fMRI at 20: has it changed the world?

The prevalence of fMRI in cognitive neuroscience research is clear, but the overall impact of the associated research in the broader scope of our scientific community, and of society, is less obvious. The first reports of fMRI garnered huge interest in many areas, giving rise to a wide range of applications and technical developments over the past 20 years. Using five primary areas, i.e. scientific impact, clinical practice, cognitive neuroscience, mental illness, and society-this essay examines the question: Has fMRI changed the world?

Evolution and current challenges in the teaching of functional MRI and functional brain imaging.

The report of any new and successful method for studying the world triggers the need to train people in the use of that method. In the case of functional magnetic resonance imaging and its use for examining human brain function in vivo, expertise is required in a greater collection of domains than usual. Development of fMRI training programs started shortly after the announcement of BOLD-based fMRI in humans. These programs had a variety of durations and primary content areas. All programs had to deal with the challenge of bringing interested researchers from a wide variety of areas-many of whom had little or no understanding of MR physics, and/or experimental psychology, and/or the nuances of data analysis and modeling-to a sufficiently detailed level of knowledge that both the funding agencies, and the existing proprietors of the technology (often radiologists or MR physicists at hospitals) would take the research proposals of new investigators seriously. Now that fMRI-based research is well established, there are new educational challenges. Some have to do with the growing list of technologies used to study human brain function in vivo. But perhaps more daunting is the challenge of training consumers of the reports and claims based on fMRI and other brain imaging modalities. As fMRI becomes influential in contexts beyond the research environment-from the clinic to the courtroom to the legislature-training consumers of fMRI-based claims will take on increasing importance, and represents its own unique challenges for education.

Developmental neural networks in children performing a Categorical N-Back Task.

The prefrontal and temporal networks subserving object working memory tasks in adults have been reported as immature in young children; yet children are adequately capable of performing such tasks. We investigated the basis of this apparent contradiction using a complex object working memory task, a Categorical n-back (CN-BT). We examined whether the neural networks engaged by the CN-BT in children consist of the same brain regions as those in adults, but with a different magnitude of activation, or whether the networks are qualitatively different. Event-related fMRI was used to study differences in brain activation between healthy children ages 6 and 10 years, and young adults (20-28 years). Performance accuracy and RTs in 10-year-olds and adults were comparable, but the performance in 6-year-olds was lower. In adults, the CN-BT was highly effective in engaging the bilateral (L>R) ventral prefrontal cortex, the bilateral fusiform gyrus, posterior cingulate and precuneus, thus suggesting an involvement of the ventral visual stream, with related feature extraction and semantic labeling strategies. In children, the brain networks were distinctly different. They involved the premotor and parietal cortex, anterior insula, caudate/putamen, and the cerebellum, thus suggesting a predominant involvement of the visual dorsal and sensory-motor pathways, with related visual-spatial and action cognitive strategies. The findings indicate engagement of developmental networks in children reflecting task-effective brain activation. The age-related pattern of fMRI activation suggests a working hypothesis of a developmental shift from reliance on the dorsal visual stream and premotor/striatal/cerebellar networks in young children to reliance on the ventral prefrontal and inferior temporal networks in adults.

Using small numbers of subjects in fMRI-based research.

The preceding speculations may not sound very novel to some ears. Indeed, when I started to describe the above ideas to any fMRI researcher who has been involved in the field for a substantial number of years, their response is invariably a comment to the effect that, "Oh yes, we are doing something like that right now in our lab." What I think they mean, of course, is that sometime in the uncertain future they might run all their own existing tasks on a few subjects, if they can get around to it and if they can find the software for some of those earlier tasks. The heart (and hard parts) of the present speculation would be creating a repository of (publicly available) interesting tasks and creating software to embody neuropsychological models and the associated imaging data. At the risk of completely losing my audience, it is the image presented in Isaac Asimov's novel Second Foundation for integrating theory and information across many different sources that is the true challenge.

Experimental design in brain activation MRI: cautionary tales.

The use of functional magnetic resonance imaging (fMRI) in cognitive neuroscience has expanded at an amazing rate in the past 10 years. Current research includes increasingly subtle and specific attempts to dissect the cognitive and emotional mechanisms called into play when humans make decisions. The present essay will briefly review some of the general considerations and domains of information needed when one designs fMRI-based experiments. However, the main theme will be the difficulties associated with designing, conducting, analyzing and interpreting such research. Functional MRI is an unusually complicated technique, and there are numerous ways for experiments to go wrong. As well as demanding exceptional care in maintaining the quality of one's own research, this makes the universal problem of evaluating other peoples' research particularly challenging.

Functional magnetic resonance imaging (fMRI) "brain reading": detecting and classifying distributed patterns of fMRI activity in human visual cortex.

Traditional (univariate) analysis of functional MRI (fMRI) data relies exclusively on the information contained in the time course of individual voxels. Multivariate analyses can take advantage of the information contained in activity patterns across space, from multiple voxels. Such analyses have the potential to greatly expand the amount of information extracted from fMRI data sets. In the present study, multivariate statistical pattern recognition methods, including linear discriminant analysis and support vector machines, were used to classify patterns of fMRI activation evoked by the visual presentation of various categories of objects. Classifiers were trained using data from voxels in predefined regions of interest during a subset of trials for each subject individually. Classification of subsequently collected fMRI data was attempted according to the similarity of activation patterns to prior training examples. Classification was done using only small amounts of data (20 s worth) at a time, so such a technique could, in principle, be used to extract information about a subject's percept on a near real-time basis. Classifiers trained on data acquired during one session were equally accurate in classifying data collected within the same session and across sessions separated by more than a week, in the same subject. Although the highest classification accuracies were obtained using patterns of activity including lower visual areas as input, classification accuracies well above chance were achieved using regions of interest restricted to higher-order object-selective visual areas. In contrast to typical fMRI data analysis, in which hours of data across many subjects are averaged to reveal slight differences in activation, the use of pattern recognition methods allows a subtle 10-way discrimination to be performed on an essentially trial-by-trial basis within individuals, demonstrating that fMRI data contain far more information than is typically appreciated.