Electrode positioning system

During each experiment, eight glass-coated tungsten microelectrodes (impedance range at 1 kHz: 0.3–0.8 MW) were inserted separately (Electrode Positioning System, AlphaOmega Engineering, Israel) through the brain and toward the STN. This was done while the monkeys’ heads were immobilized with a head-holder. The electrical activity was amplified by 5000, followed by band-pass filtered from 1 to 8000 Hz via a hardware four-pole Butterworth filter. Then the signal was sampled at 25 kHz by a 12-bit (±5 V input range) Analog/Digital (A/D) converter (Multi Channel Processor, AlphaOmega Engineering, Israel). Individual spikes were sorted online using a template-matching algorithm (Alpha Spike Detector, AlphaOmega Engineering, Israel). Each electrode could present up to three different spiking unit types. The timestamp of each detected spike was sampled at 40 kHz. An isolation score program confirmed the quality of each recorded unit^{55 (link)}. Only units exhibiting an isolation score of 0.6 or above and a stable firing rate for 9 m or more were included in the analysis. Description of the full dataset can be found in Table-S1.

During the recording sessions, the monkeys’ heads were immobilized with a head holder, and eight glass-coated tungsten microelectrodes (impedance 0.45–0.8 MΩ measured at 1000 Hz) were advanced separately by two experimenters toward the GPe. Electrodes were arranged in two towers with four electrodes per tower, thus enabling the recording of GPe neural activity from both hemispheres. Electrodes were navigated within the brain using the Electrode Positioning System (Alpha Omega Engineering, Israel). The electrical activity was amplified by 5000, high-pass filtered at 1 Hz using a hardware two-pole Butterworth filter, and low-pass filtered at 10 kHz using a hardware three-pole Butterworth filter. Raw data were sampled at 44 kHz by a 16-bit (±1.25 V input range) analog-to-digital (A/D) converter. Spiking activity was sorted online using a template-matching algorithm (SnR hardware, software version 2.0.0, Alpha Omega Engineering) by two experimenters. The data collection and analyses were not performed blind to the conditions of the experiment.
Pupil size and horizontal and vertical eye positions were recorded using an eye-tracking device (ISCAN Inc.). Raw data were sampled at 2.75 kHz. We tracked the right eye of each monkey using a camera located 50 cm in front of the monkey’s face.