Skip to main content
QUICK REVIEW

[论文解读] A Quantifiable Information-Processing Hierarchy Provides a Necessary Condition for Detecting Agency

Brett J. Kagan, Valentina Baccetti|arXiv (Cornell University)|Jan 7, 2026
Free Will and Agency被引用 0
一句话总结

本文提出一个自下而上的三阶信息处理层次结构,用于识别代理性所必需的信息前奏,其中III类(自适应、自我调节)为最高阶。

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.

研究动机与目标

  • 定义一个 substrate-independent、从下至上的框架,用于基于信息处理动态推断代理性。
  • 引入三阶有序信息处理类别(I、II、III),逐步包含记忆和适应性。
  • 论证Class III动态为跨系统的真实代理性的必要条件(而非充分条件)。
  • 用神经生理学和计算示例来说明输入如何在各类中转化。

提出的方法

  • 以最小化数学表示形式正式定义三类信息处理:Class I: R(t)=α(t)I(t)+ε(t); Class II: R(t)=T[I(t)]+ε(t)(具有固定变换T);Class III: R_t=T_t[I_t]+ε_t(通过基于过去输出的G更新自适应T)。
  • 提供案例示例:Class I 恒温器,具有外源增益调制;Class II 理想记忆电阻,具有固定非线性变换;Class III 记忆电阻生物受体,具有慢速自适应增益/偏置调制。
  • 通过分析输入–输出轨迹以及在周期驱动下IO平面的滞后环来表征记忆和自适应性;讨论通过滞后环的记忆指示与通过移动环来体现自适应。
Figure 1: Information-processing classes and their relation to memory, adaptivity, and agency.
Figure 1: Information-processing classes and their relation to memory, adaptivity, and agency.

实验结果

研究问题

  • RQ1系统在不同基质中具备代理性的必要信息条件是什么?
  • RQ2三阶信息处理层次能否将记忆和自适应性作为代理性前导特征捕捉到?
  • RQ3真实或仿真系统(恒温器、记忆电阻、记忆电阻生物受体)是否实现了每一类,它们的IO行为如何反映记忆/自适应?
  • RQ4Class III 的自适应是代理性的必要条件吗,即使它并非充分条件?
  • RQ5如何以 substrate-independent 的定量方法区分与代理性相关的动力学与仅仅是行为能力的动力学?

主要发现

  • 提出三类信息处理层次结构,III类(自适应)包含记忆与自我调节,提供代理性的可衡量前奏。
  • Class I 系统是反应型且无记忆;Class II 系统以固定算子变换输入;Class III 系统基于历史逐步自适应其变换规则。
  • 案例研究(恒温器、理想记忆电阻、记忆电阻生物受体)展示从无记忆到具记忆再到自适应调制的信息处理的演进。
  • 在IO平面轨迹中,记忆以滞后环的形式显现,并且从Class I到Class III逐步变得更复杂;在Class III中通过移动、变形的环显示自适应。
  • 该框架提供一种 substrate-independent 的定量方法来识别代理性的信息前奏,同时承认这些条件是必要的,而非充分条件,以实现真正的代理性。
Figure 2: First-order thermostat response to a square wave: instantaneous, proportional switching between two fixed output levels; no memory or adaptation.
Figure 2: First-order thermostat response to a square wave: instantaneous, proportional switching between two fixed output levels; no memory or adaptation.

更好的研究,从现在开始

从论文设计到论文写作,大幅缩短您的研究时间。

无需绑定信用卡

本解读由 AI 生成,并经人工编辑审核。