Poster #158 - Time-series Modeling of Infant-mother Head Movement Dynamics

• In Event: 2-055 - Poster Session 06
  In Poster Session: PS 06 Section - Parenting & Parent-Child Relationships

Integrative Statement

Interactions between caregivers and infants are dynamic processes that emerge adaptively based on characteristics of the environment and the individuals. Examination of regularity and differences in patterns of exchanges within the same infant-caregiver dyad over time and across different dyads sheds light on how dyad members adapt to each other over the course of an interaction. One way to study these emergent within- and between-dyad differences is to allow parameters that govern the patterns of interactions to change over time.
Head movement is an important but often overlooked component of emotion and social interaction, serving turn-taking and back-channeling functions (Hammal, Cohn, & Messinger, 2015; Hammal, Cohn, & George, 2014). We investigated the dynamics of head movements in mothers and infants during the Still Face paradigm (Tronick, et al., 1978). Participants were 24 ethnically diverse 4-month-old infants (M = 4.07) and their mothers (see Figure 1). Zface (Jeni, Cohn, & Hanade, 2015), a real-time generic software was used to track the 3 degrees of rigid head movements (i.e., pitch, yaw, and roll) for each participant. The extracted 3 degrees of freedom were then combined using Mahalanobis distance into a single measure of head movement dynamics for each measurement occasion.
In our study, we utilized a vector-autoregressive model, allowing us to establish temporal dependencies of each individual’s behavioral state on his/her previous states, and on the other individual’s previous states in a dyad. We predicted autoregressive (AR) parameters by episode (Play, Still Face and Reunion), contemporaneous mother smiling, and children’s later Richter’s scores, a continuous index of attachment security derived from expert ratings of the Strange Situation which was administered at 15 months of age. Higher (positive) values of the AR parameter indicate that previous state has a greater influence on the current state. Thus, when an individual shows a large head movement, a higher AR value means that this extreme motion would persist for longer periods of time, and lead to greater overall variability in the individual’s head position relative to his/her baseline head position. The AR parameter thus can be conceptualized in this context as an index of sustained variability in movements.
Results suggested that interaction dynamics indexed by head movements differed depending on infants’ levels of attachment security. After controlling for differences between episodes, infants who were more securely attached had more sustained variability in their head movements, possibly reflecting more variable engagement with mother and the nonsocial environment. Moreover, the relationship between mothers’ head movements and their own smiling varied with infant levels of future secure attachment. Mothers whose infants were more securely attached had less sustained variability in their head movements when they smiled compared to when they did not (see Figure 2). On the other hand, mothers whose infants were less securely attached showed comparable sustained variability in their head movements when they smiled versus not. This finding suggests that mothers of securely attached infants exhibit greater variability in their head movement patterns depending on their own affective signal than do mothers of insecurely attached infants.

Authors

• Meng Chen, The Pennsylvania State University
  Non-Presenting Author

• Sy-Miin Chow, Penn State University
  Non-Presenting Author

• Zakia Hammal, The Robotics Institute, Carnegie Mellon University
  Non-Presenting Author

• Daniel Messinger, University of Miami
  Presenting Author

• Jeffrey Cohn, University of Pittsburgh
  Non-Presenting Author