Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2025 Feb 27;46(5):564–576. doi: 10.1002/imhj.70006

Validation of the Parental Reflective Functioning Questionnaire in Finnish mothers and fathers of toddlers

Marjo Flykt 1,✉, Hannu Haarala 2, Kirsi Pyhältö 2,3, Jaana Minkkinen 4, Mirjam Kalland 2
PMCID: PMC12420897  PMID: 40017061

Abstract

Parental reflective functioning (PRF) is important in guiding parenting behavior. It is traditionally measured with interviews that are relatively time‐consuming. A self‐report measure of PRF, the Parental Reflective Functioning Questionnaire (PRFQ), has been developed but has not yet been extensively validated across different sociocultural contexts. The aim of this study was to examine the factorial structure of the PRFQ in a community‐based first‐time parent sample in Finland. We further examined the associations between PRFQ scales and parental demographic features, depression, emotional availability (EA) in parent‐child relationships, and representations of the child. The sample comprised 355 mothers and 108 fathers and was part of a larger longitudinal study. Data were collected by an online questionnaire at the child age of 2 years. PRF was measured with PRFQ, parental depression with Center for Epidemiological Studies Depression, parent‐child interaction quality with Emotional Availability Self‐Report, and parental representations of the child with a modified adjective list developed in conjunction with the Interview of Maternal/Paternal Representations. Our results showed that the factor structure of PRFQ was similar to the original sample, and parental demographics, depression, and parenting variables were associated with the PRF scales in an expected manner, supporting the validity of PRFQ across different cultures.

Keywords: depression; parental reflective functioning; parenting; questionnaire; validation; الأداء التأملي للوالدين، الاستبيان، التحقق من الصحة، الاكتئاب، التربية الوالدية; 父母反思功能, 问卷, 验证, 抑郁, 育儿; Fonction de réflexion parentale; questionnaire; validation; dépression; parentage; elterliche Reflexivfunktion; Fragebogen; Validierung; Depression; Elternschaft; 親の内省機能、質問紙、検証、うつ病、子育て; funcionamiento con reflexión del progenitor; cuestionario; validación; depresión; crianza

1. INTRODUCTION

1.1. Parental reflective functioning (PRF) and its measurement

PRF, or mentalizing, denotes the parental capacity to bear in mind both their own and the child's mental states, such as thoughts, feelings, and intentions, and to link them to behavior (Slade et al., 2005). Assessed with interview measures, such as Parent Development Interview (PDI, Slade et al., 2005), PRF is predicted by adult attachment and predicts both parenting quality and infant attachment and socioemotional development (for review, see: Slade & Sleed, 2024; Zeegers et al., 2017). Lengthy clinical interviews that must be transcribed and coded by trained coders are not always practical for large‐scale studies. Luyten and colleagues (2017) developed the Parental Reflective Functioning Questionnaire (PRFQ) to assess PRF via parent self‐report. Table 1 presents the three PRFQ subscales: Interest and curiosity in mental states (IC), Certainty of mental states (CMS), and Prementalizing modes (PM, Luyten et al., 2017). The first two dimensions (IC and CMS) describe high parental RF, delineating acknowledgment of the child's individuality and that the child has a mind of his/her own, while PM describes severely impaired reflective functioning, a failed attempt to understand or construct the relation between the mind and the outer world (Fonagy, 2008; Fonagy & Luyten, 2018; Duschinsky & Foster, 2021).

TABLE 1.

The subscales of the Parental Reflective Functioning Questionnaire (Luyten et al., 2017) and their descriptions.

Scale Description Optimal scoring
Interest and curiosity in mental states (IC) Parent's active curiosity about the child and willingness to understand the child's mental states Higher levels are more optimal
Certainty of mental states (CMS) Parent's ability to understand that although they cannot know for certain what is in their child's mind, they make efforts trying to understand the child Theoretically, the mid‐levels of certainty are expected to represent the highest PRF, but in practice, higher certainty has been found to be associated with optimal outcomes, and empirically, the variable is generally used in a linear manner
Prementalizing modes (PM) maladaptive, severely impaired reflective functioning, where realistic representations of the mind are absent, resulting in failed or more primitive ways of understanding self and others Lower levels are more optimal
Open in a new tab

Key Findings

  • The Parental Reflective Functioning Questionnaire (PRFQ) was validated in a Finnish sample.

  • All scales, especially parental prementalizing, were associated with depression, parental emotional availability, and representations of the child in an expected manner in both mothers and fathers.

  • The certainty scale also showed some curvilinear associations with child‐related parenting variables.

Statement of the relevance of the work for infant and early childhood mental health

In infant mental health practice, it is important to enhance parental reflective functioning, which is a vital aspect affecting the parent‐infant interaction quality. Culturally valid measures are needed for both clinical work and research. It is also important to validate measures among both parents since fathers need to be included more in infant mental health work.

Previous studies in Western and non‐Western countries have found the factor structure consistent with this three‐factor structure (e.g., Luyten et al., 2017; Moreira & Fonseca, 2023; Wendelboe et al., 2021; Ye et al., 2022). However, a recent Korean study reported a slightly different four‐factor structure (Lee et al., 2021), indicating that sociocultural adaptations for the PRFQ might be needed. To further validate the measure, it is important to examine its factor structure in different countries. In this study, we conducted a factor analysis of PRFQ in a Finnish community‐based sample. A previous study conducted in Finland examined the factor structure of another modification of the scale, PRFQ‐fi, developed for parents of infants (Pajulo et al., 2018). The 14‐item measure was based on most of the same original items (36) as in the current measure, and a four‐factor solution was found. However, later Finnish studies (conducted from the earlier waves of the current sample) found inadequate scale reliabilities for the PRFQ‐fi for parents of infants (Salo et al., 2021; 2022). As the PRFQ version developed by Luyten and colleagues (2017) has become highly established in the field, it is important to also study its factor structure and validity in Finland. One previous Finnish study used the PRFQ in both infants and toddlers (Lindblom et al., 2022), but the factor structure was not examined, and one of the scales (prementalizing) had inadequate reliability. Further, previous Finnish studies have included both primi‐ and multiparous parents, with data collected from only one region. In this study, we examine the factor structure of the PRFQ in Finnish first‐time parents of toddlers in a nationwide cohort. We further validate the measure by examining demographic, mental health, and parenting correlates of the PRFQ scales.

1.2. Parental characteristics and PRFQ scales

Tentative evidence suggests that paternal RF may have unique importance for child development (Benbassat & Priel, 2015; Butitta et al., 2019), but evidence on gender differences in PRF is contradictory. Some studies indicate that mothers show more optimal PRF than fathers (Benbassat & Priel, 2015; Cooke et al., 2017; Esbjørn et al., 2013), while others have failed to find differences (Katznelson, 2014). It seems that mothers may be more interested in their child's mental states than fathers (Pazzagli et al., 2018; Ruiz et al., 2020). Parents may also tend differently to girls and boys: for instance, Raley and Bianchi (2006) suggest that fathers invest more time in their sons. Studies are also scarce and contradictory on the role of child gender in parental PRF. Some studies indicate that parents of girls may show lower PRF; one study showed higher prementalizing among parents of daughters (Pazzagli et al., 2018), and another showed mothers of daughters to have lower certainty (Cooke et al., 2017). Kamza and colleagues (2024) found, however, no association between PRF and child gender.

Some studies have indicated younger parental age as a risk factor for higher prementalizing (Luyten et al., 2017) or lower certainty (Kamza et al., 2024), although others have failed to find such associations (Cooke et al., 2017). Lower education level may also be a risk factor for PRF and has been linked to higher prementalizing (Luyten et al., 2017) and lower certainty and interest and curiosity, especially among fathers (Cooke et al., 2017; Kamza et al., 2024). We were unable to locate previous studies on the role of parental marital status in PRF but examine this issue here.

Earlier studies have consistently documented that PRFQ scales are associated with parental mental health. Especially prementalizing has been linked to parental depression and other mental health symptoms (Kamza et al., 2024; Luyten et al., 2017; Moreira & Fonseca, 2023). However, the link between prementalizing and depression is not well‐established in fathers (Wendelboe et al., 2022), and even in mothers, the findings have varied in scales other than prementalizing (e.g., Luyten et al., 2017; Moreira & Fonseca, 2023).

Several findings also support that all PRFQ scales are associated with parenting features such as parental warmth, self‐efficacy, and more optimal parenting styles (Cooke et al., 2017; Moreira & Fonseca, 2023; Ye et al., 2022). Only one previous study has examined associations among PRFQ scales and maternal emotional availability (EA), finding that prementalizing was associated with lower parental, child, and dyadic EA (Luyten et al., 2017), while interest and curiosity was associated with higher child EA, and certainty with higher dyadic EA and lower hostility (Luyten et al., 2017). To our knowledge, no previous studies exist on the associations between PRF and EA in fathers. Although interview‐based PRF has been linked to less balanced representations of the child (Alismail et al., 2022; Alvarez‐Monjaras et al., 2019; Schechter et al., 2005), studies are currently lacking on the associations between PRFQ scales and parental representations or perceptions of their child.

1.3. Study aims

The purpose of this study was, first, to examine the factor structure of the Finnish version of the PRFQ in a nationally representative sample of first‐time parents. We hypothesized that our findings would result in a three‐factor structure consistent with the original PRFQ studies (Luyten et al., 2017). Second, we investigated whether PRF dimensions were associated with parental demographic features, parenting, and mental health. We hypothesized that mothers would show more optimal PRF (i.e., higher interest and curiosity (IC), and certainty of mental states (CMS), and lower prementalizing (PM)) than fathers, and older and more educated parents, as well as parents of boys, would show more optimal PRF. Due to a lack of previous research, we made no hypothesis regarding parental marital status. Finally, we examined the associations between PRF and parental depression, EA, and parental representations of the child. We hypothesized that more optimal PRF would be associated with lower depression, higher EA, and more positive representations of the child in both parents in a linear manner. For theoretical reasons, we also examine the possible curvilinear effects of the Certainty scale on the outcome variables.

2. METHODS

2.1. Participants and procedure

The study originally comprised a community‐based sample of pregnant first‐time parents from 80 communities across Finland who were invited to participate in the “Families First” study within the national system of maternity and well‐baby clinics between 2011 and 2015 (Kalland et al., 2016). The inclusion criteria comprised sufficient skills in Finnish or Swedish (the two official languages in Finland) as the questionnaires were only available in those languages. Originally, 657 mothers and 347 fathers participated, and data were collected at several time points over the child's first 2 years. Of the initial participants, 1.5% reported being Swedish‐speaking, and none reported native languages other than Finnish or Swedish.

The data in the current study comprise all families (n = 388) that participated in the follow‐up at the child age of 2 years. In 280 families, only the mother participated, in 31 families, only the father participated, and in 77 families, both parents participated. Of the children, 185 (47.7%) were boys and 201 (51.8%) were girls (information missing for two children). Due to drop‐out, the current data only included participants who reported being native Finnish speakers. The mothers were 18–46 years old (M = 29.67, SD = 4.46), and the fathers were 21–49 years old (M = 31.33, SD = 4.59). The parents were most often living together: 41.3% were cohabiting, and 57% were married. Only six participants were single parents. Regarding education level, 75.5% of mothers and 67.4% of fathers in our sample reported bachelor or master level education, whereas others had a lower education level (mostly high school or vocational school). The sample was more highly educated than the Finnish national mean, as on average, about a third of Finnish adults have a master's or bachelor's degree (Statistics Finland, 2018). Most of the children (∼96%) lived with both parents. Of mothers, 79.7% evaluated their child's health as very good and 19.7% as good, with only two mothers evaluating it as decent. All fathers evaluated their children's health as either very good (83.4%) or good (15.7%). Drop‐out analyses showed that both mothers and fathers still participating at 2 years were more often married (mothers: χ2(1) = 5.50, p = .019; fathers: χ2(1) = 10.15, p = .001) and had a higher education level (mothers: χ2(1) = 10.00, p = .002; fathers: χ2(1) = 8.70, p = .003) than those who dropped out. The ethics committee of the Finnish National Institute for Health and Well‐being approved the study plan, and the study was conducted in compliance with national and international ethical standards and accordance with the Declaration of Helsinki.

Data collection was carried out using electronic questionnaires with personal codes. Paper versions of the questionnaires were also available upon request. The public health nurses in the maternity clinics informed potential participants about the study at a regular follow‐up visit during the third trimester of pregnancy. The nurses were specifically instructed to provide equal opportunity for all first‐time parents attending the regular check‐ups to participate in the study. Either the mother, the father, or both received written information regarding the study as well as an informed consent form to be returned in a prepaid envelope. Mothers and fathers separately filled in the same questionnaires. All participants gave voluntary informed consent. Those who gave permission to be contacted were invited to the 2‐year follow‐up. The questionnaires used in this study were all obtained at the child age of 2 years.

2.2. Measures

2.2.1. Demographic factors

Demographic factors included parental and child gender, parental education level (classified as 1 = up to secondary education (high school or vocational school) and 2 = tertiary education (bachelor's or master's degree), parental age, and parental marital status (2 = cohabiting, 3 = married; due to only six single parents in the sample, we excluded single parents from the analysis).

PRF was measured with the 18‐item Parental Reflective Functioning Questionnaire (PRFQ; Luyten et al., 2017) translated into Finnish. The items are divided into three subscales of PRF on a 7‐point Likert scale (from 1 = strongly disagree to 7 = strongly agree). Based on theory and previous studies (Luyten et al., 2017), three subscales were expected in this study: Interest and curiosity in mental states (IC, e.g., “I wonder a lot about what my child is thinking and feeling”), Certainty about mental states (CMS, e.g., “I always know what my child wants”), and Prementalizing (PM, e.g., “My child sometimes gets sick to keep me from doing what I want to do”). Higher scores on the IC reflect higher PRF, higher scores on PM reflect lower PRF, and depending on the study, midpoint or high scores of CMS reflect higher PRF. The scale has shown good reliability and validity in several previous studies (e.g., Luyten et al., 2017; Kamza et al., 2024). In this study, the scale reliabilities (McDonald's omega) were adequate to good: .78 for IC, .81 for CMS, and .60 for PM for mothers and .81 for IC, .82 for CMS, and .73 for PM for fathers. Omega coefficients were used to indicate scale reliability, as they are recommended over Cronbach's alpha (Hayes & Coutts, 2020).

Parental depression was measured with the Center for Epidemiological Studies Depression Scale (CES‐D; Radloff, 1977). CES‐D is a 20‐item scale measuring depressive symptoms. The parents used a four‐point Likert scale (0 = rarely or never, 3 = almost all of the time) to indicate how they felt during the past week. The measure is widely used and has been previously validated in Finland (Konttinen et al., 2010). The scale reliabilities (McDonald's omega) were .90 for both mothers and fathers.

EA in the parent‐child relationship was measured with an Emotional Availability Self‐Report (EA‐SR; Vliegen et al., 2009). EA‐SR is a 36‐item scale with five subscales: parental intrusiveness and hostility (6 items each), mutual attunement (10 items), affective quality (5 items), and child involvement (9 items). Hostility and intrusiveness are typically combined into an adult domain, and mutual attunement and affective quality into a dyadic domain, whereas child involvement represents a child domain. Parents answered on a five‐point Likert scale (0 = disagree completely, 4 = agree completely). Total summary scores were built for each subscale and domain. The intrusiveness subscale showed low reliability, which did not increase when combined with hostility in the adult domain, so we decided to use only the hostility scale, child involvement (child domain), and dyadic domain (combination of mutual attunement and affect quality). The reliabilities (omega) for mothers were .71 for the dyadic scale, .77 for the child scale, and .84 for hostility, and for fathers, .96 for the dyadic scale, .86 for the child scale, and .85 for hostility. Higher hostility refers to more negative EA, while higher dyadic and child EA refer to more positive EA.

Parental representations of the child were measured using a modified version of a child‐related adjective scale based on a questionnaire form developed in conjunction with the Interview of Maternal/Paternal Representations (IRMAG/IRPAG) (Ammaniti et al., 1992, 1998; Pajulo et al., 2006). The measure, based on an adaptation of Osgood's semantic differentials (Osgood and Suci, 1952), originally contained 17 mixed pairs of opposite adjectives placed at opposite ends of a horizontal line, with the subjects asked to mark the point from 1 to 7 that best indicates their rating. In this study, only the child‐related scale of the measure was used, and we utilized a modified version of nine adjective pairs (easily irritable—calm, withdrawing—curious, timid—bold, hard to take care of—easy to take care of, fearful—trusting, feeble—strong, inflexible—adaptable, difficult to interpret—easy to interpret, and blank—strongly expressive). Higher values represent more positive representations (e.g., 1 = fearful and 7 = trusting). The adjectives were chosen to be suitable also for young infants, as the scale was used in earlier phases of this longitudinal study. Previously, Pajulo and colleagues (2001, 2004) found, using the original version, that representations of a child are more negative among mothers with psychosocial risk and negative representational change associated with more observed mother‐infant interaction problems and maternal postpartum depressive symptoms. In this study, we used a total mean score. Flykt and colleagues (2012) have previously shown the summary scale of the original version to show good reliability. Based on omega reliability coefficients, we decided to drop two item pairs (easily irritable—calm and hard to take care of—easy to take care of) to achieve satisfactory reliability. The scale reliabilities (omega) were .75 for mothers and .72 for fathers. The modified version has not previously been validated, but in this study, maternal representations correlated significantly with maternal depression, maternal dyadic EA, and child EA, and paternal representations correlated with paternal depression, paternal dyadic EA, and child EA (see Table 5).

TABLE 5.

Correlations between PRFQ scales and study variables.

IC a CMS PM a EA dyadic EA child a EA hostility Representations Depression a
IC a – .14* −.41*** .20*** .39*** −.09 .17** −.05
CMS .19 – −.10 .38*** .09 −.26*** .15** −.16**
PM a −.42*** −.09 – −.47*** −.33*** .25*** −.19*** .27***
EA dyadic .37*** .44*** −.53*** – .26*** −.63*** .19*** −.40***
EA child a .41*** .21* −.55*** .47*** – −.08 .27*** −.11*
EA hostility −.32*** −.30** .42*** −.61*** −.23* – −.10 .21***
Representations .09 .25** −.25** .36*** .38*** −.15 – −.15**
Depression a −.11 −.26** .23* −.46*** −.25** .31*** −.38*** –
Open in a new tab

Note: Figures above the diagonal represent mothers and figures below the diagonal represent fathers.

Abbreviations: CMS, certainty of mental states; IC, interest and curiosity; PM, prementalizing; PRFQ, Parental Reflective Functioning Questionnaire.

*p < .05, **p < .01, ***p < .001.

a

Scale not normally distributed, Spearman's Rho is used to indicate correlation.

2.3. Statistical analysis

Data were analyzed using IBM SPSS Statistics and Mplus. Of the PRF variables, only Certainty about mental states (CMS) was normally distributed. For Interest and curiosity (IC) and Prementalizing (PM), most participants reported high PRF (i.e., high IC and low PM). To answer our first questions about the factor structure of PRFQ, we first performed an Exploratory Factor Analysis (EFA) with 1–4 factors. Due to recommendations of using polychoric correlations as a basis for factor analysis in variables with Likert ordinal scale (e.g., Kiwanuka et al., 2022), we used Weighted Least Squares with Mean and Variance adjustment (WLSMV) as an estimation method and defined variables as categorical. We also ran the same analyses using continuous variables and Maximum Likelihood with Robust Standard Errors (MLR) as an estimation method, which gave similar results, but model fits were slightly better for WLSMV, and thus, we only report the results for WLSMV. Due to the small number of fathers in the sample (n = 108), both mothers and fathers were included in the same analysis to increase the number of participants. We also conducted a sensitivity analysis to see whether the results were similar among mothers only. Model fit was evaluated based on Root mean square error of approximation (RMSEA) < .06, comparative fit index (CFI) > .95, Tucker–Lewis index (TLI) > .95, and Standardized Root Mean Square Residual (SRMR) < .08 (Hu & Bentler, 1999). Models were compared based on fit indices, nested Chi‐square tests, and theoretical considerations of the factor structure.

To examine the associations between demographic factors and PRF scales, for dichotomous variables (parent gender, education level, marital status), we performed t‐tests with bootstrapped confidence intervals to account for the non‐normality of the PRF variables. To examine child gender, also parent gender was taken into account by creating four groups (mother‐son, mother‐daughter, father‐son, father‐daughter), which were compared with the Kruskal–Wallis test for IC and PM and with ANOVA for CMS. Bonferroni corrections were applied for pairwise comparisons. To examine the associations between PRF scales and parental age, parental depression, and parenting variables (EA and representations), Spearman's correlations were used for IC and PM and Pearson's correlations for CMS. To examine the quadratic associations of CMS with parental depression and parenting, we conducted a linear regression, including both the linear and quadratic terms of CMS as a predictor.

3. RESULTS

3.1. Exploratory factor analysis of PRFQ in the whole sample

First, we examined whether the original three‐factor structure of the PRFQ presented an adequate fit in the Finnish sample. For this purpose, we performed an EFA comparing factor solutions ranging from 1 to 4. One‐ and two‐factor solutions failed to show adequate model fit (1‐factor model: RMSEA = .18, CFI = .65, TLI = .61, SRMR = .20; 2‐factor model: RMSEA = .08, CFI = .94, TLI = .92, SRMR = .08). Both three‐ and four‐factor solutions showed adequate model fit (3 factor model: RMSEA = .04, CFI = .99, TLI = .98, SRMR = .03; 4‐factor model: RMSEA .03, CFI = .99, TLI = .99, SRMR = .03). Nested model comparisons showed that the two‐factor model was superior to the one‐factor model (χ2(17) = 844.86, p < .001), and the three‐factor solution was superior to the two‐factor solution (χ2(16) = 210.61, p < .001). The four‐factor solution was also superior to the three‐factor solution in nested model comparison (χ2(15) = 35.81, p = .002). As both three‐ and four‐factor solutions had adequate fit, we examined their factor loadings in more detail to determine the theoretical feasibility of the model. Tables 2 and 3 present the results for the three‐ and four‐factor solutions. For the three‐factor solution, all items loaded on the expected factors with loadings >.40. For the four‐factor solution, the factor loadings for IC and CMS remained highly similar. However, PM was divided into two factors, with all original items loading on one factor, but items 1 (“The only time I'm certain my child loves me is when s/he is smiling at me”), 13 (“When my child is fussy, s/he does that just to annoy me”), and 16 (“Often, my child's behavior is too confusing to bother figuring out”) also loading on another factor. Item 16 had a loading <.40 for both factors. Since both models had very good model fit, we chose the three‐factor model for additional analyses based on theoretical considerations.

TABLE 2.

Results of the three‐factor solution for the Parental Reflective Functioning Questionnaire.

Item Factor
IC CMS PM
3. I like to think about the reasons behind the way my child behaves and feels. .83 −.06 .02
6. I wonder a lot about what my child is thinking and feeling. .69 .003 −.15
12. I try to see situations through the eyes of my child. .66 .03 −.19
15. I try to understand the reasons why my child misbehaves. .61 −.08 −.04
9. I am often curious to find out how my child feels. .57 −.009 −.33
18. I believe there is no point in trying to guess what my child feels a .46 04 −.35
2. I always know what my child wants. −.12 .84 .03
5. I can completely read my child's mind. .01 .78 .20
17. I always know why my child acts the way s/he does. .09 .69 −.02
8. I can always predict what my child will do. −.005 .68 .24
14. I always know why I do what I do to my child. .008 .63 −.11
11. I can sometimes misunderstand the reactions of my child a −.32 .56 −.08
4. My child cries around strangers to embarrass me. .01 .20 .95
13. When my child is fussy s/he does that just to annoy me. .16 −.007 .78
10. My child sometimes gets sick to keep me from doing what I want to do. −.03 .14 .73
1.The only time I'm certain my child loves me is when he or she is smiling at me. −.13 .01 .65
16. Often, my child's behavior is too confusing to bother figuring out. −.02 −.11 .55
7. I find it hard to actively participate in make‐believe play with my child. −.06 −.05 .47
Open in a new tab

Note: Bolded values represent the items that load to the factor in question according to our analysis. N = 463.

a

Item has been reversed.

TABLE 3.

Results of the four‐factor solution for the PRFQ.

Factor
Item IC CMS PM I PM II
3. .82 −.03 .04 −.01
6. .71 .05 −.15 .03
12. .66 .002 −.07 −.22
15. .60 −.06 −.001 −.04
9. .58 .03 −.28 −.08
18. a .48 .11 −.35 .02
2. −.11 .83 −.05 −.001
5. .02 .82 .007 .22
8. −.008 .68 .11 .11
17. .10 .62 .15 −.33
14. .01 .58 −.04 −.21
11. a −.31 .53 −.09 −.07
4. −.03 .06 .97 −.06
10. −.07 .02 .71 −.03
13. .16 .00 .65 .32
1. −.14 .02 .51 .31
7. −.08 −.08 .44 .09
16. −.02 −.06 .38 .39
Open in a new tab

Note: Bolded values represent the items that load to the factor in question according to our analysis. N = 463.

Abbreviations: CMS, certainty of mental states; IC, interest and curiosity; PM, prementalizing; PRFQ, Parental Reflective Functioning Questionnaire.

a

Item has been reversed.

3.2. Sensitivity analysis: Exploratory factor analysis of PRFQ among mothers

As a sensitivity analysis, we also performed the same EFA among mothers only. The results were largely similar to the whole sample. One‐ and two‐factor solutions failed to show adequate model fit (1‐factor model: RMSEA = .18, CFI = .64, TLI = .60, SRMR = .20; 2‐factor model: RMSEA = .08, CFI = .94, TLI = .93, SRMR = .07). Both three‐ and four‐factor solutions showed adequate model fit (3‐factor model: RMSEA = .05, CFI = .98, TLI = .98, SRMR = .04; 4‐factor model: RMSEA .04, CFI = .99, TLI = .98, SRMR = .04). Nested model comparisons showed that the two‐factor model was superior to the one‐factor model (χ2(17) = 653.38, p < .001), and the three‐factor solution was superior to the two‐factor solution (χ2(16) = 139.98, p < .001). The four‐factor solution was also superior to the three‐factor solution in nested model comparison (χ2(15) = 34.94, p = .003). As both three‐ and four‐factor solutions had adequate fit, we examined their factor loadings in more detail to determine the theoretical feasibility of the model. Tables S1 and S2 present the results for the three‐ and four‐factor solutions. For the three‐factor solution, all items loaded on the expected factors with loadings > .40. For the four‐factor solution, the factor loadings for IC, CMS, and PM remained highly similar to those of the three‐factor model. Some of the items from all other factors also loaded on the new fourth factor, but none of them showed high loadings (> .40). Since both models had very good model fit, we deemed the three‐factor model superior based on theoretical considerations.

3.3. Descriptives of PRFQ and other study variables

Table 4 displays the descriptive information of the PRFQ and other study variables for mothers and fathers. Regarding PRFQ, most participants reported relatively high levels of IC, mid‐levels of CMS, and low levels of PM. Further, most reported low levels of depression and relatively high levels of parenting variables (representations of child and EA). Table 5 shows Spearman's correlations between the PRFQ scales. IC showed a small positive correlation with CMS and a strong negative correlation with PM for both parents, while CMS and PM were not correlated.

TABLE 4.

Means, standard deviations and ranges, of the PRFQ and other study variables.

Mothers Fathers
M SD Range M SD Range
IC 6.09 .73 3.17–7.00 5.85 .77 3.33–7.00
CMS 3.84 1.06 1.17–6.17 3.83 1.03 1.33–6.67
PM 1.54 .53 1.00–4.33 1.60 .62 1.00–4.17
EA: Dyadic 22.15 2.89 11–29 22.63 2.84 14–28
EA: Child 34.06 2.54 21–36 32.79 3.55 22–36
EA: Hostility 11.02 5.47 0–24 8.01 4.96 0–18
Representations 5.90 .68 3.83–7 5.76 .67 3.83–7
Depression 5.58 4.81 0–26 3.94 3.97 0–18
Open in a new tab

Abbreviations: CMS, certainty of mental states; IC, interest and curiosity; PM, prementalizing; PRFQ, Parental Reflective Functioning Questionnaire.

3.4. Parental characteristics and PRF

Our second question concerned how parental demographics are associated with the PRFQ dimensions. Regarding parental gender, the results partially supported our hypothesis about mothers showing superior PRF but only in higher IC, t(461) = 2.74, p = .007. Regarding child gender, in the four‐group analyses (mother‐son, mother‐daughter, father‐son, father‐daughter), there was a difference in IC, H(3) = 9.93, p = .019. Pairwise comparisons (Bonferroni) indicated that IC was higher among mother‐daughter dyads (p = .033) and mother‐son dyads (p = .005) than among father‐son dyads, whereas father‐daughter dyads did not differ significantly from any other pairs. Results in Table 6 show that among mothers, both low education level and younger age were associated with her more negative PRF, specifically with higher PM. Married mothers showed more positive PRF, higher IC, than cohabiting mothers. None of the demographic variables are associated with paternal PRF.

TABLE 6.

Differences in PRFQ scales by parental background factors.

Education level Marital status Age
Low High Cohabiting Married
Mothers M SD M SD t(329) p M SD M SD t(325) p R
IC 6.05 .73 6.11 .73 −.11 .56 5.97 .78 6.19 .68 −2.70 .008 * .05 a
CMS 3.81 1.10 3.83 1.05 −.58 .92 3.89 1.02 3.75 1.08 1.16 .25 −.03 b
PM 1.71 .56 1.47 .48 3.39 .002 * 1.58 .53 1.50 .48 1.47 .14 −.15 ** , a
Fathers M SD M SD t(94) p M SD M SD t(96) p R
IC 5.72 .72 5.86 .79 −.72 .48 5.81 .77 5.83 .78 −.16 .88 −.18 a
CMS 3.84 1.01 3.79 1.04 .19 .86 3.66 .92 3.84 1.07 −.88 .36 −.03 b
PM 1.69 .75 1.57 .60 .65 .53 1.71 .67 1.54 .60 1.12 .27 .13 a
Open in a new tab

Note: Bolded values refer to values that are significant at the level of p < .05.

Abbreviations: CMS, certainty of mental states; IC, interest and curiosity; PM, prementalizing; PRFQ, Parental Reflective Functioning Questionnaire.

*p < .05, **p < .01.

a

Rs = Spearman's Rho.

b

Pearson's R.

Our third question was how PRF is associated with parental depression, EA variables, and parental representations of the child. As hypothesized, in both parents, PM was positively associated with depression and EA hostility and negatively associated with dyadic and child EA and representations of the child. CMS was negatively associated with depression and hostility and positively associated with dyadic EA and representations in both parents. In fathers, it was also positively associated with child EA. IC was positively associated with dyadic and child EA in both parents. In mothers, it was also positively associated with representations, and in fathers, it was negatively associated with hostility. The quadratic association of CMS was also significant with maternal representations of the child (B = .07, SE = .03, p = .009) and mother's report of child EA (B = .22, SE = .10, p = .022), showing that mothers with low and high levels of CMS tended to report higher representations of the child and higher child EA than those scoring in the middle levels.

4. DISCUSSION

The purpose of this study was, first, to validate the factor structure of PRFQ in a sample of Finnish first‐time parents. We further examined the associations of PRFQ scales with parental demographics, mental health, and parenting variables. As hypothesized, the factor structure of the measure aligned with the original one in our sample. The associations of the PRFQ scales with parental characteristics were largely similar to earlier studies, with higher depression and less optimal parenting variables associating with less optimal PRF and especially prementalizing. In mothers, also some demographic variables are associated with PRF. The certainty scale showed curvilinear associations with child‐related parenting variables but in a surprising manner.

4.1. PRFQ factor structure

Our first question concerned the factor structure of the PRFQ. PRF is divided into three factors, as in Luyten’s study (2017), supporting its validity in a Finnish sample. The results were similar, independent of whether the analysis was conducted in the whole sample or solely among mothers. Several previous studies in different countries also support the three‐factor structure (e.g., Moreira & Fonseca, 2023; Wendelboe et al., 2021; Ye et al., 2022). However, it is noteworthy that also the four‐factor structure showed a very good model fit in our data, although based on theoretical considerations and less clear item loadings, we deemed the three‐factor model superior. One earlier Korean study (Lee et al., 2021) reported a four‐factor structure to be more solid. In their study, similarly to ours, it was the prementalizing scale that divided into two different factors in the four‐factor solution, while the Interest and curiosity, and the Certainty scales showed a more robust factor structure. Some previous studies (e.g., De Roo et al., 2019; Wendelboe et al., 2021), including an earlier Finnish study (Lindblom et al., 2022), have suggested a diminished number of items in the prementalizing scale to result in a better model fit. Similarly to an earlier Finnish study, which did not directly examine the factor structure of PRFQ (Lindblom et al., 2022), we found good reliabilities for the two dimensions of interest and curiosity, and certainty of the mental state. Unlike the previous Finnish study, also the reliability of the prementalizing scale was adequate in our study, although slightly lower. Regarding intercorrelations between scales, our results were in accordance with earlier findings (e.g., Cooke et al., 2017), showing strong negative correlations between interest and curiosity, and prementalizing scales and weak positive correlations between interest and curiosity and certainty scales. In sum, the results in this study further support the PRFQ measure as consistent in different populations. Some issues warrant further investigations since, for example, the measure has been suggested to work better among nonclinical parents and those with older children (Wendelboe et al., 2021; Corlone et al., 2023; Madsen et al., 2023). On the other hand, very high levels of prementalizing or certainty are rare in a normative sample like ours, leading to low variance. Our results thus support previous literature stating that especially the prementalizing scale must be evaluated in more studies in different populations.

4.2. Parental characteristics and PRF

Our second research question was related to the associations between PRF and parental demographics. In line with our hypothesis and some previous studies (Benbassat & Priel, 2015; Cooke et al., 2017), mothers showed higher interest and curiosity about their child's mental states than fathers. Such differences have not been found in all studies (see Katznelson, 2014), and some have indicated gender differences in other dimensions, such as prementalizing (Cooke et al., 2017; Madsen et al., 2023) or certainty (Madsen et al., 2023), which were not replicated in our study. Higher maternal interest and curiosity may be explained by fathers spending less time with their toddlers (Arnott & Meins, 2007; Pazzagli et al., 2018). Also, fathers may be more interested in exploring the environment and offering stimulation to the child, while mothers might be more interested in the child's feelings and thoughts (Grossmann et al., 2008; Ruiz et al., 2020). The cultural context may also affect gender roles. In Finland, mothers are typically the main caregivers and have long maternity leaves (9 months to 3 years), which may make it especially easy to bond with children. Fathers rarely take longer paternity leaves, and the share of childcare may thus be unequal, especially during the child's first years.

Regarding child gender, our findings differed from some earlier studies that identified girls as more vulnerable (Cooke et al., 2017; Pazzagli et al., 2018). Instead, we found that it was especially father‐son dyads where the parent showed lower interest and curiosity. This is somewhat contradictory to earlier evidence that fathers invest more in their sons than in their daughters and participate more in their care (Manlove & Vernon‐Feagans, 2002; Raley & Bianchi, 2006). On the other hand, interest in the child's mental states may represent a somewhat different relational aspect than investment in the relationship, with perhaps opposite‐sex parenting promoting more curiosity in fathers. Fathers’ lower PRF scores with sons may also be due to societal gender norms, with emotions being more acceptable regarding daughters than sons.

Our results also replicated earlier findings (Luyten et al., 2017) of younger maternal age and lower education level being associated with higher prementalizing. However, unlike previous studies that have emphasized paternal risk (Cooke et al., 2017; Kamza et al., 2024), these associations were found solely among mothers. Finally, we also found that cohabiting mothers showed lower interest and curiosity than married mothers. This aligns with previous research that the parental relationship may be less stable and maternal and child well‐being lower in cohabiting families than in married families (Brown et al., 2004; Kalland et al., 2022; Osborne et al., 2007). Unfortunately, due to the small number of single parents in our study, we could not examine whether this group would have differed from parents who were together. The role of marital status in PRF is a novel finding that requires confirmation in further research.

Our third question concerned the associations between PRF, parental depression, and parenting. Depressive symptoms were linked to higher prementalizing and lower certainty but not to interest and curiosity among both parents. This is in line with earlier research, where the strongest links were found between maternal prementalizing and depression, and especially the link to interest and curiosity has been weak (Kamza et al., 2024; Luyten et al., 2017; Moreira & Fonseca, 2023). We also found similar associations among fathers, contrary to Wendelboe and colleagues (2022), who found no association between paternal PRF and paternal depression. The results seem to suggest that Interest and curiosity toward a child's mental states may be largely independent of parental depression.

The associations between maternal EA and PRF have been examined in only one previous study (Luyten et al., 2017), with findings largely similar to ours. Prementalizing was strongly associated with higher maternal hostility and lower dyadic and child EA. Higher interest and curiosity was associated with higher maternal dyadic and child EA but not with hostility. To our knowledge, this is the first study to examine the relationships between EA and PRF in fathers. We found that all PRF scales were linked to EA, with lower prementalizing associating with more hostility and lower dyadic and child EA, and certainty and interest and curiosity associating with less hostility and higher dyadic and child EA. We also found that in mothers, certainty was associated with lower hostility and higher dyadic quality. Interestingly, it was also associated with child EA but only in a curvilinear manner, with both high and low levels of certainty associated with higher evaluations of child involvement. This is opposite to what Luyten and colleagues (2017) suggest, that is, that mid‐levels of certainty would be optimal. As our study used a self‐report of child EA, it is possible that at least the mothers with very low certainty and potentially also those with overly high certainty viewed their child in an idealized manner.

To our knowledge, this was also the first study to examine parental self‐reported representations of the child and PRF. Although our representational measure should be validated more, the results paralleled findings obtained with interview measures of PRF (for review, see Slade & Sleed, 2024), showing that prementalizing is associated with more negative representations among both mothers and fathers. Further, certainty was associated with more positive representations of the child in both parents and for mothers, this was also true for interest and curiosity. Interestingly, there was a curvilinear association between maternal certainty and representations; mothers with low and high certainty reported more positive representations of the child, similar to the results obtained regarding the child involvement scale. It is interesting that this possible idealization process occurs only in the realm of maternal perceptions of their child and no other parenting variables. More research is clearly needed, especially regarding the curvilinear nature of the certainty scale.

4.3. Strengths and limitations of the study

This was a nationally representative study with a large sample size, especially for mothers. However, due to an inadequate sample size for fathers, we were unable to perform a factor analysis among fathers only. In the current sample, the education level was also higher than the national average, as there were more dropouts in those with lower education levels. Similarly, married couples were over‐represented, and there was more drop‐out in those not married. Overall, single parents were nearly completely lacking from the sample. It is likely that the results cannot be fully generalized to unmarried couples or those with lower education levels. The results can be generalized only to native Finnish speakers, as all Swedish speakers had dropped out at this phase of the study, and none of the participants reported a native language other than Finnish or Swedish. In Finland, it is not legal to ask about participant ethnicity, and thus, we only have knowledge about participants’ native language. Also, some of the parents were in a relationship with each other, which was not taken into account in our analyses.

The study used only self‐report measures, which can be susceptible to biases and might not be able to discern unconscious thoughts and emotions the same way as structured interviews such as PDI. The results may thus not be directly comparable to those obtained with interview measures. For example, the associations between PDI interviews and PRFQ are not strong (Arkle et al., 2023; Corlone et al., 2023).

5. CONCLUSION AND IMPLICATIONS FOR FUTURE RESEARCH

The results suggest that the PRFQ factor structure is also valid in a Finnish sample, supporting its use across multiple cultures. As a further validation, PRFQ scales were associated with parental demographics, parenting, and parental depression mostly in an expected manner. Few studies exist on fathers, and our results showed that PRFQ scales revealed mostly similar associations among fathers and mothers. This supports the validity of the instrument also in fathers, although more studies are needed. While the factor structure based on a combined sample of mothers and fathers was highly similar to that based on mothers alone, more studies are required on the factor structure, specifically in fathers. It is also important to note that the current study concerned parents of toddlers. Infancy measures of PRFQ may still require additional studies; for example, the version used in this study (Luyten et al., 2017) and its recently developed infancy version PRFQ‐I (Wendelboe et al., 2021, 2022) should be compared with another Finnish infancy version of the scale, PRFQ‐fi (Pajulo et al., 2018). Future studies should also examine the associations between PRF and parenting variables in a causal manner, as some of the associations may be bidirectional.

Based on both our findings and earlier literature, especially the prementalizing scale seems to be valid in discerning a variety of parental risks. The interest and curiosity scale, on the other hand, did not seem to discern negative parental factors, such as depression or hostility, very well and may rather be an indicator of positive parenting. Interestingly, our results also revealed curvilinear associations between maternal certainty and maternal reports of child‐related parenting outcomes, but in an unexpected direction. There has been controversy about whether the scale should be used in a linear or curvilinear manner, and our results suggest that this warrants further investigation. It is important to examine whether certainty also shows curvilinear effects with maternal interview‐based measures of representations and children's actual observed interaction with their mothers to rule out whether our findings only reflect idealization processes in maternal self‐report of their children.

CONFLICT OF INTEREST STATEMENT

The authors report no conflicts of interest

Supporting information

Supporting Information

IMHJ-46-564-s001.docx (19.3KB, docx)

ACKNOWLEDGMENTS

The study was funded by the Signe and Ane Gyllenberg Foundation.

Open access publishing facilitated by Helsingin yliopisto, as part of the Wiley ‐ FinELib agreement.

Flykt, M. , Haarala, H. , Pyhältö, K. , Minkkinen, J. , & Kalland, M. (2025). Validation of the Parental Reflective Functioning Questionnaire in Finnish mothers and fathers of toddlers. Infant Mental Health Journal, 46, 564–576. 10.1002/imhj.70006

Present address

Hannu Haarala is now in the City of Vantaa, Finland.

DATA AVAILABILITY STATEMENT

Data available on request due to privacy/ethical restrictions

REFERENCES

  1. Alismail, F. , Stacks, A. M. , Wong, K. , Brown, S. , Beeghly, M. , & Thomason, M. (2022). Maternal caregiving representations of the infant in the first year of life: Associations with prenatal and concurrent reflective functioning. Infant Mental Health Journal, 43, 311–327. 10.1002/imhj.21951 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Alvarez‐Monjarás, M. , McMahon, T. J. , & Suchman, N. E. (2019). Does maternal reflective functioning mediate associations between representations of caregiving with maternal sensitivity in a high‐risk sample? Psychoanalytic Psychology, 36(1), 82–92. 10.1037/pap0000166 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Ammaniti, M. , Baumgartner, E. , Candelori, C. , Perucchini, C. , Pola, M. , Tambelli, R. , & Zambino, F. (1992). Representations and narratives during pregnancy. Infant Mental Health Journal, 13, 167–182. [DOI] [Google Scholar]
  4. Ammaniti, M. , Tambelli, R. , & Perucchini, P. (1998). From pregnancy to the postnatal period: Stability and change of the factorial structure of maternal representations. Unpublished manuscript.
  5. Arkle, P. , Larkin, F. , Wang, Y. , Lee, Y. , Fernandez, A. , Li, L. Y. , & Meins, E. (2023). Early psychosocial risk factors and postnatal parental reflective functioning. Infancy, 28, 864–881. 10.1111/infa.12552 [DOI] [PubMed] [Google Scholar]
  6. Arnott, B. , & Meins, E. (2007). Links among antenatal attachment representations, postnatal mindmindedness, and infant attachment security: A preliminary study of mothers and fathers. Bulletin of the Menninger Clinic, 71, 132–149. 10.1521/bumc.2007.71.2.132 [DOI] [PubMed] [Google Scholar]
  7. Benbassat, N. , & Priel, B. (2015). Why is fathers´ reflective function important? Psychoanalytic Psychology, 32(1), 1–22. 10.1037/a0038022 [DOI] [Google Scholar]
  8. Brown, S. L. (2004). Family structure and child well‐being: The significance of parental cohabitation. Journal of Marriage and Family, 66, 351–367. 10.1111/j.1741-3737.2004.00025.x [DOI] [Google Scholar]
  9. Butitta, K. V. , Smiley, P. A. , Kerr, M. L. , Rasmussen, H. F. , Querdasi, F. R. , & Borelli, J. L. (2019). In a father's mind: Paternal reflective functioning, sensitive parenting, and protection against socioeconomic risk. Attachment and Human Development, 21, 445–466. 10.1080/14616734.2019.1582596 [DOI] [PubMed] [Google Scholar]
  10. Corlone, C. , Milan, S. , Decoste, C. , Borelli, J. L. , McMahon, T. J. , & Suchman, N. E. (2023). A self‐report measure of parental reflective functioning: A study of reliability and validity across three samples of varying clinical risk. Infant Mental Health Journal, 44, 240–254. 10.1002/imhj.22046 [DOI] [PubMed] [Google Scholar]
  11. Cooke, D. C. , Priddis, L. E. , Luyten, P. , Kendall, G. E. , & Cavanagh, R. F. (2017). Paternal and maternal reflective functioning in the West Australian peel child health study. Infant Mental Health Journal, 38, 561–574. 10.1002/imhj.21664 [DOI] [PubMed] [Google Scholar]
  12. De Roo, M. , Wong, G. , Rempel, G. R. , & Fraser, S. N. (2019). Advancing optimal development in children: Examining the construct validity of a parent reflective functioning questionnaire. JMIR Pediatrics and Parenting, 2(1), 1–12. 10.2196/11561 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Duschinsky, R. , & Foster, S. (2021) Mentalizing and epistemic trust: The work of Peter Fonagy and colleagues at the Anna Freud Centre. Oxford Academic. 10.1093/med-psych/9780198871187.001.0001 [DOI] [Google Scholar]
  14. Esbjørn, B. H. , Pedersen, S. H. , Daniel, S. I. , Hald, H. H. , Holm, J. M. , & Steele, H. (2013). Anxiety levels in clinically referred children and their parents: Examining the unique influence of self‐reported attachment styles and interview‐based reflective functioning in mothers and fathers. British Journal of Clinical Psychology, 52, 394–407. 10.1111/bjc.12024 [DOI] [PubMed] [Google Scholar]
  15. Flykt, M. , Punamäki, R.‐L. , Belt, R. , Pajulo, M. , Biringen, Z. , Salo, S. , & Posa, T. (2012). Maternal representations and dyadic interaction among drug‐abusing and non‐using mothers and their infants. Infant Mental Health Journal, 33, 123–128. 10.1002/imhj.21313 [DOI] [PubMed] [Google Scholar]
  16. Fonagy, P. (2008). The mentalization‐focused approach to social development. In Busch F. N. (Ed.), Mentalization: Theoretical considerations, research findings and clinical implications (pp. 3–56). The Analytic Press. [Google Scholar]
  17. Fonagy, P. , & Luyten, P. (2018) Attachment, mentalization, and the self. In Livesley, W. J. , & Larstone, R. (Eds.), Handbook of personality disorders (2nd ed., pp. 123–140). Guilford Press. [Google Scholar]
  18. Grossmann, K. , Grossmann, K. E. , Kindler, H. , & Zimmermann, P. (2008). A wider view of attachment and exploration: The influence of mothers and fathers on the development of psychological security from infancy to young adulthood. In Cassidy J. & Shaver P. R. (Eds.), Handbook of attachment: Theory, research, and clinical applications (pp. 857–879). Guilford Press. [Google Scholar]
  19. Hayes, A. F. , & Coutts, J. J. (2020). Use omega rather than Cronbach's alpha for estimating reliability. but…. Communication Methods and Measures, 14, 1–24. 10.1080/19312458.2020.1718629 [DOI] [Google Scholar]
  20. Hu, L.‐T. , & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6(1), 1–55. 10.1080/10705519909540118 [DOI] [Google Scholar]
  21. Kalland, M. , Fagerlund, Å. , von Koskull, M. , & Pajulo, M. (2016). Families first: The development of a new Mentalization‐based group intervention for first‐time parents to promote child development and family health. Primary Health Care Research & Development, 17, 3–17. 10.1017/S146342361500016X [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Kalland, M. , Salo, S. , Vincze, L. , Lipsanen, J. , Raittila, S. , Sourander, J. , Salvén‐Bodin, M. , & Pajulo, M. (2022). Married and cohabiting Finnish first‐time parents: Differences in wellbeing, social support and infant health. Social Sciences, 11(4), 181. 10.3390/socsci110 [DOI] [Google Scholar]
  23. Kamza, A. , Luyten, P. , & Piotrowski, K. (2024). Psychometric evaluation of the Parental Reflective Functioning Questionnaire in Polish mothers. PLoS One, 19(4), e0299427. 10.1371/journal.pone.0299427 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Katznelson, H. (2014). Reflective functioning: A review. Clinical Psychology Review, 34, 107–177. 10.1016/j.cpr.2013.12.003 [DOI] [PubMed] [Google Scholar]
  25. Kiwanuka, F. , Kopra, J. , Sak‐Dankosky, N. , Nanyonga, R. C. , & Kvist, T. (2022). Polychoric correlation with ordinal data in nursing research. Nursing Research, 71, 469–476. 10.1097/NNR.0000000000000614 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Konttinen, H. , Silventoinen, K. , Sarlio‐Lähteenkorva, S. , Männistö, S. , & Haukkala, A. (2010). Emotional eating and physical activity self‐efficacy as pathways in the association between depressive symptoms and adiposity indicators. American Journal of Clinical Nutrition, 92, 1031–1039. 10.3945/ajcn.2010.29732 [DOI] [PubMed] [Google Scholar]
  27. Lee, Y. , Meins, E. , & Larkin, F. (2021). Translation and preliminary validation of a Korean version of the Parental Reflective Functioning Questionnaire. Infant Mental Health Journal, 42, 47–59. 10.1002/imhj.21883 [DOI] [PubMed] [Google Scholar]
  28. Lindblom, J. , Pajulo, M. , Nolvi, S. , Tervahartiala, K. , Karlsson, H. , Karlsson, L. , & Korja, R. (2022). Early pathways of maternal mentalization: Associations with child development in the FinnBrain birth cohort study. Frontiers in Psychology, 13, 1–20. 10.3389/fpsyg.2022.855190 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Luyten, P. , Mayes, L. C. , Nijssens, L. , & Fonagy, P. (2017). The Parental Reflective Functioning Questionnaire: Development and preliminary validation. Plos One, 12(5), e0176218. 10.1371/journal.pone.0176218 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Madsen, E. B. , Væver, M. S. , Egmose, I. , Krogh, M. T. , Haase, T. W. , de Moor, M. H. M. , & Karstoft, K. I. (2023). Parental reflective functioning in first‐time parents and associations with infant socioemotional development. Journal of Child and Family Studies, 32, 2140–2152. 10.1007/s10826-023-02565-5 [DOI] [Google Scholar]
  31. Manlove, E. E. , & Vernon‐Feagans, L. (2002). Caring for infant daughters and sons in dual‐earner households maternal reports of father involvement in weekday time and tasks. Infant and Child Development, 11, 305–320. 10.1002/icd.260 [DOI] [Google Scholar]
  32. Moreira, H. , & Fonseca, A. (2023). Measuring parental reflective functioning: Further validation of the Parental Reflective Functioning Questionnaire in Portuguese mothers of infants and young children. Child Psychiatry and Human Development, 54, 1042–1054. 10.1007/s10578-021-01288-2 [DOI] [PubMed] [Google Scholar]
  33. Osborne, C. , Manning, W. D. , & Smock, P. J. (2007). Married and cohabiting parents' relationship stability: A focus on race and ethnicity. Journal of Marriage and Family, 69, 1345–1366. 10.1111/j.1741-3737.2007.00451.x [DOI] [Google Scholar]
  34. Osgood, C. E. , & Suci, G. J. (1952). A measure of relation determined by both mean difference and profile information. Psychological Bulletin, 49, 251–262. 10.1037/h0062981 [DOI] [PubMed] [Google Scholar]
  35. Pajulo, M. , Savonlahti, E. , Sourander, A. , Piha, J. , & Helenius, H. (2001). Prenatal maternal representations: Mothers at psychosocial risk. Infant Mental Health Journal, 22, 529–544. 10.1002/imhj.1016 [DOI] [Google Scholar]
  36. Pajulo, M. , Savonlahti, E. , Sourander, A. , Piha, J. , & Helenius, H. (2004). Maternal representations, depression and interactive behaviour in the postnatal period: A brief report. Journal of Reproductive and Infant Psychology, 22, 91–98. 10.1080/0264683042000205954 [DOI] [Google Scholar]
  37. Pajulo, M. , Helenius, H. , & Mayes, L. (2006). Prenatal views of baby and parenthood: Association with sociodemographic and pregnancy factors. Infant Mental Health Journal, 27, 229–250. 10.1002/imhj.20090 [DOI] [PubMed] [Google Scholar]
  38. Pajulo, M. , Tolvanen, M. , Pyykkönen, N. , Karlsson, L. , Mayes, L. , & Karlsson, H. (2018). Exploring parental mentalization in postnatal phase with a self‐report questionnaire (PRFQ): Factor structure, gender differences and association with sociodemographic factors. The Finn Brain Birth Cohort Study. Psychiatry Research, 262, 431–439. 10.1016/j.psychres.2017.09.020 [DOI] [PubMed] [Google Scholar]
  39. Pazzagli, C. , Delvecchio, E. , Raspa, V. , Mazzeschi, C. , & Luyten, P. (2018). The Parental Reflective Functioning Questionnaire in mothers and fathers of school‐aged children. Journal of Child and Family Studies, 27, 80–90. 10.1007/s10826-017-0856-8 [DOI] [Google Scholar]
  40. Radloff, L. S. (1977). The CES‐D Scale: A self‐report depression scale for research in the general population. Applied Psychological Measurement, 1(3), 385–401. 10.1177/014662167700100306 [DOI] [Google Scholar]
  41. Raley, S. , & Bianchi, S. (2006). Sons, daughters, and family processes: Does gender of children matter? Annual Review of Sociology, 32, 401–421. https://www.jstor.org/stable/29737745 [Google Scholar]
  42. Ruiz, N. , Witting, A. , Ahnert, L. , & Piskernik, B. (2020). Reflective functioning in fathers with young children born preterm and at term. Attachment & Human Development, 22(1), 32–45. 10.1080/14616734.2019.1589059 [DOI] [PubMed] [Google Scholar]
  43. Salo, S. J. , Pajulo, M. , Vince, L. , Raittila, S. , Sourander, J. , & Kalland, M. (2021). Parent relationship satisfaction and reflective functioning as predictors of emotional availability and infant behavior. Journal of Child and Family Studies, 30, 1214–1228. 10.1007/s10826-021-01934-2 [DOI] [Google Scholar]
  44. Salo, S. , Lipsanen, J. O. , Sourander, J. , Pajulo, M. , & Kalland, M. (2022). Parental relationship satisfaction, reflective functioning, and toddler behavioral problems: A longitudinal study from pregnancy to 2 years postpartum. Frontiers in Psychology, 13, 904409. 10.3389/fpsyg.2022.904409 [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Schechter, D. S. , Coots, T. , Zeanah, C. H. , Davies, M. , Coates, S. W. , Trabka, K. A. , Marshall, R. D. , Liebowitz, M. R. , & Myers, M. M. (2005). Maternal mental representations of the child in an inner‐city clinical sample: Violence‐related posttraumatic stress and reflective functioning, Attachment & Human Development, 7(3), 313–331. 10.1080/14616730500246011 [DOI] [PubMed] [Google Scholar]
  46. Slade, A. , Grieneberger, J. , Bernbach, E. , Levy, D. , & Locker, A. (2005). Maternal reflective functioning, attachment, and the transmission gap: A preliminary study. Attachment and Human Development, 7(3), 283–298. 10.1080/14616730500245880 [DOI] [PubMed] [Google Scholar]
  47. Slade, A. , & Sleed, M. (2024). Parental reflective functioning on the parent development interview: A narrative review of measurement, association, and future directions. Infant Mental Health Journal, 45, 464–480. 10.1002/imhj.22114 [DOI] [PubMed] [Google Scholar]
  48. Statistics Finland . (2018). At least one‐fourth of persons with foreign background had completed tertiary level qualifications https://stat.fi/til/vkour/2018/vkour_2018_2019-11-05_tie_001_en.html
  49. Vliegen, N. , Luyten, P. , & Biringen, Z. (2009). A multimethod perspective on emotional availability in the postpartum period. Parenting: Science and Practice, 9(3‐4), 228–243. 10.1080/15295190902844514 [DOI] [Google Scholar]
  50. Wendelboe, K. I. , Smith‐Nielsen, J. , Stuart, A. C. , Luyten, P. , & Skovgaard Væver, M. (2021). Factor structure of the Parental Reflective Functioning Questionnaire and association with maternal postpartum depression and comorbid symptoms of psychopathology. PLoS One, 16(8), e0254792. 10.1371/journal.pone.0254792 [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Wendelboe, K. I. , Nielsen, J. S. , Stuart, A. C. , & Vaever, M. S. (2022). The parental Reflective Functioning Questionnaire: Infant version in fathers of infants and association with paternal postpartum mental health. Infant Mental Health Journal, 43, 921–937. 10.1002/imhj.22023 [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Ye, P. , Ju, J. , Zheng, K. , Dang, J. , & Bian, Y. (2022). Psychometric evaluation of the Parental Reflective Functioning Questionnaire in Chinese parents. Frontiers in Psychology, 13, 745184. 10.3389/fpsyg.2022.745184 [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Zeegers, M. A. J. , Colonnesi, C. , Stams, G. J. M. , & Meins, E. (2017). Mind matters: A meta‐analysis on parental mentalization and sensitivity as predictors of infant–parent attachment. Psychological Bulletin, 143, 1245–1272. 10.1037/bul0000114 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supporting Information

IMHJ-46-564-s001.docx (19.3KB, docx)

Data Availability Statement

Data available on request due to privacy/ethical restrictions

Articles from Infant Mental Health Journal are provided here courtesy of Wiley

RESOURCES