Positive and Negative Effects of Parental Conflicts on Children’s Condition and Behaviour

Literature Review Conflicts are a regular phenomenon in family life. It can be assumed that family-centred experiences are at the basis of the child’s development. Therefore, the analysis of the consequences of parental conflicts on the child is of fundamental importance. The effects of parental conflict behaviours on the child’s condition have been widely researched since the 1980’s. However, the focus of this research has predominantly been on the negative consequences of these behaviours. This includes particularly the finding that parental conflicts can lead to children’s maladjustment, which in turn results in negative effects on social, cognitive, educational and psycho-biological functions (Cummings & Davies, 2002). In this literature review, a more differentiated approach to the topic is adopted, which encompasses both positive and negative effects of parental conflict behaviours on the child’s condition and behaviour. A parental conflict is considered a disagreement that leads to a greater or lesser interaction of the parents. This interaction can vary from very positive to very negative (Goeke-Morey, Cummings, & Papp, 2007; Cummings, Goeke-Morey, & Papp, 2003). An extensive analysis of previous research in the field has drawn a distinction between constructive and destructive conflict styles, which shows that parents can adopt different conflict behaviour patterns. At the same time, all children are exposed to familial conflicts, but the majority do not display clinically significant maladjustment (Fincham, 1994). Consequently, two assumptions can be formulated: Firstly, diverse forms of parental conflicts have different effects, some being positive and some being negative (Cummings & Davies, 2002). Hence, the present literature review is based on one research question: Which positive and negative effects do constructive and destructive parental conflict behaviours have on a child’s condition and behaviour? Secondly, moderators and mediators play a major role in this relationship. Therefore, possible moderators as well as mediators will be investigated as the second phase of research of this review. In the second section of this review, the two theories on the effects of parental conflicts – the cognitive-contextual framework of Grych and Fincham (1990) and the emotional security hypothesis developed by Davies and Cummings (1994) – are explained as they comprise the cognitive, emotional and behavioural aspects of the child’s response to parental conflicts. Furthermore, the third section is about the empirical findings on this topic. Next, the fourth section comprises of an exhaustive discussion of studies relevant to the present research question, alongside with some contradictory results, in order to to integrate these into a theoretical framework. Possible methodological limitations are also discussed and practical implications are derived. And, lastly, both a conclusion and an outlook is provided on the directions of potential research in the future.

According to the authors, the child perceives a parental conflict as a stressor. Thence, based on this perception, a primary and subsequently secondary processing stage are generated, which in turn influences the child's coping behaviour. These two processing stages are influenced by contextual factors. The primary processing influences the child's affect, which interacts with both secondary processing and coping behaviours. The children's affects are influenced by their observation of the parental conflicts, by their attributional styles as well as by their behaviour, and these in turn influence the cognitive elaboration process and the child's behaviour (Grych & Fincham, 1990). Furthermore, the child's coping behaviour impacts the parental conflict. Delving deeper, the individual components of the model are explained next.
Intensity, content, duration and resolution are among the most important characteristics of parental conflicts. Grych and Fincham (1990) suggest that the effects of these characteristics are cumulative. Intense, prolonged, unsolved conflicts and those affecting the children, the family or the marriage are very stressful for children.
Regarding the context, Grych and Fincham (1990) refer to the psychological aspects of children and make a perspicuous distinction between proximal and distal contextual factors; distal contextual factors are stable (e.g. gender) or relatively stable (e.g. the perceived emotional climate). Contrastingly, proximal contextual factors indicate the child's thoughts and feelings immediately before the conflict. The distal contextual factors include the child's past experiences with conflicts, the perceived emotional climate, the child's temperament and gender. The perceived emotional climate refers to the child's perception of family relationships, especially of the quality of the parent-child relationship. A good climate can act as a buffer against various stressors. Moreover, the child's temperament can have an impact on the relationship between the parental conflict and the child's reaction in three different ways: firstly, some children respond more readily than others to stressors; secondly, their temperament affects their coping behaviour and thirdly, it influences the development of the parent-child relationship. According to Grych and Fincham (1990), gender influences the child's emotions and processing style, due to different socialisation experiences, but not their behaviour. On the other hand, proximal contextual factors include expectations of how the current conflict will develop, as well as the child's current mood. Children have expectations based on their previous experiences and on the characteristics of the current situation. A positive mood can reduce stress. Contrarily, in a negative mood, a stressor is perceived as more negative and previous negative experience is better remembered than in a positive mood (Grych & Fincham, 1990).
The model distinguishes two processes in the child's processing: in the primary processing stage, the child takes notice of the conflict and perceives its characteristics, threats and significance for himself/herself. This leads to an emotional appraisal and reaction. If the child considers the conflict as negative, important or relevant for himself/herself, the secondary processing stage of the conflict is activated, in which the child tries to understand the reasons for the conflict. The child makes a causal attribution, an attribution of responsibility and blame, and develops an efficacy expectation. Secondary processing stage presupposes advanced cognitive skills and is thus differentiated as a function of the child's developmental level. It is assumed that very young children pass only through the primary processing stage and that depending on the kind of cognitive processing, the child relies on a certain coping strategy.
Concerning various attribution and the efficacy expectation of the children, stable, internal and global casual attributions are particularly stressful for children. Attributions of responsibility and guilt trigger different emotions and change with age. For instance, younger children increasingly tend to blame themselves and therefore feel negative emotions (Grych & Fincham, 1990). According to Bandura (1982), the self-efficacy expectation indicates the subjective assessment of own skills to deal with daily difficulties and obstacles in order to successfully perform desired actions. This expectation is also age-dependent. On the one hand, younger children often exhibit "magical thinking", such as the belief that they could stop their parents' conflicts with an unusual power. Conversely, older children have more realistic estimates and expectations as well as better skills to solve conflicts (Grych & Fincham, 1990).
Concerning the coping behaviour, developed by the child based on the secondary processing stage, it is crucial to denote that Folkman and Lazarus (1980) propound there being two forms of coping behaviour which can reduce emotional arousal: 1) emotion-focused strategies through which people can regulate their feelings and 2) problem-focused strategies which are based on changing the stressful situation. Younger children are assumed to have fewer coping strategies than older ones, who have the ability to cognitively reconstruct a stressful situation and to apply emotion-focused strategies.
The Emotional Security Hypothesis of Davies and Cummings (1994) Davies and Cummings (1994) developed their emotional security hypothesis to complement the cognitive-contextual framework of Grych and Fincham (1990). While affect plays a rather secondary role in the cognitive-contextual framework, it is of great importance in the emotional security hypothesis. As reported by Bowlby (1973), emotional security is the confidence in the availability of attachment figures if an individual desires it, which influences the susceptibility to fear. This confidence is postulated to be built during the years of immaturity based on the person's experience regarding the accessibility and responsiveness of attachment figures. Thus, the emotional security hypothesis is the child's perception of the relationship between the child's parents. Emotional security is influenced by past experience with parental conflicts, the emotional bond between parents and children as well as by the quality of the marital relationship, and it is orientated towards future responses to parental conflicts. Depending on how the child perceives and handles the parental conflict, the child's emotional security increases or decreases, which in turn affects the child's well-being. So, in this way, emotional security plays the role of a mediator between the parental conflict and the child's condition. Emotional security affects the child's functional ability in three different ways: by the child's regulation of his/her emotional arousal, by his/her attempt to regulate his/her parents' emotions and by internal representations (Davies & Cummings, 1994).
Withal, there are three hypotheses about the role of the child's regulation of his/her emotional arousal: 1) a high level of emotional arousal depletes the psychological resources needed for an effective emotional regulation (Dodge, 1991;Fabes & Eisenberg, 1992). 2) emotional arousal triggers an energy that affects the child's functional ability. 3) positive and negative emotions have different effects on children's reactions to parental conflicts (Davies & Cummings, 1994). Children with negative emotions suffer more, have a reduced self-regulation capacity and judge the parental conflicts more negatively, whereas children with positive emotions confront future conflicts with a more optimistic attitude. Even if the child's efforts to regulate his parents' emotions lead to a reduction of negative emotions in the short-term, they can have long-term negative consequences for all people involved. Instead of resolving their conflict, the parents' attention is drawn to less important current educational issues and the children adopt a poor coping strategy. The more their emotional security is threatened, the more the children intervene. The internal representations, generated in the mind as internal ideas, are formed from long-term childhood experience with parental conflicts, which have an impact on the child's behaviour. Representations, which have both emotional and cognitive consequences, are activated in parental conflicts. Children can learn to use destructive and constructive approaches, based on their internal representations (Davies & Cummings, 1994).
Different factors moderate the effects of parental conflicts on the children. Firstly, the child's experience of whether his/her parents solved their conflicts in the past or not influences the effect of parental conflicts on the child, the child's emotional reaction to these, and his/ her emotional security. The resolution of a conflict (or better the extent of a conflict resolution) should be considered as a continuum between complete resolution and no resolution (Cummings, Ballard, El-Sheikh, & Lake, 1991). Additionally, it seems that the form of the conflict influences the effect of the frequency. Intensive, violent, unsolved and threatening conflicts, as well as those which revolve around the child, reduce the child's emotional security. Secondly, repeated conflicts reduce the child's emotional security, which in turn decreases his/her ability to regulate his/her emotional arousal and increases the proportion of negative emotions, leading to adjustment problems. Thirdly, the reactions change with age, as the experience with parental conflicts increases, and can vary according to gender due to a different socialisation experience. In addition, a difficult temperament of the children leads to a greater response to negative events on their behalf, coupled with a less positive perception. The parents' behaviour and the child's temperament influence each other reciprocally which adds a further layer of complexity to the overall picture (Davies & Cummings, 1994).

Definition of Constructive and Destructive Conflict Behaviours and Theoretical Considerations
A further focus is conducted here to integrate the two theoretical frameworks together, elucidate the definition of constructive and destructive conflict behaviours and develop certain hypotheses about their effects on children. The nature of the effects of parental conflicts on the child's condition is influenced, among others, by the way in which the conflict is exerted. For this reason, constructive and destructive conflict behaviours can be distinguished based on the effects of the conflict on the child's reaction, which includes behavioural, emotional and cognitive responses. There is a lack of an explicit theory, but the emotional security hypothesis provides a conceptional basis for this distinction from the child's perspective (Cummings & Davies, 2002). In line with Goeke-Morey (1999), different conflict behaviours can be distinguished on the basis of emotional security: if a conflict behaviour provokes more negative than positive emotional reactions, it is classified as destructive, as it weakens emotional security. Contrastingly, a conflict behaviour is classified as constructive if it provokes more positive than negative emotional responses, which, in turn, leads to an increased emotional security. Thus, conflict behaviours can be understood as a continuum which goes from constructive to destructive, depending on how the conflict influences the emotional security of the child (Goeke-Morey, Cummings, Harold & Shelton, 2003). Destructive conflict styles include behaviours such as verbal or corporal aggression towards the partner or objects; hostility, violence and behaviour patterns which threaten the integrity of the family. In contrast, constructive conflict styles are characterised by progress in finding a resolution and by the parents' explanations to the child (Cummings & Davies, 2002).
It is assumed that constructive conflict styles lead to optimism toward future conflict styles. On the contrary, destructive conflict styles could lead to a reduced selfregulation and to negative judgements. The more the emotional security is threatened, the more children intervene in a parental conflict (Davies & Cummings, 1994). Therefore, the more constructive a conflict is, the less the emotional security of the child should be threatened; therefore, children should intervene less in constructive parental conflicts. In their cognitive-contextual framework, Grych and Fincham (1990) discuss the importance of distal contextual factors and of the child's expectations of how the current conflict will develop (Grych & Fincham, 1990). The child's experience of constructive parental conflicts increases their emotional security and therefore, according to this theory, children expect constructive conflict behaviours of their parents also in the future. Moreover, the child's temperament, gender as well as age have to be considered as moderators. The child's temperament may influence the relationship between the parental conflict and the child's reaction in relation to several factors (Grych & Fincham, 1990): 1) in relation to the different conflict behaviours, it can be assumed that more temperamental children react more negatively than others to destructive conflicts than to constructive ones, since they respond more to negative stressors and have a less positive perception of a parental conflict (Davies & Cummings, 1994: Grych & Fincham, 1990 and 2) as temperamental children respond more to negative inputs and have different coping behaviours, they are expected to show less positive behaviours. Boys and girls experience different types of socialisation, which might influence their affective and behavioural responses to parental conflicts (Grych & Fincham, 1990). Boys react in a more aggressive way and with more withdrawal to conflict situations than girls, who are more prosocial than boys (Lindeman, Harakka & Keltikangas-Järvinen, 1997). There might also be an interaction between gender and conflict behaviour style, in a way that gender makes the differences between constructive and destructive conflict behaviour more extreme. Only older children are assumed to reach the secondary processing level, as it requires advanced cognitive skills (Grych & Fincham, 1990). This age difference could make the effects of destructive conflict behaviours even worse: Younger children are expected to react with more negative emotions, as they do not try to understand the reasons of the conflict. Causal attributions, which are variable, external and specific, should not have a negative but instead a positive effect on the reaction of older children. The cognitive-contextual framework presupposes that the secondary processing of the conflict is activated if the child considers it as negative, important or relevant to himself (Grych & Fincham, 1990). Based on this presupposition, children should not process calm discussions, uninteresting topics and positive interactions. However, it is questionable whether this is the case, as according to the social learning theory of Bandura (1977) children learn from and imitate their parents, which is not limited to negative situations. If children ascribe to themselves a high level of self-efficacy, it should influence their reactions positively. According to the theory of Grych and Fincham (1990), it can be assumed that the child's affect is different depending on the style of conflict, on their expectations of the conflict's outcome and on the nature of the attribution. Moreover, age differences with respect to coping behaviour are to be expected (Grych & Fincham, 1990). Older children are assumed to be able to apply more emotionfocused strategies than younger ones (Compas, Malcarne, & Fondacaro, 1988). In contrast, younger children are expected to intervene more than older ones, which means that they show more problem-focused coping behaviours.
This is further annotated in the modelling theory of Zimet and Jacob (2001), which acclaims that children learn basic behaviour patterns in the family, especially from the parents, as the parent-child relationship is intense and long-lasting. Specifically, parental conflict behaviour can serve as models, which are used to guide the child's development of social behaviour (Zimet & Jacob, 2001). The reason is that their internal representations are activated during parental conflicts, which are the basis for learning destructive as well as constructive conflict behaviour (Davies & Cummings, 1994). Therefore, it must be assumed that constructive parental conflict styles lead to constructive problem-solving approaches, good ability to regulate their emotions, good comprehension of other people's emotions and positive social behaviour of the child. Similarly, destructive conflict styles are expected to lead to destructive problem-solving approaches and to negative social behaviour (Davies & Cummings, 1994). Eisenberg and Fabes (1994), Dunn and Brown (1994) and Denham (1998) found supportive results: the parents' support and acceptance of the child's emotions were related to the child's ability to regulate his/her own emotions in a constructive way. The parents' willingness to discuss their emotions with their children was related to the child's attention and understanding of others' emotions.

Empirical Findings
Various studies have revealed that destructive and constructive conflict styles have different consequences on the child. Specifically, the more destructive the conflict behaviour is, the more negative its effects on the child's condition and behaviour. In Davies, Myers, Cummings and Heindel's (1999) study, children watched videos of conflicts and were witnesses of real conflicts afterwards. This study showed that the experience of destructive conflict behaviour provokes more negative reactions than exposure to constructive conflict behaviour. Goeke-Morey et al. (2003) showed videos to children (aged between 8 and 16 years old) and asked them how they would react if their parents argued in this manner. The children indicated that they would intervene or avoid a destructive conflict more than a constructive conflict. Correspondingly, they would react in a more negative emotional way and have a more negative expectation of the conflict's end in a destructive than in a constructive conflict situation. Cummings, Goeke-Morey and Papp (2003;2004) used the diary method and questionnaires to measure the parental conflict behaviour. In the former study of Cummings et al. (2003), both the mothers and fathers filled out the questionnaires, whereas in the latter, only the mothers filled them out. These authors found destructive conflict behaviour to be related to more negative emotional reactions and to lead to more aggressive behaviour than a constructive conflict style. Additionally, if the conflict was about the child itself or the marriage, the child's behaviour was more aggressive 2004). Goodman, Barfoot, Frye and Belli (1999) let parents fill out questionnaires and interviewed children. They found that the more frequent and the more aggressive the parental conflicts were, the less effective the problem-solving strategies of the children tended to be. Using the diary method and questionnaires administered to the parents, Cummings, Goeke-Morey, Papp and Dukewich (2002) found that parents' negative emotions and destructive conflict behaviours were associated with the child's insecure emotional and behavioural responses. According to McCoy, Cummings and Davies' (2009) study, in which the authors administered questionnaires to parents and observed the children's reactions to their parents' discussion, destructive conflict behaviours correlate negatively with emotional security. Davies, Martin and Cicchetti (2012) found that a low level of emotional security acted as a mediator between destructive parental conflict and children's psychological problems. The authors used questionnaires to adolescents and their primary caregivers in their first study and questionnaires and observations in their second study. In a study conducted by Du Rocher Schudlich, White, Fleischhauer and Fitzgerald (2011), parents filled out questionnaires and the researchers analysed the parents' and the children's reactions during a parental interaction. In line with other studies on the effects of parental conflicts on children, these authors also investigated children's reactions to depressive parental conflict styles. A depressive conflict behaviour is a behavioural pattern, in which parents do not argue with each other openly, but instead ignore and backstab each other. The child's reactions were negative and children were found to be more involved in the conflicts when the conflict behaviour was either destructive or depressive (Du Rocher Schudlich et al., 2011).
Nonetheless, at the other end of the spectrum there is research which suggests that the more constructive the conflict behaviour is, the more positive its effects on the child's response are. In Lopez Larrosa, Escudero and Cummings' (2009) study, parents had to fill out questionnaires and discuss different situations at home in the presence of the child. In the case of a constructive conflict behaviour, the children's intervention was minimal and they continued their own activities. Additionally, the child's reaction to calm discussions is similar to their reaction to constructive conflict behaviour  and there is also a significant relationship between constructive conflict behaviour and better problem-solving skills (Goodman et al., 1999). Parents' constructive conflict behaviour correlates positively with the child's emotional security, which in turn is positively related to the child's prosocial behaviour (McCoy et al., 2009). This finding is in line with the results of Davies et al. (2012), which did not show emotional insecurity as a mediator between constructive parental conflict and children's psychological problems. Positive emotions and constructive conflict behaviour of the parents have been shown to be related to secure and more positive emotional and behavioural responses of the child Cummings et al., 2003). Du Rocher Schudlich et al. (2011) found that in a constructive conflict situation, the infants intervened less in the parental conflict, showed fewer negative emotional reactions and played more (Du Rocher Schudlich et al., 2011). However, the results of David's (2009) study are different from the ones of the above-mentioned studies: No relation between positive marital qualities -which included positive emotional observations, constructive conflict behaviours and marital satisfaction of the mother -and positive relationships of children with their peers were found when observing the child's interactions and analysing questionnaires submitted to the mothers and teachers.
With regards to possible moderators and mediators, the child's age, gender, temperament, their past experiences with conflicts (parental conflicts and conflicts between strangers), as well as the extent of a conflict resolution, the gender of the parents, parenting practices and psychological control were investigated. The results showed that all of the above-mentioned factors moderate or mediate the relationship between parental conflicts and children's reactions. Cummings et al. (2003) found the child's gender and age to be moderators, but did not specify the nature of their influence. Easterbrooks, Cummings and Emde (1994), using observation and questionnaires, revealed that the child's reaction changes with age: older children showed more positive behaviours if their parents were in harmony. Furthermore, sadness and the motivation to flee have been found to decline with age (Davies et al., 1999). If the children were exposed to destructive conflict styles, the older ones showed less fear, sadness and motivation to flee than the younger ones. Additionally, Davies et al. (1999) found that the motivation to intervene increases with age and peaks during adolescence. These findings are in line with those of Cummings, Zahn-Waxler and Radke-Yarrow (1984), who questioned the mothers of toddlers and primary school children, and with those of Cummings, Vogel, Cummings and El-Sheikh (1989), who showed videos of angry or friendly interactions to children. These previous studies show that the nature and the intensity of the feelings experienced by children change with age: toddlers respond to parental conflicts with crying and anger (Cummings et al., 1984), six-to seven-year-old children with anger, and eight-to nine-year-old children with worry (Cummings et al., 1989). There is also a gender difference: girls responded with more fear and anger than boys (Davies et al., 1999). Howbeit, impeding the case is an interaction between the child's age and gender: sixyear-old boys and ten-year-old girls were mostly motivated to go away, whereas ten-year-old boys wanted to intervene more than the others. In contrast to these studies, Goeke-Morey et al. (2007) did not find the child's age and gender to be moderators in this process. In their study, parents used the diary method and filled out questionnaires. Their children received a description of a parental conflict situation, watched videos with various conflicts' endings and were interviewed. David (2009) andEasterbrooks et al. (1994) found that, depending on the child's temperament, the children's reactions differed: more temperamental children were more reactive and showed less positive behaviours, less anger and intervened more, whereas children with a higher level of effortful control benefited more from positive marital qualities (Easterbrooks et al., 1994). According to Du Rocher Schudlich et al.'s (2011) study, past experiences with parental conflicts and a child's temperament moderate the relationship between destructive parental conflict behaviour and the child's reaction. Destructive conflict behaviour and negative emotional reactions of the child are only significantly related if the children have often experienced parental conflicts or have a less reactive temperament. The extent of a resolution diminishes the negative effects of destructive conflict behaviours (Goeke-Morey et al., 2007). In other words, the more the conflict is resolved, the more positive the emotional reaction of the child was. There are also differences according to an interaction of specific emotions and the parents' gender. Cummings and Davies (2002) found that specific emotions of the father or the mother acted as a moderator in destructive behaviours. More precisely, the child reacted with negative emotions to the father's anger or to the mother's sadness. Goeke-Morey et al. (2003) have revealed that the father's aggression towards the mother and the mother's threats to leave the family were particularly stressful. The reason behind the different influences stemming from the interaction of specific emotions and the parents' gender remains unclear. In the study of Coln, Jordan and Mercer (2013) mothers completed questionnaires. The results showed that parenting practices and psychological control are mediators of the relation between constructive and destructive parental conflict behaviour and child's internalizing and externalizing problems. Specifically, there were direct paths from destructive parental conflict behaviour to negative parenting practices, psychological control, and child's internalizing and externalizing problems (Coln et al., 2013).

Discussion
The research question of this review was: "Which positive and negative effects do constructive and destructive parental conflict behaviours have on the child's condition and behaviour?". The studies presented confirmed that more constructive conflict behaviours have more positive effects on the children, which include positive emotional consequences (e.g. more positive emotional reactions and increased emotional security) and behavioural outcomes (e.g. more prosocial behaviours and less aggressiveness). Correspondingly, more destructive conflict behaviours have been found to result in more negative effects on the children, which have emotional implications (e.g. more negative emotional reactions and decreased emotional security) and behavioural effects (e.g. more aggressiveness and less effective problem-solving strategies).
Moreover, in the present review, the impact of possible moderators and mediators was addressed. In this respect, it was shown that the child's age, gender, temperament and past experiences with conflicts (with parental ones as well as with conflicts between strangers), as well as the extent of a conflict resolution and the gender of the parents moderate the relationship between parental conflicts and children's responses to these. Additionally, parenting practices and psychological control mediate this relationship.
Although the results of the studies generally present a consistent picture, there are some contradictory results. While the results of Goeke-Morey et al.'s (2003) study showed that calm discussions were neither classified as constructive nor destructive in terms of their emotional impact on the child, Cummings et al. (2003) found a positive relationship between calm discussions and childhood gaiety, which leads to fewer adjustment problems. This distinction could be due to the use of different methodologies: While Goeke-  directly examined children, who were watching short films about everyday conflicts of foreign couples, in the study of Cummings et al. (2003), parents used the diary method and filled out questionnaires. No relation between positive marital qualities and positive relations of children with their peers was found (David, 2009), which is contradictory to the theoretical considerations of this review and to the results obtained by both Goodman et al. (1999) and McCoy et al. (2009), showing that children learn behaviour patterns from their parents. Another unclear result of the study conducted by David (2009) is the negative relationship between parental qualities and boys' relationships with their peers. Interestingly, in the study of Goodman et al. (1999), no significant relationship between children's perceptions of parental conflicts and their ability to solve problems was found. As such, the results obtained by the studies of both David (2009) andGoodman et al. (1999) need to be replicated and further research is necessary in order to draw conclusions.
The results of the studies presented in this literature review are consistent with the emotional security hypothesis: there are differences in the emotions and in the intervening behaviour, depending on how destructive or constructive the parental conflict behaviour is, because constructive conflict behaviours increase the emotional security of children (Davies et al., 2012;Davies et al., 1999;Du Rocher Schudlich et al., 2011;Goeke-Morey et al., 2003;Goeke-Morey et al., 2007;Lopez Larrosa et al., 2009;McCoy et al., 2009). Regarding the considerations on the cognitive-contextual framework of Grych and Fincham (1990), we can deduce the following: depending on their experience with conflicts, children expect similar behaviour patterns in the future . The emotional reaction to parental conflicts changes with age (Easterbrooks et al., 1994;Davies et al., 1999;Cummings et al., 1984;Cummings et al., 1989). Contrary to the considerations that younger children should show more problem-focused coping behaviours, the results on age-dependent intervening behaviour demonstrate that the motivation to intervene increases with age and peaks in adolescence (Davies et al., 1999). An explanation for it could be that children acquire more experience of conflicts with increasing age and, therefore, they could develop alternative behaviour patterns to those of their parents. Therefore, long-term effects are also to be researched. It remains to be investigated whether calm discussions lead to a more elaborate processing of the conflict by the children and how their attributions influence children's reactions. Apart from the results obtained in the study of David (2009), other studies tend to show that children seem to learn both constructive and destructive conflict behaviours from their parents. This common result is in line with the theoretical considerations of Zimet and Jacob (2001).
In the studies examined, different methods have been used: the diary method with parents, questionnaires and surveys of children and observations of parent-child interactions. The advantage of the diary methods is the acquisition of current parental conflicts in their natural environments. However, this method also presents some disadvantages. On the one hand, what is directly measured is not the child's reactions, but instead their parental perceptions, which could differ from the child's actual reactions. On the other hand, the results obtained using the diary method can be biased by the parents' sensitisation to this topic through the training, individual interpretations, a person's definition of a conflict, as well as the motivation and self-discipline of the parents. Another method consists of questionnaires given to children's adult attachment figures. By doing so, the children's reactions can be indirectly recorded, but again this depends on the accurate perception by their parents. However, this is an economical method, as it is time-and cost-saving. Another method commonly used consists in presenting conflict situations of strangers to children and assessing their reactions to these, thus documenting the child's perception and response directly. This method is ethically acceptable because children are less likely to suffer from conflict situations of strangers than from conflict situations of their parents. However, the generalisability of conflicts between strangers to parental conflicts is limited, as parental conflicts represent a more emotional issue for children (Davies et al., 1999). All the same, the child's reactions to conflicts between strangers are assumed to represent the same response patterns which would have been provoked in a parental conflict situation (Cummings & Davies, 2002). An alternative method is the instruction to imagine those styles of conflict with the child's own parents, which requires much imagination and is abstract instead of real. Interviews of children presuppose awareness of one's own feelings and thoughts as well as the language skills to verbalise them. This could lead children to respond in a socially-desirable way. Due to these disadvantages, interviews of children could have been problematic in the study by Davies et al. (1999). It could have led to the result that only adolescents, but not the younger children, differentiated between destructive and constructive conflict styles in their assessment of future conflicts. In contrast to this assumption are the results of the study conducted by Herrera and Dunn (1997), where no correlation between constructive conflict behaviours and children's emotional understanding and language skills was found. Whether this would be true also with respect to parental conflicts, is yet to be investigated. The method of using observations of parental interactions and children's responses to these as well as of children's interactions with peers makes it possible to register and directly encode the nuances of both reactions and interactions. There is a range of situations in which observations take place. Parents' debates in the laboratory are, depending on the setting, a bit artificial and can influence the conflict behaviour of the parents. If ever the same topics are presented for discussion, as conducted in the study by Lopez Larrosa et al. (2009), they can be compared well, but might be irrelevant for the couples, thus not evoking natural and real emotions. However, interactions in natural settings are related to additional organisational expenses. The combination of different methods compensates for the disadvantages of each method and thus this strategy has been used by most authors (e.g. Cummings & Davies, 2002;Cummings et al., 2003;Cummings et al., 2004;David, 2009;Davies et al., 2012;Du Rocher Schudlich et al., 2011Easterbrooks et al., 1994Goeke-Morey et al., 2007;Goodman et al., 1999;Lopez Larrosa et al., 2009;McCoy et al., 2009).
Conflicts between parents are an everyday phenomenon in family life and there are some practical implications which can be identified from this review. The different effects of conflicts, depending on their way of being dealt with, lead to different practical implications. It is supposed that the more destructive a conflict is, the more necessary a change in the conflict style would be, specifically if it is accompanied by hostility, corporal aggression towards the partner or objects, and violence or behaviour patterns which threaten the integrity of the family. Intervention and prevention could have various emphases, such as a focus on the arguing parents, their child or the relationship between these. The goal while focusing on the parents could be to shift their argumentative style from more destructive to more constructive. This would include working on finding a resolution and building good communication skills. If the focus is on the child's reaction, his/her individual characteristics should be taken into account, such as gender, age, temperament and previous experiences with conflicts. Negative emotions, thoughts and behavioural reactions as well as their behaviour in interaction with peers should be addressed. Concerning the relationship between the arguing parents and their child, the focus of the intervention should be on the preservation of the child's emotional security. Parents could explain to the child the reason and the result of a conflict, so that the child understands it better and does not make an erroneous attribution. They can also talk to their children about their perceptions and emotions, as this contributes to both a clarification and development of children's language skills (Herrera & Dunn, 1997). The child could give his/her parents feedback with regards to how he/she has experienced their conflict and what kind of thoughts he/she has. This could bring to mind the points to improve. However, all these practical implications are dependent on the motivation of the parents and of the child.
To conclude, parental conflict behaviours can vary from very destructive to very constructive, just as the extent of the conflict's resolution, which constitutes a bandwidth from none to complete resolution. Correspondingly, the impacts of parental conflicts on the emotions and behaviours of children are considered to be a continuum from negative to positive. Both the cognitive-contextual framework of Grych and Fincham (1990) and the emotional security hypothesis of Davies and Cummings (1994) serve as theoretical frameworks for the searching and the interpretation of related studies. Since the former theory focuses on the cognitive and behavioural aspects of the effects of parental conflicts on the child's condition and behaviour and the latter theory on the emotional ones, they complement each other. The results of the studies presented in this literature review generally support these two theories: parental conflicts, depending on the conflict style, can either increase or weaken children's emotional security and children learn conflict behaviours from the parental model. There are several variables that moderate or mediate this relationship. The question of how to measure the effects of parental conflicts is an ethical, economical and scientific challenge, because the various methods present different advantages and disadvantages. Therefore, a combination of various methods is recommended. Future research should focus, on the one hand, on the replication of findings of a few studies; and on the other hand, on whether the different aspects of parents' conflict behaviours comprise individual or cumulative effects or not. In addition to the research about the impacts of constructive and destructive conflict behaviours, the effects of the depressive conflict behaviour are also to be investigated. Furthermore, long-term effects of parental conflicts on children's condition and behaviour could be examined. Foreasmuchas, the research of the effects of parental conflicts on the child and its practical implications will remain of central importance and topical in the future.