Page 93 - Science

RESEARCH | REPORT

M1 Recordings Auditory Daily Baseline (500 samples)
E1 E2
Feedback S = S = 1 2 3 4
E1
E2
1 2 3 4 19 Target 2
E1 Counts E2 Counts Tone (kHz) 10 Center
1 mV Target 1
1 ms 5
10th 50th 90th 10th 50th 90th 1 4 7
Ensemble FR Modulation Percentile Decoder = S -S +4
500 ms E1 E2
cycle Baseline: Baseline:
40 S1 7 S2
Decoder S2 6 S3
E1 E1 S3 5 S4
S4
+ + % Occupancy 20 Occupancy Gain 4
E2 E2 3
_ 2
E1 FR Modulation + _
+ E2 FR Modulation n = 16 animals 1
0 0
5 6 8 10 12 15 19 5 6 8 10 12 15 19
Tone (kHz) Tone (kHz)
Target 1 Target 2
5 kHz 6 kHz 8 kHz 10 kHz 12 kHz 15 kHz 19 kHz Ensemble 1 0.5
0.5 Downloaded from
Ensemble 2
Target 1 Neural-Auditory Mapping Target 2
Firing rate (z-score) 0 Firing rate (z-score) 0
Closed-loop
Optogenetic Stimulation
Target 1 HIT
-0.5 -0.5
-3 0 3 -3 0 3
Time (s) Time (s) http://science.sciencemag.org/
VTA
Cue (1 s) Cue (1 s)
Trial Starts Trial Ends
Achieve Stim: 2s train
Center Tone Achieve Target Tone (1-60 s) ITI (3.5 s)
14Hz, 28 pulses (10 ms)
T1
Stim
M1 Target 2 Trial Target 1 Trial ITI (3.5 s)
Cue Cue ITI (3.5s) Cue Cue Stim
19 on March 1, 2018
(kHz) 10
5
Silence
M1 Center T2 Center T1
M1M1
YFP TH VTA 1.5 s 2 s
Time
Fig. 1. Closed-loop BMI paradigm for pairing specific motor cortex 500 samples of 500-ms spike counts are collected from spontaneous
activity patterns with phasic VTA dopaminergic activity. (A)Sche- neural activity as the mouse freely behaves in the box with no task
matic of the BMI paradigm. Each mouse receives a unilateral microwire or auditory tones. Each ensemble’s firing rate modulation is defined
array implant in the motor cortex (targeted to layer V) and a as the sum of the member neurons’ normalized spike counts
contralateral optical fiber implant in the VTA. Recorded single units are (mean-centered, range-normalized) and then quantized into four
arbitrarily assigned into two ensembles (E), and the concomitant activation states. The decoder’s state is the difference between
increase (up arrow) of one ensemble’s activity and decrease (down ensemble 1’sand ensemble 2’s activation state and is mapped into
arrow) in the other ensemble’s activity drives the decoder to change one of seven tones. The stars indicate target tones. (D) BMI calibration
the auditory tone produced every 500 ms. The rare, lowest-pitch on baseline period spontaneous neural activity results in a Gaussian-like
tone triggers phasic optical stimulation to the VTA, whereas the rare, distribution over tones, such that target 1 (5 kHz) and target 2
highest-pitch tone serves as a control. Solid triangles indicate neurons (19 kHz) are rare. The mean and SEM baseline distribution for each
with positively modulated firing rate; open triangles indicate neurons session is plotted on the left, averaged over all animals. Baseline
with negatively modulated firing rate. Yellow color indicates the distributions show no change from session 1, as shown on the right.
center-pitch tone. FR, firing rate. (B) Coronal brain slice depicting (E) Ensemble 1 and 2 firing rate modulation before target 1 and target
viral infection specific to the dopaminergic cells of the VTA. The 2 hits, averaged over all recorded cells and sessions. (F)Task
immunohistochemistry labels for tyrosine hydroxylase (TH, red) and the schematic. Trial structure is the same for target 1 and target 2,
Cre-dependent fluorescent protein (YFP, yellow) are shown. (C)BMI except that a target 1 hit results in phasic VTA stimulation (2-s train of
decoder calibration. For every session (S) during the baseline period, 14 Hz pulses with 10-ms width). ITI, intertrial interval.

Athalye et al., Science 359, 1024–1029 (2018) 2 March 2018 2of6