Claims 19
From the paper Voltage dynamics of dendritic integration and back-propagation in vivo · claims file · checked 2026-10-21: ready
- fig1d-light-intensity-spike-rate
In vivo, 0.5 s pulses of blue light of increasing intensity evoke somatic spiking at increasing rates, with corresponding bAPs in the dendrites.
Optogenetic stimulation with pulses of increasing blue light intensity evoked action potentials with increasing spike rates and corresponding bAPs in the dendrites.
Figure 1d legend; Results, Dendritic voltage mapping, paragraph 1How it will be judged
Shown for one neuron (Fig. 1c-d). Somatic spikes are counted in each blue light pulse of the intensity series; the paper gives no spike detection method, so spikes are detected under two readings, (a) peaks above 5 and (b) above 4 robust standard deviations (median absolute deviation) of the high-pass filtered soma trace, a standard choice that the paper does not fix. Reproduced when, under both readings, the spike rate rises with pulse order (Spearman rho > 0 and higher in the last pulse than in the first), and the dendritic traces show events within 5 ms after most somatic spikes.
- fig1fg-dendrite-peaks-after-soma
Spike-triggered averages of dendritic fluorescence, triggered off somatic spike times, peak after the somatic spike, as expected for back-propagation.
The dendritic events peaked after the somatic spike (dashed line), as expected for back-propagation.
Figure 1f-g legend; Results, Dendritic voltage mapping, paragraph 1How it will be judged
For the Fig. 1 neuron, the spike-triggered average of each dendritic trace, triggered off the somatic spike peaks. Reproduced when the dendritic STA peak follows the somatic peak (delay > 0) at all four dendrites; partially reproduced for two or three.
- fig2c-no-trial-variation
For optogenetically evoked single spikes in anesthetized mice, bAP amplitude did not vary from trial to trial beyond baseline noise at any recording site, so bAP failures were infrequent.
For these optogenetically evoked single spikes, we did not observe trial-to-trial variation in the bAP amplitude beyond baseline noise fluctuations at any of the recording sites (see e.g., Fig 2c ), indicating that bAP failures were infrequent under these conditions.
Results, Mapping bAP propagation, paragraph 1; Figure 2cHow it will be judged
The paper reports no test. For each Fig. 2 recording site (40 pulses, 8 ms, 5 Hz), the single-spike dendritic bAP amplitudes are compared with the baseline noise. The paper does not define either quantity, so two readings are run, (a) the s.d. of single-trial bAP amplitudes against the s.d. of the same measure taken in pre-stimulus windows of the same length, and (b) the fraction of single bAPs smaller than 2 s.d. of baseline noise (a failure count). Reproduced when, under both readings, at least 90% of sites show amplitude variation no larger than 1.5 times the baseline variation and failures in under 10% of spikes; these bounds are not in the paper and the result is reported with the values.
- fig2e-half-amplitude-distance
In anesthetized mice, the bAP peak amplitude decays to 50% at a contour distance from the soma of 242 ± 30 μm (95% C.I., 30 sites, 5 neurons, 5 mice), with R2 = 0.49 for amplitude against distance.
The bAP peak amplitude decayed to 50% at a contour distance from the soma of 242 ± 30 μm (95% C.I., n = 30 recording sites, 5 neurons, 5 mice, Fig. 2e ).
Results, Mapping bAP propagation, paragraph 2; Figure 2e legendHow it will be judged
The distance tolerance is the paper's 95% C.I. (± 30 μm, 30 sites); the R2 tolerance is 10%. Amplitude is the peak of each site's mean bAP waveform (-dF/F), triggered off the soma spike peaks, using the background-robust dF/F of the Methods (Robust estimate of dF/F). The paper does not say what curve is fit or what the 50% is relative to. Readings, (a) a linear regression of amplitude on contour distance, distance where the fit falls to half its value at distance 0; (b) an exponential fit, distance ln(2) times the length constant; (c) amplitude normalized to the somatic spike amplitude of the same recording, distance where the linear fit reaches 0.5. Reproduced when, under one reading, the distance is within 212 to 272 μm and R2 within tolerance; the readings are all reported.
- fig2f-delay-range
Back-propagation delays range from 1.5 to 4 ms over contour distances from the soma of 50 to 280 μm.
Back-propagation delays ranged from 1.5 to 4 ms, over contour distances from the soma of 50-280 μm ( Fig. 2f ).
Results, Mapping bAP propagation, paragraph 2; Figure 2fHow it will be judged
Delay is the time of the peak of each site's mean bAP waveform minus the time of the somatic spike peak (Fig. 2f legend, "Peak delay"). At 500 or 1,000 Hz the paper does not say how the peak is timed below one frame; readings, (a) the sample of the maximum and (b) a parabolic fit through the maximum and its two neighbours, in both soma and dendrite. Reproduced when, under one reading, the delays of sites between 50 and 280 μm lie within 1.35 to 4.4 ms.
- fig2f-conduction-speed
At distances over 50 μm from the soma, the bAP conduction speed is 0.09 ± 0.02 m/s (95% C.I., 30 sites, 5 neurons, 5 mice), with R2 = 0.43 for delay against distance.
Conduction speeds were 0.09 ± 0.02 m/s (95% C.I., n = 30 recording sites, 5 neurons, 5 mice), consistent with prior theoretical predictions
Results, Mapping bAP propagation, paragraph 2; Figure 2f legendHow it will be judged
The speed tolerance is the paper's 95% C.I. (± 0.02 m/s, 30 sites); the R2 tolerance is 10%. From the Fig. 2f legend, a linear regression of peak delay on contour distance for sites more than 50 μm from the soma; speed is the inverse of the slope. The peak timing readings are those of fig2f-delay-range. The paper does not say whether the fit pools all sites or is made per neuron; readings, (a) one pooled fit over the sites, (b) the mean of per-neuron fits. Reproduced when, under one reading, the speed is within 0.07 to 0.11 m/s and the pooled R2 within tolerance.
- fig3d-dendrite-cross-correlation
Spontaneous voltage in pairs of dendritic branches of the same neuron is highly cross-correlated, 91 ± 3% (mean ± s.d.) in the anesthetized state and 73 ± 4% in the awake state (23 branch pairs, 3 neurons, 3 mice).
Remarkably, dendrites displayed a 91 ± 3% cross-correlation (mean ± s.d.) in the anesthetized state and 73 ± 4% in the awake state ( Fig. 3d ).
Results, Dendrite-Dendrite coupling, paragraph 2; Figure 3d legendHow it will be judged
Tolerance is the reported s.d. of the pairs. Per pair of simultaneously recorded branches, the cross-correlation of spontaneous traces; the mean over the 23 pairs per state. Fig. S9 legend takes the zero-lag value as the measure for the shot-noise bound, but the paper does not say whether the branch-pair value is taken at zero lag or at the peak over lags, nor whether spikes are kept in the traces. Readings, (a) zero-lag Pearson correlation of the full traces; (b) peak of the normalized cross-correlation within ±10 ms; (c) as (a) with spikes removed by excluding 10 ms around each somatic spike where a soma recording exists. Reproduced when, under one reading, both means are within tolerance and the anesthetized mean exceeds the awake mean.
- figs9-shot-noise-bound
Accounting for photon shot noise, the upper bounds on dendrite-to-dendrite cross-correlation are 94 ± 2% (anesthetized) and 85 ± 4% (awake), so the electrical dynamics are largely shared throughout the L2/3 apical dendritic tree.
Accounting for the influence of shot noise ( Fig. S9 , Methods), the upper bounds on the dendrite-to-dendrite cross-correlations were estimated to be 94 ± 2% (anesthetized) and 85 ± 4% (awake).
Results, Dendrite-Dendrite coupling, paragraph 2; Figure S9; Methods, Contribution of shot noiseHow it will be judged
Tolerance is the reported spread. The Methods split each branch's region of interest into two interleaved checkerboard pixel sets and take the zero-lag cross-correlation of the two signals. This needs pixel-level movies; if the dataset holds only per-branch traces the claim cannot be tested and is reported so. Reproduced when both bounds are within tolerance and each exceeds the matching fig3d mean.
- fig3i-coupling-vs-distance
Normalized dendrite-dendrite coupling amplitude decreases by 50% beyond about 200 μm of interbranch distance and stays near 50% out to 300 μm (92 pairs, 5 neurons, 5 mice), so most apical dendrites of L2/3 neurons are electrically coupled in vivo.
Figure 3i shows dendrite-dendrite coupling amplitude vs. interbranch distance, displaying a 50% decrease after ∼200 μm, and similar ∼50% coupling out to 300 μm.
Results, Dendrite-Dendrite coupling, paragraph 4; Figure 3i legendHow it will be judged
From the Results and Methods (Dendrite-dendrite coupling matrix), the stimulus-triggered average of each branch for each stimulated branch (50 repeats, 10 ms pulses, 5 Hz), its amplitude as background-robust dF/F, each row divided by the amplitude at the stimulated branch; interbranch contour distance from the reconstruction. The figure's distance bins and the bins its Wilcoxon rank-sum p-values compare are not in the text (no PDF was available). Readings, (a) bins of 0 to 100, 100 to 200, 200 to 300 μm and beyond; (b) bins of 50 μm. Reproduced when, under both readings, the mean coupling of the bins beyond 200 μm is within 0.45 to 0.55, is lower than that of the bin nearest 0 by a Wilcoxon rank-sum test at p < 0.05 (the conventional threshold; the paper prints none in the text), and pairs beyond 300 μm, if any, keep a mean coupling above 0.4.
- figs11-dendritic-spikes-rare
Strong dendrite-targeted stimuli usually evoke spikes that correspond to bAPs; in 2 of 40 trials they evoke dendritic and not somatic spikes.
Simultaneous recordings at soma and dendrites in anesthetized mice showed that these spikes usually corresponded to bAPs, although occasionally (2 of 40 trials) dendrite-targeted stimuli evoked dendritic and not somatic spikes ( Fig. S11 ).
Results, Dendrite-Dendrite coupling, paragraph 5; Figure S11How it will be judged
Tolerance 0, since the 10% default (0.2 trials) admits no count but 2. A trial is one strong dendrite-targeted stimulus with soma and dendrite recorded (anesthetized). The paper does not say what the 40 trials are; readings, (i) all strong dendrite-targeted stimuli, including those that evoked only subthreshold events (Fig. S11e), and (ii) only the trials in which a spike was detected at the soma or the dendrite. The paper gives no detection rule for spikes or dendritic spikes; readings for an event, (a) a peak above 5 and (b) above 4 robust s.d. of the trace, the standard choices used in fig1d-light-intensity-spike-rate. A dendritic spike with no somatic spike is one with no somatic spike in the 10 ms before it (a window not in the paper). Reproduced when, under one reading, 40 trials are found, exactly 2 of them have a dendritic spike with no somatic spike, and in most trials with a dendritic spike a somatic spike precedes it ("usually"); the counts of every reading are reported.
- fig4a-doublet-isi
In awake mice some cells fire spontaneous spike doublets with a narrow distribution of inter-spike intervals, 18 ± 7 ms (mean ± s.d., 216 events, 3 neurons, 3 mice).
In some cells, we observed spike doublets with a narrow distribution of inter-spike intervals (ISIs, 18 ± 7 ms, mean ± s.d., n = 216 events, 3 neurons, 3 mice, Fig. 4a ).
Results, Accelerating spike rate favors spontaneous bAP propagation, paragraph 1; Figure 4aHow it will be judged
The paper does not define a doublet. Readings, (a) a pair of somatic spikes with ISI below 75 ms (the threshold the paper uses for triplets), preceded and followed by ISIs above 75 ms; (b) any consecutive pair with ISI below 75 ms. Spike detection as in fig1d-light-intensity-spike-rate. Reproduced when, under one reading, mean, s.d. and count are within tolerance.
- fig4bc-doublet-first-fails
In spontaneous spike doublets, the first spike typically fails to propagate to distal dendrites while the second succeeds (doublet ISIs 18 ± 7 ms, 10 branches, 3 neurons, 3 mice).
Examination of individual spike doublets ( Fig. 4b , S12 ) as well as the average doublet waveforms ( Fig. 4c ) revealed that the first spike typically failed to propagate to distal dendrites, while the second spike succeeded.
Results, Accelerating spike rate favors spontaneous bAP propagation, paragraph 1; Figure 4b-c legendHow it will be judged
The paper reports no test. Doublets as in fig4a-doublet-isi, the average dendritic waveform triggered off the first somatic spike peak, per branch. bAP amplitude is the peak of the dendritic trace within 0 to 5 ms after each somatic spike peak (covering the 1.5 to 4 ms delays of Fig. 2f) minus the mean of the 5 ms before the first spike; this window is not in the paper, so a second reading of 0 to 8 ms is run. Distal is read as (a) all 10 branches and (b) branches with conduction delay > 3 ms (the Methods' split for optogenetic data). Reproduced when, under every reading, the second bAP exceeds the first in at least 70% of branches, with the median ratio of second to first over branches above 1.
- fig4d-accelerating-amplifies
In spontaneous spike trios, the third bAP is amplified relative to the second when the spike rate accelerates (ISI n-1 > 75 ms, ISI n < 75 ms) and suppressed when it decelerates (186 spike triplets, 10 branches, 3 neurons, 3 mice), so spontaneous bAP propagation is favored by an accelerating spike rate.
The third bAP in accelerating trios experienced amplification relative to the second bAP, while the third bAP in decelerating trios experienced suppression relative to the second (10 branches, 3 neurons, 3 mice).
Results, Accelerating spike rate favors spontaneous bAP propagation, paragraph 2; Figure 4d legend; Methods, Analysis of bAP failures during spontaneous activityHow it will be judged
From the Methods, somatic spikes and their dendritic events; for each dendritic compartment bAPs normalized by their mean amplitude; trios classified by the somatic ISIs at the 75 ms threshold of the Fig. 4d legend. The figure tests with a Wilcoxon signed-rank test, but its p-value is printed only in the figure and the paired quantity is not stated. Readings, (a) per trio and branch, third against second normalized amplitude, paired, within each class; (b) per branch, mean third-to-second ratio, accelerating against decelerating, paired over the 10 branches; (c) as the Methods word it ("The amplitude of the third spike in each triplet was determined and was classified"), the normalized third-spike amplitude in each class against the per-compartment mean of 1, by signed-rank test. The bAP amplitude window is that of fig4bc-doublet-first-fails, both readings. Reproduced when, under every reading, accelerating trios show third > second and decelerating trios third < second (in (b), accelerating ratio above decelerating; in (c), accelerating third above 1 and decelerating third below 1), at p < 0.05 by the Wilcoxon signed-rank test (the conventional threshold). The number of trios found is reported against the paper's 186.
- fig4fg-opto-first-bap-attenuated
With sustained (400 ms) blue light on the soma, the first bAP in each evoked spike train is highly attenuated in the dendrites relative to the soma, and subsequent bAPs propagate.
As in the spontaneous activity, the first bAP in each spike train was highly attenuated in the dendrites relative to the soma ( Fig. 4f-g ).
Results, Accelerating spike rate favors optogenetically triggered bAP propagation, paragraph 2; Figure 4f-g legendHow it will be judged
The paper reports no test. Fig. 4g shows one neuron (n = 10 trials); the claim is tested over all sustained-pulse recordings (16 trials, 7 neurons, Fig. 4h legend), and for the Fig. 4g neuron alone where it can be identified. The statement compares the dendrite with the soma, which the Methods' normalization does not; readings, (a) bAP amplitudes normalized per dendrite by the mean bAP amplitude 200 to 400 ms after stimulus onset (Methods, Analysis of bAP failures during optogenetic stimulation); (b) per spike, the dendritic bAP amplitude divided by the somatic spike amplitude of the same spike, normalized by the same ratio's 200 to 400 ms mean. Reproduced when, under both readings, in electrotonically distal dendrites (delay > 3 ms, Methods), the median normalized value of the first bAP of each train is below that of the second and later bAPs in the first 150 ms (a window not in the paper, the end of the Results' 50 to 150 ms window), and below 1, with the windows of fig4bc-doublet-first-fails.
- fig4h-biphasic-bap-amplitude
In distal dendrites, bAP amplitude during sustained optogenetic stimulation is biphasic in time after stimulus onset, suppressed for the first bAPs, enhanced in a window from 50 to 150 ms, then decreasing (16 trials, 7 neurons, 7 mice).
There was a window from 50 –150 ms after stimulus onset where bAP amplitude increased; then bAP amplitude decreased again ( Fig. 4h ).
Results, Accelerating spike rate favors optogenetically triggered bAP propagation, paragraph 2; Figure 4h legendHow it will be judged
From the Methods, spikes sorted into time bins after stimulus onset; in each bin the STA at soma and dendrite triggered off the soma spike times and its mean bAP amplitude, normalized by the 200 to 400 ms mean; distal dendrites have conduction delay > 3 ms. The bin widths and the bins the legend's Wilcoxon signed-rank tests compare are not in the text, nor are the p-values. Readings, (a) bins of 0 to 50, 50 to 150, 150 to 200 and 200 to 400 ms (the window the Results name); (b) bins of 25 ms; (c) bins of 50 ms. In each reading the first bin, and the peak bin within 50 to 150 ms, are compared over the 16 trials with the 200 to 400 ms value of 1 by a Wilcoxon signed-rank test. Reproduced when, under every reading, the first bin is below 1 and a 50 to 150 ms bin above 1, each at p < 0.05 (the conventional threshold), and the amplitude after 150 ms is below that peak.
- figs13-distal-more-modulated
The biphasic modulation of bAP amplitude is larger in dendrites further from the soma, by mean propagation delay, than in more proximal dendrites (25 branches, 7 cells, 7 mice).
This biphasic modulation of bAP amplitude was larger in dendrites further from the soma (as measured by mean propagation delay) than in more proximal dendrites ( Fig. S13 , 25 branches, 7 cells, 7 mice).
Results, Accelerating spike rate favors optogenetically triggered bAP propagation, paragraph 2; Figure S13How it will be judged
The paper reports no test. Branches split at a conduction delay of 3 ms (Methods and Fig. S13 legend), the time-binned normalized amplitudes of fig4h under each of its bin readings. Modulation is not defined; readings, (a) the peak bin within 50 to 150 ms minus the first bin; (b) the range of bin values. Reproduced when, under every reading, the mean modulation of distal branches exceeds that of proximal branches.
- fig4ijk-sensory-first-fails
Contralateral whisker air puffs in awake mice evoke bursts of spikes in which all detected dendritic spikes start as bAPs from the soma, and the first spike in the burst typically fails to propagate to distal dendrites while subsequent spikes succeed (13 trials).
Examination of the single-trial responses ( Fig. 4j ) as well as the average evoked waveforms ( Fig. 4k ) revealed that the first spike in the burst typically failed to propagate to distal dendrites, while subsequent spikes succeeded (n = 13 trials).
Results, Accelerating spike rate favors sensory evoked bAP propagation; Figure 4i-k legendHow it will be judged
The paper reports no test. Air-puff trials with a somatic burst; the average dendritic waveform triggered off the first somatic spike peak (Fig. 4k legend). bAP amplitude windows as in fig4bc-doublet-first-fails; distal read as delay > 3 ms and as all recorded dendrites. Reproduced when, under every reading, the first bAP is smaller than the mean of the subsequent bAPs of the burst in the average and in at least 9 of the 13 trials (a bound not in the paper, for "typically"), and every dendritic spike detected in the air-puff trials (spike detection as in fig1d, both readings) follows a somatic spike within the bAP windows of fig4bc-doublet-first-fails ("all detected dendritic spikes initiating as bAPs from the soma").
- fig4l-sensory-plateau
The average sensory-evoked dendritic response is a fast upstroke followed by a sustained plateau depolarization lasting about 400 ms, much longer than the 100 ms stimulus.
The average dendritic waveform showed a fast sensory-evoked upstroke, followed by a sustained plateau depolarization which lasted ∼400 ms, substantially longer than the stimulus.
Results, Accelerating spike rate favors sensory evoked bAP propagation; Figure 4l legendHow it will be judged
The stimulus-triggered average dendritic trace, triggered off air-puff onset (Fig. 4l). The paper does not define the plateau's duration; readings, (a) time from puff onset until the average falls back within 1 s.d. of the 500 ms pre-stimulus baseline; (b) full width at half of the plateau amplitude, the plateau taken as the mean from 50 to 150 ms after onset. Reproduced when, under one reading, the duration is within 360 to 440 ms and the depolarization outlasts 100 ms in both.
- figs12-spontaneous-failures-both-states
During spontaneous activity, bAP failures occur in both anesthetized and awake recordings.
We observe bAP failures in both anesthetized and awake recordings.
Figure S12 legendHow it will be judged
The paper reports no test and shows example traces only. Spontaneous recordings with soma and dendrite, split by state (anesthetized, awake). A failure is a somatic spike whose dendritic bAP amplitude, in the windows of fig4bc-doublet-first-fails, is below 2 s.d. of the dendrite's baseline noise (the failure reading (b) of fig2c-no-trial-variation; the paper gives no rule). Spike detection as in fig1d-light-intensity-spike-rate, both readings. Reproduced when, under every reading, failures are found in at least one recording of each state; their fraction per state is reported.
Pipeline
- Paper
- done (1 of 1)
- Claims
- done
- Claims check
- done (ready)
- Attempt
- todo
- Questions
- none
- Verified
- none