Track the frontier, from condensed matter to cosmology.
Search papers, labs, and topics across Lattice.
This paper critically evaluates existing biosignal world models against the requirements for intervention use, identifying three core physiological bottlenecks: temporal latency, system openness, and endogenous dynamics. These limitations hinder the ability of current models to accurately capture the causal effects of interventions on physiological systems. The authors propose research directions and an evaluation checklist to guide the development of more robust biosignal world models capable of true simulation and decision-making.
Biosignal world models fail to meet intervention use requirements due to fundamental physiological bottlenecks, necessitating a paradigm shift beyond mere prediction.
World models promise to learn how systems evolve and respond to actions, enabling simulation, planning and decision-making. Bringing this promise to biosignals requires more than accurate prediction of physiological recordings. We examine existing biosignal world models through two requirements for intervention use: a genuine action and evidence of its causal effect. We identify three fundamental bottlenecks rooted in physiology: Temporal Latency, the variable delay between an action and its observable effect; System Openness, the influence of drivers beyond the modelled intervention; and Endogenous Dynamics, the system's ongoing evolution without that intervention. Drawing on published evidence, we outline corresponding research directions and propose an Intervention-Effect Validation Checklist to guide evaluation. The aim is to help the biosignal community realise the core promise of world models by learning the latent mechanisms that govern physiological evolution and responses to interventions.