Effect of storytelling-based learning on child psychiatry psychopharmacology education among medical students: a controlled pretest-posttest study.

The identification of stable, individualized brain networks is of considerable interest in functional magnetic resonance imaging (fMRI). Though much research has relied on standardized, group-average template networks, individualized networks may better predict treatment responses, capture heterogeneity in psychiatric disorders, and optimize neuromodulation approaches for a given patient (, , , ). Precision functional mapping (PFM) involves estimating networks based on functional connectivity (FC) relationships across the cortex or entire brain and has emerged as a promising framework for this purpose. Foundational PFM studies often leveraged densely sampled data (> 2 h of resting-state fMRI from individual participants across sessions; , , ) to establish that individual-specific network organization is reproducible within a person and includes idiosyncratic features not apparent in group-average maps (, , , ). Studies also indicate that data needs are target dependent: whole connectome FC estimates can converge relatively quickly (∼30 min in adults (, ), while reliable network assignment and topographic stability can require much longer duration (∼90 min; (, ). This variability raises a central translational question: what level of individualized network characterization is achievable under the more constrained scanning conditions typical of developmental and clinical research? Adult PFM studies have utilized a variety of methods for identifying individualized networks. Methods include Infomap (), Homologous Functional Region mapping (), Template matching (), and multi-session hierarchical Bayesian model (MS-HBM (). Original research on adult PFM relied on Infomap, a computationally intensive method that derives whole-brain networks (including subcortex) by subdividing communities of vertices or voxels based on functional connectivity thresholds. Individual networks derived from Infomap often differed topologically (in spatial location) from group-average networks (), and the stability of these differences over time has been a central focus. Stability analyses often involve measuring elbow points, which are inflection points at which the rate of improvement of a metric diminishes significantly (without implying that further improvements past that point are not meaningful). Spatial stability analyses have shown elbows at around 60 min (). In addition, network topologies appear similar across resting-state or task-based MRI, allowing data concatenation to increase usable signal (, ). Stable networks may be identified across the brain, though they are most consistently observed in primary sensory cortices (, , )—including visual, somatomotor, and auditory networks. Interestingly, these areas also show stronger task-related activations when individualized networks are used instead of group templates, suggesting they are meaningfully related to individual perceptual abilities (, ). In contrast, regions susceptible to signal dropout (e.g., ventromedial prefrontal cortex, anterior medial temporal lobe), exhibit more limited stability (, ).