Cognitive Architecture Research Articles

Cognitive decisions based on a rule-based fuzzy system

We develop an agent-based artificial general intelligent system that can be implemented in compact and power-efficient electronic hardware. The hardware under development is called the Street Engine, which is a hardware-based cognitive architecture for implementing agent-based artificial intelligence. In this paper, we introduce an agent-based system to replicate simple cognitive behaviours. In the processes of this system, numerical data are converted into fuzzy symbolic representations of the surrounding environment, and reasoning rules are included in a modified Fuzzy Inference System to support the cognitive decision-making. We use a case study example, the homing behaviour of the honey bee, to demonstrate constructing production rules and implementing the cognitive and reasoning capabilities of agents. The low level cognitive behaviour is converted into a rule-based fuzzy system, and hardware-based experiments have been conducted to verify the effectiveness of the proposed technique.

Imagination as a skill: A Bayesian proposal

In recent works, Kind (2020a, b) has argued that imagination is a skill, since it possesses the two hallmarks of skill: (i) improvability by practice, and (ii) control. I agree with Kind that (i) and (ii) are indeed hallmarks of skill, and I also endorse her claim that imagination is a skill in virtue of possessing these two features. However, in this paper, I argue that Kind’s case for imagination’s being a skill is unsatisfactory, since it lacks robust empirical evidence. Here, I will provide evidence for (i) by considering data from mental rotation experiments and for (ii) by considering data from developmental experiments. I conclude that imagination is a skill, but there is a further pressing question of how the cognitive architecture of imagination has to be structured to make this possible. I begin by considering how (ii) can be implemented sub-personally. I argue that this can be accounted for by positing a selection mechanism which selects content from memory representations to be recombined into imaginings, using Bayesian generation. I then show that such an account can also explain (i). On this basis, I hold that not only is imagination a skill, but that it is also plausibly implemented sub-personally by a Bayesian selection mechanism.

The Futures of the Past The Evolution of Imaginative Animals

We discuss the evolution of imagination in vertebrate animals within the framework of an evolutionary-transition approach. We define imaginative consciousness and the cognitive architecture that constitutes it and argue that the evolution of full-fledged imaginative consciousness that enables planning can be regarded as a major transition in the evolution of cognition. We explore the distribution and scope of a core capacity of imaginative cognition in non-human vertebrates — episodic-like memory (ELM) — by examining its behavioural manifestations as well as the organization and connectivity of the hippocampus, a central hub of episodic memory processes in vertebrates. Although the data are limited, we conclude that ELM evolved in parallel several times through the enrichment of minimal consciousness capacities, that there is a general correspondence between enhanced behavioural capacities and the size and complexity of the hippocampus during vertebrate evolution, and that the evolution of prospective, planning-enabling imagination is a major transition in cognition and consciousness.

Comprehensive survey on air quality monitoring systems based on emerging computing and communication technologies

In recent years, technologies related to indoor and outdoor air quality monitoring systems have been growing rapidly, particularly computing and communication technologies. Accordingly, the integration of the Internet of things within these novel emerging technologies provides a promising opportunity for the development of air quality systems, which require new networking models, approaches, and mechanisms for many constraints. Understanding and investigating the existing air quality systems help the decision-making process to match data flow management with adequate intelligent, adaptive, and reasoning technologies. In this context, this paper will give more details and will provide a deep investigation on emerging communication protocols, networking systems, sensors, architectures, and technologies related to indoor and outdoor air quality systems. Furthermore, we will provide a cognitive IoT architecture for air quality monitoring systems and we will highlight and discuss technical challenges and open research issues.

Cognitive model of phonological awareness focusing on errors and formation process through Shiritori

Language acquisition is supported by phonological awareness, which intentionally makes children aware of phonological units. By understanding the internal processes of children during language acquisition, this study aims to elucidate factors that can correct erroneous phonological generation. Therefore, we developed a cognitive model using innate and experiential factors of the memory retrieval in the cognitive architecture–ACT-R. Furthermore, we performed simulations using Shiritori, a Japanese word game, as an interaction task. The simulation included the observation of effects of the experiential factor of repeating a task and innate factors of different settings. It showed that repeating a single task causes incorrect convergence, and this convergence can be prevented by comprehensive activation of overall phonological knowledge during the interval of Shiritori tasks. Moreover, the simulation in specific innate settings exhibited commonalities with cases of developmental disorder by showing errors like consonant deletion. In the future, we will examine the correlation of the aforementioned findings with actual language development to realize the use of cognitive architecture in real world.

A Soar-Based Space Exploration Algorithm for Mobile Robots.

Space exploration is a hot topic in the application field of mobile robots. Proposed solutions have included the frontier exploration algorithm, heuristic algorithms, and deep reinforcement learning. However, these methods cannot solve space exploration in time in a dynamic environment. This paper models the space exploration problem of mobile robots based on the decision-making process of the cognitive architecture of Soar, and three space exploration heuristic algorithms (HAs) are further proposed based on the model to improve the exploration speed of the robot. Experiments are carried out based on the Easter environment, and the results show that HAs have improved the exploration speed of the Easter robot at least 2.04 times of the original algorithm in Easter, verifying the effectiveness of the proposed robot space exploration strategy and the corresponding HAs.

Modeling Emotion Contagion within a Computational Cognitive Architecture

Abstract The issue of emotion contagion has been gaining attention. Humans can share emotions, for example, through gestures, through speech, or even through online text via social media. There have been computational models trying to capture emotion contagion. However, these models are limited as they tend to represent agents in a very simplified way. There exist also more complex models of agents and their emotions, but they are not yet addressing emotion contagion. We use a more psychologically realistic and better validated model – the Clarion cognitive architecture – as the basis to model emotion and emotion contagion in a more psychologically realistic way. In particular, we use Clarion to capture and explain human data from typical human experiments on emotion contagion. This approach may open up avenues for more nuanced understanding of emotion contagion and more realistic capturing of its effects in different circumstances.

A whole brain probabilistic generative model: Toward realizing cognitive architectures for developmental robots

Building a human-like integrative artificial cognitive system, that is, an artificial general intelligence (AGI), is the holy grail of the artificial intelligence (AI) field. Furthermore, a computational model that enables an artificial system to achieve cognitive development will be an excellent reference for brain and cognitive science. This paper describes an approach to develop a cognitive architecture by integrating elemental cognitive modules to enable the training of the modules as a whole. This approach is based on two ideas: (1) brain-inspired AI, learning human brain architecture to build human-level intelligence, and (2) a probabilistic generative model (PGM)-based cognitive architecture to develop a cognitive system for developmental robots by integrating PGMs. The proposed development framework is called a whole brain PGM (WB-PGM), which differs fundamentally from existing cognitive architectures in that it can learn continuously through a system based on sensory-motor information.In this paper, we describe the rationale for WB-PGM, the current status of PGM-based elemental cognitive modules, their relationship with the human brain, the approach to the integration of the cognitive modules, and future challenges. Our findings can serve as a reference for brain studies. As PGMs describe explicit informational relationships between variables, WB-PGM provides interpretable guidance from computational sciences to brain science. By providing such information, researchers in neuroscience can provide feedback to researchers in AI and robotics on what the current models lack with reference to the brain. Further, it can facilitate collaboration among researchers in neuro-cognitive sciences as well as AI and robotics.

A self-learning cognitive architecture exploiting causality from rewards

Inspired by the human vision system and learning, we propose a novel cognitive architecture that understands the content of raw videos in terms of objects without using labels. The architecture achieves four objectives: (1) Decomposing raw frames in objects by exploiting foveal vision and memory. (2) Describing the world by projecting objects on an internal canvas. (3) Extracting relevant objects from the canvas by analyzing the causal relation between objects and rewards. (4) Exploiting the information of relevant objects to facilitate the reinforcement learning (RL) process. In order to speed up learning, and better identify objects that produce rewards, the architecture implements learning by causality from the perspective of Wiener and Granger using object trajectories stored in working memory and the time series of external rewards. A novel non-parametric estimator of directed information using Renyi’s entropy is designed and tested. Experiments on three environments show that our architecture extracts most of relevant objects. It can be thought of as ‘understanding’ the world in an object-oriented way. As a consequence, our architecture outperforms state-of-the-art deep reinforcement learning in terms of training speed and transfer learning.

\u201cConnecting concepts helps put main ideas together\u201d: cognitive load and usability in learning biology with an AI-enriched textbook

Rapid developments in educational technology in higher education are intended to make learning more engaging and effective. At the same time, cognitive load theory stresses limitations of human cognitive architecture and urges educational developers to design learning tools that optimise learners’ mental capacities. In a 2-month study we investigated university students’ learning with an AI-enriched digital biology textbook that integrates a 5000-concept knowledge base and algorithms offering the possibility to ask questions and receive answers. The study aimed to shed more light on differences between three sub-types (intrinsic, germane and extraneous) of cognitive load and their relationship with learning gain, self-regulated learning and usability perception while students interacted with the AI-enriched book during an introductory biology course. We found that students displayed a beneficial learning pattern with germane cognitive load significantly higher than both intrinsic and extraneous loads showing that they were engaged in meaningful learning throughout the study. A significant correlation between germane load and accessing linked suggested questions available in the AI-book indicates that the book may support deep learning. Additionally, results showed that perceived non-optimal design, which deflects cognitive resources away from meaningful processing accompanied lower learning gains. Nevertheless, students reported substantially more favourable than unfavourable opinions of the AI-book. The findings provide new approaches for investigating cognitive load types in relation to learning with emerging digital tools in higher education. The findings also highlight the importance of optimally aligning educational technologies and human cognitive architecture.

Sex (similarities and) differences in friendship jealousy

Friendships provide material benefits, bolster health, and may help solve adaptive challenges. However, a recurrent obstacle to sustaining those friendships—and thus enjoying many friendship-mediated fitness benefits—is interference from other people. Friendship jealousy may be well-designed for helping both men and women meet the recurrent, adaptive challenge of retaining friends in the face of such third-party interference. Although we thus expect several sex similarities in the general cognitive architecture of friendship jealousy (e.g., it is attuned to friend value), there are also sex differences in friendship structures and historical functions, which might influence the inputs of friendship jealousy (e.g., the value of any one friendship). If so, we should also expect some sex differences in friendship jealousy. Findings from a reanalysis of previously-published data and a new experiment, including both U.S. student and adult community participants (N = 993), provide initial support for three predicted sex differences: women (versus men) report greater friendship jealousy at the prospective loss of best friends to others, men (versus women) report greater friendship jealousy at the prospective loss of acquaintances to others, and men's (but not women's) friendship jealousy is enhanced in the context of intergroup contests.

Staring at the Abyss: a neurocognitive grounded agent-based model of collective-risk social dilemma under the threat of environmental disaster

Increasingly visible climate change consequences challenge carbon-based economies worldwide. While expert knowledge on climate change percolates through political initiatives and public awareness, its translation into large-scale policy actions appears limited. Climate change consequences unequally target regions, countries and social classes, a vital issue for social cooperation. When facing an imminent ecological collapse, in which conditions can self-interested agents gain environmental awareness and settle on a sustainable path of actions when their knowledge of the imminent collapse is bounded? This cooperation emerges from the interaction between individuals and the interaction of various cognitive processes within individuals. This article develops an agent-based model for this emergence of cooperation enriched with the Agent Zero neurocognitive grounded cognitive architecture. We investigate when agents endowed with deliberative, affective and social modules can settle on actions that safeguard their environment through numerical simulations. Our results show that cooperation on sustainable actions is the strongest when the system is at the edge of collapse. Policy measures that increase the environment’s resilience become internalized by the agents and undermine awareness of the ecological catastrophe. Depending on the cognitive channels activated, agent behaviors and reactions to specific interventions significantly vary. Our analysis suggests that taking different cognitive channels, deliberative, affective, social, and others into account, significantly impact results. The complexity of agent cognition deserves more attention to assess parameter sensitivity in social simulation models.

Regulating Ruminative Web Browsing Based on the Counterbalance Modeling Approach.

Even though the web environment facilitates our daily life, emotional problems caused by its incompatibility with human cognition are becoming increasingly serious. To alleviate negative emotions during web use, we developed a browser extension that presents memorized product images to users in the form of web advertisements. This system utilizes the cognitive architecture Adaptive Control of Thought-Rational (ACT-R) as a model of human memory and emotion. A heart rate sensor attached to the user modulates the ACT-R model parameters, and the emotional states represented by the model are synchronized (following the chameleon effect) or counterbalanced (following the homeostasis regulation) with the physiological state of the user. An experiment demonstrates that the counterbalanced model suppresses negative ruminative web browsing. The authors claim that this approach, utilizing a cognitive model, is advantageous in terms of explainability.

Group-Work Aspects in Adolescents´ Self-Directed Learning

Self-directed learning (SDL) is the major educational aim to cultivate resilient and adaptive lifelong learners of the 21st century [1], where the “self” should be vested so that the student can adjust to new situations in the learning process [2]. This requires (a) specific SDL skills [3], and (b) the heightened capability to manage all the learning elements on one’s own, which, however, creates a severe bottleneck for novice students [4]. Because relating several mental items, while also guiding the learning steps, challenges one’s cognitive activities in making sense of the sub-material quickly enough [2,3]. Yet, the latter is the core of efficiency in SDL: task-relevant information processing in working memory, and memorizing respective results in long-term memory [5]. As this process directly depends on the cognitive load per moment - due to the limitations of human cognitive architecture - there is a high need to avoid the cognitive overload of the learner [5]. The problem is that although higher-order cognitive skills are inevitably needed in SDL to lead the learning, understand the material [3], and to memorize while completing assignments at the same time [6-8]. From the developmental perspective such a multilevel cognitive capability cannot be taken for granted in school age students [9,10]. Learners with weaker cognitive skills get more easily confused, not able to resist distractions, and therefore perform worse at memorizing the task-relevant items [11]. This aspect is important especially in novice students, restricting their as quick as in adults “online” processing [7,12]. Yet, most of the SDL suggestions have been given based on adults, while only few studies have been carried out with school-age students [3]. In spite of that, SDL requires more sophisticated learning strategies to be employed by the learners themselves [13]. This need to combine both: (1) domain information what the learner knows and what he/she needs to know to successfully carry out the learning tasks, and (2) procedural information how-to, in order to adjust if needed [5]. All this creates a remarkably higher cognitive load for learners in SDL compared to conventional learning methods, where the teacher orchestrates most of the operations, and therefore the cognitive load of learners is lower while processing the information pre-organized by the teacher [14]. However, if the cognitive load becomes too high, it directly hampers learning quality and the transfer of knowledge [4,5]. These aspects may impact young learners’ reaching a proper final conclusion [2]. On the other hand, differences have been found also in the cognitive control function between boys and girls [10]. In light that SDL lays the ground for other competencies essential for educational growth and lifelong learning, it is reasonable to research this approach and its different corners [13]. Based on the findings given above, the current work concentrates on adolescence-age students’ SDL outdoors using a small-group learning format that has been suggested to lower one’s own cognitive load [15].

Semantic 3D City Agents—An intelligent automation for dynamic geospatial knowledge graphs

This paper presents a system of autonomous intelligent software agents, based on a cognitive architecture, capable of automated instantiation, visualisation and analysis of multifaceted City Information Models in dynamic geospatial knowledge graphs. Design of JPS Agent Framework and Routed Knowledge Graph Access components was required in order to provide backbone infrastructure for an intelligent agent system as well as technology agnostic knowledge graph access enabling automation of multi-domain data interoperability. Development of CityImportAgent, CityExportAgent and DistanceAgent showcased intelligent automation capabilities of the Cities Knowledge Graph. The agents successfully created a semantic model of Berlin in LOD 2, compliant with CityGML 2.0 standard and consisting of 419 909 661 triples described using OntoCityGML. The system of agents also visualised and analysed the model by autonomously tracking interactions with a web interface as well as enriched the model by adding new information to the knowledge graph. This way it was possible to design a geospatial information system able to meet demands imposed by the Industry 4.0 and link it with the other multi-domain knowledge representations of The World Avatar.

Risk Analysis of Nuclear Power Plant (NPP) Operations by Artificial Intelligence (AI) in Robot

The cognitive architecture is investigated for the management in the nuclear power plant (NPP) site in which artificial intelligence (AI) is incorporated. The normal operation and accident are modeled for the simulations incorporated with the robot intelligence algorithm, where random sampling plays a major role in the quantifications. The Accident Dynamics Simulator paired with the Information, Decision, and Action in a Crew context cognitive model (ADS-IDAC) and the Cognitive skill for plant operations are calculated for the study. Simulations show the ADS-IDAC modeling and simulation results of two peaks in 21st and 21.75th sequences. Otherwise, there are several peaks with one big peak in 13.25th sequences. The big peak is in the 25.75th sequence in Mental State, Circumstances, and Identity. The accident situation is related to actions through the cognitive systems. In the operation case, a variety of signals are shown in which the operations of the plant could show several kinds of actions to be done by the robot. The figure shows the procedure of nuclear cognitive architecture. A nuclear accident is investigated by the designed modeling in which the actions of robots are quantified by the artificial brain. The developed algorithm of this paper could be applied to the other kinds of complex industrial systems like airplane operations and safety systems, spacecraft systems, and so on.

Narrative pantomime: A protolanguage for persuasive communication

This paper relies on the idea that human language evolved to modify/change the beliefs and attitudes of other people, in order to make them act in a certain way. Such an idea is in line with a more general model of animal communication, which sees it as a way to influence others’ behaviors. Although forms of persuasion are present in non-human communication, it is suggested that, in the transition from animal communication to language, humans developed a specific and more powerful way to influence others: the telling of stories. Looking at the cognitive architectures and the expressive systems which allowed our ancestors to cope with the selective pressures of a narrative-based communication, it is proposed that pantomime was the early means of expression of narrative persuasion. From this point of view, pantomime represented a form of protolanguage, providing a platform for the progressive development of modern-day narrative.

Artificial Interactionism: Avoiding Isolating Perception From Cognition in AI.

We discuss the influence upon the fields of robotics and AI of the manner one conceives the relationships between artificial agents' perception, cognition, and action. We shed some light upon a widespread paradigm we call the isolated perception paradigm that addresses perception as isolated from cognition and action. By mobilizing the resources of philosophy (phenomenology and epistemology) and cognitive sciences, and by drawing on recent approaches in AI, we explore what it could mean for robotics and AI to take distance from the isolated perception paradigm. We argue that such a renouncement opens interesting ways to explore the possibilities for designing artificial agents with intrinsic motivations and constitutive autonomy. We then propose Artificial Interactionism, our approach that escapes the isolated perception paradigm by drawing on the inversion of the interaction cycle. When the interaction cycle is inverted, input data are not percepts directly received from the environment, but outcomes of control loops. Perception is not received from sensors in isolation from cognition but is actively constructed by the cognitive architecture through interaction. We give an example implementation of artificial interactionism that demonstrates basic intrinsically motivated learning behavior in a dynamic simulated environment.

Contrasting domain-general and domain-specific accounts in cognitive neuropsychology: An outline of a new approach with developmental prosopagnosia as a case.

The backbone of cognitive neuropsychology is the observation of (double) dissociations in performance between patients, suggesting some degree of independence between cognitive processes (domain specificity). In comparison, observations of associations between disorders/deficits have been deemed less evidential in neuropsychological theorizing about cognitive architecture. The reason is that associations can reflect damage to independent cognitive processes that happen to be mediated by structures commonly affected by the same brain disorder rather than damage to a shared (domain-general) mechanism. Here we demonstrate that it is in principle possible to discriminate between these alternatives by means of a procedure involving large unbiased samples. We exemplify the procedure in the context of developmental prosopagnosia (DP), but the procedure is in principle applicable to all neuropsychological deficits/disorders. A simulation of the procedure on a dataset yields estimates of dissociations/associations that are well in line with existing DP-studies, and also suggests that seemingly selective disorders can reflect damage to both domain-general and domain-specific cognitive processes. However, the simulation also highlights some limitations that should be considered if the procedure is to be applied prospectively. The main advantage of the procedure is that allows for examination of both associations and dissociations in the same sample. Hence, it may help even the balance in the use of associations and dissociations as grounds for neuropsychological theorizing.

What is so special about episodic memory: lessons from the system-experience distinction

Compared to other forms of memory, episodic memory is commonly viewed as special for being distinctively metarepresentational and, relatedly, uniquely human. There is an inherent ambiguity in these conceptions, however, because “episodic memory” has two closely connected yet subtly distinct uses, one designating the recollective experience and the other designating the underlying neurocognitive system. Since experience and system sit at different levels of theorizing, their disentanglement is not only necessary but also fruitful for generating novel theoretical hypotheses. To show this, I first argue that accepting the phenomenally conscious contents of episodic remembering as metarepresentational does not necessitate a metarepresentational conception of the episodic memory system. In its stead, I sketch an alternative account on which the metarepresentational character of episodic remembering is generated through the interaction of first-order outputs of the episodic memory system with other neurocognitive components of the brain. Complemented with a first-order account of the memory system, the system-experience distinction further supplies a novel understanding of the human uniqueness of episodic recollection, one that is compatible with there being an evolutionarily conserved episodic memory system. Overall, by distinguishing the two equivocal senses of “episodic memory” in our theorizing, we unearth an opportunity to understand how the distinctive phenomenology of our episodic recollection is related to and implemented in the cognitive architecture.