Also a good metaphor for "aligned" AI

2-The people: you’ll be part of an ambitious and highly collaborative team studying psychosis in humans and neural networks led by @mattnour.bsky.social at @oxfordpsychiatry.bsky.social @drrickadams.bsky.social at @ucl.ac.uk and @mariaeckstein.bsky.social at DeepMind

Immediately top of the wish list. Fantastic stuff

Adverts coming soon.

If you're passionate about understanding the computational architecture of the mind and its breakdown in psychiatry, reach out now! Happy to jump on a call.

All jobs 4 years, at @ox.ac.uk

#ComputationalPsychiatry #Neuroscience #AcademicJobs #PostdocPosition #PhDPosition

Part of an incredible cross-disciplinary effort:

Preclinical: @melgaby.bsky.social ( @nature.com paper: www.nature.com/articles/s41... ) 🐭

ANNs: @drrickadams.bsky.social & @mariaeckstein.bsky.social (www.biorxiv.org/content/10.1...) 💻

Each with their own jobs opportunities ...

❗Post-doc (computational/cognitive)❗

You'll develop cutting-edge analyses bridging brain imaging, behaviour, and language.

Lots of scope to develop new methods (LLM-based) and experiments. You'll work closely with computational collaborators on the animal and neural network side (see below!).

❗Pre-doctoral fellow (clinical)❗

You'll be supported to lead a clinical fMRI study, and embedded within an academic psychiatry group studying neurobiology of psychosis.

Ideal for a junior doctor at the start of their academic journey, keen to learn! 4 years, potential PhD.

You're correct

Thanks! Great paper. As you ofc know, PEs come in many forms (reward/state; sensory/semantic, low/high level). Most tasks (mmn) measure v. low level PEs. We suggest narrative tasks might allow us to track PEs at higher levels of abstraction. No strong claims about cog maps and delusions here.

Obligatory to reference @kennethstanley.bsky.social and Joel Lehman's amazing book

www.amazon.co.uk/Why-Greatnes...