Fleur Zeldenrust
banner
fleurzeldenrust.bsky.social
Fleur Zeldenrust
@fleurzeldenrust.bsky.social
Computational Neuroscientist • Associate prof @Donders Institute/Radboud University • Member De Jonge Akademie • founder Dutch Brain Olympiad and BrainHelpDesk • website: fleurzeldenrust.nl • ORCID: 0000-0002-9084-9520
No, not explicitly, the book treats science and when and why we can trust its claims much more generally, and puts it in a historical context. But it does say that observations often fit multiple theories, and if something doesn’t fit it is often difficult to say why/what part of which theory.
January 14, 2026 at 10:04 AM
Moreover: a big dataset (or big dataset +AI) won’t magically give you a good scientific theory, actual thought work, reasoning, is needed for that.
January 14, 2026 at 8:16 AM
Was reading this: press.princeton.edu/books/hardco... Relevant for here: if there are always multiple theories that could fit with an experimental outcome and vice versa, a lot of back and forth is needed. In that sense it is silly to insist on either one order.
Why Trust Science?
Why the social character of scientific knowledge makes it trustworthy
press.princeton.edu
January 14, 2026 at 8:11 AM
Recognizable 😀
January 12, 2026 at 3:21 PM
Isn’t it both? Model based on some data/observations. Then do predictions based on this model, then use other data to test these?
January 12, 2026 at 1:02 PM
Showing something is NOT the case is much harder than showing it is, but Nishant painstakingly went through every possible analysis loophole to show this. Well done Nishant!

For everyone interested in #clustering, #heterogeneity, #classification and #modelling!
December 23, 2025 at 9:38 AM
En ook altijd zo leuk: de correcties die dan gevraagd worden: kun je even overal waar 'figure' staat het veranderen in 'Fig.'? Je betaalt ze duizenden euro's, en zelfs dat kunnen ze niet zelf? Ik krijg altijd zin om ze uit te leggen hoe 'Find and replace' werkt.
December 19, 2025 at 9:14 AM
En dan daarna weer weken niets van zich laten horen...
December 19, 2025 at 9:05 AM
Met:
🔹@fleurzeldenrust.bsky.social (@dondersinst.bsky.social, Radboud Universteit, initiatiefnemer manifest Onder Controle, lid van De Jonge Akademie)
🔹 @felienne.bsky.social (hoogleraar Computer Science Education, @vuamsterdam.bsky.social)
🔹 Titus Stahl (uhd; privacy & publieke sfeer, @rug.nl)
December 4, 2025 at 10:57 AM
We bespreken:
• afhankelijkheid van platforms
• informatiekwaliteit
• hoe tracking- en datasystemen ons gedrag sturen
• wat universiteiten en individuele wetenschappers zélf kunnen doen
• open source, decentralisatie en publieke alternatieven
December 4, 2025 at 10:57 AM
In aflevering 3 van Toekomstmakers verkennen we hoe deze afhankelijkheid is ontstaan, waarom informatie online steeds meer verschraalt (enshittification), en welke keuzes instellingen én individuele wetenschappers kunnen maken om hun digitale soevereiniteit te versterken.
December 4, 2025 at 10:57 AM
Dat beïnvloedt waar we onderzoek naar doen, hoe we lesgeven en waar we ons onderwijs op focussen. Zo verschuift invloed en controle van universiteiten naar bedrijven. Wat betekent dit voor #academischevrijheid, #informatiekwaliteit en de publieke sfeer waarin #wetenschap hoort te functioneren?
LinkedIn
This link will take you to a page that’s not on LinkedIn
lnkd.in
December 4, 2025 at 10:57 AM
As if there is a single universal scale for PhD students....they're just like humans you know, and each of them has strong and weaker points. I hate that 'top 5%' question. They could all be based in the top 5% based on a different skill or quality.
December 2, 2025 at 6:58 AM
Very cool! Made me think of this work: www.mdpi.com/2313-7673/9/... it helps if you can put the delays at the right values for the task requirements.
www.mdpi.com
November 14, 2025 at 12:49 PM