Background-The recreational drug, MDMA (3,4-methylenedioxymethamphetamine; 'Ecstasy'), is a synthetic amphetamine derivative and a serotonin neurotoxin. MDMA use is associated with cognitive dysfunction and impulsivity, but since polydrug abuse is common among users it is difficult to attribute these problems specifically to MDMA. Moreover, few studies have examined rewardrelated cognitive processes. Our aim was to examine reward-related decision-making and impulsivity among MDMA users while controlling for polydrug use via appropriate comparison groups.Methods-We examined decision-making (Iowa Gambling Task; IGT; Bechara et al., 1994), selfreported impulsivity (Multidimensional Personality Questionnaire -Brief Form [Constraint subscale]; Barratt Impulsiveness Scale; Zuckerman Sensation Seeking Scale), and drug use among 22 abstinent MDMA users, 30 other drug users, and 29 healthy non-drug controls.Results-MDMA and other drug users showed comparable patterns of decision-making and impulsivity. However, both drug groups demonstrated poorer IGT performance and elevated selfreported impulsivity relative to controls. Poorer decision-making was related to heavier drug use in the past year, heavier weekly alcohol use, and meeting lifetime substance use disorder (SUD) criteria for more drug classes. Elevated impulsivity was associated with heavier drug use, heavier weekly alcohol use, more lifetime SUDs, and higher self-reported depression levels.Conclusions-These findings contradict the idea that MDMA is specifically associated with deficient decision-making. Drug users, in general, may be at risk for decision-making deficits and elevated impulsivity. Such behaviors may represent trait factors that lead to the initiation of drug and alcohol use, and/or they may represent behavior patterns that are exacerbated by extensive use.

Insulin receptors are found throughout the brain, particularly in the hippocampus, although the impact of insulin on memory is unclear. The purpose of this study was to examine the effect of insulin on event-related potentials in response to a standard memory task and visual evoked potentials (VEPs) during exposure to a reversing checkerboard. We hypothesized that insulin would decrease P300 magnitude and latency during the presentation of previously observed stimuli, but would have no effect on VEPs. Sixteen humans participated in two euglycemic clamp studies with somatostatin performed in random order in which serum insulin levels were either suppressed (14 ± 1 pmol/l) or increased by insulin infusion (433 ± 40 pmol/l). At steady state, event-related potentials and then VEPs were collected using a 32-electrode cap. The major finding was that the P300 amplitude measured during the identification of an object as old was significantly smaller over parietal regions when insulin was infused than when no insulin was provided. Insulin was without effect on the VEPs. We conclude that insulin has region- and task-specific effects on neuronal activation. While the P300 amplitude measured during the presentation of an old object was reduced during insulin infusion, the hormone was without effect on VEPs.