Auditory Frequency Representations in Human Somatosensory Cortex

Auditory Frequency Representations in Human Somatosensory Cortex Pérez-Bellido Anne Barnes Crommett Yau

Tag layer of the DHSS Meta-Tagging Project. The article text is reproduced from the staged copy named in sourceDesc and is not altered; the tags are held separately in standOff so that overlapping and nested spans are preserved.

Auditory Frequency Representations in Human Somatosensory Cortex 2018 Cerebral Cortex (New York, NY) 10.1093/cercor/bhx255 P-rez-Bellido2018-auditory-frequency-representations-in-human P-rez-Bellido2018-auditory-frequency-representations-in-human.txt, the project's staged copy fd11a67b5c387e39d37c0e3cd6657eeab8840121 not-redistributable neuroscience

the paper's own words

the tag layer; a sentence written by the tagger about the paper, always anchored to a quotation but not itself a quotation

0 spans located by searching the stored text for the quote's exact words, allowing any whitespace between them auditory frequency representations in classically defined somatosensory cortex - whether responses to a non-preferred modality carry specific information Quotation withheld because this record is not cleared for redistribution. whether parietal responses to sound are frequency-selective and related to perception, or merely present Quotation withheld because this record is not cleared for redistribution. frequency-specific auditory responses in parietal cortex whose representational geometry correlates with offline perceptual similarity judgments, surviving conservative erosion of the parieto-temporal border Quotation withheld because this record is not cleared for redistribution. existence Quotation withheld because this record is not cleared for redistribution. auditory frequency representation in somatosensory cortex Quotation withheld because this record is not cleared for redistribution. fmri fmri Quotation withheld because this record is not cleared for redistribution. mechanisms Sounds and vibrations that share frequency content co-occur constantly in everyday experience, and this long history of co-activation builds audio-tactile associations inside frequency-processing circuits Quotation withheld because this record is not cleared for redistribution. mechanisms Predictive coding: hearing a tone evokes the neural representation of the tactile event usually associated with that tone, producing parietal activity with no mechanical stimulation at all Quotation withheld because this record is not cleared for redistribution. mechanisms Auditory signals reach parietal cortex over already-documented cortico-cortical and thalamo-cortical routes linking the auditory and somatosensory systems Quotation withheld because this record is not cleared for redistribution. subgroups of voxels in classically defined somatosensory regions of the parietal lobe respond to audition and are better suited to represent auditory frequency information Quotation withheld because this record is not cleared for redistribution. with the proposed route being an extensive history of co-activation at frequencies common to both senses, so that hearing a tone evokes the tactile representation associated with it Quotation withheld because this record is not cleared for redistribution. and the association may be feature-specific - frequency shared between the senses where stimulus duration appears to rely on modality-dedicated circuits Quotation withheld because this record is not cleared for redistribution. fMRI's temporal resolution cannot say whether the parietal auditory responses are feedforward or feedback. what this limits: the direction of the parietal auditory signal: DUE TO THE LIMITED TEMPORAL RESOLUTION OF fMRI, Quotation withheld Quotation withheld because this record is not cleared for redistribution. the task performed during the discrimination scans, designed to ensure that participants attended to stimulus frequency, did not permit a sensitive analysis of auditory discrimination performance what this limits: the analysis the discrimination scans could not support: THE TASK ... DESIGNED TO ENSURE THAT PARTICIPANTS ATTENDED TO STIMULUS FREQUENCY, DID NOT PERMIT A SENSITIVE ANALYSIS Quotation withheld because this record is not cleared for redistribution. these observations need to be confirmed in ultra-high resolution scanning and single-unit recording experiments testing a more comprehensive range of tones what this limits: the whole set of observations: THESE OBSERVATIONS NEED TO BE CONFIRMED IN ULTRA-HIGH RESOLUTION SCANNING AND SINGLE-UNIT RECORDING EXPERIMENTS TESTING A MORE COMPREHENSIVE RANGE OF TONES Quotation withheld because this record is not cleared for redistribution. the perceptual measure is a group average collected offline from mostly different people, because the in-scanner task was not designed to yield a sensitive discrimination measure what this limits: the brain-to-perception comparison: WE COMPARED THE CORTICAL DISSIMILARITY MEASURES TO THE PERCEPTUAL DISSIMILARITY JUDGMENTS TESTED OFFLINE AND AVERAGED across mostly different people Quotation withheld because this record is not cleared for redistribution. auditory frequency representations in bilateral SR eroded are sensitive to ROI size and voxel inclusion criteria, unlike those in AR eroded which are more robust to the same criteria what this limits: the SR result specifically: AUDITORY FREQUENCY REPRESENTATIONS IN BILATERAL SR ERODED ARE SENSITIVE TO ROI SIZE AND VOXEL INCLUSION CRITERIA, UNLIKE THOSE IN AR ERODED WHICH ARE MORE ROBUST Quotation withheld because this record is not cleared for redistribution. the elimination of the cutaneous-activation account rests partly on a comparison whose data are not shown what this limits: the cutaneous-activation argument: it rests on a comparison of frequency response profiles whose data are not shown Quotation withheld because this record is not cleared for redistribution. fMRI cannot say whether the parietal auditory responses are feedforward or feedback what this limits: the same scope as limitations[0] - feedforward against feedback is not resolvable at this temporal resolution Quotation withheld because this record is not cleared for redistribution. and whether frequency is explicitly tuned, temporally coded or topographically arranged in somatosensory cortex is unknown what this limits: any claim that somatosensory cortex represents frequency explicitly: EVIDENCE FOR EXPLICIT FREQUENCY TUNING is lacking, even where spike timing can signal vibration above 100 Hz Quotation withheld because this record is not cleared for redistribution. the cortical DM calculated for each ROI compared to the group perceptual DM, so that a high correlation indicates the representational geometry of the auditory stimuli in the ROI corresponds to the geometry of the stimuli in perceptual space Quotation withheld because this record is not cleared for redistribution. pairwise similarity ratings, collected outside the scanner and turned into a dissimilarity matrix Quotation withheld because this record is not cleared for redistribution. the 100 to 300 Hz range as the band where the two senses overlap, used to predict where parietal responses should stop Quotation withheld because this record is not cleared for redistribution. temporal coding of vibration frequency in somatosensory cortex, named as the mechanism that would explain the fMRI result and has not been shown Quotation withheld because this record is not cleared for redistribution. a +/-5 per cent JITTER added to the loudness-matched amplitude of every stimulus presentation eliminates: amplitude as a systematic correlate of frequency - without it, a region that appeared to encode auditory FREQUENCY could be responding to loudness instead, which is the confound the paper's whole claim turns on · outcome: held Quotation withheld because this record is not cleared for redistribution. ROI size → statistical relationship AN ANALYSIS CHOICE THAT MOVES THE RESULT: significant correlations required between 50 and 200 voxels with the tactile mask, and beyond that, including more voxels WEAKENED the relationship. The correlation is not a fixed property; it depends on where the ROI is cut. Quotation withheld because this record is not cleared for redistribution. cortical and perceptual DMs → postcentral gyrus POSITIVE CORRELATION, RESTRICTED: significant only in the postcentral and supramarginal gyri, and only under the auditory mask. The restriction IS the result - an unscoped edge would assert it across parietal cortex, which the span explicitly denies. Quotation withheld because this record is not cleared for redistribution. THE REGION'S NAME DOES NOT BOUND WHAT IT PROCESSES: robust and frequency-specific responses to sound in CLASSICALLY DEFINED somatosensory areas, against the traditional notion of modality-dedicated cortical systems Quotation withheld because this record is not cleared for redistribution. and the question is sharpened before the measurement - multimodal responses are NO LONGER DISPUTED, but whether they convey SPECIFIC representations and how they relate to perception are the open questions Quotation withheld because this record is not cleared for redistribution. with the precise gap named: auditory stimulation was already known to evoke distinct signatures in somatosensory cortex, and it REMAINED UNCLEAR whether those patterns represent specific auditory information Quotation withheld because this record is not cleared for redistribution. WHICH LOCALIZER DEFINED THE VOXELS IS ITSELF THE HYPOTHESIS TEST: separate auditory, tactile and audio-tactile masks, used to ask whether parietal voxels PRIVILEGED in their sensitivity to sound are the ones that carry the representation Quotation withheld because this record is not cleared for redistribution. AND THE ANSWER DEPENDS ON THE MASK: fewer and smaller regions show a frequency effect with the tactile mask, and the paper reads that difference as the result rather than as noise Quotation withheld because this record is not cleared for redistribution. THE SAME DATA ARE SIGNIFICANT OR NULL BY MASK, ONCE SPLIT BY HEMISPHERE: in the left somatosensory ROI the auditory mask gives r = 0.24 P < 0.05, the tactile mask r = 0.07 P = 0.25 and the audio-tactile mask r = 0.1 P = 0.15 Quotation withheld because this record is not cleared for redistribution. and the same split against perception - significant correlations ONLY with the auditory masks, with the tactile left-hemisphere mask at P = 0.07 Quotation withheld because this record is not cleared for redistribution. THE RESULT EXISTS INSIDE A WINDOW OF ANALYSIS SIZE, AND THE WINDOW DIFFERS BY MASK: the auditory mask needs more than 130 voxels, the tactile mask between 50 and 200 - BEYOND WHICH MORE VOXELS WEAKEN IT - and the audio-tactile mask only 300 to 350 Quotation withheld because this record is not cleared for redistribution. stated plainly as a limit - the parietal representations are SENSITIVE to ROI size and voxel inclusion criteria where the temporal ones are ROBUST to the same criteria Quotation withheld because this record is not cleared for redistribution. A THRESHOLD CHOSEN FOR ITS ROLE IN THE PIPELINE, NOT FOR INFERENCE: a LIBERAL uncorrected P < 0.05 used to build the masks, precisely because they only constrain later tests Quotation withheld because this record is not cleared for redistribution. THE artifact IS ELIMINATED BY A PARAMETRIC SWEEP RATHER THAN ONE CONTROL: ROIs eroded by 2 mm, then 4 mm, then 10 mm - and the auditory-mask result survives all three Quotation withheld because this record is not cleared for redistribution. with four separate design decisions listed against the same confound - 2 mm isotropic voxels, no volumetric smoothing, surface-space filtering, and conservative borders Quotation withheld because this record is not cleared for redistribution. AND THE SPATIAL ARGUMENT IS TURNED AROUND: the significant parietal subdivisions are NOT the ones closest to auditory cortex, which is what a partial-volume account would have required Quotation withheld because this record is not cleared for redistribution. ONE CONTROL EXPERIMENT ANSWERING TWO DIFFERENT OBJECTIONS: air-conduction headphones show the effect does not need vibrating earbuds, AND that session involved no touch at all, so a learned sound-vibration association cannot explain it either Quotation withheld because this record is not cleared for redistribution. and the earbud account is refuted by SOMATOTOPY - cutaneous drive in the ear canal would confine the response to the ear representation, and the responses are broadly distributed Quotation withheld because this record is not cleared for redistribution. A LOAD-BEARING COMPARISON WHOSE EVIDENCE IS NOT IN THE PAPER: tactile stimulation at matched frequencies gave response profiles that 'clearly differed' from the sound profiles - DATA NOT SHOWN Quotation withheld because this record is not cleared for redistribution. TWO NEGATIVE CONTROLS, ONE ACROSS REGIONS AND ONE WITHIN: visual cortex patterns do not correlate with the perceptual geometry, and neither do the voxels that DEACTIVATED in the localizer Quotation withheld because this record is not cleared for redistribution. THE TWO SIDES OF THE CORRELATION COME FROM DIFFERENT PEOPLE AND DIFFERENT SESSIONS: the perceptual judgments were collected offline and averaged over subjects, because the in-scanner task could not yield a sensitive discrimination measure Quotation withheld because this record is not cleared for redistribution. and the peak sits at the edge of the tested range - the somatosensory profile is bimodal with a peak at 75 Hz, THE LOWEST TESTED FREQUENCY Quotation withheld because this record is not cleared for redistribution. claim_type: existence a second reading reaching the same value. Recorded: existence — note: a second reading reaching the same value from a different sentence. Not applied - a person decides. Quotation withheld because this record is not cleared for redistribution. method: three independent localizer masks - auditory, tactile, audio-tactile - used to select voxels, so that which mask defines the data becomes itself the hypothesis test a second reading DIFFERING from the recorded value. Recorded: fmri — note: a second reading, verbatim-verified, disagreeing with the value on the record. Not applied - a person decides. Quotation withheld because this record is not cleared for redistribution. phenomenon: temporal frequency processing shared by audition and touch, traditionally described as segregated into separate parietal and temporal systems a second reading DIFFERING from the recorded value. Recorded: auditory frequency representation in somatosensory cortex — note: a second reading, verbatim-verified, disagreeing with the value on the record. Not applied - a person decides. Quotation withheld because this record is not cleared for redistribution. extends → Liang et al. (2013) Quotation withheld because this record is not cleared for redistribution. replicates → Beauchamp and Ro (2008) Quotation withheld because this record is not cleared for redistribution. contradicts → the traditional notion of modality-dedicated cortical systems in which frequency processing is segregated into parietal and temporal streams Quotation withheld because this record is not cleared for redistribution. supports → the account that parietal responses to sound reflect auditory processing rather than partial volume contamination, from the erosion sweep Quotation withheld because this record is not cleared for redistribution. distinguishes-from → cutaneous activation by vibrating earbuds, excluded by somatotopy, by frequency profile and by a headphone control experiment differs in: THE AXIS IS A NAMED CONFOUND AND THE EQUIPMENT THAT CREATES IT. 'Quotation withheld.' The alternative explanation is a property of the apparatus, not of the brain - so every study using these headphones inherits it, and only this one says so. Quotation withheld because this record is not cleared for redistribution. explains-away → the possibility that the effect is a learned association between sounds and vibrations, since the control session involved no tactile stimulation Quotation withheld because this record is not cleared for redistribution. critiques → its own perceptual comparison, which is offline, group-averaged and not trial-wise differs in: THE AXIS IS THE GRAIN OF THE COMPARISON. 'Quotation withheld.' A brain-behaviour correspondence established between a within-subject neural measure and a between-subject behavioural one is a weaker claim than it reads as - and the paper opens the paragraph with 'admittedly, our approach can be substantially improved'. Quotation withheld because this record is not cleared for redistribution. predicts → that parietal regions respond only to the low frequency range that is also experienced through touch Quotation withheld because this record is not cleared for redistribution. somatotopic —part-of→ parietal somatotopic parietal strength of the relation: 1 Audit 2026-09-28: DOUBTFUL - read against its span and kept, but the span addresses somatotopic parietal responses; part-of is loose. Quotation withheld because this record is not cleared for redistribution. spike timing —operationalised-as→ frequency spike timing frequency strength of the relation: 1 · note: the mechanism that would explain the result and has not been shown - explicit frequency tuning in somatosensory cortex REMAINS ELUSIVE Audit 2026-09-28: DOUBTFUL - read against its span and kept, but the span says somatosensory neurons signal vibration frequency in spike timing; operationalised-as is loose. Quotation withheld because this record is not cleared for redistribution. general linear model general linear model Quotation withheld because this record is not cleared for redistribution. t-test t-test Quotation withheld because this record is not cleared for redistribution. ANOVA ANOVA Quotation withheld because this record is not cleared for redistribution. multiple-comparison correction (false discovery rate) multiple-comparison correction (false discovery rate) Quotation withheld because this record is not cleared for redistribution. correlation (Pearson) correlation (Pearson) Quotation withheld because this record is not cleared for redistribution. correlation (Spearman rank) correlation (Spearman rank) Quotation withheld because this record is not cleared for redistribution. Wilcoxon signed-rank test Wilcoxon signed-rank test Quotation withheld because this record is not cleared for redistribution. bootstrap bootstrap Quotation withheld because this record is not cleared for redistribution. confidence interval confidence interval Quotation withheld because this record is not cleared for redistribution. signal detection sensitivity (d-prime) signal detection sensitivity (d-prime) Quotation withheld because this record is not cleared for redistribution. Dice coefficient (spatial overlap) Dice coefficient (spatial overlap) Quotation withheld because this record is not cleared for redistribution. representational similarity analysis representational similarity analysis Quotation withheld because this record is not cleared for redistribution. Wilcoxon signed-rank tests Wilcoxon signed-rank tests Quotation withheld because this record is not cleared for redistribution. Spearman correlation Spearman correlation Quotation withheld because this record is not cleared for redistribution. FDR corrected FDR corrected Quotation withheld because this record is not cleared for redistribution. Pure-tone auditory stimulation produced robust, frequency-specific BOLD responses in classically defined somatosensory areas of the parietal lobe, outside the bilateral temporal lobes. Quotation withheld because this record is not cleared for redistribution. Multi-voxel activation patterns for sound pairs in the parietal lobe showed a similarity structure that correlated significantly with participants' perceptual similarity judgments of the same tones. Quotation withheld because this record is not cleared for redistribution. Auditory frequency information is therefore distributed across brain regions traditionally treated as dedicated to touch, not confined to auditory cortex. Quotation withheld because this record is not cleared for redistribution. The parietal auditory representations are not a partial-volume artifact of the neighbouring temporal lobe: they survived even when the region of interest was eroded by 10 mm from the parieto-temporal border. Quotation withheld because this record is not cleared for redistribution. Within parietal cortex, significant perceptual-cortical correspondences were found only in the postcentral gyrus and supramarginal gyrus. Quotation withheld because this record is not cleared for redistribution. Parietal voxels selected by the auditory localizer carried more robust auditory frequency representations than voxels selected by the tactile or audio-tactile localizers. Quotation withheld because this record is not cleared for redistribution. The relationship between cortical and perceptual dissimilarity for the tones appears specific to temporal and parietal cortex. Quotation withheld because this record is not cleared for redistribution. The parietal responses are not driven by the vibrating piezoelectric earbuds mechanically stimulating the ear canal: significant parietal BOLD changes were still obtained with circumaural air-conduction headphones. Quotation withheld because this record is not cleared for redistribution. Frequency tuning profiles differed between the two regions: both peaked around 200 Hz, but the somatosensory region showed a second, bimodal peak at the lowest tested frequency of 75 Hz. Quotation withheld because this record is not cleared for redistribution. robust and frequency-specific responses to auditory stimulation in CLASSICALLY DEFINED somatosensory areas, against the traditional notion of modality-dedicated cortical systems Quotation withheld because this record is not cleared for redistribution. and the parietal similarity patterns correlated significantly with perceptual similarity judgments made outside the scanner Quotation withheld because this record is not cleared for redistribution. the question is sharpened first - multimodal responses are NO LONGER DISPUTED, and whether they convey specific representations is the open question Quotation withheld because this record is not cleared for redistribution. with the precise gap named: distinct response signatures were known, and it remained unclear whether they REPRESENT specific auditory information Quotation withheld because this record is not cleared for redistribution. WHICH LOCALIZER DEFINED THE VOXELS IS ITSELF THE HYPOTHESIS TEST - three masks, asking whether the voxels privileged for sound are the ones that carry the representation Quotation withheld because this record is not cleared for redistribution. A THRESHOLD CHOSEN FOR ITS ROLE IN THE PIPELINE: liberal and uncorrected at P < 0.05, because the masks only constrain later tests Quotation withheld because this record is not cleared for redistribution. THE ANSWER DEPENDS ON THE MASK: the differences in the results acquired with the tactile localizer masks compared to the auditory and audio-tactile masks imply that brain regions preferentially responsive to auditory stimulation were more likely to exhibit significant response modulation related to auditory frequency Quotation withheld because this record is not cleared for redistribution. THE SAME DATA ARE SIGNIFICANT OR NULL BY MASK once split by hemisphere - left somatosensory ROI: auditory r = 0.24 P < 0.05, tactile r = 0.07 P = 0.25, audio-tactile r = 0.1 P = 0.15 Quotation withheld because this record is not cleared for redistribution. and against perception, significant correlations ONLY with the auditory masks, with the tactile left-hemisphere mask at P = 0.07 Quotation withheld because this record is not cleared for redistribution. THE RESULT EXISTS INSIDE A WINDOW OF ANALYSIS SIZE: the auditory mask needs more than 130 voxels, the tactile mask 50 to 200 beyond which MORE VOXELS WEAKEN IT, the audio-tactile mask only 300 to 350 Quotation withheld because this record is not cleared for redistribution. stated as a limit - parietal representations are sensitive to ROI size and voxel inclusion criteria where temporal ones are robust to the same criteria Quotation withheld because this record is not cleared for redistribution. THE artifact IS ELIMINATED BY A PARAMETRIC SWEEP: ROIs eroded by 2 mm, 4 mm and 10 mm, with the auditory-mask result surviving all three Quotation withheld because this record is not cleared for redistribution. with four separate design decisions listed against the same confound - 2 mm isotropic voxels, no volumetric smoothing, surface-space filtering, conservative borders Quotation withheld because this record is not cleared for redistribution. AND THE SPATIAL ARGUMENT IS TURNED AROUND: the significant parietal subdivisions are not the ones closest to auditory cortex, as a partial-volume account would require Quotation withheld because this record is not cleared for redistribution. the tactile stimuli were confirmed inaudible - chance performance, 0.49, without physical contact with the stimulator Quotation withheld because this record is not cleared for redistribution. and the vibrating-earbud account is refuted by SOMATOTOPY - cutaneous drive in the ear canal would confine the response to the ear representation Quotation withheld because this record is not cleared for redistribution. A LOAD-BEARING COMPARISON WHOSE EVIDENCE IS NOT IN THE PAPER: matched-frequency tactile stimulation gave profiles that clearly differed from the sound profiles, DATA NOT SHOWN Quotation withheld because this record is not cleared for redistribution. ONE CONTROL EXPERIMENT ANSWERING TWO OBJECTIONS: air-conduction headphones remove the earbud account, and that session involved no touch, removing the learned-association account Quotation withheld because this record is not cleared for redistribution. TWO NEGATIVE CONTROLS, ACROSS REGIONS AND WITHIN: visual cortex patterns do not correlate with the perceptual geometry Quotation withheld because this record is not cleared for redistribution. and voxels that DEACTIVATED in the localizer carry no relationship to auditory stimulus frequency Quotation withheld because this record is not cleared for redistribution. THE TWO SIDES COME FROM DIFFERENT SESSIONS - the cortical dissimilarity measures were compared to perceptual dissimilarity judgments tested offline and averaged over subjects, because the discrimination scans did not permit a sensitive analysis of auditory discrimination performance Quotation withheld because this record is not cleared for redistribution. somatosensory ROI dissimilarity against frequency: auditory mask r = 0.40, tactile r = 0.24, audio-tactile r = 0.25, all significant bilaterally Quotation withheld because this record is not cleared for redistribution. with the higher somatosensory dissimilarity attributed to noisier patterns, and temporal patterns called qualitatively more reliable Quotation withheld because this record is not cleared for redistribution. the somatosensory frequency profile is bimodal with a peak at 75 Hz, THE LOWEST TESTED FREQUENCY Quotation withheld because this record is not cleared for redistribution. and the stimulus set was built to exclude two confounds at once - no harmonically related frequencies, and a 5% jitter on loudness-matched amplitudes Quotation withheld because this record is not cleared for redistribution. the proposed route is a history of co-activation, with the falsifiable consequence that parietal regions should respond only to the band shared with touch Quotation withheld because this record is not cleared for redistribution. and the association may be FEATURE-SPECIFIC - frequency shared between the senses where stimulus duration appears to rely on modality-dedicated circuits Quotation withheld because this record is not cleared for redistribution. explicit frequency tuning in somatosensory cortex remains elusive, and whether auditory information is carried there by a temporal code is unknown Quotation withheld because this record is not cleared for redistribution. and fMRI's temporal resolution cannot say whether the parietal auditory responses are feedforward or feedback Quotation withheld because this record is not cleared for redistribution. cortical dissimilarity matrix (DM) cortical dissimilarity matrix (DM) Quotation withheld because this record is not cleared for redistribution. eroded ROI eroded ROI Quotation withheld because this record is not cleared for redistribution. somatosensory regions somatosensory regions Quotation withheld because this record is not cleared for redistribution. auditory regions of interest auditory regions of interest Quotation withheld because this record is not cleared for redistribution. oddball event oddball event Quotation withheld because this record is not cleared for redistribution. dissimilarity dissimilarity Quotation withheld because this record is not cleared for redistribution. erosion erosion Quotation withheld because this record is not cleared for redistribution. partial volume partial volume Quotation withheld because this record is not cleared for redistribution. localizer localizer Quotation withheld because this record is not cleared for redistribution. frequency content frequency content Quotation withheld because this record is not cleared for redistribution. perceived similarity perceived similarity Quotation withheld because this record is not cleared for redistribution. stimulus amplitude stimulus amplitude Quotation withheld because this record is not cleared for redistribution. frequency frequency Quotation withheld because this record is not cleared for redistribution. dissimilarity dissimilarity Quotation withheld because this record is not cleared for redistribution. ROI ROI Quotation withheld because this record is not cleared for redistribution. mask mask Quotation withheld because this record is not cleared for redistribution. Head motion parameters Head motion parameters Quotation withheld because this record is not cleared for redistribution. drift parameters drift parameters Quotation withheld because this record is not cleared for redistribution. Ten — participants in the main fMRI experiment Quotation withheld because this record is not cleared for redistribution. 75, 130, 195, 270, and 355 Hz — the five non-harmonically-related pure tones used as auditory stimuli Quotation withheld because this record is not cleared for redistribution. mean d ’ = 5.89 — participants' sensitivity in detecting oddball events during the scans Quotation withheld because this record is not cleared for redistribution. 37 and 54 clusters — left- and right-hemisphere clusters showing significant auditory responses under the auditory localizer mask Quotation withheld because this record is not cleared for redistribution. mean r = 0.40, W = 54, P < 0.005 — relationship between frequency distance and activation-pattern dissimilarity in the bilateral somatosensory ROI with the auditory localizer mask Quotation withheld because this record is not cleared for redistribution. mean r = 0.41, W = 55, P < 0.001 — correlation between the perceptual dissimilarity matrix and the somatosensory-ROI cortical dissimilarity matrix with the auditory mask Quotation withheld because this record is not cleared for redistribution. mean r = 0.83, W = 21, P < 0.05 — relationship between perceptual similarity ratings and absolute frequency differences between tones Quotation withheld because this record is not cleared for redistribution. 0.49 ± 0.012 — chance-level tactile detection rate when subjects had no physical contact with the stimulator, confirming the tactile stimuli were inaudible Quotation withheld because this record is not cleared for redistribution. 130 voxels — ROI size the eroded somatosensory region had to exceed before perceptual-cortical correlations reached significance with the auditory mask Quotation withheld because this record is not cleared for redistribution. r = 0.24 ... r = 0.07 ... r = 0.1 — the same somatosensory data analysed with three different localizer masks - one significant, two not, so the mask decides whether there is a finding Quotation withheld because this record is not cleared for redistribution. 130 ... 50 and 200 ... 300–350 — the ROI sizes at which each mask yields a significant brain-perception correlation, with the tactile mask WEAKENING beyond its window - a result that exists only inside an analysis-size range Quotation withheld because this record is not cleared for redistribution. 10 mm — the most extreme erosion of the parieto-temporal border at which the auditory-mask result still holds - a parametric sweep on the artifact rather than a single control Quotation withheld because this record is not cleared for redistribution. 0.49 — detection of the tactile stimulator without physical contact - chance, which is what licenses treating the vibrations as inaudible Quotation withheld because this record is not cleared for redistribution. 75 Hz — the second peak of the somatosensory frequency profile, and also the lowest frequency tested - so the profile's shape is bounded by the design Quotation withheld because this record is not cleared for redistribution.

The full text of this article is not redistributable, so it is not included here. The evidence sentences are withheld with it; the tag layer and its grounding status remain. The article itself is at https://doi.org/10.1093/cercor/bhx255.