Results found here were somewhat complicated by the fact that we did not have access to 2 complete sets of selected lines maintained in Indianapolis for equivalent numbers of generations; on the other hand, we were able to observe the behavior of a nonselected control line, the HS/Ibg. With respect to DD, the line difference was slightly larger when the bidirectionally selected lines were compared to each other as opposed to the high drinking line versus the progenitor line. This may suggest that the largest response to selection was in the direction of high drinking leading to high impulsivity, rather than low drinking leading to low impulsivity. Overall, however, HAP2 differed from LAP2 with a similar magnitude as HAP1 differed from HS/Ibg. Typically, using a nonselected progenitor stock instead of another LAP line would tend to diminish our ability to detect correlated responses; the fact that we were nonetheless able to observe the DD and RT differences suggests that they are fairly robust.
