Perceptual Characterization of the Macronutrient Picture System (MaPS) for Food Image fMRI

Hunger modulates perceptions of food health but not taste in restricted eaters

Introduction

Food taste and health perceptions are of particular interest for their implications on food choice. Most in vivo food choice studies experimentally control for hunger via a set preload or fast.

Methods

To explore how hunger may interact with these perceptions to impact food decisions, we compared taste and health perceptions of sweet and savory obesogenic food items among hungry or satiated participants with varying restrained eating tendencies.

Results

In our sample of 232 adults (M BMI = 25.9; M age = 36.4 yrs), highly palatable foods were perceived as tasty but unhealthy. Tastiness ratings were high, consistent across restrained eating groups, and unassociated with hunger. Perceptions of health, however, were impacted by the interaction of restrained eating group and hunger. Amongst hungry participants only, a graded association between restrained eating group and perceptions of health emerged for both food types. Specifically, hungry and highly restrained eaters viewed sweet foods as 2.8x healthier and savory foods as 2.1x healthier than their satiated counterparts.

Discussion

Our data suggest that hunger predicts differential health perceptions, but not tastiness ratings, among restrained eaters. We argue that the generalization of food perception data-especially among different eater types-may be limited if the continuum of hunger level is experimentally constrained. Therefore, hunger is a critical dynamic to consider in explorations of food perceptions and eating behavior in restrained eaters.

Development of Chinese food picture library for inducing food cravings

Cue-induced food cravings are strong desires directed toward specific foods, usually ones with high caloric content, and can lead to overeating. However, although food cravings vary according to individual preferences for specific high-calorie food subtypes, a structured library of food craving-inducing pictures including multiple categories of high-calorie foods does not yet exist. Here, we developed and validated a picture library of Chinese foods (PLCF) consisting of five subtypes of high-calorie foods (i.e., sweets, starches, salty foods, fatty foods, and sugary drinks) to allow for more nuanced future investigations in food craving research, particularly in Chinese cultural contexts. We collected 100 food images representing these five subtypes, with four food items per subtype depicted in five high-resolution photographs each. We recruited 241 individuals with overweight or obesity to rate the food pictures based on craving, familiarity, valence, and arousal dimensions. Of these participants, 213 reported the severity of problematic eating behaviors as a clinical characteristic. Under the condition of mixing multiple subtypes of high-calorie foods, we did not observe significant differences in craving ratings for high- and low-calorie food images (p _tukey > 0.05). Then, we compared each subtype of high-calorie food images to low-calorie ones, and found craving ratings were greater for the images of salty foods and sugary drinks (ps < 0.05). Furthermore, we conducted a subgroup analysis of individuals according to whether they did or did not meet the criteria for food addiction (FA) and found that greater cravings induced by the images of high-calorie food subtypes (i.e., salty foods and sugary drinks) only appeared in the subgroup that met the FA criteria. The results show that the PLCF is practical for investigating food cravings.

Validating Visual Stimuli of Nature Images and Identifying the Representative Characteristics

This study fills a void in the literature by both validating images of nature for use in future research experiments and examining which characteristics of these visual stimuli are found to be most representative of nature. We utilized a convenience sample of university students to assess 129 different nature images on which best represented nature. Participants (n = 40) viewed one image per question (n = 129) and were asked to rate images using a 5-point Likert scale, with the anchors “best represents nature” (5) and “least represents nature” (1). Average ratings across participants were calculated for each image. Canopies, mountains, bodies of water, and unnatural elements were identified as semantic categories of interest, as well as atmospheric perspectives and close-range views. We conducted the ordinary least squares (OLS) regression and the ordered logistic regression analyses to identify semantic categories highly representative of nature, controlling for the presence/absence of other semantic categories. The results showed that canopies, bodies of water, and mountains were found to be highly representative of nature, whereas unnatural elements and close-range views were inversely related. Understanding semantic categories most representative of nature is useful in developing nature-centered interventions in behavioral performance research and other neuroimaging modalities. All images are housed in an online repository and we welcome the use of the final 10 highly representative nature images by other researchers, which will hopefully prompt and expedite future examinations of nature across multiple research formats.

Effects of testosterone administration on fMRI responses to executive function, aggressive behavior, and emotion processing tasks during severe exercise- and diet-induced energy deficit

BACKGROUND

Clinical administration of testosterone is widely used due to a variety of claimed physical and cognitive benefits. Testosterone administration is associated with enhanced brain and cognitive function, as well as mood, in energy-balanced males, although such relationships are controversial. However, the effects of testosterone administration on the brains of energy-deficient males, whose testosterone concentrations are likely to be well below normal, have not been investigated.

METHODS

This study collected functional magnetic resonance imaging (fMRI) data from 50 non-obese young men before (PRE) and shortly after (POST) 28 days of severe exercise-and-diet-induced energy deficit during which testosterone (200 mg testosterone enanthate per week in sesame oil, TEST) or placebo (sesame seed oil only, PLA) were administered. Scans were also collected after a post-energy-deficit weight regain period (REC). Participants completed five fMRI tasks that assessed aspects of: 1) executive function (Attention Network Task or ANT; Multi-Source Interference Task or MSIT; AXE Continuous Processing Task or AXCPT); 2) aggressive behavior (Provoked Aggression Task or AGG); and 3) latent emotion processing (Emotional Face Processing or EMO).

RESULTS

Changes over time in task-related fMRI activation in a priori defined task-critical brain regions during performance of 2 out of 5 tasks were significantly different between TEST and PLA, with TEST showing greater levels of activation during ANT in the right anterior cingulate gyrus at POST and during MSIT in several brain regions at REC. Changes over time in objective task performance were not statistically significant; testosterone-treated volunteers had greater self-reported anger during AGG at POST.

CONCLUSIONS

Testosterone administration can alter some aspects of brain function during severe energy deficit and increase levels of anger.

The effects of the form of sugar (solid vs. beverage) on body weight and fMRI activation: A randomized controlled pilot study

Objective

To test if sugar sweetened beverages (SSBs) and sugar sweetened solids (SSSs) have differential effects on body weight and reward processing in the brain.

Methods

In a single blind randomized controlled pilot trial (RCT), twenty participants with BMI between 20 and 40 kg/m² were randomized to consume a 20 fluid ounce soda (SSB, 248 kcal) or the equivalent in solid form (SSS; similar to thick gelatin or gummy candy) daily. At baseline and day 28, fasting body weight and fed-state BOLD fMRI of the brain were assessed. Differences in fMRI signals between views of low-fat (LF (<30%)) high sugar (HS (>30%)) food, and non-food images were calculated in brain regions implicated in energy homeostasis, taste, and reward.

Results

All participants in the SSB (6F 4M; 8 Caucasian; 36±14 y, 28.2±5.5 kg/m²; Mean±SD) and SSS (3F 7M; 6 Caucasian; 39±12; 26.3±4.4) groups completed the study. Weight change was 0.27±0.78 kg between SSB and SSS participants. Changes in the fMRI response to LF/HS foods in reward, homeostatic and taste regions tended to not be different between the groups over the four weeks. However, activation of the right substantia nigra increased following the SSB but decreased activation following the SSS in response to LF/HS foods over 28 days (-0.32±0.12). Ratings of wanting for LF/HS foods were correlated with activation in several brain regions, including the OFC.

Conclusions

Change in weight was modest between the groups in this study. Daily consumption of a SSB over 28 days led to mixed responses to LF/HS foods in areas of the brain associated with reward. Ratings of wanting are correlated with fMRI activation inside an MRI scanner.

Food Folio by Columbia Center for Eating Disorders: A Freely Available Food Image Database

Food images are useful stimuli for the study of cognitive processes as well as eating behavior. To enhance rigor and reproducibility in task-based research, it is advantageous to have stimulus sets that are publicly available and well characterized. Food Folio by Columbia Center for Eating Disorders is a publicly available set of 138 images of Western food items. The set was developed for the study of eating disorders, particularly for use in tasks that capture eating behavior characteristic of these illnesses. It contains foods that are typically eaten, as well as those typically avoided, by individuals with eating disorders. Each image has now been rated across 17 different attributes by a large general United States population sample via Amazon's Mechanical Turk (n = 1054). Ratings included subjective attributes (e.g., tastiness, healthiness, and favorable texture) as well as estimates of nutrient content (e.g., fat and carbohydrate). Each participant rated a subset of stimulus set food items (46 foods) on all 17 dimensions. Additional description of the image set is provided in terms of physical image information and accurate nutritional information. Correlations between subjective ratings were calculated and an exploratory factor analysis and exploratory cluster analysis completed. Outcomes of the factor analysis suggested foods may be described along three latent factors of healthiness, tastiness, and umami taste; the cluster analysis highlighted five distinct clusters of foods varying on these same dimensions. Descriptive outcomes indicated that the stimulus set includes a range of foods that vary along multiple dimensions and thus is likely to be useful in addressing various research questions surrounding eating behavior and cognition in healthy populations, as well as in those with eating disorders. The provision of comprehensive descriptive information allows for stimulus selection that is optimized for a given research question and promotes strong inference.

Development of MacroPics: A novel food picture set to dissociate the effects of carbohydrate and fat on eating behaviors

Emerging evidence suggests that fat and carbohydrate interact to potentiate the reward value of food (DiFeliceantonio et al., 2018). The primary goal of the current study was to develop a novel picture set to facilitate research into the effects of macronutrient composition on food choice and eating behavior. Toward this aim, we developed "MacroPics." In Experiment 1, we photographed 120-kcal portions of 60 snack foods falling into one of the three macronutrient categories: (1) mostly carbohydrate, (2) mostly fat, or (3) a combination of fat and carbohydrate. Sixty-one participants rated the images for liking, familiarity, frequency of consumption, healthiness, estimated energy content (in kcal), and expected satiation. A subset of these images consisting of 36 items was then selected in an iterative process to minimize differences in ratings between the macronutrient categories while simultaneously ensuring similar within-category variability on a number of food characteristics (e.g., energy density, portion size, retail price) and visual properties (e.g., color, complexity, visual area). In Experiment 2, an independent sample of 67 participants rated the pictures of the final 36-item MacroPics. Both experiments reveal similar participant ratings across categories for item liking, familiarity, frequency, healthiness, and estimated energy content. Protein content was higher in the fat compared to the carbohydrate and combination categories, leading to higher ratings of estimated satiety and energy density for fatty foods. Item and macronutrient category characteristics of the final MacroPics set are reported.

Neural responses to food cues in middle to older aged adults: a scoping review of fMRI studies

AIM

Understanding neural responses through functional magnetic resonance imaging (fMRI) to food and food cues in middle-older adults may lead to better treatment options to address the growing issue of malnutrition. This scoping review aimed to determine the extent, range and nature of research using fMRI, related to reward-based regions, in response to food cues in middle to older aged adults (50 years and over).

METHODS

The following databases were systematically searched in July 2019: CINAHL, CENTRAL, Embase, Dissertations and Theses, Ovid Medline, PsycINFO, PsycEXTRA, Scopus and Web of Science. Studies were eligible for inclusion if participants had a mean or median age ≥50 years, utilised and reported outcomes of either a food cue task-related fMRI methodology or resting-state fMRI. Data from included studies were charted, and synthesised narratively.

RESULTS

Twenty-two studies were included. Eighteen studies utilised a task-related design to measure neural activation, two studies measured resting state neural connectivity only and an additional two studies measured both. The fMRI scanning paradigms, food cue tools and procedure of presentation varied markedly. Four studies compared the neural responses to food between younger and older adults, providing no consensus on neural age-related changes to food cues; two studies utilised longitudinal scans.

CONCLUSION

This review identified significant extent, range and nature in the approaches used to assess neuronal activity in response to food cues in adults aged 50 years and over. Future studies are needed to understand the age-related appetite changes whilst considering personal preferences for food cues.

Development and Pilot Testing of Standardized Food Images for Studying Eating Behaviors in Children

Food images are routinely used to investigate the cognitive and neurobiological mechanisms of eating behaviors, but there is a lack of standardized image sets for use in children, which limits cross-study comparisons. To address this gap, we developed a set of age-appropriate images that included 30 high-energy-dense (ED) foods (>2.00 kcal/g), 30 low-ED foods (<1.75 kcal/g), and 30 office supplies photographed in two amounts (i.e., "larger" and "smaller"). Preliminary testing was conducted with children (6-10 years) to assess recognition, emotional valence (1 = very sad, 5 = very happy), and excitability (1 = very bored, 5 = very excited). After the initial testing, 10 images with low recognition were replaced; thus, differences between Image Set 1 and Image Set 2 were analyzed. Thirty (n = 30, mean age 8.3 ± 1.2 years) children rated Set 1, and a different cohort of 29 children (mean age 8.1 ± 1.1 years) rated Set 2. Changes made between image sets improved recognition of low-ED foods (Set 1 = 88.3 ± 10.5% vs. Set 2 = 95.6 ± 10.6%; p < 0.0001) and office supplies (83.7 ± 10.5 vs. 93.0 ± 10.6%; p < 0.0001). For the revised image set, children recognized more high-ED foods (98.4 ± 10.6%) than low-ED foods (95.6 ± 10.6%; p < 0.05) and office supplies (93.0 ± 10.6%; p < 0.0001). Recognition also improved with age (p < 0.001). Excitability and emotional valence scores were greater for high-ED foods compared with both low-ED foods and office supplies (p < 0.0001 for both). However, child fullness ratings influenced the relationship between excitability/emotional valence and category of item (p < 0.002). At the lowest fullness level, high-ED foods were rated the highest in both excitability and emotional valence, followed by low-ED foods and then office supplies. At the highest fullness level, high-ED foods remained the highest in excitability and emotional valence, but ratings for low-ED foods and office supplies were not different. This suggests that low-ED foods were more exciting and emotionally salient (relative to office supplies) when children were hungry. Ratings of recognition, excitability, and emotional valence did not differ by image amount. This new, freely available, image set showed high recognition and expected differences between image category for emotional valence and excitability. When investigating children's responsiveness to food cues, specifically energy density, it is essential for investigators to account for potential influences of child age and satiety level.

Fat preference and fat intake in individuals with and without anorexia nervosa

Fat restriction is a characteristic eating behavior among individuals with anorexia nervosa (AN), and laboratory meal studies demonstrate restricted fat intake among low-weight patients. The Geiselman Food Preference Questionnaire-I^© (FPQ) is a validated self-report measure that yields a fat preference score (FPS). Prior research reported that patients with AN had a significantly lower FPS than did healthy control (HC) participants. The goal of the current study was to compare self-reported fat preference (FPS) to fat intake (multi-item meal (MIM) study) in low-weight ANs and HCs. Specific aims were 1) to determine if the FPS differed between ANs and HCs; 2) to determine if fat and energy intakes differed between ANs and HCs; and 3) to determine if the FPS was associated with fat and energy intakes in ANs and HCs. Forty-four female AN inpatients and 48 female HCs completed the FPQ and participated in a MIM study. Compared to HCs, ANs consumed less energy (469.1 ± 397.7 vs. 856.4 ± 346.8 kcal, p < 0.001), less fat (16.4 ± 20.4 vs. 36.7 ± 18.9 g, p < 0.001), and a smaller percentage of calories from fat (22.9 ± 13.8 vs. 36.6 ± 8.0%, p < 0.001) at the MIM. Compared to HCs, ANs also had a lower FPS (79.7 ± 27.4 vs. 102.3 ± 18.9, p < 0.001). The FPS was significantly and positively correlated with caloric intake (r = 0.481, p < 0.01), total fat (r = 0.453, p < 0.01), and the percentage of calories from fat (r = 0.37, p < 0.05) in ANs as well as in HCs (kcal: r = 0.583, p < 0.001; fat: r = 0.621, p < 0.001; % fat kcal: r = 0.601, p < 0.001). The FPS is related to objective measures of energy and fat intake in patients with AN as well as in healthy individuals.

Food-Pics_Extended-An Image Database for Experimental Research on Eating and Appetite: Additional Images, Normative Ratings and an Updated Review

Our current environment is characterized by the omnipresence of food cues. The taste and smell of real foods—but also graphical depictions of appetizing foods—can guide our eating behavior, for example, by eliciting food craving and anticipatory cephalic phase responses. To facilitate research into this so-called cue reactivity, several groups have compiled standardized food image sets. Yet, selecting the best subset of images for a specific research question can be difficult as images and image sets vary along several dimensions. In the present report, we review the strengths and weaknesses of popular food image sets to guide researchers during stimulus selection. Furthermore, we present a recent extension of our previously published database food-pics, which comprises an additional 328 food images from different countries to increase cross-cultural applicability. This food-pics_extended stimulus database, thus, encompasses and replaces food-pics. Normative data from a predominantly German-speaking sample are again presented as well as updated calculations of image characteristics.