Specific temperamental and character dimensions measured by the TCI differentiated PPG's from both non-PGs and controls, indicating that the TCI may be useful in identifying social gamblers at risk for pathological gambling behaviors, i.e., persistent extreme/excessive gambling behavior beyond the norm for a representative sample population. In agreement with previous reports (Kim and Grant, 2001), PPGs scored higher on NS (novelty seeking) values than non-PGs and controls. As a personality dimension, NS is related to preference for activities that involve chance. NS correlates highly with the diagnoses of histrionic, antisocial, and borderline personality disorder. High prevalence rates of these personality disorders have been described in gamblers. Subjects with a duration of problem gambling of less than 10 years showed higher HA (harm avoidance) values compared with those with a longer period of gambling. Subjects who reported a duration of problem gambling longer than 20 years showed lower CO (cooperativeness) values compared with the other probable pathological gamblers. Those who reported a family history of pathological gambling and/or substance misuse scored higher on NS and lower on CO compared with other PPGs. Sixty actively gambling subjects (18-65 years old) were recruited for the study. All subjects, including controls, were administered the South Oak Gambling Screen (SOGS), a 20-item questionnaire used to screen for pathological gambling. A cutoff score of five or more on the SOGS indicated that the respondent was likely a probable pathological gambler. Based on SOGS results, 32 subjects were classified as non-PGs and 28 as PPGs. Sixty-five controls were classified as neither non-PGs or as PPGs. 4 tables and 35 references