Kuchling Franz, Singh Isha, Daga Mridushi, Zec Susan, Kunen Alexandra, Levin Michael
Allen Discovery Center, Tufts University, Medford, MA, USA.
J R Soc Interface. 2025 Feb;22(223):20240645. doi: 10.1098/rsif.2024.0645. Epub 2025 Feb 26.
The field of diverse intelligence explores the capacity of systems without complex brains to dynamically engage with changing environments, seeking fundamental principles of cognition and their evolutionary origins. However, there are many knowledge gaps around a general behavioural directive connecting aneural to neural organisms. This study tests predictions of the computational framework of active inference based on the free energy principle in neuroscience, applied to aneural biological processes. We demonstrate pattern recognition in the green algae using phototactic experiments with varied light pulse patterns, measuring their phototactic bias as a readout for their preferential ability to detect and adapt to one pattern over another. Results show adapt more readily to regular patterns than irregular ones and even exhibit memory properties, exhibiting a crucial component of basal intelligence. Pharmacological and electric shock-based interventions and photoadaptation simulations reveal how randomized stimuli interfere with normal photoadaptation through a structured dynamic interplay of colony rotation and calcium-mediated photoreceptor-to-flagellar information transfer, consistent with uncertainty minimization. The detection of functional uncertainty minimization in an aneural organism expands concepts like uncertainty minimization beyond neurons and provides insights and novel intervention tools applicable to other living systems, similar to early learning validations in simpler neural organisms.
多元智能领域探索了没有复杂大脑的系统与不断变化的环境动态互动的能力，旨在寻找认知的基本原理及其进化起源。然而，在连接无神经生物与神经生物的一般行为指令方面，存在许多知识空白。本研究基于神经科学中的自由能原理，测试了应用于无神经生物过程的主动推理计算框架的预测。我们通过对绿藻进行具有不同光脉冲模式的趋光实验，测量它们的趋光偏差，以此作为它们优先检测和适应一种模式而非另一种模式的能力的读数，从而证明了绿藻中的模式识别。结果表明，绿藻更容易适应规则模式而非不规则模式，甚至表现出记忆特性，展现了基础智能的一个关键组成部分。基于药理学和电击的干预以及光适应模拟揭示了随机刺激如何通过群体旋转和钙介导的光感受器到鞭毛的信息传递的结构化动态相互作用干扰正常的光适应，这与不确定性最小化相一致。在无神经生物中检测到功能性不确定性最小化，将不确定性最小化等概念扩展到了神经元之外，并提供了适用于其他生物系统的见解和新型干预工具，类似于在更简单的神经生物中的早期学习验证。
