Experimenters have often employed Visual and/or spatial stimuli to assess r
odent working memory in delayed (non)matching-to-sample or delayed response
paradigms. We present an automated task that uses olfactory stimuli to tes
t working memory in the rat. Rats were trained on a continuous delayed two-
choice working memory task in an operant olfactometer with four different d
elays (3, 30, 60, and 120 sec). The rats required a mean of 12 sessions to
achieve a criterion of 80% correct on two consecutive days with a 3-sec del
ay. Baseline data revealed a significant mnemonic function, with performanc
e decreasing at the longer delays. Despite this significant effect, rats re
ached levels of approximately 76% correct at a retention interval (delay) o
f 120 sec. When a stable baseline level of performance was reached, the eff
ect of the central muscarinic antagonist scopolamine hydrobromide was asses
sed. Scopolamine caused a dose-dependent, delay-independent decrease in per
formance. This result supports previous findings with scopolamine in other
tests of rodent working memory and demonstrates the sensitivity of the two-
choice olfactory task to this robust cognitive effect. A rodent working mem
ory paradigm bearing closer resemblance to common human and nonhuman primat
e tests of working memory than do the often employed delayed response proce
dures [e.g., delayed (non)matching to position] may be highly valuable in t
he study of the biological substrates of short-term/working memory. The pre
sent experiment demonstrates the utility of the two-choice olfactory workin
g memory task, which is sensitive to increasing delays and anti-cholinergic
treatment and is rapidly acquired by rats.