Continuous measurement of attachment behavior: A multimodal view of the strange situation procedure

Area of Science:

• Developmental Psychology
• Infant Behavior and Development
• Social-Emotional Functioning

Background:

• Infant attachment is crucial for social-emotional development, typically assessed via the time-intensive Strange Situation Procedure (SSP).
• Current SSP methods lack objective, continuous behavioral data, relying on expert interpretation.
• There is a need for objective, quantifiable measures of infant attachment behaviors.

Purpose of the Study:

• To develop and validate objective, multimodal sensor-based measures for quantifying infant attachment behaviors during the SSP.
• To assess the predictive validity of these objective measures against expert-derived attachment classifications.
• To explore the potential of inexpensive sensors for continuous infant behavior monitoring.

Main Methods:

• Forty-nine 1-year-olds and their mothers participated in the SSP.
• Multimodal data including 3D movement (Kinect sensors) and audio (LENA recordings for crying duration) were collected.
• Theoretically-informed measures (e.g., dyadic contact, infant approach velocity, crying) were derived to predict attachment outcomes.

Main Results:

• Objective multimodal measures significantly predicted 8 out of 9 expert ratings of infant attachment behaviors during SSP reunions.
• These measures explained approximately 50% of the variance in key attachment dimensions like approach/avoidance and resistance/disorganization.
• Including all objective measures predicted individual ratings with an average accuracy of one Likert scale point.

Conclusions:

• Inexpensive sensors (Kinect, LENA) can effectively quantify infant attachment behaviors within the SSP framework.
• Objective measurement holds significant promise for enhancing the understanding of infant-parent interactions and attachment.
• This approach offers a more efficient and objective alternative to traditional expert-based SSP classifications.