The results show that the group that received testosterone scored significantly lower than the group that received the placebo, on average answering 20 percent fewer questions correctly. The testosterone group also "gave incorrect answers more quickly, and correct answers more slowly than the placebo group," the authors write. The same effect was not seen in the results of the basic math tests administered to both groups. The results "demonstrate a clear and robust causal effect of [testosterone] on human cognition and decision-making," they conclude.