How Do Families Represent the Functions of Deliberate Self-Harm? A Comparison between the Social Representations from Adolescents and Their Parents

Research has recognized the importance of understanding the social representations about the functions of deliberate self-harm, particularly in the context of clinical intervention. In addition, parents can play a relevant role in the rehabilitation of adolescents with these behaviors. However, there are few studies that focused on the description and comparison of the social representations about these functions, particularly in families. This article aimed to analyze the social representations about the functions of deliberate self-harm from adolescents and their parents. We developed two sets of analyses: first we compared the social representations from adolescents without a history of deliberate self-harm and their parents, and secondly we compared the social representations about the functions of deliberate self-harm from adolescents with a history of these behaviors and their parents' social representations. Results revealed significant differences between both groups of families, implying that the groups of participants represent the functions of deliberate self-harm differently. Overall, parents emphasized interpersonal functions and devalued intrapersonal functions. These differences were heightened in the families of adolescents with deliberate self-harm. The present article provides important insights regarding the social representations about the functions of deliberate self-harm and the differences between parents' social representations and their children experiences and social representations.

The knowledge about the functions of deliberate self-harm is one of the most important factors in this context, since it can contribute to the understanding of this phenomenon's etiology, as well as to its classification, prevention, and treatment (Klonsky, 2007). Regarding treatment, understanding deliberate self-harm's functions can be an essential factor to select which treatment is most appropriate to each individual according to their experience of these behaviors, as well as to design specific intervention strategies (Bentley, Nock, & Barlow, 2014;Muehlenkamp, 2006;Nock & Prinstein, 2004;Nock, Teper, & Hollander, 2007;Washburn et al., 2012).
Family, specifically parents, have been recognized as an important factor within the context of deliberate self-harm (e.g., Arbuthnott & Lewis, 2015;Hasking, Rees, Martin, & Quigley, 2015;Mojtabai & Olfson, 2008;Santos, 2007). Family seems to occupy a central role in clinical intervention and research suggests that it is necessary to incorporate family therapy into treatments, particularly interventions that work towards strengthening communication and emotional support (Muehlenkamp, Brausch, Quigley, & Whitlock, 2013). In addition, a caring and affectionate family environment, where space for the discussion of these behaviors exist, can favor the adolescent's rehabilitation process (Arbuthnott & Lewis, 2015). Similarly, poor family functioning is related to the presence of deliberate selfharm (Crowell et al., 2008;Kelada, Hasking, & Melvin, 2016) while better family functioning is related to recovery (Kelada et al., 2016).
Understanding the functions of deliberate self-harm is crucial for supportive and effective responses to individuals' disclosures of self-harm (Muehlenkamp et al., 2013). For example, if friends and family members have an inaccurate understanding of these functions (e.g., believing the behavior to be an act of manipulation instead of a form of support-seeking), it may lead to responses that inadvertently aggravate the frequency and severity of the behaviors (Bresin, Sand, & Gordon, 2013). Hence, understanding how family members represent the functions of deliberate self-harm can be a crucial factor to promote clinical interventions and to involve the family in the treatment process.
Research has focused on the risk factors associated with parents, help-seeking from parents, interventions involving parents, and impact on parent well-being (Arbuthnott & Lewis, 2015). Also, several studies explored the views and attitudes of parents of adolescents who self-harm (Ferrey et al., 2016;McDonald, O'Brien, & Jackson, 2007;Oldershaw, Richards, Simic, & Schmidt, 2008;Rissanen, Kylm€ a, & Laukkanen, 2008, Rissanen, Kylm€ a, & Laukkanen, 2009), but did not focus on the representations of these How do families represent the functions of deliberate self-harm? S174 VOLUME 24 NUMBER S1 2020 behaviors' functions. Oldershaw et al. (2008) concluded that parents commonly suspected and spotted self-harm prior to disclosure or service contact, but also concluded that communication difficulties and underestimating significance led to delays in addressing the behavior. The study developed by Ferrey et al. (2016) found that, after the discovery of selfharm, parents described initial feelings of shock, anger and disbelief, and later reactions of stress, anxiety, feelings of guilt, and in some cases the onset or worsening of clinical depression. Also, parents frequently emphasize their difficulties, struggles, and uncertainties in understanding and coping with their child's deliberate self-harm (McDonald et al., 2007;Oldershaw et al., 2008). Regarding the functions of deliberate self-harm, it is known that these behaviors can serve diverse functions that can occur simultaneously (Lloyd-Richardson, 2008;Nock, 2009;Saraff, Trujillo, & Pepper, 2015). According to Klonsky (2007), the most frequently studied functions include: Affect Regulation, Anti-Dissociation, Anti-Suicide, Interpersonal Boundaries, Interpersonal Influence, Self-Punishment, and Sensation-Seeking. Nonetheless, there are also other less common functions, such as Autonomy (Klonsky & Glenn, 2009), Peer Bonding (Klonsky & Glenn, 2009), Revenge (Klonsky, 2007;Rabi, Sulochana, & Pawan, 2017;Rodham, Hawton, & Evans, 2004), or Self-Care (Klonsky & Glenn, 2009).
In order to systematize the many functions of deliberate self-harm, Prinstein (2004, Nock &Prinstein, 2005) proposed the Four Function Model. According to this model, deliberate selfharm serves four primary functions that differ along two dichotomous dimensions: automatic/intrapersonal versus social/ interpersonal, and positive (i.e., followed by the presentation of a favorable stimulus) versus negative (i.e., followed by the removal of an aversive stimulus) (Nock, 2008). Hence, automatic negative functions reduce tension or other negative affective states, while automatic positive functions increase or generate a desirable physiological or affective cognitive state (Kortge, Meade, & Tennant, 2013;Nock & Prinstein, 2004, Nock & Prinstein, 2005. On the other hand, social negative functions allow escape from interpersonal interactions or task demands, while social positive functions contribute to gain attention or access to materials, or to trigger some reaction from others (Kortge et al., 2013;Nock & Prinstein, 2004, Nock & Prinstein, 2005. Recently, research has recognized the importance of the interpersonal functions, although they are less common than the intrapersonal functions (Heath, Ross, Toste, Charlebois, & Nedecheva, 2009;Muehlenkamp et al., 2013).
Social representations are a modality of knowledge that produce and determine behaviors because they define the nature of the stimuli that surround us and the answers we give them (Moscovici, 1961). These representations can be understood as dynamic sets that aim at the production of social behaviors and interactions, and not only as the mere reproduction of these behaviors and interactions as reactions to external stimuli (Sampaio et al., 2012). Hence, social representations are simultaneously a product and a process (e.g., Jodelet, 1984;Valsiner, 2003) that allow us to interpret aspects of reality to further react to them (Wachelke & Camargo, 2007). Therefore, the representations about the functions of deliberate self-harm from adolescents with and without a history of these behaviors and from parents may have important implications for clinical interventions and prevention programs, particularly in terms of social support. There are several limitations to the current knowledge concerning the representations about the functions of deliberate self-harm from adolescents and parents, since most studies focused on the attitudes about deliberate self-harm and relied on samples of adolescents and/or parents that had direct contact with these behaviors. Moreover, we did not find any studies that compared the social representations and experiences from adolescents with and without deliberate self-harm and their parents. The few studies that compared the perspectives about the functions of deliberate self-harm of participants with and without a history of these behaviors focused on the views of college students (Batejan, Swenson, Jarvi, & Muehlenkamp, 2015;Bresin et al., 2013). The study from Batejan et al. (2015) concluded that the groups did not differ in their views of the relevance of intrapersonal functions, although non-injuring participants appeared to stress some interpersonal functions slightly more than individuals with a history of deliberate selfharm did. Furthermore, the study conducted by Bresin et al. (2013) concluded that there was little differentiation among functions between groups.

THE CURRENT STUDY
The objective of the current article focuses on the comparison of the social representations about the functions of deliberate selfharm from families (adolescent, mother, and father) of adolescents with and without deliberate self-harm. We developed two sets of analyses: a) the first one compares the social representations about the functions of deliberate self-harm from adolescents without a history of these behaviors and their parents' social representations; b) the second one compares the functions mentioned by adolescents with a history of deliberate self-harm and their parents' social representations about these functions. Our main goal is to explore the possible differences regarding the social representations about the several functions of these behaviors (such as Affect Regulation, Anti-Dissociation or Interpersonal Influence) and the two dimensions where these functions can be organized (interpersonal and intrapersonal).
Research has shown the global incomprehension of parents regarding the motivations and functions of deliberate self-harm (e.g., McDonald et al., 2007;Oldershaw et al., 2008). Also, a previous study (Batejan et al., 2015) concluded that participants without deliberate self-harm appeared to value some interpersonal functions more than participants with a history of these behaviors did. For the first set of analyses, we hypothesize that there will be no significant differences between adults and adolescents concerning the interpersonal dimension, and that significant differences will emerge in the intrapersonal dimension, where adolescents will emphasize these functions (H 1 ). Also, previous findings suggest that mothers maintain closer relationships with their children (e.g., Collins & Russell, 1991;Doyle, Lawford, & Markiewicz, 2009;Markiewicz, Lawford, Doyle, & Haggart, 2006;Mojtabai & Olfson, 2008;Tsai, Telzer, & Fuligni, 2013), and communicate more with their children when compared to fathers (e.g., Bhushan, 1993;Hurd, Wooding, & Noller, 1999;Noller & Bagi, 1985). Since these factors can modify and influence the building of representations, we present a second hypothesis for this set of analyses. If differences How do families represent the functions of deliberate self-harm? emerge between the parents of adolescents without deliberate self-harm, we hypothesize that mothers' social representations will be more similar to the adolescents' social representations (H 2 ).
For the second set of analyses, previous studies revealed that intrapersonal functions are more common among adolescents with deliberate self-harm (e.g., Klonsky, 2007) and that participants without these behaviors tend to value interpersonal functions (Batejan et al., 2015). Hence, the social representations based on the experience of these behaviors' functions should be different from parents' social representations. We hypothesize that adolescents with a history of deliberate self-harm will emphasize their experience of intrapersonal functions and, on the contrary, parents will value more interpersonal functions than these adolescents (H 3 ). Similarly to the first set of analyses, we defined one more hypothesis based on the assumption that mothers maintain closer relationships with their children (e.g., Collins & Russell, 1991;Doyle et al., 2009;Markiewicz et al., 2006;Mojtabai & Olfson, 2008;Tsai et al., 2013) and communicate more with their children compared to fathers (e.g., Bhushan, 1993;Hurd et al., 1999;Noller & Bagi, 1985). Hence, if differences emerge between the parents of adolescents with deliberate selfharm, we hypothesize that mothers' social representations will be more similar to the adolescents' experiences (H 4 ).

Participants
The participants in this study are part of a bigger sample collected during a doctoral thesis investigation. In order to allow the comparison of the representations of family triads, we selected families in which all three elements had completed the questionnaire (adolescent, mother, and father). Hence, the present sample consisted of a total of 609 participants: 203 adolescents, 203 mothers and 203 fathers.

Measures
Inventory of Deliberate Self-Harm Behaviors. The Inventory of Deliberate Self-Harm Behaviors is currently being validated for Portuguese adolescents and has revealed good psychometric properties. E. Duarte et al.

S177
ARCHIVES OF SUICIDE RESEARCH This inventory presents 13 different self-harm behaviors: cutting, biting, burning, pulling hair, scratching until the skin is wounded, consuming drugs with a selfaggressive intent, inserting needles in the skin, ingesting dangerous substances with a self-aggressive intent, drinking alcohol with a self-aggressive intent, banging/hitting, ingesting medication with a selfaggressive intent, ingesting medication with a suicidal intent, and attempting suicide. The respondent is asked to sign the lifetime frequency of each method of selfharm ("No," "Yes -1 Time," "Yes, 2-10 Times," "Yes, More than 10 Times").
In the current study, we also utilized this instrument to assess parents' awareness about their child's deliberate self-harm behaviors. Therefore, parents were asked to assign the lifetime frequency of each method of self-harm for their children.
Questionnaire of Representations about the Functions of Deliberate Self-Harm. This questionnaire has two versions, one for adolescents (Duarte, Gouveia-Pereira, Gomes & Sampaio, in press) and another one for adults (Duarte, Gouveia-Pereira, Gomes & Sampaio, n.d.), which were both used in the current investigation. The questionnaires were validated to Portuguese adolescents and adults and presented acceptable psychometric properties.
The questionnaire for adolescents comprises 35 items that evaluate the representations about 11 functions of deliberate self-harm, which can be categorized according to two dimensions (interpersonal and intrapersonal functions). The interpersonal dimension includes Autonomy & Toughness (e.g., item 24 "Demonstrating they are tough or strong"), Interpersonal Boundaries (e.g., item 1 "Creating a boundary between themselves and others"), Interpersonal Influence (e.g., item 7 "Seeking care or help from others"), Peer Bonding (e.g., item 11 "Trying to fit in with others"), and Revenge (e.g., item 14 "Trying to hurt someone close to them"). The intrapersonal dimension includes Affect Regulation (e.g., item 10 "Reducing their anxiety, frustration, anger, or other emotions"), Anti-Dissociation (e.g., item 27 "Inflicting pain in order to feel something"), Escape Mechanism (e.g., item 19 "Escaping from problems"), Introspective Mechanism (e.g., item 17 "Organizing their ideas"), Replacement of Suffering (e.g., item 18 "Creating physical pain to forget the psychological pain"), and Self-Punishment (e.g., item 25 "Doing it because they feel guilty").
The questionnaire for adults presents 49 items that assess all the functions aforementioned, as well as three additional intrapersonal functions. Hence, the interpersonal dimension includes Autonomy & Toughness (e.g., item 29 "Demonstrating they are autonomous or independent"), Interpersonal Boundaries (e.g., item 22 "Establishing a barrier between themselves and others"), Interpersonal Influence (e.g., item 17 "Seeking care or help from others"), Peer Bonding (e.g., item 36 "Trying to belong to a group of friends/ colleagues"), and Revenge (e.g., item 10 "Getting revenge from someone"). The intrapersonal dimension includes Affect Regulation (e.g., item 1 "Calming themselves down"), Anti-Dissociation (e.g., item 14 "Trying to feel something instead of nothing, even if it is physical pain"), Anti-Suicide (e.g., item 15 "Reacting to suicidal thoughts without attempting suicide"), Escape Mechanism (e.g., item 43 "Escaping from something that is not right"), Introspective Mechanism (e.g., item 34 "Isolating themselves in their thoughts"), Marking Distress (e.g., item 19 "Proving themselves that their emotional pain is real"), Replacement of Suffering (e.g., item 44 "Physically responding to an emotional pain"), Self-Care (e.g., item 23 "Focusing on treating the injury, which can be gratifying or satisfying"), and Self-Punishment (e.g., item 13 "Demonstrating the anger they feel for themselves").
Socio-Demographic Questionnaire. The adolescents responded to questions regarding their age, gender, nationality, education (number of flunks and school grade), the existence of siblings, and marital status of their parents. The socio-demographic questionnaire for parents included items about their age, nationality, education level, marital status, and number of children.

Procedures
This research was approved by the General Education Directorate of the Ministry of Education and Science from Portugal regarding the participation of adolescents. Three schools were contacted and informed about the goals of the investigation. After receiving the schools' administration approval, several classes were selected. In a first phase, the researcher delivered the consent forms to the students' parents, along with the parents' questionnaires. The questionnaires for parents were delivered in an envelope, along with a letter informing them that both mother and father should respond separately and give back the questionnaires in the closed envelope to their child, even if they did not complete the questionnaire. In a second phase, the students whose parents signed the consent form completed the questionnaire for adolescents. Also in this second class, the students brought back their parents' questionnaires and delivered them to the researcher. The participants were informed that their collaboration was voluntary and that all the data were anonymous and confidential. Accordingly, a random code was used to associate the adolescents' questionnaires to their parents' questionnaires.

Data Analysis
All statistical analyses were carried out using SPSS v22 software (IBM SPSS, Chicago, IL). Descriptive statistics were used to analyze socio-demographic data, as well as deliberate self-harm lifetime prevalence. Although both questionnaires that assess the representations about the functions of deliberate self-harm share 11 types of functions, the adults' questionnaire contains three additional functions. Therefore, in order to compare the experiences/representations from these two groups (adolescents and parents), we decided to exclude the functions Anti-Suicide, Marking Distress, and Self-Care from the adults' questionnaire. To examine group differences, we utilized Repeated Measures ANOVA for paired samples.

RESULTS
In the first set of analyses, we compared the social representations about the functions of deliberate self-harm from adolescents without a history of these behaviors and their parents (Table 1). Results revealed significant differences between the group of adolescents and both groups of parents, and no significant differences between mothers and fathers.
In the interpersonal dimension, adolescents presented significantly higher means in the function Interpersonal Boundaries (F ¼ 21.60, p < .001), when compared with both parents (mothers and fathers). In addition, both parents also presented significantly higher means in the functions Interpersonal Influence (F ¼ 6.72, p < .01) and Revenge (F ¼ 66.70, p < .01) when compared to the adolescents' group.
In the intrapersonal dimension, the means from the group of adolescents were significantly higher in the global intrapersonal dimension (F ¼ 22.85, p < .001) and in the functions Affect Regulation (F ¼ 38.41, p < .001), Escape Mechanism (F ¼ 15.01, p < .001), Replacement of Suffering (F ¼ 8.23, p < .001), and Self-Punishment (F ¼ 24.63, p < .001) when compared to both parents. Also, the group of mothers revealed a significantly higher mean in the function Introspective Mechanism (F ¼ 4.16, p < .05) when compared to adolescents.
Globally, these results indicate that most social representations from adolescents and parents were considerably different. However, we did not find significant differences in the global interpersonal dimension and in the functions Autonomy & Toughness, Peer Bonding, and Anti-Dissociation, indicating that the three groups had similar social representations concerning this global dimension and these functions. Also, no significant differences emerged between the representations from mothers and fathers. In a second phase, we compared the functions represented by adolescents with a history of deliberate self-harm and their parents' social representations about these functions (Table 2). Results revealed differences between adolescents and both groups of parents, as well as between mothers and fathers.
Concerning the interpersonal dimension, results showed that parents (mothers and fathers) had significantly higher means in the global interpersonal dimension (F ¼ 11.89, p < .001), and in the functions How do families represent the functions of deliberate self-harm? S180 VOLUME 24 NUMBER S1 2020 Interpersonal Influence (F ¼ 11.07, p < .001), Peer Bonding (F ¼ 10.98, p < .001), and Revenge (F ¼ 14.14, p < .001) when compared to adolescents. Also, there were marginally significant differences in the function Autonomy & Toughness (F ¼ 2.43, p < .1) where both groups of parents presented higher means when compared to adolescents.
In the intrapersonal dimension, adolescents presented significantly higher means in the function Affect Regulation compared with both groups of parents (F ¼ 42.67, p < .001) and in the function Escape Mechanism when compared with their fathers (F ¼ 4.59, p < .05). Also, the group of mothers presented significantly higher means than the group of fathers regarding Anti-Dissociation (F ¼ 7.52, p < .01). Mothers also revealed significantly higher means in the function Introspective Mechanism (F ¼ 9.40, p < .001) in comparison with adolescents and fathers. Additionally, adolescents and their mothers presented significantly higher means in the global intrapersonal dimension (F ¼ 21.72, p < .001), and in the intrapersonal functions Replacement of Suffering (F ¼ 17.49, p < .001) and Self-Punishment (F ¼ 18.63, p < .001).
In this group of families, the only function that did not reveal significant differences was Interpersonal Boundaries. These results demonstrate that the social representations from parents were considerably different from the social representations from adolescents, and also that some differences emerged between mothers and fathers.

DISCUSSION
The objective of the present investigation was to analyze and compare the social representations about the functions of deliberate self-harm from adolescents with and without a history of these behaviors and their parents. Hence, we developed two sets of analyses. The first one focused on families (adolescent, mother, and father) of adolescents without deliberate self-harm, and the second one focused on families (adolescent, mother, and father) of adolescents with deliberate self-harm.
The analyses of the representations about the functions of deliberate self-harm revealed several differences between the groups. Focusing on the families of adolescents without deliberate self-harm, in general, results showed differences between adolescents and both groups of parents. Parents emphasized two interpersonal functions (Interpersonal Influence and Revenge), while adolescents emphasized one interpersonal function (Interpersonal Boundaries) and four intrapersonal functions (Affect Regulation, Escape Mechanism, Replacement of Suffering, and Self-Punishment). We hypothesized that there would be no significant differences between adults and adolescents concerning the interpersonal dimension, and that significant differences would emerge in the intrapersonal dimension, where adolescents would emphasize these functions (H 1 ). Most of our results confirmed our hypothesis (i.e., adolescents gave more relevance to the intrapersonal dimension than parents). However, mothers and fathers valued Interpersonal Influence and Revenge, which partially contradicts our hypothesis.
Interpersonal Influence refers to the adolescent's attempt to obtain help or manipulate others, while Revenge refers to the adolescent's attempt to take revenge on someone (sometimes due to the emotional pain inflicted by that person). Comparing these functions' nature with the other interpersonal functions, it is possible to verify that these parents emphasized the two interpersonal functions where the motivations for deliberate self-harm can be directly oriented towards them as parents. Additionally, if we look at these findings taking into account the results obtained in the intrapersonal dimension, it is clear that parents tend to devalue the intrapersonal component of these behaviors (with the exception of the function Introspective Mechanism). These results imply that parents' social representations focus on interpersonal functions that can affect them directly. Regarding the adolescents' results, our findings suggest that their social representations are closer to the adolescents' experiences, mostly because they value intrapersonal functions that are more common among adolescents who selfharm (e.g., Klonsky, 2007).
Our second hypothesis for this set of analyses stated that, if differences did emerge between parents, mothers' social representations would be closer to adolescents' social representations (H 2 ). However, results partially contradicted this hypothesis, since no differences were found between the group of mothers and the group of fathers, and most differences emerged between the group of adolescents versus both groups of parents.
Regarding families of adolescents with deliberate self-harm, we could verify that the representations' differences seemed to be more accentuated than the ones found in the other group of families. Results revealed that parents emphasized four interpersonal functions (Autonomy & Toughness, Interpersonal Influence, Peer Bonding, and Revenge). On the other hand, these adolescents greatly emphasized Affect Regulation when compared with both their parents and Escape Mechanism when compared with their fathers. We defined two hypotheses for this set of analyses. First, we hypothesized that adolescents with a history of deliberate self-harm would emphasize their experience of intrapersonal functions and, oppositely, parents would value more interpersonal functions than these adolescents (H 3 ). Globally, our results confirmed this hypothesis and are in accordance with previous research that concluded that intrapersonal functions are more prevalent among adolescents with deliberate selfharm (e.g., Klonsky, 2007) and participants without these behaviors tend to value interpersonal functions (Batejan et al., 2015).
Comparing the results of these families with the results from families of adolescents without deliberate self-harm, it is possible to verify that there are greater differences between groups. In fact, except for the function Interpersonal Boundaries (which adolescents without deliberate selfharm emphasized significantly more than their parents, while no significant differences were found in families of adolescents with deliberate self-harm), all the differences between means were accentuated. Additionally, taking into account results in both groups of families, our findings indicate that the differences between the groups can be organized according to two main axes: adolescents versus adults, and interpersonal functions versus intrapersonal functions.
This first axis of differences (adolescents versus adults), refers to the fact that most differences in the two types of families appeared between adolescents and both parents (mothers and fathers). Hence, it is clear that parents represent the functions of deliberate self-harm very differently from the social representations of adolescents with and without deliberate self-harm. Also, since these parents do not have personal experiences regarding the functions of these behaviors, our findings indicate that their social representations might be built according to the stereotypes concerning this phenomenon.
However, there were some exceptions to this adults/adolescents axis in our results. In the families of adolescents without behaviors, the function Introspective Mechanism was valued by mothers, devalued by adolescents, and no differences were found regarding parents. In the families of adolescents with behaviors, only one function from the intrapersonal dimension (Affect Regulation) revealed differences between adolescents and both their parents. The other differences were found in the functions Anti-Dissociation, where mothers emphasized it more than fathers; Escape Mechanism, where adolescents emphasized it more than fathers; and Introspective Mechanism, where mothers emphasized it more than adolescents and fathers. In the two remaining functions (Self-Punishment and Replacement of Suffering), adolescents' experiences and their mothers' social representations were similar.
Our second hypothesis for the group of families with deliberate self-harm stated that, if differences did emerge between mothers and fathers, mothers' social representations would be more similar to the adolescents' social representations (H 4 ). Therefore, these similarities between mothers and adolescents with deliberate selfharm partially confirm our hypothesis. In addition, if we compare the results from the two sets of analyses, Replacement of Suffering and Self-Punishment were the only functions where the mothers' mean was inverted between the two groups (i.e., mothers of adolescents without deliberate self-harm devalued these functions, while mothers of adolescents with deliberate selfharm valued these functions along with their children). Hence, it is accurate to say that mothers' representations of these two functions were closer to the adolescents' representations. We think this similarity may be connected to the fact that mothers usually have closer relationships with their children when compared to fathers (e.g., Collins & Russell, 1991;Doyle et al., 2009;Markiewicz et al., 2006;Tsai et al., 2013). In addition, since of the 51 adolescents with a history of deliberate self-harm in our sample, 38 were female, this result may also relate to the stronger attachment and greater intimacy between mothers and daughters when compared to other parent/ child relationships (e.g., Phares, Fields, & Kamboukos, 2009;Thompson & Walker, 1984).
The second axis refers to the organization of differences in interpersonal functions versus intrapersonal functions. In general, results demonstrated that parents (especially fathers) tend to value interpersonal functions and devalue intrapersonal functions in both types of families. However, in the group of families without deliberate self-harm, this tendency was inverted in the function Interpersonal Boundaries (where adolescents valued this function more than their parents). In this function, deliberate self-harm is a means to assert one's autonomy or to make a distinction between self and others (Klonsky, 2007). Hence, we think this result may be connected to the fact that these adolescents viewed their peers' deliberate self-harm as behaviors associated with social isolation, which has been recognized as a risk factor for deliberate self-harm (e.g., Hawton, Fagg, & Simkin, 1996;Hawton & James, 2005). In fact, if we compare the means of this function in the two groups of adolescents (without deliberate self-harm, M ¼ 3.27; with deliberate self-harm, M ¼ 2.82), it is clear that adolescents without these behaviors emphasize Interpersonal Boundaries.
Overall, our research demonstrated that, in both types of families, parents' social representations tend to value interpersonal functions and devalue intrapersonal functions. These findings may suggest that adults still have the stigma and social belief that these behaviors are essentially directed towards others (i.e., a call for attention, or an attempt to obtain revenge) and have a manipulatory nature (Law, Rostill-Brookes, & Goodman, 2009), which is a stereotypical perspective. This "attention-seeking argument" (Tantan & Huband, 2009) views deliberate self-harm as behaviors that make illegitimate demands on others, and can negatively affect responses and interventions towards these behaviors. Also, since these parents did not have personal experiences regarding the functions of these behaviors, their representations were built according to these stereotypes concerning this phenomenon. This "gap" between adults' and adolescents' social representations can be an important issue to address during the design of prevention and psychoeducation programs directed to parents.
Focusing on the differences found in the families of adolescents with deliberate self-harm, our results are in accordance with data from other studies. For example, Rissanen et al. (2009), recognized the lack of information from parents about this phenomenon, and other studies found that parents of adolescents with self-harm reported difficulties, struggles, and uncertainties in understanding and coping with their child's behaviors (McDonald et al., 2007;Oldershaw et al., 2008). In fact, the misunderstanding of the functions of deliberate self-harm can have a negative impact during clinical interventions, due to the relevance of parental support (Arbuthnott & Lewis, 2015;Miner, Love, & Paik, 2016;Muehlenkamp et al., 2013).
In addition, it is important to note that in our sample only eight parents were aware of their child's deliberate self-harm history, although a total of 102 parents had children who reported having these behaviors. This discrepancy is similar to other studies (Baetens et al., 2015). This particular result underlines the need to create parental awareness regarding deliberate self-harm, in order to allow parents to identify potential self-harm behaviors in their children and encourage helpseeking behaviors.
Finally, previous research has shown that there are differences between the representations about the functions of deliberate self-harm from adolescents without a history of these behaviors and the representations of these functions from adolescents with a history of behaviors (Batejan et al., 2015). Therefore, the results obtained in the present investigation are consistent with the idea that social representations derive from the social belonging and identification of individuals to their social groups (Gouveia-Pereira, Amaral, & Soares, 1997). We conclude that, in this context, the personal experience of deliberate self-harm and the generational differences can be factors that influence the building and modification of these social representations.

Limitations and Directions for Future Research
In summary, our research provided important insight regarding the social representations of the functions of deliberate self-harm from adolescents with and without a history of these behaviors and their parents. However, there are several limitations in this research, including the exclusion of monoparental families (since we opted for the selection of paired samples that comprised families of adolescents, mothers, and fathers), and the fact that we were not able to differentiate between parents who had knowledge of their child's self-harm behaviors and those who did not, due to insufficient data.
Since understanding the representations about the functions of deliberate self-harm can be an important factor for intervention and prevention programs, and taking into account that parents are also relevant elements in these processes, further research is clearly needed in this area. For instance, the development of parental training programs or family-focused interventions that emphasize the facilitation of corrective feedback, positive interactions, and effective support may prove beneficial in reducing the occurrence of self-injurious behavior (Bentley et al., 2014) and can benefit from the knowledge of these representations.
Globally, future research could explore and describe the social representations from adults that can be important elements for the signaling of deliberate selfharm and posterior intervention, namely from teachers and other school staff, or healthcare workers. In fact, research has explored the attitudes from teachers (Evans & Hurrell, 2016;Heath, Toste, & Beettam, 2007;Heath, Toste, Sornberger, & Wagner, 2011) and healthcare workers (Karman, Kool, Poslawsky, & Van Meijel, 2015;Rees, Rapport, Thomas, John, & Snooks, 2014) regarding deliberate selfharm, but no studies have analyzed their knowledge concerning the functions of these behaviors and their possible impact in their interaction with adolescents with deliberate self-harm.
It would also be relevant to compare the social representations of adults with and without a history of deliberate selfharm. In addition, we consider it important to compare the social representations between parents that have knowledge of their child's deliberate self-harm and those who do not know, to understand how the confrontation with these behaviors can alter the representations about its functions. Furthermore, it would be interesting to study the possible relation of these representations (both from adults and adolescents) with other variables, such as psychopathology, suicidal ideation, or religious beliefs.