Properties (17)

Id 6af58ac2-36cb-4c9f-89be-ebd4cc994795

Year 2023

Abstract People with misophonia have strong aversive reactions to speciﬁc ‘‘trigger’’ sounds. Here we challenge this key idea of speciﬁcity. Machine learning was used to identify a misophonic proﬁle from a multivariate sound-response pattern. Misophonia could be classiﬁed from most sounds (traditional triggers and nontriggers) and, moreover, cross-classiﬁcation showed that the proﬁle was largely transferable across sounds (rather than idiosyncratic for each sound). By splitting our participants in other ways, we were able to show—using the same approach—a differential diagnostic proﬁle factoring in potential co-morbidities (autism, hyperacusis, ASMR). The broad autism phenotype was classiﬁed via aversions to repetitive sounds rather than the eating sounds most easily classiﬁed in misophonia. Within misophonia, the presence of hyperacusis and sound-induced pain had widespread effects across all sounds. Overall, we show that misophonia is characterized by a distinctive reaction to most sounds that ultimately becomes most noticeable for a sub-set of those sounds.

Cssclasses

purpleRed

Aliases

(Andermane et al., 2023)

Key Terms

Misophonia
Phenomenological cartography
Hyperacusis
ASMR
Visual analog scale
Scrambled sounds

Import Date 09-30-2025

Initial Source Misophonia MOC

Dv Authors Nora Andermane, Mathilde Bauer, Ediz Sohoglu, Julia Simner, Jamie Ward,

Dv DOI 10.1016/j.isci.2023.106299

Dv Status #status/reviewed

Dv Limitations • Black box nature of machine learning limits interpretability • Need for clinical validation of comorbid groups (autism, hyperacusis) • Limited generalizability across cultures and demographics • Use of only one psychoacoustic feature to explain responses

Dv Future Directions • Explore additional psychoacoustic features • Test cross-cultural generalizability • Investigate clinical groups with comorbid sound intolerances • Develop treatment strategies based on redirecting attention (e.g., visualizing sound source)

Dv Relevance • provides a interesting method of assessing misophonia through sound response profiles. the authors called it “phenomenological cartography” • shows evidence that Misophonics can be meaningfully identified by response profiles to a variety of sounds, not just trigger stimuli. • has a wealth of supplemental materials, including the stimuli used. could be a good source for materials in a future project

Checked For Zettle Notes true

Zotero Note Link Generated true

Initial Source

Misophonia

A phenomenological cartography of misophonia and other forms of sound intolerance

Status:: #status/Bare-bones-review

Metadata

Item Type: #journalArticle
Authors:: Nora Andermane, Mathilde Bauer, Ediz Sohoglu, Julia Simner, Jamie Ward,
Date Added: 1758770805000
URL: https://linkinghub.elsevier.com/retrieve/pii/S2589004223003760
DOI:: 10.1016/j.isci.2023.106299
Cite Key: andermanePhenomenologicalCartographyMisophonia2023
Tags:: #descriptive, #Diagnosis, #Misophonia, #Phenomenology,
PDF Attachments

Abstract


Background/Purpose	Background- • Misophonia is typically understood as a highly specific aversion to certain trigger sounds, with some neuroimaging evidence supporting specificity. • Contradictory evidence suggests misophonics rate a broad range of sounds as more aversive, even those not traditionally described as triggers. • There is ongoing debate about whether misophonia reflects specific sound aversions or generalized sound intolerance. • A need exists for differential diagnosis among various forms of sound intolerance (e.g., misophonia, hyperacusis, phonophobia, autism-related sensitivities, ASMR). Purpose- • To determine whether misophonia is characterized by specific responses to specific sounds or a generalized aversive profile across many sounds. • To examine whether these sound-response profiles are distinct from those associated with hyperacusis, autism spectrum traits, or ASMR. • To apply a novel machine learning-based approach (phenomenological cartography) to assess misophonia diagnosis and its relationship to other sound intolerance profiles.
Participants/samples	• N = 418 participants (222 Misophonics, 196 controls) • Misophonics: Mean age = 41.56, SD = 14.32; 175F/42M/5NB • Controls: Mean age = 33.08, SD = 14.29; 140F/55M/1NB • Misophonia group identified using Sussex Misophonia Scale (SMS); threshold > 50.5. • Separate ASMR sample: N = 254; Mean age = 23.2; 202F/45M/7NB • Inclusion/Exclusion:- Based on SMS and ASMR-15 scores; additional self-report on traits like hyperacusis and autism.
Measures	• 17 subjective response descriptors rated per sound: e.g., rage, anxiety, disgust, soothing, pleasurable, body tension, etc. • Sounds rated using 0–100 visual analog scale. • Sussex Misophonia Scale (SMS) • Autism Spectrum Quotient (AQ), Sensory Sensitivity (GSQ), Interoceptive Awareness (MAIA) • ASMR-15 scale for assessing ASMR responsiveness stimuli- • Human oral/nasal sounds (8): chewing apple, eating crisps, chewing gum, lip smacking, coughing, snoring, breathing, sniffling. • Other human-made sounds (8): basketball dribbling, chopping vegetables, pen clicking, finger tapping, nails on chalkboard, rusty swing, typing, walking in heels. • Non-human sounds (8): bird song, clock ticking, crow cawing, dog drinking, frog croaking, printer noise, rainfall, wind chimes. • Scrambled sounds (8): temporally scrambled versions of apple eating, breathing, coughing, sniffling, basketball, chopping vegetables, finger tapping, typing. Scrambling preserved spectral content but disrupted recognizability. • ASMR Triggers (8 sounds): whispering, popping, rustling, tapping, and similar stimuli identified via ASMR forums. Tested separately in the ASMR group.
Procedures	• Participants seated with headphones in a controlled environment. • Each sound clip lasted 15 seconds; participants rated it immediately after listening. • Ratings collected on a 0–100 visual analog scale across 17 descriptors (rage, anxiety, disgust, soothing, pleasurable, body tension, etc.). • The order of sounds was randomized for each participant to control for sequence effects. • Loudness of some clips was adjusted based on pilot testing to ensure equal perceived volume across stimuli. • Scrambled sounds were included to act as acoustic controls, testing reactions independent of recognizability. • In ASMR-specific testing, participants were also exposed to 8 ASMR triggers alongside the 32 main sounds. • Machine learning classifiers (random forest-based) were trained separately per sound and tested using cross-validation. • Cross-classification was applied: a classifier trained on responses to one sound (e.g., chewing apple) was tested on responses to another (e.g., chewing gum).
Results	• 30 of 32 sounds classified Misophonics vs. controls above chance (AUC > 0.5) • Overall classifier accuracy AUC = 0.925 (specificity = 0.865, sensitivity = 0.802) • Most predictive features: body tension, annoyance, anxiety, rage • Misophonics show broad sound intolerance, not just to traditional triggers • Cross-classification (i.e., using response to one sound to classify others) showed generalized response patterns • Scrambled sounds elicited strong responses, supporting general sensitivity hypothesis • Psychoacoustic features (2500–5500 Hz and 1–16 Hz modulations) only weakly associated with aversive responses, especially for natural (recognizable) sounds • Severity of misophonia correlated with number of sounds triggering classification, suggesting less specificity with more severe symptoms
Discussion/Limits	Conclusion- • Misophonia reflects a generalized pattern of sound intolerance, with greater severity linked to broader intolerance. • Phenomenological cartography enables effective classification and insight into subjective response profiles. • Misophonia is distinct from other sound sensitivity profiles like hyperacusis, autism, and ASMR. • ASMR and misophonia do not co-occur meaningfully; profiles are distinct and sometimes oppositional. • Visual imagery responses were reduced in misophonics, possibly indicating attentional focus on internal states.
Limitations::	• Black box nature of machine learning limits interpretability • Need for clinical validation of comorbid groups (autism, hyperacusis) • Limited generalizability across cultures and demographics • Use of only one psychoacoustic feature to explain responses
Future Directions::	• Explore additional psychoacoustic features • Test cross-cultural generalizability • Investigate clinical groups with comorbid sound intolerances • Develop treatment strategies based on redirecting attention (e.g., visualizing sound source)
Relevance::	• provides a interesting method of assessing misophonia through sound response profiles. the authors called it “phenomenological cartography” • shows evidence that Misophonics can be meaningfully identified by response profiles to a variety of sounds, not just trigger stimuli. • has a wealth of supplemental materials, including the stimuli used. could be a good source for materials in a future project

Notes

key terms and concepts

Misophonia - a condition characterized by intense aversive reactions to specific trigger sounds, often oral/nasal human sounds (e.g., chewing, breathing).
Phenomenological cartography - a method used to map subjective first-person responses to sounds across multiple descriptors using machine learning to detect patterns.
Hyperacusis - a condition where normal environmental sounds are perceived as excessively loud or painful.
ASMR (Autonomous Sensory Meridian Response) - a condition involving pleasurable tingling sensations triggered by certain auditory or visual stimuli.
Scrambled sounds - acoustically similar but temporally altered versions of real sounds, used to test recognition-independent responses.
Visual analog scale- a rating method where participants quantify their responses on a continuum (0–100).

my engagement notes

my questions

my comments

the authors used “random forest” to determine the features for the classifier. I have heard of the term on some stats YouTube vids but do not fully understand the concept., I should prolyl look into it? (page 4)

Bookmarks

bookmark for breaking down the phenomenological cartography of misophonia (page 3)
bookmark for Comparisons between sounds: Cross-classification (page 5)
bookmark for speculation on why some sounds are rated highly (page 6)
bookmark to the supplemental materials table (page 17)
bookmark for in depth description of the sourcing of the stimuli (page 18)
bookmark talking about the Psychoacoustic properties of the sounds (page 20)

reading notes

Severity Insights:

Higher SMS scores correlate with broader range of aversive sound responses
Machine learning performance tracks symptom severity, supporting validity of tool

Diagnostic Differentiation:

Autism-related profiles emphasize repetitive non-oral sounds
Hyperacusis linked to loudness/pain responses rather than anger/disgust
ASMR profiles reflect increased soothing/pleasurable responses, especially to scrambled and whispering sounds

Visual Experience Ratings:

Lower in misophonics, especially for unrecognizable sounds, suggesting reduced visual imagery

Implications:

Misophonia may arise from general sensory hypersensitivity, shaped by developmental/environmental factors
Individualized treatment may be required based on unique sound-response profile

image-undefined-x59-y220.png basic tldr graphic for the study as a whole page

image-4-x44-y293.png Figure 1. A heatmap depicting misophonic > control effect sizes (Cohen’s d) for 32 sounds and 17 response features page 4

image-5-x35-y316.png Figure 2. A heatmap depicting cross-classification accuracy (AUC) when a classifer trained on responses to one sound is used to predict group membership from a pattern of responses to another sound (note that the ondiagonal values are not cross-classifications) page 5

image-6-x36-y417.png Figure 3. Top: Spearman’s correlations between the psychoacoustic region of interest identified by Kumar et al.2 and phenomenological ratings to the same sounds by misophonics and controls considering scrambled and real sounds (left and right) page 6

image-8-x36-y148.png Figure 5. The ability to classify misophonics based on their response pattern to 32 different sounds (x axis) correlates with the number of misophonic triggers reported (top figure) and the severity of the misophonia (bottom) page 8700

image-9-x37-y59.png Figure 6. Heatmaps (showing Cohen’s d effect sizes) for 32 sounds (y axis) and 17 responses (x axis) for within misophonia comparisons, contrasting presence versus absence of hyperacusis (top) and high versus low scores on the SMS pain factor (bottom) page 9

image-11-x47-y238.png Heatmaps (showing Cohen’s d effect sizes) for 32 sounds (y axis) and 17 responses (x axis) for comparisons within non-misophonics, contrasting high versus low AQ scorers page 11

image-12-x32-y145.png Figure 8. Heatmaps (showing Cohen’s d effect sizes) for (top) 32 sounds (y axis) and 17 responses (x axis) and (bottom) the eight ASMR triggers, contrasting ASMR responders and non-responders (excluding misophonics) page 12

zettlekasten

2023 idea of a phenomenological cartography of Misophonia
2023 study shows that Misophonics can be identified by sound response profiles

2023 paper supports diminished role of psychoacoustic features in Misophonia