Reprint of "Methodological and reporting quality in laboratory studies of human eating behavior"

Measuring food-related inhibition with go/no-go tasks: Critical considerations for experimental design

The use of go/no-go tasks to assess inhibitory control over food stimuli is becoming increasingly popular. However, the wide variability in the design of these tasks makes it difficult to fully leverage their results. The goal of this commentary was to provide researchers with crucial aspects to consider when designing food-related go/no-go experiments. We examined 76 studies that used food-themed go/no-go tasks and extracted characteristics related to participant population, methodology, and analysis. Based on our observations of common issues that can influence study conclusions, we stress the importance for researchers to design an appropriate control condition and match stimuli between experimental conditions in terms of emotional and physical properties. We also emphasize that stimuli should be tailored to the participants under study, whether at the individual or group level. To ensure that the task truly measures inhibitory abilities, researchers should promote the establishment of a prepotent response pattern by presenting more go than no-go trials and by using short trials. Researchers should also pre-specify the criteria used to identify potentially invalid data. While go/no-go tasks represent valuable tools for studying food cognition, researchers should choose task parameters carefully and justify their methodological and analytical decisions in order to ensure the validity of results and promote best practices in food-related inhibition research.

Predicting Food Intake from Food Reward and Biometric Responses to Food Cues in Adults with Normal Weight Using Machine Learning

BACKGROUND

Eating behaviors are determined by a complex interplay between behavioral and physiologic signaling occurring before, during, and after eating.

OBJECTIVES

The aim was to explore how selected behavioral and physiologic variables separately and grouped together predicted intake of 8 different foods.

METHODS

One hundred adults with normal weight performed a food preference task combined with biometric measurements (the Steno Biometric Food Preference Task) in the fasting state. The task measured food reward as well as biometric (eye tracking, electrodermal activity, and facial expressions) responses to images of foods varying in fat content and taste. Energy intake from an ad libitum buffet of the same 8 foods as assessed in the preference task was subsequently assessed. A mixed-effects random forest approach was applied to explore how individual and combined measures of food reward and biometric responses predicted energy intake of the 8 single foods. The performance of the different prediction models was compared with the predictions from a linear model including only an intercept (naïve model) using bootstrap cross-validation.

RESULTS

Participants had a median [IQR] intake of 369 kJ [126-472 kJ] per food. Combined or separate measures of food reward or biometric responses did not predict energy intake better than the naïve model.

CONCLUSIONS

We did not find that the reward or biometric responses to food cues assessed in a clinical setting were useful in predicting energy intake of single foods. However, this study provides a framework in the field of behavioral nutrition for applying machine learning with a focus on individual predictions. This is necessary on the road toward personalized nutrition and provides great potential for handling complex data with multiple variables.This trial was registered at clinicaltrials.gov as NCT03986619.

Measuring Ostracism-Induced Changes in Consumption of Palatable Food: Feasibility of a Novel Behavioral Task

Purpose

Ostracism is a highly aversive interpersonal experience. Previous research suggests that it can increase consumption of highly palatable food in some individuals, but decrease it in others. Thus, we developed the Cyberball-Milkshake Task (CMT), to facilitate research investigating individual differences in ostracism's effects on consumption of highly palatable food. We present data on feasibility for the CMT in a sample of young adult women.

Materials and Methods

Participants were 22 women, 18-30 years old, reporting very low or very high levels of emotional eating at screening. Participants performed the CMT, which consisted of 12 trials. Each trial included: playing a round of Cyberball (a computerized game of catch with fictitious "other participants" programmed to either include or exclude the participant); viewing a chocolate image; and then consuming a participant-determined amount of milkshake. Participants subsequently played an additional inclusion and exclusion round of Cyberball, each immediately followed by questionnaires assessing current mood and recent Cyberball experience.

Results

Cyberball exclusion (vs. inclusion) was associated with large, significant increases in reported ostracism and threats to self-esteem; exclusion's effects on affect were in the expected direction (e.g., increased negative affect), but generally small and non-significant. Milkshake intake was measurable for 95% of participants, on 96% of trials. Intake decreased quadratically across trials, with a steep negative slope for low trial numbers that decreased to the point of being flat for the highest trial numbers.

Discussion

The CMT is a generally feasible approach to investigating ostracism's effects on consumption of highly palatable food. The feasibility (and validity) of the CMT may benefit from modification (e.g., fewer trials and longer rounds of Cyberball). Future research should examine whether performance on a modified version of the CMT predicts real-world behavior in a larger sample.

Interoception and obesity: a systematic review and meta-analysis of the relationship between interoception and BMI

BACKGROUND

Interoception refers to the processes by which we sense, interpret and integrate signals originating from within the body. Deficits in interoception have been linked to higher BMI and may contribute to weight gain. However, there have been conflicting findings and it is not clear how higher BMI is associated with different facets of interoception, such as interoceptive accuracy (the ability to detect internal signals) and sensibility (the tendency to attend to internal signals).

METHODS

We conducted a systematic review and meta-analysis of studies that measured interoception and BMI. We examined relationships between interoception and BMI in children and adults separately and as a function of interoceptive facet and measure. In sensitivity analyses, we tested for evidence of publication bias and whether the results were consistent when studies with a high risk of bias were excluded.

RESULTS

A total of 87 articles were eligible for inclusion. In adults (121 effects, 10,425 participants), there was cross-sectional evidence of higher BMI being associated with overall deficits in interoception (r = -0.054, 95% CI: -0.084 to -0.025) and this was consistent across sensitivity analyses. There was no statistically significant evidence of moderation by interoceptive facet or measure, although there was some variability in effect size estimates based on interoceptive facet and measures. A smaller meta-analysis limited to studies that compared participants with normal weight vs. overweight/obesity indicated poorer interoception in participants with overweight/obesity (SMD = -0.39, 95% CI -0.60 to -0.18).

CONCLUSIONS

In cross-sectional studies, deficits in interoception are associated with higher BMI. However, it remains unclear whether deficits in interoception contribute to or are a consequence of weight gain and obesity.

Testing an online measure of portion size selection: a pilot study concerned with the measurement of ideal portion size

BACKGROUND

Portion size is known to be a key driver of food intake. As consumed portions are often pre-planned, 'ideal portion size'-an individual's preferred meal size selected prior to eating-has been identified as a strong predictor of actual consumption. However, assessments of ideal portion size have predominantly relied on laboratory-based computer tasks, limiting use online. Therefore, this cross-sectional study sought to pilot test the validity of a web-based tool to measure ideal portion size.

METHODS

In an online study (N = 48), participants responded to images of a range of foods. Each food was photographed in a series of different portions and loaded into an 'image carousel' that would allow participants to change the size of the displayed portion by moving a slider left-to-right. Using this image carousel, participants selected their ideal portion size. They also completed measures of expected satiety and expected satiation and self-reported their age and body mass index (BMI). A non-parametric correlation matrix was used to explore associations between ideal portion size and identified predictors of food intake.

RESULTS

Supporting convergent validity of this measure, ideal portion size was significantly correlated with expected satiety (rs = .480) and expected satiation (rs = -.310) after controlling for effects of baseline hunger and fullness, consistent with past research. Similarly, supporting divergent validity of this measure, ideal portion size was not significantly correlated with age (rs = -.032) or BMI (rs = -.111,).

CONCLUSIONS

Pilot results support the validity of this web-based portion size selection tool used to measure ideal portion size, though further research is needed to validate use with comparisons to actual food intake.

The impact of food packaging on measured food intake: A systematic review of experimental, field and naturalistic studies

Chronic, excess energy intake contributes to overweight and obesity. Solutions are needed to guide consumers towards portion control, especially for high energy density (HED), palatable foods. Food packaging, a key element of the eating environment, offers a potential solution. To investigate whether packaging design influences measured food intake, a systematic search was undertaken in four electronic databases (Ovid Medline; Ovid PsycInfo; Ovid Embase and Web of Science) across the previous decade. This process yielded 1671 discrete papers, of which 23 articles containing 40 relevant studies were retrieved. Most (n = 36) of the manipulated packaging features influenced consumption quantity with the largest effect sizes observed for packaging which guided consumers either by on-pack cues or structural features. For example, images on the front of the pack, packaging size, as well as partitioning and resealability all helped to reduce food intake. However, individual differences and attentional focus mediate packaging effects. Overall, packaging features can help to limit intake of HED foods and increase intake of nutrient-dense foods (e.g., carrots). Future studies on packaging for portion control (downsizing) might benefit from long term, randomised control trials to test effects outside of the laboratory context and applied to everyday consumer usage.

The effect of calorie and physical activity equivalent labelling of alcoholic drinks on drinking intentions in participants of higher and lower socioeconomic position: An experimental study

OBJECTIVES

The primary objective of the study was to examine the effect of calorie labelling and physical activity equivalence labelling of alcoholic drinks on drinking intentions in participants of lower and higher socioeconomic position (SEP).

METHODS

Participants (N = 1,084) of higher and lower SEP were recruited into an online study and randomized into one of three drink label conditions; Control (standard alcohol labelling), kcal labelling (standard labelling plus drink kilocalorie information), or kcal + PACE labelling (standard labelling and kilocalorie information, plus information on physical activity needed to compensate for drink calories). After viewing drink labels, participants reported alcohol drinking intentions. Participants also completed measures of alcoholic drink energy content estimation, beliefs about how calorie labelling would affect health behaviour and support for calorie labelling of alcoholic drinks.

RESULTS

kcal labelling (d = 0.31) and kcal + PACE labelling (d = 0.38) conditions had significantly lower drinking intentions compared to the control condition (ps < .001). There was no evidence that effect of labelling condition on drinking intentions was moderated by SEP. A subset of participants also reported that they believed calorie labelling would be likely to positively change their eating and exercise behaviour. Estimates of the energy content of alcoholic drinks tended to be inaccurate and the majority of participants supported the introduction of calorie labelling on alcoholic drinks.

CONCLUSIONS

Calorie labelling of alcoholic drinks resulted in small reductions to intended drinking and testing of the effect calorie labelling has on behaviour in real-world settings is now warranted.

Effect of food variety on intake of a meal: a systematic review and meta-analysis

BACKGROUND

Many studies have shown that food variety-the presence of multiple foods and/or sensory characteristics within and across meals-increases intake. However, studies report mixed findings, and effect size remains unclear.

OBJECTIVES

A systematic review and meta-analysis were conducted to 1) synthesize data across experimental studies that examined effects of variety on total meal intake, relative to a control condition with comparatively less variety; 2) quantify support for this effect; and 3) assist in the identification of important moderating factors (registration: CRD42019153585).

METHODS

In November 2019, we searched the following databases for relevant experimental studies, published in English from 1980, with human participants: PubMed, Cochrane Library, Web of Science, ClinicalTrials.gov, PsycINFO, and OpenGrey. This search was updated in September 2020. Means, standard deviations, and sample sizes were extracted from included articles, and Hedges' g was used to calculate effect sizes. Risk of bias was assessed using the Cochrane Collaboration's tool.

RESULTS

Of 7259 references identified in an initial search, 34 articles consisting of 37 studies contained sufficient information for review, and data from 30 studies (39 comparisons) were included in the meta-analysis. Results from a random-effects model showed a significant small to medium effect of variety on intake (in weight and energy), with greater variety being associated with increased consumption (Hedges' g = 0.405; 95% CI: 0.259, 0.552). However, heterogeneity was considerable across studies (I2 = 84%), and this was unexplained by subgroup analyses based on form of variety, test foods, sensory characteristics, age, sex, and body weight.

CONCLUSIONS

Our findings support the conclusion that variety is a robust driver of food intake. However, risk of bias was high across studies, and this review highlights methodologic limitations of studies. It is recommended that further attention is given to the development of preregistered, well-powered randomized controlled studies in eating behavior research.

From famine to therapeutic weight loss: Hunger, psychological responses, and energy balance-related behaviors

Understanding physiological and behavioral responses to energy imbalances is important for the management of overweight/obesity and undernutrition. Changes in body composition and physiological functions associated with energy imbalances provide the structural and functional context in which to consider psychological and behavioral responses. Compensatory changes in physiology and behavior are more pronounced in response to negative than positive energy balances. The physiological and psychological impact of weight loss (WL) occur on a continuum determined by (i) the degree of energy deficit (ED), (ii) its duration, (iii) body composition at the onset of the energy deficit, and (iv) the psychosocial environment in which it occurs. Therapeutic WL and famine/semistarvation both involve prolonged EDs, which are sometimes similar in magnitude. The key differences are that (i) the body mass index (BMI) of most famine victims is lower at the onset of the ED, (ii) therapeutic WL is intentional and (iii) famines are typically longer in duration (partly due to the voluntary nature of therapeutic WL and disengagement with WL interventions). The changes in psychological outcomes, motivation to eat, and energy intake in therapeutic WL are often modest (bearing in mind the nature of the measures used) and can be difficult to detect but are quantitatively significant over time. As WL progresses, these changes become more marked. It appears that extensive WL beyond 10%-20% in lean individuals has profound effects on body composition and physiological function. At this level of WL, there is a marked erosion of psychological functioning, which appears to run in parallel to WL. Psychological resources dwindle and become increasingly focused on alleviating escalating hunger and food seeking behavior. Functional changes in fat-free mass, characterized by catabolism of skeletal muscle and organs may be involved in the drive to eat associated with semistarvation. Higher levels of body fat mass may act as a buffer to protect fat-free mass, functional integrity and limit compensatory changes in energy balance behaviors. The increase in appetite that accompanies therapeutic WL appears to be very different to the intense and all-consuming drive to eat that occurs during prolonged semistarvation. The mechanisms may also differ but are not well understood, and longitudinal comparisons of the relationship between body structure, function, and behavior in response to differing EDs in those with higher and lower BMIs are currently lacking.

Unpalatable food for thought: Let marketing research guide effective public obesity interventions

The prevalence of obesity is growing unabatedly despite the considerable efforts directed at the problem. Although abundant research has contributed to our understanding of the multifactorial causes of obesity, there is less attention to research that is relevant for guiding social marketers, public health professionals and policymakers in delivering public health interventions for countering and/or preventing the problem of obesity. This review offers six points for identifying and developing research relevant for guiding community-wide obesity interventions based on the idea that an applied marketing research perspective offers a better model for identifying effective interventions than more theoretical academic research. Specifically, the research guiding public health and social marketing interventions needs to (1) provide information on ultimate outcomes (weight, health and unintended consequences) more than intermediate outcomes (beliefs, attitudes and behaviour), (2) report on observations collected over the longer term, (3) use natural settings (even at a cost of internal validity), (4) endeavour to overcome observer-effects, (5) report effect sizes (rather than statistical significance) and (6) use moderator analyses to capture variation in how a population responds to interventions.

Validation of a Deep Learning System for the Full Automation of Bite and Meal Duration Analysis of Experimental Meal Videos

Eating behavior can have an important effect on, and be correlated with, obesity and eating disorders. Eating behavior is usually estimated through self-reporting measures, despite their limitations in reliability, based on ease of collection and analysis. A better and widely used alternative is the objective analysis of eating during meals based on human annotations of in-meal behavioral events (e.g., bites). However, this methodology is time-consuming and often affected by human error, limiting its scalability and cost-effectiveness for large-scale research. To remedy the latter, a novel "Rapid Automatic Bite Detection" (RABiD) algorithm that extracts and processes skeletal features from videos was trained in a video meal dataset (59 individuals; 85 meals; three different foods) to automatically measure meal duration and bites. In these settings, RABiD achieved near perfect agreement between algorithmic and human annotations (Cohen's kappa κ = 0.894; F1-score: 0.948). Moreover, RABiD was used to analyze an independent eating behavior experiment (18 female participants; 45 meals; three different foods) and results showed excellent correlation between algorithmic and human annotations. The analyses revealed that, despite the changes in food (hash vs. meatballs), the total meal duration remained the same, while the number of bites were significantly reduced. Finally, a descriptive meal-progress analysis revealed that different types of food affect bite frequency, although overall bite patterns remain similar (the outcomes were the same for RABiD and manual). Subjects took bites more frequently at the beginning and the end of meals but were slower in-between. On a methodological level, RABiD offers a valid, fully automatic alternative to human meal-video annotations for the experimental analysis of human eating behavior, at a fraction of the cost and the required time, without any loss of information and data fidelity.