{"id":"22b3c406-342a-4fc2-b4c6-b1710925e77f","arxiv_id":"2506.12405","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Single quasi-harmonic tones in Vitalic's electronic music are perceived by listeners as carrying multiple simultaneous pitches, an effect the paper argues producers use deliberately.","lead":"This paper runs listening tests on Vitalic's electronic music and finds that single synthesizer tones are often heard as carrying several pitches at once, with different listeners hearing different pitches. If confirmed, pitch-tracking and transcription tools for modern pop would need to handle tones that deliberately project ambiguous, multi-pitch percepts.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Intentionality claim lacks independent evidence: the new listening tests establish multi-pitch perception, not that Vitalic deliberately designs it, so the abstract and conclusion overstate what the data support.","rationale":"I considered the reader's weakest assumption, that the pitch-count measure may partially construct the multiplicity it measures. That concern is real for the quantitative synthetic-versus-acoustic contrast, but it does not threaten the core perceptual existence claim: Listening Test 2's open transcriptions and the authors' own acknowledgment of pitch-timbre ambiguity still support the statement that harmonic tones in this repertoire can convey multiple ambiguous pitches. The intentionality claim, by contrast, is part of the abstract's central claim and is supported only by private interviews and under-review preprints by the same group, so it is the least secure load-bearing premise. The reader's rationale does list intentionality among the flaws, but the weakest_assumption field selects the counting task, which is why I mark agreement as partial. My check is archival and documentary rather than computational because that is the evidence the paper currently lacks. Since the reader's verdict is already CONDITIONAL and this concern is one of the stated conditions, I keep the verdict unchanged.","tokens_in":16709,"tokens_out":7365,"duration_ms":93590,"concrete_test":"Ask the authors to release the raw audio or transcript of the interviews with Vitalic cited in Section 2.2, together with any production-session evidence in which pitch multiplicity is manipulated, and have an independent coder apply a pre-registered rubric for whether Vitalic explicitly describes the number of perceived pitches as a controllable design parameter. If no such primary source is produced, or if the coding does not support deliberate control, revise the abstract and conclusion to assert that multi-pitch perception occurs and is used, without asserting that it is intentional.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The listening tests in Sections 4 and 5 give independent support to the perceptual half of the central claim: single harmonic tones in these tracks can be perceived as carrying several pitches, with listener-dependent variation. They do not, however, measure production intent. The abstract nonetheless asserts that tonal fission is 'an active and intentional feature' and the conclusion states that Vitalic 'deliberately designs' such tones. The evidence cited for intent is Section 2.2's private interviews with Vitalic and the authors' own under-review preprints (Deruty et al., 2025a,c); Section 8.1 then treats 'the producer's deliberate use' as established. The public MusicRadar Tech 2014 reference concerns 'a melody inside the bass,' which does not by itself document control over the number of perceived pitches. Since the practical consequence for pitch tracking and for MIR depends on the phenomenon being deliberately designed rather than an incidental byproduct of synthesis choices, this is a real support gap. The perceptual results can stand; the 'intentional' component is the load-bearing part that needs independent verification before the abstract's stronger claim is accepted.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper argues that a single harmonic complex tone in contemporary popular music can convey multiple simultaneous pitches (which the authors call 'tonal fission') and claims that this is an active and intentional feature in the work of electronic producer Vitalic. The authors present two listening tests: (1) 59 participants reported the number of simultaneous pitches perceived in twelve short extracts (nine electronic Vitalic excerpts, one piano transcription, and two classical solo-string controls); and (2) seven expert listeners transcribed the pitches they perceived in six samples (cello, two Vitalic synthesizer tones, a 'No Fun' inharmonic tone, a distorted guitar power chord, and an 808-style bass patch) using piano matching. Signal analyses with STFT and short-term autocorrelation were used to relate perceived pitches to spectral partials and temporal peaks. The results show that the synthetic harmonic tones were perceived as carrying more simultaneous pitches than the acoustic controls, with large inter-listener variation, while the cello transcription was unanimous single-pitch. The paper concludes that pitch is not an intrinsic property of a tone but a listener-constructed, context-dependent percept, and that harmonic tones in contemporary popular music can deliberately project multiple ambiguous pitches.","tokens_in":16847,"tokens_out":9971,"duration_ms":116318,"significance":"The paper's new listening-test data are a genuine contribution: they are independent of the authors' prior signal analyses, the procedure is reported in reasonable detail, and the limitations (uncontrolled listening conditions, habituation, piano-matching precision) are acknowledged rather than hidden. If the perceptual result holds, it strengthens the case that single-f0 pitch trackers and score-based transcription are inadequate for a nontrivial share of contemporary popular music, and it provides an empirical bridge between psychoacoustic demonstrations of pitch ambiguity and real studio-produced records. The supplementary audio examples are also a useful resource for replication. The main weakness is that the strongest framing—that tonal fission is 'active and intentional'—is not established by the tests; it rests on private interviews and under-review preprints. The study is therefore better described as demonstrating that such tones can be perceived as multi-pitch, not that producers intentionally design them to be. The paper also over-generalizes from classical-string controls and a small Vitalic sample to 'contemporary popular music' at large.","major_comments":[{"comment":"The response variable in Listening Test 1 (number of perceived simultaneous pitches) may be partly constructed by the task. Section 4.1 concedes that 'the more listeners focused on the number of pitches, the more they perceived,' and Section 7.1 reports that all expert listeners found it difficult to disentangle pitch from timbre, with one listener choosing to hear the extra pitches as timbre. The transcription test provides convergent evidence, but it likewise instructs listeners to report multiple pitches. The paper should quantify test–retest reliability beyond a single anecdotal retest, add a free-description control condition, or explicitly model attention/focus; otherwise the synthetic-versus-acoustic gap in Figures 1 and 2 may reflect task framing rather than a stable property of the tones.","section":"§4.1, §7.1, Figures 1–2"},{"comment":"The logistic regression analysis is under-specified. The text states only that 'the number of perceived pitches' was treated as an ordinal variable with controls for age, listening device, and environment. Because each participant rated all extracts, the observations are not independent; a mixed-effects model with random intercepts for participant and sample (or cluster-robust standard errors) is needed. Without this, the reported p-values for musical expertise (p<0.001) and audio expertise (p=0.209) are not trustworthy, and no effect size or confidence interval is provided. Please report the full model specification or re-analyze with appropriate random effects and provide effect sizes.","section":"§4.2"},{"comment":"The acoustic control samples are classical solo strings (cello, violin) plus a piano transcription, not acoustic instruments typical of contemporary popular music. The abstract's comparison to 'acoustic counterparts' is therefore confounded with genre and cultural familiarity. Either add acoustic CPM controls (e.g., clean electric bass, acoustic guitar, vocals) or explicitly restrict the claim to 'compared with classical acoustic solo instruments.' This also bears on the generalization from a mostly Vitalic sample to 'contemporary popular music' at large, which the current data do not support.","section":"§3.3, Table 1, Figure 1"},{"comment":"The treatment of 'No Fun' (sample 1) as a special case defines 'f0' as the frequency spacing between adjacent partials. For an inharmonic tone whose partials are shifted rather than exact multiples of that spacing, this pseudo-f0 is not a fundamental frequency, and classifying perceived pitches as 'harmonics' of this spacing is arbitrary. This affects the 'Fundamental' and 'Harmonics' columns for sample 1 in Table 2 and Figure 4. Please report the raw partial frequencies and perceived pitches for this sample separately, or remove it from the harmonic-category framework.","section":"§5.3, §6.4, Table 2"}],"minor_comments":[{"comment":"The text says \"'No Fun' (sample 7)\" but should be \"sample 1.\"","section":"§5.3"},{"comment":"In the 808 paragraph, \"The results resemble those of samples 2, 7, and 14\" should presumably read \"2, 7, and 13,\" since sample 14 is the 808 itself.","section":"§5.2"},{"comment":"The phrase \"non-acoustic quasi-harmonic tones (samples 3, 7, 13, and 14)\" should be \"2, 7, 13, and 14,\" because sample 3 was not included in Listening Test 2.","section":"§6.7"},{"comment":"The abbreviation \"sple.\" should be \"sample.\"","section":"Table 2"},{"comment":"The sentence \"This definition is adopted in Section 2.3\" refers to the common-fate definition of a tone, but Section 2.3 defines partials and harmonics, not the common-fate concept; the cross-reference should be corrected.","section":"§8.1"},{"comment":"The claim that Vitalic \"confirmed in private interviews\" should be documented with dates, recordings, or a transcript; otherwise it is an unverifiable personal communication.","section":"§2.2"},{"comment":"The single participant who took the test twice and reported 2.4 then 3 pitches is anecdotal; report formal test–retest statistics or remove this datum.","section":"§4.1"},{"comment":"The statement that 38 of 72 songs \"feature instrumental tracks based on tones that evoke more than one pitch\" is based on the authors' own critical listening; label this as an informal survey rather than a measured result.","section":"§3.3"},{"comment":"The participant demographics (age, musical expertise, listening environment, monitoring system) are summarized only in percentages in the text; a table with full distributions would aid reproducibility.","section":"§4"}],"recommendation":"major_revision","confidential_remarks":"The manuscript relies heavily on the authors' own under-review preprints (Deruty et al., 2025a,c) for both intentionality and signal-analysis context. Since those works are not publicly available, the reader cannot verify the independence of the listening-test results from the earlier claims. I also note that one co-author is affiliated with Vitalic's label (Citizen Records) and that the producer is described as being in contact with the authors; a formal conflict-of-interest or disclosure statement would be appropriate. The empirical core of the paper is valuable, but the intentionality claim and the statistical analysis need to be strengthened before publication."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The genuinely new thing here is the listening-test data: 59 people counting pitches and seven experts transcribing them, with the WJS metric for consensus. That is real evidence that synthetic harmonic tones in this repertoire can be perceived as carrying several pitches, and that the perceived set varies with listener and playback conditions. The paper is also honest about its limits—uncontrolled listening conditions, habituation, pitch-timbre confusion, the difficulty of low-register pitch naming. I believe the perceptual result. If you work on pitch tracking or transcription, this is a useful reminder that a single quasi-harmonic tone can behave like a pitch distribution.\n\nThe soft spots are real but not fatal to that core result. The acoustic control is a cello suite and a Bach violin chaconne, so the synthetic-vs-acoustic contrast is tangled up with genre, register, and musical material. The pitch-count measure may partly construct the effect: Section 4.1 concedes that focusing on counting increases the count, and Section 7.1 says every expert struggled to separate pitch from timbre. That doesn't make the multi-pitch perception an artifact, but it does mean the reported counts are task-dependent. The signal analyses in Section 6 are plausible and nicely grounded in spectral versus temporal models, but they are descriptive, not statistical.\n\nThe bigger gap is the intentionality claim. The abstract says tonal fission is \"an active and intentional feature\" and the conclusion says Vitalic \"deliberately designs\" these tones. The listening tests don't measure intention. The evidence for intent is a private interview and the authors' own under-review preprints. The public MusicRadar quote about \"a melody inside the bass\" doesn't document control over the number of perceived pitches. So the strong version of the abstract claim is not independently supported. The weaker claim—that these tones do carry multiple ambiguous pitches in practice, whether or not the producer fully controls them—is what the data actually show.\n\nIf I were the editor I would send this to peer review. The empirical core deserves scrutiny, and a good referee can push for matched controls, a formal statistical test of the synthetic-versus-acoustic contrast, and publication of the raw response data. But the authors should be asked to either temper the intentionality language or support it with something more than interviews and preprints. This is a solid paper with a fixable overreach.","headline":"The listening tests are a genuinely useful demonstration that synthetic harmonic tones can carry multiple perceived pitches, but the paper's intentionality claim outruns its evidence and the acoustic controls are confounded.","tokens_in":17471,"tokens_out":1085,"would_cite":true,"duration_ms":15107,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper claims that producing several simultaneous pitches from one harmonic tone is an intentional, active feature of contemporary popular music, demonstrated through listening tests on Vitalic's music.","keywords":["tonal fission","pitch perception","harmonic complex tones","contemporary popular music","synthetic tones","pitch ambiguity","listening tests","pitch transcription"],"falsifier":"A controlled experiment presenting the same synthetic harmonic tones under a neutral instruction that never mentions counting or multiple pitches, such as asking listeners to play back only the main melody, would weaken the strong claim if single-pitch responses dominate; additionally, matching synthetic and acoustic tones for overall spectrum, loudness, and duration should make the reported multiplicity gap disappear if it is an artifact of attention rather than of tone design.","tokens_in":16455,"feed_emoji":"🎵","tokens_out":6938,"duration_ms":80818,"temperature":0.7,"pith_summary":"The paper sets out to show that a single harmonic complex tone can deliberately convey several distinct pitches at once, a phenomenon it calls tonal fission, and that this is an active feature of contemporary popular music rather than a laboratory curiosity. Using excerpts from Vitalic's music plus control sounds such as cello, distorted guitar, and an 808-style bass, the authors ran two listening tests: one where 59 listeners counted simultaneous pitches and one where seven trained listeners transcribed the pitches they heard. The synthetic harmonic tones transmitted more perceived pitches than the acoustic tones, with significant variation across listeners and across listening conditions. If the claim holds, pitch in this repertoire cannot be treated as a single fixed value attached to each tone, and pitch-tracking and transcription methods that assume one fundamental per tone will misread much of this music.","feed_headline":"Single synthetic tones can carry multiple simultaneous pitches","feed_subtitle":"Two listening tests show electronic-music tones are deliberately pitch-ambiguous, challenging single-pitch analysis.","key_machinery":"The central object is tonal fission, the perception of multiple distinct pitches from a single quasi-harmonic complex tone, measured through listening tests and linked to signal properties. The explanatory machinery is the pairing of two standard pitch-perception models: spectral modeling, which derives perceived pitches from prominent resolved partials, and temporal modeling, which derives pitch from autocorrelation peaks, analyzed here through short-term autocorrelation (STAC). The paper argues that when the two models point to different pitches, or when upper partials are loud enough to be individually audible, the tone becomes ambiguous and can convey several pitches at once. Short-term Fourier transforms, STAC, and ISO 226 equal-loudness weighting are used to connect each listener's transcribed pitches to partials and autocorrelation features, and a certainty-weighted Jaccard similarity quantifies how much listeners' pitch sets agree.","core_discovery":"The central discovery is that contemporary popular music, especially electronic music, uses tonal fission as a deliberate compositional resource: a single quasi-harmonic tone can be designed to project several ambiguous pitches simultaneously. In Listening Test 1, no sample, acoustic or electronic, produced unanimous single-pitch perception, and the electronic excerpts, particularly inharmonic ones, were associated with significantly more perceived pitches. In Listening Test 2, trained listeners transcribed multiple pitches from the synthetic and distorted tones, with pitch sets varying by listener and often concentrated on octaves or harmonics of the fundamental; only the acoustic cello produced unanimous, stable single-pitch transcriptions. Signal analysis links the extra pitches to loud upper partials, which support spectral pitch models, and to autocorrelation structure that supports a different pitch than the fundamental, consistent with temporal pitch models. The paper concludes that pitch should be understood as a listener-dependent, context-dependent set of possible pitches rather than an intrinsic property of the tone.","pith_inferences":["Unstated in the paper: the counting task in Listening Test 1 may itself inflate the reported multiplicity, since the authors note that more focused attention on counting led to more perceived pitches; a neutral task that never mentions the number of pitches could produce lower, more stable counts.","The multistability language suggests a testable extension: presenting the same tone with different verbal or visual primes, such as 'listen for the bass line' versus 'listen for the upper line,' should shift which pitch set is heard, much as Necker cube orientation can be primed.","If pitch is a probability distribution over a set of possible pitches, then a practical production metric is the effective number of pitches or the pitch ambiguity of a tone, which could be predicted from spectral and autocorrelation features and used to control ambiguity in music generation.","The paper does not directly measure whether tonal fission contributes to listener engagement or repeated listening; a controlled experiment varying pitch ambiguity while holding other musical factors constant could test the proposed satiation-delay hypothesis."],"forward_implications":["Pitch-tracking and audio-to-MIDI systems that assume one fundamental per tone will systematically misrepresent much contemporary popular music, including over half of Vitalic's discography by the paper's count.","The perceived pitch set of a tone is not fixed by the signal alone; it shifts with listening conditions, playback system, and listener expertise, so pitch analysis in this repertoire should report a distribution rather than a point estimate.","Producers can treat the number of pitches per tone as a controllable musical parameter, shaped by choosing partial loudness and autocorrelation structure, which the paper reports Vitalic does intentionally.","The phenomenon extends beyond electronic music: a distorted guitar power chord becomes a single quasi-harmonic tone conveying several pitches, suggesting that score-based analysis of rock and metal also needs revision.","If pitch is a set of possible percepts rather than a single value, then conventional definitions of a musical note, and the reification of the score, apply poorly to this body of music."],"supporting_citations":[{"why":"Establishes Vitalic's pitch-deviation conventions and the autocorrelation features that the present study reuses to connect perceived pitches to tone structure.","marker":"Deruty et al. (2025a)"},{"why":"Documents deliberate pitch uncertainty in commercial electronic music, supporting the claim that tonal fission is an intentional stylistic technique rather than an artifact.","marker":"Deruty et al. (2025c)"},{"why":"Supplies the spectral-versus-temporal distinction that structures the signal analysis of perceived pitches.","marker":"Yost (2009)"},{"why":"Demonstrates with controlled stimuli that listeners can hear multiple pitches from a single complex tone, providing the perceptual baseline the paper extends to musical context.","marker":"Moore et al. (2006)"},{"why":"Provides early evidence of ambiguous pitch organization from complex tones, cited as background for tonal fission.","marker":"Shepard (1964)"},{"why":"Source separation algorithm that produced the isolated bass and synth tracks analyzed in the listening tests.","marker":"Stöter et al. (2019)"},{"why":"Equal-loudness contour weighting applied before spectral analysis to align signal measurements with what listeners hear.","marker":"ISO (2023)"},{"why":"Spectral model of pitch formation used to associate perceived pitches with resolved partials.","marker":"Goldstein (1973)"},{"why":"Temporal and autocorrelation model of pitch used to associate perceived pitches with autocorrelation structure.","marker":"Licklider (1951)"}],"fun_headline_variants":["One tone, many pitches: electronic music's hidden ambiguity","Electronic tones are designed to sound like multiple pitches at once","Pitch ambiguity is a deliberate feature in electronic music","Synthetic tones can project several pitches simultaneously, study finds","Tonal fission: one tone, multiple pitches in contemporary pop"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that a listener's count of how many simultaneous pitches are heard is a stable, valid measurement of what the tone conveys, even though the paper itself reports that more attention to counting increased the number reported and that all expert listeners had difficulty separating pitch from timbre.","fun_headline_variants_meta":{"raw":{"variants":["One tone, many pitches: electronic music's hidden ambiguity","Electronic tones are designed to sound like multiple pitches at once","Pitch ambiguity is a deliberate feature in electronic music","Synthetic tones can project several pitches simultaneously, study finds","Tonal fission: one tone, multiple pitches in contemporary pop"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000248,"raw_usage":{"total_tokens":1519,"prompt_tokens":892,"completion_tokens":627,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":508,"completion_tokens_details":{"reasoning_tokens":546}},"tokens_in":508,"tokens_out":627,"duration_ms":6481,"temperature":1.0,"reasoning_tokens":546,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T00:51:07.906531+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A controlled experiment presenting the same synthetic harmonic tones under a neutral instruction that never mentions counting or multiple pitches, such as asking listeners to play back only the main melody, would weaken the strong claim if single-pitch responses dominate; additionally, matching synthetic and acoustic tones for overall spectrum, loudness, and duration should make the reported multiplicity gap disappear if it is an artifact of attention rather than of tone design.","supporting_citations":[{"cited_title":"C., Glasberg, B","cited_arxiv_id":null,"evidence_quote":"Demonstrates with controlled stimuli that listeners can hear multiple pitches from a single complex tone, providing the perceptual baseline the paper extends to musical context."},{"cited_title":"Normal equal-loudness level contours-ISO 226: 2023","cited_arxiv_id":null,"evidence_quote":"Equal-loudness contour weighting applied before spectral analysis to align signal measurements with what listeners hear."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Spectral model of pitch formation used to associate perceived pitches with resolved partials."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Temporal and autocorrelation model of pitch used to associate perceived pitches with autocorrelation structure."}],"review_version":2}