St John Grimbly

St John Grimbly is a researcher in the Department of Computer Science at the University of Cape Town working with Jonathan Shock, Benjamin Rosman, Ryan Smith, and Mark Solms on computational models of homeostasis, interoception, and affective agency. His work implements Solms’s brainstem homeostatic drive theory inside discrete-state partially observable Markov decision processes under active inference.

In his October 2026 paper, “Two Routes to Interoceptive Prioritisation: Preference Reweighting and Likelihood Precision in Multi-Need Homeostatic Agents” (arXiv:2610.18617), Grimbly introduced the AffectWorld benchmark, where an embodied agent must balance three coupled lethal needs (hunger, thirst, and oxygen) while learning its own interoceptive likelihood model online via Dirichlet count updates. Across 11 primary layouts and 72 pre-registered additional layouts, the study shows that urgency-based preference reweighting and precision-based visceral attention solve distinct failures and raise survival from 0.56 at baseline to 0.94 when combined.

Grimbly’s profile is Builder, representing the engineering translation of neuropsychoanalytic and active-inference principles into reproducible agent environments where architectural claims about drive and attention can be ablated directly.

Known for. AffectWorld multi-need active inference environment, two routes to interoceptive prioritisation (preference reweighting and likelihood precision)

Coverage on this site

• Mark Solms and Ryan Smith Test Two Routes to Interoceptive Prioritisation October 2026

Related researchers

• Mark SolmsUniversity of Cape TownNeuropsychoanalysis, and the brainstem account of affective consciousness Biological naturalist5 articles
• RS Ryan SmithLaureate Institute for Brain Research. University of TulsaComputational psychiatry and active inference models of emotional awareness, interoceptive inference, hierarchical models of conscious access and affect Functionalist1 article
• BR Benjamin RosmanSchool of Computer Science and Applied Mathematics, University of the WitwatersrandReinforcement learning and decision-making in multi-need agents, skill composition, active inference homeostatic prioritisation Builder1 article
• JS Jonathan ShockDepartment of Mathematics and Applied Mathematics, University of Cape TownMathematical physics and computational neuroscience of active inference, multi-need homeostatic agency, neural dynamics Builder1 article
• KF Karl FristonUniversity College London. Chief Scientist, VERSES AIThe free energy principle and active inference Functionalist24 articles
• VA Vikas AgnihotriIndependent Aerospace and Neuromorphic Systems ResearcherNeuromorphic Satellite Intelligence framework, radiation-induced energy overhead model for spiking neural hardware, constrained on-orbit STDP Builder1 article

St John Grimbly is listed with the other builders, who hold that the claim is about what the hardware does, and the philosophy is left open. The positions themselves are set out in the index of consciousness theories. Every researcher covered on this site is indexed in the directory of consciousness researchers.