The authors considered compliance with a decision aid that E. Thorp (1966) designed to minimize loss in a gambling paradigm under different levels of risk or impairment. Twenty adult men (aged 1846) completed the South Oaks Gambling Screen (SOGS; H.R. Lesieur & S. B. Blume, 1987) and the Alcohol Use Disorders Identification Test (AUDIT; J. P. Allen. D. F. Reinert, & R. J. Volk, 2001) and then played a computer blackjack program before and after ingesting alcohol. The decision aid (online Basic advice) increased players' compliance with optimal play and also increased players' willingness to wager more at high stakes. Participants attained a mean peak blood alcohol concentration (BAC) of ...

<.05, [[eta].sup.2] = .246. Participants made more total errors according to Basic before they consumed alcohol (M = 26.7%) than after they did (M = 23.3%). There are two main classes of errors according to Basic strategy. A Failure to Hit occurs when the dealer has a greater likelihood of obtaining a better total than does the player. Such errors are typically associated with player totals of 14-16, when the dealer is showing a total between 7 and 11 (Wagenaar, 1988). A Failure to Stand occurs when the dealer has a greater likelihood of busting. Such errors are rarer during casino play and are typically associated with dealer totals between 2 and 6, when the player has a total between 12 and 14. The percentages of Failure to Hit errors (7.3%), F(1, 19) = 1.510, p>