Concurrent Learning Research Articles

Mathematical Learning Disorder in School-Age Children with Attention-Deficit Hyperactivity Disorder

To explore the prevalence of mathematics disorder (MD) relative to reading disorders (RD) in school-age children with attention-deficit hyperactivity disorder (ADHD) and examine the effects of age, sex, cooccurring conduct disorder (CD), and ADHD subtype on this comorbidity. Participants were school-age children (n = 476) with confirmed DSM-IV diagnosis of ADHD. The assessment included semistructured parent and teacher interviews and standardized measures of intelligence, academic attainment, and language abilities. Based on the presence or absence of concurrent learning disorders, we compared the emerging 4 groups: ADHD-only, ADHD + MD, ADHD + RD, and ADHD + MD + RD. Overall prevalence of comorbid ADHD + MD was 18.1%. Age, sex, ADHD subtypes, or comorbid CD did not affect the frequency of MD. Children with concurrent ADHD and either MD or RD attained lower IQ, language, and academic scores than those with ADHD alone. Children with ADHD + MD + RD were more seriously impaired and demonstrated distinct deficits in receptive and expressive language. MDs are relatively common in school-age children with ADHD and are frequently associated with RDs. Children with ADHD + MD + RD are more severely impaired. These deficits simply cannot be explained as consequences of ADHD and might have unique biological underpinnings, with implications for diagnostic classification and therapeutic interventions.

Negative priming in the SRT task: Learning of irrelevant sequences is enhanced by concurrent learning of relevant sequences

We investigated negative priming (NP) in sequence learning, as demonstrated by Cock, Berry, and Buchner (2002). The NP effect in sequence learning refers to the observation that Serial Reaction Time performance becomes impaired when participants respond to a sequence they were previously ignoring. In Experiment 1, we established a NP effect with a complex second-order conditional sequence structure. Experiment 2 showed that NP occurs even when the irrelevant sequence is not synchronised with the relevant sequence. However, the results of Experiment 3 demonstrated that NP was absent when participants did not concurrently respond to a relevant sequence, while being exposed to the irrelevant sequence. These findings indicate that NP extends over complex forms of sequence learning, but requires concurrent learning of relevant sequences. We suggest that NP and the enhancement of (the expression of) learning of irrelevant sequences by concurrent learning of relevant sequences is guided by attentional processes.

Assessing and teaching children who have difficulty learning arithmetic

Arithmetic difficulties have long captured the attention of teachers and researchers, but intervention programs for assisting children are seldom successful for all. Recent Australian research suggests that this is because we have failed to recognise the complexity of arithmetic difficulties. Analysis of some 30,000 oneon-one clinical interviews conducted over three years during the Early Numeracy Research Project provided rich data for charting the pathway of young children’s number learning in four domains (Counting, Place Value, Addition and Subtraction Strategies, and Multiplication and Division Strategies), and for identifying children who were having difficulty. We describe such children as being vulnerable or at risk of not being able to take advantage of everyday classroom experiences. The data show that the combinations of domains in which children were vulnerable were diverse, and suggest that there is no single ‘formula’ for describing children who are vulnerable in number learning, or for describing the instructional needs of students. Indeed, children have learning needs that call for teachers to make individual decisions about the instructional approach for each child. Further, the diversity of children’s mathematical knowledge in the four domains suggests that knowledge in any one domain is not necessarily prerequisite for knowledge construction in another domain. This finding has implications for both intervention programs and for the way in which school mathematics is introduced to children. It seems likely that children may benefit from concurrent learning opportunities in all number domains, and that experiences in one domain should not be delayed until a level of mathematical knowledge is constructed in another domain.

Learned preferences induced by electrical stimulation of a food-related area of the parabrachial complex: Effects of naloxone

Electrical stimulation of the External Lateral Parabrachial Subnucleus (LPBe), a food-related area, induced behavioral preferences for associated stimuli in a taste discrimination learning task. Although this stimulation appeared to be ineffective to elicit standard lever press self-stimulation, it induced place preference for one of two training compartments of a rectangular maze in which animals (adult male Wistar rats) received concurrent electrical brain stimulation. In subjects that consistently showed a preference behavior in different trials, administration of the opioid antagonist naloxone (4 mg/ml/kg) blocked concurrent learning when the test was made in a new maze but not in the same maze in which animals had learned the task. These results are discussed in terms of the possible participation of the LPBe subnucleus in different natural and artificial brain reward systems.

Multi-robot concurrent learning of cooperative behaviours for the tracking of multiple moving targets

Reinforcement learning has been extensively studied and applied for generating cooperative behaviours in multi-robot systems. However, traditional reinforcement learning algorithms assume discrete state and action spaces with finite number of elements. This limits the learning to discrete behaviours and cannot be applied to most real multi-robot systems that inherently require appropriate combinations of different elementary behaviours. To address this problem, we design a distributed learning controller that integrates reinforcement learning with behaviour-based control networks. This learning controller can enable the robots to generate appropriate control policy without the need for human design or hardcoding. Furthermore, to address the problems in concurrent learning, we propose a distributed learning control algorithm to coordinate the concurrent learning processes. The distributed learning controller and learning control algorithm are applied to multi-robot tracking of multiple moving targets. The efficacy of our proposed scheme is shown through simulations.

A Team-Based CAM Project Utilising the Latest CAD/CAM and Web-Based Technologies in the Concurrent Engineering Environment

This paper describes the computer-aided manufacture (CAM) part of newly created third-year computer-aided design/computer-aided manufacture (CAD/CAM) project in the Department of Mechanical Engineering at the University of Bristol, which brings together the most current CAD/CAM practices, web-based technologies and shop-floor practices, to provide a stimulating and dynamic learning environment for the students. The project involves the design and manufacture of a crest as a feature for a chip/material-sorting machine in a concurrent learning environment. The main objective is complete product development, from design to manufacture (i.e. the product cycle), using the latest CAD/CAM technologies for the students. Students carry out product development assessment and make suitable adjustments to the design once the problems with engineering curves are identified. They especially practise the design for manufacture (DFM) technique to ensure any part designed on the CAD/CAM system can actually be machined with available machines, tools and fixtures. They also learn how to animate the machining processes of their final product as a part of product visualisation practice. In generative machining (GM), students learn the CAD/CAM approach to computer numerical control (CNC) programming, where a tool-path is generated automatically. GM also teaches the students how quickly and efficiently a design can be customised, updated and manufactured automatically when a feature of the proposed design is modified using CAD/CAM and CNC machine tools principles.

Cooperative Multi-Agent Learning: The State of the Art

Cooperative multi-agent systems (MAS) are ones in which several agents attempt, through their interaction, to jointly solve tasks or to maximize utility. Due to the interactions among the agents, multi-agent problem complexity can rise rapidly with the number of agents or their behavioral sophistication. The challenge this presents to the task of programming solutions to MAS problems has spawned increasing interest in machine learning techniques to automate the search and optimization process. We provide a broad survey of the cooperative multi-agent learning literature. Previous surveys of this area have largely focused on issues common to specific subareas (for example, reinforcement learning, RL or robotics). In this survey we attempt to draw from multi-agent learning work in a spectrum of areas, including RL, evolutionary computation, game theory, complex systems, agent modeling, and robotics. We find that this broad view leads to a division of the work into two categories, each with its own special issues: applying a single learner to discover joint solutions to multi-agent problems (team learning), or using multiple simultaneous learners, often one per agent (concurrent learning). Additionally, we discuss direct and indirect communication in connection with learning, plus open issues in task decomposition, scalability, and adaptive dynamics. We conclude with a presentation of multi-agent learning problem domains, and a list of multi-agent learning resources.

Perceptual or motor learning in SRT tasks with complex sequence structures

We investigated under which conditions sequence learning in a serial reaction time task can be based on perceptual learning. A replication of the study of Mayr (1996) confirmed perceptual and motor learning when sequences were learned concurrently. However, between-participants manipulations of the motor and perceptual sequences only supported motor learning in cases of more complex deterministic and probabilistic sequence structures. Perceptual learning using a between-participants design could only be established with a simple deterministic sequence structure. The results seem to imply that perceptual learning can be facilitated by a concurrently learned motor sequence. Possibly, concurrent learning releases necessary attentional resources or induces a structured learning condition under which perceptual learning can take place. Alternatively, the underlying mechanism may rely on binding between the perceptual and motor sequences.

On-policy concurrent reinforcement learning

When an agent learns in a multi-agent environment, the payoff it receives is dependent on the behaviour of the other agents. If the other agents are also learning, its reward distribution becomes non-stationary. This makes learning in multi-agent systems more difficult than single-agent learning. Prior attempts at value-function based learning in such domains have used off-policy Q-learning that do not scale well as the cornerstone, with restricted success. This paper studies on-policy modifications of such algorithms, with the promise of scalability and efficiency. In particular, it is proven that these hybrid techniques are guaranteed to converge to their desired fixed points under some restrictions. It is also shown, experimentally, that the new techniques can learn (from self-play) better policies than the previous algorithms (also in self-play) during some phases of the exploration.

Managing Evolutionary Method Engineering by Method Rationale

This paper explores how to integrate formal meta-models with an informal method rationale to support evolutionary (continuous) method development. While the former provides an exact and computer-executable specification of a method, the latter enables concurrent learning, expansion, and refinement of method use (instances of meta-models) and meta-models (evolution of method specifications). We explain the need for method rationale by observing the criticality of evolving method knowledge in helping software organizations to learn, as well as by the recurrent failure to introduce rigid and stable methods. Like a design rationale, a method rationale establishes a systematic and organized trace of method evolution. Method rationale is located at two levels of type-instance hierarchy depending on its type of use and the scope of the changes traced. A method construction rationale garners a history of method knowledge evolution as part of the method engineering process, which designs and adapts the method to a given organizational context. A method use rationale maintains knowledge of concrete use contexts and their history and justifies further method deployment in alternative contexts, reveals limitations in its past use, and enables sharing of method use experience. The paper suggests how a method rationale helps share knowledge of methods between method users and engineers, explores how method engineers coordinate the evolution of the existing method base through it, and suggests ways to improve learning through method rationale.

Brain temperature fluctuations during passive vs. active cocaine administration: clues for understanding the pharmacological determination of drug-taking behavior

While it is generally assumed that cocaine self-administration (SA) is determined and maintained by the pharmacological actions of cocaine in the brain, it is also a drug-motivated and drug-reinforced goal-directed behavior, which is determined by concurrent learning and behavioral performance. To dissociate the contributions of pharmacological and behavioral factors to cocaine SA, it is important to compare cocaine SA with its pharmacological copy, passive intravenous (iv) cocaine administration. This approach was employed in the present study with respect to brain temperatures, a dynamic parameter that reflects metabolic neural activity and shows consistent fluctuations during cocaine SA. Passive cocaine injections performed with the same dose/pattern as SA induced brain temperature fluctuations similar in many ways to those in behaving animals. The initial passive drug administration of a session elevated brain temperature, while subsequent repeated injections were associated with biphasic temperature fluctuations that maintained at a relatively stable plateau. Although the magnitude of these fluctuations was twofold smaller than in behaving animals, passive animals had the same pattern; brain temperatures transiently decreased after cocaine injection, then increased, and were inhibited again by the next cocaine infusion. In contrast to self-administering animals, rats exposed to passive cocaine injections had significantly lower basal temperatures and never showed gradual temperature increases preceding the initial injection. Striking differences in brain temperature dynamics seen in the beginning of a session suggest that during the development of drug-taking behavior the initial cocaine-induced neural activation becomes transformed into behavior-related “anticipatory” neural activation (motivational arousal) that fuels drug seeking and results in the initial drug intake. While this activation is triggered by drug-related cues and enhanced by the initial cocaine intake, subsequent highly cyclical cocaine intakes appear to be primarily pharmacologically determined.

Fornix Lesions Impair Context-Related Cingulothalamic Neuronal Patterns and Concurrent Discrimination Learning in Rabbits (Oryctolagus cuniculus).

Cingulothalamic neurons develop topographic patterns of cue-elicited neuronal activity during discrimination learning. These patterns are context-related and are degraded by hippocampal lesions, suggesting that hippocampal modulation of cingulothalamic activity results in the expression of the patterns, which could promote the retrieval of context-appropriate responses and memories. This hypothesis was tested by training rabbits (Oryctolagus cuniculus) with fornix lesions concurrently on two discrimination tasks (approach and avoidance) in different contexts. Because the same conditioned stimuli were used for both tasks, contextual information was critical for overcoming intertask interference during concurrent task acquisition. The lesions degraded the topographic patterns and significantly impaired concurrent learning, suggesting that hippocampal-cingulothalamic interactions and the resulting topographic patterns are critical for processing contextual information needed to defeat interference.

Application of fast haar transform and concurrent learning to tool-breakage detection in milling

In this paper, an effective monitoring approach for manufacturing processing by combining the in-place fast Haar transform and concurrent learning is described and applied to detect tool flute breakage during end milling by sensing the feed-motor current signatures. The application procedure and the effectiveness of the proposed method have been delineated by case studies; the results indicate that the proposed approach possesses an excellent potential application to tool breakage detection in milling.

The role of the external lateral parabrachial subnucleus in flavor preferences induced by predigested food administered intragastrically

A study was undertaken of the role of the external lateral parabrachial subnucleus (PBNLe) in flavor preferences induced by the intragastric administration of predigested/cephalic food. These preferences were developed using two different learning procedures, concurrent and sequential. In the concurrent procedure, two different-flavored stimuli were presented at the same time: one stimulus was paired with the simultaneous intragastric administration of partially digested food and the other with physiological saline. In the sequential learning procedure, the two stimuli were presented at alternate sessions. The results showed that PBNLe lesions blocked acquisition of concurrent learning but had no effect on the sequential procedure. In the latter case, both lesioned and control animals showed a strong preference for the gustatory stimulus paired with partially digested food. These results are interpreted in terms of a dual neurobiological system involved in the rewarding effects of visceral signals.

Mismatch negativity shows that 3–6-year-old children can learn to discriminate non-native speech sounds within two months

Using 3–6-year-old children as subjects, we describe the neural plasticity accompanying the concurrent learning of a foreign language in a natural environment. Children were monitored for 6 months as they either enrolled in schools or daycare centers where only Finnish was spoken (Control group) or as they joined a French school or a daycare center where French was spoken 50–90% of the time (Experimental group). Whereas mismatch negativity (MMN)—a brain's electrical change-detection response—for a French speech contrast was initially absent or very small in both groups, it was conspicuous 2 months after Finnish children had joined a French kindergarten. Consequently, the data suggest that youngsters can learn to distinguish non-native speech sounds in natural language environment without any special training in just a couple of months. Accordingly, these data herald the vast potential MMN may entail for studying language learning, especially in situations where behavioral responses cannot be readily elicited.

Concurrent learning of temporal and spatial sequences.

Concurrent learning of temporal and spatial sequences.

Parallel cortico-basal ganglia mechanisms for acquisition and execution of visuomotor sequences - a computational approach.

Experimental studies have suggested that many brain areas, including the basal ganglia (BG), contribute to procedural learning. Focusing on the basal ganglia-thalamocortical (BG-TC) system, we propose a computational model to explain how different brain areas work together in procedural learning. The BG-TC system is composed of multiple separate loop circuits. According to our model, two separate BG-TC loops learn a visuomotor sequence concurrently but using different coordinates, one visual, and the other motor. The visual loop includes the dorsolateral prefrontal (DLPF) cortex and the anterior part of the BG, while the motor loop includes the supplementary motor area (SMA) and the posterior BG. The concurrent learning in these loops is based on reinforcement signals carried by dopaminergic (DA) neurons that project divergently to the anterior ("visual") and posterior ("motor") parts of the striatum. It is expected, however, that the visual loop learns a sequence faster than the motor loop due to their different coordinates. The difference in learning speed may lead to inconsistent outputs from the visual and motor loops, and this problem is solved by a mechanism called a "coordinator," which adjusts the contribution of the visual and motor loops to a final motor output. The coordinator is assumed to be in the presupplementary motor area (pre-SMA). We hypothesize that the visual and motor loops, with the help of the coordinator, achieve both the quick acquisition of novel sequences and the robust execution of well-learned sequences. A computational model based on the hypothesis is examined in a series of computer simulations, referring to the results of the 2 x 5 task experiments that have been used on both monkeys and humans. We found that the dual mechanism with the coordinator was superior to the single (visual or motor) mechanism. The model replicated the following essential features of the experimental results: (1) the time course of learning, (2) the effect of opposite hand use, (3) the effect of sequence reversal, and (4) the effects of localized brain inactivations. Our model may account for a common feature of procedural learning: A spatial sequence of discrete actions (subserved by the visual loop) is gradually replaced by a robust motor skill (subserved by the motor loop).

Visual agnosia and Klüver–Bucy syndrome in marmosets ( Callithrix jacchus) following ablation of inferotemporal cortex, with additional mnemonic effects of immunotoxic lesions of cholinergic projections to medial temporal areas

Inferotemporal ablations in the New World monkey, the common marmoset ( Callithrix jacchus), produced a persistent impairment on visual discrimination learning and a florid, but transient, Klüver–Bucy syndrome. Monkeys with these ablations were impaired on acquisition of object discriminations to a high criterion and on concurrent discrimination learning, to a single high criterion across all trials. Neither the control monkeys nor the monkeys with inferotemporal ablations found acquisition more difficult when the component discriminations of a set were presented concurrently compared to consecutively, although the monkeys with inferotemporal ablations found acquisition under both these conditions somewhat more difficult than did control monkeys. This suggests that the severe impairment caused by inferotemporal ablations on concurrent learning measured across all trials is due to the need for sustained performance across a concurrent set rather than to the extra mnemonic demands of concurrent presentation. When immunotoxic lesions of the cholinergic projection to the hippocampal formation were added to the inferotemporal ablations, a further impairment on retention, and a differential impairment on concurrent, compared to consecutive, learning was observed. Previous studies have shown that lesions of the cholinergic projection to the hippocampus alone, or excitotoxic hippocampal lesions, do not affect simple visual discrimination learning. It is suggested that large inferotemporal ablations in monkeys produce a visual agnosia which causes severe ‘psychic blindness’ in the first instance, and a persistent impairment on visual discrimination learning. The hippocampus makes a contribution, which may be mnemonic, to discrimination performance after inferotemporal ablations.

CA1-specific N-methyl-D-aspartate receptor knockout mice are deficient in solving a nonspatial transverse patterning task.

In both humans and animals, the hippocampus is critical to memory across modalities of information (e.g., spatial and nonspatial memory) and plays a critical role in the organization and flexible expression of memories. Recent studies have advanced our understanding of cellular basis of hippocampal function, showing that N-methyl-d-aspartate (NMDA) receptors in area CA1 are required in both the spatial and nonspatial domains of learning. Here we examined whether CA1 NMDA receptors are specifically required for the acquisition and flexible expression of nonspatial memory. Mice lacking CA1 NMDA receptors were impaired in solving a transverse patterning problem that required the simultaneous acquisition of three overlapping odor discriminations, and their impairment was related to an abnormal strategy by which they failed to adequately sample and compare the critical odor stimuli. By contrast, they performed normally, and used normal stimulus sampling strategies, in the concurrent learning of three nonoverlapping concurrent odor discriminations. These results suggest that CA1 NMDA receptors play a crucial role in the encoding and flexible expression of stimulus relations in nonspatial memory.

An interactive dynamic model for integrating knowledge management methods and knowledge sharing technology in a traditional classroom

This paper reports an interactive dynamic model using Continuous Knowledge Management methods and Knowledge Sharing technology to integrate the acquisition of skills and relevant information (knowledge level) into diverse, individualized, concurrent learning processes in a traditional classroom setting. As opposed to a passive introduction of technology to facilitate the traditional learning processes a Web based active learning and continuous evaluation process was created which integrates objective scientific knowledge relating to course content, subjective knowledge obtained through personal interactions and empirical knowledge collected during the learning process. Knowledge Management, an emerging area of Artificial Intelligence, encompasses identifying, mapping, and managing intellectual assets to generate new knowledge for competitive advantage and for sharing of technology. The Web-based model of knowledge management discussed here allows a diverse group of learners to progressively interact and participate in the learning process, providing non-threatening self-evaluation and just-in-time individualized feedback to the learners and efficient tracking and supervision tools to the instructor. CS1003, a required general education class provides an ideal application of this model as the course draws from a diverse body of students ranging from history to math majors and from freshmen to seniors. The instructional design of this course using the interactive dynamics of Knowledge Management includes (i) provision of course archives and relevant static information as a passive repository, (ii) Web Discussion Forums, electronic chats and email communication for active learning and continuous interaction, (iii) an intelligent self-evaluation and grade reporting system for non-threatening self-testing and what-if analysis of performance, and (iv) a dynamic student feedback system including individualized supervision and anonymous feedback. Application of this instructional process enhanced the goals of the course from mere computer literacy to what the 1999 NRC Report calls Fluency in Information Technology (FIT). Three kinds of knowledge requirements are identified for FIT: (1) Contemporary skills, (2) Foundational concepts, and (3) Intellectual capabilities. This model is broadly applicable to extend the benefits of traditional classroom instruction to focus diverse intellectual abilities and interests in a collaborative learning process. Formal and informal evaluation support this claim, demonstrating that the transition from purely traditional teaching to a high degree of technology fluency can be painless, efficient and effective in preparing the students for a technology intensive information age.