Paul Bays

Our review of Bayesian efficient coding as a theory of perception is out now in TICS – with @ivntmc.bsky.social and Máté Lengyel. www.cell.com/trends/cogni...

Just published open access in 𝘗𝘴𝘺𝘤𝘩𝘰𝘭𝘰𝘨𝘪𝘤𝘢𝘭 𝘔𝘦𝘵𝘩𝘰𝘥𝘴: “𝗪𝗼𝗿𝗸𝗶𝗻𝗴 𝘄𝗶𝘁𝗵 𝗖𝗶𝗿𝗰𝘂𝗹𝗮𝗿 𝗗𝗮𝘁𝗮: 𝗔 𝗧𝘂𝘁𝗼𝗿𝗶𝗮𝗹 𝗳𝗼𝗿 𝗖𝗼𝗴𝗻𝗶𝘁𝗶𝘃𝗲 𝗮𝗻𝗱 𝗕𝗲𝗵𝗮𝘃𝗶𝗼𝗿𝗮𝗹 𝗥𝗲𝘀𝗲𝗮𝗿𝗰𝗵” 🔗 doi.org/10.1037/met0... 🙎 Co-authors: @bayslab.org, @ivntmc.bsky.social, @cjungerius.bsky.social, Xiaolu Wang, Julie de Falco, Zahara Gironés, Adam Triabhall

Feeling competent can bias visual encoding like a reward. In these tasks, observers made more confident about some stimuli prioritized encoding them: better where already accurate, worse where not, and less accurate overall. @ivntmc.bsky.social @bayslab.org #neuroskyence #PsychSciSky

Do the contents of visual #WorkingMemory deteriorate more quickly when more items are being remembered? Seems a straightforward question but we had to bring together 13 studies and >120,000 trials to get definitive answers... doi.org/10.1037/rev0... @ivntmc.bsky.social

Our lab is at @ecvp.bsky.social this week, giving one talk and presenting three posters. Come and say hi!

Our lab is now on LinkedIn: www.linkedin.com/company/bays...

Feeling really proud of this one: a review of Bayesian efficient coding with Máté Lengyel and Paul Bays (@bayslab.org) is now available online in TICS! authors.elsevier.com/a/1nF3i4sIRv...

New preprint with Sebastian Schneegans and @bayslab.org doi.org/10.31234/osf... Here, we ask whether the key limits of working memory - load and retention interval - are independent, or do they interact? Despite years of research, this question is still much debated. #psychscisky #neuroskyence 1/6

New work on sensory prediction resolves some long-standing puzzles about attenuation of self-touch journals.physiology.org/doi/full/10....

Now out in Comms Psych: the "anti-Bayesian" size-weight illusion is a consequence of the brain focusing resources on encoding typical combinations of size and weight (i.e. "efficient coding"). Explains the material-weight illusion too www.nature.com/articles/s44...

We have an opening for a postdoc to work on a collaborative project with Máté Lengyel (UCambridge Engineering) combining machine learning methods with human experiments on visual perception and memory www.jobs.cam.ac.uk/job/49299/

Out now in NHB (finally!), a review of visual working memory from a computational perspective, with @weijima01 @timothyfbrady and Sebastian Schneegans.

Nice commentary from @salinas_urgent and @mangosheikh_B on our new eLife paper

The size-weight illusion is a by-product of efficient sensory coding adapted to the combinations of volume and mass found in everyday objects. New preprint

We measured how effectively observers can reallocate working memory resources to new visual items when old ones become obsolete - people are surprisingly good at it! New paper with @ivntmc @dataforyounz @DAagtenMurphy

A critique of the psychological similarity account of working memory errors: work with @ivntmc now out in JEP:LMC

New work with Jess McMaster & others: we show swap errors (item confusions) in cued recall are not a strategic response to forgotten items, but instead occur at exactly the rate predicted by variability in recall of the cue features

New in Psych Review with Sebastian Schneegans & Jess McMaster: comparing the roles of time and space in binding features in working memory

We have an opening for a post-doc (or potentially a talented graduate RA) to research computational mechanisms of visual perception/memory using online and offline experiments - note deadline 11 Aug

Views of an object before and after a saccade may be combined even if you are aware the object has changed - new with Garry Kong, @DAagtenMurphy and Jess McMaster.

New in JOV: the ability to combine visual evidence across gaze fixations depends on a limited but flexible memory resource

New paper on the consequences of stroke for recall precision and binding in visual WM, a collaboration with Roy Kessels and colleagues at the Donders (@DondersInst)

Our new Analogue Report Toolbox can be downloaded at . It implements in MATLAB a range of methods we use in the Bays lab for analyzing and modelling behavioural responses on VWM recall tasks, including...

Our new PNAS paper reveals how the main competing models of working memory limits can all be interpreted in terms of sampling. A number of surprises, including that item limits don't require discrete representations.

New with @robthedatafiend in Psych Review: for every visual feature dimension there is a consistent upper bound on the "s.d." you can obtain by fitting errors with a normal+uniform mixture...and it just might be telling us something about neural tuning!

"Stochastic sampling provides a unifying account of working memory limits" - our new preprint:

A short note on BioRxiv on the relationship between the psychophysical scaling account of working memory by @timothyfbrady @markSchurgin and population coding models of the same:

Asymmetric competition in visual working memory: storing orientations doesn't affect memory for facial expressions, but storing expressions degrades orientation recall - new paper with Viljami Salmela on WM at different levels of the visual hierarchy:

An independent store in working memory for the locations of visual objects in relation to one another: provides a separate source of information for recalling locations and doesn't tap absolute (egocentric) WM resources - work with @DAagtenMurphy: PDF here

We have an opening for a graduate Research Assistant to join @BaysLab. More info here:

A Journal of Neuroscience "journal club" about our work on drift in working memory representations, by Ben Cuthbert & Dominic Standage

Our new review article identifies three distinct functions of transsaccadic memory

Another postdoc opening in @BaysLab: we're looking for someone with background in experimental study of eye movements to investigate visual evidence accumulation across gaze fixations. Apply here:

New paper in PLOS Comp Biol reveals how the sensory strength of a stimulus to be remembered determines how much space it takes up in working memory (corrected link)

New paper in PLOS Comp Biol reveals how the sensory strength of a stimulus to be remembered determines how much space it takes up in working memory

Our new review article sets out the latest advances in understanding how visual features are bound together in working memory:

We are looking for two new post-docs to join @BaysLab. Applications by 13 September:

New paper in J Neurosci brings together efficient coding theory and the neural resource model of visual working memory to explain why the bias and variance of recall estimates varies across stimulus space

Time from manuscript accepted to full and final publication online: 9 days. Well done to Cognitive Psychology, but why can't every journal do this?

Another challenge to discrete resource models of WM - new article in Cognitive Psychology

Reassessing the evidence from imaging data for a fixed item limit in working memory - new paper in Cerebral Cortex

As of 2017 we will be making all our published data publicly available via our publications page

The population coding model of working memory explains why people make binding ("swap") errors

New article in PLOS Computational Biology examines the fidelity with which we store information about reward

New code available for trial-by-trial estimation of mixture component probabilities in working memory tasks

New paper in eLife: activity in angular gyrus predicts the precision of episodic memories

We have an opening for a post-doc in our lab, see