iiinova — where your digital self keeps evolving

Mads Lund Pedersen @madslupe.bsky.social · 06/08/2026

Methylphenidate had a selective effect. It raised sensitivity to costs, but left sensitivity to benefits unchanged. Controls weighed costs over benefits. ADHD participants off medication did the reverse. On MPH, the balance shifted back.

Violin plot of posterior distributions showing relative sensitivity to benefits versus costs in three groups: healthy controls, ADHD on methylphenidate, and ADHD on placebo. Values above zero mean more weight on benefits than costs; below zero means more weight on costs. Healthy controls and the methylphenidate group fall below zero. The placebo group falls above zero. The methylphenidate minus placebo contrast is −0.18, with a 95% credible interval from −0.22 to −0.13.

100

Mads Lund Pedersen @madslupe.bsky.social · 06/08/2026

Off medication, the ADHD group was less sensitive to both costs and benefits, and set narrower decision thresholds. Evidence built up more slowly, and responses came before enough of it had accumulated.

100

Mads Lund Pedersen @madslupe.bsky.social · 06/08/2026

Paper out in Psychological Medicine: does methylphenidate shift how adults with ADHD weigh costs against benefits? 61 adults with ADHD (tested on and off MPH, double-blind, placebo-controlled) and 63 controls did a cost-benefit choice task, analysed with the drift diffusion model.

142

Mads Lund Pedersen @madslupe.bsky.social · 26/06/2026

The contribution of RL vs WM to this signal shifts with load: WM dominates at low set sizes, RL at high. Adding trial-by-trial CPP slope to the model further improved fit to choice and RT. @iscience-cp.bsky.social

040

Mads Lund Pedersen @madslupe.bsky.social · 26/06/2026

Using the RLWM-LBA model (@actlab.bsky.social & @annecollins.bsky.social), we found that — as in perceptual decision-making — the CPP tracks the build-up toward a choice, here reflecting latent value estimates rather than sensory evidence.

161

Mads Lund Pedersen @madslupe.bsky.social · 26/06/2026

New paper in iScience on the neural dynamics of decision-making during learning. We analyzed behavioral and EEG data from 510 participants on the RLWM task (developed by @annecollins.bsky.social & Michael J. Frank). doi.org/10.1016/j.is...

doi.org

Redirecting

2147

Reposted by Mads Lund Pedersen

ᴅʀ. ᴍᴏᴡɪɴᴄᴋᴇʟ'ꜱ @drmowinckels.io · 30/11/2025

🎄 ANNOUNCEMENT: R Package Development Advent Calendar 2025! 🎄 Starting December 1st, I'm launching a 25-day journey through modern R package development. here's why you should follow along 🧵 #rstats #RPackageAdvent2025 #OpenSource

29733

Reposted by Mads Lund Pedersen

Erik R. Frogner @erikrf.bsky.social · 17/10/2025

Excited to use my first post to share our new paper and my first first-authorship! "Linking reinforcement learning, working memory, and choice dynamics to age and symptoms of anxiety and depression in adolescence" in Developmental Cognitive Neuroscience 🎉https://doi.org/10.1016/j.dcn.2025.101626

doi.org

Redirecting

1104