Testing the effect of the directionality of verbal predictions on decision-making with experience and feedback

Verbal probabilities are either positive, and attract attention towards the outcome occurrence (e.g., “a small likelihood”) or negative, and attract attention towards the outcome non-occurrence (e.g., “unlikely”). The effect of directionality on decisions has mostly been evidenced in “context poor” studies, where participants had no other information than a verbal forecast, possibly magnifying the role of directionality. In three experiments, we varied how much information participants receive in a series of decisions in a computer game (shoot/don’t shoot) based on positive and negative predictions about the suspect’s dangerousness. In Experiment 1, participants received visual information and feedback; in Experiment 2, participants either received normal or ambiguous (blurry) visual information; and in Experiment 3, participants received ambiguous information only and no feedback. The results showed that directionality influenced decision-making (shoot/don’t shoot and response time) across experiments, even when relevant information and feedback were provided. However, the effect was larger when decision relevant information was ambiguous. Our results demonstrate that limiting knowledge might have magnified effects observed in past research but also that the effect of directionality is robust in experiential methodologies.