Figure 1.

Noradrenaline blockade diminished information gathering. A, Starting from a fully covered deck of cards, subjects were allowed to open as many cards until they felt “certain enough” to indicate whether they believed the majority of the 25 cards was yellow or blue. In a “fixed” condition, no external costs for sampling applied, but in the “decreasing” condition, a potential win of 250 points reduced by 10 points per uncovered card. B, Information gathering is decreased consequent on noradrenaline blockade (propranolol), but not consequent on a dopamine perturbation (amisulpride). This increase in impulsivity is consistently observed across both conditions, rendering the statistics of choices in the noradrenaline group closer to those of an optimal model (green diamonds) in the decreasing condition but further away in the fixed condition. The drug manipulation did not lead to a statistically significant difference in points earned (C) and did not affect choice accuracy (D; the probability of choosing the color that currently forms the majority of cards at time of decision). *p < 0.05. Data are shown as mean ± SEM.