The paper derives M β€ Q = 2ΟΞ½Οcoh for labels encoded solely by frequency, under the favourable assumption that the usable bandwidth scales with carrier frequency. This is a necessary distinguishability bound, not a general information-capacity theorem: amplitude, phase, timing, chemical identity, and spatial codes require different analyses. The paper appropriately states this scope limit and distinguishes source coherence from reader estimation error using a CramΓ©rβRao-based criterion.
The worked 30 GHz critique is the paperβs strongest section. Using a generous 1 ps proxy gives Q = 0.19; using the proposed 27 fs water-continuum estimate gives Q = 0.005. The manuscript also identifies a geometric inconsistency: a 30 Β΅m domain defined as resonant near 7.8 THz is approximately 333 wavelengths too small to function as a 30 GHz cavity. Its independent metabolic calculation reports a five-to-nine-order-of-magnitude power deficit under the examined assumptions. These arguments are conditional, however, because direct glutamate-in-water dephasing measurements at the relevant mode are absent.
The seven-criterion screen is conceptually useful because it prevents a proposal from satisfying distinguishability while ignoring readout, coupling range, energy maintenance, persistence, writability, or thermal regime. The supplied table reports that cortical gamma, sustained gamma, alpha, and hippocampal ripples pass both the paperβs C1 and perceptual-task C5 screen, while molecular, microtubule, microwave, theta, and calcium-wave entries fail C5; the manuscript correctly labels most high-frequency coherence times as estimates or upper bounds rather than direct measurements.
The central blind spot is not omission but identifiability: C5βs 0.05β0.5 s window is derived for perceptual labelling, so it cannot be transferred unchanged to working memory, motor control, synaptic plasticity, or rapid event coding. Likewise, the proposed conclusion that gamma is near the highest usable cortical frequency depends on an estimated number of coherent cycles; the authors themselves identify this inference as unresolved and acknowledge that apparent Qβfrequency regularity may be partly circular. Passing the screen also does not establish that gamma actually performs binding or addressing; the paper explicitly separates physical suitability from demonstrated biological function.
The paper is unusually transparent about uncertainty: it names missing excitation-fraction parameters, correlated-reader noise, uneven scrutiny across candidates, proxy substitution across molecular modes, and the absence of direct relevant dephasing measurements. Its open reference implementation improves arithmetic reproducibility, but reproducible calculations do not compensate for uncertain biological inputs. The most decisive tests are direct linewidth or T2 measurements for hydrated glutamate near the proposed frequency, broadband spectroscopy of hydrated glutamate above the cited measurement range, an independently specified cavity or near-field mechanism, and direct measurements of Q across frequencies rather than estimates based on cycle counts. A result such as T2 exceeding the proxy by at least 103, or a directly measured biological Q far above 100, would materially change the conclusions.
Bottom line: high value as a skeptical pre-screen for frequency-coded proposals; moderate evidential strength for the broad biological ranking; stronger, but still conditional, rejection of the specific 30 GHz labelling claim. The paperβs most defensible contribution is methodological: require measured Q, explicit channel demand, coupling range, power, persistence, writability, and thermal-regime accounting before debating exotic mechanisms.
Know what changed, what holds up, and what remains uncertain. Every Friday. No ads.