Analysis 6A — onset timing during stimulus omission

Macaque multi-area laminar electrophysiology · Exploratory analysis not publication-final

LFP frequency and temporal resolution

1,820 area×frequency cells; 114 are temporally resolved. Whether independent estimators agree on when the change happened is itself the informative quantity here.

Unresolved does not mean unmodulated. Low-frequency LFP power can change strongly while its onset remains poorly localized in time. Response magnitude and temporal resolution are separate measurements, and this page reports only the second.

B1 · P(resolved | frequency)

B2 · LOW vs HIGH, paired within session

One line per session. The pairing is what makes this inferential: each session contributes a single difference Ds, so between-session and between-subject variation cannot inflate n.

median D = +7.27 percentage points · exact exhaustive sign-flip permutation p = 8.39e-05 (262,144 assignments enumerated over 18 non-zero sessions of 21) · Holm p = 1.7e-04 replicated across sessions

B4 · Censoring, and why there is no beta onset distribution

Beta latencies are strongly left-censored. Of 36 resolved beta cells, 17 sit exactly on the lower search bound and 25 (69%) fall at or below 20 ms. A naive beta onset distribution would report a bound, not a measurement, so none is shown here and boundary-pinned values are never rendered as ordinary latencies.

bandn_resolvedn_at_lower_boundn_le_20msfrac_le_20msmedian_T_om_unpinned_ms
theta3000309
alpha5110.2149.5
beta3617250.69193
low_gamma36180.2259.5
high_gamma24020.08139.5
out_of_band10010.1152

Theta and alpha resolve too few cells (3 and 5 respectively) to support corpus latency claims. Low and high gamma are the main descriptive LFP timing regimes.