“…Using a gambling task, Wilder et al (1998) reported that medicated schizophrenic patients are not impaired with respect to implicit rule learning; however, the schizophrenic pa-303 Significance is denoted as * for differences between the classic NL and OLA group (*P<0.05, **P<0.01) and as + for differences between classic NL and control group ( + P<0.05, ++ P<0.01) tients earned much less money than controls in the game, and non-rejection of the null hypothesis may have been due to a small sample size (n=6). In other paradigms of implicit learning, for example word stem completion and lexical decision, schizophrenics were not impaired or even showed facilitated (semantic) priming (Heinrichs and Bury 1991;Schmand et al 1992;Clare et al 1993). Classic NL reduce motor speed, which fact may impede straightforward interpretation of RT data.…”