Benjamin Lowe
Cog neuro postdoc at Macquarie Uni, Sydney
Activist for a free Palestine 🇵🇸
- He’s simply a grifter
- [Not loaded yet]
- Thanks Junjie!
- And it's out now in Cortex: www.sciencedirect.com/science/arti... Summary below 🧵
- 🚨Pre-print of some cool data from my PhD days! doi.org/10.1101/2025... ☝️Did you know that visual surprise is (probably) a domain-general signal and/or operates at the object-level? ✌️Did you also know that the timing of this response depends on the specific attribute that violates an expectation?
-
View full threadI’d like to thank my co-authors (particularly Naohide and Jonny) and reviewers for helping me elevate the quality of this work 😊
- Also, I apologise for the poor figure quality in the HTML version of the article. Elsevier’s typesetting team made some nonsense changes that I did not consent to, which have somehow proved to be frustrating to fix on their end. The PDF version is fine though!
- Perhaps the coolest result was that these surprise signals were *shared across attributes*. That is, classifiers trained to decode surprise for shape could reliably do so for colour (and vice versa), after accounting for latency shifts.
- This suggests that visual surprise may operate at the bound object level and/or is a domain-general response, which is identical to the conclusion drawn from our previous work! www.sciencedirect.com/science/arti...
- We first looked at the evoked responses and found classic effects of adaptation via the constant vs. change sequence comparisons. This said, we found no evidence for visual surprise after controlling for cortical adaptation (i.e., when comparing surprising changes to neutral changes).
- Interestingly, we were still able to decode multivariate whole-scalpe representations of surprise (neutral vs. violation) separately for each attribute. Moreover, these signals were reliable from ~250 ms, suggesting that surprise is predominantly signalled after the initial feedforward sweep.
- But when does the visual system signal surprise? And do the dynamics of a surprise signal depend on which attributes (features) violate a prediction? This is important to think about, given the functionally segregated organisation of the visual system.
- Here, we recorded EEG from participants who viewed sequences of a bound object that changed in either colour or shape over four steps. Crucially, the contexts of these changes were designed to appear random (and unsurprising) or violate the established trajectory (and cause surprise).
- Predictive coding theories assert that the brain uses prior knowledge when resolving percepts. Deviations between what is predicted and sensed generate surprise signals (so-called ‘prediction errors’), which calibrate the relevant erroneous predictions.
- [Not loaded yet]
-
View full threadPerhaps the coolest result was that surprise signals were *shared across attributes*. That is, classifiers trained to decode surprise for shape could reliably do so for colour (and vice versa), after accounting for latency shifts.
- This suggests that visual surprise may operate at the bound object level and/or is a domain-general response. This is identical to the conclusions drawn from our previous work :) doi.org/10.1016/j.co...
- We first looked at the evoked responses and found classic effects of adaptation via the constant vs. change sequence comparisons. This said, we found no evidence for visual surprise after controlling for cortical adaptation (i.e., when comparing surprising changes to neutral changes).
- Interestingly, we were still able to decode whole-scalp multivariate representations of surprise (neutral vs. violation) separately for each attribute. Moreover, these signals were reliable from ~250 ms, suggesting that surprise is predominantly signalled after the initial feedforward sweep.
- But when does the visual system signal surprise? And do the dynamics of a surprise signal depend on which attributes (features) violate a prediction? This is important to think about, given the functionally segregated organisation of the visual system.
- Here, we recorded EEG from participants who viewed sequences of a bound object that changed in either colour or shape over four steps. Crucially, the contexts of these changes were designed to appear random (and unsurprising) or violate the established trajectory (and cause surprise).
- Predictive coding theories assert that the brain uses prior knowledge to predict upcoming visual events when resolving percepts. Deviations between what is predicted and sensed generate surprise signals (so-called ‘prediction errors’), which calibrate the relevant erroneous predictions.
- And it was an absolute treat to run! Thanks everyone who attended :) #ACNS2025
- Shout out to Australian researchers! 🇦🇺 (and folks in the southern hemisphere) 🌏 We're excited that our Ambassador Ben Lowe (@brainboyben.bsky.social) will be hosting a pre-conference workshop at #ACNS2025 (@acnsau.bsky.social) on getting started with PsychoPy! Sign up here👇 shorturl.at/gvnUU
- Road trippin’ to ACNS 2025, Melbourne! @matthewod.bsky.social @tvcottier.bsky.social (Plus Ella and Seri) @acnsau.bsky.social
- Maybe a bit of a downer, but I think this conversation may be of interest to a bunch of people on here: www.youtube.com/watch?v=dSbK...
- A wonderful paper!
- Super happy to share my very first first-author paper out in @sfnjournals.bsky.social! We show content-specific predictions are represented in an alpha rhythm. It’s been a beautiful, inspiring, yet challenging journey. Huge thanks to everyone, especially @peterkok.bsky.social @jhaarsma.bsky.social
- One of the most depressing phd experiences is hearing of others' advisors (the ones that are supposed to train us into good scientists) encourage the use of chatbots in lieu of their students' development. thankfully mine don't.
- @sulfaro.bsky.social literally what we were just talking about!
- I feel seen—and heard: www.theguardian.com/lifeandstyle...
- New BBS article w/ @lauragwilliams.bsky.social and Hinze Hogendoorn, just accepted! We respond to a thought-provoking article by @smfleming.bsky.social & @matthiasmichel.bsky.social, and argue that it's premature to conclude that conscious perception is delayed by 350-450ms: bit.ly/4nYNTlb
- [Not loaded yet]
- Throw hands and then give hug. Good on ya, Will!
- Looking forward to this!
- Shout out to Australian researchers! 🇦🇺 (and folks in the southern hemisphere) 🌏 We're excited that our Ambassador Ben Lowe (@brainboyben.bsky.social) will be hosting a pre-conference workshop at #ACNS2025 (@acnsau.bsky.social) on getting started with PsychoPy! Sign up here👇 shorturl.at/gvnUU
- 🚨Pre-print of some cool data from my PhD days! doi.org/10.1101/2025... ☝️Did you know that visual surprise is (probably) a domain-general signal and/or operates at the object-level? ✌️Did you also know that the timing of this response depends on the specific attribute that violates an expectation?
- Nice work @brainboyben.bsky.social!
- Thanks @bealebrains.bsky.social!
- FYI I've published similar results previously :) doi.org/10.1016/j.co...
- It’s frankly absurd that we’re at the point where this critique needed to be written
- Paper on the problems of using LLMs as replacements for human participants journals.sagepub.com/doi/full/10....
- I really like this paper. I fear that people think the authors are claiming that the brain isn’t predictive though, which this study cannot (and does not) address. As the title says, the data purely show that evoked responses are not necessarily prediction errors, which makes sense!
- 1/3) This may be a very important paper, it suggests that there are no prediction error encoding neurons in sensory areas of cortex: www.biorxiv.org/content/10.1... I personally am a big fan of the idea that cortical regions (allo and neo) are doing sequence prediction. But... 🧠📈 🧪
-
View full threadIMO (now that this canned of worms has been opened), I think the field would really benefit from moving away from evoked responses and towards pre-stimulus and/or state-based activity characterising how predictions themselves are signalled (rather their errors!)
- I think a lot of people studying neural expectation have been skeptical of literal interpretations of PC for a while now. Again, this is not same as saying the brain doesn’t integrate prior knowledge with sensory input when resolving precepts. I’m excited to see where the field goes next :)
- The envelope has been pushed forward and now we can think about what these data mean within the broader literature. It’s exciting!
- Climate science 👀
- 🚨 Neuromatch Academy Course Applications are OPEN for 2025!! 🚨 Get your application in early to be a student or teaching assistant for this year’s courses! Applications are due Sunday, March 23. Apply & learn more: neuromatch.io/courses/ #mlsky #compneurosky #ai #climatesolutions #ScienceEdu 🧪
