pith. sign in

arxiv: 2606.24367 · v1 · pith:NPLSN2OPnew · submitted 2026-06-23 · 💻 cs.SD · stat.AP

Statistical validation and full-sphere extension of a Bayesian model for human static sound localisation

Roberto Barumerli , Fabian Brinkmann , Emanuele Zanoni , Anton Hoyer , Lorenzo Picinali , Michele Geronazzo This is my paper

Pith reviewed 2026-06-25 22:26 UTC · model grok-4.3

classification 💻 cs.SD stat.AP
keywords Bayesian modelsound localisationHRTFspatial hearingparameter recoverytemplate interpolationstatistical validation
0
0 comments X

The pith

Bayesian sound localisation model recovers individual sensorimotor and spectral parameters from behavioral data and shows full-sphere HRTF coverage determines template quality more than interpolation method

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper derives an explicit likelihood function for a Bayesian model of static sound localisation that jointly infers sound direction from noisy perceptual features and individual head-related transfer function parameters. It validates this likelihood through parameter recovery experiments on simulated data and by fitting the model to behavioral responses from 33 participants, confirming that the framework identifies personal parameters reliably. The same framework is then applied to compare four HRTF template interpolation methods, establishing that full-sphere spatial coverage and high-frequency spectral fidelity are the dominant factors for template quality while the choice of interpolation algorithm is secondary. A sympathetic reader would care because the work supplies a statistical basis for both studying how humans locate sounds and selecting effective HRTF data for applications such as virtual audio.

Core claim

Deriving an explicit likelihood enables the Bayesian model to reliably identify individual sensorimotor and spectral parameters via recovery on simulated data and fitting to responses from 33 participants, and when the model compares HRTF templates it shows that full-sphere spatial coverage together with high-frequency spectral fidelity are the primary determinants of template quality while the specific interpolation algorithm is secondary.

What carries the argument

The explicit likelihood function for the Bayesian model that jointly infers sound direction and individual HRTF parameters from noisy perceptual features

If this is right

  • The validated framework can be applied to statistically address other fundamental questions in spatial hearing research.
  • HRTF template selection and evaluation can prioritize datasets with full-sphere coverage and high-frequency fidelity over choice of interpolation algorithm.
  • Model-based statistical methods become usable for both basic perceptual studies and applied tasks such as perceptual HRTF evaluation.
  • The released open-source implementation allows direct application of the likelihood-based validation to new datasets.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The parameter-recovery approach could be tested on data from dynamic or moving-source localisation tasks to check whether the same individual parameters remain identifiable.
  • Parameters recovered this way might be used to generate listener-specific HRTF predictions that improve performance in virtual-reality audio rendering.
  • Applying the comparison across additional HRTF measurement datasets could identify which acoustic measurement densities best support accurate localisation modelling.

Load-bearing premise

The derived likelihood correctly captures the noise structure in human perceptual features and the responses from the 33 participants suffice to identify the individual parameters without overfitting or bias.

What would settle it

New simulated datasets with altered noise structures yield poor recovery of the known parameters, or the fitted model fails to predict localization responses on held-out behavioral trials from the same participants.

Figures

Figures reproduced from arXiv: 2606.24367 by Anton Hoyer, Emanuele Zanoni, Fabian Brinkmann, Lorenzo Picinali, Michele Geronazzo, Roberto Barumerli.

Figure 1
Figure 1. Figure 1: Illustration of a single modelled localisation trial for a sound presented at ϕ (red dot). Bluescale map: log-posterior distribution computed by evaluating Eq. 3 for all directions of the template grid. Red dot: actual source direction ϕ. Green triangle: listener-internal model estimate (Eq. 6). Orange triangle: final listener response (Eq. 7). Front direction is indicated with a black square for reference… view at source ↗
Figure 2
Figure 2. Figure 2: HRTF measurement and interpolated grid vi￾sualised over spherical coordinates. Blue points show the original measured directions of the G.R.A.S. KEMAR HRTF dataset [38]; red circles indicate complemented directions in the bottom hemisphere where cues are ex￾trapolated in the first step of the two-step interpolation method by mirroring measured directions across the hor￾izontal plane (see Sec. 2.5). Grey po… view at source ↗
Figure 3
Figure 3. Figure 3: Patterns of spectral amplitudes cues on the median plane (left ear) originating from unprocessed G.R.A.S KEMAR HRTFs [38] (top) and four interpola￾tion methods. Each subplot is independently normalised to the maximum value to highlight spatial structure. otherwise. The interpolation to the template grid is then realised by xT = Y T xnm (16) with Y T ∈ R T ×(NSH+1)2 holding the basis functions for the posit… view at source ↗
Figure 4
Figure 4. Figure 4: replicates [PITH_FULL_IMAGE:figures/full_fig_p007_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Model predictions of localisation metrics across the range of observed behaviour, serving as posterior pre￾dictive checks for the four interpolation methods. Each panel shows a different localisation metric (lateral error, polar error, and quadrant error rate), with actual be￾havioural responses on the x-axis and simulated locali￾sation responses generated from fitted model parameters on the y-axis. The da… view at source ↗
Figure 6
Figure 6. Figure 6: Model fit quality across template interpolation methods. Points and error bars show mean ∆BIC (± stan￾dard error over 33 participants) relative to the best-fitting method for each participant. Lower values indicate better fits. Method ∆BIC Wins Very strong SHMAX -19.4 ± 10.1 19/33 17 barycentric -12.5 ± 16.7 18/33 15 SH -2.8 ± 15.3 15/33 13 [PITH_FULL_IMAGE:figures/full_fig_p010_6.png] view at source ↗
read the original abstract

Auditory models are central tools for studying spatial hearing, yet their validation typically relies on heuristic performance metrics rather than principled statistical methods. We present two contributions building on a Bayesian sound localisation model that jointly infers sound direction from noisy perceptual features and individual head-related transfer functions (HRTFs). First, we derive an explicit likelihood function and validate it through parameter recovery on simulated data and fitting to behavioural responses from 33 participants, demonstrating that the framework reliably identifies individual sensorimotor and spectral parameters. Second, we use this framework to compare four HRTF template interpolation methods, showing that full-sphere spatial coverage and high-frequency spectral fidelity are the primary determinants of template quality, while the specific interpolation algorithm is secondary. Together, these results show that standard model-based statistical methods can address both fundamental questions in spatial hearing and applied problems such as perceptual HRTF evaluation. An open-source Python implementation is released alongside this work.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The paper derives an explicit likelihood for a Bayesian model of static sound localisation that jointly infers direction from noisy perceptual features and individual HRTFs. It validates the model via parameter recovery on simulated data and by fitting to behavioural responses from 33 participants, claiming that the framework reliably identifies individual sensorimotor and spectral parameters. The model is then used to compare four HRTF template interpolation methods on full-sphere data, concluding that spatial coverage and high-frequency spectral fidelity are the primary determinants of template quality while the choice of interpolation algorithm is secondary. An open-source Python implementation is released.

Significance. If the central validation holds, the work supplies a principled statistical alternative to heuristic performance metrics for auditory models and offers a practical tool for perceptual HRTF evaluation. The open-source code release is a clear strength that supports reproducibility and enables independent checks of the likelihood and fitting procedures.

major comments (2)
  1. [Abstract / validation procedure] Abstract and validation section: Parameter recovery is shown only on data generated from the exact model; this tests numerical identifiability under correct specification but does not address whether the derived likelihood matches the noise structure in human perceptual features. This is load-bearing for the claim that the framework 'reliably identifies' parameters from the 33-participant behavioural responses.
  2. [Results / real-data fitting] Results on real-data fitting: The manuscript reports in-sample fits to the 33 participants but provides no out-of-sample validation, cross-validation, or held-out trial analysis. Without such checks it remains possible that recovered parameters overfit idiosyncrasies of this dataset, which directly affects the subsequent comparison of HRTF interpolation methods.
minor comments (1)
  1. [Results / HRTF comparison] The abstract states that 'full-sphere spatial coverage and high-frequency spectral fidelity are the primary determinants'; the corresponding results section should explicitly quantify the relative contribution of these factors versus interpolation algorithm (e.g., via effect-size tables or ablation metrics) to make the claim precise.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their constructive comments, which help clarify the scope of our validation procedures. We respond to each major comment below.

read point-by-point responses

  1. Referee: [Abstract / validation procedure] Abstract and validation section: Parameter recovery is shown only on data generated from the exact model; this tests numerical identifiability under correct specification but does not address whether the derived likelihood matches the noise structure in human perceptual features. This is load-bearing for the claim that the framework 'reliably identifies' parameters from the 33-participant behavioural responses.

    Authors: The parameter recovery experiments on simulated data generated from the model itself are intended to verify that the likelihood function is correctly implemented and that the parameters are identifiable under the model's assumptions. The likelihood was explicitly derived from the Bayesian model of perceptual features with assumed noise structure. For the 33-participant data, the model fits provide evidence that the framework can be applied to human responses, as the recovered parameters lead to coherent HRTF comparisons. We will revise the abstract to more precisely state that the validation demonstrates identifiability and applicability rather than claiming it matches the exact human noise structure. Additionally, we will include a discussion of the model assumptions and potential mismatches with human perception. revision: partial

  2. Referee: [Results / real-data fitting] Results on real-data fitting: The manuscript reports in-sample fits to the 33 participants but provides no out-of-sample validation, cross-validation, or held-out trial analysis. Without such checks it remains possible that recovered parameters overfit idiosyncrasies of this dataset, which directly affects the subsequent comparison of HRTF interpolation methods.

    Authors: We agree that out-of-sample checks would provide additional assurance against overfitting. In the revised manuscript, we will perform a cross-validation analysis by holding out a portion of the trials for each participant, refitting the model on the remaining data, and evaluating the predictive accuracy on the held-out trials. This will be reported to confirm that the parameters generalize within the dataset. revision: yes

Circularity Check

0 steps flagged

No significant circularity; validation on simulated recovery and independent participant data is self-contained

full rationale

The paper derives an explicit likelihood for a Bayesian localisation model and validates it via parameter recovery on simulated data (standard recovery test with known ground truth) plus fitting to behavioural responses from 33 participants. It then applies the fitted framework to compare HRTF interpolation methods on external data. No load-bearing step reduces by construction to fitted inputs, self-citation chains, or ansatz smuggling; the central claims rest on external benchmarks (simulations and real responses) rather than re-expressing the model inputs. This matches the default expectation of non-circularity for statistical validation work.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

Abstract-only; the model rests on domain assumptions about noisy perceptual features and the structure of HRTFs. No free parameters or invented entities are explicitly quantified in the abstract.

free parameters (1)
  • individual sensorimotor and spectral parameters
    Model infers these per listener from behavioral data; exact count and fitting procedure not stated in abstract.
axioms (1)
  • domain assumption Perceptual features are noisy observations from which direction and HRTF can be jointly inferred
    Core modeling choice stated in the abstract description of the Bayesian framework.

pith-pipeline@v0.9.1-grok · 5702 in / 1294 out tokens · 32694 ms · 2026-06-25T22:26:10.862252+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

60 extracted references · 1 canonical work pages

  1. [1]

    Auditory processing models,

    T. Dau, “Auditory processing models,” in Handbook of signal processing in acoustics, pp. 175–196, Springer, 2008

    2008

  2. [2]

    Meddis, ed., Computational models of the auditory system

    R. Meddis, ed., Computational models of the auditory system. No. 35 in Springer handbook of auditory re- search, New York ; London: Springer, 2010. OCLC: ocn495596580

    2010

  3. [3]

    FrAMBI: A Software Framework for Auditory Modeling Based on Bayesian Inference,

    R. Barumerli and P. Majdak, “FrAMBI: A Software Framework for Auditory Modeling Based on Bayesian Inference,” Neuroinformatics, vol. 23, p. 20, February 2025

    2025

  4. [4]

    Cognitive com- putational neuroscience,

    N. Kriegeskorte and P. K. Douglas, “Cognitive com- putational neuroscience,” Nature Neuroscience , vol. 21, pp. 1148–1160, September 2018

    2018

  5. [5]

    The nature and dis- tribution of errors in sound localization by human listen- ers,

    S. Carlile, P. Leong, and S. Hyams, “The nature and dis- tribution of errors in sound localization by human listen- ers,” Hearing Research, vol. 114, pp. 179–196, December 1997

    1997

  6. [6]

    A framework for testing and compar- ing binaural models,

    M. Dietz, J.-H. Lestang, P. Majdak, R. M. Stern, T. Marquardt, S. D. Ewert, W. M. Hartmann, and D. F. M. Goodman, “A framework for testing and compar- ing binaural models,” Hearing Research, vol. 360, pp. 92– 106, March 2018

    2018

  7. [7]

    When a good fit can be bad,

    M. A. Pitt and I. J. Myung, “When a good fit can be bad,” Trends in Cognitive Sciences , vol. 6, pp. 421–425, October 2002

    2002

  8. [8]

    Tutorial on maximum likelihood esti- mation,

    I. J. Myung, “Tutorial on maximum likelihood esti- mation,” Journal of Mathematical Psychology , vol. 47, pp. 90–100, February 2003

    2003

  9. [9]

    Ten simple rules for the computational modeling of behavioral data,

    R. C. Wilson and A. G. Collins, “Ten simple rules for the computational modeling of behavioral data,” eLife, vol. 8, p. e49547, November 2019

    2019

  10. [10]

    Kass and Adrian E

    R. E. Kass and A. E. Raftery, “Bayes Factors,” Journal of the American Statistical Association , vol. 90, pp. 773–795, June 1995. eprint: https://doi.org/10.1080/01621459.1995.10476572

    work page doi:10.1080/01621459.1995.10476572 1995

  11. [11]

    Spectral weighting of monaural cues for auditory local- ization in sagittal planes,

    P. Llad´ o, P. Majdak, R. Barumerli, and R. Baumgartner, “Spectral weighting of monaural cues for auditory local- ization in sagittal planes,” Trends in Hearing , 2025

    2025

  12. [12]

    AMT 1.x: A toolbox for reproducible research in auditory model- ing,

    P. Majdak, C. Hollomey, and R. Baumgartner, “AMT 1.x: A toolbox for reproducible research in auditory model- ing,” Acta Acustica, vol. 6, p. 19, 2022

    2022

  13. [13]

    Modeling sound-source localization in sagittal planes for human lis - teners,

    R. Baumgartner, P. Majdak, and B. Laback, “Modeling sound-source localization in sagittal planes for human lis - teners,” The Journal of the Acoustical Society of America , vol. 136, no. 2, pp. 791–802, 2014

    2014

  14. [14]

    A Bayesian model for human directional localization of broadband static sound sources,

    R. Barumerli, P. Majdak, M. Geronazzo, D. Meijer, F. Avanzini, and R. Baumgartner, “A Bayesian model for human directional localization of broadband static sound sources,” Acta Acustica, vol. 7, p. 12, 2023

    2023

  15. [15]

    Classifying Non-Individual Head- Related Transfer Functions with A Computational Auditory Model: Calibration And Metrics,

    R. Daugintis, R. Barumerli, L. Picinali, and M. Geronazzo, “Classifying Non-Individual Head- Related Transfer Functions with A Computational Auditory Model: Calibration And Metrics,” in ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) , (Rhodes Island, Greece), pp. 1–5, IEEE, June 2023

    2023

  16. [16]

    Perceptual evaluation of an auditory model–based sim- ilarity metric for head-related transfer functions,

    R. Daugintis, M. Geronazzo, K. C. Poole, and L. Picinali, “Perceptual evaluation of an auditory model–based sim- ilarity metric for head-related transfer functions,” The Journal of the Acoustical Society of America , vol. 159, pp. 2822–2843, Mar. 2026

    2026

  17. [17]

    Listener Acoustic Personalization Challenge—LAP24: Head-Related Transfer Function Dataset Harmonization,

    R. Daugintis, R. Barumerli, M. Geronazzo, J. Pauwels, L. Picinali, and K. C. Poole, “Listener Acoustic Personalization Challenge—LAP24: Head-Related Transfer Function Dataset Harmonization,” IEEE Open Journal of Signal Processing , vol. 6, pp. 950–964, 2025

    2025

  18. [18]

    Toward 13 Barumerli et al.: Likelihood-based validation of a Bayesia n sound localisation model Improved Consistency Between Databases of Head- Related Transfer Functions,

    H. Bahu, J.-M. Jot, T. Carpentier, M. Noisternig, M. Mihocic, P. Majdak, and O. Warusfel, “Toward 13 Barumerli et al.: Likelihood-based validation of a Bayesia n sound localisation model Improved Consistency Between Databases of Head- Related Transfer Functions,” Journal of the Audio Engineering Society, vol. 74, pp. 35–49, February 2026

    2026

  19. [19]

    HRTF Upsampling With a Generative Adversarial Network Using a Gnomonic Equiangular Projection,

    A. O. T. Hogg, M. Jenkins, H. Liu, I. Squires, S. J. Cooper, and L. Picinali, “HRTF Upsampling With a Generative Adversarial Network Using a Gnomonic Equiangular Projection,” IEEE/ACM Transactions on Audio, Speech, and Language Processing , vol. 32, pp. 2085–2099, 2024

    2085

  20. [20]

    Listener Acoustic Personalization Challenge - LAP24: Head-Related Transfer Function Upsampling,

    A. O. T. Hogg, R. Barumerli, R. Daugintis, K. C. Poole, F. Brinkmann, L. Picinali, and M. Geronazzo, “Listener Acoustic Personalization Challenge - LAP24: Head-Related Transfer Function Upsampling,” IEEE Open Journal of Signal Processing , vol. 6, pp. 926–941, 2025

    2025

  21. [21]

    On the accu- racy of finite-difference time-domain simulations of head- related transfer functions as a function of model complex- ity,

    J. Meyer, S. Prepelit ¸˘ a, and L. Picinali, “On the accu- racy of finite-difference time-domain simulations of head- related transfer functions as a function of model complex- ity,” Applied Acoustics, vol. 228, p. 110353, January 2025

    2025

  22. [22]

    Predicting the effect of headphones on the time to lo- calize a target in an auditory-guided visual search task,

    P. Llad´ o, R. Barumerli, R. Baumgartner, and P. Majdak, “Predicting the effect of headphones on the time to lo- calize a target in an auditory-guided visual search task,” Frontiers in Virtual Reality , vol. 5, June 2024

    2024

  23. [23]

    W. J. Ma, K. P. Kording, and D. Goldreich, Bayesian Models of Perception and Action: An Introduction . MIT press, 2023

    2023

  24. [24]

    An ideal-observer model of human sound localization,

    J. Reijniers, D. Vanderelst, C. Jin, S. Carlile, and H. Peremans, “An ideal-observer model of human sound localization,” Biological Cybernetics , vol. 108, pp. 169– 181, April 2014

    2014

  25. [25]

    Lateralization of large interaural delays,

    J. E. Mossop and J. F. Culling, “Lateralization of large interaural delays,” The Journal of the Acoustical Society of America, vol. 104, pp. 1574–1579, Sept. 1998

    1998

  26. [26]

    Identification of perceptually relevant methods of inter-aural time differ- ence estimation,

    A. Andreopoulou and B. F. G. Katz, “Identification of perceptually relevant methods of inter-aural time differ- ence estimation,” The Journal of the Acoustical Society of America, vol. 142, pp. 588–598, Aug. 2017

    2017

  27. [27]

    Sound localization,

    J. C. Middlebrooks, “Sound localization,” in Handbook of Clinical Neurology, vol. 129, pp. 99–116, Elsevier, 2015

    2015

  28. [28]

    Localization cues of sound sources in the upper hemisphere,

    M. Morimoto and H. Aokata, “Localization cues of sound sources in the upper hemisphere,” Journal of the Acoustical Society of Japan (E) , vol. 5, no. 3, pp. 165– 173, 1984

    1984

  29. [29]

    Virtual localization improved by scaling nonindividualized external-ear transfer functio ns in frequency,

    J. C. Middlebrooks, “Virtual localization improved by scaling nonindividualized external-ear transfer functio ns in frequency,” The Journal of the Acoustical Society of America, vol. 106, pp. 1493–1510, August 1999

    1999

  30. [30]

    Accuracy-Precision Trade-off in Human Sound Localisation,

    R. Ege, A. J. V. Opstal, and M. M. Van Wanrooij, “Accuracy-Precision Trade-off in Human Sound Localisation,” Scientific Reports , vol. 8, p. 16399, Dec. 2018

    2018

  31. [31]

    K. V. Mardia and P. E. Jupp, Directional statistics. John Wiley & Sons, 2009

    2009

  32. [32]

    Comparison of Different Egocentric Pointing Methods for 3D Sound Localization Experiments,

    H. Bahu, T. Carpentier, M. Noisternig, and O. Warusfel, “Comparison of Different Egocentric Pointing Methods for 3D Sound Localization Experiments,” Acta Acustica united with Acustica , vol. 102, pp. 107–118, Jan. 2016

    2016

  33. [33]

    3-D local- ization of virtual sound sources: Effects of visual en- vironment, pointing method, and training,

    P. Majdak, M. J. Goupell, and B. Laback, “3-D local- ization of virtual sound sources: Effects of visual en- vironment, pointing method, and training,” Attention, Perception, & Psychophysics , vol. 72, pp. 454–469, February 2010

    2010

  34. [34]

    C. M. Bishop, Pattern recognition and machine learning , vol. 4. Springer, 2006

    2006

  35. [35]

    Discrimination of interau- ral differences of level as a function of frequency,

    W. A. Yost and R. H. Dye, “Discrimination of interau- ral differences of level as a function of frequency,” The Journal of the Acoustical Society of America , vol. 83, pp. 1846–1851, May 1988

    1988

  36. [36]

    Practical Bayesian Optimization for Model Fitting with Bayesian Adaptive Direct Search,

    L. Acerbi and W. J. Ma, “Practical Bayesian Optimization for Model Fitting with Bayesian Adaptive Direct Search,” in Advances in Neural Information Processing Systems , vol. 30, Curran Associates, Inc., 2017

    2017

  37. [37]

    Model selection and psychological the- ory: A discussion of the differences between the Akaike Information Criterion (AIC) and the Bayesian Information Criterion (BIC),

    S. I. Vrieze, “Model selection and psychological the- ory: A discussion of the differences between the Akaike Information Criterion (AIC) and the Bayesian Information Criterion (BIC),” Psychological Methods , vol. 17, pp. 228–243, June 2012

    2012

  38. [38]

    The SONICOM HRTF Dataset,

    I. Engel, R. Daugintis, T. Vicente, A. O. T. Hogg, J. Pauwels, A. J. Tournier, and L. Picinali, “The SONICOM HRTF Dataset,” Journal of the Audio Engineering Society, vol. 71, pp. 241–253, May 2023

    2023

  39. [39]

    Gr¨ af, Efficient Algorithms for the Computation of Optimal Quadrature Points on Riemannian Manifolds

    M. Gr¨ af, Efficient Algorithms for the Computation of Optimal Quadrature Points on Riemannian Manifolds . Doctoral Thesis, Technische Universit¨ at Chemnitz, Germany, Feb. 2013

    2013

  40. [40]

    Quadrature Rules on Manifolds

    M. Gr¨ af, “Quadrature Rules on Manifolds.” https://www-user.tu-chemnitz.de/∼ potts/workgroup/graef/quadrature/index.php.en Feb. 2026. (accessed 2026-02-10)

    2026

  41. [41]

    Rafaely, Fundamentals of Spherical Array Processing

    B. Rafaely, Fundamentals of Spherical Array Processing . Berlin, Heidelberg, Germany: Springer, 2 ed., 2019

    2019

  42. [42]

    Zotter and M

    F. Zotter and M. Frank, Ambisonics. A Practical 3D Audio Theory for Recording, Studio Production, Sound Reinforcement, and Virtual Reality , vol. 19 of Springer Topics in Signal Processing. Cham, Switzerland: Springer Open, 2019

    2019

  43. [43]

    HRTF magnitude modeling using a non-regularized least-squares fit of spherical harmonics coefficients on incomplete data,

    J. Ahrens, M. R. P. Thomas, and I. J. Tashev, “HRTF magnitude modeling using a non-regularized least-squares fit of spherical harmonics coefficients on incomplete data,” in APSIPA Annual Summit and Conference , (Hollywood, USA), Dec. 2012

    2012

  44. [44]

    Estimation of the Optimal Spherical Harmonics Order for the Interpolation of Head-Related Transfer Functions Sampled on Sparse Irregular Grids,

    D. Bau, J. M. Arend, and C. P¨ orschmann, “Estimation of the Optimal Spherical Harmonics Order for the Interpolation of Head-Related Transfer Functions Sampled on Sparse Irregular Grids,” Frontiers in Signal Processing, vol. 2, p. 884541, September 2022

    2022

  45. [45]

    Head-related transfer function interpolat ion in azimuth, elevation, and distance,

    H. Gamper, “Head-related transfer function interpolat ion in azimuth, elevation, and distance,” J. Acoust. Soc. Am. , vol. 134, pp. EL547–EL554, Dec. 2013

    2013

  46. [46]

    Binaural Reproduction Based on Bilateral Ambisonics,

    Z. Ben-Hur, D. Alon, O. Berebi, R. Mehra, and Rafaely, “Binaural Reproduction Based on Bilateral Ambisonics,” in Advances in Fundamental and Applied Research on Spatial Audio, IntechOpen, October 2021

    2021

  47. [47]

    Perceived Target Range Shapes Human Sound-Localization Behavior,

    R. Ege, A. J. Van Opstal, and M. M. Van Wanrooij, “Perceived Target Range Shapes Human Sound-Localization Behavior,” eneuro, vol. 6, pp. ENEURO.0111–18.2019, March 2019

    2019

  48. [48]

    Fitting Mixtures of Kent Distributions to Aid in Joint Set Identification,

    D. Peel, W. J. Whiten, and G. J. McLachlan, “Fitting Mixtures of Kent Distributions to Aid in Joint Set Identification,” Journal of the American Statistical Association, vol. 96, pp. 56–63, March 2001

    2001

  49. [49]

    Spectral equalization in binaural signals rep - resented by order-truncated spherical harmonics,

    Z. Ben-Hur, F. Brinkmann, J. Sheaffer, S. Weinzierl, and B. Rafaely, “Spectral equalization in binaural signals rep - resented by order-truncated spherical harmonics,” The Journal of the Acoustical Society of America , vol. 141, pp. 4087–4096, June 2017

    2017

  50. [50]

    From ear to hand: the role of the auditory-motor loop in pointing to an auditory source,

    E. O. Boyer, B. M. Babayan, F. Bevilacqua, M. Noisternig, O. Warusfel, A. Roby-Brami, S. Hanneton, and I. Viaud-Delmon, “From ear to hand: the role of the auditory-motor loop in pointing to an auditory source,” Frontiers in Computational Neuroscience , vol. 7, Apr. 2013. 14 Barumerli et al.: Likelihood-based validation of a Bayesia n sound localisation model

    2013

  51. [51]

    Acoustic and non-acoustic factors in modeling listener-specific per - formance of sagittal-plane sound localization,

    P. Majdak, R. Baumgartner, and B. Laback, “Acoustic and non-acoustic factors in modeling listener-specific per - formance of sagittal-plane sound localization,” Frontiers in Psychology , vol. 5, April 2014

    2014

  52. [52]

    Comparison of head- related transfer functions pre-processing techniques for spherical harmonics decomposition,

    F. Brinkmann and S. Weinzierl, “Comparison of head- related transfer functions pre-processing techniques for spherical harmonics decomposition,” in AES Int. Conf. on Audio for Virtual and Augmented Reality (A V AR) , (Redmond, USA), Aug. 2018

    2018

  53. [53]

    Magnitude-Corrected and Time-Aligned Interpolation of Head-Related Transfer Functions,

    J. M. Arend, C. P¨ orschmann, S. Weinzierl, and F. Brinkmann, “Magnitude-Corrected and Time-Aligned Interpolation of Head-Related Transfer Functions,” IEEE/ACM Transactions on Audio, Speech, and Language Processing, vol. 31, pp. 3783–3799, Nov. 2023

    2023

  54. [54]

    A Bayesian framework for simultaneously modeling neural and be- havioral data,

    B. M. Turner, B. U. Forstmann, E.-J. Wagenmakers, S. D. Brown, P. B. Sederberg, and M. Steyvers, “A Bayesian framework for simultaneously modeling neural and be- havioral data,” NeuroImage, vol. 72, pp. 193–206, May 2013

    2013

  55. [55]

    The role of spatial perception in au- ditory looming bias: neurobehavioral evidence from im- possible ears,

    T. Greif, R. Barumerli, K. Ignatiadis, B. T´ oth, and R. Baumgartner, “The role of spatial perception in au- ditory looming bias: neurobehavioral evidence from im- possible ears,” Frontiers in Neuroscience, vol. 19, August 2025

    2025

  56. [56]

    The frontier of simulation-based inference,

    K. Cranmer, J. Brehmer, and G. Louppe, “The frontier of simulation-based inference,” Proceedings of the National Academy of Sciences , vol. 117, pp. 30055–30062, Dec. 2020

    2020

  57. [57]

    Perceptually enhanced spectral distance metric for head-related transfer function quality prediction,

    D. Yao, J. Zhao, Y. Liang, Y. Wang, J. Gu, M. Jia, H. Lee, and J. Li, “Perceptually enhanced spectral distance metric for head-related transfer function quality prediction,” The Journal of the Acoustical Society of America , vol. 156, pp. 4133–4152, December 2024

    2024

  58. [58]

    Toward an Improved Auditory Model for Predicting Binaural Coloration,

    T. McKenzie and F. Brinkmann, “Toward an Improved Auditory Model for Predicting Binaural Coloration,” Journal of the Audio Engineering Society , vol. 73, pp. 115– 126, April 2025

    2025

  59. [59]

    A Survey on Machine Learning Techniques for Head-Related Transfer Function Individualization,

    D. Fantini, M. Geronazzo, F. Avanzini, and S. Ntalampiras, “A Survey on Machine Learning Techniques for Head-Related Transfer Function Individualization,” IEEE Open Journal of Signal Processing, vol. 6, pp. 30–56, 2025. Conference Name: IEEE Open Journal of Signal Processing

    2025

  60. [60]

    Strong and weak head-related trans- fer functions: The eHRTF analytical framework,

    M. Geronazzo, “Strong and weak head-related trans- fer functions: The eHRTF analytical framework,” JASA Express Letters, vol. 5, p. 087202, Aug. 2025. Supplementary material S1. T emplate interpolation distortion The three panels illustrate the spectral distortion in- troduced by different template construction approaches for the G.R.A.S. KEMAR dummy head ...

    2025