#active inference

[astoldbydee]

People forget that part of Reading Comprehension is being able to infer the intended audience of a piece of text without the writer having to TELL you directly! You should be able to read something and know who the intended audience is. That is basic reading comprehension... you shouldn't need clarifications all the time, you are supposed to be able to figure that out on your own!

[compneuropapers]

Interesting Reviews for Week 46, 2025

The history and future of resting-state functional magnetic resonance imaging. Biswal, B. B., & Uddin, L. Q. (2025). Nature, 641(8065), 1121–1131.

Models of human probability judgment errors. Huang, J., & Busemeyer, J. (2025). Journal of Mathematical Psychology, 125, 102906.

A concise mathematical description of active inference in discrete time. van Oostrum, J., Langer, C., & Ay, N. (2025). Journal of Mathematical Psychology, 125, 102921.

[compneuropapers]

Interesting Reviews for Week 40, 2022

Dynamic coupling of oscillatory neural activity and its roles in visual attention. Esghaei, M., Treue, S., & Vidyasagar, T. R. (2022). Trends in Neurosciences, 45(4), 323–335.

Perceptual hue, lightness, and chroma are represented in a multidimensional functional anatomical map in macaque V1. Li, M., Ju, N., Jiang, R., Liu, F., Jiang, H., Macknik, S., … Tang, S. (2022). Progress in Neurobiology, 212, 102251.

Planning in the brain. Mattar, M. G., & Lengyel, M. (2022). Neuron, 110(6), 914–934.

A step-by-step tutorial on active inference and its application to empirical data. Smith, R., Friston, K. J., & Whyte, C. J. (2022). Journal of Mathematical Psychology, 107, 102632.

[finelinesignattwo]

[finelinesignattwo]

deadly deadly

deadly sleep.

(the nightgaunt fucks you too)

[abrahamhaskins]

Model Math Update and Alignment!

I updated the formal architecture in the preprint today from Bayesian reasoning to Variational Inference after some... rather assertive mathematical critique from ML experts on Discord. So happy the model survived that review. In fact, the math's even cleaner now!

But now I am cognitively fatigued. I'm going to go read with the chickens until my brain stops steaming.

Generative AI inherently triggers a computational failure mode in human observers—a "generative crash"—due to a lack of latent intentionalit

[raffaellopalandri]

The Point of No Return: When Understanding Becomes a Moral Commitment

There exists a threshold within the architecture of cognition beyond which knowledge ceases to function as a neutral descriptive capacity and instead assumes the form of an irreversible ethical vector, a transformation that cannot be adequately described within the vocabulary of acquisition or comprehension, since it involves not merely the integration of new information but the reconfiguration…

[raffaellopalandri]

Against Disposable Knowledge: Learning in a Civilisation of Cognitive Consumption

To speak of learning within the contemporary civilisational configuration is already to confront a contradiction so deeply normalised as to appear invisible, namely that what is ostensibly valorised as knowledge has been structurally transformed into a consumable unit, detached from the processes of embodiment, integration, and ethical consequence that would otherwise confer upon it the status of…