Parenting a Child with Learning Disabilities: Mothers’ Self-Forgiveness, Well-Being, and Parental Behaviors

Parenting a child with learning disabilities (LDs) can not only be a fulfilling experience, but also a challenging one which can increase parenting stress as well as impair parenting self-efficacy, parental psychological well-being and interactions with the child. Informed by the parenting three-term causal model and the parenting stress model, the present research explored whether self-forgiveness was inversely associated to these undesirable outcomes in mothers of a child with specific LDs. Mothers’ self-forgiveness for perceived failures in dealing with the child’s disabilities was expected to be associated with their psychological well-being and their parental behaviors and relation with the child, both directly and indirectly through parenting stress and parenting self-efficacy. The hypothesized direct and indirect associations were assumed to be moderated by partner support. Data were obtained cross-sectionally from 92 mothers (M age = 43) of children (M age = 12) with diagnosed specific LDs. Self-forgiveness, conceived as a bi-dimensional construct, related to greater psychological well-being and to more accepting and less rejecting behaviors toward the child. For the negative dimension of self-forgiveness, these relations were mediated by parenting stress and parenting self-efficacy, whereas for the positive dimension of self-forgiveness the hypothesized mediational model was supported only for mothers who reported poor support from their partner. The implications for programs designed for parents of children with LDs are noted and several avenues for future research are described. Studies examining the role of self-forgiveness for parenting failures are missing. We show that self-forgiveness may be beneficial for mothers of children with specific LDs. Self-forgiveness is associated with greater well-being and more accepting behaviors toward the child. The above relations are mediated by parenting stress and parenting self-efficacy. Studies examining the role of self-forgiveness for parenting failures are missing. We show that self-forgiveness may be beneficial for mothers of children with specific LDs. Self-forgiveness is associated with greater well-being and more accepting behaviors toward the child. The above relations are mediated by parenting stress and parenting self-efficacy.

• Studies examining the role of self-forgiveness for parenting failures are missing. • We show that self-forgiveness may be beneficial for mothers of children with specific LDs.
• Self-forgiveness is associated with greater well-being and more accepting behaviors toward the child.

•
The above relations are mediated by parenting stress and parenting self-efficacy.
Raising a child with disabilities can be a joyful, fulfilling and enriching experience. Systematic reviews of qualitative research and meta-analyses (see for example, Green et al., 2013;Hastings, 2016;Lalvani & Polvere, 2013) indicate that parents of children with disabilities report deriving important benefits from their atypical parenting experiences (e.g., personal growth, sense of purpose, family closeness). At the same time parents, especially mothers, experience the rearing of a child with impairments as exhausting and distressing, capable of threatening not only their self-image as skilful caregivers, but also their psychological well-being and their interactions with the child (Green et al., 2013;Lee, 2013).
The three-term causal model (Bornstein et al., 2018) posits that parenting practices and their effects on child adjustment are affected by parenting cognitions such as the appraisal of one's competence in the parenting role. In addition, the parenting stress model (Abidin, 1992) assumes that parenting practices and parent-child relations are affected also by parenting stress.
Informed by these theoretical frameworks, the present research explores whether self-forgiveness, a variable which is becoming the subject of a growing interest in psychology, is inversely associated with the above-mentioned undesirable outcomes in mothers of a child with specific learning disabilities (LDs). In doing so, it also examines whether the associations of self-forgiveness are moderated by perceived support provided by the father of the child.
Parenting a Child with Learning Disabilities and its Effects on Well-being and Interactions with the Child Parenting is a challenging task under any circumstances, but it becomes even more challenging when a child is diagnosed with LDs. LDs constitute a heterogeneous group of intellectual and developmental disabilities manifested by significant difficulties in the acquisition and use of listening, speaking, reading, writing, reasoning, and/or mathematical abilities (National Joint Committee on Learning Disabilities, 2016). These difficulties are thought to be due to central nervous system dysfunction and are often accompanied by secondary emotional, behavior, and social problems (e.g., impulsivity, hyperactive-inattentive behaviors, emotional regulation difficulties, social withdrawal, aggression toward others and oneself, experienced loneliness) (Blumsack et al., 1997;Dyson, 2003;Greenham, 1999;Grolnick & Ryan, 1990;La Greca & Stone, 1990;Margalit, 1998;Toro et al., 1990).
Since parents are in an ideal position to support their children's learning by offering one-on-one attention and making immediate modifications when needed (Leach & Siddall, 1990), parents of children with LDs are often involved in tutoring their child's learning activities and in delivering interventions at home in order to improve their academic performance (e.g., Donawa, 1995;Persampieri et al., 2006). The academic support provided to their children on a daily basis often represents a considerable burden for parents, increases as the child grows, and can lead parents to feel distressed and ineffective in assisting their children (Rapus-Pavel et al., 2018;Roll-Pettersson & Mattson (2007)).
Properly understanding and managing the emotional, behavioural, and learning demands of a child with LDs is a difficult and exhausting task, which puts the parents at risk for higher stress, anxiety and depression, and more dysfunctional parent-child interactions (Antshel & Joseph, 2006;Carotenuto et al., 2017;Lardieri et al., 2000). This is particularly true for mothers who typically provide most of the care to children with or without disabilities (Carlson et al., 2016;Harden, 2005;Karande & Kulkarni, 2009;Olsson & Hwang, 2006;Raley et al., 2012). Compared to mothers of typically developing children, mothers of children with an intellectual and developmental disability usually report poorer psychological well-being and physical health (for reviews see Lee, 2013;Masefield et al. (2000); Miodrag et al., 2015;Miodrag & Hodapp, 2010). They also enact more negative and less positive parental behaviors; in particular they express more negative affect, interfering control, disapproval, and hostility, and less regard, warmth, and affection toward the child (Blacher et al., 2013;Fenning et al., 2014;Green et al., 2014).
Given that parenting practices, parent-child relations, and parents' well-being have important implications for the child's development and adjustment (Bornstein, 2015;Pinquart, 2017;Nomaguchi & Milkie, 2020;van Wel et al. (2000)), it is essential to understand factors predicting them. Three variables that can affect well-being and parent-child relations of mothers having a child with LDs, but whose unique effects have never been investigated before in this kind of population, are parenting self-efficacy, parenting stress, and self-forgiveness for parenting failures.

Parenting Self-efficacy and its Effects on Well-Being and Parent-Child Interactions
According to the three-term causal model (parenting cognitions → parenting practices → child adjustment; Bornstein et al., 2018), parenting practices as well as their effects on child adjustment are affected by parenting cognitions like attributions, satisfaction, and knowledge related to parenting (for a discussion of this model in the context of parenting a child with disabilities see Johnston & Ohan, 2005). Substantial research shows that parents' beliefs about themselves and their own parenting impact their behaviors and functioning. Among these beliefs, parenting self-efficacy, broadly defined as an individual's appraisal of his or her competence in the parenting role (Coleman & Karraker, 2000), seems to be a key predictor. A number of reviews on parental self-efficacy outcomes for parents and child from the community (Albanese et al., 2019;Coleman & Karraker, 1998;Jones & Prinz, 2005) document that parents lower in parenting self-efficacy subsequently show less effective parenting, less positive parent-child relationships, and worse psychological functioning. In particular, mothers who perceive themselves as inefficacious in their parental role tend to be less empathic and responsive, more punitive, and more depressed (e.g., de Haan et al., 2009;Meunier et al. (2011);Montgomery, 2009;O'Neil et al. (2009)).
The effects of parents' efficacy beliefs on their parenting practices and functioning have also been documented in a few studies on families having a child with developmental disabilities like ADHD and autism (e.g., Hastings & Brown, 2002;Johnston & Ohan, 2005), but, as far as we know, no research has investigated them in families having a child with LDs. Similar to parents raising children having other kinds of developmental and intellectual disabilities, parents of children with LDs are likely to question their parental competence and ability to effectively handle their child's problems and to improve his/her condition. There is indeed evidence that parents of children with LDs often feel helpless, inadequate and ill prepared to cope with their child's disability (Shechtman & Gilat, 2005). Given that caring is a central component of a mother's social role (Carlson et al., 2016), perceived inadequacy in that domain may have a greater impact on mothers and their interactions with the child compared to fathers. Consistent with this line of reasoning, Hastings and Brown (2002) found that parental self-efficacy mediated the effects of behavior problems of children with autism on their mothers' (but not fathers') depression and anxiety. However, evidence linking parenting self-efficacy to parental well-being and parent-child interactions in mothers of children with LDs is lacking.

Parenting Stress and its Effects on Well-Being and Parent-Child Interactions
Parenting stress is defined as "aversive feelings that are associated with the demands of the parenting role" (Deater-Deckard (1998), p.314). According to Abidin's (1992) parenting stress model, parental behaviors and parent-child relations are negatively affected by parenting stress. Empirical data support this viewpoint as parents who are highly distressed in their parenting role have more difficulties in being sensitive to their child's needs, are more likely to respond with over-or under-involvement, and to have more negative and coercive parent-child interactions (Crnic et al., 2005;McKay et al., 1996, Moes & Frea, 2000. Parenting stress has also been identified as a predictor of caregivers' psychological well-being. In families having children with a variety of developmental disabilities, high levels of parenting stress during childhood are related to poorer relationships between parents and their adolescents and to poorer parental mental health (Enea & Rusu, 2020;Mitchell & Hauser-Cram (2010)).
Compared to parents of typically developing children, parents of children with LDs experience higher levels of stress in domains related to child caring and lower satisfaction with their parental role (e.g., Antshel & Joseph, 2006;Bonifacci et al., 2014Bonifacci et al., , 2015Carotenuto et al., 2017;Shechtman & Gilat, 2005;Smith, 2000). Overall, it seems that the child's behavior problems and difficulties in properly managing them are more important correlates of parental stress than the child's cognitive functioning, especially in middle childhood (e.g., Hassall & Rose, 2005;Hauser-Cram et al., 2001;Neece et al., 2012). Despite this evidence, the relations of parenting distress with parental well-being and quality of parent-child interactions has yet to be explored in mothers of children with LDs.
Parenting stress is related to parenting self-efficacy, but the causal direction of this link is unclear due to the heavy reliance on cross-sectional data (Jones & Prinz, 2005). Even the existing longitudinal research on the topic does not help to disentangle which of the two variables plays the antecedent role: some evidence shows that parenting selfefficacy predicts subsequent parenting stress (e.g., Harper et al., 2013), whereas other data indicate the opposite (e.g., Yap et al., 2019), thereby suggesting a possible bidirectional causal relation.

Self-Forgiveness and its Outcomes
Since parenting self-efficacy and parenting stress have significant impacts on parental behaviors and functioning as well as, indirectly, on children's development and adjustment, research investigating their psychological antecedents has potentially important clinical implications for training interventions aimed at strengthening parenting self-efficacy and diminishing parenting distress (Beresford et al., 2012;McIntyre, 2008;Sanders, 2012;Todd et al., 2014).
One possible antecedent of self-efficacy and stress, which has never been analyzed in relation to parenting, is self-forgiveness, a "coping strategy to deal with stresses that result from personal failure, guilt/shame, or general incongruence between personal values and actual behavior" (Toussaint, Webb & Hirsch, 2017, p. 88; see also Davis et al., 2015). Self-forgiveness involves reducing selfresentment, self-condemning attitudes, and negative emotions directed at the self (such as guilt and regret) as well as increasing benevolence and compassion toward the self, restoring a positive self-image, and regaining a sense of self-worth and self-trust by learning and growing from one's own mistakes or shortcomings (Enright and the Human Development Study Group (1996) ;Hall & Fincham, 2005;Pelucchi et al., 2013;Woodyatt, 2017;Woodyatt & Wenzel, 2013).
Even though no study has investigated self-forgiveness for parenting outcomes, self-forgiveness may relieve parenting stress since it entails relief from negative emotions like guilt, regret or disappointment associated with perceived parenting mistakes. Parents' narratives about their child's LDs are replete with self-blame and guilt, especially in regard to perceived bad parenting or ineffective oversight of the child's mental health (Johnson et al., 2006;Padeliadou & Chideridou, 2013), and guilt feelings are a key component of stress experienced when parenting children with disabilities (Wiener et al., 2016). Consistent with our argument that self-forgiveness may alleviate parental stress, studies using community samples show that state and trait self-forgiveness are negatively related to psychological distress (Cornish et al., 2018;Fincham & May, 2019;Liao & Wei, 2015;Toussaint et al., 2001;Webb, Hirsch, Visser, & Brewer, 2013;Webb, Phillips, Bumgarner, & Conway-Williams, 2013).
Self-forgiveness may also help parents to restore a positive self-image as a capable caregiver. Indeed, genuine self-forgiveness leads to "humble self-respect" (Dillon, 2001, p.83), that reaffirms self-value and self-trust by integrating perceived wrongdoings into a more complex view of the self. According to the Dual-Process Model of Self-Forgiveness , self-forgiveness is a moral repair strategy that restores self-image by replacing self-condemning emotions with self-affirming ones. Selfforgiveness implies personal growth, which positively affects self-efficacy (Toback et al., 2016). Consistent with this view, trait self-forgiveness has recently been shown to be positively associated with global self-efficacy (Gençoğlu et al., 2018;Kravchuk, 2021).
By helping persons who feel responsible for failures to alleviate distress resulting from negative emotions as well as to positively reframe negative beliefs directed at the self, self-forgiveness plays an important role in positively predicting persons' well-being and relations with others. State and trait self-forgiveness are indeed precursors of increased physical and mental health Toussaint, Webb & Hirsch, 2017). Accordingly, in a recent study on grieving parents, self-forgiveness was found to be a strong predictor of their psychological adjustment after loss of a child (Záhorcova et al., 2019). A few studies also documented a positive association between self-forgiveness and relationship satisfaction in close relationships Massengale et al., 2017), thereby suggesting selfforgiveness may help to protect these relations. In particular, two studies show that self-forgiveness for wrongdoings committed against a relationship partner was associated with greater relationship satisfaction reported by both partners (Pelucchi et al., 2013(Pelucchi et al., , 2015. The association was stronger for the negative dimension of self-forgiveness, entailing criticism and resentment toward the self, than for the positive dimension comprising benevolence and compassion toward the self.

The Moderating Role of Partner Support
Like self-forgiveness, social support can positively predict the self-image and well-being of parents raising a child with LDs. Experiencing a supportive social environment enhances selfefficacy and relieves stress when parenting a child with disabilities (e.g., Lu et al., 2021;Smith et al., 2001). Specifically, social support perceived by parents of children with intellectual and developmental disabilities help them to be resilient: it reduces their parental distress, improves perceived parent-child relations and has a positive impact on their mental well-being (e.g., Falk et al., 2014;Peer & Hillman, 2014).
Parents can not only obtain support outside the family (e.g., from physicians, teachers, social services, community networks), but also within the family. Romantic partner support has been shown to have positive effects on mental illness and general well-being, especially for full-time parents (e.g., Acitelli & Antonucci, 1994;Beam et al., 2011;Bodenmann et al., 2011;Feder et al., 2019). Some studies show that having a positive and supportive partner leads mothers of children with intellectual disabilities to experience less caring distress over time as well as higher perceived parenting efficacy and well-being (Ekas et al., 2010;Kersh et al., 2006;Wieland & Baker, 2010).
In light of the above evidence, as well as of the positive correlation found between self-forgiveness and social support in a variety of at-risk population (e.g., Jacinto, 2010;Webb et al., 2011;Weinberg, 2013), it is reasonable to expect that self-forgiveness interacts with partner support in relating to beneficial outcomes for parents of a child with LDs. In fact, as long as receiving support increases one's sense of personal mastery and efficacy in parenting and induces a less distressing interpretation of the child's disability due to the perceived availability of valuable resources other than the self, these positive effects are likely to amplify and strengthen the positive outcomes deriving from self-forgiveness.

Study Overview
Our study examines cross-sectionally the relations between self-forgiveness, parenting self-efficacy and stress, and well-being and interaction outcomes for mothers involved in rearing and providing learning assistance to a child with specific LDs. We explored these relations in mothers only because mothers of children with disabilities typically provide most of the care and are at higher risk of low quality of life compared to fathers (Harden, 2005;Karande & Kulkarni, 2009;Olsson & Hwang, 2006). Figure 1 depicts the assumed relations among the variables of interest.
Consistent with the three-term causal model of parenting (Bornstein et al., 2018), the parenting stress model (Abidin, 1992) and the literature previously reviewed that conceptualizes self-forgiveness as a response to failures having beneficial effects both at the individual and relational level Massengale et al., 2017;Pelucchi et al., 2013Pelucchi et al., , 2015Toussaint, Webb & Hirsch, 2017), we predicted that the more mothers forgave themselves for perceived mistakes related to rearing their child the greater their psychological well-being and satisfaction with mother-child relations, and the more accepting, rather than rejecting, behaviors toward the child (path c).
We also predicted that the above positive relations would be partially mediated by parenting stress and parenting selfefficacy experienced when providing learning assistance to the child: more self-forgiveness would be associated with greater parenting self-efficacy and less parenting stress among mothers (path a), which in turn would be associated with greater psychological well-being, greater mother-child relationship satisfaction, and more accepting and less rejecting behaviors toward the child (path b). These hypotheses are in line with self-forgiveness theory and research, which show that self-forgiveness helps to reaffirm a positive self-image and to relieve distress resulting from self-blame and remorse deriving from failures and mistakes (e.g. Cornish et al., 2018;Fincham & May, 2019;Griffin et al., 2015;Woodyatt, 2017). They are also consistent with a body of research, inspired by Bornestein et al.'s (2018) three-term model of parenting and Abidin's (1992) model of parenting stress, indicating that high parenting self-efficacy and low parenting stress positively affects parents' well-being, parent-child relations, and parental behaviors (Albanese et al., 2019;Crnic et al., 2005;McKay et al., 1996, Moes & Frea, 2000. Finally, relying on the social support literature (Ekas et al., 2010;Kersh et al., 2006;Wieland & Baker, 2010), we hypothesized that the positive relationships of selfforgiveness with parenting self-efficacy, parenting stress, and psychological well-being, parent-child relationship satisfaction and accepting rather than rejecting behaviors toward the child might be stronger for mothers experiencing higher levels of partner support. In other words, we expected that partner support might exert a moderating effect with respect to those relationships (see w 1 and w 2 ).

Participants and Procedure
Ninety-two mothers living in Northern Italy who were parenting a biological child with a diagnosed specific LD participated in the study. They were recruited through facilities specializing in diagnosis and rehabilitation of LDs (N = 25), the provincial division of the Italian Dyslexia Association (N = 20), and public schools (N = 47). At the time of data collection, most of the mothers (83%) were receiving behavioural parent training, individually or together with the child's father.
Mothers were on average 43.29 years old (SD = 6.79); most of them had a high school (56.5%) or a university (27.2%) education, were married to the child's father or cohabited with him (84.8%; 13% lived alone with their children and the remaining cohabited with a different partner from the father) and had 2 children (55.4%; M = 2.0; SD = 0.74). In most cases, only one of their children was diagnosed with LDs (83.9%); if more children received the same diagnosis, mothers provided data related to the child attending the facility through which they had been contacted.
The child with disabilities to whom they referred was a female in 54.3% of cases, averaged 12.2 years of age (SD = 3.05), and received a specific LD diagnosis about 3 years before (age at diagnosis: M = 9.01, SD = 2.24). The diagnosis was made through detailed and validated tests (ACMT Tests, Cornoldi et al., 2002;BDE-2, Biancardi et al., 2016;BVSCO-2, Tressoldi et al., 2012;DDE-2, Sartori et al., 2007 ) 1 . LDs of children were classified as severe, given that their skills were, in at least two tests evaluating the same abilities, 2 SD below the mean observed in peers having the same education level. As commonly happens in Italy, LDs were diagnosed from the third year of primary school (i.e., at 8 years). The children had no other comorbid diagnoses, such ADHD, epilepsy or other neurodevelopmental and psychological disorders.
Maternal and child age, and time since diagnosis did not correlate significantly with any of the variables investigated. However, child sex correlated weakly, but significantly, with maternal parenting.
All participants were treated according to the ethical guidelines established by the Italian Psychological Association (AIP, 2015). Consistent with local legislation and institutional requirements, formal approval of the study by an ethics panel was not required. After providing written informed consent to take part in the study, all participants received an anonymous paper-and-pencil questionnaire which they were asked to complete individually at home and return within one week.

Measures
The questionnaire included instruments measuring selfforgiveness for perceived mistakes or faults in dealing with their child's disability, stress and self-efficacy experienced while providing learning assistance to the child, psychological well-being, mother-child relationship satisfaction and accepting vs rejecting behaviors toward the child. Mothers married or cohabiting with the child's father (N = 78) also completed a measure assessing the support received from him.

Self-Forgiveness
Given the lack of measures assessing self-forgiveness in relation to parental errors, self-forgiveness for mistakes and failures in dealing with the child's disability was assessed with an eight-item measure including 3 ad-hoc items and 5 items adapted from the trait self-forgiveness subscale of the Heartland Forgiveness Scale (Thompson and Synder (2003)).
Sample items include: "If I consider what I did wrong in dealing with my child's disability, it is really hard for me to accept myself", "(…), I am angry with myself", "(…), learning from mistakes that I've done helps me get over them", "(…), with time I am understanding of myself". Items were anchored at 1 (strongly disagree) and 5 (strongly agree). An exploratory factor analysis (method: principal axis factoring; rotation: varimax) revealed two underling orthogonal factors, one negative and one positive. The negative dimension, which comprised 4 items captured self-resentment and a negative self-view and accounted for the 18.1% of the variance with factor loadings ranging from 0.39 to 0.71. The positive dimension, assessing self-benevolence, self-compassion, and self-growth, accounted for the 14.8% of the variance with factor loadings ranging from 0.43 to 0.60. To overcome the long-established fact that coefficient alpha is very sensitive to scale length, several authors recommend using the mean inter-item correlation as a much clearer measure of internal consistency (e.g., Nunnally & Bernstein, 1994;Streiner, 2003). An average inter-item correlation falling in the range between 0.15 and 0.50 is commonly judged as optimal (Clark and Watson,1995). The dimensions obtained in this study demonstrated good internal consistency, especially when measured through mean inter-item correlation which, unlike Cronbach's alpha, is not sensitive to scale length (Cronbach's alpha = 0.65 and 0.61; Mean-inter item correlation = 0.37 and 0.29 for the negative and positive dimension respectively). A composite score for each selfforgiveness dimension was obtained by averaging the items.

Parenting stress and Parenting self-efficacy
Stress and self-efficacy experienced while providing learning assistance to the child were assessed with an adapted version of the Caring stress and Ability to care subscales of the Italian version of the Adult Carer Quality of Life Questionnaire (AC-QoL) (Negri et al., 2017;Joseph et al., 2012). The AC-QoL questionnaire is a 40-item instrument that measures the overall quality of life for adult carers through eight subscales consisting of 5-items each. The Caring stress subscale assesses mental and physical stress resulting from caring, such as exhaustion and depression (e.g., "I feel worn out as a result of caring", "I feel depressed due to caring"), whereas the Ability to care subscale measures the extent to which the carer perceives him/herself as able to provide care, how the carer copes with the caring role and feels about his/her competency to care (e.g. "I am satisfied with my performance as a carer", "I can take care of the needs of the person I am caring for"). In both subscales participants indicated on a 4-point Likert-type scale ranging from 0 (never) to 3 (always) the frequency with which they felt or experienced what was reported by each item in the last two weeks. In our study, the two subscales were adapted to measure stress and selfefficacy the subjects experienced when providing learning assistance to their child with disabilities. A composite score for each subscale was obtained by averaging all items, with higher scores indicating higher stress and higher selfefficacy (Cronbach's alpha = 0.76 and 0.82; Mean-inter item correlation = 0.43 and 0.39 for Parenting stress and Parenting self-efficacy, respectively).

Psychological Well-Being
Psychological well-being experienced in the last month was measured using the Psychological General Well-Being Index (PGWBI; Chassany et al., 2004;Dupuy, 1984), which has been validated and used in many countries, including Italy (Grossi et al., 2002). The scale comprises 22 polytomous items with scores ranging from 0 to 5 and covering 6 underlying domains: anxiety (5 items, e.g. "Have you been anxious, worried, or upset during the past month?"), depressed mood (3 items, e.g. "Did you feel depressed during the past month?"), positive well-being (4 items, e.g. "How have you been feeling in general during the past month?"), self-control (3 items, e.g. "Have you been in firm control of your behavior, thoughts, emotions or feelings during the past month?"), general health (3 items, e.g. "How often were you bothered by any illness, bodily disorder, aches or pains during the past month?") and vitality (4 items, e.g. "How much energy, pep, or vitality did you have or feel during the past month?"). As the subscales are internally consistent, the 22 items have been frequently used to form an overall index for general wellbeing. A composite score for each subscale as well as for the overall index was obtained by summing the items, with higher scores indicating greater psychological well-being (Cronbach's alpha = 0.92 for the overall index and ranging from 0.72 to 0.83 for all subscales with the exception of General health = 0.54; Mean-inter item correlation = 0.32 for the overall index and ranging from 0.30 to 0.50 for all subscales with the exception of the Self-control = 0.60).

Relationship Satisfaction
Mother-child relation satisfaction was measured using the 3-item Relationship Satisfaction subscale of the Perceived Relationship Quality Component Inventory (Fletcher, Simpson & Thomas, 2000). Mothers were asked to answer each item (e.g., "How satisfied are you with your relationship?) on a Likert scale ranging from not at all (1) to extremely (10). A composite score was obtained by averaging the three items, with higher scores indicating greater satisfaction with the relationship with the child (Cronbach's alpha = 0.98; Mean-inter item correlation = 0.94).

Maternal Behaviors
The mother's behaviors toward the child were measured using the short parent version of the Parental Acceptance-Rejection Questionnaire (Parent PARQ; Rohner, 2005). The questionnaire consists of 24 items measuring parental acceptance and parental rejection. Acceptance reflects behaviors expressing warmth or affection rather than hostility or aggression toward the child (e.g. "I say unkind things to my child" (reverse scored), "I make my child feel wanted and needed"), whereas rejection reflects parental behaviors expressing indifference, neglect or undifferentiated rejection toward the child (e.g. "I am too busy to answer my child's questions", "I let my child know (s)he is not wanted") (see García-Pérez et al., 2017). Response scales were anchored at 1 (never or almost never) and 4 (every day). A composite score for each dimension was obtained by averaging the items, with higher scores indicating greater acceptance/rejection (Cronbach's alpha = 0.79 and 0.65; Mean-inter item correlation = 0.16 and 0.17 for acceptance and rejection, respectively).

Partner support
Perceived support from the child's married or cohabiting father was assessed using a support scale (Paleari et al., 2009;Paleari et al., 2002) comprising 14 mirror items, 7 for support given and 7 for support received within the relationship. Since in the present study we are interested in the support received from the partner, only the latter 7 items were used (e.g., "When I need it, my partner helps me," "My partner approves what I do"). Mothers rated their degree of agreement with each item on a 5-point Likert-type scale ranging from 1 (very strong disagreement) to 5 (very strong agreement). A composite score was obtained by averaging the items, with higher scores indicating greater support received from the partner (Cronbach's alpha = 0.90; Mean-inter item correlation = 0.43).

Data Analysis
The PROCESS 3.3 macro (Hayes, 2017) for SPSS (version 21) with 5000 bootstraps was used to examine the hypothesized mediational and conditional effects. Bootstrapping addresses the limitations of approaches proposed by Baron and Kenny (1986) and Sobel (1982), yielding results that are more accurate and less affected by data distribution and sample size (Hayes, 2009;Preacher & Hayes, 2008).
Specifically, we used Model 4 to estimate the direct and indirect effects of each self-forgiveness dimension on outcome variables (psychological well-being, mother-child relation satisfaction and behaviors toward the child) via both stress and self-efficacy experienced when caring for the child. In each estimated mediation model, a selfforgiveness dimension was entered as independent variable (X), parenting stress and parenting self-efficacy were entered as simultaneous mediators (Ms), and one of our outcome variables was entered as dependent variable (Y). Specifically, for psychological well-being we alternatively entered as dependent variable the overall index and each of its dimensions, and for maternal behaviors we alternatively entered its accepting and rejecting dimensions.
We used Model 8 to test whether the direct and indirect effects of self-forgiveness dimensions on outcome variables (psychological well-being, mother-child relation satisfaction or behaviors toward the child) were moderated (or conditioned) by partner support. In particular, Model 8 estimates any indirect effect of self-forgiveness (X) by the perceived partner support (W) on outcome variables (Y) through parenting stress and parenting self-efficacy (Ms) by considering the interaction effect of X and W on Y, the interaction effect of X and W on M, and the main effect of M on Y (see Fig. 1). As before, each maternal behavior dimension, each psychological well-being dimension and its overall index were alternatively entered as dependent variables.
All analyses were performed controlling for the selfforgiveness dimension other than the one entered as the independent variable and standardizing all variables prior to model estimation so as to be able to compare effects across different predictors.

Preliminary Considerations
Descriptive statistics and correlations among the variables are displayed in Table 1.
Self-forgiveness, parenting stress and parenting selfefficacy significantly correlated in the expected manner with well-being and behaviors dimensions, but not with motherchild relationship satisfaction that was therefore excluded from subsequent analyses. Contrary to the hypothesis according to which both forgiveness dimensions would have significantly correlated to parenting stress and parenting self-efficacy, only the negative dimension of selfforgiveness was significantly related to them.
Partner support was significantly associated with parenting self-efficacy, but not with parenting stress.

Mediational Analyses
The mediational analyses showed that parenting stress and parenting self-efficacy did not mediate the links between the positive dimension of self-forgiveness and psychological well-being or behaviors toward the child. In fact, when controlling for the negative dimension of self-forgiveness, the positive dimension was positively related to overall psychological well-being (and more specifically to depression, self-control, general health, and vitality dimensions) only directly (see Path a or c and Paths ab for the positive dimension of self-forgiveness in Table 2).
In contrast, when controlling for the positive dimension of self-forgiveness, the negative dimension was associated with the outcome variables not only directly, but also indirectly through parenting stress and parenting selfefficacy (see Path a or c and Paths ab for the negative dimension of self-forgiveness in Table 2). In fact, 26% of the total effect linking the negative dimension of selfforgiveness to overall psychological well-being was accounted for by the mediation via stress and 15% of this total effect by the mediation via self-efficacy. Examining the various dimensions of well-being, we found partial mediation for anxiety, positive well-being and vitality via both parenting stress and parenting self-efficacy, as the direct effect (path c) remained significant, and total mediation for depression, self-control, and general health via parenting stress as the direct effect (path c) was Table 1 Descriptives and Pearson's correlations among the variables (N nonsignificant. The negative dimension of forgiveness was also indirectly related to child accepting and rejecting behaviors via parenting self-efficacy (proportion of variance explained by the indirect effect was 24% and 25%, respectively) and to child accepting behaviors via caring stress (proportion of variance, 15%). The total effects on outcome variables were generally stronger for the negative dimension of self-forgiveness than for the positive dimension (see Paths c + ab in Table 2).
When we introduced child sex as a further covariate in our regression analyses, regression coefficient remained significant and did not change in strength, indicating that our results were unaffected by child sex.
In summary, the mediational analyses showed that the more the mothers were unforgiving toward themselves (negative dimension of self-forgiveness) the lower their psychological well-being. They also reported displaying more rejecting and fewer accepting behaviors towards their child. These negative outcomes were at least partly explained by the fact that mothers who were unforgiving towards the self, experienced more stress and felt less able to provide care to their child. In contrast, mothers' self-forgiveness (positive dimension) was unrelated to their stress and self-efficacy and therefore was associated with their psychological well-being only directly. Contrary to our prediction that both forgiveness dimensions would be related to behaviors toward the child through the mediation of parenting stress and self-efficacy, self-forgiveness (positive dimension) was unrelated to behaviors toward the child when controlling for parenting stress and parenting self-efficacy.

Conditional Process Analyses
Conditional process analyses were carried out with mothers married or cohabiting with the child's father (N = 78) to determine whether the direct and indirect effects of selfforgiveness on outcome variables were moderated by partner support. The only significant conditional effects found are reported in Table 3, together with the simple effects estimated at lower (1 SD below M), medium (M), and higher levels (1 SD above M) of partner support.
The analyses revealed that partner support moderated the association of the positive dimension of self-forgiveness with parenting self-efficacy and, consequently, the indirect associations of the positive dimension of self-forgiveness with psychological well-being and with behaviors toward the child. For mothers who perceived lower partner support (1 SD below M), the positive dimension of self-forgiveness had significant indirect effects, through parenting self-efficacy, on their psychological well-being and on their behaviors toward the child (see Table 3). The more these mothers were forgiving toward themselves, the more they felt able to provide learning assistance to their child, which Standardized coefficients (β) are reported out of parentheses and 95% CI derived from bootstrap resample among parentheses. Significant results are typed in bold Separate analyses were carried out for each of the outcomes listed in the table. For psychological well-being, we alternatively entered as outcome variable the overall index and each of its dimensions; for maternal behaviors we alternatively entered its accepting and rejecting dimensions PW Psychological Well-being, B

Behaviors a
Higher scores indicate greater well-being (i.e. lower anxiety and depression) in turn was positively related to more accepting and less rejecting behaviors toward the child and to greater psychological well-being (especially in terms of positive wellbeing, general health, and vitality). However, contrary to our expectation that forgiveness would have a stronger association with outcome variables at higher levels of partner support, we found no significant indirect effects among mothers with higher (1 SD above M) or moderate (M) levels of perceived support from the partner (see Table  3). Also, no direct links between the positive dimension of self-forgiveness and outcome variables were moderated by partner support for these mothers. In contrast, the indirect links between the negative dimension of self-forgiveness and outcome variables were not moderated by the level of support received from the partner. Only the direct link between the negative dimension of self-forgiveness and the vitality dimension of wellbeing was moderated by partner support (see Table 3): the more the mothers were unforgiving toward themselves the less they experienced vitality, but only if they received high or average levels of support from the partner.
As before, our findings were unaffected by child sex.

Discussion
Most extant research shows that alongside positive aspects of raising a child with intellectual and developmental disabilities, parents report difficulties and failures. The latter can challenge their image of themselves as capable caregivers, increase stress experienced when parenting, and impair their well-being and their interactions with the child, especially in the absence of protective factors helping them to be resilient (Peer & Hillman, 2014). These undesirable outcomes are more frequently observed in mothers, who fulfill the primary caregiver role, than in fathers (Miodrag & Hodapp, 2010;Hastings & Brown, 2002;Karande & Kulkarni, 2009). In order to expand knowledge of factors predicting wellbeing and parental behaviors for mothers of children with disabilities, the present research investigated whether selfforgiveness for failures and mistakes mothers feel responsible for in caring for a child with specific LDs is associated to their well-being and parental behaviors. Relying on parenting models (Bornstein et al., 2018;Abidin, 1992) and existing research on self-forgiveness outcomes (Toussaint, Webb, & Hirsch, 2017), our study provides initial evidence that mothers' self-forgiveness is associated with greater psychological well-being and more accepting and less rejecting parental behaviors through the mediation of parenting stress and parenting self-efficacy.
Consistent with our initial assumptions, we found that maternal self-forgiveness was significantly related to greater psychological well-being and to more affectionate and less hostile or indifferent parental behaviors. Specifically, the positive dimension of self-forgiveness, reflecting self-benevolence, self-compassion, and self-growth, was significantly associated with greater self-control and better general health. In contrast, the negative dimension, representing resentment and a negative self-view, was significantly associated with lower positive well-being, lower self-control, lower vitality, greater anxiety, more frequent rejecting behaviors and less frequent accepting behaviors toward the child. In line with these findings, existing research on community samples suggests that greater self-forgiveness for regretted interpersonal actions leads over time to higher self-esteem, lower general psychological distress, and more empathic attitudes (Cornish & Wade, 2015;Woodyatt & Wenzel, 2013). Similarly, selfcompassion helps people to reduce impulsivity and to practice important health-promoting behaviors (e.g., Adams & Leary, 2007;Morley et al., 2016;Sirois, 2015;Webb & Forman, 2013). In contrast, resentful and condemning thoughts and feelings are often accompanied by blame and ruminative tendencies (e.g., Wohl et al., 2008: McCullough et al., 2007 that foster maladaptive outcomes, including greater anger, lower self-esteem, lower relationship satisfaction and life satisfaction (Baumeister et al., 1994;Newman & Nezlek, 2019;Pearson et al., 2010). Also, in parents of children with intellectual and developmental disabilities greater trait guilt is significantly associated with lower subjective happiness (Findler et al., 2016), whereas trait self-compassion is linked to positive mental health outcomes (Robinson et al. (2018)).
Overall, our study showed that the negative dimension of self-forgiveness was more strongly related to well-being and parental behaviors than the positive dimension. This finding mirrors previous research on personal and relational outcomes of self-forgiveness and interpersonal forgiveness (e.g., Gordon et al., 2009;Pelucchi et al., 2013Pelucchi et al., , 2015Paleari et al., 2009), which documented stronger effects for the negative than for the positive dimension of forgiveness. Some scholars have argued (e.g., Rozin & Royzman, 2001) that, from an evolutionary perspective, being able to recognize and control negative emotions and/or situations is more adaptive than being able to recognize and control positive ones.
In a similar vein, the negative, but not the positive, dimension of forgiveness was significantly related to parenting stress and parenting self-efficacy of mothers caring for a child with LDs. As predicted on the basis of Abidin's (1992) parenting stress model and Bornstein et al.'s (2018) three-term causal model of parenting, parenting stress and parenting self-efficacy were related to psychological wellbeing and parental behaviors in mothers, even when controlling for self-forgiveness levels. Therefore, our hypothesis that parenting self-efficacy and stress would mediate the association between self-forgiveness and well-being or parental behaviors was supported for the unforgiveness dimension. As some scholars have recently shown (Moreira & Canavarro, 2018;Stillar et al., 2016), a ruminative and self-critical style of thinking makes parents particularly vulnerable to parenting stress and to low perceived selfefficacy in caring for the child.
Results from conditional process analyses nonetheless indicate that the presence of self-forgiveness can actually be significantly related to greater parenting self-efficacy and, indirectly, to better psychological well-being and parental behaviors, but only when mothers perceive themselves to be poorly supported by their partner (and father of the child). Thus, contrary to our prediction, self-forgiveness did not interact synergistically with social support in relating to mothers' positive self-image and well-being. Rather, selfforgiveness seems to have played a compensatory role: when levels of support received from the partner are not high enough to help mothers restore a positive self-image as a caregiver, self-forgiveness appears to compensate by promoting a more positive self-view. Conversely, no significant interaction was found between the association of self-forgiveness and of partner support with stress related to child caring, perhaps because receiving support can have positive and/or negative effects on the recipient's stress. In fact, a number of studies show that receiving emotional support from one's partner can lead to increased, rather than decreased, stress. This occurs because it can lead the recipient to feel indebted and dependent or to doubt his/her ability in managing the stressor (Gleason et al., 2008;Shrout et al., 2006). Reviewing their multimethod program of research, McClure et al. (2013) documented that support has beneficial or detrimental effects depending on a complex configuration of individual, relational, and contextual factors such as the perceived responsiveness of the support act, the degree to which support is matched to the recipient's desires, and the level of stress and nature of the stressor. Because mothers of children with LDs are exposed to multiple enduring stressors, many of which go beyond their relationship with the child and are difficult to control (e.g. economic strains, social stigma and exclusion), it is possible that partner support may contribute to restore mothers' positive self-image, but not to reduce their stress and improve their overall well-being. Consistent with this reasoning, in the present study we found that partner support was significantly associated with mothers' parenting self-efficacy, but not with their parenting stress or overall well-being.
Although the present results are promising, several limitations need to be born in mind when interpreting the findings. First, the study is exploratory and descriptive in nature and its finding have limited generalizability, as they are based on a relatively small sample of well-educated mothers who were not selected randomly, were for a large majority receiving behavioral parent training, and whose child had no behavioral disorders or comorbidities. This last condition is quite uncommon, given that LDs frequently occur with other neurodevelopmental disorders, which are likely to further impact parental well-being and parent-child relations. Also, the failure to assess mother psychopathology (and in particular mother LDs) means that it cannot be ruled out as a cause of the present results. Second, the correlational data obtained do not provide information on direction of effects. To explore direction of effects among the variables it will be important to collect longitudinal data as well as control for covariates related to the caregiving situation (e.g., socio-economic status, working condition and family structure). Third, plausible alternative models might exist that include further moderating effects or intervening variables. For example, with a larger sample one could test whether social support has a moderating effect not only with respect to direct and indirect outcomes of self-forgiveness, but also with respect to parenting self-efficacy or parenting distress outcomes. Further, the relation between self-forgiveness and health could be mediated by health-enhancing psychosocial variables like hope and optimism, which share with forgiveness the common goal of striving for a better future and a flexible approach to coping with problems (Toussaint, Barry, Angus, Bornfriend, & Markman, 2017). In fact, dispositional optimism has been shown to mitigate the negative impact of youth developmental and behavior disorders on mothers' well-being (Blacher and Baker (2019)). Fourth, even though the state forgiveness measure used has the merit of being multidimensional and shows good convergent validity, its internal consistency can be improved. In the absence of such improvement, measurement errors should be taken into account through statistical techniques, like structural equation modeling, which could not be used in our study due to sample size. Fifth, the reliance on selfreports in this study may have generated common method biases which future research should overcome by using multiple-informant measures (e.g., observational data for parental behaviors).
Despite these shortcomings, our study is the first to show that self-forgiveness might have beneficial outcomes for the well-being and parental behaviors of mothers caring for a child with specific LDs. This evidence helps extend the literature on strategies and resources that parents can rely on to effectively cope with their children's intellectual and/or developmental disabilities and become resilient. This literature has mainly focused on the beneficial effects of optimism, social support and coping strategies such as planful problem-solving and positive reappraisal (Peer & Hillman, 2014), whereas the implications of self-forgiveness, a response which helps to reduce self-blame and restore self-worth after perceived failures, have not been investigated. Furthermore, self-forgiveness researchers have very rarely explored the effects of self-forgiveness on parents experiencing guilt or feelings of remorse (for an exception related to the loss of a child see Záhorcová et al., 2019). Finally, the study expands evidence on the key role of parenting stress and lower parenting self-efficacy in worsening parents' well-being and behaviors towards the child, by examining them in a sample of mothers of children with LDs (Abidin, 1992;Albanese et al., 2019;Jones & Prinz, 2005).
Our study also has interesting practical implications. Parents of children with LDs are at higher risk for parenting stress, poor parenting self-efficacy, and poor overall functioning, which in turn negatively impact their children's outcomes, such as their academic, social-emotional and behavioral adjustment (Carotenuto et al., 2017;Dyson, 2003;Lardieri et al., 2000;Shechtman & Gilat, 2005;Smith, 2000). It is therefore of practical importance to identify and understand the factors promoting parental selfefficacy and well-being as this information could inform interventions for parents of children with LDs. Parenting interventions based on cognitive-behavioral models, like Riding the Rapids (Beresford et al., 2012;Todd et al., 2014) that was specifically developed for parents of children with LDs, and Triple P-Positive Parenting Program (Sanders, 2012) and Incredible Years (McIntyre, 2008), which were adapted for them, have being shown to reduce parenting stress and self-blame and improve self-efficacy through modules teaching attributions retraining or mood and stress management. In light of our results as well as studies documenting the efficacy of self-forgiveness treatment (Cornish et al., 2017;Jacinto & Edwards, 2011), an additional module on self-forgiveness promotion might be profitably included in interventions to relieve parenting stress and foster parenting self-efficacy.
Although the present findings do not provide definitive support for incorporating self-forgiveness into programs provided for parents of children with specific LDs, they point to the utility of gathering further data on the role of self-forgiveness in promoting parental well-being. Future studies might for example determine whether similar results emerge when assessing forgiveness at the dispositional level. Given that well-being is usually more strongly associated with trait than state forgiveness Toussaint, Webb & Hirsch, 2017), one can expect even stronger beneficial effects for self-forgiveness when measured at the dispositional level. Also, considering that psychological well-being is commonly defined as a combination of feeling good (or hedonia) and functioning effectively (or eudaimonia) (Huppert, 2009), future research might explore whether self-forgiveness is equally effective in promoting both components of parental well-being. Given that this study did not assess child behaviors and adjustment, future research might also examine whether mothers' self-forgiveness relates to their child's outcomes cross-sectionally and longitudinally.
It would also be helpful to replicate the present findings on more heterogeneous samples of parents, including both fathers and mothers, LGBTQ parents or single parents of children with different types of disabilities and comorbidities or without disabilities. Parent's sense of inadequacy and failure may vary greatly depending on the specific financial, physical, and psychological burdens their child's unique profile of disabilities poses to them. In this regard there is evidence that parents of children with multiple disorders, who are required to complete more difficult caregiving tasks associated with higher frequencies of difficult child behavior, experience higher levels of parental stress (Plant & Sanders, 2007). Given that fathers are becoming progressively more involved in child care and tend to cope with their child's disabilities differently from mothers (Al-Yagon, 2015), efforts to recruit and include fathers in research would add important information on interparental functioning and gender specificity in rearing a child with disabilities. In addition, considering the financial and psychological strains experienced by single parents and the specific discriminations and misconceptions suffered by LGBTQ parents of children with disabilities (Cigno & Burke, 1997;Gibson, 2018;Parish et al., 2012), the extent to which self-forgiveness has positive outcomes even in these further challenging situations merits future investigation. Finally, including community samples of fathers and mothers would also assist in determining whether selfforgiveness may have positive outcomes even for parents of children in less challenging circumstances.
In conclusion, this study made a significant contribution to the literature on mothering a child with LDs by providing initial support regarding the potential beneficial outcomes of mothers' self-forgiveness. In particular, the present findings indicate that, when measured in terms of lack of selfresentment and of a negative self-view, higher selfforgiveness for mothering failures relates to less parenting stress and more parenting self-efficacy, which in turn are associated with higher well-being and more accepting and less rejecting parental behaviors. When measured in terms of self-benevolence and a positive self-view, selfforgiveness relates to higher well-being and more accepting and less rejecting parental behaviors via parenting selfefficacy only for mothers who reported poor support from their partner.
Funding Open access funding provided by Universitã degli studi di Bergamo within the CRUI-CARE Agreement.

Compliance with Ethical Standards
Conflict of Interest The authors declare no competing interests.
Ethical Approval All procedures performed in the study were in accordance with the ethical standards of the institutional research committee (Regolamento di ateneo per l'integrità e l'etica della ricerca (University regulations for the integrity and ethics of research) -Emanato con D.R. (issued by Chancellor's Decree) Rep. n. 387/2016prot. n. 80496/I/3 del 18.7.2016e Rep. n. 431/2016prot. n. 95147/I/3 del 30.8.2016) and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.
Informed Consent Informed consent was obtained from all individual participants included in the study.
Publisher's note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons. org/licenses/by/4.0/.