Figure - uploaded by Elad Schneidman
Content may be subject to copyright.
Fig. S3. Comparison of model accuracy in describing individual learning, with memory windows 1 and 2. Each pair of connected black dots show the fraction of fitted answers (y-axis) for the same subject on the same rule for W = 1 and W = 2 (x-axis). Red error bars show the population mean and SD; this is done for all of the rules. (A) One-bit; (B) two-bit; (C) three-bit; (D) majority; (E) middle symmetry; and (F) symmetry.
Source publication
Pattern classification learning tasks are commonly used to explore learning strategies in human subjects. The universal and individual traits of learning such tasks reflect our cognitive abilities and have been of interest both psychophysically and clinically. From a computational perspective, these tasks are hard, because the number of patterns an...
Similar publications
The recognition of traffic signs in many state-of-the-art driver assistance systems is performed by statistical pattern classification methods. Traffic signs in European countries share many similarities but also vary in colour, size, and depicted symbols, making it hard to obtain one general classifier with good performance in all countries. Train...
In accelerometry-based activity monitoring, to correctly assess the level and recognize the type of physical activity performed by the monitored subject, it is essential to have the knowledge of the sensor wearing position. For consumer healthcare and lifestyle applications, such as with an activity monitor (AM), to restrict users with a fixed sens...
One of the great aspirations of machine learning is the clustering methods. It consists on categorized a set of similar data into different groups based on related properties. The clustering ensemble is used in aim to improve the performance and the stability of the unsupervised classification methods through the concept of weighting. One of the ma...
Land use pattern classification of remote sensing imagery data is imperative to research that is used in remote sensing applications. Remote sensing (RS) technologies were exploited to mine some of the significant spatially variable factors, such as land cover and land use (LCLU), from satellite images of remote arid areas in Karnataka, India. Four...
A novel instance-based algorithm for pattern classification is presented and evaluated in this paper. This new method is motivated by the challenge of pattern classifications where only limited and/or noisy training data are available. For every classification, the proposed system transforms the query data and the training templates based on their...
Citations
... The fields of AI and robotics have developed other, related models, such as the mixture-of-experts architecture [113,135]. Although this model fell out of favour due to over-parameterized and excessively constrained implementations, new work is showing its merit as a neurocognitive mechanism [23,26,136,137]. Mixture-of-experts can flexibly integrate multiple representations, including multiple levels of abstractions at once: an internal (implicit) controller selects the best expert(s) based on a 'responsibility' signal (derived from priors and prediction errors from each expert) [23,114,115] (Figure I, bottom-right). ...
Colombian drug lord Pablo Escobar, while on the run, purportedly burned two million dollars in banknotes to keep his daughter warm. A stark reminder that, in life, circumstances and goals can quickly change, forcing us to reassess and modify our values on-the-fly. Studies in decision-making and neuroeconomics have often implicitly equated value to reward, emphasising the hedonic and automatic aspect of the value computation, while overlooking its functional (concept-like) nature. Here we outline the computational and biological principles that enable the brain to compute the usefulness of an option or action by creating abstractions that flexibly adapt to changing goals. We present different algorithmic architectures, comparing ideas from artificial intelligence (AI) and cognitive neuroscience with psychological theories and, when possible, drawing parallels.
... We anticipate a variety of use cases for PsyTrack. First, experimenters training animals on binary decision-making tasks can use PsyTrack to better understand the diverse range of behavioral strategies seen in early training (Cohen and Schneidman, 2013). This will facilitate the design and validation of new training strategies, which could ultimately open the door to more complex tasks and faster training times. ...
Decision-making strategies evolve during training and can continue to vary even in well-trained animals. However, studies of sensory decision-making tend to characterize behavior in terms of a fixed psychometric function that is fit only after training is complete. Here, we present PsyTrack, a flexible method for inferring the trajectory of sensory decision-making strategies from choice data. We apply PsyTrack to training data from mice, rats, and human subjects learning to perform auditory and visual decision-making tasks. We show that it successfully captures trial-to-trial fluctuations in the weighting of sensory stimuli, bias, and task-irrelevant covariates such as choice and stimulus history. This analysis reveals dramatic differences in learning across mice and rapid adaptation to changes in task statistics. PsyTrack scales easily to large datasets and offers a powerful tool for quantifying time-varying behavior in a wide variety of animals and tasks.
... Learning necessitates changes in behavior, 25 frustrating conventional analyses that are better suited to the relatively stable behavior seen posttraining. Furthermore, behavior during early training is notoriously diverse: different subjects within the same population may utilize entirely different strategies upon encountering a novel task (Cohen and Schneidman, 2013). All research dependent upon animals consistently making accurate decisions must endure the poorly understood and resource-hungry bottleneck of animal training. ...
Understanding how animals update their decision-making behavior over time is an important problem in neuroscience. Decision-making strategies evolve over the course of learning, and continue to vary even in well-trained animals. However, the standard suite of behavioral analysis tools is ill-equipped to capture the dynamics of these strategies. Here, we present a flexible method for characterizing time-varying behavior during decision-making experiments. We show that it successfully captures trial-to-trial changes in an animal's sensitivity to not only task-relevant stimuli, but also task-irrelevant covariates such as choice, reward, and stimulus history. We use this method to derive insights from training data collected in mice, rats, and human subjects performing auditory discrimination and visual detection tasks. With this approach, we uncover the detailed evolution of an animal's strategy during learning, including adaptation to time-varying task statistics, suppression of sub-optimal strategies, and shared behavioral dynamics between subjects within an experimental population.
... Accordingly, such tasks are commonly used to explore learning strategies in humans [1][2][3][4][5], as well as clinical implications [6][7][8][9]. Recent studies have shown that performance depends on rule complexity [10] and can be predicted using models that rely on high-order features of the stimulus with individual priors [11]. Studies of learning rule-based classification in monkeys have ascribed a major and complementary role to the striatum and regions of the prefrontal-cortex (PFC) [12,13]. ...
... We recorded and analyzed dynamic changes in neural representation while monkeys learned to classify multi-cue patterns based on a rich set of potential rules. In these natural conditions of varying eight different rules, the animals' performance varied across sessions and rule types as previously shown in humans [11]. Nevertheless, we found that neurons in both the striatum and the dACC change their firing pattern to eventually represent the learned rule. ...
Primates can quickly and advantageously adopt complex rule-based behaviors. We studied acquisition of rule-based classification while recording single neurons in the dorsal-anterior-cingulate-cortex (dACC) and the Striatum. Monkeys performed trial-by-trial classification on a rich set of multi-cue patterns, allowing de-novo rule-learning with varying conceptual complexities every few days. To examine neural dynamics during the learning itself, we represent each rule with a spanning set of the space formed by the stimulus features. Because neural preference can be expressed by feature combinations, we can track neural dynamics in geometrical terms in this space, allowing a compact universal description of neural trajectories by observing changes in either vector-magnitude and/or angle-to-rule. We find that a large fraction of cells in both regions follow the behavior during learning. Neurons in the dACC mainly rotate towards the rule, suggesting an increase in selectivity that approximates the rule; whereas in the Putamen we additionally find a prominent magnitude increase, suggesting strengthening of confidence. Moreover, magnitude increases in the striatum followed rotation in the dACC, and finally, the neural policy at the end of the session predicted next-day behavior. Using this novel framework enables tracking of neural dynamics during learning and suggests differential roles of confidence and policy for the different brain regions.
... The research hypothesis premised that students in the treatment group who employed the Journey Method using Google Maps would outperform the students in the control group in memorizing the given content. Yet, at no point in this dissertation is memorization contended as a pure skill to perform better on standardized tests, in isolation of all the other necessary skills, like comprehension, recognizing patterns, and deductive reasoning (Cohen & Schneidman, 2013). ...
This mixed methods research studied the effect of the Journey Method with Google Maps, a memorization strategy for memorizing informational text, on the recallability levels in high school biology students at an urban high school in Paterson, New Jersey. This memorization technique was based on an ancient memorization strategy, called the method of loci, in which images of the information that needed to be committed to memory are placed along a real or a fictional route or a journey, hence The Journey Method. By following along that route, the brain is expected to recall those images in a sequence of stops along that route. For this study, the Google Maps application was used to create journeys or routes to place images of the biology subject content, in addition to hand-drawn maps of the routes that were familiar to the subjects of
... The A-prime score in the first, prelearning, test block (T1) was used as the participant performance baseline, and the A-prime scores in the following test blocks (T2 to T9) and learning blocks (L1 to L8) were rescaled, respectively, to reflect performance change from pre-learning. After rescaling the A-prime scores, due to the relatively small number of trials (eight) in each test block, we smoothed the data across the eight post-learning test blocks using a running average with a window of three test blocks (see similar analyses in Schneidman, 2013, andGrill-Spector, 2012). This reduced the impact of intermediate trends in the learning trajectories, while preserving generic characteristics of the trajectories. ...
As children start attending school they are more likely to face situations where they have to autonomously learn about novel object categories (e.g. by reading a picture book with descriptions of novel animals). Such autonomous observational category learning (OCL) gradually complements interactive feedback-based category learning (FBCL), where a child hypothesizes about the nature of a novel object, acts based on his prediction, and then receives feedback indicating the correctness of his prediction. Here we tested OCL and FBCL skills of elementary school children and adults. In both conditions, participants performed complex rule-based categorization tasks that required associating novel objects with novel category-labels. We expected children to perform better in FBCL tasks than in OCL tasks, whereas adults to be skilled in both tasks. As hypothesized, in early-phase learning children performed better in FBCL tasks than in OCL tasks. Unexpectedly, adults performed somewhat better in OCL tasks. Early-phase FBCL performance in the two age groups was matched, but the OCL performance of adults was higher than that of children. In late-phase learning there was only an age group main effect (adults > children). Moreover, performance in post-learning categorization tasks, that did not require label recollection, indicated that in FBCL tasks children were likely to directly learn the associations between an object and a category label, whereas in the OCL tasks they were likely to first learn which feature-dimensions were relevant. These findings shed light on developmental changes in cognitive control and learning mechanisms. Implications for educational settings are discussed.
© 2015 John Wiley & Sons Ltd.
Learning new rules and adopting novel behavioral policies is a prominent adaptive behavior of primates. We studied the dynamics of single neurons in the dorsal anterior cingulate cortex and putamen of monkeys while they learned new classification tasks every few days over a fixed set of multi-cue patterns. Representing the rules and the neuronal selectivity as vectors in the space spanned by a set of stimulus features allowed us to characterize neuronal dynamics in geometrical terms. We found that neurons in the cingulate cortex mainly rotated toward the rule, implying a policy search, whereas neurons in the putamen showed a magnitude increase that followed the rotation of cortical neurons, implying strengthening of confidence for the newly acquired rule-based policy. Further, the neural representation at the end of a session predicted next-day behavior, reflecting overnight retention. The novel framework for characterization of neural dynamics suggests complementing roles for the putamen and the anterior cingulate cortex.
Variability in drug responsivity has prompted the development of Personalized Medicine, which has shown great promise in utilizing genotypic information to develop safer and more effective drug regimens for patients. Similarly, individual variability in learning outcomes has puzzled researchers who seek to create optimal learning environments for students. “Personalized Learning” seeks to identify genetic, neural and behavioral predictors of individual differences in learning and aims to use predictors to help create optimal teaching paradigms. Evidence for Personalized Learning can be observed by connecting research in pharmacogenomics, cognitive genetics and behavioral experiments across domains of learning, which provides a framework for conducting empirical studies from the laboratory to the classroom and holds promise for addressing learning effectiveness in the individual learners. Evidence can also be seen in the subdomain of speech learning, thus providing initial support for the applicability of Personalized Learning to language.
The ability to record the joint activity of large groups of neurons would allow for direct study of information representation and computation at the level of whole circuits in the brain. The combinatorial space of potential population activity patterns and neural noise imply that it would be impossible to directly map the relations between stimuli and population responses. Understanding of large neural population codes therefore depends on identifying simplifying design principles. We review recent results showing that strongly correlated population codes can be explained using minimal models that rely on low order relations among cells. We discuss the implications for large populations, and how such models allow for mapping the semantic organization of the neural codebook and stimulus space, and decoding.
In recent years, ideas from the computational field of reinforcement learning have revolutionized the study of learning in the brain, famously providing new, precise theories of how dopamine affects learning in the basal ganglia. However, reinforcement learning algorithms are notorious for not scaling well to multidimensional environments, as is required for real-world learning. We hypothesized that the brain naturally reduces the dimensionality of real-world problems to only those dimensions that are relevant to predicting reward, and conducted an experiment to assess by what algorithms and with what neural mechanisms this "representation learning" process is realized in humans. Our results suggest that a bilateral attentional control network comprising the intraparietal sulcus, precuneus, and dorsolateral prefrontal cortex is involved in selecting what dimensions are relevant to the task at hand, effectively updating the task representation through trial and error. In this way, cortical attention mechanisms interact with learning in the basal ganglia to solve the "curse of dimensionality" in reinforcement learning.
Copyright © 2015 the authors 0270-6474/15/358145-13$15.00/0.
