Mapping the network structure of self-reported everyday executive-functioning difficulties in secondary-school and...

Executive functions support students' ability to regulate cognition, behavior, and emotion in response to academic demands, thereby providing a regulatory foundation for learning. However, little is known about how self-reported everyday executive-functioning difficulties are conditionally related, particularly in Central American educational settings. This study mapped the network structure of self-reported everyday executive-functioning difficulties in secondary-school and university student samples from Honduras.

A cross-sectional study was conducted with 2,330 students, including 810 secondary-school students and 1,520 university students. Executive functioning was assessed using the Executive Functioning Questionnaire (EFECO), an ecological self-report measure. Gaussian graphical models were estimated to examine the global network, node centrality, predictability, accuracy, and stability. A secondary exploratory comparison was also conducted between the two independently recruited samples.

The global network was dense, retaining 23 of 28 possible edges and showing high transitivity. The strongest associations were observed between inhibition and emotional control, planning and working memory, and initiative and working memory. Planning, working memory, and initiative had the highest centrality estimates in the global network. These nodes also showed high predictability (R ² = 0.62-0.68). Centrality stability was high for both strength and expected influence. The exploratory comparison between the secondary-school and university student samples did not detect statistically significant differences in overall network structure or global strength, although localized sample-specific contrasts were observed in selected edges and centrality indices.

These findings extend executive-function network research beyond performance-based tasks by characterizing conditional associations among EFECO scores in an underrepresented Central American context. Centrality estimates reflect relative connectivity rather than causal importance, and the nonsignificant global comparison does not establish network equivalence. Given the cross-sectional design, a single self-report instrument, and independently recruited nonprobability samples, longitudinal, multimethod, and multisite studies are needed to establish the stability, generalizability, and educational relevance of these network patterns.