Training response inhibition to food is associated with weight loss and reduced energy intake

Abstract

The majority of adults in the UK and US are overweight or obese due to multiple factors including excess energy intake. Training people to inhibit simple motor responses (key presses) to high-energy density food pictures reduces intake in laboratory studies. We examined whether online response inhibition training reduced real-world food consumption and weight in a community sample of adults who were predominantly overweight or obese (N = 83). Participants were allocated in a randomised, double-blind design to receive four 10-min sessions of either active or control go/no-go training in which either high-energy density snack foods (active) or non-food stimuli (control) were associated with no-go signals. Participants' weight, energy intake (calculated from 24-h food diaries), daily snacking frequency and subjective food evaluations were measured for one week pre- and post-intervention. Participants also provided self-reported weight and monthly snacking frequency at pre-intervention screening, and one month and six months after completing the study. Participants in the active relative to control condition showed significant weight loss, reductions in daily energy intake and a reduction in rated liking of high-energy density (no-go) foods from the pre-to post-intervention week. There were no changes in self-reported daily snacking frequency. At longer-term follow-up, the active group showed significant reductions in self-reported weight at six months, whilst both groups reported significantly less snacking at one- and six-months. Excellent rates of adherence (97%) and positive feedback about the training suggest that this intervention is acceptable and has the potential to improve public health by reducing energy intake and overweight.

Keywords: Cognitive training; Disinhibition; Energy intake; Food liking; Response inhibition; Weight loss.

Graphical abstract

Fig. 1

Recruitment flow diagram showing numbers of participants included in each intervention group at each stage of the study (see supplementary methods for details of each sample).

Fig. 2

Schematic of the ‘go’ and ‘no-go’ trials for the food associated response inhibition task (active condition). Healthy foods were always presented on go trials, high-energy density foods always on no-go trials (bold frame) and filler images of clothes were associated with no-go signals 50% of the time.

Fig. 3

Overview of study procedure during the 2-week pre- and post-intervention phase. Participants were also followed-up remotely one month and six months after the final research session.

Fig. 4

Change in measured weight from baseline to post-intervention (a) and change in self-reported weight from screening to follow-up (b) in each inhibition training condition. A negative change indicates weight loss from pre-to post-intervention. Error bars = standard error of the mean (SEM).

Fig. 5

Change in liking ratings from baseline to week 2 as a function of inhibition training condition. Error bars = SEM.

Fig. 6

Change in attractiveness ratings from baseline to week 2 as a function of inhibition training condition. Error bars = SEM.