Dissociating explicit and procedural-learning based systems of perceptual category learning

A fundamental question is whether people have available one category learning system, or many. Most multiple systems advocates postulate one explicit and one implicit system. Although there is much agreement about the nature of the explicit system, there is less agreement about the nature of the implicit system. In this article, we review a dual systems theory of category learning called competition between verbal and implicit systems (COVIS) developed by Ashby et al. The explicit system dominates the learning of verbalizable, rule-based category structures and is mediated by frontal brain areas such as the anterior cingulate, prefrontal cortex (PFC), and head of the caudate nucleus. The implicit system, which uses procedural learning, dominates the learning of non-verbalizable, information-integration category structures, and is mediated by the tail of the caudate nucleus and a dopamine-mediated reward signal. We review nine studies that test six a priori predictions from COVIS, each of which is supported by the data.     

The Balance-Scale Task Revisited: A Comparison of Statistical Models for Rule-Based and Information-Integration Theories of Proportional Reasoning

We propose and test three statistical models for the analysis of children’s responses to the balance scale task, a seminal task to study proportional reasoning. We use a latent class modelling approach to formulate a rule-based latent class model (RB LCM) following from a rule-based perspective on proportional reasoning and a new statistical model, the Weighted Sum Model, following from an information-integration approach. Moreover, a hybrid LCM using item covariates is proposed, combining aspects of both a rule-based and information-integration perspective. These models are applied to two different datasets, a standard paper-and-pencil test dataset (N = 779), and a dataset collected within an online learning environment that included direct feedback, time-pressure, and a reward system (N = 808). For the paper-and-pencil dataset the RB LCM resulted in the best fit, whereas for the online dataset the hybrid LCM provided the best fit. The standard paper-and-pencil dataset yielded more evidence for distinct solution rules than the online data set in which quantitative item characteristics are more prominent in determining responses. These results shed new light on the discussion on sequential rule-based and information-integration perspectives of cognitive development.     

Rule-Based Category Learning in Children: The Role of Age and Executive Functioning

Rule-based category learning was examined in 4–11 year-olds and adults. Participants were asked to learn a set of novel perceptual categories in a classification learning task. Categorization performance improved with age, with younger children showing the strongest rule-based deficit relative to older children and adults. Model-based analyses provided insight regarding the type of strategy being used to solve the categorization task, demonstrating that the use of the task appropriate strategy increased with age. When children and adults who identified the correct categorization rule were compared, the performance deficit was no longer evident. Executive functions were also measured. While both working memory and inhibitory control were related to rule-based categorization and improved with age, working memory specifically was found to marginally mediate the age-related improvements in categorization. When analyses focused only on the sample of children, results showed that working memory ability and inhibitory control were associated with categorization performance and strategy use. The current findings track changes in categorization performance across childhood, demonstrating at which points performance begins to mature and resemble that of adults. Additionally, findings highlight the potential role that working memory and inhibitory control may play in rule-based category learning.     

Perceptual learning in children during perception of the emotional component of speech utterances: The influence of age and gender

Perceptual learning was examined during perception of the emotional component of speech when a valid signal was presented against a background of noise to 7-to 17-year-old boys and girls. Accuracy of emotion recognition (AR) and response time (RT) were recorded in two consecutive test series. Analysis of variance of the results demonstrated a high significance of the series order for both the RT and the AR. The study revealed the ontogenetic features of the characteristics of perceptual learning under these conditions. The most notable changes were observed in the time of discriminating emotions. Mathematically, perceptual learning was modeled exponentially. Model parameters such as the rate of perceptual learning and the initial level of perceptual performance were estimated. It was found that the rate of learning, calculated for both the AR and the RT, had close values and similar trends of changes in both the male and the female samples. This suggests that the rate of perceptual learning is an invariant characteristic of perception of the emotional component of speech in a child. The initial level of perceptual performance increased as a function of age, more notably in the boys than in the girls. The initial level of the RT decreased with age in the girls and almost did not depend on age in the boys. However, the integral AR suggests that the age-related changes in the initial level of the integral perception of emotions are identical in girls and boys.     

Moving the Weber Fraction: The Perceptual Precision for Moment of Inertia Increases with Exploration Force

How does the magnitude of the exploration force influence the precision of haptic perceptual estimates? To address this question, we examined the perceptual precision for moment of inertia (i.e., an object''s “angular mass”) under different force conditions, using the Weber fraction to quantify perceptual precision. Participants rotated a rod around a fixed axis and judged its moment of inertia in a two-alternative forced-choice task. We instructed different levels of exploration force, thereby manipulating the magnitude of both the exploration force and the angular acceleration. These are the two signals that are needed by the nervous system to estimate moment of inertia. Importantly, one can assume that the absolute noise on both signals increases with an increase in the signals'' magnitudes, while the relative noise (i.e., noise/signal) decreases with an increase in signal magnitude. We examined how the perceptual precision for moment of inertia was affected by this neural noise. In a first experiment we found that a low exploration force caused a higher Weber fraction (22%) than a high exploration force (13%), which suggested that the perceptual precision was constrained by the relative noise. This hypothesis was supported by the result of a second experiment, in which we found that the relationship between exploration force and Weber fraction had a similar shape as the theoretical relationship between signal magnitude and relative noise. The present study thus demonstrated that the amount of force used to explore an object can profoundly influence the precision by which its properties are perceived.     

Human Decision Making Based on Variations in Internal Noise: An EEG Study

Perceptual decision making is prone to errors, especially near threshold. Physiological, behavioural and modeling studies suggest this is due to the intrinsic or ‘internal’ noise in neural systems, which derives from a mixture of bottom-up and top-down sources. We show here that internal noise can form the basis of perceptual decision making when the external signal lacks the required information for the decision. We recorded electroencephalographic (EEG) activity in listeners attempting to discriminate between identical tones. Since the acoustic signal was constant, bottom-up and top-down influences were under experimental control. We found that early cortical responses to the identical stimuli varied in global field power and topography according to the perceptual decision made, and activity preceding stimulus presentation could predict both later activity and behavioural decision. Our results suggest that activity variations induced by internal noise of both sensory and cognitive origin are sufficient to drive discrimination judgments.     

Using rule-based machine learning for candidate disease gene prioritization and sample classification of cancer gene expression data

Microarray data analysis has been shown to provide an effective tool for studying cancer and genetic diseases. Although classical machine learning techniques have successfully been applied to find informative genes and to predict class labels for new samples, common restrictions of microarray analysis such as small sample sizes, a large attribute space and high noise levels still limit its scientific and clinical applications. Increasing the interpretability of prediction models while retaining a high accuracy would help to exploit the information content in microarray data more effectively. For this purpose, we evaluate our rule-based evolutionary machine learning systems, BioHEL and GAssist, on three public microarray cancer datasets, obtaining simple rule-based models for sample classification. A comparison with other benchmark microarray sample classifiers based on three diverse feature selection algorithms suggests that these evolutionary learning techniques can compete with state-of-the-art methods like support vector machines. The obtained models reach accuracies above 90% in two-level external cross-validation, with the added value of facilitating interpretation by using only combinations of simple if-then-else rules. As a further benefit, a literature mining analysis reveals that prioritizations of informative genes extracted from BioHEL's classification rule sets can outperform gene rankings obtained from a conventional ensemble feature selection in terms of the pointwise mutual information between relevant disease terms and the standardized names of top-ranked genes.     

One Giant Leap for Categorizers: One Small Step for Categorization Theory

We explore humans’ rule-based category learning using analytic approaches that highlight their psychological transitions during learning. These approaches confirm that humans show qualitatively sudden psychological transitions during rule learning. These transitions contribute to the theoretical literature contrasting single vs. multiple category-learning systems, because they seem to reveal a distinctive learning process of explicit rule discovery. A complete psychology of categorization must describe this learning process, too. Yet extensive formal-modeling analyses confirm that a wide range of current (gradient-descent) models cannot reproduce these transitions, including influential rule-based models (e.g., COVIS) and exemplar models (e.g., ALCOVE). It is an important theoretical conclusion that existing models cannot explain humans’ rule-based category learning. The problem these models have is the incremental algorithm by which learning is simulated. Humans descend no gradient in rule-based tasks. Very different formal-modeling systems will be required to explain humans’ psychology in these tasks. An important next step will be to build a new generation of models that can do so.     

Perceptual learning: specificity versus generalization

Perceptual learning improves performance on many tasks, from orientation discrimination to the identification of faces. Although conventional wisdom considered sensory cortices as hard-wired, the specificity of improvement achieved through perceptual learning indicates an involvement of early sensory cortices. These cortices might be more plastic than previously assumed, and both sum-potential and single cell recordings indeed demonstrate plasticity of neuronal responses of these sensory cortices. However, for learning to be optimally useful, it must generalize to other tasks. Further research on perceptual learning should therefore, in my opinion, investigate first, the conditions for generalization of training-induced improvement, second, its use for teaching and rehabilitation, and third, its dependence on pharmacological agents.     

Monetary Reward Modulates Task-Irrelevant Perceptual Learning for Invisible Stimuli

Task Irrelevant Perceptual Learning (TIPL) shows that the brain’s discriminative capacity can improve also for invisible and unattended visual stimuli. It has been hypothesized that this form of “unconscious” neural plasticity is mediated by an endogenous reward mechanism triggered by the correct task performance. Although this result has challenged the mandatory role of attention in perceptual learning, no direct evidence exists of the hypothesized link between target recognition, reward and TIPL. Here, we manipulated the reward value associated with a target to demonstrate the involvement of reinforcement mechanisms in sensory plasticity for invisible inputs. Participants were trained in a central task associated with either high or low monetary incentives, provided only at the end of the experiment, while subliminal stimuli were presented peripherally. Our results showed that high incentive-value targets induced a greater degree of perceptual improvement for the subliminal stimuli, supporting the role of reinforcement mechanisms in TIPL.     

Unaffected perceptual thresholds for biological and non-biological form-from-motion perception in autism spectrum conditions

Background

Perception of biological motion is linked to the action perception system in the human brain, abnormalities within which have been suggested to underlie impairments in social domains observed in autism spectrum conditions (ASC). However, the literature on biological motion perception in ASC is heterogeneous and it is unclear whether deficits are specific to biological motion, or might generalize to form-from-motion perception.

Methodology and Principal Findings

We compared psychophysical thresholds for both biological and non-biological form-from-motion perception in adults with ASC and controls. Participants viewed point-light displays depicting a walking person (Biological Motion), a translating rectangle (Structured Object) or a translating unfamiliar shape (Unstructured Object). The figures were embedded in noise dots that moved similarly and the task was to determine direction of movement. The number of noise dots varied on each trial and perceptual thresholds were estimated adaptively. We found no evidence for an impairment in biological or non-biological object motion perception in individuals with ASC. Perceptual thresholds in the three conditions were almost identical between the ASC and control groups.

Discussion and Conclusions

Impairments in biological motion and non-biological form-from-motion perception are not across the board in ASC, and are only found for some stimuli and tasks. We discuss our results in relation to other findings in the literature, the heterogeneity of which likely relates to the different tasks performed. It appears that individuals with ASC are unaffected in perceptual processing of form-from-motion, but may exhibit impairments in higher order judgments such as emotion processing. It is important to identify more specifically which processes of motion perception are impacted in ASC before a link can be made between perceptual deficits and the higher-level features of the disorder.     

Perceptual learning and dynamic changes in primary visual cortex

Perceptual learning is the improved performance that follows practice in a perceptual task. In this issue of Neuron, Yotsumoto et al. use fMRI to show that stimuli presented at the location used in training initially evoke greater activation in primary visual cortex than stimuli presented elsewhere, but this difference disappears once learning asymptotes.     

The neural basis of perceptual learning

Perceptual learning is a lifelong process. We begin by encoding information about the basic structure of the natural world and continue to assimilate information about specific patterns with which we become familiar. The specificity of the learning suggests that all areas of the cerebral cortex are plastic and can represent various aspects of learned information. The neural substrate of perceptual learning relates to the nature of the neural code itself, including changes in cortical maps, in the temporal characteristics of neuronal responses, and in modulation of contextual influences. Top-down control of these representations suggests that learning involves an interaction between multiple cortical areas.     

Dyslexic adults can learn from repeated stimulus presentation but have difficulties in excluding external noise

We examined whether the characteristic impairments of dyslexia are due to a deficit in excluding external noise or a deficit in taking advantage of repeated stimulus presentation. We compared non-impaired adults and adults with poor reading performance on a visual letter detection task that varied two aspects: the presence or absence of background visual noise, and a small or large stimulus set. There was no interaction between group and stimulus set size, indicating that the poor readers took advantage of repeated stimulus presentation as well as the non-impaired readers. The poor readers had higher thresholds than non-impaired readers in the presence of high external noise, but not in the absence of external noise. The results support the hypothesis that an external noise exclusion deficit, not a perceptual anchoring deficit, impairs reading for adults.     

Do Basal Ganglia Amplify Willed Action by Stochastic Resonance? A Model

Basal ganglia are usually attributed a role in facilitating willed action, which is found to be impaired in Parkinson''s disease, a pathology of basal ganglia. We hypothesize that basal ganglia possess the machinery to amplify will signals, presumably weak, by stochastic resonance. Recently we proposed a computational model of Parkinsonian reaching, in which the contributions from basal ganglia aid the motor cortex in learning to reach. The model was cast in reinforcement learning framework. We now show that the above basal ganglia computational model has all the ingredients of stochastic resonance process. In the proposed computational model, we consider the problem of moving an arm from a rest position to a target position: the two positions correspond to two extrema of the value function. A single kick (a half-wave of sinusoid, of sufficiently low amplitude) given to the system in resting position, succeeds in taking the system to the target position, with high probability, only at a critical noise level. But for suboptimal noise levels, the model arm''s movements resemble Parkinsonian movement symptoms like akinetic rigidity (low noise) and dyskinesias (high noise).     

Sound Symbolism Facilitates Word Learning in 14-Month-Olds

Sound symbolism, or the nonarbitrary link between linguistic sound and meaning, has often been discussed in connection with language evolution, where the oral imitation of external events links phonetic forms with their referents (e.g., Ramachandran & Hubbard, 2001). In this research, we explore whether sound symbolism may also facilitate synchronic language learning in human infants. Sound symbolism may be a useful cue particularly at the earliest developmental stages of word learning, because it potentially provides a way of bootstrapping word meaning from perceptual information. Using an associative word learning paradigm, we demonstrated that 14-month-old infants could detect Köhler-type (1947) shape-sound symbolism, and could use this sensitivity in their effort to establish a word-referent association.     

Frequency-dependent predation, crypsis and aposematic coloration

Frequency-dependent predation may maintain or prevent colour pattern polymorphisms in prey, and can be caused by a variety of biological phenomena, including perceptual processes (search images), optimal foraging and learning. Most species are preyed upon by more than one predator species, which are likely to differ in foraging styles, perceptual and learning abilities. Depending upon the interaction between predator vision, background and colour pattern parameters, certain morphs may be actively maintained in some conditions and not in others, even with the same predators. More than one kind of predator will also affect stability, and only slight changes in conditions can cause a transition between polymorphism and monomorphism. Frequency-dependent selection is not a panacea for the explanation of variation in animal colour patterns, although it may be important in some systems.     

Involvement of the visual change detection process in facilitating perceptual alternation in the bistable image

A bistable image induces one of two perceptual alternatives. When the bistable visual image is continuously viewed, the percept of the image alternates from one possible percept to the other. Perceptual alternation was previously reported to be induced by an exogenous perturbation in the bistable image, and this perturbation was theoretically interpreted to cause neural noise, prompting a transition between two stable perceptual states. However, little is known experimentally about the visual processing of exogenously driven perceptual alternation. Based on the findings of a previous behavioral study (Urakawa et al. in Perception 45:474–482, 2016), the present study hypothesized that the automatic visual change detection process, which is relevant to the detection of a visual change in a sequence of visual events, has an enhancing effect on the induction of perceptual alternation, similar to neural noise. In order to clarify this issue, we developed a novel experimental paradigm in which visual mismatch negativity (vMMN), an electroencephalographic brain response that reflects visual change detection, was evoked while participants continuously viewed the bistable image. In terms of inter-individual differences in neural and behavioral data, we found that enhancements in the peak amplitude of vMMN1, early vMMN at a latency of approximately 150 ms, correlated with increases in the proportion of perceptual alternation across participants. Our results indicate the involvement of automatic visual change detection in the induction of perceptual alternation, similar to neural noise, thereby providing a deeper insight into the neural mechanisms underlying exogenously driven perceptual alternation in the bistable image.     

Stimulus Familiarity Affects Perceptual Restoration in the European Starling (Sturnus vulgaris)

Background

Humans can easily restore a speech signal that is temporally masked by an interfering sound (e.g., a cough masking parts of a word in a conversation), and listeners have the illusion that the speech continues through the interfering sound. This perceptual restoration for human speech is affected by prior experience. Here we provide evidence for perceptual restoration in complex vocalizations of a songbird that are acquired by vocal learning in a similar way as humans learn their language.

Methodology/Principal Findings

European starlings were trained in a same/different paradigm to report salient differences between successive sounds. The birds'' response latency for discriminating between a stimulus pair is an indicator for the salience of the difference, and these latencies can be used to evaluate perceptual distances using multi-dimensional scaling. For familiar motifs the birds showed a large perceptual distance if discriminating between song motifs that were muted for brief time periods and complete motifs. If the muted periods were filled with noise, the perceptual distance was reduced. For unfamiliar motifs no such difference was observed.

Conclusions/Significance

The results suggest that starlings are able to perceptually restore partly masked sounds and, similarly to humans, rely on prior experience. They may be a suitable model to study the mechanism underlying experience-dependent perceptual restoration.     

Characterizing the Impact of Category Uncertainty on Human Auditory Categorization Behavior

Categorization is an important cognitive process. However, the correct categorization of a stimulus is often challenging because categories can have overlapping boundaries. Whereas perceptual categorization has been extensively studied in vision, the analogous phenomenon in audition has yet to be systematically explored. Here, we test whether and how human subjects learn to use category distributions and prior probabilities, as well as whether subjects employ an optimal decision strategy when making auditory-category decisions. We asked subjects to classify the frequency of a tone burst into one of two overlapping, uniform categories according to the perceived tone frequency. We systematically varied the prior probability of presenting a tone burst with a frequency originating from one versus the other category. Most subjects learned these changes in prior probabilities early in testing and used this information to influence categorization. We also measured each subject''s frequency-discrimination thresholds (i.e., their sensory uncertainty levels). We tested each subject''s average behavior against variations of a Bayesian model that either led to optimal or sub-optimal decision behavior (i.e. probability matching). In both predicting and fitting each subject''s average behavior, we found that probability matching provided a better account of human decision behavior. The model fits confirmed that subjects were able to learn category prior probabilities and approximate forms of the category distributions. Finally, we systematically explored the potential ways that additional noise sources could influence categorization behavior. We found that an optimal decision strategy can produce probability-matching behavior if it utilized non-stationary category distributions and prior probabilities formed over a short stimulus history. Our work extends previous findings into the auditory domain and reformulates the issue of categorization in a manner that can help to interpret the results of previous research within a generative framework.