Data Mining the Brain to Decode the Mind (original) (raw)
Related papers
Representation, Pattern Information, and Brain Signatures: From Neurons to Neuroimaging
Neuron, 2018
Human neuroimaging research has transitioned from mapping local effects to developing predictive models of mental events that integrate information distributed across multiple brain systems. Here we review work demonstrating how multivariate predictive models have been utilized to provide quantitative, falsifiable predictions; establish mappings between brain and mind with larger effects than traditional approaches; and help explain how the brain represents mental constructs and processes. Although there is increasing progress toward the first two of these goals, models are only beginning to address the latter objective. By explicitly identifying gaps in knowledge, research programs can move deliberately and programmatically toward the goal of identifying brain representations underlying mental states and processes.
A Shared Vision for Machine Learning in Neuroscience
The Journal of neuroscience : the official journal of the Society for Neuroscience, 2018
With ever-increasing advancements in technology, neuroscientists are able to collect data in greater volumes and with finer resolution. The bottleneck in understanding how the brain works is consequently shifting away from the amount and type of data we can collect and toward what we actually do with the data. There has been a growing interest in leveraging this vast volume of data across levels of analysis, measurement techniques, and experimental paradigms to gain more insight into brain function. Such efforts are visible at an international scale, with the emergence of big data neuroscience initiatives, such as the BRAIN initiative (Bargmann et al., 2014), the Human Brain Project, the Human Connectome Project, and the National Institute of Mental Health's Research Domain Criteria initiative. With these large-scale projects, much thought has been given to data-sharing across groups (Poldrack and Gorgolewski, 2014; Sejnowski et al., 2014); however, even with such data-sharing i...
Ghosts in machine learning for cognitive neuroscience: Moving from data to theory.
Neuroimage, 2018
The application of machine learning methods to neuroimaging data has fundamentally altered the field of cognitive neuroscience. Future progress in understanding brain function using these methods will require addressing a number of key methodological and interpretive challenges. Because these challenges often remain unseen and metaphorically "haunt" our efforts to use these methods to understand the brain, we refer to them as "ghosts". In this paper, we describe three such ghosts, situate them within a more general framework from philosophy of science, and then describe steps to address them. The first ghost arises from difficulties in determining what information machine learning classifiers use for decoding. The second ghost arises from the interplay of experimental design and the structure of information in the brain - that is, our methods embody implicit assumptions about information processing in the brain, and it is often difficult to determine if those assumptions are satisfied. The third ghost emerges from our limited ability to distinguish information that is merely decodable from the brain from information that is represented and used by the brain. Each of the three ghosts place limits on the interpretability of decoding research in cognitive neuroscience. There are no easy solutions, but facing these issues squarely will provide a clearer path to understanding the nature of representation and computation in the human brain.
Decoding the Large Scale Structure of the Brain
Brain-imaging research has largely focused on localizing patterns of activity related to specific mental processes, but recent work has shown that mental states can be identified from neuroimaging data using statistical classifiers. We investigated whether this approach could be extended to predict the mental state of an individual using a statistical classifier trained on other individuals, and whether the information gained in doing so could provide new insights into how mental processes are organized in the brain. Using a variety of classifier techniques, we achieved cross-validated classification accuracy of 80% across individuals (chance 5 13%). Using a neural network classifier, we recovered a low-dimensional representation common to all the cognitive-perceptual tasks in our data set, and we used an ontology of cognitive processes to determine the cognitive concepts most related to each dimension. These results revealed a small organized set of large-scale networks that map cognitive processes across a highly diverse set of mental tasks, suggesting a novel way to characterize the neural basis of cognition.
2020
The thesis explores the role machine learning methods play in creating intuitive computational models of neural processing. We take the perspective that, combined with interpretability techniques, machine learning could replace human modeler and shift the focus of human effort from creating the models to extracting the knowledge from the already-made models and articulating that knowledge into intuitive representations. Automatic model-building methods can process larger volumes of data and explore more computationally complex relationships than a human modeler could. This perspective makes the case in favor of the larger role that exploratory and data-driven approach to computational neuroscience could play while coexisting alongside the traditional hypothesis-driven approach. We provide an example of how an intuitive model can be extracted from machinelearned knowledge, explore major machine learning algorithms in the context of the knowledge representation they employ, and propose a taxonomy of machine learning algorithms based on the knowledge representation that is driving their decision-making process. We exemplify the illustrated approach in the context of the knowledge representation taxonomy with three research projects that employ interpretability techniques on top of machine learning methods at three different levels of neural organization. In each case we demonstrate the applicability of the approach and present the neuroscientific knowledge it allowed us to extract. The first study (Chapter 3) explores feature importance analysis of a random forest decoder trained on intracerebral recordings from 100 human subjects to identify spectrotemporal signatures that characterize local neural activity during the task of visual categorization. The second study (Chapter 4) employs representation similarity analysis to compare the neural responses of the areas along the ventral stream with the activations of the layers of a deep convolutional neural network. The analysis allowed us to make conclusions and observations about the hierarchical organization of the human visual cortex and the similarities between the biological and an artificial system of vision. The third study (Chapter 5) proposes a method that allows test subjects to visually explore the state representation of their neural signal in real time. This is achieved by using a topology-preserving dimensionality reduction technique that allows to transform the neural data from the multidimensional representation used by the computer into a two-dimensional representation a human can grasp. Taken together, the approach, the taxonomy, and the examples, present a strong case for the applicability of machine learning methods in conjunction with interpretability techniques to automatic knowledge discovery in neuroscience. Seen from this perspective, machine learning models cease to be mere statistical black boxes and, by capturing the underlying dynamics of real life processes, reintroduce themselves as candidate models of reality.
Interpretability of Multivariate Brain Maps in Brain Decoding: Definition and Quantification
2016
Brain decoding is a popular multivariate approach for hypothesis testing in neuroimaging. Linear classifiers are widely employed in the brain decoding paradigm to discriminate among experimental conditions. Then, the derived linear weights are visualized in the form of multivariate brain maps to further study the spatio-temporal patterns of underlying neural activities. It is well known that the brain maps derived from weights of linear classifiers are hard to interpret because of high correlations between predictors, low signal to noise ratios, and the high dimensionality of neuroimaging data. Therefore, improving the interpretability of brain decoding approaches is of primary interest in many neuroimaging studies. Despite extensive studies of this type, at present, there is no formal definition for interpretability of multivariate brain maps. As a consequence, there is no quantitative measure for evaluating the interpretability of different brain decoding methods. In this paper, first, we present a theoretical definition of interpretability in brain decoding; we show that the interpretability of multivariate brain maps can be decomposed into their reproducibility and representativeness. Second, as an application of the proposed definition, we formalize a heuristic method for approximating the interpretability of multivariate brain maps in a binary magnetoencephalography (MEG) decoding scenario. Third, we pro
The Future of Cognitive Neuroscience? Reverse Inference in Focus
This article presents and discusses one of the most prominent inferential strategies currently employed in cognitive neuropsychology, namely, reverse inference. Simply put, this is the practice of inferring, in the context of experimental tasks, the engagement of cognitive processes from locations or patterns of neural activation. This technique is notoriously controversial because , critics argue, it presupposes the problematic assumption that neural areas are functionally selective. We proceed as follows. We begin by introducing the basic structure of traditional "location-based" reverse inference (§1) and discuss the influential lack of selectivity objection (§2). Next, we rehearse various ways of responding to this challenge and provide some reasons for cautious optimism (§3). The second part of the essay presents a more recent development: "pattern-decoding reverse inference" (§4). This inferential strategy, we maintain, provides an even more convincing response to the lack of selectivity charge. Due to this and other methodological advantages, it is now a prominent component in the toolbox of cognitive neuropsychology (§5). Finally, we conclude by drawing some implications for philosophy of science and philosophy of mind (§6).
Two Kinds of Reverse Inference in Cognitive Neuroscience
This essay examines the prospects and limits of 'reverse inferring' cognitive processes from neural data, a technique commonly used in cognitive neuroscience for discriminating between competing psychological hypotheses. Specifically, we distinguish between two main types of reverse inference. The first kind of inference moves from the locations of neural activation to the underlying cognitive processes. We illustrate this strategy by presenting a well-known example involving mirror neurons and theories of low-level mind-reading, and discuss some general methodological problems. Next we present the second type of reverse inference by discussing an example from recognition memory research. These inferences, based on pattern-decoding techniques, do not presuppose strong assumptions about the functions of particular neural locations. Consequently, while they have been largely ignored in methodological critiques, they overcome important objections plaguing traditional methods.