Mimicking emotions: how 3–12-month-old infants use the facial expressions and eyes of a model

Abstract

While there is an extensive literature on the tendency to mimic emotional expressions in adults, it is unclear how this skill emerges and develops over time. Specifically, it is unclear whether infants mimic discrete emotion-related facial actions, whether their facial displays are moderated by contextual cues and whether infants' emotional mimicry is constrained by developmental changes in the ability to discriminate emotions. We therefore investigate these questions using Baby-FACS to code infants' facial displays and eye-movement tracking to examine infants' looking times at facial expressions. Three-, 7-, and 12-month-old participants were exposed to dynamic facial expressions (joy, anger, fear, disgust, sadness) of a virtual model which either looked at the infant or had an averted gaze. Infants did not match emotion-specific facial actions shown by the model, but they produced valence-congruent facial responses to the distinct expressions. Furthermore, only the 7- and 12-month-olds displayed negative responses to the model's negative expressions and they looked more at areas of the face recruiting facial actions involved in specific expressions. Our results suggest that valence-congruent expressions emerge in infancy during a period where the decoding of facial expressions becomes increasingly sensitive to the social signal value of emotions.

Dynamic Advances in Emotion Processing: Differential Attention towards the Critical Features of Dynamic Emotional Expressions in 7-Month-Old Infants

“…Although the extant literature on processing of dynamic facial expressions of emotion is limited, these findings are generally in line with previous data suggesting that by 7 months of age, infants exhibit scanning patterns that take into account the critical features of dynamic emotional expressions. Soussignan et al [ 38 ] found that while 3-month-old infants exhibited similar scanning patterns across multiple expressions, 7-month-olds exhibited differential scanning of facial features across expressions (e.g., increased fixations towards the eyes of fearful, angry, and sad faces compared to happy faces, fearful compared to neutral faces, and increased fixations towards the mouth of happy faces compared to all other expressions). Similarly, adults typically allocate attention towards different facial features depending on the emotion.…”

“…Parents of infants were contacted from an internal database via phone and email regarding participation in the current study. The sample size was determined by a power analysis using G*Power 3 software (Version 3.1.9.3, Heinrich Heine University, Düsseldorf, Germany) [ 41 ], to ensure sufficient power to detect a medium effect in the within-between analysis of variance (ANOVA) interaction (given the statistical significance criterion of 0.05, based on previous data; [ 38 ]). An additional 11 participants were tested, but their data were excluded because they did not complete the task due to fussiness (leading either to poor calibration at the start of the task, or inattention during the task; n = 5), technical difficulties with the eye tracking equipment ( n = 5), or they looked at the face for less than 1000 ms for both trials showing the same emotion ( n = 1).…”

“…In one such study, Soussignan et al [ 38 ] used eye tracking to examine developmental changes in 3-, 7-, and 12- month-old infants’ scanning of basic emotions (anger, disgust, fear, sadness, happiness, and neutral) using short movie clips of 3D avatar faces. They found that 7-month-olds exhibited greater attention towards the eye region of angry compared to happy faces, as well fearful compared to neutral faces, and 12-month-olds exhibited greater scanning of the fearful eyes compared to the eyes of all other expressions.…”

Disentangling the negativity bias: 7-9-month-old crawling and non-crawling infants' responses to fearful and angry expressions.

“…In an exploratory portion of the current study, infants' affective responses to facial expressions were investigated. Affective facial reactions have been studied extensively in adults using the Facial Action Coding System (FACS), which is an anatomically based Table 1 for a list of AUs have been commonly indicated as related to specifically positive or specifically negative facial displays (Camras et al, 2007;Dondi et al, 2007;Fogel, Hsu, Shapiro, Nelson-Goens, & Secrist, 2006;Kohut, Pillai Riddell, Flora, & Oster, 2012;Longfier et al, 2016;Mendes & Seidl-De-Moura, 2014;Messinger, Cassel, Acosta, Ambadar, & Cohn, 2008;Messinger, Mahoor, Chow, & Cohn, 2009;Soussignan et al, 2017)). Using the Baby FACS system, researchers have found that infants display a constellation of AUs in response to negative experiences, such as immunization pain (Kohut et al, 2012), novel foods (Longfier et al, 2016), arm restraint, and fear-eliciting situations (Camras et al, 2007).…”

“…Although the majority of studies using Baby FACS have focused on infants' affective responses to negative situations, one study asked how infants would respond to images of facial expressions. Soussignan et al (2017) presented 3-, 7-, and 12-month-old infants with virtual models of male and female faces displaying dynamic emotional expressions. Infant expressions were hand-coded using BabyFACS for clusters of AUs representing prototypical expressions and for AUs related to general negative affective responses.…”