Pushing the Boundaries of Background Functional Connectivity for Infant fNIRS Data: Evaluating Alternative Analytical Approaches

There is increasing interest in task-based functional connectivity analyses to examine the emergence of functional networks during specific cognitive states starting early in development (e.g., infancy). However, studying functional connectivity in infants presents unique methodological challenges. Task-based neuroimaging studies must be carefully designed to collect sufficient high-quality data, while remaining sensitive to infants’ developing motor and cognitive abilities. This limits the feasibility of certain analysis techniques, particularly those requiring a high number of trials or strict adherence to contiguous trial structures. This study aims to expand established analysis approaches for infant functional connectivity studies by reanalysing an existing fNIRS dataset with a recognised signature of task-based functional connectivity. We assess the feasibility of moving away from conventional analyses that require several contiguous trials per condition - a requirement that is often impractical in infant studies. Instead, we explore whether background functional connectivity (BGFC) analyses, which utilises the residual neural response (i.e., what remains after removing task-evoked responses), can offer a more flexible and scalable approach. To do so, we evaluate three analytic strategies: (1) downsampling, which systematically reduces the number of trials included to assess how many are necessary for well-powered analyses; (2) shuffling, which randomly reorders trial-level residuals within conditions to test whether connectivity patterns are driven by trial sequence; and (3) averaging single-trial residuals, which examines whether connectivity measures can be meaningfully extracted from individual trial residuals. Our findings indicate that downsampling was informative about the number of trials needed for well-powered analyses and the timing of the functional connectivity effect. Removing trials from the start of the block was not viable, but this approach was viable when trials were removed from the end. The shuffling approach was not viable, suggesting that trial order may be important for capturing meaningful connectivity patterns. Single-trial residual analyses revealed that while connectivity can be examined toward the level of individual trials, averaging trial-level residuals before performing connectivity analyses produced more robust and interpretable results. These findings provide new insights into the methodological feasibility of background functional connectivity analyses for infant fNIRS data. By demonstrating that downsampling and averaged single-trial connectivity approaches are viable, our study highlights ways to enhance the accessibility and flexibility of task-based functional connectivity studies in infancy. These results have important implications for task design, as they suggest that strict adherence to contiguous trial structures may not be necessary, allowing researchers to design more adaptable and infant-friendly paradigms. Overall, this work contributes to advancing the methodological toolkit for studying functional brain networks in early development.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.