Using TMS to evoke slow waves during sleep May 16, 2007Posted by Johan in Neuroscience, Sleep.
During non-REM sleep, in stages 2-4, the neurons in the cortex fire in near synchrony, at a rate of about one per second. This is known as slow wave sleep, and is thought to be important for various processes, including memory consolidation (see a previous related post).
Simply put, TMS is produced by a very strong electromagnet, capable of sending out controllable pulses. If held close to the head, the pulses depolarize neurons on the surface of the cortex, thus producing a single burst of activation. TMS is most often used to disrupt processing by applying a repetitive pulse, which incapacitates that part of the cortex temporarily. This is a very powerful technique, as it allows the researcher to infer that the brain region has a causal role in the behaviour that is being measured, unlike fMRI and other neuroimaging measures where only a correlation between brain activation and behaviour can be made. However, in this study, TMS was used for a somewhat different purpose.
While the participants were in non-REM sleep, TMS was applied. EEG recordings showed the effect of this on the brain. Each TMS pulse tiggered a single slow wave, which left in its wake a drop in activity to levels similar to what is observed while recovering from sleep deprivation (this according to the authors). In terms of the stages of sleep, the activity while TMS was on was classified by independent raters as stage 4, while the recording when it was off was rated as stage 2.
It’s worth emphasising that the participants were already in non-REM sleep, where slow waves normally occur. The crucial finding in this study is that TMS made it possible to control when these slow waves occur. The same stimulation did not evoke slow waves when the participants were awake.
Unlike the normal slow waves, which can originate from various parts of the cortex, the TMS-induced slow waves all originated from the site of the coil (perhaps unsurprisingly), which was placed over sensorimotor cortex, with the coil almost straight on top of the head (see the red cross in the figure below). The investigators found no other location that could induce slow waves as reliably. However, many parts of the head could not be tested as stimulating them produced muscle movements, so this does not necessarily mean that sensorimotor cortex is crucial in producing spontaneous slow waves. The figure below shows the positioning of the coil, and the EEG activation patterns during evoked and spontaneous slow waves.
Note that while the high rate of stimulation makes the TMS slow waves seem quite unlike the spontaneous slow waves, the individual slow waves that TMS produced (compare the graphs in the red boxes to the graph in the blue box) are quite similar to those of spontaneous slow waves. The figure at the top, showing the average slow waves produced spontaneously or by stimulation, makes this point quite clear.
For anyone who knows how loud TMS can be, you might wonder how the participants managed to sleep through this study. The researchers went through some trouble to ensure that the click that the TMS produces when it discharges was imperceptible: they created an opposing waveform of the click noise that was played to the participants through headphones, thus masking the sound. A foam layer was also used to prevent bone conduction of the sound. Despite these precautions, I’m more than a little impressed by the ability of these participants to sleep soundly, knowing some guys in white coats are about to start sending pulses through your head as soon as you pass out.
More than anything else, this paper is a description of a new method. I’m not sure that we’ve learned a lot about how slow waves are produced or what purpose they serve, but with the method outlined by Massimini et al (2007), future investigators may be able to expand on this. This application of TMS allows you to vary the rate of the slow waves, and measure how this affects the various functions that slow wave sleep is believed to involve. Much like the repetitive-TMS application I outlined above, this means that causal relationships can be inferred.
Massimini, M., Ferrarelli, F., Esser, S.K., Riedner, B.A., Huber, R., Murphy, M., Peterson, M.J., & Tononi, G. (2007). Triggering sleep slow waves by transcranial magnetic stimulation. PNAS, 104, 8496-8501.