Marianne Thode Krogh
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Marianne Thode Krogh has a total of 19 co-authors. Here are the top co-authors by number of shared publications:
- Mette Skovgaard Væver
Department of Psychology University of Copenhagen Copenhagen Denmark
10 shared publications - Anne Christine Stuart
Department of Psychology University of Copenhagen Copenhagen Denmark
5 shared publications - Eva Back Madsen
Department of Psychology Center for Early Intervention and Family Studies University of Copenhagen Copenhagen Denmark
4 shared publications Ida Egmose
Centre of Excellence in Early Intervention and Family Studies (CIF), Department of Psychology, University of Copenhagen, Copenhagen K, Denmark
3 shared publications- Ida Egmose
Department of Psychology Center for Early Intervention and Family Studies University of Copenhagen Copenhagen Denmark
3 shared publications - Johanne Smith-Nielsen
Department of Psychology University of Copenhagen Copenhagen Denmark
3 shared publications
Latest Publications
- Open Access
- 1 March 2026
- Vol. 29 (2) , e70156
Evaluating Open-Source Solutions for Computerized Inference of Infant Facial Affect.
Infant affect is often expressed through facial expressions, making this modality a key source of insight into the child's well-being and social functioning. Computational inference of infant affect could critically assist both researchers and clinicians working with infant development and mitigate the need for manual coding. While many studies have explored open-source solutions in the adult domain, only the commercial Baby FaceReader 9 exists for the infant domain. To address this gap, we utilize the recently proposed, open-source infant-native action unit (AU) detection library PyAFAR (Python-based Automated Facial Action Recognition) on a sample of 71 four-month-old infants, whose facial expressions were manually annotated frame-by-frame for three minutes according to the Infant Facial Affect (IFA) coding scheme. Using these AUs as features, we classify facial affect into negative, neutral, and positive using XGBoost and Bayesian filtering, both in a multiclass and a binary setup. Our results show that AUs estimates from PyAFAR, combined with an XGBoost classification model, can distinguish positive from neutral and positive from negative affect with AUC scores of 0.78 and 0.76, respectively. This performance is essentially on par with that reported in evaluation studies of the Baby FaceReader 9, when accounting for differences in study setup. Our work indicates that the area of infant facial affect is particularly well-suited to supervised learning, given the availability of two distinct, commensurable measurement schemes that underpin the same phenomenon. Finally, we discuss how future iterations of PyAFAR may benefit from including AUs that capture more variability around infant forehead and mouth opening.
- 1 March 2026
- Vol. 29 (2) , e70156
Open AccessEvaluating Open-Source Solutions for Computerized Inference of Infant Facial Affect.
Infant affect is often expressed through facial expressions, making this modality a key source of insight into the child's well-being and social functioning. Computational inference of infant affect could critically assist both researchers and clinicians working with infant development and mitigate the need for manual coding. While many studies have explored open-source solutions in the adult domain, only the commercial Baby FaceReader 9 exists for the infant domain. To address this gap, we utilize the recently proposed, open-source infant-native action unit (AU) detection library PyAFAR (Python-based Automated Facial Action Recognition) on a sample of 71 four-month-old infants, whose facial expressions were manually annotated frame-by-frame for three minutes according to the Infant Facial Affect (IFA) coding scheme. Using these AUs as features, we classify facial affect into negative, neutral, and positive using XGBoost and Bayesian filtering, both in a multiclass and a binary setup. Our results show that AUs estimates from PyAFAR, combined with an XGBoost classification model, can distinguish positive from neutral and positive from negative affect with AUC scores of 0.78 and 0.76, respectively. This performance is essentially on par with that reported in evaluation studies of the Baby FaceReader 9, when accounting for differences in study setup. Our work indicates that the area of infant facial affect is particularly well-suited to supervised learning, given the availability of two distinct, commensurable measurement schemes that underpin the same phenomenon. Finally, we discuss how future iterations of PyAFAR may benefit from including AUs that capture more variability around infant forehead and mouth opening.
- Open Access
- 24 February 2026
- Published by Wiley in Developmental Science
- Vol. 29 (2)
Evaluating Open‐Source Solutions for Computerized Inference of Infant Facial Affect
Infant affect is often expressed through facial expressions, making this modality a key source of insight into the child's well‐being and social functioning. Computational inference of infant affect could critically assist both researchers and clinicians working with infant development and mitigate the need for manual coding. While many studies have explored open‐source solutions in the adult domain, only the commercial Baby FaceReader 9 exists for the infant domain. To address this gap, we utilize the recently proposed, open‐source infant‐native action unit (AU) detection library PyAFAR (Python‐based Automated Facial Action Recognition) on a sample of 71 four‐month‐old infants, whose facial expressions were manually annotated frame‐by‐frame for three minutes according to the Infant Facial Affect (IFA) coding scheme. Using these AUs as features, we classify facial affect into negative, neutral, and positive using XGBoost and Bayesian filtering, both in a multiclass and a binary setup. Our results show that AUs estimates from PyAFAR, combined with an XGBoost classification model, can distinguish positive from neutral and positive from negative affect with AUC scores of 0.78 and 0.76, respectively. This performance is essentially on par with that reported in evaluation studies of the Baby FaceReader 9, when accounting for differences in study setup. Our work indicates that the area of infant facial affect is particularly well‐suited to supervised learning, given the availability of two distinct, commensurable measurement schemes that underpin the same phenomenon. Finally, we discuss how future iterations of PyAFAR may benefit from including AUs that capture more variability around infant forehead and mouth opening.
- 24 February 2026
- Published by Wiley in Developmental Science
- Vol. 29 (2)
Open AccessEvaluating Open‐Source Solutions for Computerized Inference of Infant Facial Affect
Infant affect is often expressed through facial expressions, making this modality a key source of insight into the child's well‐being and social functioning. Computational inference of infant affect could critically assist both researchers and clinicians working with infant development and mitigate the need for manual coding. While many studies have explored open‐source solutions in the adult domain, only the commercial Baby FaceReader 9 exists for the infant domain. To address this gap, we utilize the recently proposed, open‐source infant‐native action unit (AU) detection library PyAFAR (Python‐based Automated Facial Action Recognition) on a sample of 71 four‐month‐old infants, whose facial expressions were manually annotated frame‐by‐frame for three minutes according to the Infant Facial Affect (IFA) coding scheme. Using these AUs as features, we classify facial affect into negative, neutral, and positive using XGBoost and Bayesian filtering, both in a multiclass and a binary setup. Our results show that AUs estimates from PyAFAR, combined with an XGBoost classification model, can distinguish positive from neutral and positive from negative affect with AUC scores of 0.78 and 0.76, respectively. This performance is essentially on par with that reported in evaluation studies of the Baby FaceReader 9, when accounting for differences in study setup. Our work indicates that the area of infant facial affect is particularly well‐suited to supervised learning, given the availability of two distinct, commensurable measurement schemes that underpin the same phenomenon. Finally, we discuss how future iterations of PyAFAR may benefit from including AUs that capture more variability around infant forehead and mouth opening.
- Open Access
- 12 May 2025
- Published by Wiley in Infant Mental Health Journal: Infancy and Early Childhood
- Vol. 46 (6) , 870-885
Profiles of parental reflective functioning and mind‐mindedness in first‐time parents of 4‐month‐old infants
Profiles of parental reflective functioning (PRF) may offer valuable insights into individual differences in PRF. Previous studies have identified PRF profiles; however, further validation is needed. This study aimed to investigate the convergent validity of the identified PRF profiles by examining their association with a related measure of parental mentalization: representational mind‐mindedness. Participants were 1168 first‐time mothers and 490 fathers living in Denmark. In an online survey, parents reported on the Parental Reflective Functioning Questionnaire—Infant Version and described their 4‐month‐old infant in a free‐text field, which was subsequently coded for the frequency of mental attributes used to describe the infant, indicating representational mind‐mindedness. Results support previous findings indicating potential at‐risk and typical PRF profiles, supporting the convergent validity of these profiles. Results showed that mothers in the High Pre‐Mentalizing profile reported the lowest levels of representational mind‐mindedness compared to the other maternal profiles. For fathers, members of Moderate Interest reported lower levels of representational mind‐mindedness compared to Moderately Low Certainty and Moderately High Certainty . The current study extends previous work by showing evidence of convergent validity of the PRF profiles, yet further research is still needed to examine the profiles’ generalizability and associations with observed infant and parent outcomes.
- 12 May 2025
- Published by Wiley in Infant Mental Health Journal: Infancy and Early Childhood
- Vol. 46 (6) , 870-885
Open AccessProfiles of parental reflective functioning and mind‐mindedness in first‐time parents of 4‐month‐old infants
Profiles of parental reflective functioning (PRF) may offer valuable insights into individual differences in PRF. Previous studies have identified PRF profiles; however, further validation is needed. This study aimed to investigate the convergent validity of the identified PRF profiles by examining their association with a related measure of parental mentalization: representational mind‐mindedness. Participants were 1168 first‐time mothers and 490 fathers living in Denmark. In an online survey, parents reported on the Parental Reflective Functioning Questionnaire—Infant Version and described their 4‐month‐old infant in a free‐text field, which was subsequently coded for the frequency of mental attributes used to describe the infant, indicating representational mind‐mindedness. Results support previous findings indicating potential at‐risk and typical PRF profiles, supporting the convergent validity of these profiles. Results showed that mothers in the High Pre‐Mentalizing profile reported the lowest levels of representational mind‐mindedness compared to the other maternal profiles. For fathers, members of Moderate Interest reported lower levels of representational mind‐mindedness compared to Moderately Low Certainty and Moderately High Certainty . The current study extends previous work by showing evidence of convergent validity of the PRF profiles, yet further research is still needed to examine the profiles’ generalizability and associations with observed infant and parent outcomes.
- Open Access
- 1 October 2024
- Published by Taylor & Francis in Parenting
- Vol. 24 (4) , 154-181
Identifying Profiles of Parental Reflective Functioning in First-Time Parents and Associations with Parental Attachment and Infant Socioemotional Adjustment
- 1 October 2024
- Published by Taylor & Francis in Parenting
- Vol. 24 (4) , 154-181
Open AccessIdentifying Profiles of Parental Reflective Functioning in First-Time Parents and Associations with Parental Attachment and Infant Socioemotional Adjustment
- 7 May 2023
- Published by Taylor & Francis in European Journal of Developmental Psychology
- Vol. 20 (5) , 889-902
Reliability and validity of representational mind-mindedness in mothers of infants
Mind-mindedness (MM) refers to the parent’s ability to treat the child as an individual with a mind of his or her own. Studies have found representational and interactional MM to predict child development, but more research is needed on the validity of representational MM in parents of infants. Therefore, we examine the reliability and validity of representational MM in 46 mothers of infants. Representational MM was measured at 4 and 11 months, and interactional MM at 11 months. The results supported the reliability of representational MM by showing a moderate rank-order stability over a 7-month-period, although the stability was due to associations between representational MM and maternal age. The results also supported the validity of representational MM by showing concurrent and longitudinal associations between representational and interactional MM. Finally, the results suggested that the capacity for appropriate mind-related comments is more stable than the tendency to use non-attuned mind-related comments.
- 7 May 2023
- Published by Taylor & Francis in European Journal of Developmental Psychology
- Vol. 20 (5) , 889-902
Reliability and validity of representational mind-mindedness in mothers of infants
Mind-mindedness (MM) refers to the parent’s ability to treat the child as an individual with a mind of his or her own. Studies have found representational and interactional MM to predict child development, but more research is needed on the validity of representational MM in parents of infants. Therefore, we examine the reliability and validity of representational MM in 46 mothers of infants. Representational MM was measured at 4 and 11 months, and interactional MM at 11 months. The results supported the reliability of representational MM by showing a moderate rank-order stability over a 7-month-period, although the stability was due to associations between representational MM and maternal age. The results also supported the validity of representational MM by showing concurrent and longitudinal associations between representational and interactional MM. Finally, the results suggested that the capacity for appropriate mind-related comments is more stable than the tendency to use non-attuned mind-related comments.
- 1 November 2019
- Published by Elsevier in Infant Behavior and Development
- Vol. 57, 101352
Does gender affect Bayley-III scores and test-taking behavior?
- 1 November 2019
- Published by Elsevier in Infant Behavior and Development
- Vol. 57, 101352
Does gender affect Bayley-III scores and test-taking behavior?
- 18 June 2018
- Published by Taylor & Francis in European Journal of Developmental Psychology
- Vol. 16 (6) , 727-738
A longitudinal study of the predictive validity of the Bayley-III scales and subtests
The purpose of the present study was to examine the predictive validity of all scales and subtests from the Bayley-III. The sample consisted of 55 normal children who were assessed at six time points between 4 and 36 months. Significant correlations ranging from .28 to .79 were found for all ages and scales/subtests, but the three main scales differed with respect to predictive validity. The Cognitive Scale was predictive from 7–10, 7–13 and 10–13 months but not from 13 months and onwards. The Language Scale was predictive from 10 months and onwards and the Motor Scale was predictive at all ages, but only with respect to adjacent and not distant assessment points.
- 18 June 2018
- Published by Taylor & Francis in European Journal of Developmental Psychology
- Vol. 16 (6) , 727-738
A longitudinal study of the predictive validity of the Bayley-III scales and subtests
The purpose of the present study was to examine the predictive validity of all scales and subtests from the Bayley-III. The sample consisted of 55 normal children who were assessed at six time points between 4 and 36 months. Significant correlations ranging from .28 to .79 were found for all ages and scales/subtests, but the three main scales differed with respect to predictive validity. The Cognitive Scale was predictive from 7–10, 7–13 and 10–13 months but not from 13 months and onwards. The Language Scale was predictive from 10 months and onwards and the Motor Scale was predictive at all ages, but only with respect to adjacent and not distant assessment points.
- 26 September 2016
- Published by Wiley in Scandinavian Journal of Psychology
- Vol. 57 (6) , 501-508
Bayley‐III: Cultural differences and language scale validity in a Danish sample
The purpose of this study was to investigate cultural differences between Danish and American children at 2 and 3 years as measured with the developmental test Bayley-III, and to investigate the Bayley-III Language Scale validity. The Danish children (N = 43) were tested with the Bayley-III and their parents completed an additional language questionnaire (the MacArthur-Bates CDI). Results showed that scores from the Danish children did not differ significantly from the American norms on the Cognitive or Motor Scale, but the Danish sample scored significantly higher on the Language Scale. A comparison of the Bayley-III Language subtests with the CDI showed that the two measures correlated significantly, but the percentile score from the CDI was significantly higher than the percentile score from the Bayley-III Language subtests. This could be because the two instruments measure slightly different areas of language development, or because the Bayley-III overestimates language development in Danish children. However, due to the limitations of the current study, further research is needed to clarify this issue.
- 26 September 2016
- Published by Wiley in Scandinavian Journal of Psychology
- Vol. 57 (6) , 501-508
Bayley‐III: Cultural differences and language scale validity in a Danish sample
The purpose of this study was to investigate cultural differences between Danish and American children at 2 and 3 years as measured with the developmental test Bayley-III, and to investigate the Bayley-III Language Scale validity. The Danish children (N = 43) were tested with the Bayley-III and their parents completed an additional language questionnaire (the MacArthur-Bates CDI). Results showed that scores from the Danish children did not differ significantly from the American norms on the Cognitive or Motor Scale, but the Danish sample scored significantly higher on the Language Scale. A comparison of the Bayley-III Language subtests with the CDI showed that the two measures correlated significantly, but the percentile score from the CDI was significantly higher than the percentile score from the Bayley-III Language subtests. This could be because the two instruments measure slightly different areas of language development, or because the Bayley-III overestimates language development in Danish children. However, due to the limitations of the current study, further research is needed to clarify this issue.
- 9 September 2016
- Published by Wiley in Scandinavian Journal of Psychology
- Vol. 57 (6) , 571-583
Effects of maternal postpartum depression in a well‐resourced sample: Early concurrent and long‐term effects on infant cognitive, language, and motor development
This study examined early and long‐term effects of maternal postpartum depression on cognitive, language, and motor development in infants of clinically depressed mothers. Participants were 83 mothers and their full‐term born children from the urban region of Copenhagen, Denmark. Of this group, 28 mothers were diagnosed with postnatal depression three to four months postpartum in a diagnostic interview. Cognitive, language, and motor development was assessed with the Bayley Scales of Infant and Toddler Development third edition, when the infants were 4 and 13 months of age. We found that maternal postpartum depression was associated with poorer cognitive development at infant age four months, the effect size being large (Cohen's d = 0.8) and with similar effects for boys and girls. At 13 months of age infants of clinical mothers did not differ from infants of non‐clinical mothers. At this time most (79%) of the clinical mothers were no longer, or not again, depressed. These results may indicate that maternal depression can have an acute, concurrent effect on infant cognitive development as early as at four months postpartum. At the same time, in the absence of other risk factors, this effect may not be enduring. The main weaknesses of the study include the relatively small sample size and that depression scores were only available for 35 of the non‐clinical mothers at 13 months.
- 9 September 2016
- Published by Wiley in Scandinavian Journal of Psychology
- Vol. 57 (6) , 571-583
Effects of maternal postpartum depression in a well‐resourced sample: Early concurrent and long‐term effects on infant cognitive, language, and motor development
This study examined early and long‐term effects of maternal postpartum depression on cognitive, language, and motor development in infants of clinically depressed mothers. Participants were 83 mothers and their full‐term born children from the urban region of Copenhagen, Denmark. Of this group, 28 mothers were diagnosed with postnatal depression three to four months postpartum in a diagnostic interview. Cognitive, language, and motor development was assessed with the Bayley Scales of Infant and Toddler Development third edition, when the infants were 4 and 13 months of age. We found that maternal postpartum depression was associated with poorer cognitive development at infant age four months, the effect size being large (Cohen's d = 0.8) and with similar effects for boys and girls. At 13 months of age infants of clinical mothers did not differ from infants of non‐clinical mothers. At this time most (79%) of the clinical mothers were no longer, or not again, depressed. These results may indicate that maternal depression can have an acute, concurrent effect on infant cognitive development as early as at four months postpartum. At the same time, in the absence of other risk factors, this effect may not be enduring. The main weaknesses of the study include the relatively small sample size and that depression scores were only available for 35 of the non‐clinical mothers at 13 months.