Eiko Fried
@eikofried
Professor of Mental Health & Data Science at Leiden University. Studying mental health problems as systems (). Building an early warning system for depression ().
Reality has entirely caught up with satire. Real headlines are no longer distinguishable from The Onion.
Great work by @robert-m-ross.bsky.social: Do People Sincerely Believe Conspiracy Theories That They Endorse? Raccoon pink/yellow (highly bizarre army conspiracy) is a yes/no sincerity check; green/orange is self-report if they answered sincerely. online.ucpress.edu/collabra/art...
... all the time (e.g. someone just asked me to send an email of which prior papers contain work on dataset X, which takes me now 20 seconds by just clicking a few times CITE and pasting).
Join our Many-Analysts Simulation Study—yes, you read that correctly—accepted as Stage 1 Registered Report in (AMPPS) @psychscience.bsky.social. We're trying to find out how researchers' simulation design choices shape methodological conclusions. Further info: osf.io/8vcxh/files/...
Hi Sanjay. /1 Indeed, Ashley also discusses this in the paper, and their (thoughtful) sim deals with this differently than the one I hacked together quickly to just demonstrate the point that you can get a moderate correlation from a mixture of 0s in smaller groups.
2/2 I was interesting for folks working with our very large EMA dataset, & Leonie in careless responding, so we put together a large team of experts in different careless responding methods, and they all applied their methods to the data. Main lesson: methods do not agree with each other.
1/2 New #openaccess Psych Methods paper: "Identifying Careless Responding in Ecological Momentary Assessment: Inconsistent Signals From Different Detection Methods in the WARN-D Data" led by @leonievogelsmeier.bsky.social. psycnet.apa.org/fulltext/202...
Our work shows that the correlations go drown drastically in higher severity samples; less than half than in healthier (not healthy) samples. Here another examples from another paper published afterwards; should add it to the blog post.
Example from the simulation (you can adjust these sliders) showing that even in a world where psychopathology domains are uncorrelated in both high symptom and low symptom groups, externalizing and internalizing domains would be correlated around 0.6.
New Guardian article links to “landmark study” on babies & screentime. URL goes to advocacy foundation that financed study. On their website I find a self hosted 110page “report” but no paper. Would love to understand if this is peer-reviewed, published research or advocacy piece.
The universe in which we all have to use outlook has to be one of the worst possible ones. Seriously, both searches and auto completion of addresses were better in the early 2000s in basically any software than 2026 Outlook.
I haven't uploaded stuff to youtube for years but apparently everything is stupid now I clicked the "suggest thumbnail" button thinking it would pick a video frame rather than doing ... whatever THIS is. Holy AI hell.
Every single humans rights org — including Israeli ones — have been telling the same story. Given that Israel still bars journalists entry to Gaza, which in itself is bonkers, how about we trust any or all human rights orgs. www.theguardian.com/world/2026/j...
i'm perpetually trying not to freak out ... hope this helps others, too:
we exclude people 'with depression' (too long discussion) at baseline because we want to forecast onset. so indeed too healthy perhaps. here a summary of the items we assess during EMA cpe.psychopen.eu/index.php/cp...
Been reading up on depression onset / relapse forecasting and found a lot of papers in tier 1 psychiatry journals with similar distributions of p-values. Doesn't speak to any individual paper of course—sometimes lucky, perhaps—but collectively, there are problems in this literature.
Just came across the MHI-5, a shortform of the SF-36. It seems ... really narrow in coverage other than item 1? Does anyone have experience with the scale?
Yes yes you've had a tough year, but please spare a thought for the PHQ-9 ... www.sciencedirect.com/science/arti... jamanetwork.com/journals/jam...
Not into soccer but after just having finished our depression forecasting competition, I find this an interesting read. England is their favorite.