Fig. 5. A, Mean changes in electrochemical signal recorded when the standard rate of milk delivery (0.2 ml/30 sec) was tripled (0.6 ml/30 sec), doubled (0.4 ml/30 sec), halved (0.1 ml/30 sec), or withheld (0.0 ml). Flow rate significantly affected signal levels recorded at 15 sec (F(4,1645) = 16.6162, p < 0.01) and 30 sec (F(4,1645) = 27.6817, p< 0.01) of the milk delivery period. At these two time points, signal levels under the 0.4 and 0.6 ml conditions did not differ, but both were significantly lower than those seen at 0, 0.1, and 0.2 ml (p < 0.01), whereas levels under the 0 ml condition were significantly higher than those at 0.2, 0.4, and 0.6 ml (p < 0.01). B, The magnitude of signal decreases associated with milk consumption did not depend on the rate of milk delivery per se but, rather, on the rate of milk delivery relative to what the animals were trained to expect. The pronounced decrease in signal seen when milk delivered at 0.6 ml/30 sec was consumed by animals trained to expect the standard reward (0.2 ml/30 sec) was significantly attenuated (F(1,393) = 12.1373, p< 0.01) in animals trained to expect milk at 0.6 ml/30 sec.