Goals and the structure of experience
Clinical Snapshot
PICO Framework
| P — Population | Cognitive agents (natural and artificial intelligence systems) |
| I — Intervention | Goal-directed state representation framework with telic states |
| C — Comparator | Traditional reinforcement learning models with separate state representation and reward functions |
| O — Outcomes | Unified account of purposeful behaviour across behavioural, phenomenological and neural dimensions |
Bottom Line
This theoretical paper introduces a novel computational framework where goal-directed behaviour emerges from the co-development of descriptive and prescriptive aspects of world models, rather than treating them as separate components. While intellectually interesting and potentially important for understanding purposeful behaviour in both natural and artificial systems, the work remains purely theoretical without empirical validation. The framework draws on Buddhist epistemology to define 'telic states' as classes of goal-equivalent experiences, offering a parsimonious account of learning through statistical divergence measures. However, clinicians should note this has no immediate practical applications in healthcare. The work may eventually inform AI development in medical settings, but requires substantial further development and testing before any clinical relevance can be established.
Key Findings
Effect Size: Not applicable - theoretical framework
Primary Outcome: Introduction of telic states as goal-equivalent experience distributions
Nnt Or Sensitivity: Framework claims to provide unified account across diverse cognitive substrates
Clinical Application
Requires significant computational development and empirical validation before clinical implementation Minimal direct relevance to Australian clinical practice; may inform future AI-assisted healthcare technologies Limited direct clinical application; primarily relevant for AI development and cognitive neuroscience research
Abstract
Purposeful behaviour is a hallmark of natural and artificial intelligence. Its acquisition is often believed to rely on world models, comprising both descriptive (what is) and prescriptive (what is desirable) aspects that identify and evaluate state of affairs in the world, respectively. Canonical computational accounts of purposeful behaviour, such as reinforcement learning, posit distinct components of a world model comprising a state representation (descriptive aspect) and a reward function (prescriptive aspect). However, an alternative possibility, which has not yet been computationally formulated, is that these two aspects instead co-emerge interdependently from an agent's goal. Here, we describe a computational framework of goal-directed state representation in cognitive agents, in which the descriptive and prescriptive aspects of a world model co-emerge from agent-environment interaction sequences, or experiences. Drawing on Buddhist epistemology, we introduce a construct of goal-directed, or telic, states, defined as classes of goal-equivalent experience distributions. Telic states provide a parsimonious account of goal-directed learning in terms of the statistical divergence between behavioural policies and desirable experience features. We review empirical and theoretical literature supporting this novel perspective and discuss its potential to provide a unified account of behavioural, phenomenological and neural dimensions of purposeful behaviours across diverse substrates. This article is part of the theme issue 'World models in natural and artificial intelligence'.
References
- 1.Amir, N., Tiomkin, S., & Langdon, A. (2026). Goals and the structure of experience. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 384(2274). https://doi.org/10.1098/rsta.2025.0004
This content is for educational purposes for healthcare professionals only and does not constitute clinical advice. Clinical decisions should be based on individual patient assessment, current guidelines, and appropriate specialist consultation. Editorial Standards · Privacy Policy · Terms of Service