{"id":"fccf9a48-2de3-4565-a4ca-71af711b44fd","arxiv_id":"2507.23462","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Using 10 kHz phase-resolved OCT during gentle brushing, the authors observed depth-dependent skin displacements, with deeper layers showing larger amplitude near 60 Hz than the surface.","lead":"This paper demonstrates a fast optical technique for watching what happens inside the skin while it is gently brushed. The first measurements suggest deeper skin layers move more than the surface at certain vibration frequencies, which could matter for understanding touch perception.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The depth-dependent 60 Hz peak in Fig. 1D could be an OCT phase-noise artifact: FFT amplitudes are not normalized for per-depth SNR, so larger deeper peaks may reflect decreasing SNR rather than larger tissue displacement.","rationale":"The reader identified the phase-to-displacement conversion and depth-layer mapping as the weakest assumptions. This stress-test sharpens that concern with a concrete mechanism: OCT phase noise increases with depth as SNR decreases, so an unnormalized FFT can produce spurious depth-dependent amplitudes. Because the paper's main new result is exactly the depth ordering of the 60 Hz peak, this artifact would directly undermine the central claim. However, the paper is explicitly framed as a single-participant method demonstration, and the reader's CONDITIONAL verdict already captures the need for validation. Therefore no verdict change is needed, but the condition should explicitly require SNR/phase-noise normalization or a no-contact control to rule out the artifact.","tokens_in":3642,"tokens_out":4925,"duration_ms":55718,"concrete_test":"Reanalyze the existing phase time series (or acquire a stationary, non-brushing control at the same ROI and OCT settings) to compute the per-depth SNR and phase-noise floor; divide the 60 Hz FFT amplitude at 0, 100, 400 µm by that floor and check whether the depth ordering 400 > 100 > 0 µm survives. If it does not, the depth-dependent skin response claim is not supported.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that deeper layers respond more strongly near 60 Hz rests on comparing FFT amplitudes of phase data at 0, 100, and 400 µm (Section III, Fig. 1D). In phase-resolved OCT, phase noise variance scales inversely with SNR^2; with depth, SNR falls due to tissue attenuation and sensitivity roll-off, raising the FFT noise floor. The paper reports no SNR profile, no phase-noise calibration, and no repeated trials. Thus the ordering (400 µm > 100 µm > 0 µm) may simply mirror the SNR roll-off rather than a mechanical property. The paper also assumes 0/100/400 µm correspond to stratum corneum/epidermis/dermis without imaging the layer boundaries or accounting for refractive-index scaling. A bulk 60 Hz vibration source (brush motor, robot arm) common to all depths would then appear depth-dependent in the unnormalized FFT. The paper's own acknowledgement of a single participant (Section I) leaves this ambiguity unresolved.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents a proof-of-concept functional Optical Coherence Tomography (fOCT) measurement of sub-surface skin displacement during gentle brushing on the dorsal hand. A custom-driven 10 kHz SD-OCT system records 2.5 s time series, and the authors use phase-resolved processing to obtain frequency spectra of pixel-wise phase changes at depths of 0, 100, and 400 µm, which they interpret as stratum corneum, epidermis, and dermis. They report a post-brushing spectral peak near 60 Hz whose amplitude is larger at deeper locations than at the skin surface, and interpret this as evidence that different skin layers respond differently to brushing.","tokens_in":3865,"tokens_out":3133,"duration_ms":33448,"significance":"If the depth-dependent signals are shown to reflect tissue mechanics rather than imaging artifacts, the technique could provide a non-invasive window into the mechanical filtering of tactile stimuli by skin layers, complementing prior surface recordings and finite-element simulations. The paper's strengths include the high 10 kHz temporal resolution with depth-resolved phase data, a clearly described brushing robot with force and speed control, and the authors' explicit acknowledgment that this is a single-participant demonstration. However, the central quantitative comparison between depths is not yet supported because no per-depth signal-to-noise characterization or layer verification is provided.","major_comments":[{"comment":"The claim that deeper skin layers show larger FFT amplitudes near 60 Hz rests on comparing phase-change FFT amplitudes at 0, 100, and 400 µm without normalizing for per-depth signal-to-noise ratio (SNR). In phase-resolved OCT, phase noise variance scales inversely with SNR^2, and SNR falls with depth due to tissue attenuation and sensitivity roll-off. The manuscript reports no SNR profile, no phase-noise calibration, and no repeated trials, so the observed ordering (400 µm > 100 µm > 0 µm) may simply mirror sensitivity fall-off rather than a mechanical property. Please provide the per-depth SNR and either normalize the amplitude spectra by the noise floor or explicitly state that the depth comparison is hypothesis-generating rather than quantitative.","section":"Section III, Fig. 1D"},{"comment":"The assignment of depths 0, 100, and 400 µm to stratum corneum, epidermis, and dermis is stated without justification or verification. The OCT depth scale is given in air (5.5 µm resolution, 3.5 mm field of view), but skin has a refractive index near 1.4, so the physical depth corresponding to 100 and 400 optical pixels may differ substantially from 100 and 400 µm. Moreover, the thickness of epidermal and dermal layers on the dorsal hand varies across individuals and locations. The authors should either use the OCT morphological images to identify layer boundaries for this participant, account for refractive-index scaling, or clearly label the depths as optical distances and avoid tissue-layer terminology.","section":"Section II, Method"},{"comment":"The conclusion that 'each skin layer responds differently to the stimulus' is drawn from a single participant and a single recording ROI, with no repeated trials or error bars. The abstract and discussion present the depth-dependent difference as a general finding, even though the introduction acknowledges the single-participant nature of the demonstration. To support the stated conclusion, the authors need at least repeated trials on the same participant and preferably multiple participants, or they must reframe the depth comparison as an illustrative observation from one case. As written, the physiological claim goes beyond what the data can establish.","section":"Section I and Section III, Results"},{"comment":"The analysis uses only the 0.5 s windows immediately before and after the brush blocks the OCT light path, but this window selection is justified after the fact and could bias the frequency comparison. The 'before' window may contain motion from the brush approaching, and the 'after' window may contain transient settling of the skin; neither is shown to be stationary or representative. Please analyze the full 2.5 s time series or provide a principled justification for the chosen windows, and show that the reported 60 Hz feature is not an artifact of the window boundaries or of non-stationarity.","section":"Section II, Method"}],"minor_comments":[{"comment":"Typo: 'comprise d' should be 'comprised' in the first sentence of the Method section.","section":"Section II, Method"},{"comment":"Typo: 'variation s' in the first paragraph should be 'variations'.","section":"Section I, Introduction"},{"comment":"The figure would be easier to interpret if the axes were labeled with explicit units (frequency in Hz on the x-axis, phase-change amplitude in radians on the y-axis, or normalized units), and if the 60 Hz peak frequency and the amplitude values at each depth were stated in the text or figure caption.","section":"Section III, Fig. 1D"},{"comment":"Reference [13] appears to be an arXiv preprint/early release; consider citing the peer-reviewed published version if one exists, so readers can more easily access the final results.","section":"References"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is a short conference contribution whose central method demonstration is plausible. The main risk is that the depth-dependent amplitude comparison in Fig. 1D is confounded by per-depth SNR roll-off, which is a standard concern in phase-resolved OCT. The authors can address this within the scope of a revision by adding SNR normalization, error bars from repeated trials, and refractive-index/ layer-boundary justification, or by softening the layer-specific claim to a hypothesis. The paper fits the WHC scope as a methods demo, but the current wording of the abstract overstates the evidential strength."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nQuick take: this is a short conference paper that shows a real capability—10 kHz phase-resolved OCT tracking of skin displacement while a robot brushes the dorsal hand—and reports a depth-dependent 60 Hz response (deeper layers larger). The new thing is the application, not the instrument; the fOCT method is their own prior work. The illustration is clean enough to be worth a look.\n\nWhat it does well: the setup is sensible (controlled brush force/speed, CT-relevant site), the imaging refresh rate is genuinely high for OCT skin work, and the authors are honest that this is a single participant and a method illustration. They also go beyond surface morphology by using phase data, which is the right direction.\n\nSoft spots, in order of importance. First, the depth comparison rests on unnormalized FFT amplitudes of phase data. Phase noise in OCT scales with SNR, which drops with depth; without an SNR profile or phase-noise calibration, the 'larger deeper peaks' at 60 Hz could partly be a noise-floor artifact. The absence of a 60 Hz peak in the before-brushing condition suggests there is a real vibration, so this is not fatal, but it means the layer-specific claim is not established. Second, the depth-to-layer assignment (0/100/400 µm to stratum corneum/epidermis/dermis) is assumed; no layer boundaries are imaged, and no refractive-index correction is mentioned. Third, single participant, no repeated trials, no error bars—fine for a demo, but it does not support a general statement about layer-dependent mechanics. The post hoc window choice is a lesser concern; it is at least explicitly defined.\n\nThe stress-test note about SNR is the main substantive concern, and it lands partially. If the paper is revised, adding per-depth SNR/phase-noise calibration and at least a few participants/trials would make the depth comparison credible. As is, the central method demonstration is plausible and the depth-dependence claim is conditional.\n\nWho this is for: tactile/haptics researchers and anyone using OCT for skin biomechanics. It deserves a serious referee as a method demonstration; the conclusion about layer-specific dynamics should be softened or supported.","headline":"A plausible method demonstration of depth-resolved OCT during brushing, but the depth-dependence claim is under-supported by single-subject, uncalibrated FFT amplitudes.","tokens_in":4320,"tokens_out":2792,"would_cite":false,"duration_ms":31979,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"By tracking optical phase changes at 10 kHz, this paper shows that gentle brushing produces a 60 Hz vibration peak inside the skin whose amplitude grows with depth.","keywords":["functional optical coherence tomography","phase-resolved OCT","skin deformation","tactile perception","brushing stimulus","skin layers","mechanotransduction","hand dorsum"],"falsifier":"Repeat the brushing experiment on a rigid, homogeneous phantom with the same surface motion and confirm that the depth-growing 60 Hz peak disappears; if it persists, the peak is an artifact of the imaging system rather than skin tissue. Independently, verify the layer–depth mapping by imaging the same skin site with histology or high-frequency ultrasound.","tokens_in":3495,"feed_emoji":"🖐️","tokens_out":5799,"duration_ms":56622,"temperature":0.7,"pith_summary":"This paper claims that functional Optical Coherence Tomography (fOCT), running at 10 kHz with phase-resolved analysis, can track tiny mechanical displacements inside the skin while a soft brush strokes the back of the hand. In data from one participant, brushing creates a vibration peak around 60 Hz whose amplitude is larger at 100 and 400 µm below the surface than at the skin surface itself. The authors interpret this as evidence that skin layers respond differently to the same tangential stimulus, and that internal skin dynamics may therefore shape tactile perception. If the finding holds, fOCT becomes a tool for linking sub-surface mechanics to individual differences in touch.","feed_headline":"Deeper skin layers flex more under gentle brushing","feed_subtitle":"Phase-resolved OCT at 10 kHz finds a 60 Hz vibration peak that grows with depth inside the skin","key_machinery":"The load-bearing object is the phase-resolved OCT signal: each pixel's optical phase over time tracks sub-wavelength tissue displacement at that depth. The system records depth-resolved complex OCT time series at 10 kHz, and an FFT of the phase at selected depths—0, 100, and 400 µm, taken to be stratum corneum, epidermis, and dermis—produces motion spectra. A brushing robot delivers a calibrated 0.2–0.4 N, 3 cm/s stroke to the hand dorsum, and the phase-to-displacement conversion is what makes minute internal vibrations measurable.","core_discovery":"The central discovery is that frequency analysis of OCT phase changes at fixed depths reveals depth-dependent mechanical responses to gentle brushing. Before the brush arrives, motion is concentrated below 30 Hz; after it passes, a distinct peak near 60 Hz appears, and the amplitude of this peak grows with depth—the stratum corneum moves least while the epidermis and dermis show larger peaks. The authors view this as direct evidence that the layered, viscoelastic structure of skin filters and redistributes brushing forces, so each layer follows a different mechanical trajectory. The paper presents this as a new application of fOCT: turning sub-surface skin into a time-resolved motion sensor.","pith_inferences":["A direct test of the mechanism would be to measure the same brushing protocol on a silicone skin phantom: if the depth-growing 60 Hz peak persists without biological tissue, the effect is optical or mechanical rather than physiological.","The 60 Hz peak may reflect a resonant mode of the skin–tissue system; if so, its frequency should shift with skin temperature, hydration, or age—an easily testable prediction the paper does not make.","The phase signal mixes genuine tissue motion with potential artifacts such as bulk movement and speckle decorrelation; separating these would strengthen any inference about layer-specific amplification.","Because the paper uses a single participant, the claim of depth-dependent responses would be more convincing if the same protocol were repeated across individuals with varied skin properties."],"forward_implications":["Tactile coding models would need to include internal strain distribution, not just surface contact, to predict neural responses.","Haptic devices could target the depth-dependent 60 Hz response, since that frequency band appears to be amplified inside the skin.","Comparing fOCT depth profiles across individuals could link mechanical filtering to individual differences in pleasantness ratings.","Adding varied stimuli and skin interventions, such as hydration, would reveal whether the observed depth-dependent response is a general property of skin mechanics."],"supporting_citations":[{"why":"Establishes the fOCT method for tracking submicrometer vibrations in depth in glabrous skin.","marker":"[14]"},{"why":"Provides the phase-resolved optical frequency domain imaging technique used to convert phase time series into motion.","marker":"[18]"},{"why":"Prior OCT-based observation of sub-surface deformation of fingertip ridges that the paper compares its depth-dependent finding to.","marker":"[13]"},{"why":"Describes the OCT system and its resolution, the imaging hardware the custom software drives at 10 kHz.","marker":"[15]"},{"why":"Shows that stroking force and speed strongly activate pleasant-touch nerve fibers, motivating the chosen brushing parameters.","marker":"[16]"},{"why":"Supports the affective-touch framework linking pleasant-touch fibers and brushing, the perceptual context of the stimulus.","marker":"[17]"}],"fun_headline_variants":["Brushing reveals depth-dependent skin motion","Deeper skin layers respond more to gentle brushing","OCT uncovers depth-specific skin strain to brushing","Brushing creates depth-dependent 60 Hz vibration in skin"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument rests on phase changes at a given pixel being true mechanical displacement of the tissue at that depth, and on the chosen depths (0, 100, 400 µm) matching stratum corneum, epidermis, and dermis for this participant.","fun_headline_variants_meta":{"raw":{"variants":["Brushing reveals depth-dependent skin motion","Deeper skin layers respond more to gentle brushing","OCT uncovers depth-specific skin strain to brushing","Brushing creates depth-dependent 60 Hz vibration in skin"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000208,"raw_usage":{"total_tokens":1335,"prompt_tokens":810,"completion_tokens":525,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":426,"completion_tokens_details":{"reasoning_tokens":465}},"tokens_in":426,"tokens_out":525,"duration_ms":5387,"temperature":1.0,"reasoning_tokens":465,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T10:42:35.766283+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Repeat the brushing experiment on a rigid, homogeneous phantom with the same surface motion and confirm that the depth-growing 60 Hz peak disappears; if it persists, the peak is an artifact of the imaging system rather than skin tissue. Independently, verify the layer–depth mapping by imaging the same skin site with histology or high-frequency ultrasound.","supporting_citations":[{"cited_title":"Tracking submicrometer vibrations in depth in the glabrous skin,","cited_arxiv_id":null,"evidence_quote":"Establishes the fOCT method for tracking submicrometer vibrations in depth in glabrous skin."},{"cited_title":"Phase-resolved optical frequency domain imaging,","cited_arxiv_id":null,"evidence_quote":"Provides the phase-resolved optical frequency domain imaging technique used to convert phase time series into motion."},{"cited_title":"Sub-surface deformation of individual fingerprint ridges during tactile interactions,","cited_arxiv_id":null,"evidence_quote":"Prior OCT-based observation of sub-surface deformation of fingertip ridges that the paper compares its depth-dependent finding to."},{"cited_title":"Revealing the morphology and function of the cochlea and middle ear with optical coherence tomography,","cited_arxiv_id":null,"evidence_quote":"Describes the OCT system and its resolution, the imaging hardware the custom software drives at 10 kHz."},{"cited_title":"Coding of pleasant touch by unmyelinated afferents in humans,","cited_arxiv_id":null,"evidence_quote":"Shows that stroking force and speed strongly activate pleasant-touch nerve fibers, motivating the chosen brushing parameters."},{"cited_title":"Discriminative and affective touch: sensing and feeling,","cited_arxiv_id":null,"evidence_quote":"Supports the affective-touch framework linking pleasant-touch fibers and brushing, the perceptual context of the stimulus."}],"review_version":1}