Primitive Behaviors Research Articles

Pan-cortical 2-photon mesoscopic imaging and neurobehavioral alignment in awake, behaving mice.

The flow of neural activity across the neocortex during active sensory discrimination is constrained by task-specific cognitive demands, movements, and internal states. During behavior, the brain appears to sample from a broad repertoire of activation motifs. Understanding how these patterns of local and global activity are selected in relation to both spontaneous and task-dependent behavior requires in-depth study of densely sampled activity at single neuron resolution across large regions of cortex. In a significant advance toward this goal, we developed procedures to record mesoscale 2-photon Ca2+ imaging data from two novel in vivo preparations that, between them, allow for simultaneous access to nearly all 0f the mouse dorsal and lateral neocortex. As a proof of principle, we aligned neural activity with both behavioral primitives and high-level motifs to reveal the existence of large populations of neurons that coordinated their activity across cortical areas with spontaneous changes in movement and/or arousal. The methods we detail here facilitate the identification and exploration of widespread, spatially heterogeneous neural ensembles whose activity is related to diverse aspects of behavior.

Pan-cortical 2-photon mesoscopic imaging and neurobehavioral alignment in awake, behaving mice

The flow of neural activity across the neocortex during active sensory discrimination is constrained by task-specific cognitive demands, movements, and internal states. During behavior, the brain appears to sample from a broad repertoire of activation motifs. Understanding how these patterns of local and global activity are selected in relation to both spontaneous and task-dependent behavior requires in-depth study of densely sampled activity at single neuron resolution across large regions of cortex. In a significant advance toward this goal, we developed procedures to record mesoscale 2-photon Ca2+ imaging data from two novel in vivo preparations that, between them, allow for simultaneous access to nearly all 0f the mouse dorsal and lateral neocortex. As a proof of principle, we aligned neural activity with both behavioral primitives and high-level motifs to reveal the existence of large populations of neurons that coordinated their activity across cortical areas with spontaneous changes in movement and/or arousal. The methods we detail here facilitate the identification and exploration of widespread, spatially heterogeneous neural ensembles whose activity is related to diverse aspects of behavior.

Motion planning for off‐road autonomous driving based on human‐like cognition and weight adaptation

AbstractDriving in an off‐road environment is challenging for autonomous vehicles due to the complex and varied terrain. To ensure stable and efficient travel, the vehicle requires consideration and balancing of environmental factors, such as undulations, roughness, and obstacles, to generate optimal trajectories that can adapt to changing scenarios. However, traditional motion planners often utilize a fixed cost function for trajectory optimization, making it difficult to adapt to different driving strategies in challenging irregular terrains and uncommon scenarios. To address these issues, we propose an adaptive motion planner based on human‐like cognition and cost evaluation for off‐road driving. First, we construct a multilayer map describing different features of off‐road terrains, including terrain elevation, roughness, obstacle, and artificial potential field map. Subsequently, we employ a convolutional neural network‐long short‐term memory network to learn the trajectories planned by human drivers in various off‐road scenarios. Then, based on human‐like generated trajectories in different environments, we design a primitive‐based trajectory planner that aims to mimic human trajectories and cost weight selection, generating trajectories that are consistent with the dynamics of off‐road vehicles. Finally, we compute optimal cost weights and select and extend behavioral primitives to generate highly adaptive, stable, and efficient trajectories. We validate the effectiveness of the proposed method through experiments in a desert off‐road environment with complex terrain and varying road conditions. The experimental results show that the proposed human‐like motion planner has excellent adaptability to different off‐road conditions. It shows real‐time operation, greater stability, and more human‐like planning ability in diverse and challenging scenarios.

Unsupervised Object Interaction Learning with Counterfactual Dynamics Models

We present COIL (Counterfactual Object Interaction Learning), a novel way of learning skills of object interactions on entity-centric environments. The goal is to learn primitive behaviors that can induce interactions without external reward or any supervision. Existing skill discovery methods are limited to locomotion, simple navigation tasks, or single-object manipulation tasks, mostly not inducing interaction between objects. Unlike a monolithic representation usually used in prior skill learning methods, we propose to use a structured goal representation that can query and scope which objects to interact with, which can serve as a basis for solving more complex downstream tasks. We design a novel counterfactual intrinsic reward through the use of either a forward model or successor features that can learn an interaction skill between a pair of objects given as a goal. Through experiments on continuous control environments such as Magnetic Block and 2.5-D Stacking Box, we demonstrate that an agent can learn object interaction behaviors (e.g., attaching or stacking one block to another) without any external rewards or domain-specific knowledge.

Bullying on Social Media: Primitive Behavior Against Prophetic Marchals From A Hadith Ahkam Perspective

This article aims to outline some prophetic instructions regarding social interaction to avoid bullying behavior. This study uses a normative qualitative method, namely analyzing hadiths related to bullying with an ahkam hadith perspective. This study concludes that bullying behavior from the perspective of hadith is: 1) The Hadith narrated by Ibn Majah generally explains how the Prophet's Hadith views bullying as leading to degrading behavior. This is obtained from the word ihtiqarwhich has a correlation in meaning with the orientation of bullying behavior itself, namely degrading. Then a complete reading of the hadith narrated by Imam Muslim adds an explanation of indications of bullying behavior such as envy, hatred, belittling and tyrannizing other people. These attitudes are the seeds of hostility and division between people; 2) The basic idea of Ibn Majah's hadith number 3203 is the value of humanism, namely the effort to maintain a humane attitude towards fellow humans in any situation and condition. In the previous comprehensive thematic study, it was also sufficient to contribute to the conclusion of the basic idea of the hadith studied, namely explaining the consequences for perpetrators of bullying both verbally and physically. This spirit is in line with the spirit of humanism which condemns all kinds of acts of harassment or degrading towards fellow human beings. Therefore, in the perspective of ahkam hadith there is a need for preventive measures in order to prevent or overcome this act of bullying; 3) The preventive measures offered by this hadith are: a) providing spiritual awareness of the importance of preventing bullying early, b) supporting cooperation and breaking the cycle of conflict, c) eliminating inferiority for victims of bullying and honing assertive skills.

Sharing with minimal regulation? Evidence from neighborhood book exchange

Informal peer-to-peer services to share or barter goods often succumb to free riding behavior because they lack the tools to enforce compliance and reciprocity. We collect unique quantitative data on a form of unregulated peer-to-peer in-kind exchange that appears internationally viable: the free exchange of books via privately owned public bookcases, also known as little free libraries. Other than previously studied honor-based exchanges, little free libraries use a non-monetary one-to-one book exchange rate.We find surprisingly limited free riding in this market. Users return 9 books for every 10 taken. An incentivized survey points to strong social norms and preferences for cooperation among owners and users as key behavioral primitives that can explain the observed high and stable level of reciprocal exchange.

Coordinated Motion Planning for Heterogeneous Autonomous Vehicles Based on Driving Behavior Primitives

Heterogeneous autonomous vehicle (HAV) coordinated motion planning must guide each vehicle out of the conflict zone based on the differences in vehicle platform characteristics. Decomposing complex driving tasks into primitives is an effective way to improve algorithm efficiency. Hence, the purpose of this paper is to complete the coordinated motion planning tasks through offline driving behavior primitive (DBP) library generation, online extension and selection of DBPs. The proposed algorithm applies dynamic movement primitives and singular value decomposition to learn driving behavior patterns from driving data, integrates them into a model-based optimization generation method as constraints, and builds a DBP library by fusing driving data and vehicle model. Based on the generated DBP library and primitive association probabilities learned from labeled driving segments via stochastic context-free grammar, the planning method completes the independent DBP extension of each vehicle in the conflict zone, generates an interaction DBP tree, and uses the mixed-integer linear programming algorithm to optimally select the primitives to be executed. This study demonstrates that the generated DBP library not only expands the types of primitives, but also distinguishes the characteristics of HAVs. We also present how to utilize the DBP libraries to obtain coordinated motion planning results with spatiotemporal information in the form of DBP extension and selection. The results obtained by real vehicle platforms and simulation show that the proposed method can accomplish coordinated motion planning tasks without relying on specific scene elements and highlight the unique motion characteristics of HAVs.

私的言語論と二次的意味

Wittgensteinʼs so-called “private language argument” denies the normativity of private definitions, and suggests that sensation words are taught through replacing primitive behaviour expressive of sensations. However, these observations fail to provide a full understanding of mental concepts of our ordinary language. In this article, the relevance of the concept of secondary sense is brought to the fore through examining the concept of particular feelings inherent in various experiences and also that of calculation in the head. It is shown that natural linguistic reactions present in the use of words in the secondary sense are essential in gaining an overview of our ordinary language.

Anticipating human presence for safer worker - robot shared workspaces

The co-existence for human and mobile robots in modern industrial environments is increasingly common. Safety primitive behaviours are typically built-in mobile robots, to ensure safety. However, when fleets of multiple robots are operating in such environments, robot path planning becomes complicated and is often left sub-optimal to avoid compromising human, equipment, or process safety. Enhanced performance can be achieved if path planning takes into account not just current human presence, but projected human movement trajectories. While this problem has received extensive attention in outdoor environments in autonomous driving contexts, its indoors workspace equivalent has received less attention. This paper presents an approach for human movement prediction in industrial work environments, based on past and current heatmap occupancy grids and convolutional neural networks. The adopted heatmap format is appropriate for dealing with privacy concerns so as to avoid individual person identification. Obtained results from a range of simulation data are presented, following by a discussion on limitations, and challenges to be handled by further work.

PATTERNS IN INTELLIGENT CONTROL SYSTEMS FOR AUTONOMOUS ENTITIES

Nowadays, the process of integrating autonomous robotic systems (ARS) and military personnel formation is actively underway. Such a new object in the armed forces (AF) should have the property of self-sufficient behavior that guarantees mission fulfillment. The problem is in the construction of information processes in control systems that ensure a predictable behavior of an autonomous artificial system within the Armed Force subdivisions similar to a human one. Currently, the solution for this problem is the development of artificial entities that use symbolic or neural network methods and models for acquiring and formalizing “knowledge”. Ontologies made it possible to make these methods and knowledge models interpretable in natural language semantics and to bring the problem of human-machine interaction to a new level, which was noted by a number of researchers. Nevertheless, it should be noted that as long as there is no cognition process model, the theoretical results obtained are not in demand in practice and are only of theoretical interest. There is a gap between primitive behavior models of artificial entities, their interaction models and expectations from practice. The purpose of this work is to propose an approach to the intellectualization of the artificial entities’ behavior by revising the logical and mathematical abstractions underlying the construction of their onboard control systems based on the pattern theory development. Such approach ensures the transfer of effective human experience to the control systems of artificial entities, as well as the compatibility of the theological and causal approach. It is shown that the choice under severe time constraints is based on behavioral patterns that reflect effective experience. Patterns form both the information structure of representations and a set of possible variants of representations. Subject’s assessments of satisfaction with the current choice situation lead to a change in a subject's interest structure, and he can choose it. There is a developed formal model of the behavioral pattern and a method of logical inference on patterns. The paper considers the problems of identification and construction of pattern models. It proposes to use four positions of information processing for these purposes. The paper proposes an approach to developing software for intelligent control systems for autonomous systems based on the pattern theory. There are the results of behavioral pattern identification solutions that use new generation training systems with voice control. The paper considers the issues of constructing the architecture of unmanned undersea vehicles with an onboard control system that uses pattern models and a pattern inference system. It also solves the problem of the behavior tactics of the node - autonomous unmanned vehicle based on the onboard sensor system data.

Learning generalizable behaviors from demonstration.

Generalizing prior experiences to complete new tasks is a challenging and unsolved problem in robotics. In this work, we explore a novel framework for control of complex systems called Primitive Imitation for Control (PICO). The approach combines ideas from imitation learning, task decomposition, and novel task sequencing to generalize from demonstrations to new behaviors. Demonstrations are automatically decomposed into existing or missing sub-behaviors which allows the framework to identify novel behaviors while not duplicating existing behaviors. Generalization to new tasks is achieved through dynamic blending of behavior primitives. We evaluated the approach using demonstrations from two different robotic platforms. The experimental results show that PICO is able to detect the presence of a novel behavior primitive and build the missing control policy.

Initialization of latent space coordinates via random linear projections for learning robotic sensory-motor sequences.

Robot kinematic data, despite being high-dimensional, is highly correlated, especially when considering motions grouped in certain primitives. These almost linear correlations within primitives allow us to interpret motions as points drawn close to a union of low-dimensional affine subspaces in the space of all motions. Motivated by results of embedding theory, in particular, generalizations of the Whitney embedding theorem, we show that random linear projection of motor sequences into low-dimensional space loses very little information about the structure of kinematic data. Projected points offer good initial estimates for values of latent variables in a generative model of robot sensory-motor behavior primitives. We conducted a series of experiments in which we trained a Recurrent Neural Network to generate sensory-motor sequences for a robotic manipulator with 9 degrees of freedom. Experimental results demonstrate substantial improvement in generalization abilities for unobserved samples during initialization of latent variables with a random linear projection of motor data over initialization with zero or random values. Moreover, latent space is well-structured such that samples belonging to different primitives are well separated from the onset of the training process.

Energies for Elastic Plates and Shells from Quadratic-Stretch Elasticity

We derive stretching and bending energies for isotropic elastic plates and shells. Through the dimensional reduction of a bulk elastic energy quadratic in Biot strains, we obtain two-dimensional bending energies quadratic in bending measures featuring a bilinear coupling of stretches and geometric curvatures. For plates, the bending measure is invariant under spatial dilations and naturally extends primitive bending strains for straight rods. For shells or naturally-curved rods, the measure is not dilation invariant, and contrasts with previous ad hoc postulated forms. The corresponding field equations and boundary conditions feature moments linear in the bending measures, and a decoupling of stretching and bending such that application of a pure moment results in isometric deformation of a unique neutral surface, primitive behaviors in agreement with classical linear response but not displayed by commonly used analytical models. We briefly comment on relations between our energies, those derived from a neo-Hookean bulk energy, and a commonly used discrete model for flat membranes. Although the derivation requires consideration of stretch and rotation fields, the resulting energy and field equations can be expressed entirely in terms of metric and curvature components of deformed and reference surfaces.

The shifting role of the cerebellum in executive, emotional and social processing across the lifespan

The cerebellum’s anatomical and functional organization and network interactions between the cerebellum and the cerebral cortex and subcortical structures are dynamic across the lifespan. Executive, emotional and social (EES) functions have likewise evolved during human development from contributing to primitive behaviors during infancy and childhood to being able to modulate complex actions in adults. In this review, we address how the importance of the cerebellum in the processing of EES functions might change across development. This evolution is driven by the macroscopic and microscopic modifications of the cerebellum that are occurring during development including its increasing connectivity with distant supra-tentorial cortical and sub-cortical regions. As a result of anatomical and functional changes, neuroimaging and clinical data indicate that the importance of the role of the cerebellum in human EES-related networks shifts from being crucial in newborns and young children to being only supportive later in life. In early life, given the immaturity of cortically mediated EES functions, EES functions and motor control and perception are more closely interrelated. At that time, the cerebellum due to its important role in motor control and sequencing makes EES functions more reliant on these computational properties that compute spatial distance, motor intent, and assist in the execution of sequences of behavior related to their developing EES expression. As the cortical brain matures, EES functions and decisions become less dependent upon these aspects of motor behavior and more dependent upon high-order cognitive and social conceptual processes. At that time, the cerebellum assumes a supportive role in these EES-related behaviors by computing their motor and sequential features. We suspect that this evolving role of the cerebellum has complicated the interpretation of its contribution to EES computational demands.

Spectral Representation of Behaviour Primitives for Depression Analysis

Depression is a serious mental disorder affecting millions of people all over the world. Traditional clinical diagnosis methods are subjective, complicated and require extensive participation of clinicians. Recent advances in automatic depression analysis systems promise a future where these shortcomings are addressed by objective, repeatable, and readily available diagnostic tools to aid health professionals in their work. Yet there remain a number of barriers to the development of such tools. One barrier is that existing automatic depression analysis algorithms base their predictions on very brief sequential segments, sometimes as little as one frame. Another barrier is that existing methods do not take into account what the context of the measured behaviour is. In this article, we extract multi-scale video-level features for video-based automatic depression analysis. We propose to use automatically detected human behaviour primitives as the low-dimensional descriptor for each frame. We also propose two novel spectral representations, i.e., spectral heatmaps and spectral vectors, to represent video-level multi-scale temporal dynamics of expressive behaviour. Constructed spectral representations are fed to Convolution Neural Networks (CNNs) and Artificial Neural Networks (ANNs) for depression analysis. We conducted experiments on the AVEC 2013 and AVEC 2014 benchmark datasets to investigate the influence of interview tasks on depression analysis. In addition to achieving state of the art accuracy in severity of depression estimation, we show that the task conducted by the user matters, that fusion of a combination of tasks reaches highest accuracy, and that longer tasks are more informative than shorter tasks, up to a point.

Using Behavioral Primitives to Model Students’ Digital Behavior

The rapid and intense development of distance learning in recent years has led to increasingly comprehensive solutions, recording students’ activity in the form of learning traces. Adapted learning systems can exploit this data to analyze students’ behavior and help them be more successful in their learning process. In this paper, we propose an approach that is new in data mining, which consists of representing the behavioral dynamics of a set of students through Behavioral Primitives. A Behavioral Primitive represents the temporal evolution of learning indicators to model the behavior of a group of students. The primitive then corresponds to a dynamic distribution, representing both the mean values and the variance of specific indicators over time. We experiment this method on a known, widely used, and open dataset (OULAD). Our results confirm the relevance of the proposed methodology to characterize students’ behaviors intelligibly and visually while respecting each student's anonymity. Our work provides a powerful and explainable tool for educational actors. It allows them to analyze learners’ behaviors and perform pedagogical actions.

Autonomous Obstacle Crossing Strategies for the Hybrid Wheeled-Legged Robot Centauro

The development of autonomous legged/wheeled robots with the ability to navigate and execute tasks in unstructured environments is a well-known research challenge. In this work we introduce a methodology that permits a hybrid legged/wheeled platform to realize terrain traversing functionalities that are adaptable, extendable and can be autonomously selected and regulated based on the geometry of the perceived ground and associated obstacles. The proposed methodology makes use of a set of terrain traversing primitive behaviors that are used to perform driving, stepping on, down and over and can be adapted, based on the ground and obstacle geometry and dimensions. The terrain geometrical properties are first obtained by a perception module, which makes use of point cloud data coming from the LiDAR sensor to segment the terrain in front of the robot, identifying possible gaps or obstacles on the ground. Using these parameters the selection and adaption of the most appropriate traversing behavior is made in an autonomous manner. Traversing behaviors can be also serialized in a different order to synthesise more complex terrain crossing plans over paths of diverse geometry. Furthermore, the proposed methodology is easily extendable by incorporating additional primitive traversing behaviors into the robot mobility framework and in such a way more complex terrain negotiation capabilities can be eventually realized in an add-on fashion. The pipeline of the above methodology was initially implemented and validated on a Gazebo simulation environment. It was then ported and verified on the CENTAURO robot enabling the robot to successfully negotiate terrains of diverse geometry and size using the terrain traversing primitives.

Autonomous Identification and Goal-Directed Invocation of Event-Predictive Behavioral Primitives

Voluntary behavior of humans appears to be composed of small, elementary building blocks, or behavioral primitives. While this modular organization seems crucial for the learning of complex motor skills and the flexible adaption of behavior to new circumstances, the problem of learning meaningful, compositional abstractions from sensorimotor experiences remains an open challenge. Here, we introduce a computational learning architecture, termed as surprise-based behavioral modularization into event-predictive structures (SUBMODES) that explores behavior and identifies the underlying behavioral units completely from scratch. The SUBMODES architecture bootstraps sensorimotor exploration using a self-organizing neural controller. While exploring the behavioral capabilities of its own body, the system learns modular structures that predict the sensorimotor dynamics and generate the associated behavior. In line with recent theories of event perception, the system uses unexpected prediction error signals, i.e., surprise, to detect transitions between successive behavioral primitives. We show that, when applied to two robotic systems with completely different body kinematics, the system manages to learn a variety of complex behavioral primitives. Moreover, after initial self exploration the system can use its learned predictive models progressively more effectively for invoking model predictive planning and goal-directed control in different tasks and environments.

Neoichnology of mygalomorph spiders: Improving the recognition of spider burrows in the geological record

The female Mygalomorphae spiders are sedentary and long-lived organisms that spend most of their lives inside their burrows. Neoichnological studies provide relevant information that can help the recognition of these structures in paleosols. Body fossils of spiders are known since the Carboniferous and burrowing is a primitive behavior in Mygalomorphae spiders. However, trace fossils attributable to ground-dwelling spiders are still poorly documented in the geological record. In this work, we examine the burrows and burrowing behavior of Rachiasintermedius Soares, 1944 (Araneae: Mygalomorphae: Nemesiidae) in its natural environment and discuss the characteristics that can be used as ichnotaxobases for recognition of fossil spider burrows. Three major architectures, straight shaft with a terminal ovoid chamber, J-shaped winding shaft with the terminal chamber, and Y-shaped with a terminal teardrop-shaped chamber, are described and compared to morphologically similar ichnogenera, like Capayanichnus, Loloichnus, Macanopsis, and Psilonichnus. Differences in burrow shape and architecture are linked with the spider's sex and ontogenetic stage. Pedipalps, chelicerae, and fangs are used for soil excavation, forming a variety of burrow wall ornaments represented by delicate sub-horizontal parallel ridges, irregular knobby micro-relief surface with soil structures attached to the wall, rounded pits, and millimeter-scale vertical striations along the burrow length. A thick inorganic clay lining covers the inner burrow wall, a feature that has not been described for spider burrows yet. These characteristics allow distinguishing spider burrows from burrows produced by other soil-dwelling arthropods. They should be used for spider burrow recognition in paleosols, mainly the millimeter-scale vertical striations that had not been documented before. The data discussed herein improve the knowledge about spider burrowing behavior and the mechanism that play the main role in preserving these burrows' features in the fossil record.

A Framework for Automatic Behavior Generation in Multi-Function Swarms.

Multi-function swarms are swarms that solve multiple tasks at once. For example, a quadcopter swarm could be tasked with exploring an area of interest while simultaneously functioning as ad-hoc relays. With this type of multi-function comes the challenge of handling potentially conflicting requirements simultaneously. Using the Quality-Diversity algorithm MAP-elites in combination with a suitable controller structure, a framework for automatic behavior generation in multi-function swarms is proposed. The framework is tested on a scenario with three simultaneous tasks: exploration, communication network creation and geolocation of Radio Frequency (RF) emitters. A repertoire is evolved, consisting of a wide range of controllers, or behavior primitives, with different characteristics and trade-offs in the different tasks. This repertoire enables the swarm to online transition between behaviors featuring different trade-offs of applications depending on the situational requirements. Furthermore, the effect of noise on the behavior characteristics in MAP-elites is investigated. A moderate number of re-evaluations is found to increase the robustness while keeping the computational requirements relatively low. A few selected controllers are examined, and the dynamics of transitioning between these controllers are explored. Finally, the study investigates the importance of individual sensor or controller inputs. This is done through ablation, where individual inputs are disabled and their impact on the performance of the swarm controllers is assessed and analyzed.