iiinova — where your digital self keeps evolving

Alkis Hadjiosif @alkismh.bsky.social · 03/10/2026

One of the very unfortunate consequences of AI is that it rendered dashes—one of my favorite means of punctuation—suspicious

140

Reposted by Alkis Hadjiosif

allen institute @alleninstitute.org · 01/04/2026

The Orange Cat Brain Atlas is here. 🧠🐈 Today, we published the first comprehensive cellular map of the orange cat brain. The new atlas reveals a single, specialized neuron responsible for behaviors like staring at walls, knocking objects off tables, and the 3am "zoomies."

10379127

Alkis Hadjiosif @alkismh.bsky.social · 08/01/2026

This moves the goalposts for what makes a mind-blowing illusion

010

Reposted by Alkis Hadjiosif

Gavin Buckingham @drgbuckingham.bsky.social · 08/01/2026

Leftmost goal isn't actually facing the camera

5375

Alkis Hadjiosif @alkismh.bsky.social · 06/01/2026

Great for MOCAP making it to last Saturday's NYT Crossword

000

Alkis Hadjiosif @alkismh.bsky.social · 12/12/2025

Reach out if you have any comments! I will put together a bluesky version ("skeetprint"?) over the weekend.

010

Alkis Hadjiosif @alkismh.bsky.social · 12/12/2025

I want to thank to the reviewers for their tough but fair feedback, which I honestly think made this work stronger. I was nervous about trying eLife's new model, but editor and reviewers made it a smooth experience. (Also: the longest revision/response letter I have written so far).

110

Alkis Hadjiosif @alkismh.bsky.social · 05/12/2025

Random thoughts while going through paper proofs - how long until the terms "clockwise" and "counterclockwise" stop making intuitive sense? As traditional clocks become less common/relevant

110

Alkis Hadjiosif @alkismh.bsky.social · 30/11/2025

Trust as in, is the linear fit shown indeed calculated based on the data? Maybe, there are many blue dots on the top-right. More color contrast would be nice! Trust as in, is a linear fit a good description of what's going on here, not so sure...

010

Alkis Hadjiosif @alkismh.bsky.social · 13/11/2025

The frequency of review requests seems to follow the "when it rains, it pours" rule

010

Reposted by Alkis Hadjiosif

Gavin Buckingham @drgbuckingham.bsky.social · 25/10/2025

For your enjoyment (by @jagarikin)

513540

Alkis Hadjiosif @alkismh.bsky.social · 16/07/2025

Non-science post: if you are among the trivia aficionados in LearnedLeague, check out today's round on Ancient Roman cuisine that my friend Max and I put together!

000

Alkis Hadjiosif @alkismh.bsky.social · 22/05/2025

I am also honored to receive a Career Development Award from the @americanheart.bsky.social to push this work forward. I want to thank both the AHA and my faculty mentors in this award: David Lin at MGH, Conor Walsh at Harvard SEAS, and Monica Perez at Shirley Ryan.

071

Alkis Hadjiosif @alkismh.bsky.social · 22/05/2025

Excited to share my recent appointment as Instructor at MGH/HMS! I will continue and expand my work dissecting the neural mechanisms behind the different components of post-stroke motor impairment. The goal/hope is that better mechanistic understanding will reveal new targets for rehabilitation.

060

Alkis Hadjiosif @alkismh.bsky.social · 15/05/2025

Correct, all mysteries in the brain will be unveiled through the *cough* prism of visuomotor rotations.

010

Alkis Hadjiosif @alkismh.bsky.social · 11/05/2025

This suggests a clear benefit for latency reduction in computer-based training that involves implicit sensorimotor learning. Critically, it means that across-study differences may be explained by differences in unmeasured feedback latencies. We thus need to measure, report, and minimize latency. 6/6

010

Alkis Hadjiosif @alkismh.bsky.social · 11/05/2025

We found that, just a 60 ms increase in latency (from 25 ms to 85 ms) decreased implicit learning by 1/3 and increased explicit learning proportionally. Implicit sensorimotor learning was much more (5-10x) sensitive to latencies in the sub- 100 ms range than to higher latencies. 5/6

110

Alkis Hadjiosif @alkismh.bsky.social · 11/05/2025

We trained participants on a visuomotor rotation (VMR) under latencies of 25, 85, or 300ms. 25ms was the minimum we could achieve in our optimized setup; 85ms is within the latencies expected on an average setup. We dissected learning into implicit and explicit components using aim reports. 4/6

100

Alkis Hadjiosif @alkismh.bsky.social · 11/05/2025

Neurophysiology suggests that learning-related mechanisms such as cerebellar LTD and cortical LTP are disrupted by similarly short latencies, so this question might be worth investigating. 3/6

100

Alkis Hadjiosif @alkismh.bsky.social · 11/05/2025

Short delays in visual feedback (under 100 ms) are common in human-computer interactions—and, while hard to perceive, they’re known to hurt performance in tasks like pointing or robotic surgery. Do these tiny lags also impair sensorimotor learning? 2/6

100

Alkis Hadjiosif @alkismh.bsky.social · 11/05/2025

Our latest paper with Maurice Smith and team is now out: Tiny visual latencies can profoundly impair implicit sensorimotor learning, in Scientific Reports rdcu.be/eldlh. A brief summary: 1/n

rdcu.be

Tiny visual latencies can profoundly impair implicit sensorimotor learning

Scientific Reports - Tiny visual latencies can profoundly impair implicit sensorimotor learning

193

Alkis Hadjiosif @alkismh.bsky.social · 09/05/2025

Is it Kobe or somewhere by the Aegean?

120

Alkis Hadjiosif @alkismh.bsky.social · 06/01/2025

Following up on my email from last year

010

Alkis Hadjiosif @alkismh.bsky.social · 02/12/2024

I had a similar experience - it was a schematic based on a photo, where you could arguably recognize the participant (one of the authors on the paper!) I guess biorxiv doesn't want to deal with permissions etc. so the simplest thing to do is avoid human photography altogether.

010

Alkis Hadjiosif @alkismh.bsky.social · 23/11/2024

My take/intuition is that, talking to a human (or at least being convinced that you are) matters by itself. The exact same responses won't be as effective if one suspects the therapist is an LLM

010

Alkis Hadjiosif @alkismh.bsky.social · 18/11/2024

Thanks for putting this together! I'd like to be added.

010

Alkis Hadjiosif @alkismh.bsky.social · 08/11/2024

Hello, world! I'm happy to join the neuro community here at bluesky. I look forward to engaging discussions.

0100