Can approach-avoidance instructions influence facial representations? A distinction between past- and future-oriented inferences

Mere instructions about a supposedly upcoming approach/avoidance training (i.e., “you will approach stimulus A and avoid stimulus B”) can influence stimuli evaluation (e.g., stimulus A is evaluated more positively). In this work, we argue that because approach/avoidance instructions are typically future-oriented (e.g., “you will approach stimulus A”), they are less powerful than past-oriented information (e.g., “you approached stimulus A”). We introduce the placebo approach/avoidance training, a procedure implementing past-oriented information that involves a bogus training without actual contingencies between stimuli and approach/avoidance actions. Experiments 1a and 1b revealed an approach/avoidance effect on visual representations when employing placebo training. Experiment 2 showed that the effect of placebo training (involving past-oriented information) is larger than the effect of approach/avoidance instructions (involving future-oriented information). Finally, Experiment 3 replicated the distinct effect of past- vs. future-oriented approach/avoidance information by controlling for the experience of approaching/avoiding stimuli. This work highlights the role of past-oriented thinking in approach/avoidance effects.