draft lab ai policy, feel free to use, modify, or discuss! todd.gureckislab.org/2026/03/06/g...
07.03.2026 02:18
👍 50
🔁 12
💬 1
📌 1
draft lab ai policy, feel free to use, modify, or discuss! todd.gureckislab.org/2026/03/06/g...
i switched to in person, no notes, paper+pencil exams graded with scantron this semester because there was no clear answer to this issue. so far result has been more genuine engagement in classes. OTOH some people dropped once they heard the evaluation required learning and remembering.
i relate
i'm kinda excited about markup languages for cooking: cooklang.org
say more!
def able to conjure educational interfaces of my dreams
We are excited to announce that the Cognitive Computational Neuroscience meeting (CCN 2026) will be held at New York University from August 3–6, 2026.
2026.ccneuro.org
🚨 #CCN2026 Proceedings submissions are open!
CCN 2026 again features an 8-page Proceedings track (alongside extended abstracts). Accepted papers will appear in CCN-Proceedings (CCN‑P) with DOIs on OpenReview.
Graphic with a teal background highlighting two upcoming Cognitive Science Society deadlines. On the left, a calendar icon shows January 19 next to the text “Deadline for nominations – The Glushko Dissertation Prize.” On the right, a calendar icon shows February 2 next to “Deadline for submissions – CogSci 2026: Cognitive Inefficiency.” Additional text notes the conference dates, July 22–25, in Rio de Janeiro, with small gear icons as decoration.
⏰ Two important #CogSci deadlines are coming up:
🗓️ Jan 19 - Glushko Dissertation Prize
Nominate an outstanding cognitive science dissertation:
cognitivesciencesociety.org/glushko-diss...
🗓️ Feb 2 - #CogSci2026 submissions
Check the guidelines and templates: cognitivesciencesociety.org/submissions/
effect size feels somewhat academic when you can readily convert even small behavioral effects into substantial $.
nice post. i don't know this data is public, or it's actual assessment, but my hunch is the effectiveness of optimized social media application algorithms for retaining human attention is 5 sigma on a chronological, friend only feeds of the early 2000s.
The cost of AI Over-Gloss
todd.gureckislab.org/2026/01/08/a...
some of the most vocal advocates for high quality advising will upload the wrong person's letter for phd admissions.
yeah, i've been white hatting some of this and i'm a little surprised at how bad most of these things are (for now), esp for multi-modal real-time stuff.
e.g., bsky.app/profile/josh...
nice! i think people should talk about the good data experiences like this, because i have run across reviewers and people on grant panels and stuff who are now completely spooked on online data quality to the point of basically flat reject it's validity, despite maybe a more nuanced set of issues
I'm exploring this issue a little and interested in where the failure points are. Please DM me.
Several people have mentioned online that they get terrible responses from online services such as Prolific, e.g., bots, LLM responses. I'm curious if anyone who has experienced that in a memorable way would mind sharing the details of their project (code, etc.).
Today is #GivingTuesday! Join us in helping to make computational science education accessible to everyone, everywhere!
Since, 2021, Neuromatch Academy and Climatematch Academy have supported nearly 14,000 learners from 135 countries.
Your support makes this possible.
neuromatch.io/donate/
it seems like not to many people that follow me are using it but here's a good summary of the upsides: towardsdatascience.com/why-im-makin...
the winner is columbia which had a box that looked like the pdf upload link but was actually a request to upload your business card (if appropriate). good captcha!
is anyone using marimo? it has some selling points over jupyter, esp. for better ai integration marimo.io
a dashboard of an experiment
we may not have agi yet, but claude code and i coded a dashboard so i can understand the geographic and demographic recruitment patterns in our Smile experiments and i'm excited how it turned out! 🤖+😎=✨ (a public version of this likely in a future Smile release, internal for now)
If you start someone on an online experiment and the subject reloads the page or browser quits -- what happens? Usually we are resigned that they just start over 😢 With the magic of Smile you can determine trials at run time and it picks up where it left off! smile.gureckislab.org/coding/steps...
"The inevitability and superfluousness of cell types in spatial cognition". Intuitive cell types are found in random artificial networks using the same selection criteria neuroscientists use with actual data. elifesciences.org/reviewed-pre... 1/2
1963 JExp. PSych -- if you can't see a significant between-subject effect with 28 subjects then might as well just accept your nulls. 🫠 (Worth nothing the effect they are referring to is a manipulation applied to only the first 5 out of ~140 total learning trials.)
CDS is hiring Faculty Fellows for 2026!
This two-year role offers time and support to pursue independent research with a minimal teaching load.
Learn more and apply by Nov 24: nyudatascience.medium.com/cds-is-hirin...
I'm recruiting a grad student! My lab at NYU (psych dept) studies the computational basis of moral cognition and aims to build AI systems that are aligned with human values. Now admitting a PhD student for Fall 2026. Apps due 12/1. sites.google.com/corp/site/sy...
