Critical behavior in a cross-situational lexicon learning scenario

The associationist account for early word learning is based on the co-occurrence between referents and words. Here we introduce a noisy cross-situational learning scenario in which the referent of the uttered word is eliminated from the context with probability gamma, thus modeling the noise produced by out-of-context words. We examine the performance of a simple associative learning algorithm and find a critical value of the noise parameter gamma(c) above which learning is impossible. We use finite-size scaling to show that the sharpness of the transition persists across a region of order tau(-1/2) about gamma(c), where tau is the number of learning trials, as well as to obtain the learning error (scaling function) in the critical region. In addition, we show that the distribution of durations of periods when the learning error is zero is a power law with exponent -3/2 at the critical point. Copyright (C) EPLA, 2012

Ontogeny of manipulative behavior and nut-cracking in young tufted capuchin monkeys (Cebus apella): a Perception-action perspective

How do capuchin monkeys learn to use stones to crack open nuts? Perception-action theory posits that individuals explore producing varying spatial and force relations among objects and surfaces, thereby learning about affordances of such relations and how to produce them. Such learning supports the discovery of tool use. We present longitudinal developmental data from semifree-ranging tufted capuchin monkeys (Cebus apella) to evaluate predictions arising from Perception-action theory linking manipulative development and the onset of tool-using. Percussive actions bringing an object into contact with a surface appeared within the first year of life. Most infants readily struck nuts and other objects against stones or other surfaces from 6 months of age, but percussive actions alone were not sufficient to produce nut-cracking sequences. Placing the nut on the anvil surface and then releasing it, so that it could be struck with a stone, was the last element necessary for nut-cracking to appear in capuchins. Young chimpanzees may face a different challenge in learning to crack nuts: they readily place objects on surfaces and release them, but rarely vigorously strike objects against surfaces or other objects. Thus the challenges facing the two species in developing the same behavior (nut-cracking using a stone hammer and an anvil) may be quite different. Capuchins must inhibit a strong bias to hold nuts so that they can release them; chimpanzees must generate a percussive action rather than a gentle placing action. Generating the right actions may be as challenging as achieving the right sequence of actions in both species. Our analysis suggests a new direction for studies of social influence on young primates learning sequences of actions involving manipulation of objects in relation to surfaces.

OBSERVING BEHAVIOR AND ATYPICALLY RESTRICTED STIMULUS CONTROL

Restricted stimulus control refers to discrimination learning with atypical limitations in the range of controlling stimuli or stimulus features In the study reported here 4 normally capable individuals and 10 individuals with Intellectual disabilities (ID) performed two-sample delayed matching to sample Sample stimulus observing was recorded with an eye tracking apparatus High accuracy scores indicated stimulus control by both sample stimuli for the 4 nondisabled participants and 4 participants with ID and eye tracking data showed reliable observing of all stimuli Intermediate accuracy scores indicated restricted stimulus control for the remaining 6 participants Their eye tracking data showed that errors were related to failures to observe sample stimuli and relatively brief observing durations Five of these participants were then given interventions designed to improve observing behavior For 4 participants the interventions resulted initially in elimination of observing failures increased observing durations and Increased accuracy For 2 of these participants contingencies sufficient to maintain adequate observing were not always sufficient to maintain high accuracy subsequent procedure modifications restored It however For the 5th participant initial improvements in observing were not accompanied by improved accuracy in apparent Instance of observing without attending accuracy improved only after an additional intervention that imposed contingencies on observing behavior Thus interventions that control observing behavior seem necessary but may not always be sufficient for the remediation of restricted stimulus control

Word sense disambiguation via high order of learning in complex networks

Complex networks have been employed to model many real systems and as a modeling tool in a myriad of applications. In this paper, we use the framework of complex networks to the problem of supervised classification in the word disambiguation task, which consists in deriving a function from the supervised (or labeled) training data of ambiguous words. Traditional supervised data classification takes into account only topological or physical features of the input data. On the other hand, the human (animal) brain performs both low- and high-level orders of learning and it has facility to identify patterns according to the semantic meaning of the input data. In this paper, we apply a hybrid technique which encompasses both types of learning in the field of word sense disambiguation and show that the high-level order of learning can really improve the accuracy rate of the model. This evidence serves to demonstrate that the internal structures formed by the words do present patterns that, generally, cannot be correctly unveiled by only traditional techniques. Finally, we exhibit the behavior of the model for different weights of the low- and high-level classifiers by plotting decision boundaries. This study helps one to better understand the effectiveness of the model. Copyright (C) EPLA, 2012

Hypoactivity of the central dopaminergic system and autistic-like behavior induced by a single early prenatal exposure to lipopolysaccharide

The aim of the present study was to evaluate the behavioral patterns associated with autism and the prevalence of these behaviors in males and females, to verify whether our model of lipopolysaccharide (LPS) administration represents an experimental model of autism. For this, we prenatally exposed Wistar rats to LPS (100 mu g/kg, intraperitoneally, on gestational day 9.5), which mimics infection by gram-negative bacteria. Furthermore, because the exact mechanisms by which autism develops are still unknown, we investigated the neurological mechanisms that might underlie the behavioral alterations that were observed. Because we previously had demonstrated that prenatal LPS decreases striatal dopamine (DA) and metabolite levels, the striatal dopaminergic system (tyrosine hydroxylase [TH] and DA receptors D1a and D2) and glial cells (astrocytes and microglia) were analyzed by using immunohistochemistry, immunoblotting, and real-time PCR. Our results show that prenatal LPS exposure impaired communication (ultrasonic vocalizations) in male pups and learning and memory (T-maze spontaneous alternation) in male adults, as well as inducing repetitive/restricted behavior, but did not change social interactions in either infancy (play behavior) or adulthood in females. Moreover, although the expression of DA receptors was unchanged, the experimental animals exhibited reduced striatal TH levels, indicating that reduced DA synthesis impaired the striatal dopaminergic system. The expression of glial cell markers was not increased, which suggests that prenatal LPS did not induce permanent neuroinflammation in the striatum. Together with our previous finding of social impairments in males, the present findings demonstrate that prenatal LPS induced autism-like effects and also a hypoactivation of the dopaminergic system. (c) 2012 Wiley Periodicals, Inc.

Modelling shared attention through relational reinforcement learning

Shared attention is a type of communication very important among human beings. It is sometimes reserved for the more complex form of communication being constituted by a sequence of four steps: mutual gaze, gaze following, imperative pointing and declarative pointing. Some approaches have been proposed in Human-Robot Interaction area to solve part of shared attention process, that is, the most of works proposed try to solve the first two steps. Models based on temporal difference, neural networks, probabilistic and reinforcement learning are methods used in several works. In this article, we are presenting a robotic architecture that provides a robot or agent, the capacity of learning mutual gaze, gaze following and declarative pointing using a robotic head interacting with a caregiver. Three learning methods have been incorporated to this architecture and a comparison of their performance has been done to find the most adequate to be used in real experiment. The learning capabilities of this architecture have been analyzed by observing the robot interacting with the human in a controlled environment. The experimental results show that the robotic head is able to produce appropriate behavior and to learn from sociable interaction.

Network-based stochastic semisupervised learning

Semisupervised learning is a machine learning approach that is able to employ both labeled and unlabeled samples in the training process. In this paper, we propose a semisupervised data classification model based on a combined random-preferential walk of particles in a network (graph) constructed from the input dataset. The particles of the same class cooperate among themselves, while the particles of different classes compete with each other to propagate class labels to the whole network. A rigorous model definition is provided via a nonlinear stochastic dynamical system and a mathematical analysis of its behavior is carried out. A numerical validation presented in this paper confirms the theoretical predictions. An interesting feature brought by the competitive-cooperative mechanism is that the proposed model can achieve good classification rates while exhibiting low computational complexity order in comparison to other network-based semisupervised algorithms. Computer simulations conducted on synthetic and real-world datasets reveal the effectiveness of the model.