Faculty of Social and Behavioural Sciences
Graduate School of Childhood Development and Education
Triadic Emotional Exchanges:
Mother-Infant-Person Facial Expressions and Maternal Depression
Research Master Educational Sciences Research Master Thesis
Student: Pigi Panagiotopoulou
Supervisors: Dr. Evin Aktar, Dr. Cristina Colonnesi Reviewer 1: Dr. Eline Möller
Reviewer 2: Drs. Milica Nikolic 05-08-2016
Abstract
Maternal depressive symptoms have been found to diminish mother’s abilities to respond to her infant’s emotional needs and to be associated with alterations in mother-child emotional communication. This limited maternal responsiveness may extend to social referencing contexts. The present study
investigates the relation between maternal depressive symptoms and shared positive and negative affect of infant facial emotional expressions, during mother-infant-person triadic interactions. A community sample of 49 mothers with their 12-month infants was observed during computerized presentations of positive vs. negative facials expressions on the screen. The study also explored whether this shared positive and negative affect changes between positive and negative facial expressions on the screen. Findings revealed that positive expressions on screen were associated with longer durations of positive and negative shared affect compared to the negative expression on the screen. In turn, maternal depressive symptoms and the interaction between screen and maternal depressive symptoms did not significantly predicted mother-infant shared affect.
Keywords: triadic interaction, shared affect, maternal depressive symptoms,
facial expressions
Triadic Emotional Exchanges:
Mother-Infant-Person Facial Expressions and Maternal Depression
Immediately after they are born, infants show an orientation bias at face patterns compared to other stimuli (Goren, Sarty, & Wu, 1975). Infants’ experiences with faces are essential in order for them to gain the ability to understand and process facial emotional expressions (Pascalis & Kelly, 2009). It has been found that maternal emotional disposition is associated with infants’ processing of facial expressions (De Haan, Belsky, Reid, Volein & Johnson, 2004). More specifically, evidence shows that maternal depression is related to infants’ attention to emotional faces (Hernandez-Reif, Field, Diego, Vera, & Pickens, 2006), and that it interferes with mother’s ability to emotionally synchronize and positively interact with her infant (Beck, 1995). According to Feinman (1982),infants start to use other’s emotional reactions to guide them when confronting a novel stimulus, like a stranger or an unfamiliar situation at the end of the first year of life. Considering this, it is important to further explore the relationship between maternal depressive symptoms and infants’ reactions to facial emotional expressions. To that end, the present study investigates the relation between maternal depressive symptoms and mother-infant simultaneous positive and negative facial expressions, during the presentation of positive (i.e. happy) and negative (i.e. angry) facial emotional expressions on the screen in mother-infant-person triadic interactions.
Infants and face processing
Infants begin to understand the differences in emotions from a very young age. Even from 10 weeks, infants are able to discriminate between their mothers' positive and negative facial expressions (Haviland & Lelwica, 1987). In addition, at the age of 3 months, they are able to discriminate and recognize happy and angry
expressions, posed not only by their mother but also by a stranger (Barrera & Maurer, 1981). In order to further explore infants’ perception of facial expressions, Barbera, Izard, Vietze and Parisi (1976) investigated the attention of twelve 4- and 6-month-old infants towards happy, anger and neutral facial expressions. They found that infants at both ages looked longer at the happy expressions than the other two conditions. Moreover, an event-related potential (ERP) study measured the brain responses of forty 7- and 12- month-old infants to investigate the developmental changes when processing happy and angry facial expressions. It was found that in both age groups infants gazed significantly longer at the happy faces compared to the angry ones (Grossmann, Striano & Friederici, 2007). These findings suggest that infants are preferably oriented to looking at happy faces and they maintain their choice across time. This preference for happy faces can be explained by the fact that angry facial expressions are considered to be a threatening stimulus that facilitates avoidance behavior both in infants (Hunnius, De Wit, Vrins & Von Hofsten, 2011) and in adults (Marsh, Ambady, & Kleck, 2005).
Studies, however, reveal that infants may not always have a preference in positive faces, as they grow older. A negativity bias in their gaze was evident in 7-month-old infants, when they were looking at positive (i.e. happy) and negative (i.e. fear) facial expressions (Kotsoni, De Haan & Johnson, 2001). One of these studies investigated maternal affective personality and infants responsiveness to happy and fearful facial expressions in a sample of sixty-nine 7-month-old infants. What was found is that infants of mothers with high positive affect gazed longer at fearful than happy faces, whereas infants of mothers with lower positive affect did not (De Haan, Belsky, Reid, Volein & Johnson, 2004). These findings suggest that infants have a preference for negative facial expressions, which can be explained by the fact that
infants are oriented to gaze longer at faces that seem more novel to them.
Nevertheless, these studies only examined fearful expressions and since fear –unlike anger– is not always considered to be a threatening stimulus (Marsh, Ambady, & Kleck, 2005), further investigation is needed on this field.
In sum, what is illustrated by these earlier findings is that infants’
competence in processing emotional faces evolves rapidly, as they are increasingly exposed to faces in daily life (Pascalis & Kelly, 2009). The more the infants are being exposed to emotional faces through face-to-face relationships, the more they learn how to develop their own emotional skills (Campos, Campos & Barrett, 1989; Markova & Legerstee, 2006). Still, although several studies have so far examined infants’ gaze during positive and negative faces, less is known about mother-infant emotional expressions, during positive and negative faces. On account of this, the present study explores mother-infant simultaneous positive and negative facial expressions, during happy and angry facial expressions on the screen. Considering that infants begin to emotionally communicate with others in face-to-face interactions with their caregiver (Tronich, 1989), it is necessary to present what previous studies have revealed about mother-infant emotional communication in dyadic interactions.
Dyadic interactions
Infant’s primary social interactions are established through face-to-face interactions called protoconversations. They involve gaze, touch, smiles and
vocalizations between the infant and the adult, which maintain their operation thanks to the emotional exchanges that take place between the two parties (Trevarthen, 1979). Due to the importance of these emotional exchanges, several studies have investigated mother-infant face-to-face interaction. It has been found that mothers’ facial emotional expressions are more positive (vs. neutral) and continue to be so, as
long as the infant remains positive and gazes at them (Cohn & Tronick, 1987; Kaye & Fogel, 1990). Similarly, infants coordinate their gaze with positive facial expression more often when interacting with the mother than with the father (Colonnesi, Zijlstra, Van der Zande, & Bögels, 2012) and they tend to become positive when their mother is positive (Cohn & Tronick, 1987).
This shared positive affect is accomplished via a bidirectional effect; where mother-infant adjust their emotional expressions and affected behavior according to each other (Tronick, 1989). This positive emotional exchange seems to be beneficial for the development of infants’ emotional skills (Markova & Legerstee, 2006).
Feldman (2007) investigated the developmental outcomes of mother-infant synchrony and found that it predicts self-regulation at infancy and childhood as well as empathy in adolescence. These findings relate shared positive affect to the present and future psychological well-being of the child and therefore, demonstrate its importance in infant’s socio-emotional development. Nonetheless, due to the fact that women may face an increased risk of developing symptoms of depression at the early years of motherhood (Gavin et al., 2005; O'hara & Swain, 1996), it is necessary to investigate mother-infant emotional communication and synchrony between depressed mothers and their infants. For this reason, in the next section we discuss how maternal
depressive symptoms are associated with mother-infant emotional communication and shared affect in dyadic interactions.
Dyadic interactions and maternal depression
Depressive symptoms refer to feelings of sadness, loneliness, emotional unavailability, withdrawal, fatigue, difficulty in concentrating, suicidal thoughts and reduced interest in daily activities (American Psychiatric Association [APA], 2013). Depression in mothers has been found to be associated with maternal avoidance
(Groh, Roisman, Ijzendoorn, Bakermans-Kranenburg & Fearon, 2012) and low warm-engaged parenting (Bayer, Sanson & Hemphill, 2006). Additionally, research shows that maternal depression is also linked to reduced emotional reactivity (Rottenberg, 2005), diminished maternal responsiveness towards her infant’s vocalizations (Bettes, 1988) and reduced sensitivity to infant’s distress (Donovan, Leavitt, & Walsh, 1998).
Maternal depressive symptoms have also been found to interfere with the mother’s ability to respond to her infant’s emotional expressions. In particular,
mother-infant synchrony of positive affect occurs less among mothers with depression compared to non-depressed (Feldman, 2003). Furthermore, evidence reveals that depressive symptoms in mothers also diminish their ability to positively interact with their infant. In a sample of 101 infants, aged from 3.5- to 5.5-month-old, Aktar et al. (in press) investigated parental depression and anxiety in relation to parent-infant face-to-face interaction. They found that parents with depressive symptoms tend to be less positive and more neutral with their infant, compared to the non-depressed. Campbell, Cohn & Meyers (1995) explored mother-infant positive and negative interaction at 2, 4, and 6 months in home-observations during three tasks: feeding, face-to-face interaction and toy play. It was observed that mothers with longer
durations of postnatal depression had more pronounced effects on their positive mood than those who were depressed for a shorter period. Finally, Cohn, Campbell, Matias, & Hopkins (1990) based in home-observations at the first trimester of the infancy, reported that mothers with depression were not only less positive, but also more negative than the reference mothers, during face-to-face interactions. All these findings demonstrate that maternal depression diminishes mother’s ability to
Although, the way that maternal depression is related to mother-infant expressions of affect has been predominantly investigated in dyadic interaction, there are very few studies that address this issue in triadic interactions. In the following section, first we present the underline processes and basic elements of triadic interactions in relation to existing data. Then we discuss the available evidence of maternal depression in triadic contexts.
Triadic interactions
As infants become gradually more engaged in interactions that involve third parties, shared affect can take the form of sharing goals with regard to an external entity (Tomasello, Carpenter, Call, Behne & Moll, 2005). Another adult interacting with the mother, a toy or a representational activity depict examples of triadic interactions (Hay, 1979; Hay & Murray, 1982). In such triadic contexts, infants coordinate their gaze with the other partner’s attention towards the third object or activity to share their awareness. This ability is called joint attention (Dunham & Moore, 1995) and it emerges between the age of 6 and 12 months (Corkum & Moore, 1998). Research has found that joint attention skills are correlated with emotion regulation (Morales, Mundy, Crowson, Neal, & Delgado, 2005), implying that their acquisition can be important for the development of infants’ emotional skills and responses.
However, even when infants have acquired joint attention skills, they have yet to learn how to use other’s signals when encountering a novel experience. When infants are between 10 and 14 months, they manage to observe and adopt other’s emotional responses and signals to determine how to react towards an unfamiliar or unexpected event. This is referred to as social referencing (Feinman, 1982) and it can also be considered as a process, where the infants begin to shape their own individual
perception of the new stimuli, based on people’s reactions and interpretations (Gewirtz & Peláez-Nogueras, 1992). Interaction with a stranger (Feinman & Lewis, 1985), toys (Mumme, Fernald & Herrera, 1996) or new situations, like crossing a visual cliff (Sorce, Emde, Campos & Klinnert, 1985) constitute the main sources of ambiguity in social referencing situations. In such cases, the infants actively seek emotional cues and signals from the parent in order to overcome the uncertainty they experience (Klinnert, Campos, Sorce, Emde & Svejda, 1983). Taking into
consideration that infants are expected to have acquired this skill by the age of 12 months, the present study explores mother-infant-person triadic interaction in 12-month-old infants by observing mother-infant facial emotional exchanges, during computerized strangers’ positive and negative facial expressions.
Triadic interactions and maternal depression
Even though, social referencing offers the infant a unique opportunity to learn how to react to unknown situations or strangers, its acquisition can be impeded in cases of infants with depressed mothers. Because of maternal emotional
unavailability, mothers with depressive symptoms are more likely to exchange less information with their infant, when s/he is searching for reference signals to overcome uncertainty. To test this theory, Stenberg (2003) examined infant’s gaze in a mother-infant-toy triadic interaction. During the experiment, ninety-six 12-month-old infants were presented with an ambiguous or an unambiguous toy and were randomly assigned to one of four conditions: mother inattentive, mother having positive facial expression and behavior, mother having negative facial expression and behavior and a control condition. It was found that infants whose mothers were limited in exchanging information (mother inattentive) expressed less interest in toy exploration and referred more to the experimenter than their mothers compared to the infants of all other
conditions. Another study (Pelaez-Nogueras, Field, Cigales, Gonzalez & Clasky, 1994) investigated whether the 3-month-old infants of depressed mothers would express more positive affect to their non-depressed nursery teachers. The results showed that when the infants were interacting with their depressed mothers they displayed more negative affect, whereas when they were interacting with their
teachers they fostered more positive patterns of communication. These findings imply that infants of depressed mothers interact less with the new stimuli and prefer
interacting more with familiar non-depressed adults than with their mothers. Moreover, the previous evidence supports the assumption that mothers with
depressive symptoms express fewer emotional responses and signals during a social referencing situation.
The very few studies that investigated triadic interactions have also found differences between infants’ behavior of depressed and non-depressed mothers. According to these, infants of depressed mothers were found to be more negative when interacting with a toy. In particular, Hart, Field, Del Valle and Pelaez-Nogueras (1998) investigated those differences in a play-toy interaction in a sample of sixty-four 12-month-old infants and their mothers. The outcome was that the daughters of depressed mothers demonstrated less positive and more negative affect. Additionally, another study examined the emotional responses of forty 5-month infants of depressed and non-depressed mothers to animated and inanimated faces (Field et al., 2007). There were four conditions: an inanimate doll’s face; an animated (moving and talking) doll’s face; a mother’s less animated face during which she would imitate her infant’s behavior; and a mother’s more animated face, during a spontaneous play-interaction. The infants of depressed mothers expressed less positive and more negative behavior in the latter condition, whereas they showed unexpectedly more
positive and less negative behavior during the mother imitation condition.
Furthermore, the infants of depressed mothers showed more positive and less negative behavior, during both inanimate and animate doll conditions. These findings may imply that infants of depressed mothers process information more easily and modulate their emotional expression in a more positive way when interacting with an unfamiliar stimulus (doll conditions) rather than when interacting with their depressed mother. To summarize, though some studies have examined mother’s and infant’s behavior during triadic interactions and mother-child synchrony in dyadic interactions, the shared affect of mother-infant facial emotional expressions in a social referencing paradigm with computerized facial expressions remains yet to be explored.
The present study
The present study investigates mother-infant-person interaction in a community sample of 49 mothers and their 12-month infants. Shared positive and negative affect regarding mother-infant facial emotional expressions is investigated, during computerized presentations of positive (i.e. happy) and negative (i.e. angry) facial expressions. More specifically, the research questions are:
a) Is the shared positive affect of mother-infant facial expressions during positive and negative faces on the screen related to maternal depressive symptoms? b) Is the shared negative affect of mother-infant facial expressions during
positive and negative faces on the screen related to maternal depressive symptoms?
c) Does the shared positive or negative affect change as a function of positive compared to negative expression on the screen?
Based on previous research regarding the relation between maternal
& Tronick, 1987), it is hypothesized that shared positive affect will be negatively associated with maternal depressive symptoms. In respect of the second research question, it is hypothesized that shared negative affect will be positively associated with maternal depressive symptoms. This derives from previous research that found mothers with depression being more negative than positive when interacting with their infant compared to non-depressed mothers (Cohn, Campbell, Matias, & Hopkins, 1990), However, due to the fact that other studies (Aktar et al., in press) found that mothers with depression were more neutral than positive with their infants, shared negative affect could be expected to be negatively associated with maternal depressive symptoms. Finally, regarding the third research question, previous literature is contradicting (De Haan, Belsky, Reid, Volein & Johnson, 2004; Grossmann, Striano & Friederici, 2007) and mostly relevant to infants’ gaze and facial processing, not mother-infant emotional communication and facial expressions. For this reason, no hypothesis is formulated.
Keep in mind that we are investigating this aspect of mother-infant emotional communication for the first time in a mother-infant-person triadic interaction during computerized facial expressions; thus, we extrapolate these findings mostly based on previous studies of dyadic interaction and facial processing findings. A better
understanding of the mother-child interaction regarding their emotional expressions during computerized facial expression is necessary. The aim is to shed light into this aspect of triadic emotional exchanges, mother-infant shared affect and the
transmission of depressive symptoms. This knowledge could influence the development of programs and the health care providers’ work, by targeting at the prevention or the reduction of depressive symptoms, the empowerment of mother-child bond and the prevention of mother-children’s later socio-emotional problems.
Method Participants
The sample consisted of 49 mothers and their children (boys: n=20, girls:
n=29). The children’s mean age was 12.1 months (SD =0.6, range=10.2-13). Mother’s
mean age was 33.7 years (SD =4.8, range=22-45). Mothers’ nationality was 69.4% Dutch, 4.1% was Romanian, German and Brazilian, 2% English and France, whereas 14.3% was from other countries. Percentages of mother’s educational level were 12.2% for primary and secondary education, 16.3% for higher professional education, 67.3% for university education and 4.1% for other educational background. Finally, regarding the socioeconomic status (SES) of the mothers, 4% had an income up to €1.000 per month, 42.9% between €1.000 and €3.000, and 38.8% had an income higher than €3.000 per month. The information about maternal depressive symptoms, and the durations of shared positive and negative affect during the happy and angry faces on the screen are included in Table 1.
Procedure
Mothers with their infants were invited to the baby lab of the Child Development and Education Department of the University of Amsterdam. The participants were recruited via the municipality of Amsterdam, as a part of a larger group of participants for a longitudinal study. During the interaction, the infant was placed in an infant seat, mounted on a table, facing the screen and the mother was sitting in a chair next to the infant, diagonally looking at the screen. Two cameras were recording the triadic interaction: one recorded the face and upper body of the baby and parent, and the other recorded the screen. Both cameras were combined and recorded digitally. All facial expressions were coded, via Observer XT 10.5 and 11.5 (Noldus, Trienes, Hendriksen, Jansen, & Jansen, 2000).
Parents were instructed to be neutral during the experiment. In cases where the infant needed attention or soothing, parents were asked to interact with their infant as they normally do, but without using any toys. The experimenter was not present during the interaction and the time of the recording was different for each case (e.g. if the baby started to cry, the recording took a longer time to be completed). The study was approved by the ethical committee and all participants signed a written consent form in order to participate.
Measures
Maternal depressive symptoms Mothers rated their depressive mood using
the Depression Anxiety and Stress Scale (DASS; Lovibond & Lovibond, 1995). DASS is a 42-item questionnaire that measures depression, anxiety and stress. The Depression scale consists of 14 items and evaluates the emotional state of dysphoria, hopelessness, devaluation of life, self-deprecation, lack of interest/involvement, anhedonia, and inertia (e.g. ‘I couldn't seem to experience any positive feeling at all’). Each item rated the extent to which the mothers had experienced the statement during the past week, on a 4-point severity/frequency scale ((0) =‘Did not apply to me at all’ to (3) =‘Applied to me very much, or most of the time’). According to the scores, 5 categories take form; normal (0-9), mild (10-13), moderate (14-20), severe (21-27) and extremely severe depression (more than 28). Cronbach's alpha was 0.849.
Observational measures:
Duration of screen facial expressions Five computerized presentations of 2
males’ and 2 women’s positive and negative facial expressions (happy, fear, angry, sad, and neutral) appeared in the screen (Van der Schalk, J., Hawk, S. T., Fischer, A. H., & Doosje, B. J., 2011). Every video started with the neutral face and continued with a random presentation of the other four facial expressions. Twenty facial
expressions appeared in the screen. Each expression began with a neutral face for 0.5 seconds and then it was followed by the onset of the expression, which was held for 5 seconds. Before each facial presentation, visual and auditory attention-getters were used (i.e. little funny cartoons with sound jumping on the screen). For our study, the happy facial expressions of the screen have been used as positive emotional
expressions and the angry as negative.
Duration of mother-infant facial expressions Mother-infant facial expressions were
coded according to three categories: positive, negative, neutral. In the present study, only the positive and negative facial expressions are going to be investigated. More specifically, the positive facial expressions refer to both closed and open smiles. Opening of the mouth, raising the corners of the lips and cheeks and constriction of the eyes are characteristics that help to identify the smiles (Ekman & Friesen, 1982; Messinger, Fogel, & Dickson, 2001). The negative facial expressions refer to sad, cry, frowns and angry faces that can be recognized by lowering the brows as well as by constriction of the eyes and opening of the mouth (Bolzani Dinehart, Messinger, Acosta, Cassel, Ambadar, & Cohn, 2005). The coders were trained for one and a half month, until they reached a minimum of 80% agreement. Inter-rater reliability was established on 15% of the tapes: Cohen’s kappa was .86 (SD=.10) for mothers’ emotional expressions and .80 (SD=.11) for infants’ emotional expressions.
Outcome variables:
Co-occurrence of mother-infant positive and negative expressions The
outcome variables are obtained by calculating the co-occurrences of mothers’ and infants’ facial expressions during positive (i.e. happy) and negative (i.e. angry) facial expressions on the screen. Co-occurrences refer to a) the shared positive affect i.e. the durations of simultaneous positive mother-infant facial expressions, during the
positive and negative facial expressions of the screen and b) the shared negative affect i.e. the durations of simultaneous negative mother-infant facial expression, during the positive and negative facial expressions of the screen.
Data analysis
The data were analyzed with multilevel regression models for repeated observations to account for the nested structure coming from repeated observations of the screen. More specifically, the shared affect of mother-infant’s facial emotional expressions is considered to be nested within the positive and negative expressions of the screen. Maternal depressive symptoms and screen were the predictors, while the shared positive and negative affect were the outcomes. The skewness and kurtosis of half of the variables indicated a normal distribution (<|1.96|) (Tabachnick & Fidell, 2007). Six outliers (>|1.96|) were found and replaced by the next more extreme value based on their distribution. The scores on the continuous variables were standardized (i.e. maternal depressive symptoms and positive and negative affect). The intercept and the predictors (i.e. maternal depressive symptoms and screen expressions) were included in the models as fixed effects. Each outcome was analyzed separately, first without predictors, then with both predictors added and finally with their interaction. Random effects (intercept and slopes) were calculated, but because they made the model fit worse (Field, 2013), they were excluded of the final model. The association between maternal depression and outcome variables (i.e. the shared positive affect during positive faces, the shared positive affect during negative faces, the shared negative affect during positive faces, and the shared negative affect during negative faces) was also investigated in the preliminary analyses.
Results Preliminary analyses
None of the correlations between maternal depressive symptoms and the outcomes was significant, indicating that maternal depressive symptoms were not significantly associated with shared affect. Shared positive affect during happy faces on the screen was significantly correlated with shared positive affect during angry faces on the screen and shared negative affect during happy faces on the screen. Moreover, shared positive affect during angry faces on the screen was significantly positively correlated with the shared negative affect during happy faces on the screen. Pearson’s product-moment correlations (two-tailed) for maternal depressive
symptoms and shared positive-negative affect during positive and negative faces on the screen are provided in Table 2.
Main analyses Shared positive affect
Regarding the multilevel regression models, the shared positive affect was predicted only by the screen (see Table 3a). In more detail, during the positive expressions on the screen there was more shared positive affect than during the negative expressions on the screen. Repeated measures showed similar findings (see Table 4a). The effect of maternal depressive symptoms was not found to be
significant, neither was the maternal depressive symptoms-screen interaction.
Shared negative affect
In the model investigating the negative shared affect (see Table 3b), the screen was the only predictor that was significantly negatively related to the predicted outcome, given the effect of the other predictors. This means that during the negative facial expressions on the screen there was less negative shared affect than during the positive expressions on the screen. Repeated measures showed similar findings (see
Table 4b). The effect of maternal depressive symptoms was again not found to be significant, neither was the interaction.
Discussion
The present study investigated the relation between maternal depressive symptoms and shared affect of mother-infant facial emotional expressions, during positive and negative facial expressions, presented on the screen. Findings suggest that mother-infant shared positive and negative affect occurred in significantly longer durations during the positive rather than during the negative expressions on the screen. In turn, no statistically significant relation was found between maternal depressive symptoms and shared positive and negative affect. This link did not change as a function of the positive or negative expression on the screen.
To begin with, the most innovative aspect of this study is the finding that more shared positive and negative affect occurred during the positive expressions on the screen. Due to the lack of evidence, there was no hypothesis formulated about this research question. The most possible explanation of this finding is that mother-infant not only emotionally react in a matching way, but they also perceived in a similar manner the strangers’ positive and negative facial expressions. It could be that both mothers and infants perceived the happy expressions on the screen as friendly and hence, they felt more comfortable with emotionally interacting with each other and with exchanging both positive and negative shared affect. Similarly, they perceived the angry faces as threatening, and thus they felt more intimidated in emotionally interacting with one another, which led to a diminished expression of shared affect. Both scenarios, however, resulted to the same outcome: more shared positive and negative affect occurred during the happy faces compared to the negative ones. Since research so far has examined infants’ gaze instead of facial expressions, future
research should also investigate infants’ facial expressions as well as mother-infant simultaneous emotional expressions towards strangers’ expressions. In this way, we would be provided with more information on this subject.
An issue that should be taken into consideration, when interpreting this finding is the ambiguity of whether both mother and infant expressed more shared affect only by looking at the happy faces on the screen or if they expressed that after they looked at each other. More specifically, it could be either that the infant
emotionally reacted to the screen and the mother synchronized with her infant’s facial emotional expression or that the infant reacted to the screen by looking at the mother and synchronized with her emotional expression. A crucial dimension to be taken into account is that shared affect is a bidirectional effect (Tronick, 1989) and that its occurrence is strongly associated with the other person’s facial expression. Bearing this in mind, it would be valuable for future knowledge to investigate shared affect during positive and negative faces combined with mother-infant gaze.
The second finding revealed that maternal depressive symptoms and
consequently their interaction with the expression on the screen were not significantly associated with mother-infant shared affect. This finding was surprising and not consistent with the hypotheses. In absence of previous evidence on the effect of maternal depression in a social referencing paradigm, we should first note that our hypotheses were formulated on dyadic interactions. During those, maternal depressive symptoms were found to be associated with less positive and more neutral maternal affect (Aktar et al., in press), or more negative maternal affect (Cohn, Campbell, Matias, & Hopkins, 1990). In triadic interactions, it has been found that mothers with depressive symptoms are limited in exchanging information with their infants
(Stenberg, 2003). Nevertheless, according to our findings it can be concluded, based on our findings that depressive symptoms in triadic interactions are not related to less positive and more negative affect as in dyadic interactions. On the contrary, triadic interactions may act protectively for the mother-infant emotional communication and hence, for the socio-emotional development of the infant. To be more specific, in spite of maternal depressive symptoms being a cause of alterations in mother-infant face-to-face interaction, when a third party comes along –in our study the strangers’ facial expressions on the screen- it diminishes those alterations. In this way, the third person or object may act protectively for the mother-infant shared affect. However, it should be noted that we only looked at shared moments of positive and negative mother-infant facial expressions; we did not investigate separate durations of mothers’ and infants’ positivity or negativity. Thus, based on the fact that diagnosis of depression predicts less positive affect from parents, but not from the infants (Aktar, in press), an amount of information could have been lost, when we focused on shared affect.
Finally, a number of limitations have to be considered for a better
interpretation of the findings. First, regarding the sample characteristics, the mothers that participated in the current study were not clinically diagnosed with depression. Although, depressive symptoms have been found to predict clinical depression (Horwath, Johnson, Klerman, & Weissman, 1992; Pine, Cohen, Cohen, & Brook, 2014), we cannot assume that depression disorder would result in similar findings. In other words, it is not known whether the relation between maternal depression and shared affect during the positive and negative faces on the screen would be different, if mothers were clinically diagnosed. We should also take into consideration that the mothers that participated in the study were highly educated, with middle or higher income, and that the prevalence in maternal depressive symptoms was low. These
characteristics of the sample indicate that the findings might not be representative of the general population of mothers with depressive symptoms. Finally, although this study investigated the relation of maternal depressive symptoms and mother-infant shared positive and negative affect in a mother-infant-face triadic interaction, it still remains unknown what would happen in a real-life social referencing situation. As a consequence, further research is needed to explore the above associations not only in computerized tasks but also in naturalistic situations. Despite these limitations, this study has explored for the first time the relation between maternal depressive
symptoms and shared positive and negative affect, during positive and negative faces on the screen. More importantly, its most basic and innovative result is that during the positive screen the mother-infant shared affect occurs significantly more than during the negative screen.
Conclusion
Although an imperative and compelling literature has examined the
association between maternal depressive symptoms and mother-infant shared affect in dyadic interactions and infants’ attention to positive and negative expressions.
Nonetheless, little is known about the relationship between depressive symptoms and mother-infant shared affect in triadic interactions, especially in a social referencing paradigm with computerized positive and negative facial expressions. This issue was addressed in this study, according to which mother-infant shared positive and negative affect occurred significantly longer during the positive than during the negative
expressions on the screen. No statistically significant relation was found between maternal depressive symptoms and the outcomes and no interaction between the screen and the maternal depressive symptoms was evident. Although, more research
is needed in this field, this study filled in the gaps by exploring mother-infant shared affect in a triadic interaction.
References
Aktar, E. (in press). How do parents’ depression and anxiety, and infants’ negative temperament relate to parent-infant face-to-face interactions?
American Psychiatric Association. (2013). Diagnostic and statistical manual of
mental disorders (5th ed). Arlington, VA: American Psychiatric Association.
Barbera, J. D., Izard, C. E., Vietze, P., & Parisi, S. A. (1976). Four-and six-month-old infants' visual responses to joy, anger, and neutral expressions. Child
Development, 47, 535-538.
Barrera, M. E., & Maurer, D. (1981). The perception of facial expressions by the three-month-old. Child Development, 52, 203-206.
Bayer, J. K., Sanson, A.V., & Hemphill, S.A. (2006). Parent influences on early childhood internalizing difficulties. Journal of Applied Developmental
Psychology, 27, 542–559.
Beck, C. T. (1995). The effects of postpartum depression on maternal-infant interaction: A meta-analysis. Nursing Research, 44, 298-305.
Bettes, B. A. (1988). Maternal depression and motherese: Temporal and intonational features. Child Development, 59, 1089-1096.
Bolzani Dinehart, L. H., Messinger, D. S., Acosta, S. I., Cassel, T., Ambadar, Z., & Cohn, J. (2005). Adult perceptions of positive and negative infant emotional expressions. Infancy, 8, 279-303.
Campbell, S. B., Cohn, J. F., & Meyers, T. (1995). Depression in first-time mothers: mother-infant interaction and depression chronicity. Developmental
Campos, J. J., Campos, R. G., & Barrett, K. C. (1989). Emergent themes in the study of emotional development and emotion regulation. Developmental Psychology,
25, 394-402.
Cohn, J. F., & Tronick, E. Z. (1987). Mother–infant face-to-face interaction: The sequence of dyadic states at 3, 6, and 9 months. Developmental Psychology, 23, 68-77.
Cohn, J. F., Campbell, S. B., Matias, R., & Hopkins, J. (1990). Face-to-face
interactions of postpartum depressed and nondepressed mother-infant pairs at 2 months. Developmental Psychology, 26, 15-23.
Colonnesi, C., Zijlstra, B. J., Van der Zande, A., & Bögels, S. M. (2012). Coordination of gaze, facial expressions and vocalizations of early infant communication with mother and father. Infant Behavior and Development, 35, 523-532.
Corkum, V., & Moore, C. (1998). The origins of joint visual attention in infants.
Developmental Psychology, 34, 28-38.
De Haan, M., Belsky, J., Reid, V., Volein, A., & Johnson, M. H. (2004). Maternal personality and infants’ neural and visual responsivity to facial expressions of emotion. Journal of Child Psychology and Psychiatry, 45, 1209-1218.
Donovan, W. L., Leavitt, L. A., & Walsh, R. O. (1998). Conflict and depression predict maternal sensitivity to infant cries. Infant Behavior and Development,
21, 505-517.
Dunham, P. J., Moore, C. (1995). Current themes in research on joint attention. In C., Moore, P. J., Dunham (Eds.), Joint attention: Its origins and role in
development, (pp. 15-28), Lawrence Erlbaum Associates, Hillsdale, New
Ekman, P., & Friesen, W. V. (1982). Felt, false, and miserable smiles. Journal of
nonverbal behavior, 6, 238-252.
Feinman, S. (1982). Social referencing in infancy. Merrill-Palmer Quarterly, 28, 445-470.
Feinman, S., & Lewis, M. (1983). Social referencing at ten months: A second-order effect on infants' responses to strangers. Child Development, 54, 878-887. Feldman, R. (2003). Infant–mother and infant–father synchrony: The coregulation of
positive arousal. Infant Mental Health Journal, 24, 1-23.
Feldman, R. (2007). Parent–infant synchrony biological foundations and
developmental outcomes. Current Directions in Psychological Science, 16, 340-345.
Field, A. (2013). Discovering statistics using IBM SPSS statistics. Sage.
Field, T., Hernandez‐Reif, M., Diego, M., Feijo, L., Vera, Y., Gil, K., & Sanders, C. (2007). Responses to animate and inanimate faces by infants of depressed mothers. Early Child Development and Care, 177, 533-539.
Gavin, N. I., Gaynes, B. N., Lohr, K. N., Meltzer-Brody, S., Gartlehner, G., & Swinson, T. (2005). Perinatal depression: A systematic review of prevalence and incidence. Obstetrics & Gynecology, 106, 1071-1083.
Gewirtz, J. L., & Peláez-Nogueras, M. (1992). Social referencing as a learned process. In S. Feinman (Ed.), Social referencing and the social construction of reality in
infancy (pp. 151-173). New York: Plenum Press.
Goren, C., Sarty, M., & Wu, P., Y., K. (1975). Visual following and pattern
discrimination of face-like stimuli by newborn infants. Pediatrics, 56, 544–549. Groh, A. M., Roisman, G. I., Van IJzendoorn, M. H., BakermansKranenburg, M. J., &
attachment for children’s internalizing symptoms: A meta-analytic study. Child
Development, 83, 591-610.
Grossmann, T., Striano, T., & Friederici, A. D. (2007). Developmental changes in infants’ processing of happy and angry facial expressions: A neurobehavioral study. Brain and Cognition, 64, 30-41.
Hart, S., Field, T., Del Valle, C., & Pelaez-Nogueras, M. (1998). Depressed mothers' interactions with their one-year-old infants. Infant Behavior and Development,
21, 519-525.
Haviland, J. M., & Lelwica, M. (1987). The induced affect response: 10-week-old infants' responses to three emotion expressions. Developmental Psychology, 23, 97-104.
Hay, D. F. & Murray, P. (1982) Giving and requesting: Social facilitation of infants’ offers to adults. Infant Behavior and Development, 5, 301–10.
Hay, D. F. (1979) Cooperative interactions and sharing between very young children and their parents. Developmental Psychology, 15, 647–53.
Hernandez-Reif, M., Field, T., Diego, M., Vera, Y., & Pickens, J. (2006). Happy faces are habituated more slowly by infants of depressed mothers. Infant Behavior
and Development, 29, 131-135.
Hunnius, S., De Wit, T. C., Vrins, S., & Von Hofsten, C. (2011). Facing threat: Infants' and adults' visual scanning of faces with neutral, happy, sad, angry, and fearful emotional expressions. Cognition and Emotion, 25, 193-205.
Kaye, K., & Fogel, A. (1980). The temporal structure of face-to-face communication between mothers and infants. Developmental Psychology, 16, 454-464.
Klinnert, M., Campos, J. J., Sorce, J. F., Emde, R. N., & Svejda, M. (1983). Emotions as behavior regulators: Social referencing in infancy. In R. Plutchik and H.
Kellerman (Eds.), Emotion: Theory, research and experience: Vol 2. Emotions
in early development (pp. 57-86). New York: Academic Press.
Kotsoni, E., De Haan, M., & Johnson, M. H. (2001). Categorical perception of facial expressions by 7-month-old infants. Perception, 30, 1115-1125.
Lovibond, S.H. & Lovibond, P.F. (1995). Manual for the Depression Anxiety Stress
Scales. Sydney: Psychology Foundation.
Markova, G., & Legerstee, M. (2006). Contingency, imitation, and affect sharing: Foundations of infants' social awareness. Developmental Psychology, 42, 132-141.
Marsh, A. A., Ambady, N., & Kleck, R. E. (2005). The effects of fear and anger facial expressions on approach-and avoidance-related behaviors. Emotion, 5, 119-124. Messinger, D. S., Fogel, A., & Dickson, K. L. (2001). All smiles are positive, but
some smiles are more positive than others. Developmental Psychology, 37, 642-653.
Morales, M., Mundy, P., Crowson, M. M., Neal, A. R., & Delgado, C. E. (2005). Individual differences in infant attention skills, joint attention, and emotion regulation behaviour. International Journal of Behavioral Development, 29, 259-263.
Mumme, D. L., Fernald, A., & Herrera, C. (1996). Infants' responses to facial and vocal emotional signals in a social referencing paradigm. Child
Development, 67, 3219-3237.
Mundy, P., Sigman, M., Ungerer, J., & Sherman, T. (1986). Defining the social deficits of autism: The contribution of non‐verbal communication measures. Journal of child psychology and psychiatry, 27, 657-669.
Noldus, L. P., Trienes, R. J., Hendriksen, A. H., Jansen, H., & Jansen, R. G. (2000). The Observer Video-Pro: New software for the collection, management, and presentation of time-structured data from videotapes and digital media files. Behavior Research Methods, Instruments, & Computers, 32, 197-206. O'hara, M. W., & Swain, A. M. (1996). Rates and risk of postpartum depression-a
meta-analysis. International Review of Psychiatry, 8, 37-54.
Pascalis, O., & Kelly, D. J. (2009). The origins of face processing in humans: Phylogeny and ontogeny. Perspectives on Psychological Science, 4, 200-209. Pearson Education.
Pelaez-Nogueras, M., Field, T., Cigales, M., Gonzalez, A., & Clasky, S. (1994). Infants of depressed mothers show less" depressed" behavior with their nursery teachers. Infant Mental Health Journal, 15, 358-367.
Rottenberg, J. (2005). Mood and emotion in major depression. Current Directions in
Psychological Science, 14, 167–170.
Sorce, J. F., Emde, R. N., Campos, J. J., & Klinnert, M. D. (1985). Maternal emotional signaling: Its effect on the visual cliff behavior of 1-year-olds.
Developmental Psychology, 21, 195-200.
Stenberg, G. (2003). Effects of maternal inattentiveness on infant social referencing. Infant and Child Development, 12, 399-419.
Tabachnick, B.G. & Fidell, L.S. (2007). Using multivariate statistics, Boston: Tomasello, M., Carpenter, M., Call, J., Behne, T., & Moll, H. (2005). Understanding
and sharing intentions: The origins of cultural cognition. Behavioral and brain
sciences, 28, 675-691.
Trevarthen, C. (1979). Instincts for human understanding and for cultural cooperation: Their development in infancy. In M. von Cranach, K. Foppa, W. Lepenies & D.
Ploog (Eds.), Human ethnology: Claims and limits of a new discipline (pp. 530-571). Cambridge University Press.
Tronick, E. Z. (1989). Emotions and emotional communication in infants. American
psychologist, 44, 112-119.
Van der Schalk, J., Hawk, S. T., Fischer, A. H., & Doosje, B. J. (2011). Moving faces, looking places: The Amsterdam dynamic facial expressions set (ADFES),
Table 1
Descriptive statistics of the measurement variables: Maternal depressive symptoms and percentage durations of facial expressions
M SD Range
Maternal depressive symptoms 1.57 2.19 0-9
Positive affect during happy faces on the screen
0.66 0.76 0-2.59
Positive affect during angry faces on the screen
0.13 0.22 0-0.75
Negative affect during happy faces on the screen
0.66 0.76 0-2.59
Negative affect during angry faces on the screen
0.02 0.08 0-0.35
Table 2
Correlations between maternal depressive symptoms and the shared positive-negative affect during the positive and negative faces on the screen
Measures 1 2 3 4 5
1. Depression - .24 .10 .24 .05
2. Positive affect during happy screen
- - .58** 1.00** -.14
3. Positive affect during angry screen
- - - .58** -.18
4. Negative affect during happy screen
- - - - -.14
5. Negative affect during angry screen
- - - - -
Notes: N=49. *p<.05; **p<.01 (two-tailed), Depression= maternal depressive
Table 3
Multilevel regressions models regarding positive and negative shared affect. 3.a. Shared positive
affect (N=49) 3.b. Shared negative affect (N=49) β SE p β SE p Intercept 1.25 0.32 < .001 1.46 0.34 < .001 Depression 0.51 0.32 0.12 0.54 0.34 0.13 Happy (vs. angry)
expressions on the screen
-0.84 0.15 < .001 -0.97 0.17 < .001
Depression*Screen -0.24 0.15 0.13 -0.27 0.17 0.13
Table 4
Multilevel regressions models regarding repeated measures and shared affect. 4.a. Shared positive
affect (N=49)
4.b. Shared negative affect (N=49)
Parameter β SE Wald Z β SE Wald Z
UN (1,1) 1.45** 0.29 4.95 1.42** 0.29 4.95
UN (2,1) 0.20** 0.06 3.24 -0.02** 0.02 -1.01
UN (2,2) 0.10** 0.02 4.95 0.01** 0.02 4.95
Notes: *p<.05; **p<.01
UN (1,1)= happy facial expressions on the screen. UN (2,2)= angry facial expressions on the screen. UN (2,1)=covariate between the two expressions on the screen.
