Claims 23
From the paper 10.1016/j.isci.2025.112767 · claims file · checked 2026-10-22: ready
- zenodo-descriptors-link
The per-cell descriptors computed from the dandiset (event rate, IEI CV, mean amplitude, spontaneous and task) agree with the per-cell table the paper deposited on Zenodo, so that table can give the cells' first occurrences.
How it will be judged
Cells matched by session and cell name (C, D, reading i). The link holds (reproduced) when at least 90% of matched cells have event rate and amplitude within 1% of the table's values (rounding), partially reproduced for 50 to 90%, not reproduced below 50%. The number of unmatched cells on either side is reported.
- cell-counts
In SMA 129, 61 and 63 cells were identified in monkeys Q, F and U, and in M1 37 and 23 cells in monkeys U and V.
In the SMA, we identified 129, 61, and 63 cells in monkeys Q, F, and U, respectively. In M1, we identified 37 and 23 cells in monkeys U and V, respectively
Results, General description of calcium activity, paragraph 2 (Table S2)How it will be judged
Accepted cells per site over the sessions listed in the opening comment. Readings (i) unique cells, each counted once (C), as Fig. 2A treats 129 Q cells over seven sessions; (ii) the sum of accepted cells over sessions. The Limitations call 23 to 129 the number imaged per session, which fits neither reading for Q (129 cells over seven sessions, Fig. 2A); the two readings are of which count is meant, so either may match. Table S2, which gives the counts, is not in the text read. Reproduced when all five counts are within tolerance under one reading.
- cells-tracked-18pct
About 18% of cells were tracked across up to three recording sessions.
We could track a small portion of cells (18%) across up to three recording sessions.
Results, General description of calcium activity, paragraph 2How it will be judged
Needs the cross-session cell identity (C). The fraction of unique cells, over all five sites, that appear in two or more sessions; the maximum number of sessions per cell is reported. Blocked when no cross-session identity is available from the dataset or the Zenodo table.
- fig1d-rate-no-change
Calcium event rates did not differ between the spontaneous and reaching task conditions in either SMA or M1.
The calcium event rates and the coefficient of variation of the inter-event intervals (IEI CV) did not differ between the spontaneous and reaching task conditions in either cortical region
Results, General description of calcium activity, paragraph 3; Figure 1DHow it will be judged
A claim of absence, on the paper's test, a Wilcoxon signed-rank test of spontaneous against task event rate, paired by cell, cells with both parts (C, D), with Benjamini-Hochberg FDR and significance at p < 0.05 (Methods, Statistical comparisons; Figure 1 legend). The grouping and the FDR family are not given; readings, groups (i) per region with monkeys pooled, (ii) per monkey and region (Q SMA, U SMA, U M1, and V M1 when its second part is taken as task); families (a) all spontaneous against task tests of the three measures of Fig. 1D-F, (b) the tests of this measure only; with D readings (i) and (ii). Reproduced when no reading finds a significant difference, not reproduced when every reading does, blocked as methods_underspecified when they differ. Median rates per condition and the effect size are reported.
- fig1e-ieicv-no-change
The IEI CV did not differ between the spontaneous and reaching task conditions in either SMA or M1.
The calcium event rates and the coefficient of variation of the inter-event intervals (IEI CV) did not differ between the spontaneous and reaching task conditions in either cortical region
Results, General description of calcium activity, paragraph 3; Figure 1EHow it will be judged
As fig1d-rate-no-change, for the IEI CV. Cells with fewer than two inter-event intervals in a part have no CV and are left out.
- fig1f-amplitude-u-sma-spont-higher
In monkey U's SMA, event amplitude was significantly higher in the spontaneous state than during the reaching task.
In the SMA, the amplitude of the events was significantly higher in the spontaneous state, compared to the reaching task recordings in monkey U
Results, General description of calcium activity, paragraph 3; Figure 1FHow it will be judged
Wilcoxon signed-rank test of spontaneous against task mean amplitude, paired by cell, for U SMA cells with both parts (C, D), FDR as in fig1d-rate-no-change, families (a) and (b), D readings (i) and (ii). Reproduced when every reading gives corrected p < 0.05 with the spontaneous median higher, not reproduced when none does, blocked as methods_underspecified when they differ.
- fig1d-rate-m1-above-sma-u
In monkey U the event rate was higher in M1 than in SMA, in both the spontaneous and the reaching task conditions.
In monkey U, we found that the event rate was higher in M1 than in SMA, both during the spontaneous and reaching task conditions
Results, General description of calcium activity, paragraph 3; Figure 1DHow it will be judged
For each condition, a Wilcoxon rank-sum test of U M1 cells against U SMA cells (C, D, readings i and ii). The Methods apply Benjamini-Hochberg FDR, the Results do not say it for these tests; readings uncorrected, and corrected over the two region comparisons of each measure. Per reading, reproduced when both tests give p < 0.05 with the M1 median higher, partially reproduced when one does; the outcome is the one every reading gives, blocked as methods_underspecified when they differ.
- fig1f-amplitude-m1-above-sma-u-task
In monkey U, event amplitude during the task was higher in M1 than in SMA.
Similarly, the amplitude of the events was higher for the task in M1, compared to SMA
Results, General description of calcium activity, paragraph 3; Figure 1FHow it will be judged
As fig1d-rate-m1-above-sma-u, for the task mean amplitude. Reproduced when every reading gives p < 0.05 with the M1 median higher, not reproduced when none does, blocked as methods_underspecified when they differ.
- fig2-q-sma-modulated-34pct
In monkey Q's SMA, 43 of 129 cells (34%) significantly changed their activity after the rewarded target appeared.
We found that in monkey Q's SMA, 34% (43/129) of cells significantly modulated their activity in response to the presentation of the rewarded target and the immediately ensuing movement
Results, Calcium traces in SMA and M1 show changes in activity, paragraph 2; Figure 2AHow it will be judged
Method Z on the seven Q SMA sessions with TTL data, each cell in its first session. Reproduced when the count of modulated cells is within tolerance under the main method; the variants of Z are reported.
- fig2c-q-sma-direction
Of the 129 Q SMA cells, 11 (9%) had task responses that were not direction-related and 32 (25%) responded more for one target position.
In 11/129 cells (9%), the activity was not direction-related, but 32/129 cells (25%) showed larger changes in activity in trials where the target was located at a specific position
Results, Calcium traces in SMA and M1 show changes in activity, paragraph 2; Figure 2B and 2CHow it will be judged
Method Z, classes as defined there. Reproduced when both counts are within tolerance.
- fig2-q-magnitude-no-difference
In Q's SMA the size of the response did not differ between cells whose activity rose and cells whose activity fell, and was similar for all target positions.
The magnitude of the responses did not differ between cells responding with an increase or decrease in firing (Mann-Whitney rank-sum test) and was similar regardless of the target position (Kruskal-Wallis test; data not shown)
Results, Calcium traces in SMA and M1 show changes in activity, paragraph 2 (data not shown)How it will be judged
A claim of absence, on the paper's tests (Methods, Alignment to behavioral events). Magnitude is the absolute change in mean z between the 1 s before and the 1 s after target onset, per modulated cell and significant condition; increase or decrease is the sign of the change (the code's Ratio1). A Mann-Whitney test of increase against decrease magnitudes, and a Kruskal-Wallis test across right, center and left, each at p < 0.05. Which values enter is not given; readings (i) one value per significant cell and condition, (ii) one value per modulated cell (mean over its significant conditions). Reproduced when no reading finds a difference in either test; not reproduced when every reading finds one in either; blocked as methods_underspecified otherwise.
- figs1-u-sma-direction
In monkey U's SMA, 6 of 14 cells (43%) were modulated at rewarded target onset, all six direction-related, and the proportions were the same when aligned to movement onset.
In monkey U's SMA, we found that 43% (6/14) cells had significant modulation of activity when the calcium traces were aligned to the appearance of the rewarded target, and in all cases (6/6) these changes in activity were related to the direction of the movement. These proportions were the same when the data were aligned to the movement onset
Results, Calcium traces in SMA and M1 show changes in activity, paragraph 3; Figure S1How it will be judged
Method Z on the U SMA session with TTL data (14 cells), for both alignments. Which session holds the TTL data is the author's to find; the Figure S1 legend is not in the text read. Reproduced when all four counts match; partially reproduced when the target-onset counts match and the movement-onset counts do not.
- figs1-u-m1-target-onset
In monkey U's M1, aligned to rewarded target onset, 4 of 6 cells (67%) changed their activity, none direction-related.
In monkey U's M1, when the calcium traces were aligned to the appearance of the rewarded target, 4/6 cells (67%) showed changes in activity, without directional relation.
Results, Calcium traces in SMA and M1 show changes in activity, paragraph 4; Figure S1How it will be judged
Method Z on the U M1 session with TTL data (6 cells), aligned on side target onset. Reproduced when both counts match.
- figs1-u-m1-movement-onset
In monkey U's M1, aligned to movement onset, all 6 cells were modulated and half of them were direction-related.
When the data were aligned to the movement onset, 100% (6/6) cells were modulated, and 50% of them showed direction-related activity
Results, Calcium traces in SMA and M1 show changes in activity, paragraph 4; Figure S1; Discussion, paragraph 7How it will be judged
Method Z on the U M1 session with TTL data, aligned on the center release. Reproduced when both counts match; partially reproduced when the modulated count matches only.
- fig3-synchronized-pairs-exist
Pairs of neurons with synchronous activity occur in both SMA and M1, in both the spontaneous and the task conditions.
Pairs of neurons exhibited neuronal synchrony in both the SMA and M1 during the spontaneous and task conditions
Results, Cells in SMA and M1 show coactivation, paragraph 1; Figure 3A to 3CHow it will be judged
Method J. Reproduced when, in each of SMA and M1 and each condition, at least one session has pairs with |Z-Jaccard| > 1.96, the paper's definition of a synchronized pair. The paper compares the proportions with no chance level; the proportion per session and the 0.05 a two-sided 1.96 threshold gives under the null are reported, not used to decide.
- fig3ab-synchrony-not-distance
Synchrony between cell pairs was not related to the distance between their centroids, in SMA or M1.
The cell pairs exhibiting synchrony were not related to the centroid distances in either SMA or M1
Results, Cells in SMA and M1 show coactivation, paragraph 1; Figure 3A and 3B; Discussion, paragraph 9How it will be judged
The paper shows a loess smoother (span 1) of Z-Jaccard on centroid distance per pair for example sessions of monkey U (SMA and M1, both conditions) and reports no test; the legend calls the panels correlation plots. Method J; distance in um (pixels times 3.2, code). Open details, with readings. The test, (i) Pearson and (ii) Spearman correlation of Z-Jaccard with distance, two-sided p < 0.05; the example sessions, which are not named, each U SMA and U M1 session (which session is meant, so any may match). Per test reading, reproduced when there is a U SMA session and a U M1 session where neither condition shows a significant correlation, not reproduced when every U session shows one in some condition; the outcome is the one both test readings give, blocked as methods_underspecified when they differ. The loess curves and the correlations of all sessions of all monkeys are reported.
- fig3c-sync-no-condition-change
The proportion of synchronized cell pairs did not differ between spontaneous and task conditions in SMA (p = 0.07) or M1 (p = 0.22).
The proportion of cells that were synchronized did not significantly differ between spontaneous and reaching tasks in either SMA or M1
Results, Cells in SMA and M1 show coactivation, paragraph 1; Figure 3CHow it will be judged
A claim of absence, on the paper's test (Methods; 1_JaccardIndexAnalysisAllSessions.R). Per session and condition, the proportion of pairs with |Z-Jaccard| > 1.96 (method J); per region, a two-sided Wilcoxon signed-rank test of spontaneous against task, paired by session, over the sessions with both conditions, at p < 0.05. Reproduced when neither region shows a significant difference; not reproduced when either does. The p-values, the median change and the code's 10-shift null for SMA are reported.
- fig4a-example-sequences-q
In an example session of monkey Q's SMA with 40 cells, 13 sequences were detected in the spontaneous condition and 17 during the task, and the specific sequences differed between the two even where they involved the same cells.
In an example imaging session in monkey Q's SMA, we detected 13 such sequences in spontaneous and 17 sequences during the task ... While some of the sequences detected involved the same cells ... the specific sequences differed
Results, Cells in SMA and M1 show sequential activation patterns, paragraph 2; Figure 4AHow it will be judged
Method S on the Q SMA session with 40 cells (Figure 4A legend); the session is not named, so every Q SMA session with 40 accepted cells is a reading of which session is meant. Per reading of S, reproduced when for some such session both counts are within tolerance and the two sets of sequences (ordered cell lists) are not the same; the outcome is the one every reading of S gives, blocked as methods_underspecified when they differ; the number of sequences found in both is reported.
- fig4b-example-three-segments-q
In another Q SMA session with 18 cells and 20 min of spontaneous recording, 8, 13 and 9 sequences were found in the two spontaneous halves and the task, and the sequences differed between all three segments.
We detected 8, 13, and 9 unique sequences during ... The detected sequences differed between all three states
Results, Cells in SMA and M1 show sequential activation patterns, paragraph 2; Figure 4B; Discussion, paragraph 10How it will be judged
Method S, spontaneous split into its first and second 10 min, on the Q SMA session with 18 cells and a spontaneous part of about 20 min; each such session is a reading of which session is meant. Per reading of S, reproduced when for some such session the three counts are within tolerance and no two segments have the same set of sequences; the outcome is the one every reading of S gives, blocked as methods_underspecified when they differ. The number of sequences shared by each pair of segments, and by all three (the preprint says none), is reported.
- fig4d-sequence-measures-vary
The number of sequences per cell, their repetition and the proportion of cells in sequences varied across conditions and imaging sites.
We found that these factors varied across conditions and across imaging sites, similar to observations made in monkey Q's SMA.
Results, Cells in SMA and M1 show sequential activation patterns, paragraph 3; Figure 4DHow it will be judged
Method S on every session and segment (spontaneous 1, spontaneous 2 where there are 20 min, task). Per segment, the number of sequences over the number of cells, the median number of repetitions of its sequences, and the fraction of cells in at least one sequence (Figure 4D legend). The paper reports no test. Reproduced when the per-site medians of each measure are not all equal across segments, and not all equal across sites; the per-session values are compared with the Zenodo Fig. 4D table (doi:10.5281/zenodo.14976005) where they match by session.
- fig4c-sequence-cells-not-clustered
Cells involved in sequences were spread across the field of view, with no preferred spatial relation between them.
Plotting the distribution of distances between pairs of cells that were involved in sequences, as well as the overall distribution of distances between all detected cells showed that there is no preferred spatial relation between cells involved in sequences
Results, Cells in SMA and M1 show sequential activation patterns, paragraph 4; Figure 4C and Figure S2How it will be judged
The paper compares distributions by eye and reports no test; Figure S2 is not in the text read. What is compared is read two ways, (i) the centroid positions of cells in sequences against those of all cells, spontaneous condition, Q SMA and U M1 (Figure 4C legend), (ii) the distances between cells of the same sequence against the distances between all cell pairs (Results). The test is open; readings a two-sample Kolmogorov-Smirnov test and a Mann-Whitney test, p < 0.05, on distances (ii) or on each centroid coordinate (i). Reproduced when no reading finds a difference; not reproduced when every reading does; blocked as methods_underspecified otherwise.
- histology-lens-placement
Lenses and prisms were placed in the arm regions of SMA and M1, in deep cortical layers, with monkey Q's SMA lens at the anterior edge of SMA.
Postmortem histological examination confirmed that the lens and prism were placed in the arm regions of SMA and M1
Results, Expression of GCaMP6f and lens localization, paragraphs 1 and 2; Figure S3AHow it will be judged
Histology, not imaging data. Judged from the MAP2 sections on Zenodo against the atlas of Paxinos et al. (2000); reproduced when each of the five lens tracks ends in deep layers of the arm region of SMA or M1 and Q's SMA lens lies at the anterior edge of SMA (Results).
- histology-gcamp-not-gabaergic
GCaMP6f expression was robust around most imaging sites, sparse in monkey U's SMA, and GCaMP6f-positive cells were not GABAergic.
We observed robust GFP expression around the lens/prism for most sites, except in monkey U's SMA, where the GFP expression was sparse ... We confirmed that GCaMP6f expression was specific for non-GABAergic neurons by performing a double immunofluorescence assay for GABA and GFP
Results, Expression of GCaMP6f and lens localization, paragraph 2; Figure S3B to S3DHow it will be judged
Histology, not imaging data. From the GFP and GABA double-label images on Zenodo, reproduced when no GFP-positive cell body is also GABA-positive; GFP density per site is described.
Pipeline
- Paper
- done (1 of 1)
- Claims
- done
- Claims check
- done (ready)
- Attempt
- todo
- Questions
- none
- Verified
- none