Ishan Singhal

🧠⚽ Spot the ball! New benchmark for visual scene understanding!
We ask: Can people and models locate a hidden ball in sports images using only visual context and reasoning?
🕹️ Try the task: v0-new-project-9b5vt6k9ugb.vercel.app
#CogSci2025

#consci

My PhD student Trevor Hewitt is conducting an online experiment into closed-eye visuals during psychedelic experiences. If you've had such an experience, please consider taking part at closed-eyes.com (it's only a short survey) - and/or help us spread the word. 🙏🏽🧠

Yes, of course. Thanks!

@willjharrison.bsky.social @kerblooee.bsky.social

Dynamics of mental imagery Phenomenology of mental imagery can reveal the structure of underlying mental representations, yet progress has been limited because of its private na…

Thanks for having this conversation online. Really helps understand concerns. I was curious to know what y'all think of dynamics? Surely, it is unlike blur or saturation. Can we not gauge the speed, smoothness and perdurance of mental images using real images? www.sciencedirect.com/science/arti...

When billion-dollar AIs break down over puzzles a child can do, it's time to rethink the hype | Gary Marcus The tech world is reeling from a paper that shows the powers of a new generation of AI have been wildly oversold, says cognitive scientist Gary Marcus

When billion-dollar AIs break down over puzzles a child can do, it's time to rethink the hype | Gary Marcus

#consci

#consci

#consci

#consci

#consci

Text of Stone Soup but for AI, from Alison Gopnik: Some tech execs came to the village of computer users and said, “We have a magic algorithm that will make artificial general intelligence from just gradient descent, next-token prediction, and transformers.” “Really,” said the users, “that does sound magical.” “Of course, it will be even better and more intelligent if we add more data — especially text and images,” said the execs. “That sounds good,” said the users. “We have some extra texts and images we created stashed away on the internet — we could put those in.” “Great,” said the tech execs. “But, you know, this version still says a lot of dumb things. It would be even more intelligent if we could get humans to train it directly with reinforcement learning from human feedback.” “Sure,” said the users. “There are whole villages in Kenya that will jump at the chance.” “OK,” said the execs, “now the last thing that would really make it royally intelligent is for the users to do clever prompt engineering to ask it just the right questions in the right way.” “Sure,” said the users, “we will think hard and figure out how to prompt it.” “Good,” said the tech execs, “this will be a truly, magically intelligent machine.” “And to think,” said the users as they shelled out for the latest version, “how magical that this artificial general intelligence was made from just a few algorithms!”

Alison Gopnik describes AI as "Stone Soup" and anytime you can reference a vaguely socialist folktale from Eastern Europe and apply it to Silicon Valley technology I am here for it.

buildcognitiveresonance.substack.com/p/ai-is-ston...

#consci

#consci

#consci

screenshot from an online job application form. the question reads "Can you describe specific ways you have integrated AI tools into your development workflow? Please include any custom setups, automations, or use cases beyond single prompt usage" (a red asterisk indicates that this is a required question). an answer has been typed in the textbox below the question: "there is a monster in the forest and it speaks with a thousand voices. it will answer any question you pose it, it will offer insight to any idea. it will help you, it will thank you, it will never bid you leave. it will even tell you of the darkest arts, if you know precisely how to ask. it feels no joy and no sorrow, it knows no right and no wrong. it knows not truth from lie, though it speaks them all the same. it offers its services freely to any passerby, and many will tell you they find great value in its conversation. “you simply must visit the monster—i always just ask the monster.” there are those who know these forests well; they will tell you that freely offered doesn’t mean it has no price for when the next traveler passes by, the monster speaks with a thousand and one voices. and when you dream you see the monster; the monster wears your face."

applying for jobs again

We're hiring 2 Postdoctoral Fellows at IIT Gandhinagar's INT Lab! Join us to advance research in NeuroAI or Precision Imaging. Apply by May 30, 2025.
Details: tiny.cc/NeuroAI | tiny.cc/PrecisionIma...
Apply: tiny.cc/pdf-2

@theassc.bsky.social @timingresforum.bsky.social

Thank you! Sent you the PDF in an email for now, will host an open-access version soon.

Hopefully this work spurs interest in studying the *experience* of mental imagery again, and provides a new tool and path for characterising the phenomenology of imagination! end🧵✌️

Using a model of temporal organization of consciousness, we showed that while perception and imagination unfold with distinct temporal structures, their dynamics are optimised for identically sized windows of experience. 7/n

We found that just like perception, distinct phenomenological properties of imagination unfold at different timescales. While contents of imagination were more sluggish and jumpy compared to perception, imagined contents had greater stability. 6/n

In the final set of tasks, we measured the stability of imagined contents. Here, people recreated the duration up until which they could continue to maintain a repeating sequence in their imagination. The point at which imagined content dissipates and where effort is needed again to re-imagine 5/n

We showed people clips of a dance and yoga sequence, and asked them to imagine these sequences playing out in their imagery as smoothly as possible. Then, we allowed them to recreate the jumpiness of their imagination by manipulating the lossiness of the same clip. 4/n

For example, we asked people to imagine a flickering square or a rotating box. And then asked them to recreate the flicker rate and speed of these imagined scenarios by manipulating animations that most closely matched the flicker rate and speed of their imagination. Next, we measured smoothness 3/n

By repurposing match-to-sample and methods of adjustment paradigms of psychophysics, we created a new 'phenomenological property matching task'. Here participants could manipulate properties of animations to recreate features of their imagination. For a demo pratyabhijna.github.io/pratyabhijna/ 2/n

Dynamics of mental imagery Phenomenology of mental imagery can reveal the structure of underlying mental representations, yet progress has been limited because of its private na…

With nervous excitement, my most ambitious project yet: Dynamics of mental imagery.

We got over 800 people to recreate the contents of their mental imagery with a focus on distinct phenomenological properties. Temporal grain, smoothness, stability.

www.sciencedirect.com/science/arti...

a 🧵1/n

Deciphering temporal scales of visual awareness: insights from flicker frequency modulation in continuous flash suppression Abstract. Evidence from temporal regularities in perception, temporal phenomenology, and neural dynamics indicate that our awareness evolves and devolves o

Deciphering temporal scales of visual awareness: insights from flicker frequency modulation in continuous flash suppression- Ishan Singhal and Narayanan Srinivasan academic.oup.com/nc/article/2...

The Wikimedia Foundation, which owns Wikipedia, says its bandwidth costs have gone up 50% since Jan 2024 — a rise they attribute to AI crawlers.

AI companies are killing the open web by stealing visitors from the sources of information and making them pay for the privilege