DANDI:000005

Reproducing Yu et al. (2016) from dandiset 000005 22

Notebook reproduce_yu_2016.ipynb

• ✕Not reproducedResults, Chronic extracellular recordings from somatosensory thalamus, paragraph 1; Figure 1b,d; Figure 5b,c

  VPM neurons respond to active touch with a sharp, short-latency increase in spike rate (0.6 +- 0.5 spikes per touch, median 0.4; latency 3.2 +- 1.0 ms, median 2.5 ms).

  Neurons responded to active touch during behavior with an increase in spike rate (Fig. 1b). The peristimulus time ... histogram aligned to touch onset showed a sharp peak in spike rate (mean ± s.d., 0.6 ± 0.5 spikes per touch; median, 0.4 spikes per touch), with short latency after touch onset (mean ± s.d., 3.2 ± 1.0 ms; median, 2.5 ms)

  Results, Chronic extracellular recordings from somatosensory thalamus, paragraph 1; Figure 1b,d; Figure 5b,c

  Over the 29 VPM units, spikes per touch (0 to 25 ms after minus 0 to 25 ms before onset, principal-whisker touches with ITI of at least 25 ms) average 0.403, median 0.268, outside 0.6 +- 0.093 and 0.4 +- 0.04. 25 units pass the latency exclusion; the latency has mean 5.64 ms and median 2.0 ms with the bin's left edge (O9 a) and 6.14 and 2.5 ms with its centre (O9 b), one unit at 71.5 ms. What the answer changed. The answer of 2026-10-20 says the paper does not fix the bin position, and the criterion now runs both; this attempt already showed both, the spikes per touch fail under either, and the outcome is unchanged. The measured latencies are those of O9 b, the reading under which the median matches. The earlier attempt found this claim reproduced under its own criterion (a fraction of units with a one-sided test at p < 0.01, 0.83); the new criterion tests the paper's spikes per touch and latency, and the files' one principal whisker per file lowers the spikes per touch (11 of 20 units in files with several whiskers respond most to another whisker; each unit's best whisker gives a mean of 0.485, a selection biased upward that the criterion does not make).

  How the test differed from the paper’s method

  • The files give one principal whisker per VPM file, though the session description says each unit has its own; the file's principal whisker is used for every unit.
• ✕Not reproducedResults, Dynamics of L4 fast-spiking interneurons, paragraphs 1 and 2; Figure 4d-h

  In whole-cell L4 FS interneurons (n = 7), touch depolarizes the membrane by 7.0 +- 5.2 mV with onset 4.0 +- 0.6 ms, and whisking depolarizes it by 2.7 +- 2.1 mV toward threshold.

  Membrane potential depolarized (mean ± s.d., 7.0 ± 5.2 mV; median, 6.7 mV; n = 7 neurons, Fig. 4d,f) ... Response latencies were short for membrane potential depolarization (onset ... 4.0 ± 0.6 ms; median, 4.2 ms, n = 7 neurons; Fig. 4d,f) ... after onset of whisking, the membrane potential of FS neurons depolarized closer to spike threshold (mean ± s.d., 2.7 ± 2.1 mV; median, 2.4 mV, n = 7 neurons; Fig. 4e–h)

  Results, Dynamics of L4 fast-spiking interneurons, paragraphs 1 and 2; Figure 4d-h

  5 whole-cell FS cells (paper 7). Touch amplitude 5.56 mV and whisking depolarization 2.54 mV (onset windows; 2.24 forward filter; 1.47 to 2.55 over the O10 and O11 readings, within tolerance in 12 of 16) are within tolerance; the threshold-distance reading gives 1.01 mV, outside. The touch latency with dVm/dt at the sampling rate averages 1.88 ms (median 1.35), outside 4.0 +- 0.4. In the two cells at 0.00 and 0.30 ms the first crossing is a single sample of dVm/dt just above the threshold (0.96 against 0.71 and 3.05 against 1.85 mV/ms), in averages whose Vm lies 0.26 to 0.46 mV above the pre-touch mean at onset; subtracting the baseline's mean dVm/dt, which decides nothing, gives 1.76 ms. So the touch values are not all within tolerance. New claim.

  How the test differed from the paper’s method

  • The PV-ChR2 cell, FS in Supplementary Table 2, has spikes 0.44 ms wide here and falls outside the 0.35 ms rule for lines without a tag; adding it is shown as a variant.
• ✕Not reproducedResults, Dynamics of L4 fast-spiking interneurons, paragraph 3; Figure 5b,d,e

  L4 excitatory neurons respond mostly to touch (0.3 +- 0.4 spikes per touch, n = 85) and only weakly to whisker movement; their rate does not change significantly between nonwhisking and whisking (0.4 to 0.6 spikes/s, P = 0.20).

  Excitatory neurons responded mostly to touch (mean ± s.d., 0.3 ± 0.4 spikes per touch; median, 0.1 spikes per touch; n = 85 neurons; Fig. 5b,d) and showed weak response to whisker movement ... (nonwhisking ... 0.4 ± 0.6 spikes per s; median, 0.1 spikes per s; whisking ... 0.6 ± 0.9 spikes per s; median, 0.2 spikes per s; n = 85, P = 0.20; Fig. 5d,e). In contrast to VPM and L4 FS neurons, L4 excitatory neurons responded selectively to touch

  Results, Dynamics of L4 fast-spiking interneurons, paragraph 3; Figure 5b,d,e

  What the paper reports

  • nonwhisking against whisking spike rate over L4 excitatory neurons: p = 0.2 (n = 85 neurons)

  The 97 excitatory cells fire more during whisking than without it under every filter and O11 reading (signed-rank p = 9.1e-07 zero-phase and 8.7e-06 forward for the main epochs, at most 8.7e-06 over all; higher in 50 to 58), so the paper's test finds a difference where the paper found none (P = 0.20). Spikes per touch 0.320 is within tolerance; the nonwhisking (1.53 to 1.59) and whisking (3.38 to 3.64 spikes/s) rates are above 0.4 and 0.6. Cell-attached cells fire more than whole-cell ones (2.45 against 0.70 spikes/s without whisking), and the whole-cell cells, whose detection is clean, still fire above the paper's 0.4 spikes/s. The cell-attached detection has 5.0% of interspike intervals under 1.5 ms in the median cell, but a 2 ms dead time removes only 8.1% of the cell-attached E spikes (mean rate 6.20 to 5.70 spikes/s over all time), so double detection does not explain the high rates. The earlier attempt also found it not reproduced, under a touch-selectivity index the paper does not use (one-sided p = 0.074 and 0.090); the same outcome now comes from the paper's own test of whisking against nonwhisking rate.

  How the test differed from the paper’s method

  • The group holds 97 cells against 85, with the excitatory cells of ION-cut mice and 26 VGAT-ChR2 cells that have too few cortical pulses to be tagged; FS cells that the rules cannot find (the fourth PV cell, and 5 of the 29 VGAT-ChR2 FS cells of Supplementary Table 2 that no test here finds driven) may be among them. The 26 untested cells include a whole-cell cell with 11,087 spikes 0.37 ms wide and a cell-attached cell with 16,350 spikes, both likely FS.
• ✕Not reproducedResults, Layer 4 FS interneurons suppress reafference in L4, paragraph 2; Figure 8e; Supplementary Figure 13b

  PV interneuron suppression does not change mean touch-evoked spiking of L4 excitatory neurons significantly (0.27 to 0.32 spikes per touch, P = 0.09); its effect is heterogeneous, with weakly responding neurons responding more and strongly responding neurons unchanged or less.

  PV interneuron suppression also modulated touch-evoked spiking in L4 excitatory neurons, but they did so in a heterogeneous manner (mean ± s.e.m., from 0.27 ± 0.06 spikes per touch to 0.32 ± 0.06 spikes per touch, n = 40 neurons, P = 0.09, Wilcoxon signed-rank test; Fig. 8e and Supplementary Fig. 13). Neurons that responded weakly to touch tended to respond more strongly with PV interneuron suppression.

  Results, Layer 4 FS interneurons suppress reafference in L4, paragraph 2; Figure 8e; Supplementary Figure 13b

  What the paper reports

  • Wilcoxon signed-rank test, p = 0.09 (n = 40 neurons)

  Spikes per touch (4 to 20 ms after minus 0 to 16 ms before, ITI of at least 10 ms) rise with the light from 0.339 to 0.448 (40 cells, signed-rank p = 0.044, control in the light span) or from 0.304 to 0.442 (41 cells, p = 0.0091, control anywhere), so the test finds a change. Cells below the median control response rise by 0.123 on average, cells above it by 0.095. New claim.

• ✕Not reproducedResults, Layer 4 FS interneurons suppress reafference in L4, paragraph 3; Figure 8f,g

  Suppressing PV interneurons improves touch detection (AUC) for L4 excitatory neurons with weak touch responses and diminishes it for neurons with robust touch responses.

  Consistent with the heterogeneous changes in spikes per touch, suppression of PV interneurons improved AUC for neurons with weak responses to touch and diminished AUC for neurons with robust responses to touch (Fig. 8f,g).

  Results, Layer 4 FS interneurons suppress reafference in L4, paragraph 3; Figure 8f,g

  With the fixed seeds (29 and 31), cells whose AUC rises with the light have a lower control touch-evoked rate than cells whose AUC falls, 5.93 against 10.87 spikes/s (reading a, 23 and 12 cells) and 8.14 against 8.18 (reading b, 17 and 16). Over 20 further seeds, reading (a) holds in 19 and reading (b) in 8 (rising group 8.14 to 8.53 against falling group 7.86 to 8.18 spikes/s), so under (b) the sign of a 0.04 spikes/s difference is set by the resampling and the reading does not match; the criterion needs both readings. No rank-sum test separates the groups with the fixed seeds (p = 0.37 and 1, reported only). Changed after verification round 1 from reproduced, on the seed check the verifier asked for. New claim.

  How the test differed from the paper’s method

  • Where an outcome depends on the random seed, a reading counts as matching only when it matches under most of 20 seeds, a rule of this attempt.
• ✕Not reproducedResults, Layer 4 FS interneurons suppress reafference in L4, paragraph 4; Figure 8h

  Neurons whose AUC rose with PV suppression gained touch-evoked spiking (3.4 to 8.7 spikes/s); neurons whose AUC fell kept their touch-evoked response (7.3 to 7.0 spikes/s) but had higher baseline rates (0.9 to 2.6 spikes/s).

  L4 excitatory neurons with increased AUCs showed a population increase in touch-evoked responses on average (from 3.4 ± 0.8 spikes per s to 8.7 ± 1.9 spikes per s, P = 0.0001). Neurons with reduced AUC showed unchanged touchevoked responses as a population (from 7.3 ± 1.5 spikes per s to 7.0 ± 1.4 spikes per s, P = 0.98, average 0–250 ms after touch onset) ... but elevated baseline spike rates (from 0.9 ± 0.2 spikes per s to 2.6 ± 0.6 spikes per s, mean ± s.e.m., P = 0.0002; Fig. 8f–h).

  Results, Layer 4 FS interneurons suppress reafference in L4, paragraph 4; Figure 8h

  What the paper reports

  • touch-evoked rate, control against eNpHR, neurons with increased AUC: p = 10⁻⁴ (n = 19 neurons (Fig. 8h))
  • touch-evoked rate, control against eNpHR, neurons with reduced AUC: p = 0.98 (n = 21 neurons (Fig. 8h))
  • baseline rate, control against eNpHR, neurons with reduced AUC: p = 2 × 10⁻⁴ (n = 21 neurons (Fig. 8h))

  With the fixed seeds the readings differ. Under (a) the increased-AUC group's response rises (9.99 to 15.16 spikes/s, p = 4.9e-05), the reduced-AUC group's does not change (15.98 to 17.85, p = 0.11) and its baseline rises (5.11 to 12.08, p = 0.0049), with group sizes 23 and 12 and rates outside tolerance; under (b) the reduced-AUC group's response changes (12.37 to 14.95, p = 0.021). Over 20 further seeds the three tests hold in 9 seeds under (a) and in none under (b), where the reduced-AUC group's response changes in every seed (p = 0.015 to 0.045), so neither reading matches under most seeds and the claim is not reproduced. Group sizes range from 15 to 22 (increased) and 11 to 19 (reduced) over seeds. Changed after verification round 1 from blocked (methods underspecified), on the seed check. New claim.

  How the test differed from the paper’s method

  • Where an outcome depends on the random seed, a reading counts as matching only when it matches under most of 20 seeds, a rule of this attempt.
• ◐Partially reproducedOnline Methods, Behavior, paragraph 2

  Mice touched the pole often when it was within reach (lick trials, 5.7 +- 5.3 touches per trial over 4,147 trials) and rarely when it was distal (0.9 +- 3.4 over 4,173 trials); performance was 0.72 +- 0.11 in cortical and 0.71 +- 0.07 in VPM recordings.

  Mice touched the pole at a proximal location multiple times (number of touches ... Mice touched the pole at a distal location less often (number of touches, 0.9 ± 3.4, mean ± s.d., 4,173 trials) and withheld licking. Average performance (proportion of correct trials) was 0.72 ± 0.11 (mean ± s.d., during the recordings of 111 cells) for cortical recordings and 0.71 ± 0.07 for VPM recordings (during the recordings of 29 cells).

  Online Methods, Behavior, paragraph 2

  Under light reading (x), every trial, with all sessions (b) and principal-whisker touches (i), mice touch the pole 6.04 times per lick trial (9,794 trials) and 0.85 times per no-lick trial (8,223 trials), both within tolerance; no other reading puts both counts within tolerance. Under light reading (y), trials without light only, the same recordings give 5.74 and 0.80 (8,219 and 6,713 trials), and the no-lick count is outside tolerance. The matching trial counts are more than twice the paper's 4,147 and 4,173, and no reading gives the paper's counts (the nearest, 5,212 and 5,149, come from reading a with every trial). Performance averages 0.681 under (x) and 0.694 under (y) over all 122 cortical cells, within tolerance (0.703 and 0.718 without the flagged mice, a variant that no longer decides), and 0.585 over the 29 VPM units under both (the VPM sessions have no light), outside 0.71 +- 0.071. New claim.

  How the test differed from the paper’s method

  • The 111 control cortical cells of reading (a) cannot be identified, as the files do not mark the ION-cut mice; every cortical cell counts, as in every other group, and leaving out the cells of the 4 flagged mice is a variant.
• ◐Partially reproducedResults, Chronic extracellular recordings from somatosensory thalamus, paragraphs 1 and 2; Figure 1c,e; Figure 5d,e; Supplementary Figure 2b,c

  VPM neurons fire at 5.1 +- 5.5 spikes/s without whisker movement or touch; most (25 of 29) increase their rate during whisking without touch, and the mean rate rises from 5.1 to 13.5 spikes/s. Three of 29 do not increase with whisking but still respond to touch.

  Typical VPM neurons showed significant spike rates (mean ± s.d., 5.1 ± 5.5 spikes per s; median, 3.3 spikes per s; n = 29 neurons), even in the absence of whisker movement and touch. ... Most VPM neurons increased their spike rate during whisking in the absence of touch (25 out of 29 neurons ... The average spike rate increased severalfold after onset of whisker movement (from 5.1 ± 5.5 spikes per s to 13.5 ± 12.7 spikes per s; P = 4 × 10−6; n = 29 neurons, Wilcoxon signed-rank test

  Results, Chronic extracellular recordings from somatosensory thalamus, paragraphs 1 and 2; Figure 1c,e; Figure 5d,e; Supplementary Figure 2b,c

  What the paper reports

  • nonwhisking against whisking spike rate over VPM neurons: Wilcoxon signed-rank test, p = 4 × 10⁻⁶ (n = 29 neurons)

  The whisking rate exceeds the nonwhisking rate over the 29 units under every filter (O1), sample (O2) and touch reading (O11) (signed-rank p = 5.2e-08 zero-phase and 7.1e-08 forward for the main epochs; higher in 28). Significant increases number 21 to 25 over the 16 readings, outside 25 +- 2.5 under some. Mean nonwhisking rate 3.74 to 3.96 spikes/s and median 1.71 to 1.86, outside tolerance under every reading; mean whisking rate 11.01 to 12.09, within 13.5 +- 2.36 under some readings only. The earlier attempt found this reproduced under its own criterion (a phase-modulation test, 0.88 of units); the new criterion asks for the paper's rates, which are lower here.

• ◐Partially reproducedResults, Membrane-potential dynamics of L4 excitatory neurons, paragraph 2; Figure 2f

  In L4 excitatory neurons (37 whole-cell), resting Vm spans -76 to -45 mV (-63 +- 8 mV); every neuron depolarizes after touch (latency 4.8 +- 0.8 ms, amplitude 6.0 +- 2.9 mV), while whisking causes only a small depolarization or hyperpolarization (1.2 +- 2.1 mV).

  Across the excitatory population, the resting membrane potential (Vm) spanned a large range (−76 to −45 mV, mean ± s.d., −63 ± 8 mV; median, −64 mV, n = 37 neurons ... Touches evoked short-latency depolarizations (latency: mean ± s.d., 4.8 ± 0.8 ms; median, 4.7 ms; amplitude: mean ± s.d., 6.0 ± 2.9 mV; median, 5.7 mV; n = 37 neurons) for all neurons (Fig. 2f) ... Whisking induced small depolarization or hyperpolarization (mean ± s.d., 1.2 ± 2.1 mV; median, 1.2 mV; Fig. 2f)

  Results, Membrane-potential dynamics of L4 excitatory neurons, paragraph 2; Figure 2f

  51 whole-cell excitatory cells (the paper has 37 control cells); 48 have at least 5 touches, and all 48 have a positive touch PSP and a latency. Resting Vm mean -61.69 mV (O4 a; -61.89 O4 b) is within tolerance under every reading. The touch amplitude (4.78 mV) and the latency (7.68 ms with dVm/dt at the sampling rate, median 4.90) are outside tolerance and do not depend on the whisking readings. A diagnostic shows that the latency rule picks single-sample excursions of dVm/dt in 5 cells (under 1 ms) and late crossings in others (up to 101 ms); subtracting the baseline's mean dVm/dt, which decides nothing, gives 7.48 ms. Under the 16 combinations of O10, O11 and the two filters, the whisking change with the onset windows (reading a) ranges from 0.65 to 0.87 mV and is within 1.2 +- 0.35 in 2 (zero-phase, O10 a, O11 b-i 0.87 mV and b-ii 0.86 mV); the measured value, 0.83 mV, is the first reading the criterion lists (zero-phase, O10 a, O11 a-i), outside tolerance, and with the epochs (reading b) it is 1.08 to 1.14 mV, within tolerance in all 16. Without the 8 cells of the flagged mice the amplitude is 5.54 mV and the latency 4.97 ms, both within; that variant does not decide. Amplitudes go down to 0.01 mV, so that every neuron depolarizes is a weak test. Trials of the PV-ChR2 cell held near 0 mV are left out, and its resting Vm is -38.8 mV. The 26 VGAT-ChR2 cells with too few pulses to tag count as excitatory and include a whole-cell cell with 11,087 spikes 0.37 ms wide, likely FS, so the group probably holds some FS cells. New claim.

  How the test differed from the paper’s method

  • The group includes the whole-cell cells of ION-cut mice, which the files do not mark, and 4 more PV-IRES-Cre whole-cell cells than Supplementary Table 2 counts.
  • Under O11 the 250 ms touch-free rule of whisking onsets uses the touches of the whiskers the reading names; the main rule (every tracked whisker bounding the epochs, principal-whisker touches in the 250 ms rule) is shown beside them.
  • Trials whose median Vm is above -20 mV (13 trials of the PV-ChR2 cell, held near 0 mV by current) are left out at rest, a standard choice.
• ◐Partially reproducedResults, Membrane-potential dynamics of L4 excitatory neurons, paragraph 2; Figure 2g,h

  Touch moves the membrane potential of L4 excitatory neurons much closer to spike threshold than whisking does (shift 6.1 +- 3.2 mV against 1.1 +- 2.0 mV).

  Compared to whisking, touch caused a larger shift of membrane potential toward spike threshold (whisking, 1.1 ± 2.0 mV; touch, 6.1 ± 3.2 mV; P = 1 × 10−7, Wilcoxon signed-rank test; Fig. 2g,h)

  Results, Membrane-potential dynamics of L4 excitatory neurons, paragraph 2; Figure 2g,h

  What the paper reports

  • touch against whisking shift toward threshold: Wilcoxon signed-rank test, p = 10⁻⁷ (n = 37 neurons)

  Over 50 whole-cell excitatory cells the touch shift exceeds the whisking shift in 43 or 44 under every filter and O11 reading (signed-rank p = 9e-11 to 1.3e-10). Means, whisking 0.84 to 1.16 mV (within 1.1 +- 0.33 under every reading), touch 4.60 to 4.67 mV (outside 6.1 +- 0.61). The two threshold readings O3 give the same values (a cell's median threshold differs by at most 0.0089 mV and 8 of 41,620 spikes change threshold). 17 cells have no spikes and take the default threshold of -42 mV. With the held trials of the PV-ChR2 cell left out, the main whisking shift is 0.84 mV, within tolerance, and the touch shift 4.62 mV, outside. New claim.

• ◐Partially reproducedResults, Dynamics of L4 fast-spiking interneurons, paragraph 2; Figure 5d,e

  Most L4 FS interneurons (23 of 26) increase their spike rate during whisking without touch and 3 of 26 decrease it; over the population the rate roughly doubles (to 21.1 +- 15.8 spikes/s).

  most L4 FS neurons significantly increased their spike rate during whisker movement without touch (23 of 26 neurons ... and a small subset (3 of 26 neurons ... decreased spike rate. Overall, after onset of whisking ... and the spike rate doubled (mean ± s.d., 21.1 ± 15.8 spikes per s; median, 14.6 spikes per s, n = 26 neurons, P = 2.7 × 10−5; Fig. 5d)

  Results, Dynamics of L4 fast-spiking interneurons, paragraph 2; Figure 5d,e

  What the paper reports

  • nonwhisking against whisking spike rate over FS neurons: p = 2.7 × 10⁻⁵ (n = 26 neurons)

  Over 17 FS cells the whisking rate exceeds the nonwhisking rate under every filter, sample and O11 reading (signed-rank p = 0.0002 to 0.0005). 13 or 14 cells increase and 1 to 3 decrease (rank-sum p < 0.05), so the increase count is outside 23 +- 2.3 under every reading. Mean whisking rate 19.02 to 19.17 spikes/s with the zero-phase filter (within 21.1 +- 3.1) and 17.77 to 17.91 with the forward filter (outside). New claim.

• ◐Partially reproducedResults, Inhibition is driven by thalamus, paragraph 1; Supplementary Figure 9a-f

  Thalamic photoinhibition hyperpolarizes L4 excitatory neurons and reduces their Vm fluctuations, with an onset latency of 4.2 ms, longer than the 2.8 ms latency of direct cortical GABAergic activation.

  photoinhibition of thalamus hyperpolarized excitatory neurons and reduced the amplitudes of their membrane voltage fluctuations (Supplementary Fig. 9a–e). Onset latency of Vm hyperpolarization induced by thalamus inactivation was 4.2 ms. This was longer than the 2.8-ms latency measured with direct cortical GABAergic activation (Supplementary Fig. 9f)

  Results, Inhibition is driven by thalamus, paragraph 1; Supplementary Figure 9a-f

  13 whole-cell excitatory cells with thalamic light (paper 12). Vm is lower with light in 12 (median -4.02 mV, signed-rank p = 0.00049) and Vm variance above 10 Hz lower in 10 (median ratio 0.55), under both window readings. With dVm/dt at the sampling rate the onset latency of the grand average is 0.10 ms for both the thalamic light (13 cells) and the cortical tag pulses (14 cells, now that every pulse of a train counts), against 4.2 and 2.8 ms, so neither latency is within tolerance and the thalamic one is not longer. The crossing at 0.1 ms is one sample that drops by 0.061 mV (thalamic) and 0.064 mV (cortical) and returns at the next sample, a transient at light onset; the first crossing 2 ms or more after onset, which decides nothing, is at 4.9 ms (thalamic) and 5.2 ms (cortical), but against different thresholds (-0.0206 and -0.0557 mV per sample; 1 and 3 of 250 baseline samples cross them), so it does not compare the two. The 1 ms bins (5.0 against 3.0 ms) and the 0.5 ms grand averages, which start to fall at about 4.5 to 5 ms (thalamic) and about 3 ms (cortical), put the thalamic onset later, as in the paper; neither is the criterion's rule and neither decides. New claim.

  How the test differed from the paper’s method

  • The latency rule for laser onset is the touch rule turned downward, an assumption the claims file states. The crossing at 0.1 ms is a one-sample transient at light onset.
  • 20 kHz averages are brought to 10 kHz by the mean of each pair of samples (standard).
• ◐Partially reproducedResults, Layer 4 FS interneurons suppress reafference in L4, paragraph 7; Supplementary Figure 13a,c

  Yellow light in PV-IRES-Cre mice expressing eNpHR3.0 strongly inhibits most L4 PV interneurons (3 of 4); their rate falls 3 ms after laser onset and L4 excitatory activity rises 2 ms later.

  First, the spike rate of L4 PV interneurons decreased 3 ms after laser onset, and activity in L4 neurons increased only 2 ms later (Supplementary Fig. 13c) ... Four L4 PV interneurons recorded. 3 out of 4 are strongly inhibited by eNpHR stimulation.

  Results, Layer 4 FS interneurons suppress reafference in L4, paragraph 7; Supplementary Figure 13a,c

  3 PV cells are identified by their inactivation (the paper has 4), and all 3 are strongly inhibited (light rate below the lower bound of the control 95% interval in nonwhisking and whisking periods). The count cannot fail, since the cells are found by their inactivation, and a fourth PV cell that the light did not inhibit would be classed excitatory. The parts that can fail fail: in the grand average PSTH aligned to laser onset the dPSTH/dt rule finds no fall for the 3 PV cells and a rise at 48.0 ms (bin edge, O9 a) or 48.5 ms (centre, O9 b) for the 42 excitatory cells, so neither latency is within tolerance under either bin position. New claim.

  How the test differed from the paper’s method

  • The latency rule for laser onset is the Methods' spike-latency rule, an assumption the claims file states. The PV grand average rises in the first 3 ms after onset (66.5 spikes/s at 1 ms) before falling to 0, possibly an onset artifact.
• ◐Partially reproducedResults, Layer 4 FS interneurons suppress reafference in L4, paragraph 1; Figure 8b-d; Supplementary Figure 13b

  Partially suppressing PV (FS) interneurons increases the spike rate of L4 excitatory neurons during nonwhisking (0.6 to 1.6 spikes/s) and whisking (0.8 to 2.7 spikes/s).

  Partially suppressing PV (FS) interneurons changed the responses of excitatory neurons (Fig. 8b–e; P = 1 × 10−6, Kruskal-Wallis test). L4 excitatory neurons increased their spike rate during both nonwhisking periods (from 0.6 ± 0.1 spikes per s to 1.6 ± 0.3 spikes per s; mean ± s.e.m., n = 40 neurons, P = 6 × 10−5) and whisking periods(from 0.8 ± 0.2 spikes per s to 2.7 ± 0.5 spikes per s, P = 2 × 10−6

  Results, Layer 4 FS interneurons suppress reafference in L4, paragraph 1; Figure 8b-d; Supplementary Figure 13b

  What the paper reports

  • rates of excitatory neurons across conditions: Kruskal-Wallis test, p = 10⁻⁶ (n = 40 neurons)
  • nonwhisking rate, control against eNpHR: p = 6 × 10⁻⁵ (n = 40 neurons)
  • whisking rate, control against eNpHR: p = 2 × 10⁻⁶ (n = 40 neurons)

  In the 42 excitatory cells of PV-IRES-Cre mice (paper 40) the light raises the nonwhisking rate (32 to 36 cells, p at most 7.4e-05) and the whisking rate (31 to 34 cells, p at most 0.0003) under every filter, window and O11 reading (Kruskal-Wallis p = 2.7e-05 to 0.00035, reported only). The rates are higher than the paper's (control 1.18 to 1.32 rising to 2.26 to 2.55 spikes/s without whisking, 3.36 to 4.31 rising to 5.69 to 6.63 with it, against 0.6 to 1.6 and 0.8 to 2.7), outside tolerance; 34 of the 42 cells are cell-attached, whose rates are high throughout the dandiset. The measured values are zero-phase, control in the light span, main epochs. The earlier attempt found this reproduced from 7 whole-cell cells inside the light (6 of 7 rising, p = 0.016); the new criterion uses all excitatory cells and also tests the paper's rates, which are not matched.

• ◐Partially reproducedResults, Layer 4 FS interneurons suppress reafference in L4, paragraph 1; Figure 8d

  With PV interneurons suppressed, L4 excitatory neurons become driven by whisker movement (whisking rate above nonwhisking, P = 0.02), while in control trials they are not (P = 0.55); their selectivity for touch is normally maintained in part by FS interneurons.

  L4 excitatory neurons were instead driven by whisker movements (P = 0.02, Fig. 8d), implying that their selectivity for touch was normally maintained in part by FS interneurons.

  Results, Layer 4 FS interneurons suppress reafference in L4, paragraph 1; Figure 8d

  What the paper reports

  • nonwhisking against whisking rate in eNpHR trials (Fig. 8d): p = 0.02 (n = 40 neurons)
  • nonwhisking against whisking rate in control trials (Fig. 8d, printed in the figure): p = 0.55 (n = 40 neurons)

  With the light, whisking rates exceed nonwhisking rates (31 to 33 of 42 cells, p = 1.7e-06 to 3.7e-05 over the filter, window and O11 readings), as the paper reports (P = 0.02), but in control trials they do too (27 to 35 of 42, p = 6.7e-06 to 0.00013), where the paper found none (P = 0.55). The earlier attempt found it not reproduced by a touch-selectivity index the paper does not use; the new criterion is the paper's test of whisking against nonwhisking rate, whose first part holds.

• ✓ReproducedResults, Chronic extracellular recordings from somatosensory thalamus, paragraph 2; Supplementary Figure 2d-f

  The spike rate of VPM neurons is modulated by the phase of whisker movement.

  The spike rate was modulated by the phase of the movement (Supplementary Fig. 2d).

  Results, Chronic extracellular recordings from somatosensory thalamus, paragraph 2; Supplementary Figure 2d-f

  25 to 27 of 29 units have r2 > 0.25 for a sinusoid fit of rate in 12 phase bins under both filters and every O11 reading (25 zero-phase and 27 forward for the main epochs). With spikes moved to the next trial, 16 of 29 units also pass, so the r2 threshold marks more than half of the units on a null; the criterion does not ask for a null, and the units' own spikes give a higher r2 (p = 0.0015). New claim.

• ✓ReproducedResults, Membrane-potential dynamics of L4 excitatory neurons, paragraph 2; Supplementary Figure 4

  The effective reversal potential of the touch response of L4 excitatory neurons is more depolarized than resting Vm, while that of whisking is close to rest and more hyperpolarized than that of touch.

  of the touch response was more depolarized than the resting Vm and sometimes crossed the spike threshold ... In contrast, the effective reversal potential for whisker movement was close to the resting Vm, below the spike threshold and more hyperpolarized than that for touch

  Results, Membrane-potential dynamics of L4 excitatory neurons, paragraph 2; Supplementary Figure 4

  Per neuron (47 or 48 with both fits) the touch reversal (median -50.0 mV, the mean of the middle two, -50.23 and -49.77 mV) lies above the whisking reversal (median -57.4 to -60.9 mV over the filters and the O10 and O11 readings) in 38 to 40 neurons and above the median resting Vm (-63.9 mV). Pooled fits give -31.90 mV for touch and -51.6 to -56.3 mV for whisking. New claim.

• ✓ReproducedResults, Dynamics of L4 fast-spiking interneurons, paragraph 2; Supplementary Figure 7d,g

  Spikes of L4 FS interneurons are coupled to whisking phase.

  Spikes of L4 FS neurons were also coupled to whisking phase (Supplementary Fig. 7).

  Results, Dynamics of L4 fast-spiking interneurons, paragraph 2; Supplementary Figure 7d,g

  15 or 16 of 17 FS cells have r2 > 0.25 under both filters and every O11 reading (16 for the main epochs; 5 and 3 with the double fit). With spikes moved to the next trial 9 of 17 also pass, which the criterion does not ask for. New claim.

• ✓ReproducedResults, Dynamics of L4 fast-spiking interneurons, paragraph 3; Figure 5b,d; Supplementary Figure 7

  Touch- and whisking-evoked spikes are typically multiplexed in individual L4 FS neurons, as in VPM neurons.

  Touch- and whisking-evoked spikes were typically multiplexed in individual L4 FS neurons (Fig. 5b,d and Supplementary Fig. 7), as in VPM neurons.

  Results, Dynamics of L4 fast-spiking interneurons, paragraph 3; Figure 5b,d; Supplementary Figure 7

  11 or 12 of 17 FS cells (0.65 to 0.71) have spikes per touch with a 95% interval above zero and a whisking rate that differs from the nonwhisking rate, under every filter, sample and O11 reading; VPM units give 18 to 22 of 29 (0.62 to 0.76). The earlier attempt also found it reproduced (13 of 20 tagged cells touch-responsive and 16 modulated by phase, by its own measures); the outcome is the same under the paper's measures and the claims file's FS rule.

• ✓ReproducedResults, Layer 4 FS interneurons suppress reafference in L4, paragraph 6; Supplementary Figure 14b,c

  With PV interneurons suppressed, most mice (6 of 8) are more likely to report touch with the pole out of reach (false alarms on no-lick trials).

  with PV interneuron suppressed, most mice (6 of 8) were more likely to report touches with the pole out of reach (Supplementary Fig. 14). ... In 8 mice tested, 6 showed a significant decrease of performance during Nolick trials with PV interneuron inactivation

  Results, Layer 4 FS interneurons suppress reafference in L4, paragraph 6; Supplementary Figure 14b,c

  Of 9 PV-IRES-Cre mice with the light (the paper tested 8), 6 show a significant decrease of no-lick performance under reading (b)(i), per session (a mouse counting when any session is significant) and without the continuity correction, which meets 6 of 8. The other readings give 5 ((a)(i), trials pooled, no correction), 4 ((a)(ii), pooled, Yates) and 5 ((b)(ii), per session, Yates); pooled, no mouse shows a significant increase under either test. Mice have 1 to 11 sessions, so reading (b) gives more chances to a mouse with many sessions, and one session of anm244024 shows a significant rise (p = 0.042) while another shows a fall. What the answer changed. The answer of 2026-10-20 says the paper does not state a continuity correction, and the criterion now runs the test without it (what Matlab's crosstab returns) and with Yates's correction. The claim had been blocked as methods underspecified because the two disagreed; under the updated criterion the reading without the correction matches, so it is reproduced. New claim.

  How the test differed from the paper’s method

  • The dandiset holds 9 PV-IRES-Cre mice with the eNpHR light where the paper tested 8; which mouse the paper left out is not stated, and all 9 are counted.
• ⊘BlockedResults, Dynamics of L4 fast-spiking interneurons, paragraph 1; Figure 5b,c· Methods underspecifiedverifier disagreed at first

  L4 FS interneurons fire at 9.3 +- 8.5 spikes/s without whisker movement and fire 1.9 +- 1.0 spikes per touch, with a spike latency of 5.3 +- 2.1 ms.

  FS interneurons were active during baseline conditions, even in the absence of whisker movements (mean ± s.d., 9.3 ± 8.5 spikes per s; median, 6.4 spikes per s; n = 26 neurons ... and spike rates increased after touch (mean ± s.d., 1.9 ± 1.0 spikes per touch; median, 1.9 spikes per touch; n = 26 neurons ... and spiking (mean ± s.d., 5.3 ± 2.1 ms; median, 4.5 ms, n = 24 neurons; Fig. 5c)

  Results, Dynamics of L4 fast-spiking interneurons, paragraph 1; Figure 5b,c

  The four FS readings of the criterion give different outcomes. Under (a)(i), driven at p < 0.001 and inside the boundary, 17 FS cells have a principal whisker (paper 26); the nonwhisking rate 8.53 spikes/s (8.52 to 8.72 over the O11 readings and filters) and the latency 5.23 ms (bin edge, O9 a) or 5.73 ms (centre, O9 b) over 13 cells are within tolerance, and spikes per touch 1.50 is outside 1.9 +- 0.2, so two of three values match (partially reproduced); (b)(i) gives the same cells. Without the boundary, (a)(ii) has 25 cells (8.73 spikes/s, 1.48 spikes per touch, latency 6.16 or 6.66 ms) and (b)(ii) 28 (8.71, 1.39, 10.64 or 11.14 ms), with only the nonwhisking rate within tolerance (not reproduced). The counts against Supplementary Table 2 are 4 VGAT-ChR2 whole-cell FS cells under every reading (paper 5) and 12, 20, 12 and 23 cell-attached ones under (a)(i), (a)(ii), (b)(i) and (b)(ii) (paper 24, 14 control and 10 ION-cut), so the readings without the boundary come closer to the paper's count. The tag test now counts every pulse of the 5 and 10 Hz trains; it drives the same 24 VGAT-ChR2 cells as the earlier rule of isolated pulses. Changed after verification round 1 from partially reproduced, as the criterion's rule for readings that disagree requires. New claim.

  How the test differed from the paper’s method

  • The light-driven test (one-sided binomial test of a spike 0 to 5 ms after each pulse of the 5 and 10 Hz trains, previous pulse more than 60 ms earlier, against the same 5 ms 50 ms earlier) and the waveform measures (peak-to-trough time and peak to trough ratio of the 300 Hz high-passed mean spike, larger deflection made positive) are this attempt's, as the paper gives neither; every FS group includes FS cells of ION-cut mice, which the files do not mark.
  • 26 VGAT-ChR2 cells have fewer than 10 short pulses at any cortical site, so no tag test can be run, and they count as excitatory.
• ⊘BlockedResults, Inhibition is driven by thalamus, paragraph 2; Figure 6d,e; Supplementary Figure 9j· Other

  Photoinhibiting the thalamus reduces the spike rates of L4 FS interneurons during nonwhisking (to 4% of control), whisking (to 13%) and touch (to 32%), so FS interneurons are driven by thalamic input.

  Photoinhibition reduced spike rates in FS interneurons during nonwhisking (to 4% of control ... and whisking epochs (to 13% of control ... as well as in response to touch (to 32% of control ... FS interneurons were thus driven by input from the thalamus.

  Results, Inhibition is driven by thalamus, paragraph 2; Figure 6d,e; Supplementary Figure 9j

  What the paper reports

  • nonwhisking spike rate with and without thalamic photoinhibition: p = 5 × 10⁻⁴ (n = 13 neurons)
  • whisking spike rate with and without thalamic photoinhibition: p = 6 × 10⁻⁵ (n = 15 neurons)
  • touch response with and without thalamic photoinhibition: p = 10⁻⁴ (n = 14 neurons)

  Too few cells under the claim's own rule. Only 5 FS cells by the claims file's rule were recorded with thalamic light, all from anm260070, and a two-sided signed-rank test of 5 cells cannot go below p = 0.0625 (0.062 to 0.12 over every reading). The rates fall with the light in 5 of 5 cells without whisking and 4 of 5 with it, to 0.204 and 0.273 of control and spikes per touch to 0.410 (population means, all powers, control in the same span, main epochs); over the 32 combinations of filter, window, power and O11 reading the fractions are 0 to 0.32 without whisking and 0.15 to 0.29 with it. The one-sided variant gives p = 0.031 for the nonwhisking rate and spikes per touch and 0.062 for the whisking rate, so it would not pass either. Without the waveform boundary (9 cells, a reading of fs-touch-response only) all three two-sided tests pass (p = 0.012, 0.0078 and 0.0039) and the fractions are 0.199, 0.292 and 0.462, outside tolerance, so the best outcome it could give is partially reproduced. The earlier attempt found this reproduced from 9 tagged cells and their overall rate (median ratio 0.40); the new criterion applies the waveform boundary, which leaves 5 cells.