Human attachment as a multi-dimensional control system: A computational implementation

Area of Science:

• Psychology
• Computational Modeling
• Developmental Science

Background:

• Attachment theory describes emotional bonds, typically between child and caregiver.
• Previous computational models often used a categorical approach, focusing on caregiver proximity.
• Recent theories suggest attachment is multidimensional, involving various relationship domains.

Purpose of the Study:

• To propose and test a dimensional perspective (DP) for computational modeling of attachment.
• To develop an agent-based model informed by the DP to simulate attachment behaviors.
• To evaluate the validity of the DP by comparing model outputs to established attachment literature.

Main Methods:

• Developed a DP-informed agent-based model of child-mother attachment.
• Modeled child and mother as point-agents in a two-dimensional space.
• Simulated interactions based on the Strange Situation Procedure for avoidant and ambivalent dimensions.

Main Results:

• The DP model generated dynamical patterns of approach and exploration.
• These simulated patterns align with descriptions found in attachment literature.
• The model confirmed the implementability and validity of the dimensional perspective.

Conclusions:

• The dimensional perspective offers a robust framework for computational attachment modeling.
• Agent-based modeling can effectively simulate complex attachment dynamics.
• This research validates the DP approach for understanding multidimensional attachment phenomena.