Revalidation of the Curiosity and Exploration Inventory-II (CEI-II) using network analysis

Abstract

Background: Curiosity is crucial across various domains, from clinical to educational fields, and holds potential for psychological interventions. Accurate definition and assessment of curiosity are essential for understanding its role and utility. Traditional approaches like factor analysis may not fully capture the construct’s nuances. Objectives: This study aims to reassess the psychometric properties of the Curiosity and Exploration Inventory-II (CEI-II) using Network Analysis. Methods: A total of 849 Spanish undergraduate students participated in the study. Descriptive analysis, partial-correlation network analysis with gender invariance testing, and Bayesian network model estimation were conducted. Results: The findings indicate that the CEI-II is best conceptualized as a stable, one-dimensional model, consistent with prior research. The partial-correlation network exhibited moderate density and was invariant in structure, centrality measures, and edge strength across genders, although global strength differed. The Bayesian network identified key pathways for designing interventions based on curiosity. Conclusions: While the results revealed three distinct item groupings based on centrality measures—challenge-seeking (specific curiosity), novelty-seeking (diversive curiosity), and a combination of both—the empirical evidence supported a stable unidimensional network structure. Items related to specific curiosity showed stronger interconnections, highlighting their importance in fostering curiosity-driven behaviors. These insights suggest that interventions targeting key items may enhance curiosity, and accounting for gender differences could further improve effectiveness.