Neil Cohn

Congrats, Neil! Looks incredible.

That sounds exciting! Most of the book is about the structure of pictures but I do talk about sign language and deaf language learning in various places too. Thanks for your interest!

I'm starting a comics club at my son's school, so I'm excited to learn more about comics' formal visual language and to see how #deafkids in particular engage with it.

OK, I need to get my hands on this. <3

An old jungle comics page - simple looking comics page, but it's very unclear which panel follows from the next in a few places

An old jungle comics page - simple looking comics page, but it's very unclear which panel follows from the next in a few places

A slide with images from Neil Cohn's visual linguist work - showing how panel alignment affects reader flow. If panels align, you want to read in the direction of alignment...

1/2 An activity you can try! I do a brief discussion of unclear compositions: sharing these from Jungle Comics, a bit from @neilcohn.bsky.social's studies on layout readability, then students draw intentionally "Bad" layouts - where reading order isn't clear from design & 2nd student tries to solve!

Comic describing the history of a local shopping center as a former textile factory in Tilburg, Netherlands

Comic describing the history of a local shopping center as a former textile factory in Tilburg, Netherlands

One of our local shopping centers is a converted textile factory, so they made a comic to describe the history of the building and its conversion!

“Speaking With Pictures pushes boundaries of how we disseminate scientific content in unprecedented ways. Cohn’s innovative use of the comic format to illustrate theory and research is truly unique. The format allows him to illustrate complex arguments as they are being made. It is a must read for cognitive scientists interested in the psychology of visual languages and anyone who loves comics.” – Joe Magliano, Professor of Educational Psychology, Georgia State University, USA

More praise for Speaking with Pictures, my upcoming graphic novel on language, cognition, and visual communication. I'm counting down the days until this finally gets out and I can't wait...👀 visuallanguagelab.com/sip

And yet LLMs have been let loose in government and education.

Today in AI sucks, stop using it

Research takes time and will never simultaneously test the commercial models as they come out. I suspect they’ll do just as poorly. I’ve been tracking these studies for a decade and they’ve hardly improved

Our university is in the process of allowing AI use in assessment, a supposedly pragmatic response to its ubiquity. This means I can't ban it on my comics course. This might bolster students' desire to avoid it.

I kind of want to go back and edit all of my master's essays to add that comics are not only a legitimate form of literature but that we now have evidence proving they are an innately human one.

The least shocking thing. But don't worry, linear story telling is only a feature of comics!! /s

Considering that I’ve demonstrated LLMs can’t handle any of the subtextual engines in prose, this isn’t surprising, but it remains shocking how not good at it the technology is. The Future!

Narrative and spatial relationships are two fundamental aspects of human cognition. LLMs are consistently terrible at both and have no obvious path to improving.

Not shocking, to be honest.

I messed around with this in a far less scientific fashion and: yeah.

Beyond Single Frames: Can LMMs Comprehend Implicit Narratives in Comic Strip? Xiaochen Wang, Heming Xia, Jialin Song, Longyu Guan, Qingxiu Dong, Rui Li, Yixin Yang, Yifan Pu, Weiyao Luo, Yiru Wang, Xiangdi Meng, Wenjie Li, Zhifang Sui. Findings of the Association for Computatio...

Here's an interesting new study exploring whether LLMs are able to understand the narrative sequencing of comics and... even the best AI models are *terrible* at it for pretty much all tasks that were analyzed aclanthology.org/2025.finding...

Thanks a lot!

I forgot the most important: good luck to @cogirmak.bsky.social !

This is really interesting and I will definitively have a look! A long time ago, when I was a student, I did my project on this topic for a Semiotics module, at a much smaller scale. Funnily enough, it was about compiling and writing about all the different types of movements in the Tintin books.

♥️

Features of graphic motion cues uncovered through a corpus of 300+ comics

She additionally lead the annotation of a 300+ comics from around the world, and @cogirmak.bsky.social's analysis of them uncovered various abstract patterns involved in how motion events are encoded in different motion cues www.degruyterbrill.com/document/doi...

Proportions of routes depicted by comics from different types of languages

She also examined comics directly for how they use motion cues. @cogirmak.bsky.social first used a corpus of 85 comics and showed that the depiction of motion events varies based on the structures of the languages spoken by those authors www.degruyterbrill.com/document/doi...

Perceived speed ratings of different motion cues

In experiments that compared how people perceive the speed implied by these different motion cues, @cogirmak.bsky.social found that background lines and suppletion lines seem faster than normal motion lines, which mostly indicate direction, not speed journalofcognition.org/articles/10....

Different graphic cues showing motion events

In a review paper for Cognitive Science, @cogirmak.bsky.social showed that studies overall suggest that motion lines are not based on perception or metaphors, but are encoded as part of a visual lexicon that requires exposure and familiarity onlinelibrary.wiley.com/doi/full/10....

Dissertation of Irmak Hacimusaoglu

Congrats to @cogirmak.bsky.social whose dissertation is now printed and ready to be defended in a few weeks! She researched the visual depiction of motion events as part of our broader TINTIN Project, so here’s a little thread of her work… www.visuallanguagelab.com/tintin

Perceived speed ratings of how fast different objects move when using different motion cues

In experiments that compared how people perceive the speed implied by these different motion cues, @cogirmak.bsky.social found that background lines and suppletion lines seem faster than normal motion lines, which mostly indicate direction, not speed journalofcognition.org/articles/10....

Different cues used to depict motion, including a) postures, b) polymorphism, c) circumfixing lines, d) motion lines, e) suppletion lines, and f) backfixing lines

In a review paper for Cognitive Science, @cogirmak.bsky.social showed that studies overall suggest that motion lines are not based on perception or metaphors, but are encoded as part of a visual lexicon that requires exposure and familiarity onlinelibrary.wiley.com/doi/full/10....

Our special issue on ‘Parallelism in the Architecture of Language’ is now published! Huge thanks to all contributing authors, reviewers, Eva, Neil, and the topiCS journal team for their excellent work 📚🗣️🧠🤖🙏🏻
onlinelibrary.wiley.com/toc/17568765...