Parent training effects on emotion recognition in mothers rearing children with attention-de cit/hyperactivity disorder: A randomized controlled functional magnetic resonance imaging study

Kai Makita (  kai@u-fukui.ac.jp ) University of Fukui: Fukui Daigaku https://orcid.org/0000-0002-7699-9485 Akiko Yao University of Fukui: Fukui Daigaku Koji Shimada University of Fukui: Fukui Daigaku Ryoko Kasaba University of Fukui: Fukui Daigaku Takashi X Fujisawa University of Fukui: Fukui Daigaku Yoshifumi Mizuno University of Fukui: Fukui Daigaku Akemi Tomoda University of Fukui: Fukui Daigaku


Introduction
There is a strong link between developmental disabilities in children and parenting di culties (Woodman et al., 2015). Attentionde cit/hyperactivity disorder (ADHD), one of the most commonly diagnosed neurodevelopmental disorders in childhood, has a relatively high incidence (approximately 5%) (American Psychiatric Association, 2013). Children with ADHD have functional problems in social and academic areas (Marshall et al., 2014). In addition, ADHD symptoms in children can negatively affect the caregivers' psychological state, parenting behavior, and parent-child relationships (Hutchison et al., 2016;Pelham et al., 1998). These negative in uences could result in inappropriate parental responses, such as maltreatment, creating risk factors for the later development of conduct problems in the affected children (Chronis-Tuscano et al., 2008;Chronis et al., 2007).
Parent training (PT) is a psychosocial and support intervention that promotes a nurturing environment for child care (parental understanding and response), widely known for its effectiveness (Daley et al., 2014) and recommended for parents of children with ADHD (Posner et al., 2020). The PT program for ADHD includes essential information for parents on the characteristics of the disability and treatment options (Iwasaka, 2012;Zwi et al., 2011). Studies using PT for mothers rearing ADHD children have reported that PT signi cantly improved their stress and parenting self-e cacy and style (reduced overly harsh responses, lack of consistency, and ineffective limit-setting) ( A previous neuroimaging study reported that tasks that measure social skills and abilities, such as mind reading tasks ("Reading the mind in the eyes task [RMET] (Baron-Cohen et al., 2001)), are negatively in uenced by stress (Nolte et al., 2013) or depressive symptoms and may, therefore, be useful biomarkers of caregivers' depression/stress (Shimada et al., 2018 Shimada et al., 2018). However, it is unclear whether improving a caregiver's stress by PT also improves the activation of these areas.
Herein, we focused on the effects of PT on parenting stress in mothers of children with ADHD and examined its effects at the neurological level. We hypothesized that PT would improve qualitative stress and unsound parenting practice index, as determined using parenting stress and discipline questionnaires. Moreover, we hypothesized that functional changes in the brains of mothers, especially in the regions involved in judging other's mental status, could be revealed by blood oxygenation level-dependent (BOLD) fMRI during the RMET.
Understanding the effects of PT at a functional neural level could potentially aid the development of more precise treatment strategies and biomarkers that could evaluate treatment effects.

Participants
Thirty mothers (mean age=38.72 years; standard deviation [SD]=4.46 years) agreed to participate in this study. The inclusion criteria were as follows: mothers caring for ≥1 school-aged child (6~12 years old) diagnosed with ADHD in hospitals or pediatric clinics. The diagnosis of ADHD was performed using the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (American Psychiatric Association, 2013). The exclusion criteria were as follows: mothers who had participated in other PT programs within two months before the time of enrolment, and changes in the child's medication status during the study.
All participants had normal or corrected-to-normal vision with no known medical, neurological, or psychiatric history (based on self-report questionnaires) and met the safety requirements for participation in an MRI study. The Ethics Committee of the University of Fukui approved the study protocol; all procedures were conducted in accordance with the Declaration of Helsinki. All participants provided written informed consent.

Study design
Using a permuted-block randomization procedure as a computer-based random number generator, mothers were randomly allocated to two groups: mothers who attended a 13-week PT session (PT group), and mothers who did not attend such a session until the end of the study (non-PT group). In the PT group, mothers were taught and trained on the following contents by two trained clinical psychologists ). In the experiment, the participants' psychological characteristics were measured using psychological questionnaires, and their brain activities during emotion/gender judgement were examined using fMRI. Among the participants (30 mother-child dyads), the data of seven mothers did not reach postbaseline evaluations; thus, the e cacy analysis included 13 dyads who received PT and 10 dyads who did not ( Figure 1).
---- Figure 1 insert around here Questionnaires for psychological testing The Interpersonal Reactivity Index (IRI) (Davis, 1983;Sakurai, 1988), which comprises four subscales (empathic concern, personal distress, perspective taking, and fantasy), was used to measure the mothers' empathic abilities, and the Beck Depression Inventory-II (BDI-II) (Beck et al., 1996;Kojima et al., 2002) was used to measure the mothers' current depression. To measure maternal parenting stress, we used the Parental Stress Index (PSI) (Abidin, 1995;Narama et al., 1999), which includes the child and parent stressor domains. In both domains, higher scores indicate that the rater scored the child's behavior more negatively or that the rater negatively assessed themselves or the surrounding environment in parenting. The Parenting Scale (PS) (Arnold et al. 1993;Itani 2010) was used to examine the mothers' parenting disciplines in laxness and overreactivity; this refers to the tendency to respond permissively (by passive and inconsistent parenting practices) or harshly (by overly authoritarian parenting practices), respectively, to child misbehavior. During the study, the BDI-II, PSI, and PS scores were measured twice (Time 1 and Time 2), and used to assess the effects of PT.

Task and procedures
In the MRI study, mothers completed the revised RMET version (Baron-Cohen et al., 2001; Shimada et al., 2018). The task consisted of 36 trials. In each trial, the participant was presented with a black and white photograph showing the eye region of a human face with two words placed below (one of which correctly identi ed the person's mindset or gender). The participants were asked to indicate the word they felt best described what the person in the photograph was thinking or feeling (in this study "ToM" condition) or the gender (male or female) depicted (in this study "GeN" condition) by pressing one of two buttons in the scanner (Current Design, Philadelphia, USA). The In the imaging data processing, for MRI signal equilibrium, the rst four volumes were discarded from the analysis. The remaining 240 images were preprocessed and analysed using the Statistical Parametric Mapping software (SPM ver.12, Wellcome Trust Center for Neuroimaging, London, UK) implemented in MATLAB R2019b (MathWorks, Natick, USA). Following slice timing correction and spatial realignment, T1-weighted anatomical images were co-registered to functional images and segmented using a new segmentation algorithm with diffeomorphic anatomical registration through the exponentiated lie algebra (DARTEL) technique (Ashburner, 2007).
Functional images were then spatially normalized into the Montreal Neurological Institute template, re-sampled into a spatial resolution of 3 × 3 × 3 mm³, and spatially smoothed with a 6 mm full width at half-maximum Gaussian kernel.
After MRI data pre-processing, we rst performed individual analyses to identify task-related activity. In this analysis, a design matrix containing two regressors of interest (ToM and GeN conditions) and six motion parameters, as nuisance regressors, was created. Then, two contrast images (ToM and GeN) were obtained for each participant. Brain activation during each condition was modelled with a general linear model, by using a box-car function convolved with the canonical hemodynamic response function. the PT group, we conducted an SPM one-way analysis of variance (ANOVA) with a within-subject design to compare separately the activities within the PT and non-PT groups under both conditions. The statistical threshold was set at p<0.005, uncorrected at the peak level, and at p<0.05 at the cluster level, family-wise error-corrected for multiple comparisons over the whole brain. The Neuromorphometrics atlas (http://www.neuromorphometrics.com/) was used to identify the anatomical localization of signi cant clusters.

Questionnaires and behavioral task data analyses
Fisher's exact tests and individual t-tests were performed to compare the clinical and demographic characteristics between groups. Questionnaire scores and behavioral performance during the RMET were compared using mixed-model repeated measures ANOVAs with time as the withinsubject factor (pre-and post-PT duration: Time 1 & Time 2, and groups as the between-group factor (PT and non-PT).
For the RMET performance, the intervention effects were assessed on reaction times (RTs) and correct answer rate (accuracy) separately for the ToM and GeN conditions. To further investigate signi cant effects, Bonferroni-corrected pair-wise comparisons were performed. Statistical analyses were performed using IBM SPSS version 24 (IBM Corp., Armonk, NY, USA).

Demographics and assessment outcomes at Time 1
There were no signi cant differences between the PT and non-PT groups regarding age; BDI-II, STAI, PSI, PS, and IRI scores; years of education; and handedness at Time 1 (Table 1; Supplementary Material 2, Table S1, children's data in both groups). ----  Notes: Accuracy, correct answer rate (%); PT group, mothers planned to be enrolled in the parent training intervention; non-PT group, mothers who were not enrolled in the parent training intervention after the end of the study.

Imaging results
Regarding the between-and within-group differences of the regional brain activation (  Table 3), whereas no such difference was found in the non-PT group. Given the few studies investigating the intervention effect of PT on brain function in mothers caring for children with ADHD, and to control the balance between Type I and II error risks, we explored the potential effects of PT by using p<0.005, uncorrected with a cluster size of k ≥40 voxels (Lieberman & Cunningham, 2009;Nishiyama et al., 2015). We observed bilateral clusters of increased activation in the calcarine cortex, left inferior temporal gyrus, and right inferior occipital gyrus (Supplementary Material 3, Table S2). Other tests did not show any signi cant differences. Notes: The threshold was set at p < 0.005, uncorrected at the peak level and p < 0.05, family-wise error-corrected for multiple comparisons at the cluster level. Locations were de ned using the SPM neuromorphometrics atlas.

Discussion
In our study, mothers enrolled in the PT program exhibited less parenting stress as measured by the PSI, including both the child and parent domains. PS scores revealed that mothers in the PT group adopted fewer inappropriate parenting practices with respect to overreactivity. These results suggest that PT promotes the mothers' understanding of the behavior in children with ADHD, thus reducing their stress and improving parenting disciplines. Our ndings are in-line with those of previous studies using PT for mothers with ADHD Regarding the changes in regional brain activation, within-group comparisons showed signi cantly increased activation in the occipital part of the fusiform gyrus during others' mental state judgement from facial features only in the PT group during the ToM condition. the changes in brain activity in the PT group might re ect their increased sensitivity to the other's mental states.
Altogether, our ndings suggest that PT interventions in mothers provide them with a better understanding of their children, helping to create a more nurturing environment, and reducing parenting stress, subsequently improving activation in the fusiform gyrus during socioemotional processing.
This study has some limitations. First, it included a relatively small sample size and lacked a control group comprising mothers of typically developed children. Second, as all psychometric measures were self-reported, subjective bias might be present. Lastly, because of the single-center study design, the participants' demographic variability might be limited. Future research to address these limitations is warranted.

Conclusions
In summary, our ndings suggest that PT promotes mothers' understanding of problematic behavior in children with ADHD and might enhance mothers' ability to process socioemotional information (recognize other's emotional expression), which could help build a more nurturing environment, reduce parenting stress and maladaptive parenting style, and improve the mother-child relationship. Flow diagram of this study.
Abbreviations: Parent training (PT) group, mothers planned to be enrolled in the parent training intervention, non-PT group, mothers who were not enrolled in parent training intervention after the end of the study.

Figure 2
Regions showing signi cant differences between Time 1 and Time 2 in the PT group during the RMET.
The statistical threshold was set at p < 0.005, uncorrected at the peak level, and p < 0.05, family-wise error-corrected for multiple comparisons at the cluster level. Locations were defined using the neuromorphometrics atlas implemented in SPM.
Abbreviations: PT group, mothers enrolled in the parent training intervention. PT, parent training, RMET, Reading the Mind in the Eyes Test.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. MakitaSupplementaryMaterial.doc AuthorChecklist.pdf