Maternal and paternal insightfulness and reaction to the diagnosis in families of preschoolers with Autism spectrum disorder: associations with observed parental sensitivity and inter-parent interaction

ABSTRACT

Sensitivity among parents of children with Autism Spectrum Disorder (ASD) is based on parental insightfulness and on resolution regarding the child’s diagnosis. This has been supported in studies of mothers, and we examined whether the same is true regarding fathers. Also, we asked whether parents’ Insightfulness and Resolution tap general mentalization and therefore also be expressed in parent-parent interactions. Eighty preschooler boys with ASD and both of their parents participated. As expected, fathers who were more insightful and mothers who were more resolved were more sensitive. Contrary to expectations, no associations were found between fathers’ resolution and mothers’ insightfulness and their sensitivity. Associations were found between parental insightfulness and resolution and positive parent-parent interaction. The findings are the first to demonstrate the insightfulness-sensitivity link among fathers of children with ASD. Also, they suggest that both insightfulness and resolution involve general mentalization that is evident both in parent-child and parent-parent interactions.

KEYWORDS:

• Autism spectrum disorder
• insightfulness
• resolution of child diagnosis
• fathers
• mentalization

Sensitivity is based on the parent’s capacity for insightfulness, which involves “seeing things from the child’s point of view” (Oppenheim & Koren-Karie, 2002). In the case of children who have a significant developmental diagnosis such as Autism Spectrum Disorder (ASD), sensitivity is also influenced by the degree to which parents have come to terms, or “resolved” their emotional reactions to their child’s diagnosis (Marvin & Pianta, 1996). A few studies provided support for these ideas with regard to mothers (Oppenheim et al., 2010; Oppenheim, Dolev, et al., 2009; Sher-Censor & Shahar-Lahav, 2022), but none involved fathers. In fact, one limitation of most research on parent-child interactions in ASD is its almost sole focus on the mother-child dyad (Crowell et al., 2019). Therefore, the study’s first goal was to examine whether the findings linking insightfulness with sensitivity among mothers of preschoolers with ASD also apply to fathers. The second goal was to examine whether the findings linking the resolution of mothers of preschoolers with ASD to their sensitivity also apply to fathers. The third goal was to examine whether paternal and maternal insightfulness and resolution are associated only with parental sensitivity when interacting with the child, or whether they reflect a more general capacity for mentalization and therefore also expressed in parent-parent interaction, outside the parent’s interaction with the child.

Parental insightfulness and sensitivity

Parental insightfulness involves “seeing things from the child’s point of view” and empathically understanding the motives underlying the child’s experience (Oppenheim & Koren-Karie, 2002). Insightful parents focus on the motives underlying the child’s behavior, show acceptance regarding challenging or disappointing behavior, hold a multidimensional view of the child, and are flexible in revising their representation of the child and of themselves in light of new information. Insightfulness is thought to provide the foundation for sensitive parental behavior, namely accurate reading of the child’s signals and the appropriate responsiveness to them (Ainsworth et al., 1978). This was supported in studies linking maternal insightfulness and sensitivity in typically developing (TD) children (Koren-Karie et al., 2002), children with Intellectual Disability (Feniger-Schaal et al., 2019), and, most pertinent to the present study, children with ASD (Oppenheim et al., 2010). It was also found regarding constructs related to sensitivity. For example, Yuval-Adler (2010) found that insightfulness was associated with foster mothers’ sensitive guidance of emotion dialogues, and Martinez-Torteya et al. (2018) found in a sample of mothers at heightened risk for parenting difficulties based on their adverse childhood histories, that insightful mothers showed more positive parental behavior when interacting with their children than non-insightful mothers (see also Guédeney et al., 2014). Finally, in a sample of children with ASD that assessed maternal synchrony, an aspect of sensitivity, Hutman et al. (2009) found that insightful mothers were more synchronous when interacting with their children than non-insightful mothers.

Although fathers play an increasingly active role in parenting in recent decades (Bakermans‐Kranenburg et al., 2019), there is very little research on paternal insightfulness and particularly on its links with paternal sensitivity. One study with TD toddlers reported that fathers were as likely as mothers to show insightfulness, but the study did not assess sensitivity (Marcu et al., 2016). Even when considering mentalization constructs related to insightfulness, such as paternal Reflective Functioning or Mind Mindedness (Zeegers et al., 2017), very few studies with fathers have been done. Zeegers et al. (2017) meta-analysis found no studies of paternal mentalization and sensitivity, and since that meta-analysis only two additional relevant studies have been conducted: In a study of TD children, Buttitta et al. (2019) found that paternal Reflective Functioning (RF) was associated with observed paternal “socioemotional support” during a teaching task, and Dinzinger et al. (2023) similarly found that first-time fathers’ RF was associated with their sensitivity. Thus, even when broadening the scope to mentalization measures other than insightfulness, only very few studies involved fathers. With respect to paternal mentalization and sensitivity in children with ASD, no studies have been conducted. The first goal of the study, therefore, was to examine whether, as is the case with mothers, insightful fathers of children with ASD are more sensitive than non-insightful fathers.

Parental resolution

As mentioned above, an important additional factor likely to impact parental sensitivity in the case of children with an ASD diagnosis is the degree to which parents have come to terms with their child’s diagnosis. This involves adjusting their view of the child, including their feelings and expectations, in light of the diagnosis the child received (Sher-Censor & Shahar-Lahav, 2022). This adjustment is referred to as resolution because receiving a diagnosis is seen as a crisis, followed by a period of grieving the loss of the wished for typically-developing child (Marvin & Pianta, 1996). Parents who are more resolved regarding their grief reactions experience over time less intense negative feelings and more positive feelings about the child. They integrate the reality of the child’s diagnosis into their view of the child, so that it is not dominated by the child’s diagnosis and is more nuanced, including strengths and positive aspects. Parents who are less resolved may be overwhelmed by anger or despair or, conversely, appear emotionally detached; overly focus on the past (e.g. anger about how the diagnosis was delivered) or future; or show confused or disorganized thinking about the diagnosis.

Because Marvin and Pianta’s (1996) model was rooted in attachment theory, initial studies examined the association between maternal resolution and child attachment. Marvin and Pianta (1996) found that mothers classified as resolved were more likely to have children who were securely attached to them. A few subsequent studies provided further support to these early findings (Barnett et al., 1999, 2006; Shah et al., 2011) including one study involving children with ASD (Oppenheim, Dolev, et al., 2009). Marvin and Pianta (1996) explained that by seeing the child beyond the diagnosis, parents who are resolved can respond in a more sensitive and appropriate way to the child’s signals and communications, thus supporting the development of a secure attachment. Two studies among mothers supported this hypothesis (Dolev et al., 2016; Feniger-Schaal & Oppenheim, 2012) but no studies with fathers linking resolution and sensitivity have been conducted. The second goal of the study, therefore, was to bridge this gap.

Insightfulness and resolution as measures of parental mentalization

The third goal of the study, as mentioned briefly above, was to examine whether parental insightfulness and resolution are associated only with parents’ behavior when interacting with their children (a “narrow” view), or whether they are an expression of a more general mentalization capacity (Fonagy et al., 2016; Slade, 2005) which will also be expressed outside the parent-child relationship, such as when parents interact with one another (a “broad” view). Whereas insightfulness is clearly a measure of mentalization because it involves the parent’s thinking about the motives, thoughts, and feelings underlying the child’s behavior (Oppenheim & Koren-Karie, 2002; Zeegers et al., 2017), resolution was not originally conceptualized within the mentalization framework. However, examination of the definition of resolution provided by Marvin and Pianta (1996) reveals that it clearly involves mentalization. For example, integral to resolution is the parents’ understanding of their initial emotional reactions to receiving the diagnosis and how these changed over time, how their emotional reactions impacted their view of the child, and how, as their emotions changed so did their view and expectations of the child. Another aspect of mentalization integral to resolution is the parents’ awareness of shifts in their cognitive processes (e.g. from an excessive search for the causes or even self-blame to focusing primarily on the present and what can be done to help the child).

Surprisingly, there is only very little empirical evidence that can help differentiate between the “narrow” and “wide” views regarding mentalization mentioned above. In a sample of twins in which parental RF was assessed, Tohme (2014) found significant correlations between the RF scores for the two twins. Koren-Karie and Markman-Gefen (2016) examined the insightfulness of foster mothers regarding two of the children under the mothers’ care that differed in the level of challenge they presented (high vs. low). The findings showed consistency of insightfulness (or lack thereof) for the two children. Thus, the little research available regarding the consistency of mentalization across relationships provides some support to the “broad” view.

Importantly, while both insightfulness and resolution appear to have in common mentalization processes they are also distinct constructs: Insightfulness focuses on understanding the motives underlying the child’s behavior in the present, and does not necessarily involve discussion of negative emotions. In contrast, resolution focuses on the parents’ thoughts and feelings around a difficult moment in the past in which the parents received a diagnosis for the child. It is therefore possible for a parent to be insightful (e.g. by speaking in an organized and coherent way about the motives underlying the child’s behavior), but unresolved (e.g. be flooded by anger or pain when talking about the child’s diagnosis). Conversely, a parent may be non-insightful (e.g. express disappointment, detachment, or anger when talking about the child’s thoughts and feelings in a given moment), but be resolved (e.g. have an organized and emotionally-regulated narrative regarding the emotional reaction to receiving the diagnosis for the child and the parent’s subsequent adaptation). The common and distinct aspects of insightfulness and resolution lead to expect partial overlap between them, as was found by Oppenheim, Dolev, et al. (2009).

The present study

The first goal of the study was to examine whether mothers’ and fathers’ insightfulness with respect to their children with an ASD diagnosis would be associated with their sensitivity when interacting with the child. The hypothesis was that mothers and fathers who are more insightful will be more sensitive than those who are less insightful. The second goal was to examine whether mothers’ and fathers’ resolution with respect to their children’s diagnosis would also be associated with their sensitivity when interacting with the child. The hypothesis was that mothers and fathers who are more resolved will be more sensitive than those and less resolved. Finally, the third goal was to examine whether parents’ insightfulness and resolution would be associated only with their sensitivity during parent-child interactions, supporting the “narrow” view described above, or also with their interaction with the other parent during a parent-parent discussion regarding the child, supporting the “broad” view. The hypothesis was that the “broad” view will receive support.

Three potential covariates were assessed. Children’s level of cognitive delay and the severity of their symptoms were likely to impact their interactions with their parents and possibly impact parental insightfulness and resolution and were therefore measured. In addition, parents’ Broad Autism Phenotype (BAP; Rubenstein & Chawla, 2018), sub-clinical characteristics reflecting Autism traits that are more prevalent in family members of individuals diagnosed with ASD was also measured, because it could impact parents’ behavior and their insightfulness and resolution.

Method

Participants

Eighty preschooler boys with ASD and their parents participated in the study, which was part of a larger project (Oppenheim et al., 2023). Only boys were included because the ASD diagnosis is more common in boys at a 4:1 ratio (American Psychiatric Association, 2013). Children’s ages ranged between 29 and 68 months (M = 49.45, SD = 11.03). Mothers’ age ranged between 25 and 47 years (M = 36.62, SD = 5.24) and fathers’ age ranged between 29 and 57 years (M = 39.80, SD = 6.29). Mothers and fathers had a mean of 15.31 (SD = 3.11) and 15.09 (SD = 2.79) years of education respectively. Thirty-six of the children (45%) were first born. After preschool hours, children were cared for by their mothers in 80.0% of the families, by a relative or professional caregiver in 11.7%, and by fathers in 8.3%. Families were recruited through treatment centers for children with ASD and through social media. Inclusion criteria were being male, a diagnosis of ASD known for at least 6 months, child age between 2.5 and 6 years, no known additional medical diagnoses, and biological and married parents.

Procedure

The study included two laboratory visits: The first included the Autism Diagnostic Observation Schedule (ADOS; Lord et al., 2000), a parental conversation task for assessing parent-parent interaction, filming of interactions for the Insightfulness Assessment (IA; Oppenheim & Koren-Karie, 2002; see description below), and an observation of the family which was not part of this study. The second laboratory visit, conducted 1–2 weeks later, included the IA and the Reaction to Diagnosis Interview (RDI; Marvin & Pianta, 1996) for assessing parental resolution with respect to the child’s diagnosis. The RDI was administered to both parents simultaneously and individually. The visit also included an assessment of children’s cognitive functioning, and observations of the children interacting with each of their parents (separately) to assess parental sensitivity. Parental questionnaires were completed during the laboratory visits. The study received approval from the University of Haifa IRB (400/16). All parents signed consent forms.

Child and parent control measures

Autism Diagnostic Observation Schedule (ADOS; Lord et al., 2000)

The ADOS is a standardized and widely used assessment for diagnosing individuals with ASD through interactions with a trained professional. It consists of 4 modules of which one is administered based on the child’s level of expressive language. Thirty-five children were assessed using module 1, 39 using module 2, and 6 using module 3. The ADOS diagnostic algorithm was used (Gotham et al., 2007) and a Calibrated Severity Score (CSS; Gotham et al., 2009) was calculated (range: 4–10; see also Table 1). Reliability of the blind assessors on 17 cases (21% of the sample) was 87.4% agreement (range: 78% to 94%).

Table 1. Descriptive statistics of study variables and comparisons of mothers and fathers.

	Mothers	Fathers
Study Variables	M (SD)	M (SD)	t value
Insightfulness	5.61 (1.51)	5.04 (1.75)	t (78) = 2.68, p < .001
Resolution	4.17 (1.17)	4.07 (1.29)	t (78) < 1, ns
Sensitivity	4.97 (.84)	5.04 (.87)	t (76) < 1, ns
Parent-parent interaction	.011 (1.00)
Control Measures
Parental ^aBAPQ	1.42 (.58)	2.65 (.50)	t (78) = 1.48, p = .14
Child ^bCSS	7.01 (1.58)
Child IQ	68.01 (25.01)

^aBAPQ – Broad Autism Phenotype Questionnaire.

^bCSS – ADOS Calibrated Severity Score.

Social Communication Questionnaire (SCQ; Rutter et al., 2003)

The SCQ is a 40-item parental questionnaire about children’s autism symptoms. The SCQ items are scored as present (1) or absent (0) and summed, with the cut-off for ASD for young children ≥ 11 (Wiggins et al., 2015).

Children’s ASD diagnosis was confirmed based on the ADOS, the SCQ, and clinical judgement of a licensed psychologist with extensive experience evaluating and diagnosing children with ASD. In all cases the diagnoses that the children received in the community were confirmed.

Assessments of children’s cognitive functioning

The Wechsler Preschool and Primary Scale of Intelligence – Third Edition (WPPSI – III; Wechsler, 2002) is a widely used and standardized cognitive assessment for children between the ages of 2:06 and 7:07 years which yields an IQ score and was used for verbal children. Non-verbal children were assessed using the Mullen Scales of Early Learning (MSEL; Mullen, 1995), also a widely used and standardized assessment of children’s cognitive functioning. Due to the very low score of many of the children on the four MSEL scales (below 1%ile), a standardized Developmental Quotient (DQ) score was calculated for each scale by dividing children’s mental age by their chronological age and multiplying the result by 100. Children’s overall DQ score was calculated by averaging the four DQ scales.

The broad autism phenotype questionnaire

The Broad Autism Phenotype Questionnaire (BAPQ: Hurley et al., 2007). The BAPQ was completed by each parent to assesses parents’ sub-clinical Autism related traits. It includes 36 items rated on a 1 (very rarely) to 6 (very often) scale tapping three dimensions: Aloofness, rigidity, and pragmatic language difficulties. Internal consistency of the scale was high (α = 0.90 for mothers and α = 0.86 for fathers).

Study measures

The Insightfulness Assessment (IA; Oppenheim & Koren-Karie, 2002)

In the IA, the parent was shown three two-minute video segments: In the first segment both parents were asked to complete questionnaires in the child’s presence; in the second the child had a snack with his parents; and in the third the child played with an examiner without the parents’ presence. Each parent was shown the three video segments and asked what s/he thought “went through his/her child’s head, what he was thinking and feeling?,” if the behaviors observed were typical, and how s/he felt when s/he watched the segment. Following the responses to each of the segments the parent was asked if “in general” the behavior observed was typical of the child and what the parent felt observing the child. Throughout the interview the parent was asked to support his/her statements with examples from the observation and from everyday life.

Coding parental insightfulness

The interviews were transcribed verbatim, and all identifying information was removed. Because in Hebrew the gender of the speaker is revealed in the grammatical form of the verbs, all interviews were transcribed in the female voice so that the coders were blind to parental gender. Next, transcripts were coded on ten 1–9 scales: Insight into child’s motives (Attempts to understand the thoughts and feelings which may underlie child’s behavior), Openness (Adjusting and changing the view of the child based on new information arising from the video-taped observations), Complexity in description of child (Referring to a wide range of behaviors and characteristics including positive but also challenging aspects, and considering the context in which the behavior took place), Maintenance of focus on child (Keeping the focus of discussion on the child and not digressing to irrelevant issues or the parent’s feelings), Richness of description of child (Full, comprehensive, and vivid responses), Coherence of thought (Development of ideas in a consistent, connected, clear, and logical way), Acceptance/warmth (Acceptance and tolerance of the full range of child’s behaviors), Anger (Current anger towards the child expressed directly or by blaming the child), Concern (Maternal concerns about the child or the self are a central, repetitive theme throughout the interview), Separateness from child (Child is seen as a separate person with needs and wishes that are different or even contradictory to the parent’s). The IA scales are usually used in order to classify the transcripts into insightfulness categories, but several studies (e.g. Koren-Karie & Getzler-Yosef, 2019; Oppenheim et al., 2001; Siller et al., 2018) used an Insightfulness continuous composite score in order to increase statistical power. The score was based on the mean of the following scales: Insight, Complexity, Acceptance, Openness, Richness, and Coherence, and was used in the study analyses.

The IA coding was conducted by the second author (the co-developer of the IA) and a trained reliable coder. Both were only exposed to the de-identified transcripts and consequently blind to any additional information about the families. One half of the transcripts coded by each coder was of mothers and the other half was of fathers, and in no case did a coder code both members of a couple. Twenty percent of the transcripts were double coded. Intra Class Correlations of the IA scales ranged between .62 and 1.0 with a median of .86. Disagreements were resolved through discussion. One paternal interview was lost due to equipment failure.

Reaction to Diagnosis Interview (RDI; Marvin & Pianta, 1996)

The RDI is a semi-structured interview regarding parents’ feelings and thoughts about their child’s diagnosis. Parents are asked to remember when they first observed difficulties in their child’s development; How they felt when they were given the diagnosis for their child and what were their thoughts; How their feelings and thoughts changed from the day they received the child’s diagnosis until the present; and whether they have thoughts about the causes for their child’s ASD. Interviews were transcribed verbatim in the female voice for the reasons described above. The RDI is usually classified dichotomously as Resolved or Unresolved, with each category defined by a set of criteria (Marvin & Pianta, 1996). Based on Wachtel and Carter (2008) and in order to increase statistical power, the RDI criteria were converted into a set of 1–7 rating scales, with higher scores reflecting more resolved responses (Slonim et al., 2023). The scales were: Positive change in the feelings regarding the diagnosis (openness regarding the negative feelings in response to receiving the diagnosis coupled with reduction in these feelings over time), Coherence (consistency and logical sequence in parental response and providing supportive examples), Regulated emotional expression (talking about a wide range of emotional experiences without being flooded or detached), Present orientation (talk is focused primarily on the present without excessive attention to the past or future), Absence of search for reasons for the diagnosis (parent is not overly focused on finding the causes for the child’s diagnosis or on self-blame), Sense of competence (parent is involved in problem-solving the practical challenges associated with the child’s diagnosis and feels competent in addressing these issues), Overall resolution (overall score indicating the cognitive and emotional adaptation to the reality of the child’s diagnosis). De-identified transcripts were coded by blind and independent coders. Inter-rater reliability was computed on 20% of the transcripts and averaged ICC = .87. Because the scales were significantly correlated (r = .32 to r = .74, mean r = .5(1, a mean Resolution composite score was computed by averaging the scale scores.

Parental sensitivity

Parental sensitivity was assessed from two 7-minute observations with each parent: Free play (with toys) and social play (without toys). The parent and child were instructed to “play together as they usually do at home.” Order of parents and order of type of play were counterbalanced. Sensitivity was assessed using the 1–7 sensitivity scale which is part of the Emotional Availability Scales (Biringen et al., 2014). Parental sensitivity is defined as the parent’s attunement and responsiveness to the child’s signals, while expressing appropriate affect and warmth. The scale is based on the following elements: balanced and genuine affect; clarity of perceptions and appropriate responsiveness; awareness of timing; creativity and flexibility in interaction; acceptance; appropriate amount of interaction; and resolution of conflict situations.

The interactions were coded by two trained, blinded, and independent coders. To ensure the independence of the maternal and paternal coding, each child’s interactions with their mother and father were coded by a different coder. Each coder coded half of the interactions with the mothers and half of the interactions with the fathers so as not to confound coder with the parent being father or mother. Inter-rater reliability was established between the coders and a primary coder based on 25% of the sample and was good (ICC = .81). Parents’ sensitivity in the Free and Social Play observations were significantly correlated (r = .28, p < .001 and r = .45, p < .001 for mothers and fathers respectively) and were therefore averaged.

Parent-parent interaction

Parent-parent interaction was assessed in a conversation task developed for this study in which parents were asked to hold three 5-minute conversations regarding their child (Slonim et al., 2023). Each conversation was initiated by the experimenter entering the room, presenting the parents with the instructions, and then leaving the room for 5 minutes. During the first segment parents were requested to share with one another “something from the recent past that made each one of you happy regarding your child.” During the second segment, parents were requested to share with one another “something from the recent past that is difficult or challenging for each one of you regarding your child,” and during the third segment parents were requested to “discuss an issue regarding their child on which they disagree.” Parents were told that they have five minutes for each conversation. The conversations were video-taped and transcribed verbatim in the female voice for the reasons described above. In addition to the verbatim, the transcripts included indications of the following behaviours: nodding, shaking one’s head, rolling one’s eyes, heavy sighs, hand movement indicating dismissal of the other person’s comment, turning one’s back toward the partner, looking away from the partner for more than 10 seconds, crying, significant change in tone of voice (e.g. laughing or whispering), engagement with one’s cellular telephone, and additional unusual behaviors (e.g. talking to the camera).

The transcripts were coded using a coding system based on Darwiche (2017) and adapted to this task. The coding system included the following scales ranging from 1 (“low”) to 7 (“high”); mid-point scores were assigned when one partner was high on the dimension tapped by the scale and the other was low: Emotional togetherness – positive emotional connection between the parents, expressed in empathy, validation, and warmth; Competition - competition between the partners regarding who is a more knowledgeable, important, and competent; Endorsement - Endorsement of the other parent’s point of view or behavior; Positive mutual investment – Both parents are positively invested in the conversation; Positive representation of the child – Both parents speak about the child in a rich, multi-dimensional, and accepting way; Accommodation - Parents deal with parenting disagreements in a respectful and cooperative manner; Hostility/Anger - Hostility or tension between the parents; Openness - Willingness to hear and understand the partner’s perspective and change one’s behaviors accordingly. On all scales higher scores represent positive parental interaction, except for the Competition and Hostility/Anger scales in which high scores represent negative parental interaction. Two blind independent coders coded the transcripts and inter-rater reliability assessed on 35% (n = 28) of the transcripts using Intraclass Correlation Coefficients ranged between 0.82 and 0.96 with a median of .92. Coding disagreements were resolved by consensus. The two scales in which high scores reflected negative parental communication (Competition, Hostility/Anger) were reversed so that high scores would reflect positive parent-parent interaction. Correlations between the scales were strong and significant (ranging between r = 0.57 to r = 0.83, p < 0.01), and therefore an overall inter-parent interaction score was computed by averaging the scales, with higher scores reflecting more positive parent-parent interaction. Support for the validity of this task was provided by its associations with parents’ self-reported co-parenting and observations of the parents during mother-father-child interactions (Slonim et al., 2023).

Data analysis plan

Descriptive statistics of the study variables were examined first, including comparisons of maternal and paternal scores. Second, the associations between the study variables and demographic and control variables were examined. Third, to examine the hypotheses the parental insightfulness and resolution are expressed in parental sensitivity, bivariate correlations between each parents’ insightfulness and resolution and the parent’s sensitivity were examined. The same correlations except with parents’ interaction with one another were also computed to examine whether the insightfulness and resolution scores are associated only with parent-child interactions or also with parent-parent interactions. Finally, regression analyses to predict each parent’s sensitivity from the parent’s insightfulness and resolution while controlling for the appropriate covariates were conducted. An identical regression only predicting parent-parent interaction was also conducted.

Results

Table 1 provides the descriptive statistics of the study variables and comparisons between mothers and fathers. Because one paternal IA interview was lost (as mentioned above) the family was dropped from the study and the analyses refer to 79 families. Mothers were more insightful than fathers, but there were no differences in parental resolution and sensitivity between mothers and fathers, nor differences in the BAPQ scores. Next the associations between the study variables (maternal and paternal insightfulness, resolution, sensitivity, and positive parent-parent interaction) and demographic and control variables (parental education, parental BAPQ, child age, birth order, IQ and CSS) were examined. None were significant with regard to the insightfulness and resolution of both mothers and fathers except that mothers with higher BAPQ scores had lower resolution scores (r = −.27, p = .01). Maternal and paternal sensitivity were positively associated with child IQ (r = .22, p < .05 and r = .24, p < .05 for mothers and fathers respectively), and positive parent-parent interaction was positively associated with maternal and paternal education (r = .23, p < .05 and r = .25, p < .05 for mothers and fathers respectively). Therefore, IQ was controlled in the analyses involving sensitivity and parental years of education was controlled in the analyses involving positive parent-parent interaction.

Predicting sensitivity and positive parent-parent interaction from insightfulness and resolution

Bivariate correlations between maternal and paternal insightfulness and resolution and maternal and paternal sensitivity as well as positive parent-parent interaction are presented in Table 2. As can be seen, maternal sensitivity was associated with maternal resolution but not insightfulness, and paternal sensitivity was associated with paternal insightfulness but not resolution. Positive parent-parent interaction was associated with both parents’ insightfulness and resolution.

Table 2. Associations between the study variables (N = 79).

	Maternal Sensitivity	Paternal sensitivity	Parent-parent interaction
Maternal Insightfulness	−.07	.23*	.45**
Paternal Insightfulness	.11	.35**	.30**
Maternal Resolution	.29**	.04	.23*
Paternal Resolution	.12	.15	.18*

Note: *p < .05. **p < .01.

Next, hierarchical regression analyses were conducted to predict each parent’s sensitivity from the parent’s insightfulness and resolution while controlling for child IQ. The regression model predicting maternal sensitivity revealed that, after controlling for child IQ in the first block (ΔR² = .05, F_{(1, 75)} = 3.97, p = .05), maternal resolution and insightfulness which were entered in the second block explained an additional marginally significant portion of the variance (ΔR² = .07, F_{(2, 73)} = 2.80, p = .06). Examination of the beta weights revealed that insightfulness was not a significant predictor (β = −.11, t = −1.05, p = .29) but resolution was significant (β = .26, t = 2.26, p = .02). The final regression model was significant (F_(3,73) = 3.25, p = .026).

The regression model predicting paternal sensitivity revealed that, after controlling for child IQ in the first block (ΔR² = .06, F_{(1, 77)} = 4.91, p = .03), paternal resolution and insightfulness which were entered in the second block explained an additional significant portion of the variance (ΔR² = .13, F_{(2, 75)} = 6.12, p = .003). Examination of the beta weights revealed that insightfulness was a significant predictor (β = .34, t = 3.18, p = .002) but resolution was not significant (β = .07, t = .67, p = .50). The final regression model was significant (F_(3,75) = 5.94, p = .001).

Finally, a hierarchical regression predicting positive parent-parent interaction from insightfulness and resolution of both parents while controlling for parental years of education was conducted. The regression revealed that, after controlling for paternal and maternal years of education in the first block (ΔR² = .10, F_{(2, 71)} = 4.04, p = .02), maternal and paternal resolution and insightfulness which were entered in the second block explained an additional significant portion of the variance (ΔR² = .21, F_{(4, 67)} = 5.38, p = .001). The final regression model was significant (F_(6,67) = 5.27, p < .001). Examination of the beta weights revealed that only maternal insightfulness was a significant predictor (β = .34, t = 3.05, p = .003). Maternal resolution (β = .12, t = 1.13, p = .26), paternal insightfulness (β = .14, t = 1.28, p = .20), and paternal resolution (β = .04, t =.36, p = .71) were not significant.

Discussion

The study sought to examine whether the findings linking insightfulness and resolution of mothers of preschoolers with ASD to their sensitivity (Oppenheim et al., 2010; Oppenheim, Dolev, et al., 2009) also apply to fathers. With regard to insightfulness, we found the expected association with sensitivity for fathers, with those who were more insightful showing higher levels of sensitivity when interacting with their children compared to less insightful fathers. The expected association between insightfulness and sensitivity was not found for mothers, however. With regard to resolution, we found the expected association for mothers, with those who were more resolved showing higher levels of sensitivity than those less resolved, but no such association was found for fathers. Finally, the four measures of maternal and paternal insightfulness and resolution were each associated with parent-parent positive interaction, supporting the idea that the mentalization capacity involved in these measures is expressed not only in the interaction with the child but also beyond, in parent-parent interaction. Finally, when all four measures were considered together, maternal insightfulness emerged as the most salient predictor of parent-parent interaction.

The findings regarding the insightfulness-sensitivity link among fathers are important for several reasons. First, this is the first study of paternal insightfulness (or mentalization more generally) with regard to children with ASD: Fathers rated relatively high on insightfulness, namely capable of seeing things from their children’s point of view, appeared to translate their insightfulness to sensitive responsiveness when interacting with their children. It is also of note that paternal insightfulness, like maternal insightfulness, was unrelated to children’s IQ or the severity of their symptoms. Thus, as is the case with mothers, it does not seem to be that fathers’ insightfulness is necessarily diminished when children have lower cognitive capacities or have more severe symptoms. Rather, insightfulness appears to reflect fathers’ (as well as mothers’) efforts to make meaning and empathically understand the motives underlying their child’s behavior (Oppenheim et al., 2001).

In contrast to the findings regarding fathers, no association was found between mothers’ insightfulness and their sensitivity. This was puzzling: As mentioned above, in a previous study using the same measures of insightfulness and sensitivity, the insightfulness-sensitivity link was not only significant but had a large effect size (partial eta squared = .24; Oppenheim et al., 2010). Thus, there is equivocal evidence for the insightfulness-sensitivity link in the case of mothers of children with ASD, with one study supporting the association and the current study failing to support it. Also, there are no studies using other assessments of maternal mentalization (e.g. Reflective Functioning, Mind Mindedness) with regard to children with ASD to help understand the contrasting findings. More research is therefore clearly needed.

While no links between maternal insightfulness and sensitivity were found, interesting gender differences were found, with mothers showing higher levels of insightfulness than fathers. It is possible that mothers were more insightful than fathers because they were more engaged than fathers in everyday after-school caregiving: For example, 80% of the children spent after-preschool hours with their mothers and only 8.3% spending that time with their fathers. This increased involvement may have provided mothers with more opportunities to gain the deep and contextualized understanding of the child needed for insightfulness. It is also possible that the gender differences reflect general differences between mothers and fathers not unique to parenting children with ASD. In a study comparing maternal and paternal insightfulness with respect to TD children (Marcu et al., 2016), mothers were more likely to be insightful than fathers, although the difference was not significant. A few additional studies of TD children using other measures of parental mentalization found that mothers showed higher levels of RF than fathers (e.g. Cooke et al., 2017; Esbjørn et al., 2013; Ruiz et al., 2020), although the overall picture regarding gender differences in RF is not simple: Vismara et al. (2021) found no such differences and Decarli et al. (2023) found differences favoring fathers.

Interestingly, while mothers were more insightful than fathers no differences in sensitivity between mothers and fathers were found. This was consistent with the findings of Bentenuto et al. (2020) who also found no differences between maternal and paternal sensitivity in a sample with children with ASD. Interestingly, these findings contrast with studies comparing maternal and paternal sensitivity with TD children that found that mothers were more sensitive than mothers (Hallers-Haalboom et al., 2017; Lovas, 2005). Perhaps the special needs of children with ASD (and possibly other special needs: See De Falco et al., 2009 regarding fathers of children with Down Syndrome) draw fathers in and increase their levels of sensitivity to be comparable to those of mothers. This is consistent with Biringen et al. (2022) conclusion that in the context of neuro-divergent children fathers “step up to the plate.” Finally, the lack of a difference between maternal and paternal sensitivity, combined with the difference in insightfulness favoring mothers, suggests that the two assessments, although theoretically thought to be linked (Oppenheim & Koren-Karie, 2013), reflect somewhat distinct windows into parenting. This may be the case for parents in general, or for those with children with special needs such as ASD in particular.

We turn now to the findings regarding resolution. As mentioned above, paternal resolution was not associated with sensitivity but maternal resolution was associated with sensitivity. The finding regarding maternal resolution replicates the Oppenheim, Dolev, et al. (2009) results regarding children with ASD, and the studies mentioned earlier regarding children with other diagnoses (Feniger-Schaal & Oppenheim, 2012; Sher-Censor & Shahar-Lahav, 2022). Importantly, similarly to the findings regarding insightfulness, resolution was unrelated to children’s IQ or the severity of their symptoms, consistent with the notion that resolution reflects the parent’s adaptation (Barnett et al., 2006) and “coming to terms” with the diagnosis, to a degree irrespective of how severely the child is impacted by the disorder. The lack of association between resolution and sensitivity among fathers was contrary to expectations. Possibly, fathers’ resolution is expressed in aspects other than sensitivity. For example, studies of father-child interactions highlighted activation (Paquette, 2004) and challenging support (Grossmann et al., 2008) as central dimensions of father-child interaction, perhaps more than sensitivity. These and additional dimensions of fathers’ interactions with their children should be considered in future studies of paternal resolution.

The study’s final goal was to examine whether parental insightfulness and resolution are expressed only in the context of parent-child interaction (the “narrow” view), or whether insightfulness and resolution are an expression of a more general capacity for mentalization that would also be expressed in parent-parent interaction (the “broad” view). The findings support the latter. All four measures of mothers’ and fathers’ insightfulness and resolution were associated with parent-parent interaction, so that higher levels of insightfulness and resolution were associated with more positive interaction. Perhaps insightful parents are not only insightful with respect to their child but also insightful with respect to their parenting partners and capable of seeing things from the partner’s point of view. Furthermore, perhaps developing a perspective regarding one’s emotional responses, their change over time, and their relations with thoughts and behavior inherent to resolution was also applied by the parents when communicating with their partners about their child. These characteristics of insightfulness and resolution may have supported the positive emotional connection, willingness to hear and understand the partner’s perspective, and dealing with parenting disagreements respectfully and cooperatively that were part of the positive parent-parent interaction aggregate score.

Finally, it is noteworthy that when all four parental scores (i.e. insightfulness and resolution of mothers and fathers) were considered together, the strongest predictor of co-parental communication was mothers’ insightfulness. Particularly due to the puzzling lack of association between maternal insightfulness and sensitivity discussed above, this finding strengthens the confidence in the maternal insightfulness measure. It also suggests that, at least when discussing the child, mothers have a particularly important role in shaping parent-parent interaction. Two caveats are important to mention regarding these findings. First, it is unknown whether parents’ communication about topics other than their child would also be related to their insightfulness and resolution. Second, parent-parent interaction was coded using dyadic scales, and scores were not given to each of the partners separately. Therefore, unlike the assessments of parental sensitivity which were separate for each parent, we do not know how each parent behaved when interacting with one another, only the resulting interaction process.

Additional general limitations should also be noted. First, the study involved only boys, and the extent to which the findings apply to girls as well is unknown. Second, all of the measures were concurrent and therefore the direction of effects between insightfulness/resolution and sensitivity/parent-parent interaction could not be determined. Third, the study examined the mother- and father-child dyads separately, even though more complex effects, such as that of one parent on the interaction of the other parent with the child may also be in play. These were beyond the scope of the present study. Finally, the study focused on only one aspect of parental behavior, sensitivity, and other aspects of parental behavior might be equally informative. Specifically, the sensitivity scale used in this study is part of the Emotional Availability Scales (EAS, Biringen et al., 2014). Although it is the EA scale most closely aligned with attachment and insightfulness (Oppenheim & Koren-Karie, 2013) and was therefore used in this study, the EAS also includes additional parental scales. In particular, studies have shown that the parental structuring EA scale, which combines sensitivity with appropriate means of control, predicts parent/child outcomes better than sensitivity (Biringen et al., 2022; Clark et al., 2021), and therefore studies of parental insightfulness and resolution which will include this scale are needed. Furthermore, the EAS also include the Responsiveness and Involving child scales, which are particularly relevant for children with ASD because they touch on core difficulties associated with this disorder (responding to overtures of others and initiating interactions with others, Adamson et al., 2019; Mundy, 2018) and future studies should assess these dimensions as well.

Three strengths of the study should also be mentioned. First, the study measures were based on coding transcripts (insightfulness, resolution, and parent-parent interaction) as well as observations (sensitivity), and none involved parental-reports. Second, the study involved fathers who are often neglected in studies of parent-child interactions, particularly with children with special needs (Rankin et al., 2019). Finally, children’s IQ and symptom severity scores were not based on parental reports but rather assessments by an examiner, so that these scores and the study measures were independent.

Beyond their importance for research on parent-child interaction in autism the findings are important for clinical work with parents of children with ASD. In particular, the new findings regarding paternal insightfulness highlight the importance of encouraging the inclusion of fathers in intervention work with families with young children with ASD, and realizing that fathers are an under-used resource for such interventions (Rankin et al., 2019).

Abbreviations

ASD: Autism Spectrum Disorder; TD: Typically Developing; IA: Insightfulness Assessment; RDI: Reaction to Diagnosis Interview.

Acknowledgments

We thank the participating families for their time and for sharing their experience with us as well as the research assistants involved in the study.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by an Israel Science Foundation grant 155/15 and by the Harry and Sadie Lasky Foundation.