Large-scale automated synthesis of human functional neuroimaging data
TL;DR: An automated brain-mapping framework that uses text-mining, meta-analysis and machine-learning techniques to generate a large database of mappings between neural and cognitive states is described and validated.
Abstract: The rapid growth of the literature on neuroimaging in humans has led to major advances in our understanding of human brain function but has also made it increasingly difficult to aggregate and synthesize neuroimaging findings. Here we describe and validate an automated brain-mapping framework that uses text-mining, meta-analysis and machine-learning techniques to generate a large database of mappings between neural and cognitive states. We show that our approach can be used to automatically conduct large-scale, high-quality neuroimaging meta-analyses, address long-standing inferential problems in the neuroimaging literature and support accurate 'decoding' of broad cognitive states from brain activity in both entire studies and individual human subjects. Collectively, our results have validated a powerful and generative framework for synthesizing human neuroimaging data on an unprecedented scale.
Citations
More filters
[...]
TL;DR: Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis.
Abstract: Machine Learning is the study of methods for programming computers to learn. Computers are applied to a wide range of tasks, and for most of these it is relatively easy for programmers to design and implement the necessary software. However, there are many tasks for which this is difficult or impossible. These can be divided into four general categories. First, there are problems for which there exist no human experts. For example, in modern automated manufacturing facilities, there is a need to predict machine failures before they occur by analyzing sensor readings. Because the machines are new, there are no human experts who can be interviewed by a programmer to provide the knowledge necessary to build a computer system. A machine learning system can study recorded data and subsequent machine failures and learn prediction rules. Second, there are problems where human experts exist, but where they are unable to explain their expertise. This is the case in many perceptual tasks, such as speech recognition, hand-writing recognition, and natural language understanding. Virtually all humans exhibit expert-level abilities on these tasks, but none of them can describe the detailed steps that they follow as they perform them. Fortunately, humans can provide machines with examples of the inputs and correct outputs for these tasks, so machine learning algorithms can learn to map the inputs to the outputs. Third, there are problems where phenomena are changing rapidly. In finance, for example, people would like to predict the future behavior of the stock market, of consumer purchases, or of exchange rates. These behaviors change frequently, so that even if a programmer could construct a good predictive computer program, it would need to be rewritten frequently. A learning program can relieve the programmer of this burden by constantly modifying and tuning a set of learned prediction rules. Fourth, there are applications that need to be customized for each computer user separately. Consider, for example, a program to filter unwanted electronic mail messages. Different users will need different filters. It is unreasonable to expect each user to program his or her own rules, and it is infeasible to provide every user with a software engineer to keep the rules up-to-date. A machine learning system can learn which mail messages the user rejects and maintain the filtering rules automatically. Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis. Statistics focuses on understanding the phenomena that have generated the data, often with the goal of testing different hypotheses about those phenomena. Data mining seeks to find patterns in the data that are understandable by people. Psychological studies of human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people (concept learning, skill acquisition, strategy change, etc.).
13,246 citations
TL;DR: A meta-analytic summary of the neuroimaging literature on human emotion finds little evidence that discrete emotion categories can be consistently and specifically localized to distinct brain regions, and finds evidence that is consistent with a psychological constructionist approach to the mind.
Abstract: Researchers have wondered how the brain creates emotions since the early days of psychological science. With a surge of studies in affective neuroscience in recent decades, scientists are poised to answer this question. In this target article, we present a meta-analytic summary of the neuroimaging literature on human emotion. We compare the locationist approach (i.e., the hypothesis that discrete emotion categories consistently and specifically correspond to distinct brain regions) with the psychological constructionist approach (i.e., the hypothesis that discrete emotion categories are constructed of more general brain networks not specific to those categories) to better understand the brain basis of emotion. We review both locationist and psychological constructionist hypotheses of brain-emotion correspondence and report meta-analytic findings bearing on these hypotheses. Overall, we found little evidence that discrete emotion categories can be consistently and specifically localized to distinct brain regions. Instead, we found evidence that is consistent with a psychological constructionist approach to the mind: A set of interacting brain regions commonly involved in basic psychological operations of both an emotional and non-emotional nature are active during emotion experience and perception across a range of discrete emotion categories.
1,702 citations
Cites background from "Large-scale automated synthesis of ..."
...…“machine”) learning, could in principle be used to discover which kinds of psychological categories are most consistently and specifically respected in the brain (Mitchell et al. 2008; Yarkoni et al. 2011), potentially providing new tests of locationist versus constructionist approaches to emotion....
[...]
TL;DR: Evidence is provided that theta band activities over the midfrontal cortex appear to reflect a common computation used for realizing the need for cognitive control, and frontal theta is a compelling candidate mechanism by which emergent processes, such as 'cognitive control', may be biophysically realized.
1,589 citations
TL;DR: A quantitative, coordinate-based meta-analysis of 206 published fMRI studies investigating neural correlates of SV identifies two general patterns of SV-correlated brain responses, which appear to constitute a "valuation system," carrying a domain-general SV signal and potentially contributing to value-based decision making.
1,519 citations
Cites background or methods from "Large-scale automated synthesis of ..."
...Chi-square results are converted to z-scores and thresholded at a false discovery rate level of 0.05 to control for multiple comparisons (see Yarkoni et al., 2011 for full details)....
[...]
...(Related points have been noted previously; Yarkoni et al., 2011)....
[...]
...Recently a new and exciting approach to coordinate-based synthesis of neuroimaging literature has been proposed, based on automated text mining and coordinate extraction (Yarkoni et al., 2011)....
[...]
TL;DR: Emerging evidence suggests that atypical engagement of specific subdivisions of the insula within the salience network is a feature of many neuropsychiatric disorders.
Abstract: The brain is constantly bombarded by stimuli, and the relative salience of these inputs determines which are more likely to capture attention. A brain system known as the 'salience network', with key nodes in the insular cortices, has a central role in the detection of behaviourally relevant stimuli and the coordination of neural resources. Emerging evidence suggests that atypical engagement of specific subdivisions of the insula within the salience network is a feature of many neuropsychiatric disorders.
1,484 citations
References
More filters
[...]
TL;DR: Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis.
Abstract: Machine Learning is the study of methods for programming computers to learn. Computers are applied to a wide range of tasks, and for most of these it is relatively easy for programmers to design and implement the necessary software. However, there are many tasks for which this is difficult or impossible. These can be divided into four general categories. First, there are problems for which there exist no human experts. For example, in modern automated manufacturing facilities, there is a need to predict machine failures before they occur by analyzing sensor readings. Because the machines are new, there are no human experts who can be interviewed by a programmer to provide the knowledge necessary to build a computer system. A machine learning system can study recorded data and subsequent machine failures and learn prediction rules. Second, there are problems where human experts exist, but where they are unable to explain their expertise. This is the case in many perceptual tasks, such as speech recognition, hand-writing recognition, and natural language understanding. Virtually all humans exhibit expert-level abilities on these tasks, but none of them can describe the detailed steps that they follow as they perform them. Fortunately, humans can provide machines with examples of the inputs and correct outputs for these tasks, so machine learning algorithms can learn to map the inputs to the outputs. Third, there are problems where phenomena are changing rapidly. In finance, for example, people would like to predict the future behavior of the stock market, of consumer purchases, or of exchange rates. These behaviors change frequently, so that even if a programmer could construct a good predictive computer program, it would need to be rewritten frequently. A learning program can relieve the programmer of this burden by constantly modifying and tuning a set of learned prediction rules. Fourth, there are applications that need to be customized for each computer user separately. Consider, for example, a program to filter unwanted electronic mail messages. Different users will need different filters. It is unreasonable to expect each user to program his or her own rules, and it is infeasible to provide every user with a software engineer to keep the rules up-to-date. A machine learning system can learn which mail messages the user rejects and maintain the filtering rules automatically. Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis. Statistics focuses on understanding the phenomena that have generated the data, often with the goal of testing different hypotheses about those phenomena. Data mining seeks to find patterns in the data that are understandable by people. Psychological studies of human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people (concept learning, skill acquisition, strategy change, etc.).
13,246 citations
TL;DR: The data allow us to reject alternative accounts of the function of the fusiform face area (area “FF”) that appeal to visual attention, subordinate-level classification, or general processing of any animate or human forms, demonstrating that this region is selectively involved in the perception of faces.
Abstract: Using functional magnetic resonance imaging (fMRI), we found an area in the fusiform gyrus in 12 of the 15 subjects tested that was significantly more active when the subjects viewed faces than when they viewed assorted common objects. This face activation was used to define a specific region of interest individually for each subject, within which several new tests of face specificity were run. In each of five subjects tested, the predefined candidate “face area” also responded significantly more strongly to passive viewing of (1) intact than scrambled two-tone faces, (2) full front-view face photos than front-view photos of houses, and (in a different set of five subjects) (3) three-quarter-view face photos (with hair concealed) than photos of human hands; it also responded more strongly during (4) a consecutive matching task performed on three-quarter-view faces versus hands. Our technique of running multiple tests applied to the same region defined functionally within individual subjects provides a solution to two common problems in functional imaging: (1) the requirement to correct for multiple statistical comparisons and (2) the inevitable ambiguity in the interpretation of any study in which only two or three conditions are compared. Our data allow us to reject alternative accounts of the function of the fusiform face area (area “FF”) that appeal to visual attention, subordinate-level classification, or general processing of any animate or human forms, demonstrating that this region is selectively involved in the perception of faces.
7,059 citations
"Large-scale automated synthesis of ..." refers methods in this paper
...Second, we used NeuroSynth to replicate previous findings of visual category-specific activation in regions such as the fusiform face are...
[...]
TL;DR: Functional anatomical work has detailed an afferent neural system in primates and in humans that represents all aspects of the physiological condition of the physical body that might provide a foundation for subjective feelings, emotion and self-awareness.
Abstract: As humans, we perceive feelings from our bodies that relate our state of well-being, our energy and stress levels, our mood and disposition. How do we have these feelings? What neural processes do they represent? Recent functional anatomical work has detailed an afferent neural system in primates and in humans that represents all aspects of the physiological condition of the physical body. This system constitutes a representation of 'the material me', and might provide a foundation for subjective feelings, emotion and self-awareness.
4,673 citations
TL;DR: This paper shows that the accuracy of learned text classifiers can be improved by augmenting a small number of labeled training documents with a large pool of unlabeled documents, and presents two extensions to the algorithm that improve classification accuracy under these conditions.
Abstract: This paper shows that the accuracy of learned text classifiers can be improved by augmenting a small number of labeled training documents with a large pool of unlabeled documents. This is important because in many text classification problems obtaining training labels is expensive, while large quantities of unlabeled documents are readily available.
We introduce an algorithm for learning from labeled and unlabeled documents based on the combination of Expectation-Maximization (EM) and a naive Bayes classifier. The algorithm first trains a classifier using the available labeled documents, and probabilistically labels the unlabeled documents. It then trains a new classifier using the labels for all the documents, and iterates to convergence. This basic EM procedure works well when the data conform to the generative assumptions of the model. However these assumptions are often violated in practice, and poor performance can result. We present two extensions to the algorithm that improve classification accuracy under these conditions: (1) a weighting factor to modulate the contribution of the unlabeled data, and (2) the use of multiple mixture components per class. Experimental results, obtained using text from three different real-world tasks, show that the use of unlabeled data reduces classification error by up to 30%.
3,123 citations
"Large-scale automated synthesis of ..." refers background in this paper
...This regularization prevents rare activations or terms from degrading accurac...
[...]
TL;DR: The authors conducted a quantitative meta-analysis of 668 sets of activation coordinates in Talairach space reported in 24 primary studies of n-back task variants manipulating process (location vs. identity monitoring) and content (verbal or nonverbal) of working memory.
Abstract: One of the most popular experimental paradigms for functional neuroimaging studies of working memory has been the n-back task, in which subjects are asked to monitor the identity or location of a series of verbal or nonverbal stimuli and to indicate when the currently presented stimulus is the same as the one presented n trials previously. We conducted a quantitative meta-analysis of 668 sets of activation coordinates in Talairach space reported in 24 primary studies of n-back task variants manipulating process (location vs. identity monitoring) and content (verbal or nonverbal) of working memory. We found the following cortical regions were activated robustly (voxelwise false discovery rate = 1%): lateral premotor cortex; dorsal cingulate and medial premotor cortex; dorsolateral and ventrolateral prefrontal cortex; frontal poles; and medial and lateral posterior parietal cortex. Subsidiary meta-analyses based on appropriate subsets of the primary data demonstrated broadly similar activation patterns for identity monitoring of verbal stimuli and both location and identity monitoring of nonverbal stimuli. There was also some evidence for distinct frontoparietal activation patterns in response to different task variants. The functional specializations of each of the major cortical components in the generic large-scale frontoparietal system are discussed. We conclude that quantitative meta-analysis can be a powerful tool for combining results of multiple primary studies reported in Talairach space. Here, it provides evidence both for broadly consistent activation of frontal and parietal cortical regions by various versions of the n-back working memory paradigm, and for process- and content-specific frontoparietal activation by working memory.
2,960 citations
"Large-scale automated synthesis of ..." refers result in this paper
...For working memory, the forward inference map revealed the most consistent associations in the dorsolateral prefrontal cortex, anterior insula and dorsal medial frontal cortex, replicating previous finding...
[...]