A stable, distributed code for cue value in mouse cortex during reward learning
- Center for the Neurobiology of Addiction, Pain and Emotion, University of Washington, United States
- Anesthesiology and Pain Medicine, University of Washington, United States
- Department of Biological Structure, University of Washington, United States
- Department of Pharmacology, University of Washington, United States
Figures
Figure 1 with 2 supplements
Electrophysiology and calcium imaging during olfactory Pavlovian conditioning.
(A) Trial structure in Pavlovian conditioning task. (B) Timeline for mouse training. (C) Mean (+/− standard error of the mean (SEM)) lick rate across mice () on each trial type for each odor set …
Figure 1—figure supplement 1
Anticipatory licking during the electrophysiology sessions.
(A) Mean anticipatory licks (change from baseline) for the CS+ and CS50 from odor set A (left) and B (right) for every session, color-coded by mouse. and in each odor set for a main effect of …
Figure 1—figure supplement 2
Similar neural activity in prelimbic area using electrophysiology and calcium imaging.
(A) Heatmap of the normalized activity of each neuron recorded with electrophysiology in prelimbic area (PL), aligned to each of the six odors. All columns sorted by mean firing 0 - 1.5s following …
Figure 2 with 4 supplements
Graded cue and lick coding across the recorded regions.
(A) Location of each recorded neuron relative to bregma, projected onto one hemisphere. Each neuron is colored by common-coordinate framework (CCF) region. Numbers indicate total neurons passing …
Figure 2—figure supplement 1
Task-related neural activity across brain regions.
(A) For each of the 5 mice in the electrophysiology experiment, the number of neurons recorded in each region. (B) Heatmap of the normalized activity of each neuron ( trials per cue). All columns …
Figure 2—figure supplement 2
Identification of cue and lick cells with GLM.
(A) Mean variance explained (fraction) by linear models in each region for each session (x) and the mean (+/− SEM) across those sessions. (B) Mean (+/− SEM) activity of neurons encoding cues, licks, …
Figure 2—figure supplement 3
Validation of variance cutoff for variable coding.
(A) Fraction of neurons encoding cues, licks, and rewards in each region when varying the unique variance cutoff used (how much model performance drops when removing that variable). (B) As in (A), …
Figure 2—figure supplement 4
Comparing proportions of cue and lick neurons across regions.
(A) Fraction of neurons in each region classified as coding cues (left), licks (middle), or both (right), as well as estimated fraction(± 95% CI) with random effect of session (see Methods). Data …
Figure 3 with 2 supplements
Robust value encoding and decoding among cue cells.
(A) Normalized activity of an example value cell with increasing modulation for cues with higher reward probability.(B) For the same neuron, model-fit cue kernel for the original value model and …
Figure 3—figure supplement 1
Schematic of value model shuffles.
(A) For each of the 153 cue coding models, the value taken on by the variable cue kernel on trials corresponding to each of the six cue types. Values were 0, 0.5, or 1. Also, the fraction of cue …
Figure 3—figure supplement 2
Population analysis of value coding schemes.
(A) Projecting the activity (0 to 2.5s from odor onset) of all value and value-like cells onto the coding dimensions maximally separating CS− and CS+ (x-axis) and CS− and CS50 (y-axis). X marks …
Figure 4 with 3 supplements
Widespread cue value coding.
(A) Fraction of neurons in each region and region group classified as value cells (blue) and other cue neurons (gray), as well as fraction (± 95% CI) estimated from a linear mixed effects model with …
Figure 4—figure supplement 1
Relative proportions of value and value-like cells across regions.
(A) Additional cue value (left) or value-like (right) neurons in region on y-axis compared to region on x-axis as a fraction of all neurons, for regions with non-overlapping 95% confidence …
Figure 4—figure supplement 2
Value coding as a proportion of cue cells.
(A) Fraction of cue neurons in each region classified as coding value (left) or value-like (right), as well as estimated fraction(± 95% CI) with random effect of session (see Methods). (B) …
Figure 4—figure supplement 3
Comparing prefrontal cortex (PFC) and striatum.
(A) Fraction of neurons in each region and region group classified as coding cues (left), licks (middle), or both (right), as well as estimated fraction(± 95% CI) with random effect of session (see …
Figure 5
A subset of cue cells incorporate reward history.
(A) Coefficient weight (± standard error from model fit) for reward outcome on the previous 10 trials of any type (left) and on the previous 10 trials of the same cue type (right) for the ‘trial …
Figure 6
Acquisition of conditioned behavior and cue encoding in prefrontal cortex (PFC).
(A) Training schedule for five of the mice in the calcium imaging experiment. An additional three were trained only on odor set A. (B) Mean (± SEM) licking on early (first 60) and late (last 60) …
Figure 7 with 1 supplement
Cue and lick coding is stable across days.
(A) Standard deviation fluorescence from example imaging plane. (B) Masks (randomly colored) for all tracked neurons from this imaging plane. (C) Deconvolved spike rate on every CS+ trial from all …
Figure 7—figure supplement 1
Correlation across days in prelimbic area (PL).
(A) Cumulative distribution of percentile of correlation for the activity of a given neuron with its own activity on the subsequent day compared to its correlation with the activity of all other …
Figure 8
Stable cue coding across separately trained odor sets.
(A) Normalized activity of all pixels in the imaging plane following CS+ presentation on the third day of each odor set (A3 and B3, days 5 and 6 of training). (B) Fraction of neurons coding for …
Additional files
-
Supplementary file 1
Statistics related to Figure 2G and Figure 2—figure supplement 4.
Bonferroni-corrected p-values from region contrast in generalized linear mixed-effects model.
- https://cdn.elifesciences.org/articles/84604/elife-84604-supp1-v1.docx
-
Supplementary file 2
Statistics related to Figure 3.
Top: Bonferroni-corrected p-values from pairwise comparisons between the decoding accuracy of each group of neurons at each time point with their performance at baseline and with the other neuron groups at that time point. Middle, Bottom: Bonferroni-corrected p-values for pairwise comparisons of bootstrapped distributions (1000 samples) of decoding performance using increasing numbers of neurons in each group.
- https://cdn.elifesciences.org/articles/84604/elife-84604-supp2-v1.docx
-
Supplementary file 3
Statistics related to Figure 4.
Top, Middle: Bonferroni-corrected p-values from region contrasts in generalized linear mixed-effects model. Bottom: Bonferroni-corrected p-values for pairwise comparisons of bootstrapped distributions (1000 samples) of decoding performance using value cells from each region.
- https://cdn.elifesciences.org/articles/84604/elife-84604-supp3-v1.docx
-
Supplementary file 4
Statistics related to Figure 5.
Top: Bonferroni-corrected p-values for pairwise comparisons of bootstrapped distributions (5000 samples) of the slope of population activity of each group of neurons across CS50 trials of increasing value. Bottom: Bonferroni-corrected p-values from region contrasts in generalized linear mixed-effects model.
- https://cdn.elifesciences.org/articles/84604/elife-84604-supp4-v1.docx
-
Supplementary file 5
Statistics related to Figure 2G and Figure 2—figure supplement 4.
Bonferroni-corrected p-values from region contrast in generalized linear mixed-effects model.
- https://cdn.elifesciences.org/articles/84604/elife-84604-supp5-v1.docx
-
MDAR checklist
- https://cdn.elifesciences.org/articles/84604/elife-84604-mdarchecklist1-v1.pdf
