A Quantifiable Information-Processing Hierarchy Provides a Necessary Condition for Detecting Agency
🤖 AI Summary
This work addresses the longstanding challenge in defining agency, which has often relied on high-level descriptions that are difficult to quantify and lack operational criteria. The authors propose a bottom-up, substrate-independent hierarchical framework for information processing that classifies systems into three categories based on the temporal complexity of their input transformations: reactive (I), fixed-memory (II), and rule-adaptive (III). These categories serve as necessary informational conditions for identifying agency. Through information-theoretic analysis, system modeling, and illustrative examples—including thermostats and receptor-type memristors in both neural and computational contexts—the framework effectively distinguishes varying levels of information processing across diverse systems. This approach provides the first quantifiable and operationally applicable standard for evaluating the ethical and functional attributes of potential agentic systems.
Application Category
📝 Abstract
As intelligent systems are developed across diverse substrates - from machine learning models and neuromorphic hardware to in vitro neural cultures - understanding what gives a system agency has become increasingly important. Existing definitions, however, tend to rely on top-down descriptions that are difficult to quantify. We propose a bottom-up framework grounded in a system's information-processing order: the extent to which its transformation of input evolves over time. We identify three orders of information processing. Class I systems are reactive and memoryless, mapping inputs directly to outputs. Class II systems incorporate internal states that provide memory but follow fixed transformation rules. Class III systems are adaptive; their transformation rules themselves change as a function of prior activity. While not sufficient on their own, these dynamics represent necessary informational conditions for genuine agency. This hierarchy offers a measurable, substrate-independent way to identify the informational precursors of agency. We illustrate the framework with neurophysiological and computational examples, including thermostats and receptor-like memristors, and discuss its implications for the ethical and functional evaluation of systems that may exhibit agency.
Problem
Research questions and friction points this paper is trying to address.
agency
information processing
quantifiability
intelligent systems
system hierarchy
Innovation
Methods, ideas, or system contributions that make the work stand out.
information-processing hierarchy
agency detection
adaptive systems
quantifiable framework
substrate-independent
🔎 Similar Papers
No similar papers found.
