Incidental memory for three types of cartoons was compared: original cartoo
ns, literal translations of the originals, and weird cartoons created by in
serting incongruous material into the literal translations. In Experiment 1
, the three types of cartoons were mixed together in lists. In Experiment 2
, each list contained only two cartoon types. In both experiments, original
cartoons were remembered better than the literal and the weird cartoons, w
hereas the literal and the weird cartoons were equally well remembered. Til
e detection of incongruities, or attempts to resolve those incongruities, c
annot adequately explain the observed humor effects. The results were also
inconsistent with both rehearsal and distinctiveness interpretations. Rathe
r, humor per se appears to support good memory performance. perhaps partici
pants elaborated or gave sustained attention to humorous material at the ex
pense of less humorous material.