Rapid Tuning of Auditory “What” and “Where” Pathways by Training

Rapid Tuning of Auditory “What” and “Where” Pathways by Training Du He Arnott Ross Wu Li Alain

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.

Rapid Tuning of Auditory “What” and “Where” Pathways by Training 2015 Cerebral Cortex (New York, NY) 10.1093/cercor/bht251 Du2015-rapid-tuning-of-auditory-what-and-where-path Du2015-rapid-tuning-of-auditory-what-and-where-path.txt, the project's staged copy d0e261f848c72df45392fe35b878b63a9618f873 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 whether improvement in the first hour of training is task familiarisation or a real change in what the perceptual system is tuned to Quotation withheld because this record is not cleared for redistribution. whether 45 minutes of training on one acoustic cue produces gains and brain activity specific to that cue rather than general to the task Quotation withheld because this record is not cleared for redistribution. a double dissociation in which listeners trained on pitch or on location gain specifically from that cue and recruit ventral or dorsal areas respectively, with the stimuli identical across groups Quotation withheld because this record is not cleared for redistribution. existence Quotation withheld because this record is not cleared for redistribution. rapid feature-specific auditory perceptual learning Quotation withheld because this record is not cleared for redistribution. electrophysiology electrophysiology Quotation withheld because this record is not cleared for redistribution. mechanisms the training session acts as a priming episode that biases top-down attention toward the trained acoustic cue, so that cue is processed as more salient in the later untrained test Quotation withheld because this record is not cleared for redistribution. mechanisms the implicit attentional bias makes the trained spectral or spatial difference more salient than the untrained one, which differentially recruits the ventral or dorsal auditory stream Quotation withheld because this record is not cleared for redistribution. training may act as a priming task that biases attention to the trained cue, making it more salient later even without instruction Quotation withheld because this record is not cleared for redistribution. the selectivity survives the delay to the MEG session and the absence of feedback there Quotation withheld because this record is not cleared for redistribution. training changes how the two cues combine, not just how well each is used Quotation withheld because this record is not cleared for redistribution. the animal precedent is that receptive fields shift within minutes of a task, and much less when the same sounds are heard passively Quotation withheld because this record is not cleared for redistribution. A repetition/familiarity effect cannot be fully excluded as a partial cause of the behavioural and neural changes. Quotation withheld because this record is not cleared for redistribution. The study cannot say whether these rapid source-activity changes are precursors of longer-term plasticity; that needs combined MEG and fMRI work. Quotation withheld because this record is not cleared for redistribution. the dissociation is a bias in recruitment rather than an all-or-none separation of the two streams Quotation withheld because this record is not cleared for redistribution. the untrained comparison group comes from a previous study rather than being run alongside these two Quotation withheld because this record is not cleared for redistribution. the frequency group's gain for the trained one-semitone cue did not improve across training blocks, unlike the location group's Quotation withheld because this record is not cleared for redistribution. repetition priming remains a possible partial contributor, which the paper argues against rather than excludes Quotation withheld because this record is not cleared for redistribution. the what and where stream division, applied to audition Quotation withheld because this record is not cleared for redistribution. HRTFs, used to render spatial separation over headphones Quotation withheld because this record is not cleared for redistribution. the finding that task relevance, not exposure, drives rapid receptive-field change Quotation withheld because this record is not cleared for redistribution. the first training block, before any training, compared between the two groups at the 0-semitone / 0-degree baseline eliminates: a pre-existing difference between the two groups as the source of the later divergence · outcome (stated by the tagger, no quotation): held Quotation withheld because this record is not cleared for redistribution. comparison against participants trained on the OTHER feature and against an untrained group from an earlier study eliminates: improvement caused by mere exposure to the stimuli rather than by training on the specific feature · outcome (stated by the tagger, no quotation): held Quotation withheld because this record is not cleared for redistribution. analysing the GAIN attributable to frequency or spatial separation rather than raw accuracy eliminates: familiarity with the same stimuli across the training and the recording session as an explanation of the effect · outcome (stated by the tagger, no quotation): held Quotation withheld because this record is not cleared for redistribution. double dissociation here means a RELATIVE bias in recruitment, not categorical separation - the paper says the group difference was not all or none Quotation withheld because this record is not cleared for redistribution. the gain measure is always a DIFFERENCE from the same-pitch same-location baseline, which is what makes it independent of stimulus familiarity Quotation withheld because this record is not cleared for redistribution. spatial separation is simulated over headphones by a HRTF chosen per participant, so location is a rendered cue rather than a physical one Quotation withheld because this record is not cleared for redistribution. feature-specific learning is contrasted throughout with procedural learning, defined as strategy change from task familiarisation Quotation withheld because this record is not cleared for redistribution. the ventral and dorsal streams are imported as established what and where pathways, and the paper's dissociation is read against that prior assignment rather than establishing it Quotation withheld because this record is not cleared for redistribution. method: 45 minutes of training on either fundamental-frequency or spatial separation, then MEG during an identical concurrent-vowel task for both groups a second reading DIFFERING from the recorded value. Recorded: electrophysiology — 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: rapid feature-specific perceptual learning within a single session a second reading DIFFERING from the recorded value. Recorded: rapid feature-specific auditory perceptual learning — 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. builds-on → Du et al. 2011 Quotation withheld because this record is not cleared for redistribution. extends → Alain et al. 2007; Alain, Campeanu, et al. 2010; Ben-David et al. 2011; Spierer et al. 2007 Quotation withheld because this record is not cleared for redistribution. contradicts → the view that first-hour improvement is procedural learning without feature specificity Quotation withheld because this record is not cleared for redistribution. extends → behavioural reports of within-session specificity, by supplying the neurophysiological evidence they lacked status: the paper's own words: remains to be determined Quotation withheld because this record is not cleared for redistribution. confirms → animal work showing receptive fields change within minutes of a task and much less during passive listening Quotation withheld because this record is not cleared for redistribution. builds-on → Du et al. 2011, whose untrained group supplies the additivity baseline Quotation withheld because this record is not cleared for redistribution. supports → the ventral what and dorsal where division of auditory cortical processing Quotation withheld because this record is not cleared for redistribution. procedural learning —distinct-from→ perceptual sensitivity procedural learning perceptual sensitivity strength of the relation: 1 · note: CRITERION, stated in advance: procedural learning predicts comparable effects across differently trained groups, perceptual change predicts feature-specific gains. The rival is given its own prediction Quotation withheld because this record is not cleared for redistribution. ventral —contrasted-with→ dorsal ventral dorsal strength of the relation: 0.8 · note: SCOPED: the paper calls the group difference a bias in recruitment and not all or none, so the weight is deliberately below 1 Quotation withheld because this record is not cleared for redistribution. IFG —part-of→ ventral IFG ventral strength of the relation: 0.9 Audit 2026-09-28: DOUBTFUL - read against its span and kept, but the span lists IFG among regions more active in one group; part-of 'ventral' is loose. Quotation withheld because this record is not cleared for redistribution. IPL —part-of→ dorsal IPL dorsal strength of the relation: 0.9 Audit 2026-09-28: DOUBTFUL - read against its span and kept, but the span lists parietal regions as posterior and dorsal; part-of 'dorsal' is loose. Quotation withheld because this record is not cleared for redistribution. gain in accuracy —distinct-from→ repetition priming gain in accuracy repetition priming strength of the relation: 0.9 · note: CRITERION: the measure is a difference from a baseline present in both phases, so stimulus familiarity cancels Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but the span argues the gain is unlikely to be repetition priming; distinct-from is loose. Quotation withheld because this record is not cleared for redistribution. ANOVA ANOVA Quotation withheld because this record is not cleared for redistribution. post hoc test post hoc test Quotation withheld because this record is not cleared for redistribution. ANOVA ANOVA Quotation withheld because this record is not cleared for redistribution. Monte Carlo simulation (multiple-comparison correction) Monte Carlo simulation (multiple-comparison correction) Quotation withheld because this record is not cleared for redistribution. correlation correlation Quotation withheld because this record is not cleared for redistribution. standard error of the mean standard error of the mean about this quote: This quote ends with the stray word 'In', the first word of the next sentence. It is not trimmed, because the quote without it occurs more than once in the paper's text and could not be tied to one place. The claim is carried by the sentence before 'In'. Quotation withheld because this record is not cleared for redistribution. repeated-measures ANOVA repeated-measures ANOVA Quotation withheld because this record is not cleared for redistribution. paired t -tests paired t -tests Quotation withheld because this record is not cleared for redistribution. Pearson correlation Pearson correlation Quotation withheld because this record is not cleared for redistribution. family-wise corrected family-wise corrected Quotation withheld because this record is not cleared for redistribution. Groups trained on a spectral versus a spatial cue showed a double dissociation in brain activity, ventral areas for the spectral group and dorsal areas for the spatial group, despite identical stimuli. Quotation withheld because this record is not cleared for redistribution. The frequency-trained group gained more from a pitch difference alone than from a location difference alone. Quotation withheld because this record is not cleared for redistribution. The group difference in cortical source activity appeared within roughly 100 ms of sound onset, which the authors read as a pre-stimulus attentional bias toward the trained feature. Quotation withheld because this record is not cleared for redistribution. Because both groups had identical stimulus-response demands, the difference cannot easily be explained as procedural learning. Quotation withheld because this record is not cleared for redistribution. After training, the sum of the gains from each cue alone was considerably larger than the gain from both cues together. Quotation withheld because this record is not cleared for redistribution. The two groups improved equally with practice overall, so the feature-specific effect is not an artefact of one group simply learning more. Quotation withheld because this record is not cleared for redistribution. The authors call this the first demonstration of a feature-specific gain in human brain activity after brief training. Quotation withheld because this record is not cleared for redistribution. both groups improved with practice during the training session Quotation withheld because this record is not cleared for redistribution. the frequency-trained group gained more from a pitch difference and the location-trained group more from a spatial difference Quotation withheld because this record is not cleared for redistribution. the group by trial-type interaction was large and each group beat the other on its own trained cue Quotation withheld because this record is not cleared for redistribution. no group difference remained when both cues were available at once Quotation withheld because this record is not cleared for redistribution. the frequency-trained group recruited inferior frontal and temporal areas more strongly Quotation withheld because this record is not cleared for redistribution. the location-trained group recruited parietal and posterior temporal areas more strongly Quotation withheld because this record is not cleared for redistribution. the two networks differed although both groups heard exactly the same stimuli Quotation withheld because this record is not cleared for redistribution. right inferior frontal activity at 80-120 ms correlated with individual gain from the pitch cue Quotation withheld because this record is not cleared for redistribution. in both trained groups the gain from the trained cue was almost as large as the gain from both cues together Quotation withheld because this record is not cleared for redistribution. CONTROL NULL: no group by block interaction, so both groups improved equally overall and the specificity is not one group learning more Quotation withheld because this record is not cleared for redistribution. CONTROL: the same-pitch same-location condition appeared in both phases and showed little repetition priming Quotation withheld because this record is not cleared for redistribution. ASYMMETRY: across training blocks the location group improved its use of the trained cue while the frequency group did not Quotation withheld because this record is not cleared for redistribution. SCOPE STATED: the group difference is a bias in recruitment rather than an all-or-none separation Quotation withheld because this record is not cleared for redistribution. procedural learning procedural learning Quotation withheld because this record is not cleared for redistribution. trial type trial type Quotation withheld because this record is not cleared for redistribution. synthetic aperture magnetometry (SAM) synthetic aperture magnetometry (SAM) Quotation withheld because this record is not cleared for redistribution. procedural learning procedural learning Quotation withheld because this record is not cleared for redistribution. HRTF HRTF Quotation withheld because this record is not cleared for redistribution. double dissociation double dissociation Quotation withheld because this record is not cleared for redistribution. training group training group Quotation withheld because this record is not cleared for redistribution. trial type trial type Quotation withheld because this record is not cleared for redistribution. source activity source activity Quotation withheld because this record is not cleared for redistribution. bottom-up sensory input bottom-up sensory input Quotation withheld because this record is not cleared for redistribution. HRTF coefficient HRTF coefficient Quotation withheld because this record is not cleared for redistribution. gain in accuracy gain in accuracy Quotation withheld because this record is not cleared for redistribution. semitone semitone Quotation withheld because this record is not cleared for redistribution. MEG MEG Quotation withheld because this record is not cleared for redistribution. R 2 = 0.58 — variance shared between right inferior frontal gyrus source activity at 80-120 ms and the behavioural gain from the pitch cue Quotation withheld because this record is not cleared for redistribution. R 2 = 0.63 — variance shared between right inferior parietal lobule source activity at 80-120 ms and the behavioural gain from the location cue Quotation withheld because this record is not cleared for redistribution. about 40% accuracy (chance 25%) — baseline double-vowel identification accuracy of both groups in the first training block Quotation withheld because this record is not cleared for redistribution. t 22 = 4.17, P < 0.001 / t 22 = 6.18, P < 0.001 — between-group tests confirming each group benefited more than the other from its own trained cue Quotation withheld because this record is not cleared for redistribution. F 2,44 = 33.32, P < 0.001 — group-by-trial-type interaction, the statistical core of the feature-specific behavioural effect Quotation withheld because this record is not cleared for redistribution. 45-min — length of the single training session that produced the effects Quotation withheld because this record is not cleared for redistribution. all t 11 > 2.20, P < 0.05 — improvement of the Location group's use of the 90-degree separation cue from block 1 to blocks 2-4 Quotation withheld because this record is not cleared for redistribution. 45 min — duration of the training session that produced the effect Quotation withheld because this record is not cleared for redistribution. 40% against 25% chance — baseline accuracy before training, equal in both groups Quotation withheld because this record is not cleared for redistribution. F 2,44 = 33.32 — the group by trial-type interaction carrying the behavioural dissociation Quotation withheld because this record is not cleared for redistribution. R squared = 0.58 — correlation between right inferior frontal activity and individual pitch-cue gain Quotation withheld because this record is not cleared for redistribution. 24 — participants, split into two groups of twelve Quotation withheld because this record is not cleared for redistribution. 95% or better — single-vowel accuracy every participant reached before training began Quotation withheld because this record is not cleared for redistribution. 1088 μL — minimum cluster size for a reported activation, with family-wise correction 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/bht251.