cognitive network science
Definition
Cognitive network science is an approach that encodes relationships between concepts to give structure to human knowledge through associative links between concepts, ideas, and words. Drawing on graph theory and complex systems science, it provides a means to visualise and quantify the web of relationships among concepts, with nodes representing individual ideas expressed through words. A specific instantiation of this framework, Textual Forma Mentis Networks, reconstructs the semantic, syntactic, and emotional patterns embedded in text by combining AI-based syntactic parsing with psychological valence and emotion data, without relying on simple word co-occurrence. Applied to psychometric instruments such as the 42-item Depression Anxiety and Stress Scale, this framework can identify semantic communities corresponding to emotional dysregulation, emotional exhaustion, physical distress, and tension states in ways that align with psychometric factor structures in non-random ways.
Sources: Stanghellini et al. (2024)
Related Terms
- network psychometrics (1 shared article)
- network clustering (1 shared article)
- text analysis (1 shared article)
- psychometric measurements (1 shared article)
- semantic framing (1 shared article)
Applications
Cognitive Network Science and Network Psychometrics
Cognitive network science and network psychometrics are bridged through the concept of semantic loadings, which quantify how clusters of semantically related concepts distribute across psychometric factors identified by item ratings. In the analysis of the DASS, semantic loadings derived from Textual Forma Mentis Networks were found to predict factor structures identified by Exploratory Graph Analysis, outperforming purely psychometric methods in capturing the organisation of depression, anxiety, and stress dimensions.
Sources: Stanghellini et al. (2024)
Cognitive Network Science and Semantic Memory
Cognitive network science connects to semantic memory through the theoretical grounding of Textual Forma Mentis Networks in the idea that reading a psychometric item activates interconnected concepts in semantic memory that then influence how a respondent rates that item. The TFMN structure extracted from questionnaire texts is treated as a partial model of the semantic memories assumed to be broadly shared across individuals completing a given scale, a property justified by the face validity requirements psychometricians impose during scale construction.
Sources: Stanghellini et al. (2024)
