"First, an escape response [to electric shock] was set up in two groups of rats under conditions that prevented, insofar as possible, its conditioning to environmental cues. Then, a buzzer was paired with shock under conditions that prevented the animals from making any consistent escape response. The rats were then tested to see if they would make the conditioned escape response to the buzzer without being shocked. The test scores of one group were 70 percent positive, and of the other 84.5 percent. The control groups were trained and tested in the same way except that in no case was the buzzer ever paired with shock. Instead of buzzer-shock trials two groups received shock only; one, buzzer only; and one, buzzer or shock in random order. The percentages of positive test trials were 9.0 and 34.2 for shock-only, 10.8 for buzzer-only, and 13.7 for random buzzer or shock. These figures are all significantly less than those of the corresponding experimental groups. Further results, such as difference in response latencies on test trials, tendencies to extinction, and observational records, indicate that the buzzer had more drive power for the experimental than for the control groups." (PsycINFO Database Record (c) 2006 APA, all rights reserved)