Classical and Bayesian inference in neuroimaging: theory

A Bayesian hierarchical framework for spatial modeling of fMRI data

Applications of functional magnetic resonance imaging (fMRI) have provided novel insights into the neuropathophysiology of major psychiatric, neurological, and substance abuse disorders and their treatments. Modern activation studies often compare localized task-induced changes in brain activity between experimental groups. Complementary approaches consider the ensemble of voxels constituting an anatomically defined region of interest (ROI) or summary statistics, such as means or quantiles, of the ROI. In this work, we present a Bayesian extension of voxel-level analyses that offers several notable benefits. Among these, it combines whole-brain voxel-by-voxel modeling and ROI analyses within a unified framework. Secondly, an unstructured variance/covariance matrix for regional mean parameters allows for the study of inter-regional (long-range) correlations, and the model employs an exchangeable correlation structure to capture intra-regional (short-range) correlations. Estimation is performed using Markov Chain Monte Carlo (MCMC) techniques implemented via Gibbs sampling. We apply our Bayesian hierarchical model to two novel fMRI data sets: one considering inhibitory control in cocaine-dependent men and the second considering verbal memory in subjects at high risk for Alzheimer's disease.

Bayesian decoding of brain images

This paper introduces a multivariate Bayesian (MVB) scheme to decode or recognise brain states from neuroimages. It resolves the ill-posed many-to-one mapping, from voxel values or data features to a target variable, using a parametric empirical or hierarchical Bayesian model. This model is inverted using standard variational techniques, in this case expectation maximisation, to furnish the model evidence and the conditional density of the model's parameters. This allows one to compare different models or hypotheses about the mapping from functional or structural anatomy to perceptual and behavioural consequences (or their deficits). We frame this approach in terms of decoding measured brain states to predict or classify outcomes using the rhetoric established in pattern classification of neuroimaging data. However, the aim of MVB is not to predict (because the outcomes are known) but to enable inference on different models of structure-function mappings; such as distributed and sparse representations. This allows one to optimise the model itself and produce predictions that outperform standard pattern classification approaches, like support vector machines. Technically, the model inversion and inference uses the same empirical Bayesian procedures developed for ill-posed inverse problems (e.g., source reconstruction in EEG). However, the MVB scheme used here extends this approach to include a greedy search for sparse solutions. It reduces the problem to the same form used in Gaussian process modelling, which affords a generic and efficient scheme for model optimisation and evaluating model evidence. We illustrate MVB using simulated and real data, with a special focus on model comparison; where models can differ in the form of the mapping (i.e., neuronal representation) within one region, or in the (combination of) regions per se.

Neural correlates of emotional working memory in patients with mild cognitive impairment

Emotional stimuli can have beneficial effects on memory in healthy aged subjects and partly on patients with dementia. So far, no experimental study has explored the effects of memory for emotional stimuli in patients with mild cognitive impairment (MCI), a concept that describes a transitional state between normal aging and dementia. The present fMRI study explored working memory for emotional stimuli in 16 patients with amnestic MCI (aMCI) and 16 healthy aged participants. Subjects performed an n-back task (2-back) with neutral, positive, and negative emotional pictures. The analysis focused on target processing. Results showed that groups did not differ in working memory performance. In healthy aged participants emotional targets had no significant impact on working memory. In patients with aMCI a negativity bias was observed, indicating that negative targets were better remembered compared to neutral and positive targets. Regarding fMRI results, both groups showed an increase in functional activity in prefrontal and lateral parietal brain regions associated with target processing. As a key result, we observed significant group by emotion interaction effects in the precuneus. Healthy aged participants showed a signal decrease in the left precuneus for positive compared to neutral targets. The precuneus deactivation in healthy aged participants may indicate a disengagement of self-referential processes towards task-related processes. Patients with aMCI revealed a signal increase in the right precuneus for negative compared to neutral targets. This increase in precuneus activity, combined with a behavioural facilitation effect, may indicate a mechanism to compensate disease related processes in aMCI.

Power calculation for group fMRI studies accounting for arbitrary design and temporal autocorrelation

When planning most scientific studies, one of the first steps is to carry out a power analysis to define a design and sample size that will result in a well-powered study. There are limited resources for calculating power for group fMRI studies due to the complexity of the model. Previous approaches for group fMRI power calculation simplify the study design and/or the variance structure in order to make the calculation possible. These approaches limit the designs that can be studied and may result in inaccurate power calculations. We introduce a flexible power calculation model that makes fewer simplifying assumptions, leading to a more accurate power analysis that can be used on a wide variety of study designs. Our power calculation model can be used to obtain region of interest (ROI) summaries of the mean parameters and variance parameters, which can be use to increase understanding of the data as well as calculate power for a future study. Our example illustrates that minimizing cost to achieve 80% power is not as simple as finding the smallest sample size capable of achieving 80% power, since smaller sample sizes require each subject to be scanned longer.

Diffusion-based spatial priors for imaging

We describe a Bayesian scheme to analyze images, which uses spatial priors encoded by a diffusion kernel, based on a weighted graph Laplacian. This provides a general framework to formulate a spatial model, whose parameters can be optimized. The application we have in mind is a spatiotemporal model for imaging data. We illustrate the method on a random effects analysis of fMRI contrast images from multiple subjects; this simplifies exposition of the model and enables a clear description of its salient features. Typically, imaging data are smoothed using a fixed Gaussian kernel as a pre-processing step before applying a mass-univariate statistical model (e.g., a general linear model) to provide images of parameter estimates. An alternative is to include smoothness in a multivariate statistical model (Penny, W.D., Trujillo-Barreto, N.J., Friston, K.J., 2005. Bayesian fMRI time series analysis with spatial priors. Neuroimage 24, 350–362). The advantage of the latter is that each parameter field is smoothed automatically, according to a measure of uncertainty, given the data. In this work, we investigate the use of diffusion kernels to encode spatial correlations among parameter estimates. Nonlinear diffusion has a long history in image processing; in particular, flows that depend on local image geometry (Romeny, B.M.T., 1994. Geometry-driven Diffusion in Computer Vision. Kluwer Academic Publishers) can be used as adaptive filters. This can furnish a non-stationary smoothing process that preserves features, which would otherwise be lost with a fixed Gaussian kernel. We describe a Bayesian framework that incorporates non-stationary, adaptive smoothing into a generative model to extract spatial features in parameter estimates. Critically, this means adaptive smoothing becomes an integral part of estimation and inference. We illustrate the method using synthetic and real fMRI data.

Model-free analysis of brain fMRI data by recurrence quantification

We propose a novel model-free univariate strategy for functional magnetic resonance imaging (fMRI) studies based upon recurrence quantification analysis (RQA). RQA is an auto-regressive method, which identifies recurrences in signals without any a priori assumptions. The performance of RQA is compared to that of univariate statistics based on a general linear model (GLM) and probabilistic independent component analysis (P-ICA) for a set of simulated and real fMRI data. RQA provides an appealing alternative to conventional GLM techniques, due to its exclusive feature of being model-free and of detecting potentially both linear and nonlinear dynamic processes, without requiring signal stationarity. The overall performance of the method compares positively also with P-ICA, another well-known model-free algorithm, which requires prior information to discriminate between different spatio-temporal processes. For simulated data, RQA is endowed with excellent accuracy for contrast-to-noise ratios greater than 0.2, and has a performance comparable to that of GLM for t(CNR)>or=0.8. For cerebral fMRI data acquired from a group of healthy subjects performing a finger-tapping task, (i) RQA reveals activations in the primary motor area contra-lateral to the employed hand and in the supplementary motor area, in agreement with the outcome of GLM analysis and (ii) identifies an additional brain region with transient signal changes. Moreover, RQA identifies signal recurrences induced by physiological processes other than BOLD (movement-related or of vascular origin). Finally, RQA is more robust than the GLM with respect to variations in the shape and timing of the underlying neuronal and hemodynamic responses which may vary between brain regions, subjects and tasks.

Distinct causal influences of parietal versus frontal areas on human visual cortex: evidence from concurrent TMS-fMRI

It has often been proposed that regions of the human parietal and/or frontal lobe may modulate activity in visual cortex, for example, during selective attention or saccade preparation. However, direct evidence for such causal claims is largely missing in human studies, and it remains unclear to what degree the putative roles of parietal and frontal regions in modulating visual cortex may differ. Here we used transcranial magnetic stimulation (TMS) and functional magnetic resonance imaging (fMRI) concurrently, to show that stimulating right human intraparietal sulcus (IPS, at a site previously implicated in attention) elicits a pattern of activity changes in visual cortex that strongly depends on current visual context. Increased intensity of IPS TMS affected the blood oxygen level-dependent (BOLD) signal in V5/MT+ only when moving stimuli were present to drive this visual region, whereas TMS-elicited BOLD signal changes were observed in areas V1-V4 only during the absence of visual input. These influences of IPS TMS upon remote visual cortex differed significantly from corresponding effects of frontal (eye field) TMS, in terms of how they related to current visual input and their spatial topography for retinotopic areas V1-V4. Our results show directly that parietal and frontal regions can indeed have distinct patterns of causal influence upon functional activity in human visual cortex.

The Different Neural Correlates of Action and Functional Knowledge in Semantic Memory: An fMRI Study

Previous reports suggest that the internal organization of semantic memory is in terms of different "types of knowledge," including "sensory" (information about perceptual features), "action" (motor-based knowledge of object utilization), and "functional" (abstract properties, as function and context of use). Consistent with this view, a specific loss of action knowledge, with preserved functional knowledge, has been recently observed in patients with left frontoparietal lesions. The opposite pattern (impaired functional knowledge with preserved action knowledge) was reported in association with anterior inferotemporal lesions. In the present study, the cerebral representation of action and functional knowledge was investigated using event-related analysis of functional magnetic resonance imaging data. Fifteen subjects were presented with pictures showing pairs of manipulable objects and asked whether the objects within each pair were used with the same manipulation pattern ("action knowledge" condition) or in the same context ("functional knowledge" condition). Direct comparisons showed action knowledge, relative to functional knowledge, to activate a left frontoparietal network, comprising the intraparietal sulcus, the inferior parietal lobule, and the dorsal premotor cortex. The reverse comparison yielded activations in the retrosplenial and the lateral anterior inferotemporal cortex. These results confirm and extend previous neuropsychological data and support the hypothesis of the existence of different types of information processing in the internal organization of semantic memory.

Comparison of fMRI statistical software packages and strategies for analysis of images containing random and stimulus-correlated motion

The objectives of this study were to use computer-generated phantoms containing real subject motion to: (1) compare the sensitivity of four commonly used fMRI software packages and (2) compare the sensitivity of three statistical analysis strategies with respect to motion correction. The results suggest that all four packages perform similarly in fMRI statistical analysis with SPM2 having slightly higher sensitivity. The most sensitive analysis technique was to perform motion correction and include the realignment parameters as regressors in the general linear model. This approach applies to all four packages examined and can be most beneficial when stimulus-correlated motion is present.

Local linear discriminant analysis (LLDA) for group and region of interest (ROI)-based fMRI analysis

A post-processing method for group discriminant analysis of fMRI is proposed. It assumes that the fMRI data have been pre-processed and analyzed so that each voxel is given a statistic specifying task-related activation(s), and that individually specific regions of interest (ROIs) have been drawn for each subject. The method then utilizes Local Linear Discriminant Analysis (LLDA) to jointly optimize the individually-specific and group linear combinations of ROIs that maximally discriminates between groups (or between tasks, if using the same subjects). LLDA tries to linearly transform each subject's voxel-based activation statistics within ROIs to a common vector space of ROI combinations, enabling the relative similarity of different subjects' activation to be assessed. We applied the method to data recorded from 10 normal subjects during a motor task expected to activate both cortical and subcortical structures. The proposed method detected activation in multiple cortical and subcortical structures that were not present when the data were analyzed by warping the data to a common space. We suggest that the method be applied to group fMRI data when warping to a common space may be ill-advised, such as examining activation in small subcortical structures susceptible to mis-registration, or examining older or neurological patient populations.

Parametric spherical deconvolution: inferring anatomical connectivity using diffusion MR imaging

The human brain forms a complex neural network with a connectional architecture that is still far from being known in full detail, even at the macroscopic level. The advent of diffusion MR imaging has enabled the exploration of the structural properties of white matter in vivo. In this article we propose a new forward model that maps the microscopic geometry of nervous tissue onto the water diffusion process and further onto the measured MR signals. Our spherical deconvolution approach completely parameterizes the fiber orientation density by a finite mixture of Bingham distributions. In addition, we define the term anatomical connectivity, taking the underlying image modality into account. This neurophysiological metric may represent the proportion of the nerve fibers originating in the source area which intersect a given target region. The specified inverse problem is solved by Bayesian statistics. Posterior probability maps denote the probability that the connectivity value exceeds a chosen threshold, conditional upon the noisy observations. These maps allow us to draw inferences about the structural organization of the cerebral cortex. Moreover, we will demonstrate the proposed approach with diffusion-weighted data sets featuring high angular resolution.

Cortical Plasticity in Amyotrophic Lateral Sclerosis: Motor Imagery and Function

Background. Cortical networks underlying motor imagery are functionally close to motor performance networks and can be activated by patients with severe motor disabilities. Objective. The aim of the study was to examine the longitudinal effect of progressive motoneuron degeneration on cortical representation of motor imagery and function in amyotrophic lateral sclerosis. Methods. The authors studied 14 amyotrophic lateral sclerosis patients and 15 healthy controls and a subgroup of 11 patients and 14 controls after 6 months with a grip force paradigm comprising imagery and execution tasks using functional magnetic resonance imaging. Results. Motor imagery activated similar neural networks as motor execution in amyotrophic lateral sclerosis patients and healthy subjects in the primary motor (BA 4), premotor, and supplementary motor (BA 6) cortex. Amyotrophic lateral sclerosis patients presented a stronger response within premotor and primary motor areas for imagery and execution compared to controls. After 6 months, these differences persisted with additional activity in the precentral gyrus in patients as well as in a frontoparietal network for motor imagery, in which activity increased with impairment. Conclusion. The findings suggest an ongoing compensatory process within the higher order motor-processing system of amyotrophic lateral sclerosis patients, probably to overcome loss of function in primary motor and motor imagery-specific networks. The increased activity in precentral and frontoparietal networks in motor imagery might be used to control brain-computer interfaces to drive communication and limb prosthetic devices in patients with loss of motor control such as severely disabled amyotrophic lateral sclerosis patients in a locked-in-like state.

Postprandial glucagon-like peptide-1 (GLP-1) response is positively associated with changes in neuronal activity of brain areas implicated in satiety and food intake regulation in humans

Postprandial glucagon-like peptide-1 (GLP-1) secretion can act as a meal termination signal in animals and humans. We tested the hypothesis that the postprandial changes in plasma GLP-1 concentrations are associated with changes in the human brain activity in response to satiety by performing a post-hoc analysis of a cross-sectional study of neuroanatomical correlates of hunger and satiation using (15)O-water positron-emission tomography (PET). Forty-two subjects (22M/20F, age 31+/-8 years) spanning a wide range of adiposity (body fat: 7-44%) were included in this analysis. Outcome measures included changes in PET-measured regional cerebral blood flow (rCBF) and plasma concentrations of GLP-1, glucose, insulin, and free-fatty acids (FFA), elicited by the administration of a satiating amount of a liquid formula meal. The peak postprandial increases in plasma GLP-1 concentrations were correlated with increases in rCBF in the left dorsolateral prefrontal cortex (including the left middle and inferior frontal gyri), previously implicated in PET studies of human satiation, and the hypothalamus, previously implicated in the regulation of food intake in animal and human studies, both before and after adjustment for sex, age, body fat, and changes in plasma glucose, insulin, and serum FFA concentrations. The postprandial GLP-1 response is associated with activation of areas of the human brain previously implicated in satiation and food intake regulation.

Analysis of a large fMRI cohort: Statistical and methodological issues for group analyses

The aim of group fMRI studies is to relate contrasts of tasks or stimuli to regional brain activity increases. These studies typically involve 10 to 16 subjects. The average regional activity statistical significance is assessed using the subject to subject variability of the effect (random effects analyses). Because of the relatively small number of subjects included, the sensitivity and reliability of these analyses is questionable and hard to investigate. In this work, we use a very large number of subject (more than 80) to investigate this issue. We take advantage of this large cohort to study the statistical properties of the inter-subject activity and focus on the notion of reproducibility by bootstrapping. We asked simple but important methodological questions: Is there, from the point of view of reliability, an optimal statistical threshold for activity maps? How many subjects should be included in group studies? What method should be preferred for inference? Our results suggest that i) optimal thresholds can indeed be found, and are rather lower than usual corrected for multiple comparison thresholds, ii) 20 subjects or more should be included in functional neuroimaging studies in order to have sufficient reliability, iii) non-parametric significance assessment should be preferred to parametric methods, iv) cluster-level thresholding is more reliable than voxel-based thresholding, and v) mixed effects tests are much more reliable than random effects tests. Moreover, our study shows that inter-subject variability plays a prominent role in the relatively low sensitivity and reliability of group studies.

Episodic encoding is more than the sum of its parts: an fMRI investigation of multifeatural contextual encoding

Episodic memories are characterized by their contextual richness, yet little is known about how the various features comprising an episode are brought together in memory. Here we employed fMRI and a multidimensional source memory procedure to investigate processes supporting the mnemonic binding of item and contextual information. Volunteers were scanned while encoding items for which the contextual features (color and location) varied independently, allowing activity elicited at the time of study to be segregated according to whether both, one, or neither feature was successfully retrieved on a later memory test. Activity uniquely associated with successful encoding of both features was identified in the intra-parietal sulcus, a region strongly implicated in the support of attentionally mediated perceptual binding. The findings suggest that the encoding of disparate features of an episode into a common memory representation requires that the features be conjoined in a common perceptual representation when the episode is initially experienced.

Progressive changes of pre- and post-synaptic dopaminergic biomarkers in conscious MPTP-treated cynomolgus monkeys measured by positronemission tomography

Positron emission tomography (PET) is a useful technique for the consecutive investigation of the relationship between changes in neurotransmission biomarkers and behavioral signs in animal models of Parkinson's disease (PD). In this study, we aimed to investigate the threshold of dopamine (DA) neuron damage for the appearance of tremor by observing the longitudinal changes of pre- and post-synaptic DA biomarkers in awake monkeys using PET with multiple tracers. Three cynomolgus monkeys were treated with MPTP every 3-6 weeks until tremor was observed. Brain uptake of [11C]PE2I, [beta-11C]DOPA, and [11C]raclopride for DA transporter (DAT), DOPA utilization, and DA D2 receptor were measured using PET as a single set in awake condition. Sets of PET scans were repeated in parallel with continuous behavioral estimation. The pre-synaptic biomarkers of DA neuron in the striatum decreased [11C]PE2I binding and [beta-11C]DOPA uptake in an MPTP dose-dependent manner. Tremor was not observed until striatal [11C]PE2I binding was reduced to about 15% of the pretreatment level and [beta-11C]DOPA uptake was reduced to about 34%. DA D2 receptor measured by [11C]raclopride was not significantly changed throughout the experiment. Our results revealed that it is possible to quantitatively define the threshold of the onset of behavioral PD signs by monitoring spontaneous motor activity, and in vivo PET with DAT marker can be a biomarker for early diagnosis at the presymptomatic stage of PD and for high-risk groups.

Improved methods for fMRI studies of combined taste and aroma stimuli

Previous neuroimaging studies of the cortical representation of gustatory and olfactory stimuli have often delivered tastants to the mouth in very small quantities or stimulated olfaction orthonasally. In studies of retro-nasal olfaction, swallowing was generally delayed to reduce head motion artefacts. The present fMRI study aims to improve upon such methodological limitations to allow investigation of the cortical representation of flavour (taste and aroma combination) as it typically occurs during the consumption of liquid foods. For this purpose we used (1) a novel, automated, sprayed stimulus delivery system and a larger volume of liquid sample (containing sweet tastants and banana/pear aroma volatiles) to achieve more extensive stimulation of the oral cavity taste receptors, (2) a pseudo-natural delivery paradigm that included prompt swallowing after each sample delivery to obtain physiological retro-nasal olfactory stimulation, (3) fMRI acquisition with wide brain coverage and double-echo EPI to improve sensitivity. We validated our paradigm for the delivery of volatiles using atmospheric pressure chemical ionisation mass spectrometry. This showed that the main retro-nasal delivery of volatiles in the paradigm occurs immediately after the swallow. Several brain areas were found to be activated, including the insula, frontal operculum, rolandic operculum/parietal lobe, piriform, dorsolateral prefrontal cortex, anterior cingulate cortex, ventro-medial thalamus, hippocampus and medial orbitofrontal cortex.

Implicit learning deficits in dyslexic adults: An fMRI study

It is assumed that several neuropsychological impairments characterize the cognitive profile of individuals with developmental dyslexia (DD). Phonological and visual processing are often impaired as well as auditory processing, attention, and information processing speed. Although reports in the literature on implicit learning abilities are contradictory, recent neurological and physiological data suggest that these abilities are deficient in individuals with DD. To evaluate implicit learning we administered a classical version of the serial reaction time task (SRTT) related to sequence learning. Using functional magnetic resonance imaging we investigated brain activation patterns associated with implicit learning deficits in 14 adults with DD matched with 14 normal readers. SRTT results indicated the absence of implicit learning in the DD group and different activations between groups mainly in SMA, inferior parietal areas and cerebellar lobule 6. These results can be interpreted in the light of the different capacities for the two groups to build an internal model to guide movements. Further, they explain DD individuals' difficulty in domains not directly related to reading ability.

Remapping in human visual cortex

With each eye movement, stationary objects in the world change position on the retina, yet we perceive the world as stable. Spatial updating, or remapping, is one neural mechanism by which the brain compensates for shifts in the retinal image caused by voluntary eye movements. Remapping of a visual representation is believed to arise from a widespread neural circuit including parietal and frontal cortex. The current experiment tests the hypothesis that extrastriate visual areas in human cortex have access to remapped spatial information. We tested this hypothesis using functional magnetic resonance imaging (fMRI). We first identified the borders of several occipital lobe visual areas using standard retinotopic techniques. We then tested subjects while they performed a single-step saccade task analogous to the task used in neurophysiological studies in monkeys, and two conditions that control for visual and motor effects. We analyzed the fMRI time series data with a nonlinear, fully Bayesian hierarchical statistical model. We identified remapping as activity in the single-step task that could not be attributed to purely visual or oculomotor effects. The strength of remapping was roughly monotonic with position in the visual hierarchy: remapped responses were largest in areas V3A and hV4 and smallest in V1 and V2. These results demonstrate that updated visual representations are present in cortical areas that are directly linked to visual perception.

Intermediate phenotypes and genetic mechanisms of psychiatric disorders

Genes are major contributors to many psychiatric diseases, but their mechanisms of action have long seemed elusive. The intermediate phenotype concept represents a strategy for characterizing the neural systems affected by risk gene variants to elucidate quantitative, mechanistic aspects of brain function implicated in psychiatric disease. Using imaging genetics as an example, we illustrate recent advances, challenges and implications of linking genes to structural and functional variation in brain systems related to cognition and emotion.

On-Line Attentional Selection From Competing Stimuli in Opposite Visual Fields: Effects on Human Visual Cortex and Control Processes

We used fMRI to investigate competition and on-line attentional selection between targets and distractors in opposite visual hemifields. Displays comprised a high-contrast square-wave grating, defined as target by its orientation, presented alone (unilateral) or with a similar distractor of orthogonal orientation in the opposite hemifield (bilateral displays). The target appeared unpredictably on the left or right, precluding anticipatory attention to one side. We found greater activation in target-contralateral superior occipital gyrus for unilateral than for bilateral displays, indicating suppression of the target’s visual representation by distractor presence despite the competing distractor projecting to a different occipital hemisphere. Several frontal and parietal regions showed greater activation for bilateral than unilateral trials, suggesting involvement in on-line attentional selection. This was particularly pronounced for regions in bilateral intraparietal sulcus (IPS), which also showed greater functional coupling with occipital cortex specifically on bilateral trials that required selection plus some repetition-suppression effects when target side was repeated, but again only on bilateral trials requiring selection. Our results indicate that competition between visual stimuli in opposite hemifields can influence occipital cortex, and implicate IPS in resolution of this competition by selection.

Variational free energy and the Laplace approximation

This note derives the variational free energy under the Laplace approximation, with a focus on accounting for additional model complexity induced by increasing the number of model parameters. This is relevant when using the free energy as an approximation to the log-evidence in Bayesian model averaging and selection. By setting restricted maximum likelihood (ReML) in the larger context of variational learning and expectation maximisation (EM), we show how the ReML objective function can be adjusted to provide an approximation to the log-evidence for a particular model. This means ReML can be used for model selection, specifically to select or compare models with different covariance components. This is useful in the context of hierarchical models because it enables a principled selection of priors that, under simple hyperpriors, can be used for automatic model selection and relevance determination (ARD). Deriving the ReML objective function, from basic variational principles, discloses the simple relationships among Variational Bayes, EM and ReML. Furthermore, we show that EM is formally identical to a full variational treatment when the precisions are linear in the hyperparameters. Finally, we also consider, briefly, dynamic models and how these inform the regularisation of free energy ascent schemes, like EM and ReML.

Effects of hormone therapy on brain morphology of healthy postmenopausal women: a Voxel-based morphometry study

OBJECTIVE

Estrogens are known to be protective in age-associated cognitive changes in humans and in neurodegeneration in animal models. The aim of this study was to evaluate the potential effects of estrogen therapy (ET) on human gray matter volume in vivo.

DESIGN

Forty healthy postmenopausal women underwent three-dimensional high-resolution magnetic resonance imaging: 17 were never treated, 16 were currently receiving ET, and 7 had had ET in the past. Voxel-based morphometry (VBM) with SPM2 was used, according to an optimized protocol, to compare women under past and current ET to those never treated. Significance threshold was set at P = 0.01, corrected by false discovery rate.

RESULTS

Voxel-based morphometry indicated that estrogen use was associated with greater gray matter volumes in the whole group of treated women, which included the cerebellum (cluster size, Z coordinates: 5,527; 5.15; -14 -54 -10), the amygdaloid-hippocampal complex (left: 19; 3.55; -22 -4 -18; right: 45; 3.61; 16 -6 -16), and extended to the frontal, temporal, parietal, and occipital neocortex. The comparison current ET versus past ET use showed that women who underwent treatment in the past had greater volumes of gray matter compared to women under current treatment.

CONCLUSIONS

ET might slow down age-related gray matter loss in postmenopausal women. The structures that exhibited greater volume in association with ET included the cerebellar and cerebral cortices and, typically involved in Alzheimer's disease, the medial temporal structures and the temporoparietal junction.

Bayesian estimation of evoked and induced responses

We describe an extension of our empirical Bayes approach to magnetoencephalography/electroencephalography (MEG/EEG) source reconstruction that covers both evoked and induced responses. The estimation scheme is based on classical covariance component estimation using restricted maximum likelihood (ReML). We have focused previously on the estimation of spatial covariance components under simple assumptions about the temporal correlations. Here we extend the scheme, using temporal basis functions to place constraints on the temporal form of the responses. We show how the same scheme can estimate evoked responses that are phase-locked to the stimulus and induced responses that are not. For a single trial the model is exactly the same. In the context of multiple trials, however, the inherent distinction between evoked and induced responses calls for different treatments of the underlying hierarchical multitrial model. We derive the respective models and show how they can be estimated efficiently using ReML. This enables the Bayesian estimation of evoked and induced changes in power or, more generally, the energy of wavelet coefficients.

The Somatotopic Organization of Cytoarchitectonic Areas on the Human Parietal Operculum

The secondary somatosensory cortex (SII) of nonhuman primates is located on the parietal operculum. In the monkey, electrophysiological and connectivity tracing studies as well as histological investigations provide converging evidence for 3 distinct cortical areas (SII, PV, and VS) within this region, each of which contains a complete somatotopic map. Although the equivalency of the parietal operculum as the location of SII between humans and nonhuman primates is undisputed, the internal organization of the human SII region is still largely unknown. Based on their topography, we have previously argued that the cytoarchitectonic areas OP 1, OP 4, and OP 3 may constitute the human homologues of areas SII, PV, and VS, respectively. To test this hypothesis, we here examined (using functional magnetic resonance imaging) the somatotopic organization of the human parietal operculum by applying tactile stimulation to the skin at 4 different locations on either side of the body (face, hands, trunk, and legs). The locations of the resulting activation foci were then compared with the cytoarchitectonic maps of this region. Data analysis revealed 2 somatotopic body representations on the lateral operculum in areas OP 1 and OP 4. The functional border between these 2 body maps was defined by a mirror reversal in the somatotopic arrangement and coincided with the cytoarchitectonically defined border between these 2 areas. This somatotopic arrangement closely matches that described for SII and PV in nonhuman primates. The data also suggested a third somatotopic map located deeper inside the Sylvian fissure in area OP 3. Based on the observed topographic arrangement and their functional response characteristics, we conclude that cytoarchitectonic areas OP1, OP 4, and OP 3 on the human parietal operculum constitute the human homologues of primate areas SII, PV, and VS, respectively.

Less activation of the left dorsolateral prefrontal cortex in response to a meal: a feature of obesity

BACKGROUND

In an exploratory positron emission tomography study of postprandial regional cerebral blood flow, which is a marker of neuronal activity, obese men differed from lean men in several brain regions, including the prefrontal cortex. The subjects received a meal proportional to their body size; therefore, the meal volume was different for each person.

OBJECTIVE

We investigated whether differences in the brain responses of obese and lean men to a meal represent satiety or feelings of gastric distension.

DESIGN

We studied 9 lean (x +/- SD body fat: 15 +/- 5%; age: 33 +/- 10 y) and 9 obese (body fat: 31 +/- 4%; age: 32 +/- 10 y) men given a fixed amount (400 mL) of a liquid meal. We compared their results with those in 11 lean (body fat: 16 +/- 5%; age: 35 +/- 8 y) and 11 obese (body fat: 33 +/- 5%; age: 28 +/- 5 y) previously studied men given a meal proportional to their body size. We performed analyses by using a two-level, random-effects approach in the STATISTICAL PARAMETRIC MAPPING software package and a significance level of P < or = 0.001, uncorrected for multiple comparisons.

RESULTS

Compared with lean men, obese men had consistently less postprandial activation in the left dorsolateral prefrontal cortex, irrespective of meal size.

CONCLUSION

Because the dorsolateral prefrontal cortex has been implicated in the inhibition of inappropriate behavior, satiety, and meal termination, differential responses of neuronal activity to food intake in this area may contribute to a propensity for obesity or to the difficulty in losing weight experienced by obese men.

Brain Mechanisms Mediating Auditory Attentional Capture in Humans

The ability to detect and preferentially process salient auditory stimuli, even when irrelevant to a current task, is often critical for adaptive behavior. This stimulus-driven allocation of processing resources is known as "attentional capture." Here we used functional magnetic resonance imaging in humans to investigate brain activity and behavioral effects related to such auditory attentional capture. Participants searched a sequence of tones for a target tone that was shorter or longer than the nontarget tones. An irrelevant singleton feature in the tone sequence resulted in behavioral interference (attentional capture) and activation of parietal and prefrontal cortices only when the singleton was associated with a nontarget tone (nontarget singleton) and not when associated with a target tone (target singleton). In contrast, the presence (vs. absence) of a singleton feature in the sequence was associated with activation of frontal and temporal loci previously associated with auditory change detection. These results suggest that a ventral network involving superior temporal and inferior frontal cortices responds to acoustic variability, regardless of attentional significance, but a dorsal frontoparietal network responds only when a feature singleton captures attention.

Attentional preparation for a lateralized visual distractor: behavioral and fMRI evidence

Attending to the location of an expected visual target can lead to anticipatory activations in spatiotopic occipital cortex, emerging before target onset. But less is known about how the brain may prepare for a distractor at a known location remote from the target. In a psychophysical experiment, we found that trial-to-trial advance knowledge about the presence of a distractor in the target-opposite hemifield significantly reduced its behavioral cost. In a subsequent functional magnetic resonance imaging experiment with similar task and stimuli, we found anticipatory activations in the occipital cortex contralateral to the expected distractor, but no additional target modulation, when participants were given advance information about a distractor's subsequent presence and location. Several attention-related control structures (frontal eye fields and superior parietal cortex) were active during attentional preparation for all trials, whereas the left superior prefrontal and right angular gyri were additionally activated when a distractor was anticipated. The right temporoparietal junction showed stronger functional coupling with occipital regions during preparation for trials with an isolated target than for trials with a distractor expected. These results show that anticipation of a visual distractor at a known location, remote from the target, can lead to (1) a reduction in the behavioral cost of that distractor, (2) preparatory modulation of the occipital cortex contralateral to the location of the expected distractor, and (3) anticipatory activation of distinct parietal and frontal brain structures. These findings indicate that specific components of preparatory visual attention may be devoted to minimizing the impact of distractors, not just to enhancements of target processing.

Cortical correlates of TMS-induced phantom hand movements revealed with concurrent TMS-fMRI

We studied an amputee patient who experiences a conscious sense of movement (SoM) in her phantom hand, without significant activity in remaining muscles, when transcranial magnetic stimulation (TMS) is applied at appropriate intensity over the corresponding sector of contralateral motor cortex. We used the novel methodological combination of TMS during fMRI to reveal the neural correlates of her phantom SoM. A critical contrast concerned trials at intermediate TMS intensities: low enough not to produce overt activity in remaining muscles; but high enough to produce a phantom SoM on approximately half such trials. Comparing trials with versus without a phantom SoM reported phenomenally, for the same intermediate TMS intensities, factored out any non-specific TMS effects on brain activity to reveal neural correlates of the phantom SoM itself. Areas activated included primary motor cortex, dorsal premotor cortex, anterior intraparietal sulcus, and caudal supplementary motor area, regions that are also involved in some hand movement illusions and motor imagery in normals. This adds support to proposals that a conscious sense of movement for the hand can be conveyed by activity within corresponding motor-related cortical structures.

Estimation efficiency and statistical power in arterial spin labeling fMRI

Arterial spin labeling (ASL) data are typically differenced, sometimes after interpolation, as part of preprocessing before statistical analysis in fMRI. While this process can reduce the number of time points by half, it simplifies the subsequent signal and noise models (i.e., smoothed box-car predictors and white noise). In this paper, we argue that ASL data are best viewed in the same data analytic framework as BOLD fMRI data, in that all scans are modeled and colored noise is accommodated. The data are not differenced, but the control/label effect is implicitly built into the model. While the models using differenced data may seem easier to implement, we show that differencing models fit with ordinary least squares either produce biased estimates of the standard errors or suffer from a loss in efficiency. The main disadvantage to our approach is that non-white noise must be modeled in order to yield accurate standard errors, however, this is a standard problem that has been solved for BOLD data, and the very same software can be used to account for such autocorrelated noise.

Two approaches to repetition suppression

Repetition suppression refers to the phenomenon that prior processing of stimuli (or stimulus attributes) decreases activation elicited by processing subsequent stimuli with identical attributes. We present two complementary approaches to identify regions that show repetition suppression for subsequent sentences with either identical: (1) sentence forms or (2) speakers. The first categorical approach simply compares sentences that are presented in Same and Different blocks. The second factorial approach operationally defines repetition suppression as decreased activation for the subsequent Same stimulus relative to its preceding sentence. To account for nonspecific time confounds, this approach tests for a repetition x condition (Same or Different) interaction. Surprisingly, the two approaches revealed different results: Only the categorical analysis detected sentence repetition effects in multiple regions within a bilateral frontotemporal system that has previously been implicated in sentence processing. These discrepancies might be due to the different efficiencies with which the particular contrasts were estimated or spurious differences in stimuli or attentional set that could not be entirely controlled within a single subject. Finally, we combined the two approaches in a [global null] conjunction analysis.

Combined permutation test and mixed-effect model for group average analysis in fMRI

In group average analyses, we generalize the classical one-sample t test to account for heterogeneous within-subject uncertainties associated with the estimated effects. Our test statistic is defined as the maximum likelihood ratio corresponding to a Gaussian mixed-effect model. The test's significance level is calibrated using the same sign permutation framework as in Holmes et al., allowing for exact specificity control under a mild symmetry assumption about the subjects' distribution. Because our likelihood ratio test does not rely on homoscedasticity, it is potentially more sensitive than both the standard t test and its permutation-based version. We present results from the Functional Imaging Analysis Contest 2005 dataset to support this claim.

Sound alters activity in human V1 in association with illusory visual perception

When a single brief visual flash is accompanied by two auditory bleeps, it is frequently perceived incorrectly as two flashes. Here, we used high field functional MRI in humans to examine the neural basis of this multisensory perceptual illusion. We show that activity in retinotopic visual cortex is increased by the presence of concurrent auditory stimulation, irrespective of any illusory perception. However, when concurrent auditory stimulation gave rise to illusory visual perception, activity in V1 was enhanced, despite auditory and visual stimulation being unchanged. These findings confirm that responses in human V1 can be altered by sound and show that they reflect subjective perception rather than the physically present visual stimulus. Moreover, as the right superior temporal sulcus and superior colliculus were also activated by illusory visual perception, together with V1, they provide a potential neural substrate for the generation of this multisensory illusion.

Differential hippocampal and prefrontal-striatal contributions to instance-based and rule-based learning

It is a topic of current interest whether learning in humans relies on the acquisition of abstract rule knowledge (rule-based learning) or whether it depends on superficial item-specific information (instance-based learning). Here, we identified brain regions that mediate either of the two learning mechanisms by combining fMRI with an experimental protocol shown to be able to dissociate both learning mechanisms. Subjects had to learn object-position conjunctions in several trials and blocks. In a learning condition, either objects (Experiment 1) or positions (Experiment 2) were held constant within-blocks. In contrast to a control condition in which object-position conjunctions were trial-unique, a performance increase within and across-blocks was observed in the learning condition of both experiments. We hypothesized that within-block learning mainly relies on instance-based processes, whereas across-block learning might depend on rule-based mechanisms. A within-block parametric fMRI analysis revealed a learning-related increase of lateral prefrontal and striatal activity and a learning-related decrease of hippocampal activity in both experiments. By contrast, across-block learning was associated with an activation modulation in distinct prefrontal-striatal brain regions, but not in the hippocampus. These data indicate that hippocampal and prefrontal-striatal brain regions differentially contribute to instance-based and rule-based learning.

Dynamic causal modelling of evoked responses in EEG/MEG with lead field parameterization

Dynamical causal modeling (DCM) of evoked responses is a new approach to making inferences about connectivity changes in hierarchical networks measured with electro- and magnetoencephalography (EEG and MEG). In a previous paper, we illustrated this concept using a lead field that was specified with infinite prior precision. With this prior, the spatial expression of each source area, in the sensors, is fixed. In this paper, we show that using lead field parameters with finite precision enables the data to inform the network's spatial configuration and its expression at the sensors. This means that lead field and coupling parameters can be estimated simultaneously. Alternatively, one can also view DCM for evoked responses as a source reconstruction approach with temporal, physiologically informed constraints. We will illustrate this idea using, for each area, a 4-shell equivalent current dipole (ECD) model with three location and three orientation parameters. Using synthetic and real data, we show that this approach furnishes accurate and robust conditional estimates of coupling among sources and their orientations.

Illuminating the BOLD signal: combined fMRI–fNIRS studies

Functional magnetic resonance imaging (fMRI) is currently combined with electrophysiological methods to identify the relationship between neuronal activity and the blood oxygenation level-dependent (BOLD) signal. Several processes like neuronal activity, synaptic activity, vascular dilation, blood volume and oxygenation changes underlie both response modalities, that is, the electrophysiological signal and the vascular response. However, accessing single process relationships is absolutely mandatory when aiming at a deeper understanding of neurovascular coupling and necessitates studies on the individual building blocks of the vascular response. Combined fMRI and functional near-infrared spectroscopy studies have been performed to validate the correlation of the BOLD signal to the hemodynamic changes in the brain. Here we review the current status of the integration of both technologies and judge these studies in the light of recent findings on neurovascular coupling.

A Bayesian approach to determining connectivity of the human brain

Recent work regarding the analysis of brain imaging data has focused on examining functional and effective connectivity of the brain. We develop a novel descriptive and inferential method to analyze the connectivity of the human brain using functional MRI (fMRI). We assess the relationship between pairs of distinct brain regions by comparing expected joint and marginal probabilities of elevated activity of voxel pairs through a Bayesian paradigm, which allows for the incorporation of previously known anatomical and functional information. We define the relationship between two distinct brain regions by measures of functional connectivity and ascendancy. After assessing the relationship between all pairs of brain voxels, we are able to construct hierarchical functional networks from any given brain region and assess significant functional connectivity and ascendancy in these networks. We illustrate the use of our connectivity analysis using data from an fMRI study of social cooperation among women who played an iterated "Prisoner's Dilemma" game. Our analysis reveals a functional network that includes the amygdala, anterior insula cortex, and anterior cingulate cortex, and another network that includes the ventral striatum, orbitofrontal cortex, and anterior insula. Our method can be used to develop causal brain networks for use with structural equation modeling and dynamic causal models.

Bayesian spatio-temporal approach for EEG source reconstruction: conciliating ECD and distributed models

Characterizing the cortical activity sources of electroencephalography (EEG)/magnetoencephalography data is a critical issue since it requires solving an ill-posed inverse problem that does not admit a unique solution. Two main different and complementary source models have emerged: equivalent current dipoles (ECD) and distributed linear (DL) models. While ECD models remain highly popular since they provide an easy way to interpret the solutions, DL models (also referred to as imaging techniques) are known to be more realistic and flexible. In this paper, we show how those two representations of the brain electromagnetic activity can be cast into a common general framework yielding an optimal description and estimation of the EEG sources. From this extended source mixing model, we derive a hybrid approach whose key aspect is the separation between temporal and spatial characteristics of brain activity, which allows to dramatically reduce the number of DL model parameters. Furthermore, the spatial profile of the sources, as a temporal invariant map, is estimated using the entire time window data, allowing to significantly enhance the information available about the spatial aspect of the EEG inverse problem. A Bayesian framework is introduced to incorporate distinct temporal and spatial constraints on the solution and to estimate both parameters and hyperparameters of the model. Using simulated EEG data, the proposed inverse approach is evaluated and compared with standard distributed methods using both classical criteria and ROC curves.

MEG source localization under multiple constraints: An extended Bayesian framework

To use Electroencephalography (EEG) and Magnetoencephalography (MEG) as functional brain 3D imaging techniques, identifiable distributed source models are required. The reconstruction of EEG/MEG sources rests on inverting these models and is ill-posed because the solution does not depend continuously on the data and there is no unique solution in the absence of prior information or constraints. We have described a general framework that can account for several priors in a common inverse solution. An empirical Bayesian framework based on hierarchical linear models was proposed for the analysis of functional neuroimaging data [Friston, K., Penny, W., Phillips, C., Kiebel, S., Hinton, G., Ashburner, J., 2002. Classical and Bayesian inference in neuroimaging: theory. NeuroImage 16, 465-483] and was evaluated recently in the context of EEG [Phillips, C., Mattout, J., Rugg, M.D., Maquet, P., Friston, K., 2005. An empirical Bayesian solution to the source reconstruction problem in EEG. NeuroImage 24, 997-1011]. The approach consists of estimating the expected source distribution and its conditional variance that is constrained by an empirically determined mixture of prior variance components. Estimation uses Expectation-Maximization (EM) to give the Restricted Maximum Likelihood (ReML) estimate of the variance components (in terms of hyperparameters) and the Maximum A Posteriori (MAP) estimate of the source parameters. In this paper, we extend the framework to compare different combinations of priors, using a second level of inference based on Bayesian model selection. Using Monte-Carlo simulations, ReML is first compared to a classic Weighted Minimum Norm (WMN) solution under a single constraint. Then, the ReML estimates are evaluated using various combinations of priors. Both standard criterion and ROC-based measures were used to assess localization and detection performance. The empirical Bayes approach proved useful as: (1) ReML was significantly better than WMN for single priors; (2) valid location priors improved ReML source localization; (3) invalid location priors did not significantly impair performance. Finally, we show how model selection, using the log-evidence, can be used to select the best combination of priors. This enables a global strategy for multiple prior-based regularization of the MEG/EEG source reconstruction.

Segregation of visceral and somatosensory afferents: an fMRI and cytoarchitectonic mapping study

Ano-rectal stimulation provides an important model for the processing of somatosensory and visceral sensations in the human nervous system. In spite of their anatomical proximity, the anal canal is innervated by somatosensory afferents whereas the rectum is innervated by the visceral nervous system. In a functional magnetic resonance (fMRI) experiment, we examined the cerebral responses to pneumatic balloon distension of these two structures to test whether somatosensory and visceral stimulation elicited distinct brain activations in spite of their spinal convergence. The specificity of the identified activations was analyzed by Bayesian mixed effects modeling. Activations in the parietal operculum were also compared to the location of cytoarchitectonically defined areas OP 1-4, which are part of the secondary somatosensory cortex (SII), to analyze whether the SII region was activated by anal and/or rectal stimulation. The lowest segregation between visceral and somatosensory stimuli was in the insular cortex, which supports the interpretation of the insula as an integrative region, receiving input from different sensory modalities. The most distinct segregation was found in the fronto-parietal operculum. Here the activations following anal and rectal stimulation were not only functionally but also anatomically distinct. Anal sensations were processed similar to other somatosensory stimuli in the SII cortex (area OP 4). Rectal afferents on the other hand were not processed in SII. Rather, they evoked activation at a more anterior location on the precentral operculum. These results demonstrate a functionally and anatomically distinct processing of somatosensory and visceral afferents in the human cerebral cortex.

Encoding and the durability of episodic memory: a functional magnetic resonance imaging study

Memories vary in their durability even when encoding conditions apparently remain constant. We investigated whether, under these circumstances, memory durability is nonetheless associated with variation in the neural activity elicited during encoding. Event-related functional magnetic resonance imaging data were acquired while volunteers semantically classified visually presented words. Using the "remember/know" procedure, memory for one-half of the words was tested after 30 min and for the remaining half after 48 h. In several regions, including left hippocampus and left dorsal inferior frontal gyrus (IFG), activity at encoding differed depending on whether items were later recollected regardless of study-test delay. Delay-selective effects were also evident, however. Recollection after 48 h was associated with enhanced activity in bilateral ventral IFG, whereas recollection after 30 min was associated with greater fusiform activity. Thus, there is a relationship between the neural activity elicited by an event as it is encoded and the durability of the resulting memory representation.

Cortical activity in multiple motor areas during sequential finger movements: An application of independent component analysis

Multiple cortical regions such as the supplementary motor area (SMA), premotor cortex (PM), and primary motor cortex (M1) are involved in the sequential execution of hand movements, but it is unclear how these areas collaborate in the preparation and execution of ipsilateral and contralateral hand movements. In this study, we used right-handed subjects to examine the spatial distribution and temporal profiles of motor-related activity during visually cued sequential finger movements by applying independent component analysis (ICA) to event-related functional magnetic resonance imaging (fMRI) signals. The particular merit of the ICA method is that it allows brain activity in individual subjects to be elucidated without making a priori assumptions about the anatomical areas that are activated or the temporal profile of activity. By applying ICA, we found that (1) the SMA contributed to both the preparation and execution of movements of the right and left hand; (2) the left M1 and dorsal premotor cortex (PMd) contributed to both the preparation and execution of movements of the right and left hand, whereas the right M1 and PMd contributed mainly to the execution of movements of the left hand; (3) pre-SMA areas were activated in some subjects in concert with the posterior parietal and prefrontal cortex; and (4) fMRI signals over superficial cortical draining veins could be distinguished from cortical activation. We suggest that ICA is useful for categorizing distributed task-related activities in individual subjects into several spatially independent activities that represent functional units in motor control.

A method for generating reproducible evidence in fMRI studies

Insights into cognitive neuroscience from neuroimaging techniques are now required to go beyond the localisation of well-known cognitive functions. Fundamental to this is the notion of reproducibility of experimental outcomes. This paper addresses the central issue that functional magnetic resonance imaging (fMRI) experiments will produce more desirable information if researchers begin to search for reproducible evidence rather than only p value significance. The study proposes a methodology for investigating reproducible evidence without conducting separate fMRI experiments. The reproducible evidence is gathered from the separate runs within the study. The associated empirical Bayes and ROC extensions of the linear model provide parameter estimates to determine reproducibility. Empirical applications of the methodology suggest that reproducible evidence is robust to small sample sizes and sensitive to both the magnitude and persistency of brain activation. It is demonstrated that research findings in fMRI studies would be more compelling with supporting reproducible evidence in addition to standard hypothesis testing evidence.

High Binaural Coherence Determines Successful Sound Localization and Increased Activity in Posterior Auditory Areas

Our brain continuously receives complex combinations of sounds originating from different sources and relating to different events in the external world. Timing differences between the two ears can be used to localize sounds in space, but only when the inputs to the two ears have similar spectrotemporal profiles (high binaural coherence). We used fMRI to investigate any modulation of auditory responses by binaural coherence. We assessed how processing of these cues depends on whether spatial information is task relevant and whether brain activity correlates with subjects' localization performance. We found that activity in Heschl's gyrus increased with increasing coherence, irrespective of whether localization was task relevant. Posterior auditory regions also showed increased activity for high coherence, primarily when sound localization was required and subjects successfully localized sounds. We conclude that binaural coherence cues are processed throughout the auditory cortex and that these cues are used in posterior regions for successful auditory localization.

Language network specializations: an analysis with parallel task designs and functional magnetic resonance imaging

Although the classical core regions of the language system (Broca's and Wernicke's areas) were defined over a century ago, it took the advent of functional imaging to sharpen our understanding of how these regions and adjacent parts of the brain are associated with particular aspects of language. One limitation of such studies has been the need to compare results across different subject groups, each performing a different type of language task. Thus, this study was designed to examine overlapping versus segregated brain activations associated with three fundamental language tasks, orthography, phonology and semantics performed by the same subjects during a single experimental session. The results demonstrate a set of primarily left-sided core language regions in ventrolateral frontal, supplementary motor, posterior mid-temporal, occipito-temporal and inferior parietal areas, which were activated for all language tasks. Segregated task-specific activations were demonstrated within the ventrolateral frontal, mid-temporal and inferior parietal areas. Within the inferior frontal cortex (Broca's regional complex), segregated activations were seen for the semantic and phonological tasks. These findings demonstrate both common and task specific activations within the language system.

Wortformgedächtnis als intuitive Statistik in Sprachen mit unterschiedlicher Konsistenz

Zusammenfassung. Die Ergebnisse einer Reihe von sprachvergleichenden experimentellen Studien zur Funktionsweise des Wortformgedächtnisses werden vor dem Hintergrund nonlinearer dynamischer Computermodelle der visuellen Worterkennung zusammengefasst. Insgesamt stützen diese Befunde die allgemeine Hypothese, dass das Wortformgedächtnis sich sensibel an die statistischen Regelmäßigkeiten des Schriftsprachsystems anpasst. Insbesondere wird gezeigt, dass nicht die (binäre) Regelmäßigkeit einer Schriftsprache, sondern ihre graduelle Konsistenz die Worterkennungsleistung mitbestimmt.