Mindfulness for Children with Autism Spectrum Disorder: The Effect on Face and Emotion Recognition
Master’s thesis
Graduate School of Child Development and Education University of Amsterdam
S. M. Burger
Supervisor and first reader: mw. dr. E.I. de Bruin Second reader: mw. dr. M. Majdandzic Amsterdam, december 2018
Abstract
Children with Autism Spectrum Disorder (ASD) show impairments in face and emotion recognition, skills important for social interaction. This study focused on the effect of a Mindfulness Based Program (MBP) for children and adolescents with ASD on face and emotion recognition. It was hypothesized that the MBP would improve central coherence and theory of mind and therefore children would become more accurate and faster in face and emotion recognition. The MBP was a 9-week program delivered by child and family mental health care professionals. Forty-seven children diagnosed with ASD (age 8-23, 80.9% boys) participated. Face and emotion recognition were measured pre-intervention, post-intervention and at 2-month follow-up with the subtests ‘Face Recognition’ and ‘Identification of Facial Expressions’ of the ANT2.1. Repeated measures ANOVA showed that children were faster in face recognition after the training and faster in emotion recognition two months after the training, but not directly after the training. The number of errors children made remained stable, indicating that the increase in speed was not at the cost of accuracy. It is recommended that future research focuses on how children recognize faces and emotions before and after a MBP, to gain more insight into the mechanisms of change.
Keywords: autism spectrum disorder, mindfulness, face recognition, emotion
recognition
Samenvatting
Kinderen met Autisme Spectrum Stoornis (ASS) zijn vaak beperkt in het herkennen van gezichten en emoties; vaardigheden die belangrijk zijn voor sociale interactie. Deze studie onderzocht het effect van een mindfulness programma (MBP) voor kinderen en adolescenten met ASS op emotie- en gezichtsherkenning. Verwacht werd dat de MBP centrale coherentie en theory of mind zou verbeteren, waardoor kinderen sneller en accurater zouden zijn in emotie- en gezichtsherkenning. De MBP was een 9-weekse training, gegeven door
hulpverleners in de GGZ, gespecialiseerd in mindfulness trainingen. Zevenenveertig kinderen met ASS (8-23 jaar, 80.9% jongens) namen deel. Emotie- en gezichtsherkenning werden gemeten voor, na en twee maanden na de training met de subtesten ‘Face recognition’ en ‘Identification of Facial Expressions’ van de ANT2.1. Repeated measures ANOVA toonde aan dat kinderen sneller waren in gezichtsherkenning na de training en sneller waren in emotieherkenning twee maanden na de training, maar niet direct na de training. Het aantal fouten dat kinderen maakte veranderde niet, hetgeen aangeeft dat de toegenomen snelheid niet ten koste ging van accuraatheid. Voor vervolgonderzoek wordt aangeraden zich te focussen
op de manier waarop kinderen emoties en gezichten herkennen voor en na een MBP, om meer inzicht te krijgen in de mechanismes van verandering.
Keywords: autisme spectrum stoornis, mindfulness, gezichtsherkenning,
Mindfulness for Children with ASD: The Effect on Face and Emotion Recognition Children with Autism Spectrum Disorder (ASD) have impairments in their social interactions and communication and show stereotyped patterns of behaviour, interests and activities (American Psychiatric Association, 2013). All children with ASD experience these symptoms, but there is great variability in the extent to which they
experience them and no two children with ASD are the same (Dimitriou, Leonard, Karmilhoff-Smith, Johnson, & Thomas, 2015). Besides these core symptoms, individuals with ASD often have problems with the recognition of faces and the recognition of emotions expressed on faces (Serra et al., 2003; Uljarevic & Hamilton, 2013). Since face recognition (FR) and facial emotion recognition (FER) are necessary for nonverbal communication (De Sonneville et al., 2002), social interaction (Harms, Martin, & Wallace, 2010) and adaptive functioning (Trevisan & Birmingham, 2016), impairment in these skills can lead to difficulties in social situations (e.g. not recognizing someone, not seeing in what emotional state someone is).It is therefore important to study ways in which children with ASD could improve their face and emotion recognition skills. Mindfulness might have a positive effect on face and emotion
recognition and therefore, this study focuses on the question whether mindfulness training affects face and emotion recognition in children and adolescents with ASD.
The ability to recognize faces is an important skill, necessary for social interaction (Petrakova, Sommer, Junge & Hildebrandt, 2018). In typical development, the older children get, the faster they become in recognizing faces (De Sonneville et al., 2002) and this skill keeps developing into
adulthood. It’s argued that an important factor in this development is the transition from feature-based face processing to holistic face processing (Mondloch, Geldart, Maurer, & Le Grand, 2003). Young children mainly use a feature-based face processing strategy: to recognize a face, they focus on different features of the face, such as the nose or the mouth and use these features to identify someone. Most adults and older children, on the other hand, use a holistic strategy for face recognition and focus on the spacing of the features of a face. Looking at the face as a ‘whole’ is less time consuming than looking at individual features, and is therefore a quicker method to recognize a face. Researchers do not agree on the age on which children start processing faces at a holistic manner; some findings indicate that children use a holistic face processing strategy from 10 years old (Mondloch et al., 2003), others already find holistic processing in 6-year-olds (Petrakova et al., 2018) or even in 4-year-olds (Freire & Lee, 2001). Besides the transition from a feature-based to a holistic face recognition strategy, the development of more general cognitive abilities, such as visual attention and working memory, could play a role in the development of face recognition (Kinnunen, Korkman, Laasonen, & Lahti-Nuuttila, 2013).
The ability to recognize emotions through facial expressions, FER, is also important for social interaction (Harms et al., 2010). The development of this skill starts young; infants of 3 months old can already discriminate between different emotional expressions of their parents (Walker-Andrews, Krogh-Jespersen, Mayhew, & Coffield, 2011). The development of FER differs per emotion. At 5 years old, children are nearly as accurate as adults in
& Maurer, 2010), and between age 6 and 8 they reach this level for anger and sadness as well (Durand, Gallay, Seigneuric, Robichon, & Baudouin, 2007). The ability to recognize fear, disgust and surprise keeps developing in adolescence (Lawrence, Campbell, & Skuse, 2015). De Sonneville et al. (2002) did not only focus on the accuracy of
recognizing emotions, but also on speed. Besides an improvement in accuracy in the recognition of sadness, anger and fear, they found that children from 7 to 10 became significantly faster in recognizing these emotions and they had not yet reached adults levels of speed at age 10. Like for FR, it is theorized that children become more accurate and faster in FER because they make a transition from feature-based processing to holistic processing (De Sonneville et al., 2002).
Most children with ASD do not develop FR and FER at the same rate as children without ASD. For FR, children with ASD need more time to recognize a face (Serra et al., 2003) and they make more errors (Annaz, Karmiloff-Smith, Johnson, & Thomas, 2009) than children without ASD. For FER, the meta-analysis of Uljarevic and Hamilton (2013) concluded that there is a general impairment in FER in individuals with ASD, independent of age or IQ. In the recognition of happiness, individuals with ASD only marginally performed worse than individuals without ASD, but in the
recognition of fear, anger, disgust, surprise and sadness, individuals with ASD
performed significantly worse. There are several potential explanations why children with ASD are impaired in their face and facial emotion recognition. One theory that explains this impairment is the theory of central
coherence. This theory states that individuals with ASD have a tendency to focus on
details instead of focusing on ‘the big picture’ (Happé & Frith, 2006), while typically developing children and adults process information globally, focussing on context and on meaning. This processing style, or weak central coherence, can be advantageous in certain situations, when details are of importance (e.g. when drawing or identifying individual notes in a chord). However, it can also interfere with general and social functioning (Loth, Gómez, & Happé, 2008) and explain several difficulties that individuals with ASD encounter, such as hypersensitivity or lack of generalization (Happé & Frith, 2006). Due to this weak central coherence, children with ASD might focus on the details of a face, and not develop a holistic strategy for face and emotion recognition. Supporting this idea, Dimitriou et al. (2015) did not find an ‘inversion effect’ in children (age 5-11) with ASD in a face recognition task: the children were equally good in recognizing faces that were upside-down as recognizing upside faces. This implies that the children use a feature-based processing style, instead of a holistic style, that is not disturbed by inversion. It is also found that children with ASD show greater attention to the mouth, then to the eyes, while typically developing children look mostly at the eyes (Annaz et al., 2009). However, there are also studies that do find that children with ASD use a holistic face processing strategy (Wiegelt, Koldewyn, & Kanwisher, 2012). Scherf, Behrmann, Minshew, and Luna (2008) for example did find the inversion-effect in children with ASD. Furthermore, Wolf et al. (2008) found that children with ASD were better in identifying a face part (e.g. eyes or mouth) when the face part was presented in the context of a whole face, instead of presented separately, what implies that children with ASD process faces holistically.
Besides central coherence, theory of mind might also play a role in face and emotion recognition (Trevisan &
Birmingham, 2016). Theory of mind is the ability to infer other people’s mental states, beliefs, desires, emotions and intentions (Kim et al., 2016). In order to infer these mental states, beliefs, desires, emotions and intentions, people use observable cues, such as actions or facial expressions (Meinhardt-Injac, Daum, Meinhardt, & Persike, 2018). To correctly estimate one’s emotions, one must therefore correctly observe and recognize the emotion expressed on a face. Meinhardt-Injac et al. (2018) found that in healthy students, face recognition abilities predicted theory of mind-level. On the other hand, in order to determine which emotion is expressed on a person’s face, an individual first has to be aware that that person has internal states, which is a component of theory of mind (Lee et al., 2014). Scores on a theory of mind test were also found to
predict the ability to recognize facial expressions. A deficit in theory of mind is argued to be a core feature of ASD (Baron-Cohen, 2000) that explains the social problems that individuals with ASD encounter. Since the development of this theory, a lot of research has confirmed that most individuals with ASD have an
impairment in their theory of mind (e.g. Baron-Cohen, Leslie, & Frith, 1985; Loukusa, Makinen, Kuusikko-Gauffin, Ebeling, & Moilanen, 2014). This impairment in theory of mind might contribute to difficulties with emotion recognition. Without insight and some understanding of other people experiencing emotions, recognizing facial emotions accurately could be more challenging.
Mindfulness might have an effect on face and emotion recognition. Originally deriving from Buddhist tradition,
mindfulness practice is more and more used in Western clinical psychology (Williams & Kabat-Zinn, 2011). Starting with Kabat-Zinn who developed Mindfulness Based Stress Reduction (MBSR) and treated patients with chronic pain (Kabat-Zinn, 1990),
mindfulness is now used as part of treatment for various populations, including
individuals with depression, high levels of anxiety or an addiction (Goldberg et al., 2018). Mindfulness is also used in the treatment of children with ASD and research has shown that it can lead to a decrease in social communication problems,
internalizing, externalizing and attention problems, rumination, stress and an improvement in emotional well-being (De Bruin, Blom, Smit, Van Steensel, & Bögels, 2015; Ridderinkhof, de Bruin, Blom, & Bogels, 2018). In mindfulness practice, participants practice in meditations to focus their attention on the present moment, on purpose and without judgement (Kabat-Zinn, 2001). Furthermore, participants try to develop an accepting and compassionate attitude towards experiences (Segal, Williams, & Teasdale, 2013).
Previous research has shown that the face recognition abilities of healthy
adolescents improved after a mindfulness training (Wimmer, Bellingrath, & Von Stockhausen, 2016) and English, Wisener, and Bailey (2018) found a positive
correlation between trait mindfulness and the ability to recognize facial emotions.
Mindfulness practice might also help children with ASD improve their FR and FER, for several reasons. First, mindfulness practice could help children with ASD to see the face as a ‘whole’ and use a holistic face processing strategy, instead of looking at individual components. In mindfulness meditations, participants learn to shift between widening and narrowing their
attention (Ridderinkhof et al., 2018). This might help children with ASD to widen their attention when they look at a face, and look at the whole face instead of focusing on individual features. This might improve their face and emotion recognition, since their face processing strategy would be more holistic.
Another way in which mindfulness practice could improve emotion recognition, concerns theory of mind. In mindfulness meditations, participants focus, amongst other things, on their own emotions: they learn to become more aware of how and where they feel emotions in their body. By doing this, participants create more
awareness of their own emotions. At the same time, however, they gain more insight in emotions in general. This could help them to be more aware of the emotions that others can experience (Block-Lerner, Adair, Plumb, Rhatigan, & Orsillo, 2007), which is a part of theory of mind. More awareness in what emotion someone expresses, might make it easier to recognize the emotion expressed on a face. Tan, Lo, and Macrea (2014) found that adults who did a brief meditation
exercise, were better at recognizing emotions when looking at imagines of eyes, displaying subtle expressions, than adults who had not done the meditation exercise. Accordingly, mindfulness could improve facial emotion recognition, by making participants more aware of emotional processes, what could make it easier to recognize an emotion expressed on a face.
The current study focused on the effect of a 9-week mindfulness-based program (MBP) on face and emotion
recognition in children and adolescents with ASD. It was expected that children with ASD would become faster and more accurate in recognizing faces and emotions after the MBP.
Method Participants
Participants in this study were Dutch children with ASD (N=47, 80.9% boys) recruited at different treatment centres, varying in age from 8 till 23 (M=13.35; SD= 3.35). Children were diagnosed with ASD (10.6% autism, 23.4% Asperger’s syndrome, 42.6% Pervasive Developmental Disorder-Not Otherwise Specified (PDD-NOS)) according to Diagnostic and Statistical Manual of Mental Health Disorders guidelines (4th ed., text revised; DSM-IV-TR; American Psychiatric Association, 2000) in the treatment centre where they received the mindfulness training or at another diagnostic centre. Most children attended regular education (90.24%) and all children had an estimated IQ ≥ 80, based on clinical judgement. For the nine children who had undergone an IQ-test, mean total IQ was 111.00 (SD=13.46). Of the total 47 participants, 41 completed the tasks at the pre-intervention occasion of measurement, 40 completed the tasks at the
post-intervention measurement and 34 participants at the 2-month follow-up. Procedure
Data was used from a larger non-experimental longitudinal study assessing the effectiveness of an MPB for children and their parents (Ridderinkhof et al., 2018). The study was approved by the Medical Ethics Committee of the Academic Medical Centre (AMC). Children with ASD and their parents were recruited at difference treatment centres and placed on a waiting list. Exclusion criteria were insufficient mastery of de Dutch language of parent or child; comorbid conduct disorder; current suicidal risk or psychotic disorder and current participation in another ongoing psychological
intervention. Children followed the 9-week MYmind training and their parents the
Waitlist Pre-intervention Post-intervention 2-month follow-up 1-year follow-up MBP
Figure 1. Measurement occasions.
parallel Mindful Parenting training. There were five measurement occasions: Eight weeks before the start of the training (waitlist), one week before the start of the training (pre-intervention), directly after the start of the training (post-intervention), a 2-month follow-up and a 1-year follow-up (see Figure 1). At these measurement occasions, participating children and their parents filled in online questionnaires at home on social communication problems, emotional and behavioural functioning, parenting and mindful awareness (see Ridderinkhof et al. (2018) for more details). For the
computerized task, the children came to the treatment centre. This study only used the data of the computerized task. Only the data of the pre-intervention, post-intervention and 2-month follow-up were used in this study, since only few participants completed the computerized task during the waiting list (N = 12) and at the 1-year follow-up (N = 18). All parents and children of 12 years and above signed informed consent.
Mindfulness Based Program The MBP that the participants received was the MYmind program; Mindfulness for Youngsters with ASD and their parents (De Bruin et al., 2015). During nine weeks, participants had weekly 1,5-hour sessions at one of the two participating treatment centres. There were different groups for adolescents (12 years and older; 6-8 participants per group) and children (8-12 years; 4-6 participants per group). At the same time, one or both parents followed the
parent sessions. On average, children participated in 8.06 sessions (SD = 1.36) with 40.4% completing all nine
sessions. Mothers participated on average in 5.66 sessions (SD = 3.61), fathers in 4.62 (SD = 4.01). Eight weeks after the last session, the parent- and the child/adolescent-group had a booster-session together, which was followed by 70.2% of the children and 48.9% of the parents.
Each session in the child/adolescent-group started with an overview of the coming session and continued with some educational theory, mindfulness exercises, inquiry discussions and the discussion of homework exercises. The mindfulness exercises were based on Mindfulness Based Cognitive Therapy (MBCT; Segal et al., 2013) and MBSR (Kabat-Zinn, 1982) and were for example breathing meditations, yoga, sound meditation or a body-scan. The sessions themes subsequently were; Week 1 What is attention?; Week 2 Attention for the body; Week 3 Attention for the breath; Week 4 Attention for stress; Week 5 Attention for external distractors; Week 6 Attention for internal distractors; Week 7 Attention for feelings (of oneself/other); Week 8 Attention for changes; Week 9 Attention for
mindfulness after training. During the week, participants were given homework, which consisted of meditation practices, diary registrations and reading handouts and could be completed in 15 (children) or 30 (parents) minutes.
The mindfulness training was delivered by child and family mental health care professionals who had received a teacher training of eight days for the MYmind training and had at least one year of experience with mindfulness. Weekly intervision was held between parent and child trainers and there was monthly supervision. Treatment integrity was
monitored by using the MYmind-Treatment Adherence and Competence Scale
(MYmind-TACS; Ridderinkhof et al., 2018). Sessions were videotaped and a trained research assistant rated the level of
adherence to the program of the trainer on a scale from 0 to 2 and the level of competence of the trainer (the extent to which the trainer showed mindfulness trainer skills, such as a curious attitude) on a scale from 1 to 5. The interrater reliability was excellent for the adherence scale (absolute agreement = 100%) and good for the competence scale (absolute agreement = 73.5%). For the MYmind child program, mean adherence score was 1.97 (SD = 0.09) and mean competence score was 4.86 (SD = 0.18). For the MYmind parent program, mean
adherence score was 1.80 (SD = 0.29) and mean competence score was 4.63 (SD = 0.26). This shows that the trainers closely followed the program and did not deviate from the protocol.
Instruments
Baseline speed: To obtain a baseline speed for pressing the response key, the subtest ‘Baseline Speed’ (BS) of the computerized Amsterdam
Neuropsychological Tasks (ANT2.1; De Sonneville, 2005) was used. In this subtest, 32 times a cross presented on the screen changed into a square. Participants had to press the response key as fast as possible with their preferred hand as soon as they saw the square.
Face recognition: To examine the face recognition skills of the children, the subtest ‘Face Recognition’ of the
computerized ANT2.1(De Sonneville, 2005) was used. This subtest measured the speed and accuracy of recognizing neutral faces. Participants were shown a colour picture of a neutral face of a boy, girl, man or woman, the target face, for 2.5 seconds.
Subsequently, four pictures of faces, matching the gender and age of the target face, were presented (see Figure 2). Participants had to answer whether or not one of these pictures was the target face. Forty of these trials were shown, of which 20 contained the target face. The number of times the participant correctly responded ‘yes’ (hits), correctly responded ‘no’ (correct rejections), incorrectly responded ‘no’ (misses) and incorrectly responded
‘yes’(false alarms) were registered, as well as the reaction time of the hits.
Figure 2. Example of four pictures
represented in the FR task. Reprinted from “Facial identity and facial emotions: Speed, accuracy, and processing strategies in
children and adults”, by L.M.J De Sonneville et al., 2002, Journal of Clinical and
Experimental Neuropsychology, 24, p. 203.
Emotion recognition: To examine the emotion recognition skills of the
children, the subtest ‘Identification of Facial Expressions’ (IFE) of the computerized ANT2.1 (De Sonneville, 2005) was used. In this subtest, children were asked whether a face showed a specific emotion. A picture of a face, that showed happiness, anger, fear, contempt, disgust, surprise, sadness or shame (see Figure 3) was presented and participants answered whether the face showed that specific emotion by pressing the ‘yesʼ- or the ‘no’-key. Participants first completed 40 trials where the target emotion was happiness, of which 20 faces showed happiness and 20 did not. Subsequently, participants completed 40 trials where the target emotion was anger, of which 20 faces showed anger and 20 did not. Reaction time and the number of hits, correct rejections, misses and false alarms were registered for happiness and anger separately.
Figure 3. Example of pictures presented in
the IFE task, showing from left to right sadness, anger and happiness. Reprinted from “Facial identity and facial emotions: Speed, accuracy, and processing strategies in children and adults”, by L.M.J De Sonneville et al., 2002, Journal of Clinical and
Experimental Neuropsychology, 24, p. 203.
Data analysis
Since not all participants completed all tests at all three measurement occasions, some data were missing. To improve power, the data imputation method Expectation Maximization (EM) was used in SPSS. Little’s MCAR test showed that data were not missing at random for the variables FR false alarms (χ2
(8) = 16.503, p = .036), IFE
happy reaction time (χ2
(8) = 15.560, p = .049) and IFE angry false alarms (χ2(9) = 21.292, p = .011), so for these variables data could not be imputed. For the other
variables, data was predicted using scores at pre-intervention, post-intervention and 2-month follow-up. With EM estimated data on BS, FR reaction time, FR misses, IFE happy misses and false alarms, IFE angry reaction time and misses were added to the dataset.
To form a level of accuracy on face recognition and emotion recognition, total number of errors was calculated by adding false alarms and misses for each variable at each measurement occasion separately. Inspection of the data showed 19 univariate outliers (z > 3.29); one on pre-intervention and one on post-intervention BS, one on follow-up FR misses, one on
pre-intervention and one on follow-up FR false alarms, one on pre-intervention and one on up IFE happy misses, one on follow-up IFE happy false alarms, one on pre-intervention IFE happy reaction time, one on pre-intervention and one on follow-up IFE happy misses, one on follow-up IFE happy false alarms, one on post-intervention and one on follow-up IFE happy errors, one on pre-intervention, post-intervention and FU of IFE angry reaction time, one on
pre-intervention and post-pre-intervention IFE angry misses, one on post-intervention and follow-up of IFE angry false alarms and one on IFE angry errors. Outliers were replaced by z 3.29.
To find out whether the mindfulness training had an effect on the face and
emotion recognition of children with ASD, a repeated measurement ANOVA was
conducted for three moments of
measurement: the pre-intervention, post-intervention and 2-month follow-up measurements. Thirteen separate analyses
were performed, with the dependent variable 1) BS, 2) FR reaction time, 3) FR errors, 4) FR misses, 5) FR false alarms, 6) IFE happy reaction time, 7) IFE happy errors, 8) IFE happy misses, 9) IFE happy false alarms, 10) IFE angry reaction time, 11) IFE angry errors, 12) IFE angry misses and 13) IFE angry false alarms. The independent variable was time. A p-value ≤ .05 was taken as statistically significant and a p-value
between .05 and .1 as borderline significant, to prevent dismissing results that seem clinically meaningful (Hackshaw &
Kirkwood, 2011). Assumptions of normality and sphericity were tested. For the variables BS, FR misses, false alarms and errors, IFE happy misses, false alarms and errors and IFE angry misses, false alarms and reaction time data were not normally distributed at pre-intervention, post-intervention and 2-month follow-up. For IFE angry errors the pre-intervention and post-intervention data distribution was not normal. This means that the normality assumption of the repeated measurement ANOVA was violated. However, if the sample is large enough, the test is robust to violation of this assumption. (Agresti & Franklin, 2012). Mauchly’s test for sphericity was nonsignificant, except for FR false alarms (χ2 = 7.03, p = .030), IFE happy false alarms (χ2 = 6.53, p = .038), IFE happy errors (χ2 = 9.81, p = .007), IFE angry reaction time (χ2
= 21.92, p < .001) and IFE angry misses (χ2 = 21.21, p < .001). Huynh-Feldt correction was used for FR false alarms (ε = .838), IFE happy false alarms (ε = .913) and IFE happy errors (ε = .864).
Greenhouse-Geiser correction was used for IFE angry reaction time (ε = .722) and IFE angry misses (ε = .727).
Results
Results of the repeated measures ANOVA are presented in Table 1. As effect
size, partial η2 is reported – in social/clinical areas of psychology .01 is considered small, .06 medium and .14 large (Cohen, 1988). Baseline speed
The baseline speed of the participants did not change from pre-intervention to post-intervention or follow-up (F(2, 92) = 2.167,
p = .120).
Face recognition
The repeated measures ANOVA showed a small, borderline significant decrease of the reaction time for face recognition from pre-intervention to post-intervention and follow-up (F(2, 92) = 2.423,
p = .094). No significant difference was
found between the total number of errors participants made at pre-intervention, post-intervention and follow-up (F(2, 48) = .493,
p = .614) in the recognition of faces, neither
for misses (F(2, 92) = 0.029, p = .972) and false alarms (F(1.88, 40.23) = 3.642, p = .388) separately.
Emotion recognition – happiness The repeated measures ANOVA showed a significant effect of time on the speed of recognizing a happy facial
expression (F(2, 46) = 5.147, p = .01) with partial η2=.18, which is considered a large effect (Cohen, 1988). Post-hoc test with Bonferroni correction revealed that participants were significantly faster in recognizing a happy facial expression at follow-up than at pre-intervention (p = .011), but no significant difference was found from pre-intervention to post-intervention (p = .504) or from post-intervention to follow-up (p = .224). The ANOVA with Huynh-Feldt correction (ε = .864) revealed no significant change from pre-intervention to post- intervention or follow-up (F(1.73, 79.50) = 1.416, p = .248) in the number of errors participants made in recognizing happy facial expressions. However, number of false
Table 1.
Repeated Measures ANOVA on BS, FR and IFE.
Note. *p < .1; a Huynh-Feldt correction (ε > .75) was used; b Greenhouse-Geiser correction (ε < .75) was used. Reaction time is in milliseconds.
alarms taken separately, the effect of time appeared to be significant (F(1.83, 84.03) = 3.642, p = .035), with a medium effect size (η2 = .07). Participants’ false alarms
decreased from pre-intervention to post-intervention with borderline significance (p = .09) and significantly increased from post-intervention to follow-up (p = .02). No
significant effect was found from pre-intervention to follow-up (p = 1.00). The number of misses that participants made did not change significantly (F(2, 92) = 0.581, p = .561) from pre-intervention to
post-intervention or follow-up. pre-intervention post-intervention 2-month
follow-up Variable M SD M SD M SD N F df (time, error) p Partial η2 BS 314.73 47.73 313.30 74.24 330.74 71.97 47 2.167 2, 92 .120 .05 Face Recognition FR reaction time* 1224.56 288.56 1220.02 349.81 1151.14 296.30 47 2.423 2, 92 .094* .05 FR errors 3.48 2.82 4.00 3.63 4.04 3.63 25 0.493 2, 48 .614 .02 FR misses 2.83 2.33 2.75 2.30 2.78 2.19 47 0.029 2, 92 .972 .00 FR false alarmsa 1.24 1.05 1.36 1.66 0.98 1.53 25 0.931a 1.88, 40.23 .388 .04
Emotion recognition - happiness
IFE happy reaction
time* 616.04 124.83 581.08 91.53 546.17 110.32 24 5.147 2, 46 .010* .18
IFE happy errorsa 1.88 1.69 1.49 1.54 1.85 2.07 47 1.416a 1.73,
79.50 .248 .03 IFE happy misses 0.73 0.89 0.83 0.97 0.68 0.74 47 0.581 2, 92 .561 .01 IFE happy false alarmsa* 1.14 1.11 0.70 1.06 1.19 1.69 47 3.624 1.83, 84.03 .035* .07
Emotion recognition – anger
IFE angry reaction timeb*
863.78 231.52 869.15 254.96 793.87 153.79 47 3.401b 1.44, 66.40
.054* .07
IFE angry mistakes 6.36 5.59 4.23 2.81 4.64 4.18 22 2.236 2, 42 .119 .10
IFE angry missesb 2.46 3.15 2.42 2.77 1.82 2.03 47 1.247b 1.45, 66.87 .285 .03 IFE angry false alarms 3.72 4.68 2.05 1.81 2.89 3.12 22 2.035 2, 42 .143 .09
Emotion recognition – anger
A borderline significant effect was found of time on the speed of recognizing an angry facial expression (F(1.44, 66.40) = 5.147, p = .054). Post-hoc test with Bonferroni correction showed that
participants became significantly faster in recognizing an angry face from
pre-intervention to follow-up (p = .01), but not from pre-intervention to post-intervention (p = 1.00) or post-intervention to follow-up (p = .208). The number of errors participants made decreased from pre-intervention to post-intervention and from pre-intervention to follow-up, but the decrease was not statistically significant (F(2, 42) = 2.236, p = .119). No effect was found on misses
(F(1.45, 66.87) = 1.247, p = .285) and false alarms (F(2, 42) = 2.035, p = .143)
separately either.
Discussion
The aim of the current study was to examine the effect of a MBP on face and emotion recognition in children and
adolescents with ASD. The results show that children were faster in recognizing faces after the MBP (with borderline significance), whereas the accuracy remained stable. Children were also faster in recognizing expressions of happiness and anger (with borderline significance) two months after the MBP, but not directly after the training. Similarly to the recognition of faces, accuracy in the recognition of emotions remained unchanged. Therefore, the hypothesis that children with ASD would become more accurate and faster in recognizing faces and emotions is partly confirmed: children became faster, but did not make fewer errors. Results will be discussed in detail below.
Face recognition
For face recognition children became faster, while making the same number of errors. The increased speed in face
recognition after the MBP could indicate that children process faces in a more holistic manner after the mindfulness training. This is a faster strategy to recognize faces than a feature based strategy. However, the current study did not examine which face
recognition strategies children used and can therefore not determine if children used a more holistic face processing strategy after the MBP.
Possibly, children did not improve in their accuracy, since they already made few errors before the training, leaving not much room for improvement. In accordance with this idea, Wimmer et al. (2016) found that 10-year olds only improved their face recognition accuracy after a MBSR training on a hard face recognition task and not on an easy face recognition task. In the hard face recognition task, the faces that were not the same as the target face looked more similar to the target face than in the easy task. Possibly the current study measured no change in accuracy due to a ceiling effect of the used instrument.
Emotion recognition
With respect to facial emotion
recognition, children were faster two months after the training and the number of errors that participants made, did not change. It was expected that the MBP would improve children’s theory of mind, which would make them better in recognizing emotions. This hypothesis was not confirmed by the results. The current results are in contrast with findings of English et al. (2018), who found a positive relation between trait mindfulness and general ability to recognize emotions in female undergraduate students. The results are in line however, with findings
from Melloni et al. (2013), who found no difference in the emotion recognition abilities of adults who participated in an MBSR training and adults who did not participate. However, Melloni et al. (2013) found no improvement in the time
participants needed to recognize a facial expression, but in the current study the children became faster.
Interestingly, children became faster in emotion recognition two months after the MBP, but not directly after the MBP. After the last training session, children were encouraged to keep practicing mindfulness and they made a personal meditation plan. Perhaps these two months of extra practice were necessary to develop an effect on emotion recognition.
Some differences were found
between the recognition of happy and angry expressions. Participants were slower and made more errors in the recognition of angry expressions than when recognizing happy expressions. This is in line with findings from De Sonneville et al. (2002), who measured emotion recognition with the IFE task of the ANT in typically developing children. They found that children were less accurate and slower when recognizing angry faces than when recognizing happy faces. Similar results were found by Greimel et al. (2014) who measured emotion recognition in adolescents and adults with ASD. In the IFE task of the ANT, happy faces all have an open mouth with the teeth visible. This might make it relatively easy to identify the happy faces. In addition, eye-tracking studies found that individuals with ASD focus mainly on the mouth when looking at faces (Gross, 2008). Since the happy faces might be more easily recognized by the mouth than the angry faces, focusing on the mouth might be more effective when recognizing
happiness than anger.
For the recognition of happy
expressions, children made significantly less false alarms directly after de MBP; they less often answered that a face showed a happy expression, when it did not. This corresponds with a less impulsive response style (Van de Weijer-Bergsma, Formsma, De Bruin, & Bögels, 2012). Mindfulness training is found to reduce impulsivity scores of children with Attention Deficit Hyperactivity Disorder (ADHD; Van der Oord, Bögels, & Peijnenburg, 2012) and of children with ASD and externalizing symptoms (Bögels, Hoogstad, Van Dun, Schutter, & Restifo, 2008). Possibly the impulsivity of the participants in this study also decreased. Reacting less impulsively, children would less often directly decide that someone shows a happy facial expression, when the person does not. However, the decrease of false alarms was only found for the
recognition of happiness, not for the recognition of faces or angry expressions, and was not observed anymore eight weeks after the training.
Limitations, strengths and future research recommendations
The current study had some strong points:objective neuropsychological measures for face and emotion recognition were used and a follow-up measurement was included. The treatment integrity of the MBP was evaluated and competence and
adherence scores were high. However, some limitations should be taken into
consideration. The number of participants was relatively small, limiting the power of the statistical tests.A control group, that did not receive the MBP, was not included. Therefore it is not possible to rule out the influence of maturation (Hoyle, Harris, & Judd, 2002); children might have become faster in face and emotion recognition
because they grew older and naturally
developed their face and emotion recognition skills. It was intended to measure face and emotion recognition abilities of the children while they were on the waiting list, so that could be noted whether children’s face and emotion recognition abilities changed during the waiting list period, but only 12
participants took part in this measurement and therefore this measurement was not included in the analysis. Another limitation is that the children participating in the study were relatively high functioning children with ASD, attending regular education (90.24%) and having average to high levels of intelligence (mean IQ = 111.00). Children with ASD of average intelligence are found to score better on the face recognition and identification of emotion tasks than children with an intelligence below average
(Rommelse et al., 2015). This makes the results less generalizable to children with ASD with lower intelligence or attending special education. It is recommended that future research not only focuses on if face and emotion recognition are affected by mindfulness training, but also how. Looking at the process of face recognition and whether children process faces and facial expressions more holistically after
mindfulness training, will give more insight into the mechanisms of change.
Conclusion
In conclusion, the current study showed that after the MYmind mindfulness training, children and adolescents with ASD became faster in recognizing faces and emotions, whereas the accuracy (number of errors) remained unchanged.
References
Agresti, A., & Franklin, C. (2012). Statistics: The art and science of learning from data (3rd ed.). Boston, MA: Pearson Education.
American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). Washington, DC: Author. Annaz, D., Karmiloff-Smith, A., Johnson, M. H., &
Thomas, M. S. (2009). A cross-syndrome study of the development of holistic face recognition in children with autism, Down syndrome, and Williams syndrome. Journal of Experimental Child Psychology, 102, 456-486. doi:10.1016/j.jecp.2008.11.005 Baron-Cohen, S. (2000). Theory of mind and autism:
A review. International Review of Research in Mental Retardation: Autism, 23, 169 –184. doi:10.1016/S0074-7750(00)80010-5 Baron-Cohen, S., Leslie, A. M., & Frith, U. (1985).
Does the autistic child have a “theory of mind”?. Cognition, 21, 37-46.
doi:10.1016/0010-0277(85)90022-8
Block‐Lerner, J., Adair, C., Plumb, J. C., Rhatigan, D. L., & Orsillo, S. M. (2007). The case for mindfulness‐based approaches in the
cultivation of empathy: Does nonjudgmental, present‐moment awareness increase capacity for perspective‐taking and empathic
concern?. Journal of Marital and Family Therapy, 33, 501-516.
doi:10.1111/j.1752-0606.2007.00034.x Bögels, S., Hoogstad, B., Van Dun, L., De Schutter,
S., & Restifo, K. (2008). Mindfulness training for adolescents with externalizing disorders and their parents. Behavioural and Cognitive Psychotherapy, 36, 193-209. doi:10.1017/S1352465808004190 Cohen, J. (1988). Statistical power analysis for the
behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum Associates
De Bruin, E. I., Blom, R., Smit, F. M., van Steensel, F. J., & Bögels, S. M. (2015). MYmind: Mindfulness training for youngsters with autism spectrum disorders and their parents. Autism, 19, 906-914. doi:10.1177/1362361314553279 De Sonneville, L. M. J. (2005). Amsterdamse
Neuropsychologische Taken:
wetenschappelijke en klinische toepassingen [Amsterdam Neuropsychological Tasks: Scientific and clinical applications].
De Sonneville, L. M. J., Verschoor, C. A.,
Njiokiktjien, C., Op het Veld, V., Toorenaar, N., & Vranken, M. (2002). Facial identity and facial emotions: speed, accuracy, and processing strategies in children and adults. Journal of Clinical and Experimental Neuropsychology, 24, 200-213.
doi:10.1076/jcen.24.2.200.989
Dimitriou, D., Leonard, H. C., Karmiloff‐Smith, A., Johnson, M. H., & Thomas, M. S. (2015). Atypical development of configural face recognition in children with autism, Down syndrome and Williams syndrome. Journal of Intellectual Disability Research, 59, 422-438. doi:10.1111/jir.12141
Durand, K., Gallay, M., Seigneuric, A., Robichon, F., & Baudouin, J. Y. (2007). The development of facial emotion recognition: the role of configural information. Journal of
Experimental Child Psychology, 97, 14-27. doi:10.1016/j.jecp.2006.12.001
English, L. H., Wisener, M., & Bailey, H. N. (2018). Childhood emotional maltreatment, anxiety, attachment, and mindfulness: associations with facial emotion recognition. Child Abuse and Neglect, 80, 146-160.
doi:10.1016/j.chiabu.2018.02.006
Freire, A., & Lee, K. (2001). Face recognition in 4-to 7-year-olds: processing of configural, featural, and paraphernalia
information. Journal of Experimental Child Psychology, 80, 347-371.
doi:10.1006/jecp.2001.2639
Gao, X., & Maurer, D. (2010). A happy story: developmental changes in children’s sensitivity to facial expressions of varying intensities. Journal of Experimental Child Psychology, 107, 67-86.
doi:10.1016/j.jecp.2010.05.003 Goldberg, S. B., Tucker, R. P., Greene, P. A.,
Davidson, R. J., Wampold, B. E., Kearney, D. J., & Simpson, T. L. (2018). Mindfulness-based interventions for psychiatric disorders: a systematic review and
meta-analysis. Clinical Psychology Review, 59, 52-60. doi:10.1016/j.cpr.2017.10.011
Greimel, E., Schulte-Rüther, M., Kamp-Becker, I., Remschmidt, H., Herpertz-Dahlmann, B., & Konrad, K. (2014). Impairment in face processing in autism spectrum disorder: a developmental perspective. Journal of Neural Transmission, 121, 1171-1181. doi:10.1007/s00702-014-1206-2
Gross, T. F. (2008). Recognition of immaturity and emotional expressions in blended faces by children with autism and other
developmental disabilities. Journal of Autism and Developmental Disorders, 38, 297-311. doi:10.1007/s10803-007-0391-3
Hackshaw, A., & Kirkwood, A. (2011). Interpreting and reporting clinical trials with results of borderline significance. British Medical Journal, 343,3340–3344.
doi:10.1136/bmj.d3340
Happé, F., & Frith, U. (2006). The weak coherence account: detail-focused cognitive style in autism spectrum disorders. Journal of Autism and Developmental Disorders, 36, 5-25. doi:10.1007/s10803-005-0039-0
Harms, M. B., Martin, A., & Wallace, G. L. (2010). Facial emotion recognition in autism spectrum disorders: a review of behavioral and neuroimaging studies. Neuropsychology Review, 20, 290-322.
doi:10.1007/s11065-010-9138-6 Hoyle, R. H., Harris, M. J., & Judd, C. M.
(2002). Research methods in social relations. Wadsworth: Cengage Learning
Kabat-Zinn, J. (1982). An outpatient program in behavioral medicine for chronic pain patients based on the practice of mindfulness
meditation: theoretical considerations and preliminary results. General Hospital Psychiatry, 4, 33-47.
doi:10.1016/0163-8343(82)90026-3 Kabat-Zinn, J. (1990). Full catastrophe living: Using
the wisdom of your body and mind to face stress, pain and illness. New York: Delacorte Kabat-Zinn, J. (2001) Mindfulness meditation for
everyday life. London: Piatkus Books. Kim, E., Kyeong, S., Cheon, K. A., Park, B., Oh, M.
K., Chun, J. W., ... & Song, D. H. (2016). Neural responses to affective and cognitive theory of mind in children and adolescents with autism spectrum disorder. Neuroscience Letters, 621, 117-125.
doi:10.1016/j.neulet.2016.04.026
Kinnunen, S., Korkman, M., Laasonen, M., & Lahti-Nuuttila, P. (2013). Development of face recognition in 5-to 15-year-olds. Journal of Cognition and Development, 14, 617-632. doi:10.1080/15248372.2012.689389 Lawrence, K., Campbell, R., & Skuse, D. (2015).
Age, gender, and puberty influence the development of facial emotion
recognition. Frontiers in Psychology, 6. doi:10.3389/fpsyg.2015.00761
Lee, S. B., Koo, S. J., Song, Y. Y., Lee, M. K., Jeong, Y. J., Kwon, C., ... & An, S. K. (2014). Theory of mind as a mediator of reasoning and facial emotion recognition: findings from 200 healthy people. Psychiatry
Investigation, 11, 105-111. doi:10.4306/pi.2014.11.2.105
Loth, E., Gómez, J. C., & Happé, F. (2008). Event schemas in autism spectrum disorders: the role of theory of mind and weak central coherence. Journal of Autism and Developmental Disorders, 38, 449-463. doi:10.1007/s10803-007-0412-2 Loukusa, S., Mäkinen, L., Kuusikko‐Gauffin, S.,
Ebeling, H., & Moilanen, I. (2014). Theory of mind and emotion recognition skills in children with specific language impairment, autism spectrum disorder and typical development: group differences and connection to knowledge of grammatical morphology, word‐finding abilities and verbal working memory. International Journal of Language and Communication Disorders, 49, 498-507.
doi:10.1111/1460-6984.12091
Meinhardt-Injac, B., Daum, M. M., Meinhardt, G., & Persike, M. (2018). The two-systems account of theory of mind: testing the links to social-perceptual and cognitive abilities. Frontiers in Human Neuroscience, 12.
doi:10.3389/fnhum.2018.00025
Melloni, M., Sedeño, L., Couto, B., Reynoso, M., Gelormini, C., Favaloro, R., ... & Ibanez, A. (2013). Preliminary evidence about the effects of meditation on interoceptive sensitivity and social cognition. Behavioral and Brain Functions, 9(47), 1-6.
doi:10.1186/1744-9081-9-47
Mondloch, C. J., Geldart, S., Maurer, D., & Le Grand, R. (2003). Developmental changes in face processing skills. Journal of Experimental Child Psychology, 86, 67-84.
doi:10.1016/S0022-0965(03)00102-4 Petrakova, A., Sommer, W., Junge, M., &
Hildebrandt, A. (2018). Configural face perception in childhood and adolescence: an individual differences approach. Acta Psychologica, 188, 148-176.
doi:10.1016/j.actpsy.2018.06.005
Ridderinkhof, A., de Bruin, E. I., Blom, R., & Bögels, S. M. (2018). Mindfulness-based program for
children with autism spectrum disorder and their parents: direct and long-term
improvements. Mindfulness, 9, 773–791. doi:10.1007/s12671-017-0815-x
Rommelse, N., Langerak, I., van der Meer, J., de Bruijn, Y., Staal, W., Oerlemans, A., & Buitelaar, J. (2015). Intelligence may moderate the cognitive profile of patients with ASD. PLoS One, 10, 1-17.
doi:10.1371/journal.pone.0138698
Scherf, S. K., Behrmann, M., Minshew, N., & Luna, B. (2008). Atypical development of face and greeble recognition in autism. Journal of Child Psychology and Psychiatry, 49, 838-847. doi:10.1111/j.1469-7610.2008.01903.x Segal, Z. V., Williams, M. G., & Teasdale, J. D.
(2013). Mindfulness-based cognitive therapy for depression. New York: The Guilford Press
Serra, M., Althaus, M., De Sonneville, L. M. J., Stant, A. D., Jackson, A. E., & Minderaa, R. B. (2003). Face recognition in children with a pervasive developmental disorder not otherwise specified. Journal of Autism and Developmental Disorders, 33, 303-317. doi:10.1023/A:1024458618172
Tan, L. B., Lo, B. C., & Macrae, C. N. (2014). Brief mindfulness meditation improves mental state attribution and empathizing. PLoS One, 9, 1-5.
doi:10.1371/journal.pone.0110510 Trevisan, D. A., & Birmingham, E. (2016). Are
emotion recognition abilities related to everyday social functioning in ASD? A meta-analysis. Research in Autism Spectrum Disorders, 32, 24-42.
doi:10.1016/j.rasd.2016.08.004
Uljarevic, M., & Hamilton, A. (2013). Recognition of emotions in autism: a formal
meta-analysis. Journal of Autism and
Developmental Disorders, 43, 1517-1526. doi:10.1007/s10803-012-1695-5
Van de Weijer-Bergsma, E., Formsma, A. R., de Bruin, E. I., & Bögels, S. M. (2012). The effectiveness of mindfulness training on behavioral problems and attentional functioning in adolescents with ADHD. Journal of Child and Family Studies, 21, 775-787.
doi:10.1007/s10826-011-9531-7
Van der Oord, S., Bögels, S. M., & Peijnenburg, D. (2012). The effectiveness of mindfulness training for children with ADHD and
mindful parenting for their parents. Journal of Child and Family Studies, 21, 139-147. doi:10.1007/s10826-011-9457-0
Walker-Andrews, A. S., Krogh-Jespersen, S., Mayhew, E. M., & Coffield, C. N. (2011). Young infants' generalization of emotional expressions: effects of
familiarity. Emotion, 11, 842–851. doi:10.1037/a0024435
Weigelt, S., Koldewyn, K., & Kanwisher, N. (2012). Face identity recognition in autism spectrum disorders: a review of behavioral
studies. Neuroscience and Biobehavioral Reviews, 36, 1060-1084.
doi:10.1016/j.neubiorev.2011.12.008 Williams, J. M. G., & Kabat-Zinn, J. (2011).
Mindfulness: diverse perspectives on its meaning, origins, and multiple applications at the intersection of science and
dharma. Contemporary Buddhism, 12, 1-18. doi:10.1080/14639947.2011.564811 Wimmer, L., Bellingrath, S., & von Stockhausen, L.
(2016). Cognitive effects of mindfulness training: results of a pilot study based on a theory driven approach. Frontiers in Psychology, 7.
doi:10.3389/fpsyg.2016.01037
Wolf, J. M., Tanaka, J. W., Klaiman, C., Cockburn, J., Herlihy, L., Brown, C., ... & Schultz, R. T. (2008). Specific impairment of face‐ processing abilities in children with autism spectrum disorder using the Let's Face It! skills battery. Autism Research, 1, 329-340. doi:10.1002/aur.56.
