Toby Wise

Pleased to see this paper now out in @CellReports! Take a look if you're interested in how people use interactive cognitive maps to avoid danger cell.com/cell-reports/f…

🚨Calling early to mid-career women in academia We’re conducting research @warwickmed to explore the challenges faced by women in academia. Your insights will help us shape strategies to support career progression and retention. Survey password is 4321 warwick.co1.qualtrics.com/jfe/form/SV_54…

I’m excited to share our new, open-access paper published today in @NatMentHealth, “Latent mechanisms of language disorganization relate to specific dimensions of psychopathology”. nature.com/articles/s4422…

Preprint for "Consolidation of Sequential Planning" with @evanrussek and @NeilBurgess10. We demonstrate the neural representations supporting sequential planning and the effect of memory consolidation upon them: biorxiv.org/cgi/content/sh…. As talk, see also: youtube.com/watch?v=0QcNtz…

🚨 New preprint on the role of eating disorder symptoms in accelerating response vigor during foraging decisions 💫 w/ Sam Hall-McMaster 📜Link: osf.io/preprints/psya… 💻Code and data: github.com/ozika/ed-forag… Funded by @AvHStiftung @mpib_berlin

What happens when our stream of consciousness turns inward, towards the body? Our new fMRI study of 536 individuals finds that 'body-wandering' is associated with distinct patterns of brain connectivity, physiology, affect, and mental health: biorxiv.org/content/10.110…

@shaharnitzan @paul_b_sharp We found something which might be evidence of counterfactual state updating using MEG in our model-based avoidance task (unchosen terminal states were reactivated post-outcome) - very indirect evidence though! science.org/doi/full/10.11…

@paul_b_sharp Interesting. Maybe a more preliminary question would be whether humans engage in latent counterfactual updating for states they could have visited but chose not to. We know this happens for actions, but Im not sure if anyone ever tried to show that at the level of states ..

The 2-step task could be solved by using a cog map to assign credit to non-taken action post rare transition. This differs from using a cog map to plan action values at decision time. Does existing work test which is used, & do we think the former is an instance of planning?

@paul_b_sharp @marcelomattar Reminds me also of this paper showing that MF strategies can mimic "MB" behaviour in the two-step task journals.plos.org/ploscompbiol/a…

Thanks Marcelo! It's been great to see groups like yours demonstrate how NNs find clever / cheaper strategies relative to assumed model-based solutions. Along these lines, I developed a delayed planning task, where subjects can choose to engage control via selecting an action (where initial engagement of control presumably reflects planning formation) versus relinquish control (if they correctly infer actions at that stage have equivalent value) and it took us years to realize identical-valued strategies to the assumed model-based ones that are computationally cheaper. Makes me think we should use NNs for model discovery on our tasks during piloting more often!

Null findings are important ‼️ Excited to share my latest work finding model-based planning does NOT improve following ketamine 💉, antidepressant 💊 or iCBT 🧑‍💻 treatments, 🧵👇 🔗 osf.io/preprints/psya…

We've also created a Python package that implements the method, alongside documentation and examples. github.com/tobywise/FACSI… 4/n

In this preprint we refine the method and describe it in detail. Essentially, we train Lasso-regularised regression models to predict sum scores from individual items. The result is a short scale with a set of weights to optimally predict true sum scores. 3/n

🌟New preprint! 🌟 We describe a method for generating short versions of questionnaire measures using (very simple) machine learning (w/ @NuraSidarus) osf.io/vwmkd 1/n

Bit of a last minute announcement, but I have a project in the @KCL_MRCDTP catalogue - deadline is next week! See the link for more info, and get in touch if you want to know more kcl-mrcdtp.com/project/self-a…

Happy to share our latest publication, out now in @CognitionJourn, where we find that metacognitive self-beliefs are updated in a suprisingly domain-specific manner. sciencedirect.com/science/articl…

Just out! 🗞️ * How RL and WM interact when actions can lead to reward or to punishment * A striking lack of learning and retention diffs as a function of depr/anxiety and rumin symptoms * An RL but not S-R module captured retention patterns psyarxiv.com/82pyz

🚨PhD position available in Cognitive Neuroscience / Computational Psychiatry with us @ChariteBerlin funded by @dfg_public on mood-cognition interactions in depression, more infos here: karriere.charite.de/en/job-vacanci… Get in touch if you have questions about the position.

Two more weeks to apply for this position! Please help me by spreading the news to anyone you know who might be interested. Thanks!!

🚨My Tübingen lab has 2 new job openings🚨: Neuroimaging postdoc on real-time fMRI to uncover the role of neural noise on behaviour: devcompsy.org/wp-content/upl… Computational modelling postdoc, co-supervised w Peter Dayan: large datasets & mental health: devcompsy.org/wp-content/upl…

*New* paper out today in World Psychiatry! "Where do neurodevelopmental conditions fit in transdiagnostic psychiatric frameworks? Incorporating a new neurodevelopmental spectrum" 🧵👇 (1) @QM_SBBS @QMULSciEng onlinelibrary.wiley.com/doi/10.1002/wp…

@DoneganKelly Thanks Kelly! It was of course very much inspired by your brilliant work!

Didn’t think the two-step task could get much better… but it has 👏🏼 Super interesting to see my work replicated & extended! Nice demo of how📱& modelling can help our mechanistic framework of compulsivity 🆒🆒🆒 @toby_wise