Perception-action Cycles Research Articles

An ecological psychology perspective in teaching Chinese online

Abstract In this paper, we first introduce three ecological principles for designing Chinese language online teaching and learning activities drawing from ecological psychology. The first principle – perception and action cycles in an ecosystem, advocates for situating online learning in a physical environment and creates activities with potentials to engage perception-action cycles. With the second principle – intention and attention merge in an ecosystem, we advocate for designing complex, flexible and dynamic activities and providing scaffolding for attunement. The third – meaning-making and values-realizing coincide in an ecosystem, calls for designing activities that allow for care-taking of oneself, each other, and the environment and cultivate curiosity and mindfulness for values-realizing and individual meaning-making. To highlight ways to tap into learners’ interest and connect to people and places, we then share two example activities with both low-technology and high-technology options.

Human uniqueness in using tools and artifacts: flexibility, variety, complexity

The main goal of this paper is to investigate whether (and how) humans are unique in using tools and artifacts. Non-human animals exhibit some impressive instances of tool and artifact-use. Chimpanzees use sticks to get termites out of a mound, beavers build dams, birds make nests, spiders create webs, bowerbirds make bowers to impress potential mates, etc. There is no doubt that some animals modify and use objects in clever and sophisticated ways. But how does this relate to the way in which humans make and use objects to achieve their goals? To answer this question, this paper first presents a taxonomy of artifacts, identifying four overlapping categories, namely embodied, perceptual, cognitive, and affective artifacts. It then discusses definitions of animal tool-use, arguing that we need a more liberal approach, one that goes beyond the use of tools that are embedded in occurrent perception-action cycles. This paper ends by analysing how instances of animal tool and artifact-use can be classified according to the four identified categories, concluding that some animals use embodied, perceptual, cognitive, and affective artifacts. In this sense, humans are thus not unique in the kinds of tools and artifacts we use. What is unique, however, is our unprecedented flexibility and openness to deeply incorporate a large variety of complex tools and artifacts into our embodied, perceptual, cognitive, and affective systems.

Motor development: from dynamical system approach to a neo-developmental application

The aim of this study was to present a dynamical view of motor development and a few factors affecting the developmental course and rate of motor changes. The acquisition of motor skills involves one’s exploration of many body segmental configurations, followed by the selection of those that are most useful to achieve the task demands. Exploration and selection are attained through searching and the evolution and dissolution of the stable dynamical regions of coordination modes for a given task. Development then involves an interaction between the learner and environment, dissolving relatively successful dynamics in search of even more rewarding dynamics for the new task, based upon repetitive perception-action cycles. Developmental changes can occur naturally, but can also be deterministically influenced by, for example, teachers enhancing the probability of systematic change in performance over time. This influence can be through the planned introduction of interventions and providing information that drives the learner to transition to stable modes of performance and search for new body configuration dynamics required for the new task. Teachers play an important role in guiding learners through this complex developmental journey.

The Development From Adaptive to Cognitive Radar Resource Management

This article has presented a walk along the cognitive spectrum in the context of radar resource management. It has shown how radar resource management techniques, developed over the last decades, have focused on increasing the complexity of perception-action cycles in order to increase the operational performance of the radar. These techniques can be built upon in the context of cognitive radar, for example, through mission-oriented QoS management or stochastic control. Although basic perception-action cycles exist in operational radars, there are a number of challenges and drawbacks that must be considered before a high level of cognition can be deployed. Care should be taken in the drive toward cognitive radars, as the ultimate goal is not to have a cognitive radar for the sake of being cognitive, but to increase the radar functionality and operational performance

Emotion as an emergent phenomenon of the neurocomputational energy regulation mechanism of a cognitive agent in a decision-making task

Biological agents need to complete perception-action cycles to perform various cognitive and biological tasks such as maximizing their wellbeing and their chances of genetic continuation. However, the processes performed in these cycles come at a cost. Such costs force the agent to evaluate a tradeoff between the optimality of the decision making and the time and computational effort required to make it. Several cognitive mechanisms that play critical roles in managing this tradeoff have been identified. These mechanisms include adaptation, learning, memory, attention, and planning. One of the often overlooked outcomes of these cognitive mechanisms, in spite of the critical effect that they may have on the perception-action cycle of organisms, is “emotion.” In this study, we hold that emotion can be considered as an emergent phenomenon of a plausible neurocomputational energy regulation mechanism, which generates an internal reward signal to minimize the neural energy consumption of a sequence of actions (decisions), where each action triggers a visual memory recall process. To realize an optimal action selection over a sequence of actions in a visual recalling task, we adopted a model-free reinforcement learning framework, in which the reward signal—that is, the cost—was based on the iteration steps of the convergence state of an associative memory network. The proposed mechanism has been implemented in simulation and on a robotic platform: the iCub humanoid robot. The results show that the computational energy regulation mechanism enables the agent to modulate its behavior to minimize the required neurocomputational energy in performing the visual recalling task.

Cognitive Risk Control for Anti-Jamming V2V Communications in Autonomous Vehicle Networks

The future of intelligent transportation system (ITS) is expected to be composed of connected and autonomous vehicles (CAVs), the development of which will have great impact on people's everyday life. Unfortunately, this progress will be accompanied by all kinds of potential threats and attacks rising in CAV network. As a legacy from traditional wireless networks, jamming attack is still one of the major and serious threats to vehicle-to-vehicle (V2V) communications. In this paper, we investigate the anti-jamming V2V communication in CAV networks through power control in conjunction with channel selection. Bringing into play a brain-inspired research tool called cognitive dynamic system (CDS), the general structure of cognitive risk control (CRC) is well-tailored to analyze and address the jamming problem. Specifically, power control is carried out first using reinforcement learning, the result of which is then examined by a module called task-switch control. Based on the risk assessment, a multi-armed bandit (MAB) problem is formulated to perform the channel-selection process when necessary. Through continuous perception-action cycles (PACs), the feature of predictive adaptation is realized for the legitimate vehicle in its behavioral interactions with the jammer. Simulation results have shown that the proposed method has desirable performance in terms of several evaluation metrics.

An Updated Model of Chronic Ankle Instability.

Lateral ankle sprains (LASs) are among the most common injuries incurred during participation in sport and physical activity, and it is estimated that up to 40% of individuals who experience a first-time LAS will develop chronic ankle instability (CAI). Chronic ankle instability is characterized by a patient's being more than 12 months removed from the initial LAS and exhibiting a propensity for recurrent ankle sprains, frequent episodes or perceptions of the ankle giving way, and persistent symptoms such as pain, swelling, limited motion, weakness, and diminished self-reported function. We present an updated model of CAI that aims to synthesize the current understanding of its causes and serves as a framework for the clinical assessment and rehabilitation of patients with LASs or CAI. Our goal was to describe how primary injury to the lateral ankle ligaments from an acute LAS may lead to a collection of interrelated pathomechanical, sensory-perceptual, and motor-behavioral impairments that influence a patient's clinical outcome. With an underpinning of the biopsychosocial model, the concepts of self-organization and perception-action cycles derived from dynamic systems theory and a patient-specific neurosignature, stemming from the Melzack neuromatrix of pain theory, are used to describe these interrelationships.

Hierarchical fully adaptive radar

By emulating the cognitive perception-action cycle believed to be at the core of animal cognition, cognitive radars promise to improve radar performance over standard systems. The fully adaptive radar (FAR) framework provides a generalised approach to implementing a single cognitive perception-action cycle for radar systems, but complex adaptive problems necessitate the interaction of multiple perception-action cycles. This study describes the general form of the hierarchical FAR (HFAR) framework. The HFAR framework is applied to a single-target tracking, sensor fusion problem, and real-time experimental results demonstrate the efficacy of the proposed architecture for handling problems of varying scales in a consistent, adaptive fashion.

Afterword: a view from enaction

Enaction is the process whereby a living organism brings about its own Umwelt or lived-world of experience. The prototype example is the “the world of the tick” as described by von Uexküll. This is a nice case of simplexity – achieving an impressive result by apparently simple means. Thus, the modest tick, blind and deaf and only capable of crawling slowly, achieves the task of catching a mammal – thousands of times bigger and faster – and getting to suck its blood. The means in question involve chaining three perception-action cycles: sensing butyric acid (emitted by the sweat glands of mammals) which causes the tick to drop; crawling on a rough surface until finding a smooth surface (in context, the bare skin of the mammal); and sucking a liquid at 37 °C underneath the surface (in context, the blood of the mammal). The notion of simplexity is attractive, especially to those biologists who still retain « a feeling for the organism ». However, the concept of simplexity has its limitations; it does not explain anything, it is essentially an appreciation which comes after the event. Indeed one might even go so far as to say that simplexity is not so much a solution, but rather a problem which itself requires explanation. After commenting on each of the texts in this volume, I found no substantial indications that “simplexity” does actually explain anything. My conclusion is that “simplexity”, far from being an explanation, is rather an alert flag for situations which are certainly interesting, but which are in need of further study and explanation. In this sense, “simplexity” is indeed a useful heuristic – although perhaps not in quite the way that was originally intended.

Young children’s exploration of the indoor playroom space in center-based childcare

Exploration plays a key role in the development of children. While exploring, children develop new skills by perceiving and acting upon the possibilities for action that are specified in the environment. This study examined the relations between young children’s exploration during free play and the spatial characteristics of the indoor playroom space in childcare centers, using an observation scheme based on Gibson’s theory of perception-action affordances. The study was carried out in mixed-age groups, with ages ranging between 11 and 48 months. The results showed that depth of exploration of space was positively related to the use of tables and activity centers, and also to the child’s task-orientation as rated by the caregivers. Breadth of exploration revealed a reversed pattern of relations. The findings indicate that studying children’s exploration of indoor playroom space as affordances-guided perception-action cycles can contribute to a more comprehensive understanding of the role of spatial attributes in children’s exploratory play.

Incremental change in the set of coactive cortical assemblies enables mental continuity

This opinion article explores how sustained neural firing in association areas allows high-order mental representations to be coactivated over multiple perception-action cycles, permitting sequential mental states to share overlapping content and thus be recursively interrelated. The term “state-spanning coactivity” (SSC) is introduced to refer to neural nodes that remain coactive as a group over a given period of time. SSC ensures that contextual groupings of goal or motor-relevant representations will demonstrate continuous activity over a delay period. It also allows potentially related representations to accumulate and coactivate despite delays between their initial appearances. The nodes that demonstrate SSC are a subset of the active representations from the previous state, and can act as referents to which newly introduced representations of succeeding states relate. Coactive nodes pool their spreading activity, converging on and activating new nodes, adding these to the remaining nodes from the previous state. Thus, the overall distribution of coactive nodes in cortical networks evolves gradually during contextual updating. The term “incremental change in state-spanning coactivity” (icSSC) is introduced to refer to this gradual evolution. Because a number of associated representations can be sustained continuously, each brain state is embedded recursively in the previous state, amounting to an iterative process that can implement learned algorithms to progress toward a complex result. The longer representations are sustained, the more successive mental states can share related content, exhibit progressive qualities, implement complex algorithms, and carry thematic or narrative continuity. Included is a discussion of the implications that SSC and icSSC may have for understanding working memory, defining consciousness, and constructing AI architectures.

Incremental Change in the Set of Coactive Cortical Assemblies Enables Mental Continuity

This opinion article explores how sustained neural firing in association areas allows high-order mental representations to be coactivated over multiple perception-action cycles, permitting sequential mental states to share overlapping content and thus be recursively interrelated. The term “state-spanning coactivity” (SSC) is introduced to refer to neural nodes that remain coactive as a group over a given period of time. SSC ensures that contextual groupings of goal or motor-relevant representations will demonstrate continuous activity over a delay period. It also allows potentially related representations to accumulate and coactivate despite delays between their initial appearances. The nodes that demonstrate SSC are a subset of the active representations from the previous state, and can act as referents to which newly introduced representations of succeeding states relate. Coactive nodes pool their spreading activity, converging on and activating new nodes, adding these to the remaining nodes from the previous state. Thus, the overall distribution of coactive nodes in cortical networks evolves gradually during contextual updating. The term “incremental change in state-spanning coactivity” (icSSC) is introduced to refer to this gradual evolution. Because a number of associated representations can be sustained continuously, each brain state is embedded recursively in the previous state, amounting to an iterative process that can implement learned algorithms to progress toward a complex result. The longer representations are sustained, the more successive mental states can share related content, exhibit progressive qualities, implement complex algorithms, and carry thematic or narrative continuity. Included is a discussion of the implications that SSC and icSSC may have for understanding working memory, defining consciousness, and constructing AI architectures.

Functional distance in human gait transition

The emerging understanding of the behavioral transitions that accompany the ascending and descending method of limits is in terms of “functional distance” — the degree to which a perceiver is disengaged from ordinary exploratory activities. Increasing functional distance results in negative hysteresis in contrast to the classical positive hysteresis more typical of ongoing activity. In the present study of human gait transitions on a treadmill, the functional distance between a perceiver and ordinary exploratory activities was manipulated in two ways: (1) “Active” participants, walking or running on a treadmill, were asked to anticipate the gait that would be required if treadmill speed were increased or decreased; and (2) “passive” participants, standing off a moving treadmill, were asked to report the gait they would use if they were on the treadmill at its current speed. As expected, the increase of functional distance from (1) to (2) reduced the amount of classical hysteresis and promoted negative hysteresis, that is, a lower transition speed for walk-to-run transitions (ascending trials) than for run-to-walk transitions (descending trials). These results complement empirical findings in other behavioral transition experiments. More broadly, they signify the role of perception-action cycles for grounding natural on-going perception. In particular, they support the assertion that perception and action are intertwined and that lack of information about an impending action has consequences for perceptual judgments.

Learning Plants: Semiosis Between the Parts and the Whole

In this article, I explore plant semiosis with a focus on plant learning. I distinguish between the scales and levels of learning conceivable in phytosemiosis, and identify organism-scale learning as the distinguishing question for plant semiosis. Since organism-scale learning depends on organism-scale semiosis, I critically review the arguments regarding whole-plant functional cycles. I conclude that they have largely relied on Uexkullian biases that have prevented an adequate interpretation of modern plant neurobiology. Through an examination of trophic growth in plant roots, I expose some conceptual difficulties in attributing functional cycles to whole-plants. I conclude that the mapping of resource areas in the root system is a learning activity requiring higher-scale sign activity than is possible at the cellular scale, strongly suggesting the presence of organism-scale functional cycles. I do, however, question whether all perception-action cycles in organisms are accompanied with organism-scale semiosis.

Perception-Action Cycles of Persons with Stroke Hemiplegia: Focus on the function of abdominal muscles

Perception-Action Cycles of Persons with Stroke Hemiplegia: Focus on the function of abdominal muscles

Unnerving Intelligence

A variety of definitions of intelligence have been offered that are not exclusively human-centric. One, in particular, is taken as a starting point: end-directed behavior marked by the making of meaningful distinctions made possible by perception–action cycles. Specific examples of end-directed behavior are examined for evidence of three aspects of the kind of agency embodied in perception–action cycles, namely, prospectivity, retrospectivity, and flexibility. The chosen examples of behavior are nicely layered but otherwise unremarkable. Nonetheless, they all have an unexpected twist that challenges the kinds of explanations of intelligence qua agency that are rooted in neurobiology.

Embodied Object Recognition using Adaptive Target Observations

In this paper, we study object recognition in the embodied setting. More specifically, we study the problem of whether the recognition system will benefit from acquiring another observation of the object under study, or whether it is time to give up, and report the observed object as unknown. We describe the hardware and software of a system that implements recognition and object permanence as two nested perception-action cycles. We have collected three data sets of observation sequences that allow us to perform controlled evaluation of the system behavior. Our recognition system uses a KNN classifier with bag-of-features prototypes. For this classifier, we have designed and compared three different uncertainty measures for target observation. These measures allow the system to (a) decide whether to continue to observe an object or to move on, and to (b) decide whether the observed object is previously seen or novel. The system is able to successfully reject all novel objects as “unknown”, while still recognizing most of the previously seen objects.

Schema generation in recurrent neural nets for intercepting a moving target

The grasping of a moving object requires the development of a motor strategy to anticipate the trajectory of the target and to compute an optimal course of interception. During the performance of perception-action cycles, a preprogrammed prototypical movement trajectory, a motor schema, may highly reduce the control load. Subjects were asked to hit a target that was moving along a circular path by means of a cursor. Randomized initial target positions and velocities were detected in the periphery of the eyes, resulting in a saccade toward the target. Even when the target disappeared, the eyes followed the target's anticipated course. The Gestalt of the trajectories was dependent on target velocity. The prediction capability of the motor schema was investigated by varying the visibility range of cursor and target. Motor schemata were determined to be of limited precision, and therefore visual feedback was continuously required to intercept the moving target. To intercept a target, the motor schema caused the hand to aim ahead and to adapt to the target trajectory. The control of cursor velocity determined the point of interception. From a modeling point of view, a neural network was developed that allowed the implementation of a motor schema interacting with feedback control in an iterative manner. The neural net of the Wilson type consists of an excitation-diffusion layer allowing the generation of a moving bubble. This activation bubble runs down an eye-centered motor schema and causes a planar arm model to move toward the target. A bubble provides local integration and straightening of the trajectory during repetitive moves. The schema adapts to task demands by learning and serves as forward controller. On the basis of these model considerations the principal problem of embedding motor schemata in generalized control strategies is discussed.

TOWARDS GRASP-ORIENTED VISUAL PERCEPTION FOR HUMANOID ROBOTS

A distinct property of robot vision systems is that they are embodied. Visual information is extracted for the purpose of moving in and interacting with the environment. Thus, different types of perception-action cycles need to be implemented and evaluated. In this paper, we study the problem of designing a vision system for the purpose of object grasping in everyday environments. This vision system is firstly targeted at the interaction with the world through recognition and grasping of objects and secondly at being an interface for the reasoning and planning module to the real world. The latter provides the vision system with a certain task that drives it and defines a specific context, i.e. search for or identify a certain object and analyze it for potential later manipulation. We deal with cases of: (i) known objects, (ii) objects similar to already known objects, and (iii) unknown objects. The perception-action cycle is connected to the reasoning system based on the idea of affordances. All three cases are also related to the state of the art and the terminology in the neuroscientific area.

Exploratory learning structures in artificial cognitive systems

The major goal of the COSPAL project is to develop an artificial cognitive system architecture, with the ability to autonomously extend its capabilities. Exploratory learning is one strategy that allows an extension of competences as provided by the environment of the system. Whereas classical learning methods aim at best for a parametric generalization, i.e., concluding from a number of samples of a problem class to the problem class itself, exploration aims at applying acquired competences to a new problem class, and to apply generalization on a conceptual level, resulting in new models. Incremental or online learning is a crucial requirement to perform exploratory learning. In the COSPAL project, we mainly investigate reinforcement-type learning methods for exploratory learning, and in this paper we focus on the organization of cognitive systems for efficient operation. Learning is used over the entire system. It is organized in the form of four nested loops, where the outermost loop reflects the user-reinforcement-feedback loop, the intermediate two loops switch between different solution modes at symbolic respectively sub-symbolic level, and the innermost loop performs the acquired competences in terms of perception–action cycles. We present a system diagram which explains this process in more detail. We discuss the learning strategy in terms of learning scenarios provided by the user. This interaction between user (’teacher’) and system is a major difference to classical robotics systems, where the system designer places his world model into the system. We believe that this is the key to extendable robust system behavior and successful interaction of humans and artificial cognitive systems. We furthermore address the issue of bootstrapping the system, and, in particular, the visual recognition module. We give some more in-depth details about our recognition method and how feedback from higher levels is implemented. The described system is however work in progress and no final results are available yet. The available preliminary results that we have achieved so far, clearly point towards a successful proof of the architecture concept.