Figure 1
(A) Three-armed alcohol novelty bandit task schematic. Displays reward and non-reward presentations as well as the novel insertion of an alcohol cue. (B) P3a and P3b averaged across all participants and cue types with highlighted time windows of analysis prior to (pre-insertion) and one trial after novel insertions (post-insertion) with error bars +/– SEM.
Figure 2
(A) Choice behavior averaged across both stimulus types between groups over trials since a novel insertion for the novel, best alternative, and worst alternative stimuli. Note the increase in exploratory behavior on trial 1 after a novel insertion. (B) Probability of selecting the novel stimulus (exploration) minus probability of selecting the best alternative (exploitation) on the post-insertion trial. Hazardous drinkers explored alcohol stimuli more often than controls, and controls explored non-alcohol stimuli more often than hazardous drinkers. (C) BONUS estimates with hazardous drinkers having higher BONUS values for alcohol cues than controls. (D) IEV estimates with no significant differences between groups or cue types.
Figure 3
Probability of explore minus exploit behavior by P3a, contrasted by group and cue type with significant paired comparison marked. For hazardous drinkers but not controls, an increase in the P3a is associated with an increase in the exploration of alcohol cues relative to non-alcohol cues.
Figure 4
BONUS and IEV parameters by P3a amplitude split by group and cue type with significant paired comparisons marked. For hazardous drinkers but not controls, an increase in the P3a is associated with the BONUS values for alcohol cues relative to non-alcohol cues. Additionally in hazardous drinkers but not controls, the P3a decreases as the IEV of alcohol cues increases relative to non-alcohol cues.
Figure 5
BONUS and IEV parameters by P3b amplitude split by group and cue type with significant paired comparison marked. The IEV of alcohol cues increases with P3b amplitude in hazardous drinkers relative to controls.