Attachment Security and Infants’ Neural and Attentional Processing of Maternal Emotion Signals

• In Event: 2-189 - Internal Working Models in the Making? Attachment and Social Information Processing During Early Development

Integrative Statement

Background: Faces are a key channel for transmitting important signals related to attachment within caregiver-infant interaction, such as signals of affiliation, threat, and anger. Studying infants’ attentional and neural responses to faces during the first year may therefore reveal early markers of attachment formation and caregiver-infant relationship. Initial results indeed suggest that the development of secure infant attachment is associated with relatively greater attention to emotional facial expressions (e.g., Peltola et al., 2015; Taylor-Colls & Fearon, 2015). In the present study, we were specifically interested in whether neural and attentional responses to emotional signals displayed on the face of the attachment figure (the mother) at 7 months of age are associated with attachment security at 12 months of age.
Methods: Infants participated in an event-related potential (ERP) measurement and an attention task at 7 months (n = 90). Attachment was assessed with the Strange Situation Procedure (SSP) at 12 months (n = 69). ERPs were measured with a 128-electrode Geodesic Sensor Net to pictures of the infant’s mother expressing angry and happy facial emotions. The ERP data were analyzed for the occipitotemporal N290 and P400 components (reflecting activity of face-sensitive brain areas such as the fusiform gyrus) and the frontocentral Nc component (indicating the amount of attention allocated to different stimuli). After the ERP task, eye-movement based attention to the mother’s and a stranger’s angry and happy faces was assessed with an Overlap task (Peltola et al., 2015) to examine whether infants’ attention to facial signals of the attachment figure and an unfamiliar person differ as a function of attachment security. Complete ERP and SSP data were available from 61 infants, and attention and SSP data from 50 infants. Infants were grouped as secure (54%) and insecure (46%; i.e., avoidant, resistant, and disorganized) for the analyses.
Results: No effects of attachment were observed in the N290 and P400 ERP components. In the Nc component, an Emotion x Attachment interaction was observed, which was due to securely attached infants showing larger Nc amplitudes (i.e., indicative of greater attention) to the mother’s happy than angry faces, p = .02, d = .45, whereas insecurely attached infants did not show differences in their Nc amplitudes to the mother’s happy and angry faces, p = .42, d = .14. In the Overlap task data, a marginal Familiarity x Attachment interaction was observed, which was due to insecurely attached infants showing less attention to the mother’s vs. the stranger’s face, p < .001, d = .76, whereas securely attached infants showed equal amount of attention to the mother’s vs. the stranger’s face, p = .13, d = .26.
Conclusions: The current results extend the small number of studies on the relations between attachment and social information processing in infancy (Biro et al., 2015; Peltola et al., 2015) and suggest that insecure infant attachment is associated with reduced neural differentiation of maternal emotion signals and a greater tendency to divert attention away from the mother’s vs. stranger’s emotional faces.

Authors

• Mikko J. Peltola, Tampere University
  Presenting Author

• Szilvia Biro, Leiden University
  Non-Presenting Author

• Renske Huffmeijer, Leiden University
  Non-Presenting Author

• Lenneke Alink, Leiden University
  Non-Presenting Author

• Marian J Bakermans-Kranenburg, Vrije Universiteit Amsterdam
  Non-Presenting Author

• Marinus H van IJzendoorn, Erasmus University Rotterdam
  Non-Presenting Author