Neural recordings and spike detection
This protocol is extracted from research article:
The somatosensory cortex receives information about motor output
Sci Adv, Jul 10, 2019; DOI: 10.1126/sciadv.aaw5388

EMG signals were amplified using amplifiers (AB-611J, Nihon Kohden) with a gain of ×1000 to 2000 and were sampled at 2000 Hz in Monkey T and 1000 Hz in Monkey C. Temporal filtering of the signals was carried out with a second-order Butterworth band-pass filter (1.5 to 60 Hz). The signals were rectified and computed in 5-ms bins corresponding to the sampling rates of the motion capture system. A smoothed curve of the signals was then calculated using a moving window process with a window length of 11 bins.

ECoG signals were amplified using a multichannel amplifier (Plexon MAP system, Plexon) with a gain of ×1000 and sampled from each electrode at 2000 Hz in Monkey T and 1000 Hz in Monkey C. Temporal filtering of the signals was carried out with a second-Butterworth band-pass filter (1.5 to 240 Hz). We computed short-time fast Fourier transform on moving 100-ms windows of the preprocessed signals. We used a 200-Hz frequency step size to match the sampling rate to that of the motion capture system. We computed power normalized to the averaged power in each session and calculated an averaged power in high-γ bands (high-γ 1, 60 to 120 Hz; high-γ 2, 120 to 180 Hz). SLiR analysis showed that high-γ power of ECoG signals in M1 encoded the kinematic variables (fig. S2). Moreover, the same analysis indicated that high-γ power of ECoG signals in S1 encoded the kinematic variables immediately before cortical activity (fig. S2). As a neuronal ensemble activity of M1 neurons encoded the kinematic variables (40), we used the high-γ power of ECoG signals as representative of neural activity in cortical areas in the analyses. Data from one electrode of Monkey C were not used for the analysis because of high noise (black circle in fig. S1B).

Neuronal signals of peripheral afferents were initially amplified using the same multichannel amplifier with a gain of ×20,000 and sampled from each electrode at 40 kHz. We extracted filtered waves (150 to 8000 Hz) above an amplitude threshold that was determined by the “auto-threshold algorithm” of the software. We sorted the thresholded waves using semiautomatic sorting methods (Offline Sorter, Plexon), followed by manual verification and correction of these clusters if needed. When the interval of the consecutive spikes was less than 1 ms, the second spikes were removed. To obtain the instantaneous firing rate, we convolved the inversion of the interspike interval with an exponential decay function whose time constant was 50 ms. We computed the firing rate in 5-ms bins, corresponding to the sampling rates of the motion capture system. When we examined the modality of recorded units, we identified some units as muscle spindles, tendon organs, and cutaneous receptors by moving the forelimb, tapping over the muscle belly, and brushing the skin (39).

Note: The content above has been extracted from a research article, so it may not display correctly.



Q&A
Please log in to submit your questions online.
Your question will be posted on the Bio-101 website. We will send your questions to the authors of this protocol and Bio-protocol community members who are experienced with this method. you will be informed using the email address associated with your Bio-protocol account.



We use cookies on this site to enhance your user experience. By using our website, you are agreeing to allow the storage of cookies on your computer.