Area 2 of primary somatosensory cortex encodes kinematics of the whole arm (DANDI:000127)

Claims

From the paper Area 2 of primary somatosensory cortex encodes kinematics of the whole arm

Reproducing Chowdhury, Glaser & Miller (2020) from dandiset 000127 6

Notebook reproduce_chowdhury_2020.ipynb

Partially reproducedFigure 10
Unlike in the two-workspace experiment, many neurons had no consistent relationship with arm kinematics across active and passive movements, shown by negative pseudo-R2 for the model trained on both conditions.
19 of 50 neurons scored on active trials and 23 of 50 on passive trials have negative pseudo-R2 for the model trained on both conditions, but about as many (17 and 25) are negative when trained within one condition. Only 5 and 3 neurons are positive within a condition and negative across both. Training on both costs a little on active trials and nothing on passive trials.
How the test differed from the paper’s method
- One session from one monkey; the paper has four sessions from two.
- The whole-arm model is hand plus joint kinematics, as for fig9-whole-arm-fits-better.
- Pseudo-R2 for one trial type is computed on a repeat's pooled held-out predictions, against that type's own mean firing; the paper does not say which mean it used.
Partially reproducedFigure 11
The consistency of the whole-arm model, its pseudo-R2 when trained on both conditions, correlated negatively with a neuron's separability index.
Scored on active trials, model consistency correlates negatively with separability (Spearman -0.38, 95% interval -0.60 to -0.11). Scored on passive trials the correlation is -0.13 and on all trials together 0.14; neither can be told apart from zero.
How the test differed from the paper’s method
- One session from one monkey; the paper has four sessions from two.
- The whole-arm model is hand plus joint kinematics, as for fig9-whole-arm-fits-better.
ReproducedFigure 8Bverifier disagreed at first
In the 120 ms after movement onset, average hand speed was similar in active reaches and passive bumps.
The speed traces of active and passive trials overlap in the 120 ms window and separate afterwards. Passive movements are 13% faster on average (0.207 against 0.183 m/s), half a standard deviation, which is statistically clear; the paper's "similar" is qualitative.
How the test differed from the paper’s method
- One session from one monkey; the paper has four sessions from two.
ReproducedResults, active/passive section
A linear discriminant classifier predicted whether a movement was active or passive 89% of the time from whole-arm kinematics in the analysis window.
A linear discriminant tells active from passive trials 90% of the time from joint angles and velocities, against 58% from hand kinematics and 53% by chance. The paper reports 89% averaged over four sessions.
How the test differed from the paper’s method
- One session from one monkey; the paper has four sessions from two.
- Joint angles and velocities stand in for the hand and elbow marker kinematics, which are not in the dandiset.
ReproducedFigure 8E
Many neurons had an above-chance separability index, the accuracy of a linear discriminant predicting the trial type from that neuron's firing.
20 of 50 neurons have a separability index above chance at p < 0.05, with a median of 0.56 and a highest value of 0.80.
How the test differed from the paper’s method
- One session from one monkey; the paper has four sessions from two.
- Neurons firing below 1 Hz in the window (15 of 65) are left out; the paper states no criterion.
ReproducedFigure 9
In the active/passive experiment the whole-arm model again tended to out-perform the hand-only model, but for many more neurons than in the two-workspace experiment the difference was not significant.
The whole-arm stand-in is significantly better than the hand-only model for 15 of 50 neurons and worse for none; 35 (70%) show no significant difference, against 17% in the paper's two-workspace experiment. That comparison uses the paper's number.
How the test differed from the paper’s method
- One session from one monkey; the paper has four sessions from two.
- The whole-arm model is hand plus joint kinematics; the marker positions the paper's model uses are not in the dandiset.
- The hand-only model is two-dimensional, the fits carry a small ridge penalty, neurons below 1 Hz are left out, and no Bonferroni correction is applied.

Not tested in a notebook 4

4 data missing from the dandiset

BlockedFigure 4· Data missing from the dandiset
In the two-workspace experiment, the whole-arm model predicted neural activity better than the hand-only model for 238 of 288 neurons, with no significant difference for the other 50.
The dandiset holds one session of the active/passive experiment. None of the eight two-workspace sessions is in it.
BlockedFigure 5· Data missing from the dandiset
A whole-arm model trained on both workspaces explained as much neural variance as models trained within a single workspace.
The two-workspace sessions are not in the dandiset.
BlockedFigure 6· Data missing from the dandiset
The whole-arm model predicted the shape of directional tuning curves in the two workspaces better than the hand-only model for 138 of 288 neurons, and worse for 2.
The two-workspace sessions are not in the dandiset.
BlockedFigure 7· Data missing from the dandiset
The whole-arm model predicted the shifts of preferred direction between workspaces better than the hand-only model, with a circular variance accounted for of 0.75 against 0.57.
The two-workspace sessions are not in the dandiset.

Findings about the data

What attempts found about the dandiset itself, for its owner and the archive.

Data missingThe dandiset holds one of the twelve sessions analysed in the paper, and none from the two-workspace experiment, so four of the paper's main findings cannot be tested from it.
Data missingThe positions of the arm markers, which the paper's whole-arm model is built on, are not included. Joint angles and muscle lengths derived from them are.
Metadata errorThe text descriptions of hand position and hand velocity give centimetres while the stated unit is metres. The stated unit, with the file's conversion factor, is correct.
Selection not documentedThe dandiset's metadata names the paper without saying which part of it the data covers. It was packaged for the Neural Latents Benchmark, not as the paper's data release.

Notebooks

Every registered entry that uses this dandiset. An entry can hold several notebooks, and can use more than one dandiset.

Reproduction Active and passive arm movements in somatosensory area 2 (Chowdhury et al. 2020)

Tests the active/passive findings of Chowdhury, Glaser and Miller (2020) on dandiset 000127, which holds one of the paper's twelve sessions. It compares hand and arm kinematics between active reaches and passive bumps, measures how well single neurons tell the two apart, and fits hand-only and whole-arm models of each neuron's firing. The paper's two-workspace experiment cannot be tested because those sessions are not in the dandiset. All files

Reproducing Chowdhury, Glaser & Miller (2020) from dandiset 000127
reproduce_chowdhury_2020.ipynbExecuted, with outputs · Source

Verification

Round 1
2026-10-13
Agreed with 5 of 6 outcomes. Independent agent run by CatalystNeuro (same harness, separate context, did not write the notebook). Reviewed: First draft, with the author's conclusions withheld. Only the six claims that were tested.
- fig8-hand-speed-similar: the verifier’s outcome was “Reproduced”. The author's draft said partially reproduced because passive movements are 13% faster. The verifier judged the paper's qualitative "similar" to hold, as the traces overlap in the window. The author adopted that.
Round 2
2026-10-13Agreed with 10 of 10 outcomes. The same independent agent, second round. Reviewed: The revised notebook with the author's conclusions. For the four blocked claims the verifier checked the dandiset's file list and metadata on the archive itself and found two files, the training and test splits of one active/passive session of one monkey.

Method differences

How the tests differed from the paper’s method or data, and the claims each applies to.

One session from one monkey; the paper has four sessions from two.fig8-hand-speed-similar fig8-arm-kinematics-separable fig8-neurons-separable fig9-whole-arm-fits-better fig10-inconsistent-across-conditions fig11-separability-vs-consistency
The whole-arm model is hand plus joint kinematics, as for fig9-whole-arm-fits-better.fig10-inconsistent-across-conditions fig11-separability-vs-consistency
Joint angles and velocities stand in for the hand and elbow marker kinematics, which are not in the dandiset.fig8-arm-kinematics-separable
Neurons firing below 1 Hz in the window (15 of 65) are left out; the paper states no criterion.fig8-neurons-separable
The whole-arm model is hand plus joint kinematics; the marker positions the paper's model uses are not in the dandiset.fig9-whole-arm-fits-better
The hand-only model is two-dimensional, the fits carry a small ridge penalty, neurons below 1 Hz are left out, and no Bonferroni correction is applied.fig9-whole-arm-fits-better
Pseudo-R2 for one trial type is computed on a repeat's pooled held-out predictions, against that type's own mean firing; the paper does not say which mean it used.fig10-inconsistent-across-conditions

Provenance

Attempt record: 000127-chowdhury-2020-01
Paper: Area 2 of primary somatosensory cortex encodes kinematics of the whole arm
Date: 2026-10-13
Made by: An agent, run by CatalystNeuro
Dandiset version: 0.220113.0359
Harness: Claude Code
Model: not recorded
Protocol: pilot-0

Other dandisets that name the same paper

← Dandiset 000121 Dandiset 000130 →