Burst and tonic response modes in thalamic neurons during sleep and wakefulness.

Thalamic neurons can exhibit two distinct firing modes: tonic and burst. In the lateral geniculate nucleus (LGN), the tonic mode appears as a relatively faithful relay of visual information from retina to cortex. The function of the burst mode is less understood. Its prevalence during slow-wave sleep (SWS) and linkage to synchronous cortical electroencephalogram (EEG) suggest that it has an important role during this form of sleep. Although not nearly as common, bursting can also occur during wakefulness. The goal of this study was to identify conditions that affect burst probability, and to compare burst incidence during sleeping and waking. LGN neurons are extraordinarily heterogenous in the degree to which they burst, during both sleeping and waking. Some LGN neurons never burst under any conditions during wakefulness, and several never burst during slow-wave sleep. During wakefulness, <1% of action potentials were associated with bursting, whereas during sleep this fraction jumps to 18%. Although bursting was most common during slow-wave sleep, more than 50% of the bursting originated from 14% of the LGN cells. Bursting during sleep was largely restricted to episodes lasting 1-5 s, with approximately 47% of these episodes being rhythmic and in the delta frequency range (0.5-4 Hz). In wakefulness, although visual stimulation accounted for the greatest number of bursts, it was still a small fraction of the total response (4%, 742 bursts/17,744 cycles in 93 cells). We identified two variables that appeared to influence burst probability: size of the visual stimuli used to elicit responses and behavioral state. Increased stimulus size increased burst probability. We attribute this to the increased influence large stimuli have on a cell's inhibitory mechanisms. As with sleep, a large fraction of bursting originated from a small number of cells. During visual stimulation, 50% of bursting was generated by 9% of neurons. Increased vigilance was negatively correlated with burst probability. Visual stimuli presented during active fixation (i.e., when the animal must fixate on an overt fixation point) were less likely to produce bursting, than when the same visual stimuli were presented but no fixation point present ("passive" fixation). Such observations suggest that even brief departures from attentive states can hyperpolarize neurons sufficiently to de-inactivate the burst mechanism. Our results provide a new view of the temporal structure of bursting during slow-wave sleep; one that supports episodic rhythmic activity in the intact animal. In addition, because bursting could be tied to specific conditions within wakefulness, we suggest that bursting has a specific function within that state.