Retraining of automatic action tendencies in individuals with obesity: A randomized controlled trial.
Level 2 - randomized trial
Individual randomized controlled trial.
PubMed 29596869 · doi:10.1016/j.appet.2018.03.016
What was done
A randomized controlled trial evaluated cognitive approach-avoidance training in 189 patients with obesity admitted to a psychosomatic inpatient clinic while participating in a nutrition advice program. The active group trained to push a joystick away in response to unhealthy food images (avoidance) and pull it toward themselves for positive images (approach). The control group received sham training with mixed approach-avoidance actions across both picture categories. Outcomes assessed before and after training included approach-avoidance bias, body mass index (BMI), eating pathology questionnaires, and food-specific implicit associations via a Single-Target Implicit Association Test (ST-IAT).
What was found
The abstract provides no numerical figures or p-values. The active training group showed improved approach-avoidance bias relative to sham controls, an effect that generalized to untrained stimuli. However, the intervention produced no significant changes in food-specific ST-IAT scores, eating pathology questionnaires, or BMI.
Why it matters
While computerized cognitive retraining can successfully alter specific motor-action tendencies toward food cues in a laboratory task, these changes do not readily translate into clinical improvements in eating behavior or weight.
Limits
The study evaluated an inpatient sample undergoing concurrent nutritional counseling, which limits generalizability to free-living populations. The abstract omits exact numerical results, intervention duration, and long-term follow-up beyond the post-training assessment. The intervention failed to impact primary clinical endpoints such as eating pathology or BMI.