Introduction
The lifetime prevalence of depressive disorder is between 10 and 15 % (Lépine and Briley
2011) and is associated with significant impairments in many domains of functioning. Depression in mothers and fathers is associated with emotional and behavioral problems in their children (e.g., Connell and Goodman
2002; Goodman
2007; Goodman et al.
2011; Ramchandani et al.
2005; Weitzman et al.
2011). Depression may undermine parents’ capacity to parent and the quality of parent–child interactions (Goodman
2007; Goodman and Gotlib
1999; Ramchandani and Psychogiou
2009). Mothers with depression may express fewer positive emotions, have less child-oriented and more parent-oriented concerns (Dix et al.
2004), and exhibit behavior toward their children that is more hostile and less positive and engaging compared to mothers without depression (e.g., Lovejoy et al.
2000). Depressed fathers similarly may display more negative and less positive parenting behaviors (Nath et al.
2015; Wilson and Durbin
2010). For example, depressed fathers reported that they smacked their 12-month-old children more (41 vs 13 %) (Davis et al.
2011) and spent less time with their children on positive activities (Paulson et al.
2006), compared to fathers without depression. Although a link between parental depression and parenting is established, the mechanisms that contribute to impaired parenting remain poorly understood and understudied. Therefore, a key gap in the literature is an understanding of the mechanisms that potentially underpin depression and parenting difficulties. An emergent literature (e.g., Diedrich et al.
2014; Krieger et al.
2013; Kuyken et al.
2010) on the importance of self-compassion in depression suggests a promising avenue of research.
According to cognitive theory, people at risk for depression tend to have attentional biases to negative material, negative beliefs, upsetting images and memories, ruminative thinking, and avoidant behaviors that can be easily triggered and escalate into depression (Beck and Haigh
2014). Finding ways to manage these cognitive tendencies may enable those at risk for depression to be more resilient, including their role as parents. Self-compassion is typically defined as recognizing and being open to one’s suffering, being kind toward oneself, and seeing one’s experience as part of the larger human experience (Barnard and Curry
2011; Feldman and Kuyken
2011; Neff
2003a). It provides an alternative to the negative thinking, avoidance, and rumination that characterize depression. Self-compassion potentially enables a person to respond to negative thoughts and feelings with kindness, equanimity, and patience. This helps them de-center from negative thinking and feelings, which in turn prevents the escalation and maintenance of depressive thinking and behavior (Feldman and Kuyken
2011). There is a growing literature that suggests that self-compassion may be an adaptive strategy for managing the types of negative thoughts and feelings described above (Barnard and Curry
2011). There is now also evidence that among people with a history of depression, self-compassion may help them to overcome problematic patterns of avoidance and rumination (Diedrich et al.
2014; Krieger et al.
2013).
Self-compassion has also been found to be associated with intrinsic motivation, health behaviors like exercise, body image, personal responsibility, more caring and supportive relationship behavior (Neff et al.
2007), as well as their bio-behavioral correlates (Breines et al.
2014; Klimecki et al.
2013a,
b). Self-compassion was also impaired among people suffering depression and was associated with differences in emotion regulation (Diedrich et al.
2014; Krieger et al.
2013). Interventions like mindfulness-based cognitive therapy that cultivate self-compassion can prevent depressive relapse/recurrence, and there is some evidence that outcomes are mediated through learning self-compassion (Kuyken et al.
2010). Thus, it is likely that self-compassion may also be linked to parenting behaviors. For example, self-compassion may enable parents to recognize, allow, and de-center from negative thoughts so they can be more aware of and less reactive to negative thoughts. This may increase their ability to respond sensitively to their children’s needs (Kabat-Zinn and Kabat-Zinn
1998). Parents with greater self-compassion might increase their capacity to respond adaptively to triggers and decrease the frequency of over-learned automatic patterns of thinking and behaving. These processes may manifest as an increased ability to respond to the challenges of parenting in ways that are more sensitive and resilient. For example, a woman trying unsuccessfully to console her crying infant during the night might have the self-judging thought “I am a bad mother.” In depression, this might spiral into a sense of helplessness and hopelessness. However, an individual with greater self-compassion might respond to the same thought with “wait, where did that come from, that’s a very negative thought, I must be very tired and need some sleep. I’ll ask my partner if he can do the next feed and take some naps with my child tomorrow.” Parents who embody self-compassion may also model resilience for their children. It is possible that children learn self-compassion expressed by their parents, a hypothesis that would be interesting to test. Thus, for example, a child who sees his or her parent respond with self-compassion may learn to respond in the same way through observational learning. Another possibility is that self-compassion may be translated into more positive parenting, which in turn may protect children from the negative effects of parental depression. Parents who score high on self-compassion may display more responsiveness and warmth. Therefore, it is likely that parents’ self-compassion is linked both to parenting and children’s outcomes.
The preschool years are a key developmental period for several reasons. The effects of depression might be more evident on young children given young children’s dependence on their parents. Preschool children have to meet key goals (e.g., learn to regulate their emotions, behave appropriately in structured settings, establish and maintain positive peer relationships) to make a successful transition to school. Depression may limit parents’ capacity to model and scaffold the necessary skills for their children’s attainment of these goals. Furthermore, elevated levels of emotional and behavioral problems in the child may tax the insufficient inner resources of the depressed parent, and the parent–child dyad may then become entrenched in a negative cycle of interaction that reinforces each other’s negative behavior (Patterson
1982). Childhood emotional and behavioral problems can be persistent and predict adverse future outcomes including poor academic achievement, aggressive and antisocial behavior, as well as adult psychiatric disorders (Campbell
1995; Egger and Angold
2006; Kim-Cohen et al.
2003).
In this paper, we examined data from two studies to explore whether self-compassion may be associated with internal states related to parenting (more adaptive attributions), emotional regulation in the face of challenges (coping well with children’s negative emotions), expressed emotion (more positive and fewer critical comments about their children), and parents’ ability to respond to their children sensitively. In addition, we explored whether there was an association between self-compassion among parents and lower levels of emotional and behavioral problems that they reported in their children.
Results
Range, means, and standard deviations for all measures for study 1 are presented in Table
1. The overall self-compassion scale correlated with both positive (
r = −0.45,
p < 0.01) and negative attributions (
r = −0.42,
p < 0.05). There were no significant associations between self-compassion and sensitive responding (
r = 0.29,
p = 0.127). Table
2 shows that when controlling for child gender, parent education, and depressive symptoms, the associations between self-compassion and positive (
β = −0.45,
p < 0.05) and negative (
β = −0.42,
p < 0.05) attributions remained significant. The effect size for positive attributions was
f
2 = 0.25, and for negative attributions, it was
f
2 = 0.19, which indicate medium to large effects.
Table 1
Range, means, and standard deviations for study variables (study 1)
Self-compassion | 33 | 37–88 | 61.30 (13.26) |
Depressive symptoms | 38 | 0–32 | 11.21 (9.17) |
Parental sensitive responding across three tasks | 34 | 6–19 | 13.18 (3.42) |
Positive attributions | 37 | (−4)–(5) | 0.38 (2.45) |
Negative attributions | 37 | (−5)–(2) | −1.05 (1.55) |
Table 2
Hierarchical linear regressions showing the associations of parents’ self-compassion with parenting after controlling for children’s gender and parents’ education and depressive symptoms (study 1)
Step 1 | 31 | 0.16 | | 31 | 0.09 | | 30 | 0.15 | |
Children’s gender (1 = female, 2 = male) | | | −0.09 | | | −0.29 | | | 0.37 |
Parents’ education (0 = no qualification or diploma, 1 = degree or postgraduate) | | | 0.31 | | | −0.05 | | | 0.24 |
Parents’ depressive symptoms | | | 0.27 | | | 0.19 | | | −0.13 |
Step 2 | | 0.17a
| | | 0.15a
| | | 0.07 | |
Children’s gender (1 = female, 2 = male) | | | −0.07 | | | −0.27 | | | 0.35 |
Parents’ education (0 = no qualification or diploma, 1 = degree or postgraduate) | | | 0.31 | | | −0.05 | | | 0.26 |
Parents’ depressive symptoms | | | 0.09 | | | 0.02 | | | −0.03 |
Parents’ self-compassion | | | −0.45a
| | | −0.42a
| | | 0.29 |
Results
Table
3 presents participants’ characteristics. The mean age of the fathers and mothers was 39 and 36 years, respectively. The majority of participants were married or living together, had higher than average levels of education, and their ethnic background was White British. There was roughly equal numbers of boys and girls. Range, means, and standard deviations for study variables at times 1 and 2 are presented in Table
4. Table
5 presents correlations among study variables. Self-compassion correlated negatively with depressive symptoms (times 1 and 2) for mothers and fathers. Mothers’ self-compassion correlated with fewer critical (times 1 and 2) and more positive comments (time 1), fewer distressed (times 1 and 2), and more problem-focused reactions (time 1) while fathers’ self-compassion correlated with fewer distressed reactions (times 1 and 2). There were significant negative correlations of mothers’ self-compassion with child externalizing (times 1 and 2) and internalizing problems (time 1). Fathers’ self-compassion correlated with lower levels of child internalizing symptoms (times 1 and 2).
Table 3
Characteristics of participants (study 2)
Child gender at time 1 | | | |
Male N (%) | | | 75(47 %) |
Age (mean and SD) at time 1 | 38.8 (0.6) | 36.4 (0.5) | 3.9 (0.8) |
Marital statusa at time 1 |
Married or cohabitating | 148 (95 %) | |
Divorced, separated or single | 9 (5 %) | |
Parent educationb
N (%) at time 1 |
No qualifications, GCSE’s and A levels | 52 (33 %) | 49 (35 %) | |
Diploma or equivalent | 14 (9 %) | 7 (5 %) | |
Degree | 43 (27 %) | 44 (31 %) | |
Postgraduate degree | 48 (31 %) | 41 (29 %) | |
Ethnicityc at time 1 |
White British | 138 (95 %) | 130 (95 %) | |
Number of participants at time 1 | 160 | 146 | |
Number of participants with depression at time 1 | 40 (25 %) | 50 (34 %) | |
Number of participants at time 2 | 106 | 98 | |
Number of participants with depression at time 2 | 33 (31 %) | 37 (38 %) | |
Table 4
Range, means, and standard deviations for study variables (study 2)
Self-compassion (time 1) | 120 | 33–119 | 81.61 (17.33) | 133 | 34–122 | 85.15 (17.19) |
Current depressive symptoms (time 1) | 144 | 0–20 | 3.06 (3.55) | 157 | 0–21 | 3.69 (4.44) |
Current depressive symptoms (time 2) | 98 | 0–16 | 2.77 (3.24) | 104 | 0–20 | 3.21 (3.85) |
Critical comments (time 1) | 144 | 0–12 | 1.60 (2.36) | 158 | 0–15 | 1.37 (2.16) |
Critical comments (time 2) | 93 | 0–13 | 1.74 (2.43) | 105 | 0–9 | 1.54 (1.98) |
Positive comments (time 1) | 144 | 0–30 | 8.53 (5.14) | 158 | 0–25 | 7.56 (5.10) |
Positive comments (time 2) | 93 | 0–20 | 8.01 (4.87) | 105 | 0–18 | 7.36 (4.34) |
Distressed reactions (time 1) | 142 | 1–5.92 | 2.67 (.77) | 149 | 1–4.75 | 2.64 (.74) |
Distressed reactions (time 2) | 92 | 1.08–5.08 | 2.80 (.77) | 98 | 1.25–5 | 2.63 (.83) |
Problem focused reactions (time 1) | 142 | 3.50–7 | 5.85 (.59) | 148 | 3–7 | 5.62 (.68) |
Problem focused reactions (time 2) | 92 | 2.75–7 | 5.73 (.69) | 98 | 3.25–6.83 | 5.60 (.75) |
Internalizing symptoms (time 1) | 141 | 0–30 | 6.98 (5.64) | 149 | 0–36 | 6.96 (5.98) |
Externalizing symptoms (time 1) | 142 | 0–29 | 9.86 (6.76) | 148 | 0–40 | 10.46 (6.85) |
Internalizing symptoms (1½ -5) (time 2) | 63 | 0–26 | 6.03 (5.83) | 69 | 0–23 | 5.84 (4.78) |
Externalizing symptoms (1½ - 5) (time 2) | 63 | 0–36 | 8.14 (6.78) | 69 | 0–30 | 9.35 (6.76) |
Internalizing symptoms (6–18) (time 2) | 29 | 0–10 | 4.00 (2.98) | 29 | 0–13 | 3.45 (3.55) |
Externalizing symptoms (6–18) (time 2) | 29 | 0–25 | 7.07 (6.16) | 29 | 0–30 | 5.38 (6.59) |
Table 5
Correlations of self-compassion with parenting and child outcomes for mothers and fathers (study 2)
Depressive symptoms (time 1) | −0.50b
| −0.45b
|
Depressive symptoms (time 2) | −0.43b
| −0.54b
|
Critical comments (time 1) | −0.28b
| −0.15 |
Critical comments (time 2) | −0.30b
| −0.10 |
Positive comments (time 1) | 0.19a
| −0.02 |
Positive comments (time 2) | 0.18 | 0.11 |
Distressed reactions (Time 1) | −0.43b
| −0.36b
|
Distressed reactions (time 2) | −0.29a
| −0.32b
|
Problem-focused reactions (time 1) | 0.26b
| 0.11 |
Problem-focused reactions (time 2) | 0.12 | 0.20 |
Internalizing symptoms (time 1) | −0.20a
| −0.20a
|
Externalizing symptoms (time 1) | −0.29b
| −0.18 |
Internalizing symptoms (time 2) | −0.20 | −0.35b
|
Externalizing symptoms (time 2) | −0.31b
| −0.15 |
Mothers’ and fathers’ reports on children’s internalizing (time 1: r = 0.64, p < 0.01, and time 2: r = 0.53, p < 0.01) and externalizing problems (time 1: r = 0.54, p < 0.01, and time 2: r = 0.63, p < 0.01) correlated significantly as expected. We averaged mothers’ and fathers’ z-scores for each variable and used these four mean z-scores (child internalizing and externalizing problems at time 1 and child internalizing and externalizing problems at Time 2) for all subsequent analyses.
Table
6 presents the results of the hierarchical linear regressions for parenting outcomes. After controlling for confounders (child gender, parent education, and depressive symptoms), mothers’ self-compassion was associated with fewer critical comments (
β = −0.33,
p < 0.05). For fathers, after controlling for confounders, self-compassion was associated with fewer distressed (
β = −0.32,
p < 0.01) and more problem-focused reactions (
β = 0.29,
p < 0.05). The effect sizes for critical comments (
f
2 = 0.10), distressed reactions (
f
2 = 0.10), and problem solving (
f
2 = 0.06) were small to medium.
Table 6
Hierarchical linear regressions showing the longitudinal associations of parents’ self-compassion with parenting after controlling for children’s gender and parents’ education and depressive symptoms (study 2)
Mothers | 71 | | | 71 | | | 69 | | | 69 | | |
Step 1 | | 0.13a
| | | 0.02 | | | 0.18b
| | | 0.00 | |
Children’s gender (0 = female, 1 = male) | | | −0.25a
| | | −0.04 | | | −0.23a
| | | 0.03 |
Mothers’ education (0 = no degree, 1 = degree) | | | −0.22 | | | 0.02 | | | 0.20 | | | 0.03 |
Mothers’ depressive symptoms (time 1) | | | 0.06 | | | −0.13 | | | 0.31b
| | | 0.08 |
Step 2 | | 0.08a
| | | 0.04 | | | 0.02 | | | 0.02 | |
Children’s gender (0 = female, 1 = male) | | | −0.21 | | | −0.07 | | | −0.22 | | | 0.01 |
Mothers’ education (0 = no degree, 1 = degree) | | | −0.24a
| | | 0.04 | | | 0.19 | | | 0.04 |
Mothers’ depressive symptoms (time 1) | | | −0.09 | | | −0.03 | | | 0.23 | | | 0.16 |
Mothers’ self-compassion (time 1) | | | −0.33a
| | | 0.22 | | | −0.17 | | | 0.17 |
Fathers | 87 | | | 87 | | | 82 | | | 82 | | |
Step 1 | | 0.03 | | | 0.00 | | | 0.13a
| | | 0.04 | |
Children’s gender (0 = female, 1 = male) | | | −0.07 | | | −0.01 | | | 0.06 | | | −0.18 |
Fathers’ education (0 = no degree, 1 = degree) | | | 0.14 | | | −0.07 | | | 0.21 | | | 0.10 |
Fathers’ depressive symptoms (time 1) | | | 0.08 | | | −0.05 | | | 0.28a
| | | 0.10 |
Step 2 | | 0.00 | | | 0.01 | | | 0.08b
| | | 0.06a
| |
Children’s gender (0 = female, 1 = male) | | | −0.07 | | | 0.01 | | | 0.02 | | | −0.14 |
Fathers’ education (0 = no degree, 1 = degree) | | | 0.14 | | | −0.10 | | | 0.29a
| | | 0.03 |
Fathers’ depressive symptoms (time 1) | | | 0.09 | | | 0.01 | | | 0.14 | | | 0.23 |
Fathers’ self-compassion (time 1) | | | 0.03 | | | 0.14 | | | −0.32b
| | | 0.29a
|
In the final models, among confounders, only parent education was significant. Mothers’ higher education was associated with fewer critical comments (β = −0.24, p < 0.05) while fathers’ higher education was associated with more distressed reactions (β = 0.29, p < 0.05). When we controlled for multiple comparisons, the associations were significant for mothers’ critical comments and fathers’ distressed reactions.
When the dependent variable was children’s outcomes (Table
7), the results showed that after controlling for confounders (children’s gender, baseline problems, and mothers’ education and depressive symptoms), mothers’ self-compassion was not significantly associated with children’s outcomes. For fathers, self-compassion was inversely correlated with internalizing problems even when controlling for children’s gender, baseline emotional symptoms, and fathers’ education and depressive symptoms (
β = −0.22,
p < 0.05). The effect size was small to medium (
f
2 = 0.07). However, when we controlled for multiple comparisons, this association was no longer significant. For both parents, baseline child outcomes (time 1) were positively associated with emotional and behavioral problems at time 2 (
p < 0.001).
Table 7
Hierarchical linear regressions showing the longitudinal associations of parents’ self-compassion with children’s emotional and behavioral problems at time 2 after controlling for children’s gender and baseline problems (time 1) and parents’ education and depressive symptoms (study 2)
Mothers | 67 | | | 68 | | |
Step 1 | | 0.58b
| | | 0.45b
| |
Children’s gender (0 = female, 1 = male) | | | −0.04 | | | −0.07 |
Mothers’ education (0 = no degree, 1 = degree) | | | −0.05 | | | −0.16 |
Mothers’ depressive symptoms (time 1) | | | 0.08 | | | 0.12 |
Children’s outcomes (time 1) | | | 0.72b
| | | 0.59b
|
Step 2 | | 0.00 | | | 0.00 | |
Children’s gender (0 = female, 1 = male) | | | −0.04 | | | −0.07 |
Mothers’ education (0 = no degree, 1 = degree) | | | −0.05 | | | −0.16 |
Mothers’ depressive symptoms (time 1) | | | 0.07 | | | 0.10 |
Children’s outcomes (time 1) | | | 0.72b
| | | 0.58b
|
Mothers’ self-compassion (time 1) | | | −0.008 | | | −0.05 |
Fathers | 72 | | | 73 | | |
Step 1 | | 0.61b
| | | 0.44b
| |
Children’s gender (0 = female, 1 = male) | | | −0.11 | | | −0.11 |
Fathers’ education (0 = no degree, 1 = degree) | | | 0.15 | | | 0.09 |
Fathers’ depressive symptoms (time 1) | | | 0.19a
| | | 0.19a
|
Children’s outcomes (time 1) | | | 0.75b
| | | 0.60b
|
Step 2 | | 0.00 | | | 0.04a
| |
Children’s gender (0 = female, 1 = male) | | | −0.11 | | | −0.14 |
Fathers’ education (0 = no degree, 1 = degree) | | | 0.16 | | | 0.14 |
Fathers’ depressive symptoms (time 1) | | | 0.16 | | | 0.10 |
Children’s outcomes (time 1) | | | 0.75b
| | | 0.55b
|
Fathers’ self-compassion (time 1) | | | −0.05 | | | −0.22a
|
Discussion
Study 2 suggests an association of self-compassion with parenting behaviors in parents with a history of depression. Greater self-compassion was associated with lower levels of mothers’ child-directed criticism and fathers’ distressed reactions. While the study revealed significant longitudinal associations between fathers’ self-compassion and lower levels of emotional problems among their children, these associations become nonsignificant in the regression analysis.
General Discussion
Our findings provide support for the hypothesis that self-compassion may be associated with lower levels of mothers’ child-directed criticism and fathers’ distressed reactions. Contrary to the study hypotheses, there were no significant associations of self-compassion with parents’ positive comments and problem-focused responses. While correlational analysis revealed significant negative associations of fathers’ self-compassion with children’s internalizing problems, these associations became nonsignificant in the regression analysis. Before discussing the implications of these findings, we review methodological issues raised by the work. A strength is that parenting was examined using observations, questionnaires, and interviews. Both studies included samples of parents with a research diagnosis of a history of depression and study 2 included a sufficient sample of fathers for an analysis stratified by parental gender. Triangulation across two samples allowed us to test whether self-compassion was associated with parenting behaviors in two samples of parents, including some with current and/or a history of depression for whom parenting can be challenging. However, both studies also had limitations. Study 1 was cross-sectional and study 2, while having more than one time point, was correlational; therefore, no causal inferences can be drawn. In both studies, the sample included a relatively small number of parents with a history of depression, and so we may have lacked sufficient power to identify all important relationships of parenting with self-compassion. Furthermore, there were only seven fathers with a current depressive illness in study 2, so we could not examine whether the associations were more evident in currently depressed versus all parents with a history of depression. The sample was more educated than a general community sample, which limits generalizability. Effect sizes were small to medium, suggesting that self-compassion is one out of many other factors that predict these parenting behaviors. Study 2 recruited mothers through fathers’ participation, which may have biased the findings. Finally, the large number of statistical tests conducted increased the possibility of type I error, which we addressed by correcting the level of significance using Bonferroni. Nonetheless, the pattern of findings is similar across both samples, which suggests that the association of self-compassion with these parenting attributes is likely to be replicable.
Overall, the findings provide some support for our hypothesis that individual differences in self-compassion are associated with fewer fathers’ distressed reactions to their children’s negative emotions. Previous studies have found that parents who use positive responses to cope with their children’s negative emotions have children who are better able to regulate their emotions than parents who react negatively and increase the child’s arousal (Eisenberg et al.
1996; Fabes et al.
2001,
2002). Parents who display fewer distressed reactions may be more able to validate their children’s emotions and convey the message that these emotions are understandable, while children may learn to express rather than suppress these emotions (Fabes et al.
2001). In turn, children may become better able to regulate these emotions in similar situations, which may decrease their risk of emotional problems. When we adjusted for multiple comparisons on multivariate analysis, the link of self-compassion with fathers’ problem-focused responses became nonsignificant. Nonetheless, our findings provide promising avenues for future research using larger scale and experimental/interventional designs.
Our study suggests that higher levels of self-compassion are associated with lower levels of maternal criticism. Previous studies have found that depressed parents are more critical and less positive (Barnes et al.
2007; Psychogiou et al.
2013; Rogosch et al.
2004) and that high EE is associated with children’s emotional and behavioral problems (e.g., Cartwright et al.
2011; Nelson et al.
2003; Peris and Baker
2000; Psychogiou et al.
2007). Although our study could not explore the mechanisms that may explain this association, it is possible that greater self-compassion may enable parents to recognize, be kind toward and de-center from negative thoughts and the impulse to react with criticism. It enables parents to respond to both their negative thoughts and their children’s needs more adaptively. The pattern of associations across self-compassion and parenting behaviors, both positive and negative, converges on this explanation. This could be tested empirically among parents at risk for depression randomized to interventions that either seek to increase self-compassion or build resilience through an alternative mechanism (e.g., cognitive therapy). If future research suggests that increasing self-compassion reduces EE, interventions that can target either or both might improve parent and child outcomes.
Significant correlations between self-compassion and lower levels of children’s emotional and behavioral problems were found. While it should be interpreted with caution, study 2 indicates that fathers’, but not mothers’, self-compassion was significantly associated with lower levels of internalizing problems among their children at time 2 when controlling for depressive symptoms and earlier child internalizing symptoms. However, when we controlled for multiple comparisons, this association became nonsignificant. Given the significant longitudinal associations in the correlational analysis, perhaps an important direction for future research is to establish if parents’ self-compassion is related to lower levels of child psychopathology and if so, how and why such an association occurs. As we discussed earlier, children of parents with greater self-compassion may receive more positive parenting or they may observe and adopt a parent’s capacity to self-soothe in difficult times. Gilbert et al. (
2006) have suggested that children exposed to criticism may develop self-evaluative scripts viewing their parents as hostile and become oversensitive and vulnerable to actual and perceived rejection. In contrast, children not exposed to criticism may develop a sense of being worthy of support and in times of stress, they may be better able to self-soothe. An important direction for future research is to test these hypotheses.
The findings have potential implications for future research and interventions that target depressed parents and their children. Interventions that aim to increase parents’ self-compassion may result in decreased parental depressive symptoms and/or improved child mental health. Kuyken et al. (
2010) found that increases in self-compassion through mindfulness-based cognitive therapy among people with a history of depression were associated with fewer depressive symptoms 15 months later. In a qualitative study, parents who took part in mindfulness-based cognitive therapy reported that they were better able to recognize their own needs and regulate their emotions. They also reported that they were more emotionally available and empathetic toward their children (Bailie et al.
2012). Given that fathers’ self-compassion was significantly associated with parenting, parenting programs should encourage the attendance of both parents.
In the future, these findings need to be replicated with other populations (e.g., parents with bipolar, schizophrenia, or substance abuse/dependence disorders) to examine their specificity to depression and with other dimensions of parenting behaviors. There is also a need for research designs that include both depressed mothers and fathers to uncover similarities and differences in the associations of self-compassion with parenting.