Recordings from noradrenergic locus coeruleus (LC) neurons in behaving rats and monkeys revealed that these cells decrease tonic discharge during sleep and also during certain high arousal behaviors (grooming and consumption) when attention (vigilance) was low. Sensory stimuli of many modalities phasically activated LC neurons. Response magnitudes varied with vigilance, similar to results for tonic activity. The most effective and reliable stimuli for eliciting LC responses were those that disrupted behavior and evoked orienting responses. Similar results were observed in behaving monkeys except that more intense stimuli were required for LC responses. Our more recent studies have examined LC activity in monkeys performing an "oddball" visual discrimination task. Monkeys were trained to release a lever after a target cue light that occurred randomly on 10% of trials; animals had to withhold responding during non-target cues. LC neurons selectively responded to the target cues during this task. During reversal training, LC neurons lost their response to the previous target cue and began responding to the new target light in parallel with behavioral reversal. Cortical event-related potentials were elicited in this task selectively by the same stimuli that evoked LC responses. Injections of lidocaine, GABA, or a synaptic decoupling solution into the nucleus paragigantocellularis in the rostral ventrolateral medulla, the major afferent to LC, eliminated responses of LC neurons to sciatic nerve stimulation or foot- or tail-pinch. This indicates that certain sensory information is relayed to LC through the excitatory amino acid (EAA) input from the ventrolateral medulla. The effect of prefrontal cortex (PFC) activation on LC neurons was examined in anesthetized rats. Single pulse PFC stimulation had no pronounced effect on LC neurons, consistent with our findings that this area does not innervate the LC nucleus. However, trains of PFC stimulation substantially activated most LC neurons. Thus, projections from the PFC may activate LC indirectly or through distal dendrites, suggesting a circuit whereby complex stimuli may influence LC neurons. The above results, in view of previous findings for postsynaptic effects of norepinephrine, are interpreted to reveal a role for the LC system in regulating attentional state or vigilance. The roles of major inputs to LC from the ventrolateral and dorsomedial medulla in sympathetic control and behavioral orienting responses, respectively, are integrated into this view of the LC system. It is proposed that the LC provides the cognitive complement to sympathetic function.