Learned fear: a cause of arrhythmia onset in the presence of digitalis.

In earlier studies we have shown that guinea pigs exposed to signal-shock pairs develop digitalis toxicity earlier than control pigs on a test day when shocks are not delivered. Presenting subjects with signal-shock pairs is known to produce learned changes in autonomic tone thought to reflect fear. However, we were unable to find evidence of such changes in that model. A recent report extended our work on psychosomatic digitalis toxicity to the rabbit. Although that animal has been extensively used in studies of visceral learning, that study did not provide sufficient data to rigorously conclude that visceral learning had taken place. In this report, we show that rabbits which have learned that a signal accurately predicts the occurrence of shock develop digitalis-toxic arrhythmias more often during the signal and significantly earlier than other rabbits given prior exposure to equal numbers of signals and shocks, never explicitly paired. The use of this latter control group indicates that rabbits exposed to signal-shock pairs have learned to associate the signal with its consequences; independent evidence of learning exists in the fact that these rabbits showed a signal-locked bradycardia on their training day. However, bradycardia did not appear to be the mechanism for the early elicitation of digitalis toxicity on the test day when ouabain was infused during probes with the signal alone. These data may have clinical significance in their indication that factors in the external milieu can precipitate digitalis-toxic arrhythmias in individuals that would otherwise have no evidence of digitalis toxicity.