iiinova — where your digital self keeps evolving

Jan R. Wessel @janw.bsky.social · 06/10/2026

Has that ever stopped any of these rocket surgeons?

110

Jan R. Wessel @janw.bsky.social · 02/09/2026

Couldn't be prouder of Darcy or happier to have her return to the Midwest! Great news for UWM.

191

Reposted by Jan R. Wessel

SfN Journals @sfnjournals.bsky.social · 12/08/2026

#JNeurosci: Findings from Leach et al. provide evidence that the frontoparietal cortex influences behaviors, dynamically adjusting communication between brain networks as decisions progress. doi.org/10.1523/JNEUROSCI.0594-26.2…

Diagram of cognitive task with color cues, probe, and feedback. Buttons labeled Female, Male, City, Landscape.

0169

Reposted by Jan R. Wessel

Society for Psychophysiological Research (SPR) @therealspr.bsky.social · 15/08/2026

🌟SPR Science Spotlight Saturday🌟 Learn how Bruce Bartholow studies brain mechanisms that predict risk of alcohol use disorder. Be sure to register for the 2026 annual meeting in Halifax! spr.joynmeeting.com/conference/

021

Jan R. Wessel @janw.bsky.social · 08/08/2026

This shows 3 things: - Trigger failures are a real, distinct category of response inhibition failure - They show distcint behavioral & EEG patterns predicted by existing TF theories - They reflect a genuine failure to initiate cognitive control. I hope this approach can be useful for years to come!

110

Jan R. Wessel @janw.bsky.social · 08/08/2026

We also tested the long-standing hypothesis that stop-signal TF result from attentional failure, but find little evidence to support this claim. Prestimulus alpha power and stimulus-related visual N1 ERPs were identical between TF and successful stop trials. If anything, non-TF trials were abnormal.

100

Jan R. Wessel @janw.bsky.social · 08/08/2026

Then, we dug deeper into EEG signatures. In line with the single-trial correlation analysis, the binary split shows that stop-signal trigger failures do not have a stop-signal P3! To us, that is the strongest evidence that there are indeed trials in which no inhibition ever takes place.

160

Jan R. Wessel @janw.bsky.social · 08/08/2026

So we then split the failed stop-trials into "likely TF" and "likely non-TF" trials. We first compared behavior. Strikingly, unlike non-TF trials, TF trials are not faster than typical go-trials. That makes sense, since go-speed shouldn't matter if you don't ever initiate inhibition!

100

Jan R. Wessel @janw.bsky.social · 08/08/2026

Then, we used Gaussian mixture modeling to test whether p(TF) is unimodally distributed, or results from two generative processes. If trigger failures and 'speed failures' are distinct, binary outcomes, then one would expect a highly bimodal mixture distribution. That's exactly the case.

110

Jan R. Wessel @janw.bsky.social · 08/08/2026

So then we went and checked whether these single-trial probabilities actually reflected something that's "real". And indeed, trial-wise p(TF) was highly correlated with spatiotemporal clusters of trial-wise EEG after the stop-signal. Fronto-central activity was particularly anticorrelated with p(TF)

110

Jan R. Wessel @janw.bsky.social · 08/08/2026

But Stein didn't stop there. He also derived an analytical solution to single-trial TF probability using the BEESTS model parameters and likelihood ratio testing. Despite being fundamentally different approaches (numerical v analytic), jackknifing and LR return highly correlated parameter estimates.

100

Jan R. Wessel @janw.bsky.social · 08/08/2026

So I suggested Stein try jackknifing: run the model once with each failed stop-trial held out. If mean TF rate goes down, the trial is likely a trigger failure (and vice versa). Stein first developed are more efficient implementation of BEESTS to enable this computationally intensive approach.

110

Jan R. Wessel @janw.bsky.social · 08/08/2026

However, the big limitation of the BEESTS model is that it cannot identify which individual trials are trigger failures. Therefore, TF are essentially hypothetical entities. To test whether they are real, qualitatively distinct trials from other inhibitory failures, we need a trial-wise method.

100

Jan R. Wessel @janw.bsky.social · 08/08/2026

In the aughts, Dora Matzke (and later Andrew Heathcote) developed a computational model ("BEESTS") that can capture the mean rate of these trigger failure trials in behavior. This model has been superbly productive in studies that correlate per-subject TF rates with brain activity and other factors.

100

Jan R. Wessel @janw.bsky.social · 08/08/2026

However, people have long speculated that there may be a 2nd type of inhibitory failure: trigger failure trials, on which no inhibition process is ever initiated. This can be inferred from inhibition functions with >0 intercepts - i.e., inhibition fails even if a stop-signal is presented right away.

100

Jan R. Wessel @janw.bsky.social · 08/08/2026

For four decades, the dominant behavioral-computational model in the field of response inhibition has been the horse-race model. This model holds that response inhibition fails for only one reason: because the inhibition process was too slow to intercept the response.

100

Jan R. Wessel @janw.bsky.social · 08/08/2026

New preprint from the lab. We think we have a solution (two, actually) to a long-standing problem in the field of response inhibition. Using two different mathematical approaches, we think we can identify long-hypothesized inhibitory 'trigger failure' trials. More in🧵 www.biorxiv.org/content/10.6...

biorxiv.org

Two types of response inhibition failures in humans

Response inhibition is a key control function that allows humans to perform safe, goal-directed behaviors. The dominant behavioral-computational model holds that response inhibition fails when the inh...

13718

Jan R. Wessel @janw.bsky.social · 31/07/2026

Come be our colleague. DM me in case of questions!

021

Reposted by Jan R. Wessel

Caterina Gratton @caterinagratton.bsky.social · 23/07/2026

A special highlight of our work on precision fMRI of the prefrontal cortex

0348

Jan R. Wessel @janw.bsky.social · 22/07/2026

We're excited to welcome three new post-docs to the lab this month: Dr. Chenye Bao (University of Missouri) Dr. Gizem Filiz (Washington University in St. Louis) Dr. Veronica Zarr (University of Utah) Exciting times ahead!

2221

Reposted by Jan R. Wessel

Irene Echeverria-Altuna @irenetxeberria.bsky.social · 16/07/2026

Interested in timing, motor control, attention, working memory and/or stuttering? ⌛️ Curious to see how we try to bring these together in one study? 🧐 Check out our most recent, hot-off-the-press manuscript, born out of fun cross-field and cross-lab discussions ✨ Very excited to see this out ☀️🍾

1124

Reposted by Jan R. Wessel

Atsushi Kikumoto @kikumotoatsushi.bsky.social · 09/07/2026

We're hiring a full-time Lab Manager for the Cognitive Flexibility Dynamics Lab at the University of Maryland, studying cognitive control via brain recording, stimulation, and computational modeling. Ideal for research experience before grad school: cfd-lab.com/join

0117

Reposted by Jan R. Wessel

Christian Schnell @neuroschnell.bsky.social · 05/07/2026

On my way to Barcelona to attend the #Fens forum. Let me know if you are there and would like to meet. Very happy to chat about @plosbiology.org and learn about your research.

053

Jan R. Wessel @janw.bsky.social · 03/07/2026

The final version of this is now out at Cortex: www.sciencedirect.com/science/arti... Thanks much to @russpoldrack.org and two other reviewers for highly constructive feedback on the earlier version, and especially to @annemscheel.bsky.social, who suggested Cortex and edited the paper.

sciencedirect.com

1225

Jan R. Wessel @janw.bsky.social · 26/06/2026

Merely reading the words "the damn wife" will have me muttering Dennehy lyrics all weekend (again). Thanks in advance from my damn wife.

010

Reposted by Jan R. Wessel

Debbie Yee @debyee.bsky.social · 23/06/2026

Thrilled to be featured in the @thetransmitter.bsky.social Liftoff Series! Grateful for the opportunity to share more about our launching lab's ethos and the big questions surrounding mechanisms of affect and mental health we aim to tackle in the coming years! 🧠

1308

Jan R. Wessel @janw.bsky.social · 18/06/2026

We have a new Cognitive Control Collaborative website: www.cognitivecontrol.net Six labs, more than 20 trainees, six years of collaboration (and counting). Consider joining us, many of us are looking for new grad students in the fall!

cognitivecontrol.net

Cognitive Control Collaborative

02010

Jan R. Wessel @janw.bsky.social · 12/06/2026

However, crucially, conscious error detection rates did not differ at all between placebo and atx. As such, we can safely conclude that autonomic nervous system activity indexes – but does not causally contribute to – conscious error detection. The End.

140

Jan R. Wessel @janw.bsky.social · 12/06/2026

First, we found that the pupil response once again tracked conscious error awareness. Moreover, atomoxetine indeed significantly changed this phasic pupil dilation response to errors. So the causal manipulation of phasic autonomic reactivity to errors worked.

110

Jan R. Wessel @janw.bsky.social · 12/06/2026

Here, we tested whether this blunted orienting response to errors causally contributes to conscious error detection (as predicted by some theories) or not. 30 healthy older adults performed an error awareness on 80mg atomoxetine (an sNRI to modulate autonomic responsivity) and on placebo.

100

Jan R. Wessel @janw.bsky.social · 12/06/2026

Years ago, we found that healthy older adults show impairments in subjective error awareness: they don't commit more errors, but fail to notice more of them. This deficit was significantly correlated with changes in pupil dilation response to errors across the life-span. tinyurl.com/3y5e942z

tinyurl.com

100

Reposted by Jan R. Wessel

Melinda Sabo @sabomelinda.bsky.social · 04/06/2026

#PuG crowd: interested in cognitive control? Come to our symposium “Cognitive Control: Domain-General, Domain-Specific, or Neither?” happening on Friday at 2 pm (room: HS07). We have a fantastic line-up: @annalschubert.bsky.social, Linda Bräutigam, @sarahdepue.bsky.social, Daxun Zhu, and myself. 👇

2147

Jan R. Wessel @janw.bsky.social · 27/05/2026

Alright awesome, thanks! I will let you guys know if she ends up going the Neuroscience route down the line! (fingers crossed 😀)

000

Jan R. Wessel @janw.bsky.social · 25/05/2026

Does anybody have a favorite book recommendation to get an incoming college Biology student excited about (cognitive/clinical) neuroscience?

130

Jan R. Wessel @janw.bsky.social · 24/05/2026

As a child, I always thought scientists learned by thinking hard until they have a great idea or insight. As a scientist, I mostly learn by realizing exactly how every great idea or insight I think I have is actually dumb and trivial. Anyway I just spend Saturday night in the lab for no reason.

050

Jan R. Wessel @janw.bsky.social · 11/05/2026

That HFA "more commonly" is thought of as "predominantly represent[ing] summed local spikes" seems like a straw man to me. The general notion that the LFP largely reflects (E)PSPs always seemed to me like the dominant one. The HFA~spike association seem(ed) only popular within one specific subfield

120

Jan R. Wessel @janw.bsky.social · 08/05/2026

I'm not gonna read that, obviously, but if you give me the word count I am confident I can write this article virtually word for word based on the combination of headline and outlet alone

010

Jan R. Wessel @janw.bsky.social · 06/05/2026

Now you hold on a second! What if instead of having to orient yourself for 5 minutes to figure out how a given research group organizes their data, you spend 5 hours making sure you have the right dependencies, software versions, and programming knowledge to use the One Universal Perfect Format?

010

Jan R. Wessel @janw.bsky.social · 06/05/2026

What? I personally love feeling like a dumb asshole every other day because the entire scientific ecosystem has been 'optimized' by and for aspiring google engineers that have never done a single experiment in their life.

220

Jan R. Wessel @janw.bsky.social · 06/05/2026

You know what's nice? Downloading a .mat file from the OSF and then loading it into MATLAB. What ever happened to that?

240

Jan R. Wessel @janw.bsky.social · 06/05/2026

Here's a list of things I'd rather do than try to open NWB data: - Chew on a ball of tinfoil - Drive an icepick into my eye socket - Take a full gainer into a cesspool - Watch James Harden play basketball List to be continued...

160