Learning to sham feed: behavioral adjustments to loss of physiological postingestional stimuli.

The progressive increase in intake of a concentrated (0.8 M sucrose) solution seen when rats are first exposed to the sham-feeding procedure can be prevented by interspersing two real-feeding tests between each sham-feeding test. Under these conditions, sham intake is significantly larger than real intake but significantly smaller than intake on the fifth consecutive sham-feeding test. This result indicates that there is a learned negative-feedback signal based on the association of the taste and postingestive effects of 0.8 M sucrose which extinguishes under consecutive sham-feeding tests. Analysis of the rate of ingestion during the tests revealed that the conditioned negative-feedback signal operates during the first 6 min of a sham-feeding test that follows real-feeding tests. The effect of the absence of an unconditional negative-feedback signal appears from approximately 6 to approximately 20 min during a sham-feeding test.