Groups of rats were trained on either delayed matching or nonmatching to position tasks, then divided into four subgroups and given the following bilateral lesions: (a) SHAM [vehicle injection into the nucleus basalis magnocellularis (NBM) and dorsal noradrenergic bundle (DNAB)], (b) DNAB (6-hydroxydopamine lesion of the DNAB, vehicle into the NBM), (c) NBM (quisqualic acid lesion of the NBM, vehicle into the DNAB) and (d) DUAL (neurotoxin lesions of both DNAB and NBM). Following postoperative recovery, the DUAL lesion subjects were slightly impaired, but by the seventh day of testing all groups were performing at similar levels. This strongly suggests that quisqualate lesions of the NBM are not sufficient to produce severe and lasting mnemonic disorders resembling those seen in Alzheimer's disease (AD). These data also indicate that the noradrenergic system may not be of critical importance with respect to cognition. It was reasoned that an additional anticholinergic treatment might exacerbate an underlying deficiency. All groups were injected, peripherally, with the cholinergic antagonist scopolamine (0-0.5 mg/kg). This drug dose-dependently disrupted performance in all groups. Moreover, the highest dose had a marked effect in the DUAL group, impairing performance even when no mnemonic burden was present (at zero delay). The results suggest that cholinergic NBM and noradrenergic DNAB lesions produce only transient mnemonic deficiencies. A combination of the two can be disruptive, but longer term task (or reference) memory is the primary process affected, and only under certain conditions. The implication of these findings to research concerning animal models relating to Alzheimer's disease is discussed.