1. This study investigates the behavioral conditions in which dopamine (DA) neurons of substantia nigra and adjoining areas A8 and A10 respond with impulses to visual and auditory trigger stimuli eliciting immediate arm- and eye-movement reactions. 2. In a formal task, the rapid opening of the door of a small, food-containing box located at eye level ahead of the animal served as visible and audible trigger stimulus. Most DA neurons on the contralateral side responded to this stimulus with a short burst of impulses with median onset latency of 50 ms and duration of 90 ms (75% of 164 neurons). Similar responses were seen in a comparable fraction of DA neurons during ipsilateral task performance, suggesting that responses were not specific for the limb being used. 3. When the sensory components of the door opening stimulus were separated, DA neurons typically responded in a similar manner to the moving visual stimulus of the opening door, the low-intensity sliding noise of the opening door, and the 1-kHz sound of 90-92 dB intensity emitted from a distant source at the onset of door opening. Responses to each component alone were lower in magnitude than to all three together. 4. In a variation of the task, a neighboring, identical food box opened in random alternation with the other box but without permitting animals to reach out (asymmetric, direct-reaction go/no-go task). With each sensory component, DA neurons typically responded both to opening of go and no-go boxes. Responses were enhanced when stimuli elicited limb movements in go trials. 5. Monkeys reacted to door opening with target-directed saccadic eye movements in the majority of both go and no-go trials. Neuronal responses were equally present during the occasional absence of eye movements. Thus responses were not specific for the initiation of individual arm or eye movements. 6. Neuronal responses were absent when the same stimuli occurred outside of the behavioral task with target-direct arm and eye movements lacking. This shows that responses were not of purely sensory nature but were related to the capacity of the stimulus for eliciting behavioral reactions. 7. In a variation of the go/no-go task, an instruction light illuminated 2-3 s before door opening prepared the animal to perform the reaching movement on door opening or to refrain from moving (asymmetric, instruction-dependent go/no-go task). Phasic neuronal responses to both instruction lights occurred in a limited number of DA neurons in one monkey when the task was used infrequently, largely disappearing with more regular testing. Responses to instructions resembled those to door opening in terms of latency, duration, and magnitude. This suggest that DA neurons may also respond to novel stimuli with properties uncertain to the animal. 8. In a different reaching task, the offset of a distant sound stimulus was used for eliciting arm and eye movements toward the food box. None of the 16 DA neurons tested in one monkey showed responses to this trigger signal, although most of them responded to door opening in separate sessions. Thus responses were related to the appearance of a stimulus and not only to its capacity to elicit behavioral reactions. 9. These data suggest that the large majority of DA neurons respond in a similar and stereotyped manner to salient visual and auditory stimuli with both appetitive properties and the capacity for eliciting immediate behavioral reactions. A common factor contributing to these responses appears to be motivational arousal. As stimuli effective for driving DA neurons induce an expectancy for obtaining an object of high interest, DA neurons would be able to participate in the setting of a state during which behavioral reactions occur. In this way, they would be engaged in the initiation process without triggering each individual reaction.
