This dataset was generated to assess the different neural responses in the macaque frontal cortex during the observation of actions. This study focused on neural responses of Action Observation/Execution neurons (AOENs, also known as mirror neurons), which are neurons that respond to both the execution and the observation of an action. To find AOENs, subjects performed two tasks during each recording session, and only neurons responding to both tasks were included in the analyses: 

A) Action execution task: the subject performed a visually guided delayed grasping task in which he grasped an object after a variable delay time indicated by a go cue.

B) Action observation task: the subject passively observed short action videos in which either a monkey or a human performed the action execution task.

Additional control videos were also shown to assess neural action observation responses in more detail (for example, a video in which the hand was replaced by an ellipse that moved towards a graspable object, and a video in which the ellipse moved on a scrambled background).

The study was conducted in six male macaques that were implanted with micro-electrode arrays in frontal brain areas. Monkey 1, 2, 3, and 4 had multiple Utah arrays implanted in two or three of the following brain areas: ventral premotor (PMv) area F5c, dorsal premotor (PMd) area F2, or the primary motor cortex (M1). Monkey 5 and 6 were implanted with movable electrodes above the frontal cortex, and recordings in this dataset include the following brain areas: ventral premotor area F5p and the areas in the pre-arcuate sulcus (preAS, including the frontal eye field and area 45B).

Overall, this study showed that AOENs in PMd and M1 displayed highly phasic neural responses during specific epochs of the action video and, in addition, strong responses to a moving object. Moreover, the population dynamics in PMv, PMd and M1 showed a shared representation between action execution and action observation, with an overlap that was as large as the overlap between action execution and passive viewing of ellipse movements. AOENs were only scarcely present in ventral premotor area F5p and preAS areas, suggesting a less important role of these brain areas in action observation.

Description of the data and file structure

1. Description of dataset

The Data.zip folder contains the data of each monkey in their respective folders (Monkey 1, 2a, 2b, 3, 4, 5, and 6). Note that Monkey 2 has two folders (a and b) because he was implanted twice (once in each hemisphere) due to an implant failure after the first implantation. Each of the subject folders has two task folders that contain the data files for that specific task. The following .mat files correspond to the filtered data acquired during the study.
The name of each file includes the task, the animal name, date of collection and hour of collection. For each .mat file, there is a corresponding .nex or.nex5 file that contains the neural activity of individual neurons. For files that were generated after 2020, there is an extra .mat file named *_ReplayedMUA.mat. This file contains the complete multi-unit activity, since only a part was recorded and stored in the original .mat file due to a change in the settings during recording of the signal.

• Animal names: Monkey 1 = Isaac, Monkey 2 = Sky, Monkey 3 = Loki, Monkey 4 = Vino, Monkey 5 = Abu ('a' abbreviated), and Monkey 6 = Sheldon ('shel' abbreviated).
• Tasks: For each recording session, there is an action execution file and an action observation file (see .txt files labeled 'fnames' in each monkey's task folder to have a full list of the generated files per monkey and per task). There are two separate tasks which are listed here under there common name:
  a. Action execution task --> 'vgrasp'
  b. Action observation task --> 'mirror' or 'mirrormovies'

General structure of the data folder:
Data.zip
| - subject name
| | - task name
| | | - testfnames.txt (list of all files in that subfolder)
| | | - [Task][Subject][YYYYMMDD][HHMM][SessionID].mat (file with behavioral and neural data)
| | | - [Task][Subject][YYYYMMDD][HHMM][SessionID]sorted.nex (file with neural activity of individual neurons; .nex for Monkey 1, 5, and 6 and .nex5 for Monkey 2, 3, and 4)
| | | - [Task][Subject][YYYYMMDD]*[HHMM]_[SessionID]_ReplayedMUA.mat (file with correct multi-unit activity; only for Monkey 3 and 4)
| | | - [subject]_F5p.txt (list of all files in that subfolder, with for each file the electrodes that recorded neural activity of individual neurons in F5p; only for Monkey 5 and 6)
| | | - [subject]_preAS.txt (list of all files in that subfolder, with for each file the electrodes that recorded neural activity of individual neurons in preAS; only for Monkey 5 and 6)

1. File list

A. files of the Action execution task
(.mat file, corresponding .nex/.nex5 file, and if needed, an additional **ReplayedMua .mat file).
The naming convention is slightly different between subjects.
Monkey 1-4: [task][subject][YYYYMMDD][HHMM][SessionID]
Monkey 5-6: [subject abbreviated][task][SessionID][DD_MM_YY]*[HHMM]

 i. vgrasp_[Subject]_[YYYYMMDD]_[HHMM]_[SessionID].mat   

Example: vgrasp_Loki_20210601_0953_A.mat (3 files of Monkey 1, 3 files of Monkey 2a, 3 files of Monkey 2b, 6 files of Monkey 3, 3 files of Monkey 4, 17 files of Monkey 5, and 9 files of Monkey 6)

Description: These files contain the behavioral and neural data recorded while the subject performed the action execution task. The .mat file can be opened in MATLAB. The variables included in each .mat file are listed below.

Content .mat files of Monkey 1-4: Each .mat file contains two variables that are necessary for the analysis of the data: cerebusDataA and tnsTrials.
Some .mat files have an additional variable 'cerebusDataB' that contains the same structure as cerebusDataA, but for additional electrodes. When numbering electrodes, electrodes of cerebusDataB follow those of cerebusDataA. So if cerebusDataA contains 128 electrodes and cerebusDataB 64 electrodes, the electrodes of cerebusDataB will be electrode 129 until 192.
All timings are in milliseconds and relative to the start of the task.
A) cerebusDataA contains the neural data
- elecSpikes; timestamps of each electrode. _0 in the name indicates that it contains multi-unit activity. Each timestep is the timing of a spike event, i.e., crossing the set threshold (in milliseconds).
- Cue; timing of the Go Cue (milliseconds) used to align the neural data to the task data.
B) tnsTrials contains the task data
- Index; trial number
- Start; start of the trial (milliseconds)
- Stop; stop of the trial (milliseconds)
- Answer; answer is 1 if the animal performed the trial correctly.
- TargetObject; sphere that needed to be grasped [0:3]
- Light; moment the external light turned on and illuminated the object (milliseconds)
- LiftHand; moment the subject started to reach for the object (milliseconds)
- Target; moment the subject pulled the correct sphere (milliseconds)
- Cue; two values that indicate when the cue light went on and off (milliseconds). The second value is the Go Cue.

Content .mat files of Monkey 5-6: Each .mat file contains one variable called 'c.trial', which contains the behavioral (dsp_data) and the neural data (cerebus_data) per trial.
A) cerebus_data
- spikeEvents: timestamps of each electrode (multi-unit activity). Each timestep is the timing of a spike event, i.e., crossing the set threshold (milliseconds).
B) dsp_data
- Start; start of the trial (milliseconds)
- Stop; stop of the trial (milliseconds)
- Answer; answer is 1 if the animal performed the trial correctly.
- Condition; sphere that needed to be grasped [0:3]
- LiftHand; moment the subject started to reach for the object (milliseconds)
- PullObject; moment the subject pulled the correct sphere (milliseconds)
- Cue; two values that indicate when the cue light went on and off (milliseconds). The second value is the Go Cue.
- spikeEvents: timestamps of each electrode that had an offline sorted individual neuron (single unit activity). Each timestep is the timing of a spike event, i.e. crossing the set threshold (milliseconds).

 ii. vgrasp_[Subject]_[YYYYMMDD]_[HHMM]_[SessionID]_ReplayedMUA.mat 

Example: vgrasp_Loki_20210601_0953_A_ReplayedMUA.mat (6 files of Monkey 3, 3 files of Monkey 4)

Description: For files that were generated after 2020 (Monkey 3 and 4), there is an extra .mat file named *_ReplayedMUA.mat. This file contains one variable 'cerebusDataA'. Use this variable instead of the 'cerebusDataA' of the general .mat file (i) when the *_ReplayedMUA.mat file is present. The variable 'elecSpikes' has the same structure in both .mat files.

Content .mat file:
A) cerebusDataA contains the neural data
- elecSpikes; timestamps of each electrode. _0 in the name indicates that it is multi-unit activity. Each timestep is the timing of a spike event, i.e. crossing the set threshold (in milliseconds).

 iii. vgrasp_[Subject]_[YYYYMMDD]_[HHMM]_[SessionID]_sorted.nex 

Example: vgrasp_Loki_20210601_0953_A_cer1001_sorted.nex5 (3 files of Monkey 1, 3 files of Monkey 2a, 3 files of Monkey 2b, 6 files of Monkey 3, 3 files of Monkey 4, 17 files of Monkey 5, and 9 files of Monkey 6)

Description: The .nex and .nex5 files contain the neural data of offline sorted individual neurons, that were manually sorted with the 'Plexon Offline Sorter' software. Individual electrodes can have data from multiple sorted neurons.

Content .nex /.nex5 file: timesteps (in milliseconds) of the sorted neurons per electrode.

B. files of the Action observation task

The data files of the Action observation task are similarly structured as the files of the Action execution task. So for each subject there will be a .mat file, a corresponding .nex/.nex5 file, and if needed an additional *_ReplayedMua mat file per recording session. Since the study focused on finding AOENs, each recording session includes an Action execution data file and an Action observation data file. Therefore, the number of files is identical in the two task folders per subject.

 i. mirrormovies_[Subject]_[YYYYMMDD]_[HHMM]_[SessionID].mat
 ii. mirrormovies_[Subject]_[YYYYMMDD]_[HHMM]_[SessionID]_ReplayedMUA.mat 
 iii. mirrormovies_[Subject]_[YYYYMMDD]_[HHMM]_[SessionID]_sorted.mat

Description: These files are similarly structured as the Action execution task and contain the same variables that also apply to this task (Index, Start, Stop, Answer, elecSpikes, spikeEvents). Other variables are listed below.

Content .mat file of Monkey 1-4:
B) tnsTrials
- Stimulus; name of the video that is shown
(POV = point-of-view of the monkey = Viewpoint 1; Side = filmed from the side = Viewpoint 2).
Each stimulus name is paired with its corresponding name in the paper:
Human Grasp = [viewpoint]_Human_Grip
Human touch = [viewpoint]_Human_Fist
Monkey = [viewpoint]_Obi_Monkey
Static = [viewpoint]_Human_Nogo
Ellipse = [viewpoint]_Object_Ellipse
SCR background = [viewpoint]_SCR_Ellipse
Static Interaction = [viewpoint]_Human_staticgrip
Static Approach = [viewpoint]_Human_StaticApproach
Human_Disappear videos were not used in this study.
- PhotoEvents; timing of photoevents (in milliseconds), 3 in a correct trial: (1) start of the trial, (2) start of the video, (3) end of the video.

Content .mat file of Monkey 5-6:
B) dsp_data
- Condition; number of the video shown
(POV = point-of-view of the monkey = Viewpoint 1; Side = filmed from the side = Viewpoint 2).
Each stimulus name is paired with its corresponding name in the paper:
Human Grasp = [viewpoint]_human_grip.avi
Human Touch = [viewpoint]_human_fist.avi
Monkey = [viewpoint]_obi.avi
Static = [viewpoint]_nogo.avi
Ellipse = [viewpoint]_obj_ellipsa.avi
SCR background = [viewpoint]_scr_ellipse.avi
- PhotoEvents; timing of photoevent, i.e., the start of the trial (in milliseconds)

C. testfnames.txt files (for each task per subject)

Description: Each .txt file contains a list of all recording files used per task and per subject. Below each file name is one letter, indicating which recorded electrodes (also called channels or ch abbreviated) belong to which brain area.
All data shown in the study are from either the primary motor cortex (M1), the dorsal premotor cortex (PMd), the ventral premotor area F5p, or the pre-arcuate areas (preAS). However, not all electrodes that are included in these data files are implanted in one of these areas. The recorded data may also contain data from electrodes implanted in the ventral premotor area F5c (PMv, see related public dataset: DOI: 10.5061/dryad.kwh70rzc7).
The electrode numbering is the same in the action execution (vgrasp) and the action observation task (mirrormovies).
Below is a list of the possible letters in the .txt files and the corresponding channel configurations:

E
pmv = [1:96];
pmd = [97:128];
F
pmv = [1:64];
pmd = [65:128];
C
pmv = [97:128];
pmd = [1:96];
U
pmv = [1:96];
m1 = [97:192];
R
m1 = [1:96];
S
pmd = [1:96];
Q
m1 = [1:64,97:128];
pmd = [65:96,129:192];
pmv = [193:256];
O
pmv = [1:64,97:128];
m1 = [65:96,129:192];
pmd = [193:256];

Note: if the letter is an 'o' or 'z' (in Monkey 5 and 6), the electrode configuration is written down in additional txt files (see D. below).

D. other .txt files (for each task per subject; only Monkey 5 and 6)
The naming convention is: [Monkey][area].txt

example: _Abu_F5p.txt

Description: These .txt files have a similar content as the 'testfnames.txt' files. They list all file names in that folder, but below each file name are the electrodes listed that have offline sorted neurons (in the corresponding .nex file) in the brain area that is in the .txt name. For example, in 'Monkey6/Action execution task/_Sheldon_F5p.txt' the first few lines read:
shelvgraspC_21_03_17_09h22
16 20 24 45 54 77 85 87 89 90 91 93 94
Thus, in this specific .nex file, there will be 13 offline sorted neurons (one on ch16, one on ch20, etc.). Each folder will have two of these .txt files: one for the offline sorted neurons in F5p and one for the offline sorted neurons in preAS.

Missing data codes: None

Code/Software

MATLAB required for opening .mat files

To open .nex or .nex5 files: Software: NeuroExplorer

SHARING/ACCESS INFORMATION

Links to publications that use the data: De Schrijver S, Decramer T, Janssen P. Action observation responses in macaque frontal cortex. (2026) NeuroImage: Reports. https://doi.org/10.1016/j.ynirp.2026.100319