In the acute FN-439 or aCSF condition, the aCSF-treated animals displayed escalated responding on the fourth post-dependence test session (t (4) = 2.18, p<0.05). The mean (±S.E.M) ethanol consumption during the first three post-dependent self-administration sessions (data not graphically shown) for aCSF-treated animals (.61 ± .19, .64 ± .19 and .47 ± .18, respectively) and animals treated with FN-439 (.61 ± .14, .4 ± .11 and .54 ± .2, respectively) showed no significant differences over the sessions when analyzed with a one-way repeated measures ANOVA (F (2, 8) = 0.214, p > 0.05) anF (2, 10) = 0.383, p > 0.05, respectively). Although the three sessions are not significantly different from each other, averaging reduces variability, which could enhance the probability of rejecting the null hypothesis. Therefore, to protect against the commission of a Type I error, in addition to the two-way ANOVA that incorporated the averaged data from post-vapor sessions 1-3, a two-way ANOVA was conducted that only used the final session prior to escalation instead of the averaged data. When the two-way mixed model ANOVA was used to
