Data from: Dynamics of mesoscale brain network during decision-making learning revealed by chronic, large-scale single-unit recording
Data files
Sep 24, 2025 version files 2.05 GB
-
README.md
4.47 KB
-
shared_data.zip
2.05 GB
Abstract
Associating unfamiliar stimuli with appropriate behavior through experience is crucial for survival. While task-relevant information has been found to be distributed across multiple brain regions, how regional nodes in this distributed network reorganize their functional interactions throughout learning remains to be elucidated. Here, we performed chronic, large-scale single-unit recording across 10 cortical and subcortical regions using ultra-flexible microelectrode arrays in mice performing a visual decision-making task and tracked mesoscale functional network dynamics throughout learning. Task learning reshaped interregional functional connectivity, leading to the emergence of a subnetwork involving visual and frontal regions during the acquisition of correct No-Go responses. This reorganization was accompanied by a more widespread representation of visual stimulus across regions, and a region’s network rank strongly predicted its peak timing of visual information encoding.
Dataset DOI: 10.5061/dryad.cnp5hqcj2
Description of the data and file structure
Dataset: SpkAndBehav
SubFolder name: Early (early training stage data of learning group); Expert(expert training stage data of learning group); Fruitless(fruitless learning group)
Each mat file in the 3 folders contains behavior data and spike time data from one session of one mouse performing the visual go/no task, and can be opened in MATLAB.
Variables:
log.res_vector: stimulus and outcome of each trial.
sign of the number means correct/wrong. + is correct, - is wrong.
number is visual stimulus identity. 1 is the go stimulus, 2 is the no go stimulus.
1 = hit , 2 = correct rejection; -1 = miss; -2 = false alarm.
log.rsp_vector: response latency of each trial, relative to stimulus onset.
log.lick_time_vector: time of detected licks in each trial, relative to stimulus onset.
spk_time_in_trial: contains spike time data from all the units in this session, aligned to stimulus on set.
spk_time_in_trial{K}{N}: spike time of unit K, in trial N.
unit_tag: brain region identity of each unit.
=============================================================
Dataset: OptoManipulationBehav
SubFolder name: OFC stim (OFC stimulus period manipulation); OFC res(OFC response period manipulation); V2M stim (V2M stimulus period manipulation); V2M res(V2M response period manipulation); Ctrl stim (control group stimulus period manipulation); Ctrl res(control group response period manipulation);
Each mat file in the 6 folders contains behavior data from one session of one mice performing the visual go/no task, and can be opened in MATLAB.
Variables:
log.res_vector: stimulus and outcome of each trial.
sign of the number means correct/wrong. + is correct, - is wrong.
number is visual stimulus identity. 1 is the go stimulus, 2 is the no go stimulus.
1 = hit , 2 = correct rejection; -1 = miss; -2 = false alarm.
log.rsp_vector: response latency of each trial, relative to stimulus onset.
log.lick_time_vector: time of detected licks in each trial, relative to stimulus onset.
=============================================================
m files:
ana_GNGS3_TSPEDyn_Sig02_TS_Shared.m: the main analysis used in the paper.
A scipt written specifically for the shared data in the paper“Dynamics
of mesoscale brain network during decision-making learning revealed by
chronic, large-scale single-unit recording”.
It requires shared data which contains "log", "spk_time_in_trial",'unit_tag'
It also requires the function "TSPE.m" and "parfor_progress.m" to run.
Input parameters:
input_fn: filename to read.
t1: time before stimulus, in seconds. should be a negative value.
t2: stimulus period length，in seconds. should be a positive value.
t3: response window length, in seconds. should be a positive value.
t2 + t3 - t1 should not exceed 3.8.
eff_TSPE_window: in ms. the time window that algorithm considers
cross-correlation to be effective.
period_window: in ms. in each window the regional connection matrix
"connection_between_region" is calculated.
core_num: should be set according to number of cpu in local machine. If
a very large number is set, MATLAB would return an error and tell you the
approriate number.
Output parameters:
The script will automatically save a file，named accoring to input file.
the variable "C_reg_collection_full" contains the desired data.
"C_reg_collection_full" is a n x t cell, n is trial number, t is time window number.
In the cells corresponding to hit trials and cr trials there will be a 10 x 10
matrix containing the raw regional connection strength of brain network,
in the sequence indicated by "r_name": for example, data in the 2nd row,
3rd conlumn indicate functional output from V2M to V2L.
Similarly, “C_reg_collection_outputsum_ranked_full” ,
“C_reg_collection_inputsum_ranked_full” and
"C_reg_collection_ranked_among_all_full" contains rankized data from
"C_reg_collection_full" .(see paper for more details)
Trial identity can be retrieved from behav_log.res_vector and
effective_range. 1 is for hit, 2 is for cr, -1 is for miss, -2 is for fa.
Command to repeat analysis in the paper:
ana_GNGS3_TSPEDyn_TS_Shared('C57#xxx-dxxx_SpkAndBehav-Shared.mat',-0.4,0.8,2,20,200,xx)
Behavioral experiments were controlled using custom Matlab (MathWorks) scripts together with digital I/O devices (Arduino Uno R3, Arduino) for trial events and triggers. Neural signals were amplified and filtered with the SpikeGadgets 1024-channel recording system (SpikeGadgets). Task-related behavioral events were converted to TTL pulses and simultaneously recorded by the SpikeGadgets system to ensure synchronization with neural data. Data analyses were conducted using custom Matlab scripts.