Show newer

Ensuite on attaque le gros morceau ! La hard-core cherche à comprendre comment les humains perçoivent la hauteur, la puissance et le timbre d’un son. mycore.core-cloud.net/index.ph

Show thread

Puis une digression sur la physiologie du système auditif humain, depuis la périphérie jusqu'au cortex, et sa modélisation. mycore.core-cloud.net/index.ph

Show thread

D'abord les bases de la psychophysique : comment approcher scientifiquement la question de la perception humaine et, plus pragmatiquement, comment construire une bonne expérience de psychoac qui donne une mesure valide ? mycore.core-cloud.net/index.ph

Show thread

Comme avant chaque rentrée j'ai mis à jour mon cours de et pour @lemansuniv... et surtout les illustrations ! Tout est disponible sur mon site dbao.leo-varnet.fr/enseignemen ! Petit aperçu des chapitres ⬇️
@psycholinguistics @cognition @cogsci

@enroweb on peut également programmer des alertes par mot-clef directement dans (juste pour éviter la dépendance à Google)

Avis à la communauté @psycholinguistics : les 21-22 novembre auront lieu à l' les Journées Perception Sonore (co-organisées par la Société Française d'Acoustique ). Inscriptions gratuites pour les étudiant.e.s en master et doctorant.e.s. sites.google.com/view/jps-2023

In this conference paper for we replicate Ahumada's seminal experiment from 1975 on tone-in-noise perception using our own fastACI toolbox (github.com/aosses-tue/fastACI), and we analyze the data obtained by an artificial listener from the (also ) Auditory Modeling Toolbox on the task. It's very satisfactory to have a network of interconnected toolboxes working together... And of course all analyses can be reproduced easily using the commands listed in the article. You can even run the experiment on yourself to replicate Ahumada's original results!
hal.science/hal-04186363v1/doc

New blogpost: what to do about Discussion sections of papers, which are often an excuse for overhyped and p-hacked conclusions.
deevybee.blogspot.com/2023/08/

Bon on sent bien que Vertigo n'avait pas encore été réalisé à l'époque où a rédigé ce texte...

Show thread

Une réflexion intéressante de dans "Sens et non-sens" : la s'apparente à la psychanalyse en ce qu'elle décrit (le plus souvent) la subjectivité des personnage "de l'intérieur". Au contraire, le se rapproche plutôt de la démarche . "Si le cinéma veut nous montrer un personnage qui a le vertige, il ne devra pas essayer de rendre le paysage intérieur du vertige, comme Daquin dans Premier de Cordée et Malraux dans Sierra de Terruel ont voulu le faire. Nous sentirons beaucoup mieux le vertige en le voyant de l’extérieur, en contemplant ce corps déséquilibré qui se tord sur un rocher, ou cette marche vacillante qui tente de s’adapter à on ne sait quel bouleversement de l’espace. Pour le cinéma comme pour la psychologie moderne, le vertige, le plaisir, la douleur, l’amour, la haine sont des conduites"

"The notes themselves—the sound waves, that is—are heard alike by musician and nonmusician and are universally acknowledged to be real in the purely physical sense. But what about the melody itself? Is it something real or does it exist only in the mind or imagination […] The musician does not need to have "faith" that there is a melody, nor does he have to accept the existence of the melody on some scriptural authority; he obviously has a direct experience of the melody itself. And once the melody is heard, it is impossible ever again to doubt it."
(Richard , The Is Silent)

📢I am looking for a postdoc to work on the cognitive neuroscience of self-beliefs at
@InstitutCerveau
🤩🧠✨ - details here: marionrouault.weebly.com/uploa informal inquiries are welcome & sharing the info is appreciated 🙏

The @DataColada team has been sued by Francesca Gino for exposing the fraudulent data underlying four of her papers (she claims defamation). The legal defense could get expensive. Please join us in providing financial support for their defense.

gofundme.com/f/uhbka-support-d

When someone who grew up with the Pipes screensaver starts designing real world infrastructure

@alfora @futurebird @MostlyHarmless My favourite Excel “feature” is that biologists ended up having to rename some genes because Excel “autocorrects” them, leading to detectable errors in over 30% of a sample of published journal papers:

“What is not funny though is having genomics spreadsheets suffer from auto-conversion of gene names like SEPT8, DEC1 and MARCH3 into dates, a problem first characterised in 2004. A 2016 article on this topic led the Human Gene Name Consortium to change many of these gene names to be less susceptible to autocorrect. Despite this, our work here shows that gene name autocorrect errors continue to accumulate in supplementary genomics spreadsheet files at a rapid pace.” — journals.plos.org/ploscompbiol

How big problem it is that cross-validation (CV) is biased? I briefly discuss some points on this.

0. Unbiasedness has a special role in statistics, and too often there are dichotomous comments that something is not valid or is inferior because it's not unbiased. However, often the non-zero bias is negligible, and by modifying the estimator we may even increase bias but reduce the variance a lot, providing an overall improved performance.

Pólya’s conjecture is the usual example of a statement that’s true for many cases, but breaks down for some large value of n. In that example, n=906150257. However, I think I have a new favourite: gcd(n^17 + 9, (n+1)^17 + 9) = 1 for all n < 8424432925592889329288197322308900672459420460792433 (More counterexamples at oeis.org/A010034)

Show older
Qoto Mastodon

QOTO: Question Others to Teach Ourselves
An inclusive, Academic Freedom, instance
All cultures welcome.
Hate speech and harassment strictly forbidden.