Data from: Mesostriatal dopamine is sensitive to changes in specific cue-reward contingencies
Data files
Apr 29, 2024 version files 730.23 MB
-
ANCCR_photometry_simulation_all_params.mat
-
ANCCR_photometry_simulation_best_params.mat
-
ANCCR_simulate_ChR2_control.mat
-
ANCCR_simulate_ChR2_degradation.mat
-
ANCCR_simulate_halo.mat
-
ChR2_data.mat
-
context_data.mat
-
eopn3_data.mat
-
halo_data.mat
-
nac_dlight_data.mat
-
README.md
-
vta_gcamp_data.mat
Abstract
Learning causal relationships relies on understanding how often one event precedes another. To gain an understanding of how dopamine neuron activity and neurotransmitter release change when a retrospective relationship is degraded for a specific pair of events, we used outcome-selective Pavlovian contingency degradation in rats. Two cues were paired with distinct food rewards, one of which was also delivered in the absence of either cue. Conditioned responding was attenuated for the cue-reward contingency that was degraded. Dopamine neuron activity in the midbrain and dopamine release in the ventral striatum in response to the cue and subsequent reward were attenuated during degraded versus non-degraded trials, and contingency degradation also abolished the trial-by-trial history dependence of dopamine responses at the time of trial outcome. This profile of changes in cue- and reward-evoked responding is not easily explained by a standard reinforcement learning model. An alternative model based on learning causal relationships was better able to capture evoked dopamine responses during contingency degradation, as well as conditioned behavior following optogenetic manipulations of dopamine during noncontingent rewards. Our results suggest that mesostriatal dopamine encodes the contingencies between meaningful events during learning.
README: Mesostriatal dopamine is sensitive to changes in specific cue-reward contingencies
https://doi.org/10.5061/dryad.q573n5tr1
This dataset includes two types of behavioral data. First, there are conditioned port entry rates from five separate cohorts of rats: those that underwent fiber photometry recordings using GCaMP6f in ventral tegmental area (VTA) dopamine neurons, those that underwent fiber photometry recordings using dLight1.2 in the nucleus accumbens (NAc) core, those that underwent a context manipulation, those that underwent optogenetic inhibition of VTA dopamine neurons, and those that underwent optogenetic inhibition of dopamine release. Second, there are behavioral measures derived from videos: conditioned head velocities and distances between head and mid-tail derived from DeepLabCut, and conditioned rates of rearing and rotating derived from video hand-scoring.
This dataset also includes fiber photometry data taken from rats that underwent recordings using GCaMP6f in VTA dopamine neurons and dLight1.2 in the NAc core.
This dataset also includes simulated data from the ANCCR model (Jeong et al., 2022, Science).
Description of the data and file structure
All data are stored as .mat files and can be open and manipulated using MATLAB. Each non-simulation data file contains structure arrays that are indexed to individual rats.
The following is a description of which figures are derived from which data set:
- "context_data" are presented in figures 1B and 1C.
- "eopn3_data" are presented in figures 5F and 5G.
- "halo_data" are presented in figures 5C and 5D.
- "ChR2_data" are presented in figures 6C-E and S10.
- "vta_gcamp_data" are presented in figures 1B-D, 2B-D, 4A, 4C, and 4D.
- "nac_dlight_data" are presented in figures 1B-D, 3B-D, and 4B-D.
- "ANCCR" data files are presented in figure 7.
The following list includes definitions of abbreviations used in the variable names:
training = acquisition phase
degradation = contingency degradation phase
CR = conditioned response
CS = conditioned stimulus
nondeg = nondegraded trial
deg = degraded trial
diff = trial - baseline
coef - regression coefficient
int = regression intercept
rwd = reward
omm = ommision
noncon = noncontingent
CT = cycle-to-trial
halo = halorhodopsin group
cont = control group
FR1 = fixed-ratio 1
cre_neg = TH-Cre negative
cre_post = TH-Cre positive
dist - distance
SRC = successor representation contingency
PRC = predecessor representation contingency
The following is a description of how the the behavioral data are derived:
- All individual conditioned response data are stored as the session-averaged port entry rates during the cue period with ("diff") and without ("cue") a pre-trial baseline subtracted.
- All individual regression coefficients and intercepts are computed from the final session of contingency degradation.
- All CT ratios are computed from the final session of contingency degradation.
- In the file "eopn3_data.mat," all lever pressing data from the FR1 schedule are session-averaged.
- In the file "ChR2_data.mat," all head velocity data represent averages for sessions 1 and 12 of post-acquisition. All other variables are session averages only from session 12 of post-acquisition.
The following is a description of how the photometry data are derived:
- All individual gcamp and dlight data are presented as session-averaed df/f during the final session of contingency degradation.
- All individual regression coefficients and intercepts are computed from the final session of contingency degradation.
The following is a description of how the ANCCR simulations are derived:
- In the file "ANCCR_photometry_simulation_all_params.mat," the simulations are organized in a 7 dimensional cell array. Dimension 1 represents the 20 iterations per parameter combination. Dimension 2 represents the k parameter. Dimension 3 represents the alpha parameter. Dimension 4 represents the alpha_r parameter. Dimension 5 represents the T ratio parameter. Dimension 6 represents the threshold parameter. Dimension 7 represents the w parameter.
- In the file "ANCCR_photometry_simulation_best_params.mat," each cell array represents 100 iterations over the bet fitting parameter combination.
- In the files "ANCCR_simulate_halo.mat," "ANCCR_simulate_ChR2_control," and "ANCCR_simulate_ChR2_degradation," each variable is a trial x iteration matrix, where the first 500 trials are from the acquisition phase and the next 500 trials are from contingency degradation.