Elsevier

Psychiatry Research

Volume 219, Issue 1, 30 September 2014, Pages 166-170
Psychiatry Research

Probing behavioral responses to food: Development of a food-specific go/no-go task

https://doi.org/10.1016/j.psychres.2014.04.053Get rights and content

Abstract

The ability to exert self-control in the face of appetitive, alluring cues is a critical component of healthy development. The development of behavioral measures that use disease-relevant stimuli can greatly improve our understanding of cue-specific impairments in self-control. To produce such a tool relevant to the study of eating and weight disorders, we modified the traditional go/no-go task to include food and non-food targets. To confirm that performance on this new task was consistent with other go/no-go tasks, it was given to 147 healthy, normal weight volunteers between the ages of 5 and 30. High-resolution photos of food or toys were used as the target and nontarget stimuli. Consistent with expectations, overall improvements in accuracy were seen from childhood to adulthood. Participants responded more quickly and made more commission errors to food cues compared to nonfood cues (F(1,140)=21.76, P<0.001), although no behavioral differences were seen between low- and high-calorie food cues for this non-obese, healthy developmental sample. This novel food-specific go/no-go task may be used to track the development of self-control in the context of food cues and to evaluate deviations or deficits in the development of this ability in individuals at risk for eating problem behaviors and disorders.

Introduction

Self-control, or resistance to temptation, has been studied in the context of social, developmental and cognitive psychology. This ability can be operationally defined as the capacity to accomplish goal directed behavior in the face of salient, competing inputs and actions (Casey et al., 2005). A prominent component of self-control is the ability to suppress inappropriate behaviors in favor of appropriate ones, often termed impulse control (Casey et al., 1997, Rothbart et al., 2001, Spinrad et al., 2007), and can be measured by a number of self-report assessments and behavioral tasks (Mobbs et al., 2008, Somerville et al., 2011, Grose-Fifer et al., 2013). The classic go/no-go task measures the ability to maintain behavioral control in the face of interfering stimuli, by measuring an individual׳s speed and accuracy when instructed to respond to a frequent target (go trial) and withhold response to a rare nontarget (no-go trial). Modifications of the classic go/no-go task (Durston et al., 2003, Hare et al., 2005, Hare et al., 2008, Tottenham et al., 2011) have been successfully used to measure motivationally-driven behavior using different subsets of salient cues, such as emotional faces, in examining emotion regulation across the developmental spectrum (Casey et al., 1997, Casey et al., 2007, Casey et al., 2011). In clinical populations, such as individuals diagnosed with an eating disorder, clinically relevant stimulus sets (i.e. appetitive food cues) may be particularly useful to measure self-control in the context of the disorder. Impulsivity has been associated with food intake among healthy-weight individuals (Lindroos et al., 1997, Hays et al., 2002, Guerrieri et al., 2007, Guerrieri et al., 2008, Savage et al., 2009). While viewing advertising logos for food, healthy-weight children show enhanced recruitment of brain regions associated with impulse control relative to obese children (Bruce et al., 2013). Individuals with bulimia nervosa and binge-eating disorder (Nasser et al., 2004) display greater impulsivity scores on self-report measures compared to healthy controls, and, using fMRI, patients with bulimia fail to engage self-regulatory circuits to the same degree as healthy controls (Marsh et al., 2009). Clinically, patients with anorexia nervosa display a remarkable ability to control their food intake behavior (Mayer et al., 2012) and demonstrate enhanced ability to delay monetary (e.g. non-food) rewards (Steinglass et al., 2012).

When studying eating disorders, it is important to be able to distinguish behavioral differences in food-specific self-control from characteristic developmental differences in overall impulsivity. Eating disorders often first manifest during adolescence, but not uncommonly persist into adulthood, whereas impulse control, gradually improves from childhood to early adulthood. Additionally, subjective sensitivity to appetitive cues in our environment greatly influences our ability to exert self-control. Sensitivity to environmental cues has been shown to differ across development and within populations and can drive differences seen in self-control, where overall impulsivity measures are unable to capture this variability.

Others have modified the traditional go/no-go task to examine impulse control using food cues (Batterink et al., 2010, Mobbs et al., 2011, Jasinska et al., 2012, Loeber et al., 2012, Meule et al., 2012, Loeber et al., 2013); however, the tasks vary in design, and results are inconsistent across studies. Some using vegetables (go target) and desserts (no-go non-target) have found positive associations between commission errors to high-calorie food (dessert) cues and body mass index (BMI) (Batterink et al., 2010), however, the lack of a neutral control condition make it difficult to establish whether the association was due to lower inhibitory control in general or was food-mediated. Others have used food and nonfood words in hungry healthy (Loeber et al., 2013) and obese (Mobbs et al., 2011, Loeber et al., 2012) populations to show diminished inhibitory control in response to food-associated cues. The use of words as targets and nontargets, however, greatly limit the application of these tasks. For example, when food cues are words rather than pictures, developmental differences in reading ability and abstract thought will influence performance. Such tasks would be difficult to use early in development when changes in eating habits begin to emerge. Recently, variants of the go/no-go task using pictures have found rate of commission errors to be associated with aspects of unhealthy eating (specifically, emotional eating) (Jasinska et al., 2012), as well as faster reactions times on a modified go/no-go-‘XY’ attention task when food pictures were presented behind the “go” targets (Meule et al., 2012). These findings are confounded, however, by a lack of neutral comparison condition or homogeneity of food stimuli (only high-calorie desserts), respectively. Without these comparisons, it is difficult to determine the extent to which the decrease in inhibitory control is driven by food and more specifically types of food that are commonly associated with unhealthy eating behaviors.

While these studies exemplify how psychological tasks can be modified to measure impulses relevant to eating behavior, the following study attempts to address potential limitations of existing food go/no-go tasks. We introduce an upgraded food go/no-go task that uses pictures of low- and high-calorie food stimuli and interesting nonfood stimuli to test the specific influence of appetitive food cues on self-control. Based on previous studies (Casey et al., 1997, Durston et al., 2002, Somerville et al., 2011), we hypothesized that overall impulse control would increase across age groups and that participants would exhibit more behavioral interference to the appetitive food cues compared to the neutral nonfood cues, especially to the high-calorie food cues.

Section snippets

Participants

Participants were 147 normal-weight, healthy volunteers (96 females) between the ages of 5 and 30 years. Height was measured in centimeters to the nearest 0.1 cm using a wall-mounted stadiometer (Detecto 3PHTROD-WM), and weight was measured in light clothing on either a beam balance scale (Detecto) or a digital medical scale (HealthOMeter 349KLX) to the nearest 0.1 kg. Participants were asked to remove their shoes for both measurements. BMI was calculated as weight (kg) divided by height (m)

Results

Data from 146 participants (39 children, 49 adolescents and 58 adults) were included in the analyses. Data from one adult were excluded due to a misunderstanding of the instructions, where he only responded to the presence of fruits. Demographic information for each age group is presented in Table 1. There was a main effect of age group in raw BMI scores (F(2,121)=25.49, P<0.001), where post-hoc t-tests revealed that children significantly differed from both adolescents (t(73)=5.65, P<0.001)

Discussion

The current study introduces a task that utilizes both appetitive and neutral cues to examine behavioral inhibition in the face of rewarding food cues relative to nonfood cues. Our food-specific go/no-go task manifests the developmentally-expected age differences in task performance and demonstrates an effect of stimulus type, namely food versus toy targets, on behavior.

Reaction times were expectedly slower in children compared to adolescents and adults, consistent with their stage of brain and

Acknowledgments

The research reported in this article was supported by funding from the Diabetes and Endocrinology Research Center 5P30 DK063608-10 and R56/R01 DK097399, and in part by a Medical Scientist Training Program grant from the National Institute of General Medical Sciences of the National Institutes of Health T32GM007739. We would like to thank Eve Vagg of the Digital Communications Department at the New York State Psychiatric Institute for her amazing photography, Janet Schebendach for her food

References (35)

Cited by (0)

View full text