course of Faculdade de Medicina da Universidade de São Paulo at the time of data collection.The following inclusion criteria were considered: to have diagnosis of SLI according to the international criteria (primary language change and normal cognitive development), and presenting fairly intelligible speech and oral language, so that the responses could be faithfully transcribed.
The CG was composed of 40 children without language changes (18 girls and 22 boys) and 20 children per age group (7 and 8 years).The CG was recruited in two public elementary schools in social-economic field, a similar area of the city of São Paulo.The inclusion criterion for the CG was to present linguistic performance within the normal range.The language assessment for inclusion of the participants in the CG included the following: phonology, phonological awareness, reading, and writing (8)(9)(10)(11) .

Application of Conflict Resolution Test
The children selected for the survey were evaluated.This evaluation included the Conflict Resolution Test (2) ( Appendix 1).For the children of the GC, the test was applied at school, in a quiet and convenient location, where the evaluator was able to apply the test in each child individually.The test was applied at LIF-DLA to the GP, before the therapy of the children.The test lasted for approximately 15 minutes.
The five stories that make up the Conflict Resolution Test (2) , composed of three sentences each, which were translated (translation-back translation) and only the name of the characters were replaced by frequent names in Brazilian Portuguese language.
To facilitate story comprehension, 15 figures of hypothetical situations of the conflict were designed.Three figures were elaborated for each story, with each one representing a phrase.
Initially, the children received the following statement:

INTRODUCTION
The specific language impairment (SLI) is diagnosed when there is a delay of language that does not occur due to anyother condition (1) .Several studies report social cognition deficits in SLI (2)(3)(4) .
Social cognition involves the understanding of the social world, including understanding of knowledge, thinking, thoughts, emotions, and other's point of views, in addition to the knowledge about roles and social relationships.A type of social interaction that characterizes the cognitive development is conflict resolution.To resolve conflicts it is necessary to understand other's point of view, as well as develop and express strategies to resolve disputes, so that there is a need for language and cognitive abilities (2) .
Language competence is an important prerequisite for the implementation of the different social rules, but it is not enough.The basic social knowledge is also necessary for effectiveness in communication (2,4) .
Deficits in children with SLI are possibly a result of non-specific and general limitations of working memory and processing capacity.These limitations generate difficulty in conversations and can also lead to difficulties in properly understanding the interactions and social situations, as well as the mind and thought of the interlocutors.For all these reasons, children with SLI experience just a few social experiences, which further contributes to the social deficits observed in this population (5) .The difficulty in performing tasks that require skills associated to executive functions and the difficulty of putting the caller in place, probably contribute to the social and cognitive deficits in children with SLI (6,7) .
Conflict resolution skills can provide us with rich and varied information about the social, cognitive, and language skills of children with SLI.Thus, this study aimed to verify the conflict resolution skills of children aged between 7 and 8 years old with SLI and check for correlation between the therapy time and performance on the task of conflict resolution.

Ethical aspects
This research was approved by the Ethics Committee for review of research projects of Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo (USP), under number 060207.All those responsible for children participating in this study, signed an informed consent prior to the conduction of the study.

Casuistry
Sixty children participated in this research (37 girls and 33 boys) with ages ranging from 7 to 8 years and 11 months.The participants were divided into two groups, namely con-For further analysis, the answers were recorded on a digital recorder of Panasonic RR-US380 ® .
The responses were analyzed to determine the presence or absence of the conflict resolution strategies.Then, the answers were organized in 28 strategies showed in the study, on which this work has been based on (2) (Chart 1).Also, the resolution strategies of the stories have been translated into Brazilian Portuguese (translation-back translation) language.
The strategies were grouped into five levels, which form an evolutionary scale of development for conflict resolution skills.The responses of the children were scored according to the proposed levels (Chart 1).
The therapy time for each child was evaluated for the RG, to check if the intervention time was correlated to the speechlanguage score using the Conflict Resolution Test.
representing thesituations of the conflict were showed to the children to facilitate their comprehension.
After presenting each hypothetical situation, the children were instructed to retell the story to check the understanding of events.The children who did not understand the hypothetical situations of the conflict were supposed to be excluded, but all children in both groups showed agood comprehension about the five stories.
Right after the re-count of each story, the following question was asked: "If you were him [evaluate the points to the character of the story] what would you do?"In this case, children should report what the protagonist could do in this particular situation, hence providing just one strategy for solving the problem.
At the end of the test, each child provided five strategies to solve all those different problems presented in the five stories.

Data analysis
A significance level of 0.05 (5%) was set for this work.The following tests were used to perform the statistical analysis: multivariate analysis of variance (MANOVA), Mann-WhitneyU Test, and Pearson's Correlation.

RESULTS
To check the age and group effects in conflict resolution, weights to each level were assigned and the total score obtained by each child was calculated.Table 1 presents the results of the descriptive statistics for the total score per group and by age.Legends: SD = standard deviation calculated for each child by dividing the number of strategies of each level by the total number of strategies used by the child.To determine the differences by age group and between the groups, we used the Mann-Whitney U test.In the CG, 7-year-old children used the level 1 most of the time (M=0.21±0.2),level 2 (M=0.52 0.25), and level 3 (M=0.21±0.24),and 8-year-old children mostly used level 2 (M=0.52±0.25)and level 3 (M=0.4±0.23)(Table 3).There was a difference for level 1 (most commonly used by children of 7 years old; p=0.001) and level 3 (most commonly used by children of 8 years old; p=0.015).
It was observed that the score was higher for the GC and for the children of 8 years old.To verify that such effects are significant, MANOVA was applied.To the variable total score, were observed the effect of group F(1.52)=23.918p<0.001 but not by age F(1.52)=3.875p=0.054.Interaction between the variables was not observed.
The modal development level for each child, in which larger number of strategies were used (Table 2), was also determined.The Mann-Whitney U test was used to determine differences in developmental level and age-group.In this analysis, only difference between both the groups (p<0.001) was verified.It was found that the CG had more modal development level (M=2.30±0.597)as compared with the RG (M=0.45±0.48).
The proportion of the strategies used was also examined in each of the proposed five levels.The results were It is observed that in the RG, children of both 7 and 8 years old, most frequently used level 1 (7 years old: M=0.31±0.36;8 years old: M=0.51±0.27)and level 2 (7 years old: M=0.45±0.34;8 years old: M=0.35±0.23).For this group, only one difference was observed at level 0 (p=0.020), in which children of 7 years old used more strategies than the children of 8 years old (Table 4).
In Table 5, we present the comparison results between the groups by age and level, regarding the proportion of strategies used.There was a difference at level 0 for children aged 7 years, and this difference predominated the strategies at this level for the RG.For the children of 8 years old, there was a difference for level 1 and 3, and that for the CG 3-level strategies predominated and to the RG at 1 level.
Pearson's Correlation was performed to check if there was a correlation between the time of language therapy and the performance of children in conflict resolution.No correlation was found between the length of therapy and the total score in the competition (p=0.961;r=0.012) (Figure 1).

DISCUSSION
It was expected that children with language disorder presented worse performance when compared to children with study shows that children with SLI presents social-cognitive skills development delay and find it difficult to troubleshoot it properly, a result that supports several research studies with this population (2,4,17,18) .
The fact that children with SLI presents significant difficulties in everyday social activities, such as troubleshooting, contributes to their unpopularity with peers of their age, associated to friendships, relationships, and interactions (19,20) .In addition, social and school difficulties experienced by children with SLI for life long, contributes to low self-esteem of this population, which is a characteristic observed in several studies (20)(21)(22) .
Although social-cognitive deficits in children with SLI are quite clear, but it is not yet possible to establish the exact correlation between language and social cognition.It is known that the children having difficulties in communication with language delay experiences interference in social relations and, as a result, there is an aggravation in their social-cognitive deficits.Once the children begin to interact less as compared to their peers with normal language development, there are less chances to apply the social rules (5,23) .
There is a possibility that future research work could also investigate the correlation between mind theory and executive functions -conflict resolution in SLI -because all these skills appear to have a relation with each other.If this correlation is in fact confirmed, it can significantly influence the therapeutic interventions (6,7) .
In both normal language development and SLI, there was no difference between the total score of children of 7 years old, which indicates that children of this age group resolve conflicts in the same manner.It was expected that the performance between the age groups remained constant, because the linguistic and the social-cognitive skills did not significantly develop between these age groups.
The score was not found to be different among the children of 7 years old with normal development who most frequently used level 1 strategies, and children of 8 years old children used mostly level 3 strategies.Probably, the children of 8 years old managed to consistently use cooperative strategies, because in addition to the greater experience with social situations, it is just about at this age that children effectively understands the thoughts as being interpretations, which may vary between individuals and situations.It is also at this time that children acquire the ability to think about multiple possibilities and to answer "what if" questions, which is an important skill for troubleshooting (24,25) .
A study conducted in 2008 showed evidences that the interaction between theory of mind and executive duties continues to develop by the end of adolescence.As there is an important correlation between these skills and the ability to troubleshoot, it is believed that the troubleshooting skills develops considerably after the age of 8 years (26) .
Among children with SLI, the only difference between the levels of responses was at level 0, which was used more by 7-year-old children.It is not possible to say for sure the reason of this difference, but it probably occurred because the children of 8 years old have more experience with social situations as compared with children of 7 years old.normal language development in conflict resolution, due to language delay, processing difficulties, social/cognitive deficits, and the empathy difficulty that are characteristic of SLI (7,(12)(13)(14)(15)(16) .
In fact, children with SLI had a total score significantly lower than children with normal language development using the Conflict Resolution Test Therefore, the findings of this Hence, it is likely that the evolution of social cognition skills and, consequently, the evolution of the ability to solve problems, are associated to individual social experiences in conjunction with the evolution of language and the ability to abstract thinking, both for children with SLI and children with normal language development.
In the Conflict Resolution Test, children with SLI most frequently used unilateral strategies (level 2) and physical solutions (level 1), while children with normal development of language used cooperative strategies (level3), in addition to the strategies of level 1 and level 2. Thus, in addition to the difference in the total score, the children also showed lower level with SLI development modal (RG=1.45;CG=2.30).
This data demonstrates that although the two groups used unilateral strategies (level 2) to solve the problems presented, children of this age group also often use strategies that involve physical intervention (level 1).However, children of 8 years old with normal language development are already able to utilize cooperative strategies (level 3) consistently.
According to the results, children of 8 years old with SLI still often use aggressive strategies (verbal or physical) to solve problems.Unlike the results found in this survey, the majority of studies that evaluated the children with SLI did not only note aggressive behavior, but also behaviors of withdrawal and avoidance (23,27) .It is likely that children with SLI end up using strategies like cursing and hitting for not being able to use conflict resolution strategies with more sophistication and effectiveness.Possibly, over the years, they realize that this kind of behavior is not effective and end up choosing strategies such as to withdraw or avoid each other.
Significant correlation between performance on the task of conflict resolution and language therapy was not observed, probably because the social cognitive skills development involves linguistic and social aspects.Although the speech therapy focuses on the linguistic aspects, it is an artificial context and is not by itself, able to meet the social deficits often observed in the population with SLI.
The results of this research suggest that the deficits observed in the population with socio-cognitive SLI may not arise solely from the language disorder.So, children with SLI would benefit from therapeutic intervention strategies that target the use of language in different social contexts, and increase the opportunities for children to improve social communication skills in real situations.This type of intervention would favor the interactions of these children, because success in social tasks and conflict resolution is attributed to the establishment and maintenance of positive interpersonal relationships.
It is likely that older children and teenagers, uses more sophisticated strategies more frequently, so it would be interesting to increase the research sample, adding more age groups in it.In this manner, it will be possible to draw a detailed profile for the evolution of conflict resolution skills.

CONCLUSION
This study demonstrated that children with SLI between the age of 7 and 8 years face difficulties in solving problems, because they predominantly use physical, non-effective, and unilateral strategies to solve avast majority of conflicts experienced by them.
In general, as for the comparisons studied between the age groups, it was found that the children of 7 and 8 years old (either with normal language development or SLI), uses similar strategies to resolve the conflicts.Similarly, quantitative difference was not observed among them for the task of conflict resolution.
The language disorder was the only variable that determined the performance for conflict resolution, because the RG exhibited lower performance when compared with the CG.In addition, there was no correlation between the performance in the race and language therapy, which suggested that language therapy is not enough to suit the social-cognitive deficit by itself as observed in population with SLI.

Figure 1 .
Figure 1.Correlation between the length of therapy and the total score in the Conflict Resolution Test Chart 1. Conflict resolution strategies for hypothetical situations.

Table 1 .
Description of the total score per group and age Legends: CG = control group; RG = research group; SD = standard deviation

Table 2 . Level of modal development per group and age
* Significant values (p≤0.05)Mann-Whitney U Test Legends: CG = control group; RG: research group; SD: standard deviation

Table 3 .
Proportion of strategies used in accordance with the five levels as proposed by age group for the control group

Table 4 .
Proportion of strategies used according to the five levels proposed per age group for the research group Legends: SD = standard-deviation.

Table 5 .
Inferential statistics to compare the proportion of strategies used by Research and Control groups according to five levels proposed by age group * Significant values (p≤0.05)Mann-Whitney U Test.