The enactive approach to cognitive science has undergone a bio-phenomenologically inspired “normative turn” by characterizing an organism’s activity as motivated by intrinsic value, where this value is grounded in adaptive self-production under precarious conditions. However, efforts in the field of artificial life to model this enactive conception of life have unwittingly revealed a case of what can be called the hard problem of efficacy (HPE) : how could any intrinsic value as such make an effective difference to an organism’s behavior, in particular if bodily activity is purely determined by valueless material-organizational factors? First, this theoretical challenge of the HPE is formulated in the context of the enactive account of motivated activity. Then, by critically analyzing Schrödinger’s work on the methodological principles that define the scientific world image, it is argued that they can be revised to allow solutions to the HPE. This involves placing a limit on Schrödinger’s principle of understandability. The key move is to operationalize this limit with the concept of irruption : an organism’s motivations can make a physical difference to its bodily activity, but only indeterminately so, akin to a breakdown of its material-organizational constraints. Irruptions can thereby indirectly facilitate behavior-switching as well as long-term self-organization of adaptive behavior. Finally, it is proposed that the efficacy of motivated activity has its own specific energy cost due to the disordering effect of irruptions, which provides a new perspective on agency and the notion of mental work.

Donald Hebb proposed in his 1949 book The Organization of behavior that cell assemblies organized by temporally-asymmetric excitation form the basis of cognition. This basic idea has inspired a large body of research in neuroscience, and to a lesser extent in artificial intelligence. The modern manifestation of Hebb's principle is Spike-Timing Dependent Plasticity (STDP), and though we have a large body of experimental work investigating STDP, there is still little understanding of how networks of spiking neurons organize themselves into complex functional circuits, even though some progress has been made with models such as Liquid State Machines. Networks popular in artificial intelligence (e.g. MLPs) and in artificial life (e.g. CTRNNs) tend to eschew Hebb's insight and use error-backpropagation by gradient descent, in the case of AI, or an a-temporal Hebbian learning rule based on the outer product of neural activities, in the case of AL. Both of these approaches have greater interpretability than Spiking Neural Networks (SNNs), but both lack the mechanism that Hebb claimed was fundamental to cognition. This paper proposes to use complex-valued neurons (CVNs) to address this limitation, simultaneously promoting biological interpretation and computational tractability. The CVNs encode the firing rate and spike-time of a spiking neuron in the magnitude and angle, respectively, of a complex number. We also introduce an unsupervised piecewise-linear STDP learning rule compatible with CVNs, which for brevity we call complex-valued STDP (CVSTDP).We demonstrate both learning through error-backpropagation, and the spontaneous formation and dissolution of cell assemblies via the CVSTDP rule.

We propose to designate as dynamic interactive artificial intelligence (dAI) a cross-section of existing work in artificially designed and artificially evolved systems meant for minimal forms of interaction with human users. This approach borrows principles from artificial life and human movement science to avoid pitfalls of traditional AI. Counter to tradition, it prioritizes user-machine inter-dependence over autonomy. It starts small and relies on incremental growth instead of trying to implement advanced complete functionality. It assumes a perceptual ontology founded on movement coordination rather than object classification. Its development process is better described as reverse self-organization rather than reverse engineering. dAI can be viewed as a precursor to or pre-condition for enactive AI and an alternative to traditional frameworks grounded on information representation. We then give examples from our work in human movement science where we have used minimal dynamic interactive agents to induce specific beneficial effects in human participants’ movement skills. We also show how dAI can be exploited by both connectionist and symbolic AI.

We provide conceptual clues for one promising Artificial Life (ALife) route to Artificial Intelligence (AI) based on the notion of habit. We draw from an enactive approach that considers habits as the building blocks for mental life and, consequently, as the foundation for a science of mind. By taking this standpoint, this approach departs from the conventional view of intelligence in AI, which is based on “higher-order” cognitive functions. The first part of the paper addresses the idea of taking habits as the foundation for modeling intelligent behavior. This requires us to consider the so-called “scaling up” problem and rethink the concept of intelligence that still pervades in mainstream cognitive science. In the second part, we present the enactive approach to habits, emphasizing their adaptive and complex nature, as well as their fundamental role in guiding behavior. Finally, we acknowledge some limitations in the current enactive models of habits: either they are disembodied and decoupled, but allow for a rich landscape of attractors, or they are embodied and coupled, but remain too minimal. We propose a bridge between existing models and point to the need to go beyond the individual to include a social domain. We conclude that to better model intelligent behavior, embodied and situated agents must be capable of developing an increasingly complex network of habits from which an intelligent self emerges.

Social network analysis and agent-based modeling are two approaches used to study biological and artificial multi-agent systems. However, so far there is little work integrating these two approaches. Here we present a first step toward integration. We developed a novel approach that allows the creation of a social network on the basis of measures of interactions in an agent-based model for purposes of social network analysis. We illustrate this approach by applying it to a minimalist case study in swarm robotics loosely inspired by ant foraging behavior. For simplicity, we measured a network’s inter-agent connection weights as the total number of interactions between mobile agents. This measure allowed us to construct weighted directed networks from the simulation results. We then applied standard methods from social network analysis, specifically focusing on node centralities, to find out which are the most influential nodes in the network. This revealed that task allocation emerges and induces two classes of agents, namely foragers and loafers, and that their relative frequency depends on food availability. This finding is consistent with the behavioral analysis, thereby showing the compatibility of these two approaches.

It has recently been demonstrated that a Hopfield neural network that learns its own attractor configurations, for instance by repeatedly resetting the network to an arbitrary state and applying Hebbian learning after convergence, is able to form an associative memory of its attractors and thereby facilitate future convergences on better attractors. This process of structural self-optimization has so far only been demonstrated on relatively small artificial neural networks with random or highly regular and constrained topologies, and it remains an open question to what extent it can be generalized to more biologically realistic topologies. In this work, we therefore test this process by running it on the connectome of the widely studied nematode worm, C. elegans , the only living being whose neural system has been mapped in its entirety. Our results demonstrate, for the first time, that the self-optimization process can be generalized to bigger and biologically plausible networks. We conclude by speculating that the reset-convergence mechanism could find a biological equivalent in the sleep-wake cycle in C. elegans .

Referential communication is a ”representation-hungry” behavior, and the bee waggle dance is a classical example of referential communication in nature. We used an evolutionary robotics approach to create a simulation model of a minimalist example of this situation. Two structurally identical agents engage in embodied interaction such that one of them can find a distant target in 2D space that only the other could perceive. This is a challenging task: during their interaction the agents must disambiguate translational and communicative movements, allocate distinct behavioral roles (sender versus receiver), and switch behaviors from communicative to target seeking behavior. We found an evolutionary convention with compositionality akin to the waggle dance, correlating duration and angle of interaction with distance and angle to target, respectively. We propose that this behavior is more appropriately described as interactive mindshaping, rather than as the transfer of informational content.

Artificial life has been developing a behavior-based perspective on the origins of life, which emphasizes the adaptive potential of agent-environment interaction even at that initial stage. So far this perspective has been closely aligned to metabolism-first theories, while most researchers who study life’s origins tend to assign an essential role to RNA. An outstanding challenge is to show that a behavior-based perspective can also address open questions related to the genetic system. Accordingly, we have recently applied this perspective to one of science’s most fascinating mysteries: the origins of the standard genetic code. We modeled horizontal transfer of cellular components in a population of protocells using an iterated learning approach and found that it can account for the emergence of several key properties of the standard code. Here we further investigated the diachronic emergence of artificial codes and discovered that the model’s most frequent sequence of amino acid assignments overlaps significantly with the predictions in the literature. Our explorations of the factors that favor early incorporation into an emerging artificial code revealed two aspects: an amino acid’s relative probability of horizontal transfer, and its relative ease of discriminability in chemical space.

Dexterous assistive devices constitute one of the frontiers for hybrid human-machine systems. Manipulating unstable systems requires task-specific anticipatory dynamics. Learning this dynamics is more difficult when tasks, such as carrying liquid or riding a horse, produce unpredictable, irregular patterns of feedback and have hidden dimensions not projected as sensory feedback. We addressed the issue of coordination with complex systems producing irregular behaviour, with the assumption that mutual coordination allows for non-periodic processes to synchronize and in doing so to become regular. Chaos control gives formal expression to this: chaos can be stabilized onto periodic trajectories provided that the structure of the driving input takes into account the causal structure of the controlled system. Can we learn chaos control in a sensorimotor task? Three groups practiced an auditory-motor synchronization task by matching their continuously sonified hand movements to sonified tutors: a sinusoid served as a Non-Interactive Predictable tutor (NI-P), a chaotic system stood for a Non-Interactive Unpredictable tutor (NI-U), and the same system weakly driven by the participant’s movement stood for an Interactive Unpredictable tutor (I-U). We found that synchronization, dynamic similarity, and causal interaction increased with practice in I-U. Our findings have implications for current efforts to find more adequate ways of controlling complex adaptive systems.

Cooperation in scale-free networks has proven to be very robust against removal of randomly selected nodes (error) but highly sensitive to removal of the most connected nodes (attack). In this paper we analyze two comparable types of node removal in which the removal selection is based on tournaments where the fittest (raids) or the least fit (battles) nodes are chosen. We associate the two removals to two types of Maya warfare offences during the Classic period. During this period of at least 500 years, political leaders were able to sustain social order in spite of attack-like offences to their social networks. We present a computational model with a population fluctuation mechanism that operates under an evolutionary game theoretic approach using the Prisoner's Dilemma as a metaphor of cooperation. We find that paradoxically battles are able to uphold cooperation under moderate levels of raids, although raids do have a strong impact on the network structure. We infer that cooperation does not depend as much on the structure as it does on the underlying mechanism that allows the network to readjust. We relate the results to the Maya Classic period, concluding that Mayan warfare by itself cannot entirely explain the Maya political collapse without appealing to other factors that increased the pressures against cooperation.