We examined the effects of sex steroids on cognitive functioning by exogenously manipulating circulating T levels in a group of healthy young men. Thirty-two men were randomized to receive 8 wk of treatment including: 1) im T enanthate 100 mg/wk plus daily oral placebo (T); 2) im placebo/wk plus 125 μg daily oral levonorgestrel (LNG); 3) im T enanthate 100 mg/wk plus 125 μg daily oral LNG (T + LNG); 4) im placebo/wk plus daily oral placebo. Cognitive functions were assessed at baseline and twice during treatment. Serum T and E2 levels were significantly increased in the T and T + LNG groups compared with baseline (P < 0.01) and T levels were significantly decreased in the LNG group (P < 0.05). Verbal memory significantly decreased in the LNG group (P < 0.01) and was maintained by coadministration of T in the T + LNG group. Divided attention was unaffected in the LNG group but improved significantly in the T + LNG group. In summary, decreased serum T levels induced by LNG or direct effects of the progestin, LNG, adversely affects verbal memory in normal young men. These results suggest that short-term changes in sex steroid levels have effects on cognitive function in healthy young men.