Roberts S
Behav Neurosci. 1987 Feb;101(1):120-30. doi: 10.1037//0735-7044.101.1.120.
Gibbs and Ng (1976, 1977) proposed a three-stage model of long-term memory formation on the basis of results from a one-trial-learning passive-avoidance procedure using chicks. Chisquare tests show that much of the evidence for this model (Gibbs & Ng, 1976, 1977, 1979, 1984a, 1984b), involving transient drops in retention and drug effects, is less variable than would be expected by chance. This should reduce belief in the model.