Chronic, wireless recordings of large-scale brain activity in freely moving rhesus monkeys

Abstract

Advances in techniques for recording large-scale brain activity contribute to both the elucidation of neurophysiological principles and the development of brain-machine interfaces (BMIs). Here we describe a neurophysiological paradigm for performing tethered and wireless large-scale recordings based on movable volumetric three-dimensional (3D) multielectrode implants. This approach allowed us to isolate up to 1,800 neurons (units) per animal and simultaneously record the extracellular activity of close to 500 cortical neurons, distributed across multiple cortical areas, in freely behaving rhesus monkeys. The method is expandable, in principle, to thousands of simultaneously recorded channels. It also allows increased recording longevity (5 consecutive years) and recording of a broad range of behaviors, such as social interactions, and BMI paradigms in freely moving primates. We propose that wireless large-scale recordings could have a profound impact on basic primate neurophysiology research while providing a framework for the development and testing of clinically relevant neuroprostheses.

Figures

Figure 1. Recording Cubes and Primate Headcap

(a) Schematic drawings and photographs of actual microwire cubes, showing the wire density and adjustment mechanisms. The 3rd generation is a 10×10 array that is fitted with bundled microwires (right panel). (b) Photographs of final arrangements of connectors, showing a view from the top of the headcap of a monkey implanted with 2nd gen cubes (left) and a similar view of the VLSBA module on monkey implanted with 3rd gen cubes (middle). The right panel shows the wireless module mounted on the headcap. (c) Layered schematic drawing showing the modular headcap assembly, with the VLSBA connector organizer as the active module. (d) Schematic drawing showing the modular headcap wireless assembly on the right panel. Both headcaps are 3D printed.

Figure 2. Large-scale activity Recordings

(a) Representative mean waveforms for 1,874 neurons over a center-out task. Intermittently colored to indicate array of origin. Green highlight: neurons from left hemisphere M1. Orange highlight: left hemisphere S1. Blue highlight: right hemisphere M1. Yellow highlight: right hemisphere S1. (b) Population raster plot depicts 1 second window of neuronal spiking activity for a single recording session (556 neurons). The vertical color key identifies the recorded areas. 128-channel Plexon system was used for each recorded area. (c) This panel shows unit isolation clustering using the first two principal components of a subset (50) of channels ) from Monkey O.

Figure 3. Recording Longevity

Neuronal recording yield over time measured by quantifying the average number of single units recorded per connector (32 channels). All the data from the 5 monkeys used in these experiments, and three other monkeys (T, I, and Cl), were used. Four-year sample waveforms, identified by date, from Monkey M’s left hemisphere M1 connector surround the figure showing mean sample waveforms. Highlighted area corresponds to the waveform’s standard deviation.

Figure 4. Large Scale Wireless System

Schematic diagram showing the information flow of the wireless recording system, beginning with spike waveforms to spikes transmitted across ISM radio band to the client. (a) Exploded diagram of wireless transceivers located inside the headcap. (b) Wireless bridges showing the main components of bidirectional communication and the connections to the client computer. (c) Photograph of the radio transceiver with scale. (d) Photograph of the wireless bridge, with scale. (e) Screenshot of the wireless client with four channels visible, demonstrating the PCA sorting method and graphical user interface.

Figure 5. Wireless Brain Machine Interface

(a) Screenshot of monkey with wireless system performing a task using pure brain control. See Supplementary Video 1. (b) Performance metric increase in a single week of training. (c) Example average traces the y-coordinate of the cursor for trials with target at 0 degrees (North) of orientation, showing average constant movements despite variable postures during task interaction, as well as time-to-target. Peaks indicate reached target. (d) Peri-event time histograms centered on target onset for 212 units, grouped at each column with cursor control paradigms of basic hand control (HC) with joystick, brain control with joystick (BCWH), and sole brain control (BCWOH), respectively. Color key shows unit groupings per area, and include LS1, LM1, LSMA, RM1, and RSMA. Each unit was normalized per condition using z-score method.

Figure 6. Freely Behaving Recordings and Future Arrays

(a) Screenshots of videos used for coding of behaviors. Two frames for three sample behaviors are shown. See Supplementary Video 3. (b) Plot of the first three principal components of neural activity for 6 different behaviors, color coded per behavior, shown as filled in circles. PCA data was used for SVM classification (see Table 5). (c) PETHs for behaviors from a single monkey, calculated from 107 M1 neurons centered around behavior onset. Legend for spectrograms placed on the bottom right corner to reduce figure clutter. Each unit was normalized for the entire session (all behaviors shown) using the z-score method. (d) 864 channel array assembly modules. Left panel shows the bottom view with guiding tube density. Right panel shows the top view with a depiction of the movable assemblies. (e) Photograph of fully assembled array with guiding tubes. Note the high density connector attached. (f) Connector and cap design. Left panel shows schematic for high density connectors. Right panel shows a redesign of the cap which should allow for the management of 10,000 channels without increasing the cap footprint. (g) Panel shows photograph of plastic cap mounted with a single connector. The fully assembled cap will host 12 connectors, each with 864 channel capacity.