Lightnews — Scholar-powered news

Veli-Matti Karhulahti @mkarhulahti.bsky.social · 2h

the larger point being: an effect shouldn't be the discovery as such & hence failure to replicate shouldn't be hurtful -- if diverse effects collectively represented larger theoretical programs, single replications would have less power & be more natural ingredients in slow progress

Veli-Matti Karhulahti @mkarhulahti.bsky.social · 3h

As years go by I'm increasingly for the view that this is a very psych thing. The program of discovering "effects" as if they were chemical elements is super weak & almost always lacks interest in underlying explanation that'd require long-term theoretical curiously

2

Berna Devezer @devezer.bsky.social · 1d

I believe science is much more like art than many scientists are willing to consider. I have a feeling that it would free us to do better science if we embraced this view. I may be mistaken but that's where I am mentally.

2 6 27

Veli-Matti Karhulahti @mkarhulahti.bsky.social · 3d

Many journals that I didn't know of earlier, thanks for sharing! the platform's naming is unfortunate tho -- gold OA usually refers to APC'd journals so the name might push some authors away unintentionally

2 1

Julia M. Rohrer @dingdingpeng.the100.ci · 4d

I really strongly feel that some fields of research would profit if researchers stopped collecting online data for some time and instead maybe just read a bit outside of their field.

The ‘harm hypothesis’ strikes me as being deeply rooted in contemporary WEIRD values rather than being the result of a specific ‘evolved’ or ‘innate’ instinct or psychological mechanism. And indeed the literature cited to support it seems to suggest this.

Costello & Acerbi cite 5 papers in the paragraph above to support the model:

Stewart-Williams et al., 2024: the sample here consists of Prolific users mostly in the UK.

FeldmanHall et al., 2016: the samples were MTurk users in the US and volunteers in the UK.

Curry et al., 2004: the sample is convicted offenders in Texas in 1991.

Graso et al., 2023: US MTurk users again.

Graso & Reynolds, 2024: this is a review paper which does make some cross-cultural claims, but when you check the references you can see some important limitations. For example, they write that “Across cultures, women were perceived as less powerful than men but were seen more positively,” and when you check the reference it goes to Glick et al., 2004, which samples from 16 nations. However, when you read that paper they note in the methods that, “Most samples consisted primarily of college students participating for extra credit.”

4 8 46

Veli-Matti Karhulahti @mkarhulahti.bsky.social · 5d

Doesn't look bright on the other side either. Pretty sure we're reaching the end of an era with peer review as now is -- curious to see which major stakeholder comes out as first with the guts to try something else, risking their stock value

bsky.app/profile/mkar...

Veli-Matti Karhulahti @mkarhulahti.bsky.social · 20d

Made a new record this week re how many invitations had to be sent to obtain three good reviewers: 49 -- i don't use any automated senders so each person was individually selected by me, and won't doubt that all 46 refused to review for the good reason of being busy

1

Veli-Matti Karhulahti @mkarhulahti.bsky.social · 5d

of course there's nothing more claimable in stats vs other epistemic activities -- perhaps stats jargon should also incorporate more often "we argue" instead of claiming to confirm and falsify

1

Veli-Matti Karhulahti @mkarhulahti.bsky.social · 5d

epistemically, stats are easy to transform into "claims" whereas describing, organising, theorising & many other kinds of work are often left to make "arguments" -- a key rhetoric diff when it comes to press release and discovery

1 2

Veli-Matti Karhulahti @mkarhulahti.bsky.social · 5d

Major journals using statistics as a proxy for impact & quality these days nicely reflects academia's attention economy at the cost of being interested in explaining things

Joe Bak-Coleman @jbakcoleman.bsky.social · 5d

My hottest academic take is that we shouldn’t be using statistics in the vast majority of papers.

1 4

Joe Bak-Coleman @jbakcoleman.bsky.social · 5d

My hottest academic take is that we shouldn’t be using statistics in the vast majority of papers.

7 1 43

Ingo Rohlfing @ingorohlfing.bsky.social · 5d

How randomisation has changed the British Academy’s approach to research funding
blogs.lse.ac.uk/impactofsoci...
Another funding org exploring randomization of awards among proposal exceeding a certain threshold.
The pool of applicants and recipients was more diverse after the policy change 1/

1 3 4

Veli-Matti Karhulahti @mkarhulahti.bsky.social · 9d

Oh didn't know about that, based quick google seems to be this, looks nice!

doi.org/10.1080/1536...

Rethinking Traditional Methods of Survey Validation

It is commonly believed that self-report, survey-based instruments can be used to measure a wide range of psychological attributes, such as self-control, growth mindsets, and grit. Increasingly, su...

doi.org

1 1

Veli-Matti Karhulahti @mkarhulahti.bsky.social · 9d

in the end, it comes to the older philosophical question of the addiction field, ie when should behaviour for person Y be considered clinical/diagnosable-- outside clear cases, one's usually forced to rely on some cultural norms of deviancy, which is tough for scientists to accept & manage

2

Veli-Matti Karhulahti @mkarhulahti.bsky.social · 9d

some ppl have tried to overcome this by adding extra items asking whether X use is causing them significant harm (with ref to the DSM premise that disorders should cause distress/impairment) but that's ofc missing the point

1

Veli-Matti Karhulahti @mkarhulahti.bsky.social · 9d

in this field's case a key problem is that common items don't tap directly into health but rather ask about aspects of "use" -- they might form some constructs of intensive use but we still wouldn't know how that's related to health (if at all)

1 1

Veli-Matti Karhulahti @mkarhulahti.bsky.social · 9d

It's a lovely paper -- if I recall right, the point was in the optics: it might be able to measure something but assuming such something (here "addiction") is bad by default cannot be assessed by psychometrics alone

2 3

Veli-Matti Karhulahti @mkarhulahti.bsky.social · 12d

Whenever I see papers like these the lack of both hypotheses & tests are striking -- science happens in interesting ways when ppl who are genuinely interested do their best

1 1

Veli-Matti Karhulahti @mkarhulahti.bsky.social · 12d

unfortunately all from the same domain of tech use psych (but would indeed expect PCI to have many completed stage2s from other thematic psych areas too)

Veli-Matti Karhulahti @mkarhulahti.bsky.social · 12d

Here a few from my PCI account (recommend/review/author) -- should've least 1 of the listed tests:
doi.org/10.1080/1606...
doi.org/10.1525/coll...
doi.org/10.1098/rsos...
doi.org/10.1525/coll...
doi.org/10.1080/1606...
maybe this too (even tho the tests are a bit different)
doi.org/10.1098/rsos...

The Harmful Dysfunction Analysis applied to the concept of behavioral addiction: toward a new theoretical framework

The study’s principal aim was to explore the usefulness of the Harmful Dysfunction Analysis (HDA) in identifying individuals with pathological social media use (PSMU) and potentially other behavior...

doi.org

2

Veli-Matti Karhulahti @mkarhulahti.bsky.social · 13d

PCIRR should have ~100 of these, as all stage1s are public & almost all have open data too, the tests are common -- how many do you need?

1

Veli-Matti Karhulahti @mkarhulahti.bsky.social · 16d

--that said, Australia seems to have hard time with theirs

www.theguardian.com/media/2025/s...

Australia’s under 16s social media ban could extend to Reddit, Twitch, Roblox and even dating apps

Lego Play and Steam among the unexpected additions to the list that includes Facebook, Instagram, TikTok, YouTube and X

www.theguardian.com

Veli-Matti Karhulahti @mkarhulahti.bsky.social · 16d

Finally a job opening for an analytic philosopher

2

Veli-Matti Karhulahti @mkarhulahti.bsky.social · 16d

Very interesting if EU sets social media age limits. Relative to how impactful such decisions (both limiting & not) can be for human development and society, the scale of genuinely dedicated research feels rather small.

www.politico.eu/article/von-...

Von der Leyen calls for minimum age to access social media

Commission chief throws weight behind calls to ban social media use for kids.

www.politico.eu

1 1 2

Veli-Matti Karhulahti @mkarhulahti.bsky.social · 19d

It's amazing how it's still possible to make these processes so difficult -- noticed the same when trying to reject AI & unsubscribe from other sites they enforce re-login, while all other changes/engagement allowed directly (i think this explicitly violates GDPR)

Melanie Richter-Montpetit @mellimo.bsky.social · 20d

Academia.edu is changing its Terms and Conditions, and they're so outrageous (AI training; signing over ownership to them etc) I closed my account today.

A thread with some practical tips if you want to delete your account (5 steps):

2

Veli-Matti Karhulahti @mkarhulahti.bsky.social · 19d

Did the same -- didn't even remember I had an account there but apparently created one some 15 years ago & it had scraped most of my info + pubs somehow

1

Veli-Matti Karhulahti @mkarhulahti.bsky.social · 19d

In the past had this discussion over simulations & imo the answer is no -- yet would see it still compatible with Chang via cybernetics: we operate by feedback loops in an ever-changing world so both can be valid at the same time, representation & lack thereof (=what matters is the adaptive process)

1 1

Veli-Matti Karhulahti @mkarhulahti.bsky.social · 20d

I recall there was a website with a list but not sure if it's updated (or where it was). Can of course recommend PCIs, which are all diamond: peercommunityin.org

Peer Community In - free peer review & validation of preprints of articles

PCI is a non-profit open science organization of scientists to evaluate, recommend and publish research preprints in free open access

peercommunityin.org

1 1

Veli-Matti Karhulahti @mkarhulahti.bsky.social · 20d

I don't think anyone should ever feel bad for not taking a review, there's zero obligation to; alas, i think we'll see some change to these academic structures in the near future of 10-20 years (got some thoughts but too long for here)

2

Veli-Matti Karhulahti @mkarhulahti.bsky.social · 20d

as a disclaimer, had three very similar papers on the desk this fall which kinda exhausted the most immediate trusted ppl on the topic (n=6) but still, there's certainly an increasing trend of reviewer fatigue out there

Veli-Matti Karhulahti @mkarhulahti.bsky.social · 20d

Made a new record this week re how many invitations had to be sent to obtain three good reviewers: 49 -- i don't use any automated senders so each person was individually selected by me, and won't doubt that all 46 refused to review for the good reason of being busy

2 5