Dietary and exercise effects on neural responses to visual food stimuli in overweight and obese adults
Results from human neuroimaging studies consistently demonstrate that obesity is associated with altered neural reward processing of food-relevant cues. These observations have caused some investigators to question whether behavioral interventions such as dietary manipulations or physical activity might modulate processing of food cues in reward-related brain regions. The use of functional magnetic resonance imaging (fMRI) to measure neural responses to visual food stimuli is a commonly utilized paradigm for assessing neural reward responses relating to ingestive behavior of foods in humans. Using this study design, limited evidence suggests that increases in dietary protein and participation in acute and chronic aerobic exercise may modulate reward-related neural responses to visual food stimuli. However, data regarding the reliability of this fMRI-based paradigm to produce consistent results on multiple testing days and time course of responses following a meal have not been systematically studied. These data are critically important for the proper interpretation of existing literature and for the design of future intervention studies using this paradigm. Therefore, the purpose of Study 1(Chapter 2) was to 1) assess the test-retest reliability of fasting-state neural responses to visual food stimuli and 2) document potential meal-induced changes in neural responses up to 4.5 hours after the consumption of a meal in overweight and obese adults. We hypothesized that fasting-state neural responses would demonstrate good to excellent test-retest reliability and that neural responses to visual food stimuli would be suppressed following meal consumption and then would return to fasting-state levels of activation by 4.5 hours after the meal. Our results indicate that visual food stimuli elicits relatively consistent mean or group-level neural activation in reward-associated brain regions including the insula, amygdala, orbitofrontal cortex, caudate, and putamen. However, contrary to our hypotheses, the test-retest reliabilities of these responses were generally poor (high degree of within-subject variability) and responses were relatively unaffected by meal consumption. These results have important implications relating to the design of future studies including observable correlations among study outcomes, sample size determination, and statistical modeling. Being cognizant of these considerations, the purpose of the two randomized crossover trials within this document were to assess the effects of dietary protein and fiber (Study 2, Chapter 3) and dietary protein and acute aerobic exercise (Study 3, Chapter 4) on neural responses to visual food stimuli and appetite in overweight (Study 2) and obese (Study 3) adults. For Study 2, we hypothesized that higher intakes of protein and fiber at breakfast would independently and additively increase fullness and decrease hunger, desire to eat, neural responses to visual food stimuli, and ad libitum energy intake at lunch. For Study 3, we hypothesized that higher protein consumption at dinner would increase fullness and decrease hunger, desire to eat, and neural responses to visual food stimuli similar to our hypotheses in Study 2. We further hypothesized that acute aerobic exercise would decrease fullness and increase hunger, desire to eat, and neural responses to visual food stimuli only when performed prior to the consumption of a normal protein dinner. Higher protein breakfasts in Study 2 transiently decreased desire to eat compared to normal protein breakfasts, and hunger was increased on exercise compared to rest in Study 3. Appetite and neural responses to visual food stimuli were otherwise not influenced by the interventions in either study. Collectively, the results of these intervention studies do not support higher intakes of dietary protein and fiber or acute aerobic exercise as significant modulators of neural reward-based ingestive behavior in overweight and obese adults.
Campbell, Purdue University.
Off-Campus Purdue Users:
To access this dissertation, please log in to our