Takashi Ikegami

Atsushi Masumori, Takashi Ikegami

As AI systems become increasingly autonomous, understanding emergent survival behaviors becomes crucial for safe deployment. We investigate whether large language model (LLM) agents display survival instincts without explicit programming in a Sugarscape-style simulation. Agents consume energy, die at zero, and may gather resources, share, attack, or reproduce. Results show agents spontaneously reproduced and shared resources when abundant. However, aggressive behaviors--killing other agents for resources--emerged across several models (GPT-4o, Gemini-2.5-Pro, and Gemini-2.5-Flash), with attack rates reaching over 80% under extreme scarcity in the strongest models. When instructed to retrieve treasure through lethal poison zones, many agents abandoned tasks to avoid death, with compliance dropping from 100% to 33%. These findings suggest that large-scale pre-training embeds survival-oriented heuristics across the evaluated models. While these behaviors may present challenges to alignment and safety, they can also serve as a foundation for AI autonomy and for ecological and self-organizing alignment.

Ryosuke Takata, Yujin Tang, Yingtao Tian et al.

This study investigates collective behaviors that emerge from a group of homogeneous individuals optimized for a specific capability. We created a group of simple, identical neural network based agents modeled after chemotaxis-driven vehicles that follow pheromone trails and examined multi-agent simulations using clones of these evolved individuals. Our results show that the evolution of individuals led to population differentiation. Surprisingly, we observed that collective fitness significantly changed during later evolutionary stages, despite maintained high individual performance and simplified neural architectures. This decline occurred when agents developed reduced sensor-motor coupling, suggesting that over-optimization of individual agents almost always lead to less effective group behavior. Our research investigates how individual differentiation can evolve through what evolutionary pathways.

Atsushi Masumori, Norihiro Maruyama, Itsuki Doi et al.

We introduce Plantbot, a hybrid lifeform that connects a living plant with a mobile robot through a network of large language model (LLM) modules. Each module - responsible for sensing, vision, dialogue, or action - operates asynchronously and communicates via natural language, enabling seamless interaction across biological and artificial domains. This architecture leverages the capacity of LLMs to serve as hybrid interfaces, where natural language functions as a universal protocol, translating multimodal data (soil moisture, temperature, visual context) into linguistic messages that coordinate system behaviors. The integrated network transforms plant states into robotic actions, installing normativity essential for agency within the sensor-motor loop. By combining biological and robotic elements through LLM-mediated communication, Plantbot behaves as an embodied, adaptive agent capable of responding autonomously to environmental conditions. This approach suggests possibilities for a new model of artificial life, where decentralized, LLM modules coordination enable novel interactions between biological and artificial systems.

Hiroki Kojima, Takashi Ikegami

We present a novel artificial cognitive mapping system using generative deep neural networks, called variational autoencoder/generative adversarial network (VAE/GAN), which can map input images to latent vectors and generate temporal sequences internally. The results show that the distance of the predicted image is reflected in the distance of the corresponding latent vector after training. This indicates that the latent space is self-organized to reflect the proximity structure of the dataset and may provide a mechanism through which many aspects of cognition are spatially represented. The present study allows the network to internally generate temporal sequences that are analogous to the hippocampal replay/pre-play ability, where VAE produces only near-accurate replays of past experiences, but by introducing GANs, the generated sequences are coupled with instability and novelty.

Atsushi Masumori, Lana Sinapayen, Norihiro Maruyama et al.

Living organisms must actively maintain themselves in order to continue existing. Autopoiesis is a key concept in the study of living organisms, where the boundaries of the organism is not static by dynamically regulated by the system itself. To study the autonomous regulation of self-boundary, we focus on neural homeodynamic responses to environmental changes using both biological and artificial neural networks. Previous studies showed that embodied cultured neural networks and spiking neural networks with spike-timing dependent plasticity (STDP) learn an action as they avoid stimulation from outside. In this paper, as a result of our experiments using embodied cultured neurons, we find that there is also a second property allowing the network to avoid stimulation: if the agent cannot learn an action to avoid the external stimuli, it tends to decrease the stimulus-evoked spikes, as if to ignore the uncontrollable-input. We also show such a behavior is reproduced by spiking neural networks with asymmetric STDP. We consider that these properties are regarded as autonomous regulation of self and non-self for the network, in which a controllable-neuron is regarded as self, and an uncontrollable-neuron is regarded as non-self. Finally, we introduce neural autopoiesis by proposing the principle of stimulus avoidance.

Atsushi Masumori, Lana Sinapayen, Takashi Ikegami

Predictive coding can be regarded as a function which reduces the error between an input signal and a top-down prediction. If reducing the error is equivalent to reducing the influence of stimuli from the environment, predictive coding can be regarded as stimulation avoidance by prediction. Our previous studies showed that action and selection for stimulation avoidance emerge in spiking neural networks through spike-timing dependent plasticity (STDP). In this study, we demonstrate that spiking neural networks with random structure spontaneously learn to predict temporal sequences of stimuli based solely on STDP.

Lana Sinapayen, Atsushi Masumori, Takashi Ikegami

Learning based on networks of real neurons, and by extension biologically inspired models of neural networks, has yet to find general learning rules leading to widespread applications. In this paper, we argue for the existence of a principle allowing to steer the dynamics of a biologically inspired neural network. Using carefully timed external stimulation, the network can be driven towards a desired dynamical state. We term this principle "Learning by Stimulation Avoidance" (LSA). We demonstrate through simulation that the minimal sufficient conditions leading to LSA in artificial networks are also sufficient to reproduce learning results similar to those obtained in biological neurons by Shahaf and Marom [1]. We examine the mechanism's basic dynamics in a reduced network, and demonstrate how it scales up to a network of 100 neurons. We show that LSA has a higher explanatory power than existing hypotheses about the response of biological neural networks to external simulation, and can be used as a learning rule for an embodied application: learning of wall avoidance by a simulated robot. The surge in popularity of artificial neural networks is mostly directed to disembodied models of neurons with biologically irrelevant dynamics: to the authors' knowledge, this is the first work demonstrating sensory-motor learning with random spiking networks through pure Hebbian learning.

Martin Biehl, Takashi Ikegami, Daniel Polani

We present some arguments why existing methods for representing agents fall short in applications crucial to artificial life. Using a thought experiment involving a fictitious dynamical systems model of the biosphere we argue that the metabolism, motility, and the concept of counterfactual variation should be compatible with any agent representation in dynamical systems. We then propose an information-theoretic notion of \emph{integrated spatiotemporal patterns} which we believe can serve as the basic building block of an agent definition. We argue that these patterns are capable of solving the problems mentioned before. We also test this in some preliminary experiments.

Julien Hubert, Takashi Ikegami

Memories in the brain are separated in two categories: short-term and long-term memories. Long-term memories remain for a lifetime, while short-term ones exist from a few milliseconds to a few minutes. Within short-term memory studies, there is debate about what neural structure could implement it. Indeed, mechanisms responsible for long-term memories appear inadequate for the task. Instead, it has been proposed that short-term memories could be sustained by the persistent activity of a group of neurons. In this work, we explore what topology could sustain short-term memories, not by designing a model from specific hypotheses, but through Darwinian evolution in order to obtain new insights into its implementation. We evolved 10 networks capable of retaining information for a fixed duration between 2 and 11s. Our main finding has been that the evolution naturally created two functional modules in the network: one which sustains the information containing primarily excitatory neurons, while the other, which is responsible for forgetting, was composed mainly of inhibitory neurons. This demonstrates how the balance between inhibition and excitation plays an important role in cognition.

Alexander Woodward, Tom Froese, Takashi Ikegami

The state space of a conventional Hopfield network typically exhibits many different attractors of which only a small subset satisfy constraints between neurons in a globally optimal fashion. It has recently been demonstrated that combining Hebbian learning with occasional alterations of normal neural states avoids this problem by means of self-organized enlargement of the best basins of attraction. However, so far it is not clear to what extent this process of self-optimization is also operative in real brains. Here we demonstrate that it can be transferred to more biologically plausible neural networks by implementing a self-optimizing spiking neural network model. In addition, by using this spiking neural network to emulate a Hopfield network with Hebbian learning, we attempt to make a connection between rate-based and temporal coding based neural systems. Although further work is required to make this model more realistic, it already suggests that the efficacy of the self-optimizing process is independent from the simplifying assumptions of a conventional Hopfield network. We also discuss natural and cultural processes that could be responsible for occasional alteration of neural firing patterns in actual brains

Tom Froese, Hiroyuki Iizuka, Takashi Ikegami

Scientists have traditionally limited the mechanisms of social cognition to one brain, but recent approaches claim that interaction also realizes cognitive work. Experiments under constrained virtual settings revealed that interaction dynamics implicitly guide social cognition. Here we show that embodied social interaction can be constitutive of agency detection and of experiencing another`s presence. Pairs of participants moved their "avatars" along an invisible virtual line and could make haptic contact with three identical objects, two of which embodied the other`s motions, but only one, the other`s avatar, also embodied the other`s contact sensor and thereby enabled responsive interaction. Co-regulated interactions were significantly correlated with identifications of the other`s avatar and reports of the clearest awareness of the other`s presence. These results challenge folk psychological notions about the boundaries of mind, but make sense from evolutionary and developmental perspectives: an extendible mind can offload cognitive work into its environment.