Gatsby Computational Neuroscience Unit
Alexandra House, 17 Queen Square, LONDON, WC1N 3AR, UK, Tel: +44 (0) 20 7679 1176, Fax +44 (0) 20 7679 1173
B10 Seminar Room, Alexandra House, 17 Queen Square
London, WC1N 3AR
Supported by The Gatsby Foundation
Attention, Behavioral Flexibility and Decision: Functions of the Noradrenergic Locus Coeruleus System in Monkey
University of Pennsylvania, USA
Previous work indicates that the noradrenergic locus coeruleus (LC) system plays an important role in regulation of sleep and waking. We find activity patterns of these neurons in monkeys that indicate additional roles during the waking state. During good performance in a target detection task, LC neurons are selectively activated by target cues and not by other task events, and tonic firing rates were moderate (~ 1-2 spikes/sec). In contrast, high levels of tonic LC activity were associated with reduced LC phasic responses, and poor task performance. These patters of LC activity are termed the phasic and tonic modes, respectively. Computational modeling revealed that electrotonic coupling among LC neurons can provide a mechanism for producing these modes of LC activity, which may in turn regulate task performance. In our model, high electrotonic coupling results in the phasic mode (moderate tonic LC activity and robust phasic responses to task-defined target stimuli). This pattern of LC activity, in turn, facilitates the processing of target stimuli while responses to distractors are reduced. In contrast, low coupling results in the tonic LC mode (high tonic LC activity and reduced phasic responding), and consequently increased behavioral responding to distractors (e.g., false alarms). This leads us to hypothesize that the LC may promote the processing of task-defined stimuli ("selective attention") in the phasic mode, versus a broader sampling of stimuli in the environment (high behavioral flexibility) in the tonic mode. Inputs to the LC from the orbital and cingulate cortices were identified in monkey using tract-tracing. Such inputs are likely sources for the selective LC responses to target stimuli, and for inputs that regulate electrotonic coupling among LC cells. Additional recent studies of LC activity in a two alternative forced choice task reveal that phasic responses of LC neurons are more closely linked to behavioral responding than to stimulus presentation, and precede behavioral responses by about 230 msec. Responses did not occur for stimuli that the animal foveated but failed to respond to, nor for lever responses that occurred outside of the task context. These findings indicate that phasic LC responses are driven by decision processes, and we propose that they serve to facilitate ensuring behavioral responses. These findings extend our understanding of the phasic LC mode in selective attention, indicating a role in selective response facilitation rather than regulating sensory processes.