Poster Presentation The International Congress of Neuroendocrinology 2014

Anabolic-androgenic steroids impair decision making on a rodent version of the Iowa Gambling Task (#147)

Kathryn G Wallin 1 , Ruth I Wood 2
  1. Neuroscience Graduate Program, University of Southern California, Los Angeles, California, United States
  2. Cell and Neurobiology, Keck School of Medicine, Los Angeles, CA, United States

High-dose anabolic-androgenic steroids (AAS) induce aggression and dependence, but potential effects on cognition are understudied. In this regard, high testosterone levels correlate with financial risk taking, and AAS have been implicated in dopamine dysfunction in prefrontal cortico-striatal circuitry. The present study determined if AAS impair economic decision making. Male Long-Evans rats were treated chronically with testosterone (7.5 mg/kg) or vehicle, and tested for decision-making ability on a rodent version of the Iowa Gambling Task (IGT). This operant task required rats to choose between four levers, analogous to the four decks of cards in the human IGT. The levers were associated with different reward probability, reward magnitude on wins, and time-out duration on losses. Two levers were advantageous (205 pellets/30-minute session); two levers were disadvantageous (122 or 125 pellets). Rats were tested for 22 days, and we compared selection of each lever by testosterone- and vehicle-treated rats over the last 3 days. By RM-ANOVA, there was a significant interaction of drug (testosterone vs vehicle) with lever selection (F3,16=4.67, p<0.05). Testosterone-treated rats chose the two advantageous levers significantly less (14.2±3.5% of trials) than vehicle controls (62.5±10.2% of trials, p<0.05). As a result, testosterone-treated rats earned fewer pellets than vehicle controls (vehicle: 152.8±9.3 pellets, testosterone: 122.2±1.1, p<0.05). Testosterone-treated rats preferred the disadvantageous lever with a large reward magnitude (4 pellets/win) but a high cost of uncertainty and time-outs (percent of responses: vehicle 29.3±10.1%, testosterone 70.5±11.6%; p<0.05). During time-outs, testosterone-treated rats made significantly more responses on the inactivated lever (vehicle 0.05±0.01 responses/second, testosterone 0.18±0.02, p<0.05). This study suggests that testosterone impairs economic decision making by shifting preference to large magnitude rewards (pellets/win) despite high costs, resulting in fewer pellets earned over time. Supported by NIDA RO1-DA029613 to RIW.