Tdt workstation

Three experimental conditions were tested: Off-DBS, Traditional DBS (tDBS), and Closed-Loop (CL-DBS). tDBS consisted of a continuous biphasic pulse train (133Hz, 700μA, 80μs/phase, no interphase gap, monopolar C2 cathodic first). The pulse train for CL-DBS was similar but was triggered on only when a real-time measure of beta amplitude in the bipolar LFP activity from contacts 1–3 exceeded a pre-determined threshold level. LFP C1-3 was bandpass filtered (9–20Hz, chosen based on beta peak observed in LFP power spectral density, Fig. 1A), rectified and low-pass filtered (400ms moving average) to produce the beta amplitude signal (Fig. 1B). This signal is referred to as ‘beta’ for simplicity and because the bandpass filter peaked in the beta range, though the filter range does include much of the alpha (8–13Hz) band. CL-DBS included a 250ms ramp up/down when triggered on/off, respectively. This methodology is similar to Little et al. [15 (link)]. Recording, online processing and stimulation were programmed using a TDT workstation (Tucker Davis Technologies) operating at ~25kHz sampling rate. There was a 5 sample point delay (0.205 ms) delay between the detection of a beta threshold crossing and the change in stimulation output. The beta LFP and stimulus signals were saved for analysis (Fig. 1G)