Brain network dynamics associated with intentional weight loss in older adults.

Current obesity research suggests a link between obesity and challenges in executive functions such as decision-making, inhibitory control, and reward valuation, but what separates older individuals who are vulnerable to these deficiencies from those who are resilient remains unknown. We have mounting evidence that suggests disruptions in brain networks are a key factor. However, most of the brain connectivity analyses performed in the obesity literature estimate a static, unchanging brain network for each experimental scanning session. In this paper, we conduct a state-based dynamic functional connectivity analysis on functional magnetic resonance imaging (fMRI) data collected during a food-cue task from 58 older adults with obesity. Our intent was to build upon an original investigation of two functional networks (FN1 and FN2) found in an independent study that were associated with 18-month weight loss in older adults. We used a novel Hidden semi-Markov modeling (HSMM) approach to estimate five network states traversed by the individuals in this study, as well as the most probable sequence of states for each participant. We quantified the occupancy time, dwell time, and transition probabilities across states and compared these between older adults in a weight loss intervention who successfully lost 5% of their body weight at 6 months and those who did not. As in our prior study, adults with obesity spent much of their time in a network state where regions in FN1 and FN2 were largely segregated regardless of weight-loss success. However, we found that older adults who successfully lost weight after 6 months spent more total time in additional network states where insula and limbic structures were well integrated. On the contrary, those who were unsuccessful at losing weight after 6 months, spent more time in a state with strong connections between insula and motor structures, with limbic structures mostly segregated from the other regions. This recognition that weight management involves complex temporal shifts in brain states could lead to the development of new treatment options that complement current behavioral methods for weight management.