Dan Bennett

Lecturer in Multisensory Human Computer Interaction, Bristol University
Human autonomy & agency, psychology, enactivism

@comic-sans-soleil.bsky.social on bluesky

Lecturer in Multisensory HCI at Bristol University

Prev: Post-doc w @elisamekler at IT University of Copenhagen

"Really bad media can exorcise your semiotic ghosts" William Gibson - The Gernsback Continuum

Website
danbennettdev.github.io
Dan Bennett boosted:
Natasha 🇪🇺Natasha_Jay@tech.lgbt
2025-01-17

A comic about desire paths

Ok so… There’s a nice little park on a corner, and a path running directly to one edge of the park. That’s the first image, but there’s 12 other images here. In the second image, there is now a desire path running diagonally across the park. Next image, the path has been blocked by a park bench that the council has put in as a deterrent. Unfazed, in image 4 the path now deftly avoids the park bench. In images 5 and 6 the process is repeated, this time with a strategically placed bin which the desire path also avoids. Presumably in a fit of rage, the council plants an entire hedge to stop people creating new desire paths but eventually this too is defeated and a new path find a way. Finally, the council admits defeat and turns the desire path into a proper paved pathway. After a little while, a new desire path curving off this new paved pathway begins to emerge …
Dan Bennett boosted:
2025-01-17

Published today, the 2nd issue of the @tekhne_collab journal edited by Yann Leguay & Marie Lechner, who asked me to write a short article on audio technologies and extractivism which evolved into a meditation on planetary and media-technical magnetisms.

Thanks to Yann and Marie for putting together such an engaging issue and inviting me to be part of it!

tekhne.website/journal/issue-2

Screenshot of the essay title on a phone sccreen. materia prima written in glitchy text on neon green, with the title "Bluetooth Extractions: Geomagnetism & Metallurgy" above a satellite photo of the Bayan Obo rare earth mine in Mongolia.
Dan Bennett boosted:
2024-11-12

The new issue of @ACM ToCHI is now out.
dl.acm.org/doi/full/10.1145/36

It includes the paper of mine I'm most proud of to date: Beyond Intrinsic Motivation: The Role of Autonomous Motivation in User Experience

We try to push beyond measures of immediate experience to make good on the promise of SDT to address more eudaimonic, life-embedded aspects of motivation.

Thread here
hci.social/@dan_bennett/112976

2024-11-11

The new issue of @ACM ToCHI is now out.
dl.acm.org/doi/full/10.1145/36

It includes the paper of mine I'm most proud of to date: Beyond Intrinsic Motivation: The Role of Autonomous Motivation in User Experience

We try to push beyond measures of immediate experience to make good on the promise of SDT to address more eudaimonic, life-embedded aspects of motivation.

Thread here
hci.social/@dan_bennett/112976

Dan Bennett boosted:
2024-10-19

still amazes me how beautifully our cumulative iconicity measure pulls apart distributions & predicts where words fall in both iconicity ratings and guessability experiments — we thought it had to work like this theoretically, but it's another thing to see it work out so neatly

:OpenAccess: paper doi.org/10.1162/opmi_a_00162
blog: ideophone.org/anatomy-of-iconi

Two panel image showing, on the left, a so-called beeswarm plot of colourful dots that look like a christmas tree, and on the right, the same dots pulled apart by an independent measure, namely cumulative iconicity.

The christmas tree represents people's iconicity ratings: how well they think form and meaning fit together, when hearing a word and seeing its translation. The dots are coloured by cumulative iconicity rating, and it is already visible that lighter dots (higher ratings) appear higher up the tree.

The pulled apart version shows how strongly predictive cumulative iconicity is. If you're a "4" or "5" (4-5 distinct structural correspondences between form and meaning), you'll always end up near the top: people will rate you as highly iconic.Two-panel figure. On the left, two point clouds showing how ideophones (dots) are distributed in terms of guessability in a binary forced choice task. Across studies, most are upwards of 50% guessable. Panel B shows how one aspect of those same distributions is neatly predicted by cumulative iconicity: the higher an ideophone scores in cumulative iconicity (from 0 to 4), the higher it is in guessability. What this means is that an independently derived measure of cumulative iconicity is strongly predictive of how easy it is to choose the right meaning for an ideophone.
2024-10-19

Elisabeth Camp's response to Nguyen's book on agency and games. Consider this an endorsement I guess.

researchgate.net/publication/3

Print out of a document with lines outlined in green-blue ink and annotated with the words "Yes!!", "Yessss!!", "YES"
2024-10-05

Print-purchasing indulgence today in the amazingly curated www.rovaeditions.co.uk
Apt quote from Left Cultures:
"the problem with stuff on the internet is that a person (or their personal algorithm) must already be looking for it"

Dan Bennett boosted:
2024-10-04

Want to get rid of AI overview from your Google searches?

Just make this your default search page on your browser: udm14.com/

you're welcome

Dan Bennett boosted:
2024-10-04

For firefox users this add-on does the trick:

addons.mozilla.org/en-GB/firef

2024-10-01

May the universe bless those rare reviewers who turn in a good quality review the very first day the requests are sent out!

Dan Bennett boosted:
Dr. Jürgen Knödlsederjknodlseder@astrodon.social
2024-09-28

Green growth will fix it 🤡 #degrowth

A line graph showing the historic trend of CO2 emissions since 1900 together with future pathways compatible with warmings of 1.5, 1.7 and 2 degrees. In this representation the pathways are almost vertical, illustrating the speed with which emissions need to be reduced to limit global warming to a sustainable level.
Dan Bennett boosted:
Mike ElganMikeElgan
2024-09-28

New MIT research has found that AI-edited photos can reliably induce false memories of events personally experienced.

The scientists called them "synthetic human memories."

machinesociety.ai/p/new-ai-tri

Dan Bennett boosted:
Katarzyna Stawarzkstawarz@hci.social
2024-09-27

I'm recruiting a post-doc with good technical skills to develop interactive prototypes as part of our co-design process and later turn them into something we can deploy in participants homes. Deadline: Oct 27. Start: ASAP. More details: krb-sjobs.brassring.com/TGnewU

#HCI #HCIjobs

Dan Bennett boosted:
Jan B. VornhagenVornhagenJB@hci.social
2024-09-24

On a personal note:
I will be submitting my thesis end of this year and am looking for a position for next year. If you are (or know someone who is) looking for a postdoc in HCI (games) research with expertise in quant and quali analysis and some experience in sci-comm & Open Science, happy to have a chat/be pointed towards applications.

Dan Bennett boosted:
2024-09-24

‘Using GPT-4 to generate 100 words consumes up to 3 bottles of water.’

I’d like to see the climate cost flagged in simple terms like this every time we are about to use AI. Like a calorie counter for the climate. A pop-up AI warning label, like nutritional information on food packaging. If nothing else, it might make us individually aware of the collective cost of using AI frivolously.

#AI
#ClimateDiary

tomshardware.com/tech-industry

2024-09-20

@alarith congratulations!

2024-09-19

@VornhagenJB @zotero amazing. Is this the new zotero update? I haven't updated yet.

If Boox implement a version of this I will be a customer for life.

Dan Bennett boosted:
@yabellini@fosstodon has movedyabellini@fosstodon.org
2024-09-19

LinkedIn enables, by default, your consent to use your data to train AI.

Here is the direct link if you want to deactivate:

linkedin.com/mypreferences/d/s

2024-09-17

@elisamekler cool, for myself I think the activity of theorising can be window dressing as much as the use of it, and the line between them is thin enough that I'm comfy with the shorthand. I guess maybe not what you had in mind here tho.

Copy paste to all platforms: DIY music promo habits.

Client Info

Server: https://mastodon.social
Version: 2025.07
Repository: https://github.com/cyevgeniy/lmst