Temporal autocorrelation in univariate linear modeling of FMRI data

Functional topography of the secondary somatosensory cortex for nonpainful and painful stimulation of median and tibial nerve: an fMRI study

Functional magnetic resonance imaging (fMRI) was used to study the cortical activity of the bilateral secondary somatosensory cortex (SII) during nonpainful (motor threshold) and painful electrical stimulation of median and tibial nerves. fMRI recordings were performed in eight normal young adults. The aim was at evaluating the working hypothesis of a spatial segregation of nonpainful and painful populations not only in the "hand" representation of SII [Ferretti, A., Babiloni, C., Del Gratta, C., Caulo, M., Tartaro, A., Bonomo, L., Rossini, P.M., Romani, G.L., 2003. Functional topography of the secondary somatosensory cortex for nonpainful and painful stimuli: an fMRI study. NeuroImage 20, 1625-1638.] but also in its "foot" representation. Results showed that, in both "hand" and "foot" representations of bilateral SII, the activity elicited by the painful stimulation was localized more posteriorly with respect to that elicited by the nonpainful stimulation. A fine spatial analysis of the SII responses revealed a clear somatotopic organization in the bilateral SII subregion especially reactive to the nonpainful stimuli (i.e., segregation of the hand and foot representations). In contrast, it was not possible to disentangle the "hand" and "foot" representations of SII for painful stimuli. These results extended to the SII "foot" representation previous evidence of a spatial segregation in the SII "hand" representation of subregions for the painful and nonpainful stimuli. Furthermore, they suggest that noxious information is not somatotopically represented in human bilateral SII, at least as inferred from fMRI data at 1.5 T.

Optimal spatial regularisation of autocorrelation estimates in fMRI analysis

In the General Linear Model (GLM) framework for the statistical analysis of fMRI data, the problem of temporal autocorrelations in the residual signal (after regression) has been frequently addressed in the open literature. There exist various methods for correcting the ensuing bias in the statistical testing, among which the prewhitening strategy, which uses a prewhitening matrix for rendering the residual signal white (i.e., without temporal autocorrelations). This correction is only exact when the autocorrelation structure of the noise-generating process is accurately known, and the estimates derived from the fMRI data are too noisy to be used for correction. Recently, Worsley and co-workers proposed to spatially smooth the noisy autocorrelation estimates, effectively reducing their variance and allowing for a better correction. In this article, a systematic study into the effect of the smoothing kernel width is performed and a method is introduced for choosing this bandwidth in an "optimal" manner. Several aspects of the prewhitening strategy are investigated, namely the choice of the autocorrelation estimate (biased or unbiased), the accuracy of the estimates, the degree of spatial regularisation and the order of the autoregressive model used for characterising the noise. The proposed method is extensively evaluated on both synthetic and real fMRI data.

Cognitive modulation of pain-related brain responses depends on behavioral strategy

Interactions of pain and cognition have been studied in humans and animals previously, but the relationship between such behavioral interactions and brain activity is unknown. We aimed to show using functional MRI (fMRI) how a cognitively demanding task (Stroop) modulates pain-related brain activations and conversely, how pain modulates attention-related activity. Reaction time data indicated two types of pain responders: subjects in the A group had a faster Stroop reaction time when pain was concomitant to the attention task, while those in the P group had a slower Stroop performance during painful stimulation. fMRI data obtained during Stroop performance with and without noxious stimulation were subjected to region of interest analyses. We first tested whether brain activity during painful median nerve stimulation was modulated by cognitive load. We next tested whether brain activity during the high conflict cognitive task was modulated by pain. Pain-related activity in three regions, primary (S1), and secondary (S2) somatosensory cortices, and anterior insula, was attenuated by cognitive engagement, but this effect was specific to the A group. Pain-related activations in the caudal and rostral anterior cingulate cortex (ACC) and ventroposterior thalamus were not modulated by cognitive load. None of the areas showing attention-related responses, including bilateral dorsolateral prefrontal and posterior parietal cortices, were modulated by pain. These findings suggest that cortical regions associated with pain can be modulated by cognitive strategies. Furthermore, the distinction of behavioral subgroups may relate to cognitive coping strategies taken by patients with chronic pain.

Left inferior prefrontal cortex activity reflects inhibitory rather than facilitatory priming

Functional neuroimaging has demonstrated reduced activation correlated with behavioral priming effects, a finding generally interpreted in terms of facilitated retrieval of target items in the context of related primes. Without a neutral prime, however, one cannot separate facilitatory effects of related primes from inhibitory effects of unrelated primes. Here we report an auditory semantic priming paradigm with congruent (''The boy bounced the BALL''), neutral (''The next item is BALL''), and incongruent (''Pasta is my favorite kind of BALL'') sentence trials. As previously reported, reduced left inferior prefrontal cortex activation was observed for congruent relative to incongruent trials; however, the neutral condition allowed us to show that the effect arose from increased activation in the incongruent condition rather than reduced activation for congruent trials. Our results suggest that the left inferior prefrontal cortex inhibits interference from prepotent representations in order to select a task-appropriate target, and is consistent with its broader role in behavioral inhibition.

Interactions between decision making and performance monitoring within prefrontal cortex

Our ability to judge the consequences of our actions is central to rational decision making. A large body of evidence implicates primate prefrontal regions in the regulation of this ability. It has proven extremely difficult, however, to separate functional areas in the frontal lobes. Using functional magnetic resonance imaging, we demonstrate complementary and reciprocal roles for the human orbitofrontal (OFC) and dorsal anterior cingulate cortices (ACd) in monitoring the outcome of behavior. Activation levels in these regions were negatively correlated, with activation increasing in the ACd and decreasing in the OFC when the selected response was the result of the participant's own decision. The pattern was reversed when the selected response was guided by the experimenter rather than the participant. These results indicate that the neural mechanisms underlying the way we assess the consequences of choices differ depending on whether we are told what to do or are able to exercise our volition.

Morphology and the internal structure of words

Morphology is the aspect of language concerned with the internal structure of words, and languages vary in the extent to which they rely on morphological structure. Consequently, it is not clear whether morphology is a basic element of a linguistic structure or whether it emerges from systematic regularities between the form and meaning of words. Here, we looked for evidence of morphological structure at a neural systems level by using a visual masked priming paradigm and functional MRI. Form and meaning relations were manipulated in a 2 x 2 design to identify reductions in blood oxygenation level-dependent signal related to shared form (e.g., corner-corn), shared meaning (e.g., idea-notion), and shared morphemes (e.g., boldly-bold, which overlapped in both form and meaning). Relative to unrelated pairs (e.g., ozone-hero), morphologically related items reduced blood oxygenation level-dependent signal in the posterior angular gyrus bilaterally, left occipitotemporal cortex, and left middle temporal gyrus. In the posterior angular gyrus, a neural priming effect was observed for all three priming conditions, possibly reflecting reduced attentional demands rather than overlapping linguistic representations per se. In contrast, the reductions seen in the left occipitotemporal cortex and left middle temporal gyrus corresponded, respectively, to main effects of orthographic and semantic overlap. As neural regions sensitive to morphological structure overlapped almost entirely with regions sensitive to orthographic and semantic relatedness, our results suggest that morphology emerges from the convergence of form and meaning.

Changing Brain Networks for Visuomotor Control With Increased Movement Automaticity

Learning a motor skill is associated with changes in patterns of brain activation with movement. Here we have further characterized these dynamics during fast (short-term) learning of a visuomotor skill using functional magnetic resonance imaging. Subjects ( n = 15) were studied as they learned to visually track a moving target by varying the isometric force applied to a pressure plate held in the right hand. Learning was confirmed by demonstration of improved performance and automaticity (the relative lack of need for conscious attention during task execution). We identified two distinct, time-dependent patterns of functional changes in the brain associated with these behavioral changes. An initial, more attentionally demanding stage of learning was associated with the greatest relative activity in widely distributed, predominantly cortical regions including prefrontal, bilateral sensorimotor, and parietal cortices. The caudate nucleus and ipsilateral cerebellar hemisphere also showed significant activity. Over time, as performance improved, activity in these regions progressively decreased. There was an increase in activity in subcortical motor regions including that of the cerebellar dentate and the thalamus and putamen. Short-term motor-skill learning thus is associated with a progressive reduction of widely distributed activations in cortical regions responsible for executive functions, processing somatosensory feedback and motor planning. The results suggest that early performance gains rely strongly on prefrontal-caudate interactions with later increased activity in a subcortical circuit involving the cerebellum and basal ganglia as the task becomes more automatic. Characterization of these changes provides a potential tool for functional “dissection” of pathologies of movement and motor learning.

A general mechanism for perceptual decision-making in the human brain

Findings from single-cell recording studies suggest that a comparison of the outputs of different pools of selectively tuned lower-level sensory neurons may be a general mechanism by which higher-level brain regions compute perceptual decisions. For example, when monkeys must decide whether a noisy field of dots is moving upward or downward, a decision can be formed by computing the difference in responses between lower-level neurons sensitive to upward motion and those sensitive to downward motion. Here we use functional magnetic resonance imaging and a categorization task in which subjects decide whether an image presented is a face or a house to test whether a similar mechanism is also at work for more complex decisions in the human brain and, if so, where in the brain this computation might be performed. Activity within the left dorsolateral prefrontal cortex is greater during easy decisions than during difficult decisions, covaries with the difference signal between face- and house-selective regions in the ventral temporal cortex, and predicts behavioural performance in the categorization task. These findings show that even for complex object categories, the comparison of the outputs of different pools of selectively tuned neurons could be a general mechanism by which the human brain computes perceptual decisions.

fMRI activation maps based on the NN-ARx model

The most significant progresses in the understanding of human brain functions have been possible due to the use of functional magnetic resonance imaging (fMRI), which when used in combination with other standard neuroimaging techniques (i.e., EEG) provides researchers with a potential tool to elucidate many biophysical principles, established previously by animal comparative studies. However, to date, most of the methods proposed in the literature seeking fMRI signs have been limited to the use of a top-down data analysis approach, thus ignoring a pool of physiological facts. In spite of the important contributions achieved by applying these methods to actual data, there is a disproportionate gap between theoretical models and data-analysis strategies while trying to focus on several new prospects, like for example fMRI/EEG data fusion, causality/connectivity patterns, and nonlinear BOLD signal dynamics. In this paper, we propose a new approach which will allow many of the abovementioned hot topics to be addressed in the near future with an underlying interpretability based on bottom-up modeling. In particular, the theta-MAP presented in the paper to test brain activation corresponds very well with the standardized t test of the SPM99 toolbox. Additionally, a new Impulse Response Function (IRF) has been formulated, directly related to the well-established concept of the hemodynamics response function (HRF). The model uses not only the information contained in the signal but also that in the structure of the background noise to simultaneously estimate the IRF and the autocorrelation function (ACF) by using an autoregressive (AR) model with a filtered Poisson process driving the dynamics. The short-range contributions of voxels within the near-neighborhood are also included, and the potential drift was characterized by a polynomial series. Since our model originated from an immediate extension of the hemodynamics approach [Friston, K.J., Mechelli, A., Turner, R., Price C.J. (2000a). Nonlinear responses in fMRI: the balloon model, volterra kernels, and other hemodynamics. NeuroImage 12, 466-477.], a natural interpretability of the results is feasible.

Changes in connectivity profiles define functionally distinct regions in human medial frontal cortex

A fundamental issue in neuroscience is the relation between structure and function. However, gross landmarks do not correspond well to microstructural borders and cytoarchitecture cannot be visualized in a living brain used for functional studies. Here, we used diffusion-weighted and functional MRI to test structure-function relations directly. Distinct neocortical regions were defined as volumes having similar connectivity profiles and borders identified where connectivity changed. Without using prior information, we found an abrupt profile change where the border between supplementary motor area (SMA) and pre-SMA is expected. Consistent with this anatomical assignment, putative SMA and pre-SMA connected to motor and prefrontal regions, respectively. Excellent spatial correlations were found between volumes defined by using connectivity alone and volumes activated during tasks designed to involve SMA or pre-SMA selectively. This finding demonstrates a strong relationship between structure and function in medial frontal cortex and offers a strategy for testing such correspondences elsewhere in the brain.

Statistical parametric mapping for event-related potentials: I. Generic considerations

In this paper, we frame the strategy and motivations behind developments in statistical parametric mapping (SPM) for the analysis of electroencephalogram (EEG) data. This work deals specifically with SPM procedures for the analysis of event-related potentials (ERP). We place these developments in the larger context of integrating electrophysiological and hemodynamic measurements of evoked brain responses through the fusion of EEG and fMRI data. In this paper, we consider some fundamental issues when selecting an appropriate statistical model that enables diverse questions to be asked of the data and at the same time retains maximum sensitivity. The three key issues addressed in this paper are as follows: (i) should multivariate or mass univariate analyses be adopted, (ii) should time be treated as an experimental factor or as a dimension of the measured response variable, and (iii) how to form appropriate explanatory variables in a hierarchical observation model. We review the relative merits of the different options and explain the rationale for our choices. In brief, we motivate a mass univariate approach in terms of sensitivity to region-specific responses. This involves modeling responses at each voxel or space bin separately. In contradistinction, we treat time as an experimental factor to enable inferences about temporally distributed responses that encompass multiple time bins. In a companion paper, we develop statistical models of ERPs in the time domain that follow from the heuristics established here and illustrate the approach using simulated and real data.

Resting fluctuations in arterial carbon dioxide induce significant low frequency variations in BOLD signal

Carbon dioxide is a potent cerebral vasodilator. We have identified a significant source of low-frequency variation in blood oxygen level-dependent (BOLD) magnetic resonance imaging (MRI) signal at 3 T arising from spontaneous fluctuations in arterial carbon dioxide level in volunteers at rest. Fluctuations in the partial pressure of end-tidal carbon dioxide (Pet(CO(2))) of +/-1.1 mm Hg in the frequency range 0-0.05 Hz were observed in a cohort of nine volunteers. Correlating with these fluctuations were significant generalized grey and white matter BOLD signal fluctuations. We observed a mean (+/-standard error) regression coefficient across the group of 0.110 +/- 0.033% BOLD signal change per mm Hg CO(2) for grey matter and 0.049 +/- 0.022% per mm Hg in white matter. Pet(CO(2))-related BOLD signal fluctuations showed regional differences across the grey matter, suggesting variability of the responsiveness to carbon dioxide at rest. Functional magnetic resonance imaging (fMRI) results were corroborated by transcranial Doppler (TCD) ultrasound measurements of the middle cerebral artery (MCA) blood velocity in a cohort of four volunteers. Significant Pet(CO(2))-correlated fluctuations in MCA blood velocity were observed with a lag of 6.3 +/- 1.2 s (mean +/- standard error) with respect to Pet(CO(2)) changes. This haemodynamic lag was adopted in the analysis of the BOLD signal. Doppler ultrasound suggests that a component of low-frequency BOLD signal fluctuations is mediated by CO(2)-induced changes in cerebral blood flow (CBF). These fluctuations are a source of physiological noise and a potentially important confounding factor in fMRI paradigms that modify breathing. However, they can also be used for mapping regional vascular responsiveness to CO(2).

Evaluation of mixed effects in event-related fMRI studies: impact of first-level design and filtering

With the introduction of event-related designs in fMRI, it has become crucial to optimize design efficiency and temporal filtering to detect activations at the 1st level with high sensitivity. We investigate the relevance of these issues for fMRI population studies, that is, 2nd-level analysis, for a set of event-related fMRI (er-fMRI) designs with different 1st-level efficiencies, adopting three distinct 1st-level filtering strategies as implemented in SPM99, SPM2, and FSL3.0. By theory, experiments, and simulations using physiological fMRI noise, we show that both design and filtering impact the outcome of the statistical analysis, not only at the 1st but also at the 2nd level. There are several reasons behind this finding. First, sensitivity is affected by both design and filtering, since the scan-to-scan variance, that is the fixed effect, is not negligible with respect to the between-subject variance, that is the random effect, in er-fMRI population studies. The impact of the fixed effects error on the sensitivity of the mixed effects analysis can be mitigated by an optimal choice of er-fMRI design and filtering. Moreover, the accuracy of the 1st- and 2nd-level parameter estimates also depend on design and filtering; especially, we show that inaccuracies caused by the presence of residual noise autocorrelations can be constrained by designs that have hemodynamic responses with a Gaussian distribution. In conclusion, designs with both good efficiency and decorrelating properties, for example, such as the geometric or Latin square probability distributions, combined with the "whitening" filters of SPM2 and FSL3.0, give the best result, both for 1st- and 2nd-level analysis of er-fMRI studies.

Detection and detrending in fMRI data analysis

This article addresses the impact that colored noise, temporal filtering, and temporal detrending have on the fMRI analysis situation. Specifically, it is shown why the detection of event-related designs benefit more from pre-whitening than blocked designs in a colored noise structure. Both theoretical and empirical results are provided. Furthermore, a novel exploratory method for producing drift models that efficiently capture trends and drifts in the fMRI data is introduced. A comparison to currently employed detrending approaches is presented. It is shown that the novel exploratory model is able to remove a major part of the slowly varying drifts that are abundant in fMRI data. The value of such a model lies in its ability to remove drift components that otherwise would have contributed to a colored noise structure in the voxel time series.

Altered cerebellar functional connectivity mediates potential adaptive plasticity in patients with multiple sclerosis

BACKGROUND

The cerebellum is of potential interest for understanding adaptive responses in motor control in patients with multiple sclerosis because of the high intrinsic synaptic plasticity of this brain region.

OBJECTIVE

To assess the relative roles of interactions between the neocortex and the cerebellum using measures of functional connectivity.

METHODS

A role for altered neocortical-cerebellar functional connectivity in adaptive responses to injury from multiple sclerosis was tested using 1.5 T functional magnetic resonance imaging (fMRI) during figure writing with the dominant right hand in patients with predominantly early relapsing-remitting multiple sclerosis.

RESULTS

Patients (n = 14) showed a more bihemispheric pattern of activation in motor cortex than healthy controls (n = 11). Correlations between task related signal changes in neocortical and cerebellar regions of interest were used as a measure of functional connectivity. Healthy controls showed strong functional connectivity between the left motor cortex and the right cerebellar dentate nucleus. Significant connectivity between the left primary motor cortex and the right dentate was not found in patients. However, patients had significant connectivity between the left premotor neocortex and the ipsilateral (left) cerebellar cortex (crus I), which was not found in healthy controls.

CONCLUSIONS

Changes in apparent cerebellar-neocortical functional connectivity may mediate potentially adaptive changes in brain motor control in patients with multiple sclerosis. Similar changes in the cerebellum and premotor cortex have been reported in the healthy brain during motor learning, suggesting that common mechanisms may contribute to normal motor learning and motor recovery after injury from multiple sclerosis.

Constrained linear basis sets for HRF modelling using Variational Bayes

FMRI modelling requires flexible haemodynamic response function (HRF) modelling, with the HRF being allowed to vary spatially and between subjects. To achieve this flexibility, voxelwise parameterised HRFs have been proposed; however, inference on such models is very slow. An alternative approach is to use basis functions allowing inference to proceed in the more manageable General Linear Model (GLM) framework. However, a large amount of the subspace spanned by the basis functions produces nonsensical HRF shapes. In this work we propose a technique for choosing a basis set, and then the means to constrain the subspace spanned by the basis set to only include sensible HRF shapes. Penny et al. showed how Variational Bayes can be used to infer on the GLM for FMRI. Here we extend the work of Penny et al. to give inference on the GLM with constrained HRF basis functions and with spatial Markov Random Fields on the autoregressive noise parameters. Constraining the subspace spanned by the basis set allows for far superior separation of activating voxels from nonactivating voxels in FMRI data. We use spatial mixture modelling to produce final probabilities of activation and demonstrate increased sensitivity on an FMRI dataset.

Multilevel linear modelling for FMRI group analysis using Bayesian inference

Functional magnetic resonance imaging studies often involve the acquisition of data from multiple sessions and/or multiple subjects. A hierarchical approach can be taken to modelling such data with a general linear model (GLM) at each level of the hierarchy introducing different random effects variance components. Inferring on these models is nontrivial with frequentist solutions being unavailable. A solution is to use a Bayesian framework. One important ingredient in this is the choice of prior on the variance components and top-level regression parameters. Due to the typically small numbers of sessions or subjects in neuroimaging, the choice of prior is critical. To alleviate this problem, we introduce to neuroimage modelling the approach of reference priors, which drives the choice of prior such that it is noninformative in an information-theoretic sense. We propose two inference techniques at the top level for multilevel hierarchies (a fast approach and a slower more accurate approach). We also demonstrate that we can infer on the top level of multilevel hierarchies by inferring on the levels of the hierarchy separately and passing summary statistics of a noncentral multivariate t distribution between them.

Precision of the CASL-perfusion MRI technique for the measurement of cerebral blood flow in whole brain and vascular territories

PURPOSE

To analyze the precision of cerebral blood flow (CBF) measurements made with continuous arterial spin labeling(CASL) perfusion magnetic resonance imaging (MRI) over experimentally relevant intervals.

MATERIALS AND METHODS

CASL perfusion MRI measurements of CBF on a 1.5-T GE Signa magnet were repeated in young healthy male and female subjects at one hour and one week. Precision of the measurement was evaluated at both time intervals.

RESULTS

CASL perfusion MRI measurements of CBF yielded within-subject coefficients of variation (wsCV) of 5.8% for global and 13% for individual vascular regions when measurements were repeated within one hour. Differences in these values represent the error in post-processing. Global and regional CBF measurements over one week yielded wsCVs of 13% and 14%, respectively. At one week, error secondary to physiologic variability affected global and regional measurements to the same degree and masked the software post-processing error seen at one hour. The magnitude of the difference in repeated measures correlated with the magnitude of the measurement.

CONCLUSION

CASL perfusion MRI CBF measurements are accurate and precise. Variability over longer periods of time appears attributable to physiologic factors. Repeatability of the CASL measurement is sensitive to the magnitude of the measurement. This should be taken into account when studies requiring repeated measures involve subjects with significant variability in CBF.

Distinct portions of anterior cingulate cortex and medial prefrontal cortex are activated by reward processing in separable phases of decision-making cognition

BACKGROUND

Choosing between actions associated with uncertain rewards and punishments is mediated by neural circuitry encompassing the orbitofrontal cortex, anterior cingulate cortex (ACC), and striatum; however, the precise conditions under which these different components are activated during decision-making cognition remain uncertain.

METHODS

Fourteen healthy volunteers completed an event-based functional magnetic resonance imaging protocol to investigate blood-oxygenation-level-dependent (BOLD) responses during independently modeled phases of choice cognition. In the "decision phase," participants decided which of two simultaneous visually presented gambles they wished to play for monetary reward. The gambles differed in their magnitude of gains, magnitude of losses, and the probabilities with which these outcomes were delivered. In the "outcome phase," the result of each choice was indicated on the visual display.

RESULTS

In the decision phase, choices involving large gains were associated with increased BOLD responses in the pregenual ACC, paracingulate, and right posterior orbitolateral cortex compared with choices involving small gains. In the outcome phase, good outcomes were associated with increased BOLD responses in the posterior orbitomedial cortex, subcallosal ACC, and ventral striatum compared with negative outcomes. There was only limited overlap between reward-related activity in ACC and orbitofrontal cortex during the decision and outcome phases.

CONCLUSIONS

Neural activity within the medial and lateral orbitofrontal cortex, pregenual ACC, and striatum mediate distinct representations of reward-related information that are deployed at different stages during a decision-making episode.

Using fMRI to quantify the time dependence of remifentanil analgesia in the human brain

To understand and exploit centrally acting drugs requires reliable measures of their time course of action in the human brain. Functional magnetic resonance imaging (fMRI) is able to measure noninvasively, drug-induced changes in task-related brain activity. Here, we have characterized, in a specific region of the brain, the time of onset of action and the half-life of action of a clinically relevant dose of a potent opioid analgesic agent, remifentanil. These times were established from the temporal variation of the amplitude of the blood oxygen level-dependent response in the insular cortex contralateral to a painfully hot thermal stimulus, in volunteers receiving a remifentanil infusion. The insular cortex has repeatedly been reported as activated by noxious thermal stimulation. The times of onset and offset of drug action were each characterized by a half-life for changes in fMRI signal from within the insula. These characteristic times agreed with the observed drug-induced analgesia and previous pharmacokinetic-pharmacodynamic measurements for remifentanil. We have successfully measured, for the first time using fMRI, temporal pharmacological parameters for a CNS-active drug based on its effect on task-related activity in a specific brain region. Comparison of the time course of regional brain activity with pain perception could reveal those regions engaged in drug-induced analgesia.

Towards an understanding of gait control: brain activation during the anticipation, preparation and execution of foot movements

While a detailed understanding of brain activity with hand movements has developed, less is known about the functional anatomy of motor control for foot movements. Here we have used fMRI to define brain activity associated with unilateral foot extension and flexion, component movements of gait. We studied brain responses to visually cued active and passive movements and periods of either preparation (before active movement) or anticipation (before passive movement) with a pseudo-randomized block design. A mixed-effects (n = 12) contrast of the active movement condition vs. rest identified brain activation in regions including the medial wall of the primary sensorimotor cortex, consistent with expected somatotopy. Medial wall activation during passive movement vs. rest was less intense and localized to the same region. Frontal and association cortices were more active during preparation or anticipation periods than during the movements themselves. A contrast of preparation to move vs. active movement showed significant activation in the medial frontal and frontopolar gyri and the precuneus. Contrast of the anticipation of movement with the passive movement condition revealed activation in the dorsal premotor cortex and precuneus. Our study thus provides evidence for somatotopy in multiple functional regions in the motor control network. The anterior prefrontal activity is involved in the preparation for cued movement with distinct regions of the medial motor cortex (including SMA and CMA) preferentially involved in motor program planning and execution. This direct characterization of brain activation patterns associated with foot movements promises use of fMRI for the functional analysis of pathologies of gait.

Probabilistic independent component analysis for functional magnetic resonance imaging

We present an integrated approach to probabilistic independent component analysis (ICA) for functional MRI (FMRI) data that allows for nonsquare mixing in the presence of Gaussian noise. In order to avoid overfitting, we employ objective estimation of the amount of Gaussian noise through Bayesian analysis of the true dimensionality of the data, i.e., the number of activation and non-Gaussian noise sources. This enables us to carry out probabilistic modeling and achieves an asymptotically unique decomposition of the data. It reduces problems of interpretation, as each final independent component is now much more likely to be due to only one physical or physiological process. We also describe other improvements to standard ICA, such as temporal prewhitening and variance normalization of timeseries, the latter being particularly useful in the context of dimensionality reduction when weak activation is present. We discuss the use of prior information about the spatiotemporal nature of the source processes, and an alternative-hypothesis testing approach for inference, using Gaussian mixture models. The performance of our approach is illustrated and evaluated on real and artificial FMRI data, and compared to the spatio-temporal accuracy of results obtained from classical ICA and GLM analyses.

Fully Bayesian spatio-temporal modeling of FMRI data

We present a fully Bayesian approach to modeling in functional magnetic resonance imaging (FMRI), incorporating spatio-temporal noise modeling and haemodynamic response function (HRF) modeling. A fully Bayesian approach allows for the uncertainties in the noise and signal modeling to be incorporated together to provide full posterior distributions of the HRF parameters. The noise modeling is achieved via a nonseparable space-time vector autoregressive process. Previous FMRI noise models have either been purely temporal, separable or modeling deterministic trends. The specific form of the noise process is determined using model selection techniques. Notably, this results in the need for a spatially nonstationary and temporally stationary spatial component. Within the same full model, we also investigate the variation of the HRF in different areas of the activation, and for different experimental stimuli. We propose a novel HRF model made up of half-cosines, which allows distinct combinations of parameters to represent characteristics of interest. In addition, to adaptively avoid over-fitting we propose the use of automatic relevance determination priors to force certain parameters in the model to zero with high precision if there is no evidence to support them in the data. We apply the model to three datasets and observe matter-type dependence of the spatial and temporal noise, and a negative correlation between activation height and HRF time to main peak (although we suggest that this apparent correlation may be due to a number of different effects).

Functional asymmetry for auditory processing in human primary auditory cortex

Structural asymmetries in the supratemporal plane of the human brain are often cited as the anatomical basis for the lateralization of language predominantly to the left hemisphere. However, similar asymmetries are found for structures mediating earlier events in the auditory processing stream, suggesting that functional lateralization may occur even at the level of primary auditory cortex. We tested this hypothesis using functional magnetic resonance imaging to evaluate human auditory cortex responses to monaurally presented tones. Relative to silence, tones presented separately to either ear produced greater activation in left than right Heschl's gyrus, the location of primary auditory cortex. This functional lateralization for primary auditory cortex is distinct from the contralateral dominance reported for other mammals, including nonhuman primates, and may have contributed to the evolution of a unique role for the left hemisphere in language processing.

Model assessment and model building in fMRI

Model quality is rarely assessed in fMRI data analyses and less often reported. This may have contributed to several shortcomings in the current fMRI data analyses, including: (1) Model mis-specification, leading to incorrect inference about the activation-maps, SPM[t] and SPM[F]; (2) Improper model selection based on the number of activated voxels, rather than on model quality; (3) Under-utilization of systematic model building, resulting in the common but suboptimal practice of using only a single, pre-specified, usually over-simplified model; (4) Spatially homogenous modeling, neglecting the spatial heterogeneity of fMRI signal fluctuations; and (5) Lack of standards for formal model comparison, contributing to the high variability of fMRI results across studies and centers. To overcome these shortcomings, it is essential to assess and report the quality of the models used in the analysis. In this study, we applied images of the Durbin-Watson statistic (DW-map) and the coefficient of multiple determination (R(2)-map) as complementary tools to assess the validity as well as goodness of fit, i.e., quality, of models in fMRI data analysis. Higher quality models were built upon reduced models using classic model building. While inclusion of an appropriate variable in the model improved the quality of the model, inclusion of an inappropriate variable, i.e., model mis-specification, adversely affected it. Higher quality models, however, occasionally decreased the number of activated voxels, whereas lower quality or inappropriate models occasionally increased the number of activated voxels, indicating that the conventional approach to fMRI data analysis may yield sub-optimal or incorrect results. We propose that model quality maps become part of a broader package of maps for quality assessment in fMRI, facilitating validation, optimization, and standardization of fMRI result across studies and centers. Hum. Brain Mapping 20:227-238, 2003.

CNS Activation by Noxious Heat to the Hand or Foot: Site-Dependent Delay in Sensory But Not Emotion Circuitry

Recently, functional magnetic resonance imaging has been used as a novel method of evaluating the CNS response to noxious stimuli. In a previous study, a prolonged noxious thermal stimulus applied to the dorsum of the hand produced more than one hemodynamic response that was temporally segregated. The two major responses displayed activation in primary sensory regions (classic pain circuitry) and regions involved in emotion (reward/aversion circuitry), respectively. In the current study, we applied the same thermal stimulus separately to the dorsum of the left foot and the dorsum of the left hand in the same subjects and compared the hemodynamic responses to evaluate the effects of conduction distance on CNS activation within these two segregated systems. After stimulus delivery to the foot, the hemodynamic response in primary sensory networks occurs after a delay of 3.6 ± 1.3 s as compared with the response after hand stimulation. The relative delay of the hemodynamic response in reward/aversion regions is not significantly different between hand and foot stimulation (0.6 ± 2.1 s). These results within the primary sensory system are consistent with the greater conduction distance of the peripheral nerves from the hand versus the foot. The observation that the response within the reward/aversion pathways occurs with the same rapid temporal characteristics after either hand or foot stimulation supports the notion that the circuitry involved in the evaluation of aversive stimuli is rapid in onset and probably represents a major protective mechanism for survival.

Advances in functional and structural MR image analysis and implementation as FSL

The techniques available for the interrogation and analysis of neuroimaging data have a large influence in determining the flexibility, sensitivity, and scope of neuroimaging experiments. The development of such methodologies has allowed investigators to address scientific questions that could not previously be answered and, as such, has become an important research area in its own right. In this paper, we present a review of the research carried out by the Analysis Group at the Oxford Centre for Functional MRI of the Brain (FMRIB). This research has focussed on the development of new methodologies for the analysis of both structural and functional magnetic resonance imaging data. The majority of the research laid out in this paper has been implemented as freely available software tools within FMRIB's Software Library (FSL).

Functional connectivity: studying nonlinear, delayed interactions between BOLD signals

Correlation analysis has been widely used in the study of functional connectivity based on fMRI data. It assumes that the relevant information about the interactions of brain regions is reflected by a linear relationship between the values of two signals at the same time. However, this hypothesis has not been thoroughly investigated yet. In this work, we study in depth the information shared by BOLD signals of pairs of brain regions. In particular, we assess the amount of nonlinear and/or nonsynchronous interactions present in data. This is achieved by testing models reflecting linear, synchronous interactions against more general models, encompassing nonlinear, nonsynchronous interactions. Many factors influencing measured BOLD signals are critical for the study of connectivity, such as paradigm-induced BOLD responses, preprocessing, motion artifacts, and geometrical distortions. Interactions are also influenced by the proximity of brain regions. The influence of all these factors is taken into account and the nature of the interactions is studied using various experimental conditions such that the conclusions reached are robust with respect to variation of these factors. After defining nonlinear and/or nonsynchronous interaction models in the framework of general linear models, statistical tests are performed on different fMRI data sets to infer the nature of the interactions. Finally, a new connectivity metric is proposed which takes these inferences into account. We find that BOLD signal interactions are statistically more significant when taking into account the history of the distant signal, i.e., the signal from the interacting region, than when using a model of linear instantaneous interaction. Moreover, about 75% of the interactions are symmetric, as assessed with the proposed connectivity metric. The history-dependent part of the coupling between brain regions can explain a high percentage of the variance in the data sets studied. As these results are robust with respect to various confounding factors, this work suggests that models used to study the functional connectivity between brain areas should in general take the BOLD signal history into account.

Quantitative NumART2* mapping in functional MRI studies at 1.5 T

Quantitative mapping of the effective transverse relaxation time, T2* and proton density was performed in a motor activation functional MRI (fMRI) study using multi-echo, echo planar imaging (EPI) and NumART2* (Numerical Algorithm for Real time T2*). Comparisons between NumART2* and conventional single echo EPI with an echo time of 64 ms were performed for five healthy participants examined twice. Simulations were also performed to address specific issues associated with the two techniques, such as echo time-dependent signal variation. While the single echo contrast varied with the baseline T2* value, relative changes in T2* remained unaffected. Statistical analysis of the T2* maps yielded fMRI activation patterns with an improved statistical detection relative to conventional EPI but with less activated voxels, suggesting that NumART2* has superior spatial specificity. Two effects, inflow and dephasing, that may explain this finding were investigated. Particularly, a statistically significant increase in proton density was found in a brain area that was detected as activated by conventional EPI but not by NumART2* while no such changes were observed in brain areas that showed stimulus correlated signal changes on T2* maps.

Stimulus symmetry affects the bisection of figures but not lines: evidence from event-related fMRI

Many geometric shapes retain their symmetry when bisected, but appear asymmetrical when misbisected. We have previously shown that this correspondence can guide the accuracy and speed of perceptual bisection (Landmark) judgments. Using event-related fMRI, here we examined whether the behavioural effects of symmetry are also evident at the neural level. The data showed that the presence/absence of symmetry modulates the activity of right anterior cingulate gyrus, an area associated with a variety of higher level attentional functions. A previous visual half-field study also showed that bisected lines are apprehended more quickly and accurately than misbisected lines in right, but not left, visual field. We were able to localise this advantage to right superior temporal gyrus. Significantly, we found no evidence that symmetry played a role in apprehending the midpoint of the line stimuli traditionally used to assess visual neglect. The data clarify the effects of visual symmetry on bisection behaviour, and highlight novel dissociations within the neural systems thought to underline Landmark performance.

Functional topography of the secondary somatosensory cortex for nonpainful and painful stimuli: an fMRI study

The regional activity of the contralateral primary (SI) and the bilateral secondary (SII) somatosensory areas during median nerve stimulations at five intensity levels (ranging from nonpainful motor threshold to moderate pain) was studied by means of functional magnetic resonance imaging (fMRI). The aim was to characterize the functional topography of SII compared to SI as a function of the stimulus intensity. Results showed that the galvanic stimulation of the median nerve activated the contralateral SI at all stimulus intensities. When considered as a single region, SII was more strongly activated in the contralateral than in the ipsilateral hemisphere. When a finer spatial analysis of the SII responses was performed, the activity for the painful stimulation was localized more posteriorly compared to that for the nonpainful stimulation. This is the first report on such a SII segregation for transient galvanic stimulations. The activity (relative signal intensity) of this posterior area increased with the increase of the stimulus intensity. These results suggest a spatial segregation of the neural populations that process signals conveyed by dorsal column-medial lemniscus (nonpainful signals) and neospinothalamic (painful signals) pathways. Further fMRI experiments should evaluate the functional properties of these two SII subregions during tasks involving sensorimotor integration, learning, and memory demands.

Time-series analysis of MRI intensity patterns in multiple sclerosis

In progressive neurological disorders, such as multiple sclerosis (MS), magnetic resonance imaging (MRI) follow-up is used to monitor disease activity and progression and to understand the underlying pathogenic mechanisms. This article presents image postprocessing methods and validation for integrating multiple serial MRI scans into a spatiotemporal volume for direct quantitative evaluation of the temporal intensity profiles. This temporal intensity signal and its dynamics have thus far not been exploited in the study of MS pathogenesis and the search for MRI surrogates of disease activity and progression. The integration into a four-dimensional data set comprises stages of tissue classification, followed by spatial and intensity normalization and partial volume filtering. Spatial normalization corrects for variations in head positioning and distortion artifacts via fully automated intensity-based registration algorithms, both rigid and nonrigid. Intensity normalization includes separate stages of correcting intra- and interscan variations based on the prior tissue class segmentation. Different approaches to image registration, partial volume correction, and intensity normalization were validated and compared. Validation included a scan-rescan experiment as well as a natural-history study on MS patients, imaged in weekly to monthly intervals over a 1-year follow-up. Significant error reduction was observed by applying tissue-specific intensity normalization and partial volume filtering. Example temporal profiles within evolving multiple sclerosis lesions are presented. An overall residual signal variance of 1.4% +/- 0.5% was observed across multiple subjects and time points, indicating an overall sensitivity of 3% (for axial dual echo images with 3-mm slice thickness) for longitudinal study of signal dynamics from serial brain MRI.

General multilevel linear modeling for group analysis in FMRI

This article discusses general modeling of multisubject and/or multisession FMRI data. In particular, we show that a two-level mixed-effects model (where parameters of interest at the group level are estimated from parameter and variance estimates from the single-session level) can be made equivalent to a single complete mixed-effects model (where parameters of interest at the group level are estimated directly from all of the original single sessions' time series data) if the (co-)variance at the second level is set equal to the sum of the (co-)variances in the single-level form, using the BLUE with known covariances. This result has significant implications for group studies in FMRI, since it shows that the group analysis requires only values of the parameter estimates and their (co-)variance from the first level, generalizing the well-established "summary statistics" approach in FMRI. The simple and generalized framework allows different prewhitening and different first-level regressors to be used for each subject. The framework incorporates multiple levels and cases such as repeated measures, paired or unpaired t tests and F tests at the group level; explicit examples of such models are given in the article. Using numerical simulations based on typical first-level covariance structures from real FMRI data we demonstrate that by taking into account lower-level covariances and heterogeneity a substantial increase in higher-level Z score is possible.

Dynamical components analysis of fMRI data through kernel PCA

In parallel with standard model-based methods for the analysis of fMRI data, exploratory methods--such as PCA, ICA, and clustering--have been developed to give an account of the dataset with minimal priors: no assumption is made on the data content itself, but the data structure is assumed to show some properties (decorrelation, independence) that allow for the detection of structures of interest. In this paper, we present an alternative that tries to take into account some relevant knowledge for the analysis of the dataset, e.g., the experimental paradigm, while keeping the flexibility of exploratory methods: we use a prior temporal modeling of the data that characterizes each voxel time course. Two implementations are proposed: one based on the General Linear Model, the other one on more flexible short-term predictors, whose complexity is controlled by a Minimum Description Length approach. However, our main concern here is the construction of a multivariate model; the latter is performed with the help of a kernel PCA method that builds a redundant representation of the data through the nonlinearity of the kernel. This allows for a refinement in the description of the (temporal) patterns of interest. In particular, this helps in the characterization of subtle variations in the response to different experimental conditions. We illustrate the usefulness of nonlinearity through the analysis of a synthetic dataset and show on a real dataset how it helps to interpret the experimental results.

Empirical analyses of null-hypothesis perfusion FMRI data at 1.5 and 4 T

Functional magnetic resonance imaging (fMRI) based on arterial spin labeling (ASL) perfusion contrast is an emergent methodology for visualizing brain function both at rest and during task performance. Because of the typical pairwise subtraction approach in generating perfusion images, ASL contrast manifests different noise properties and offers potential advantages for some experimental designs as compared with blood oxygenation-level-dependent (BOLD) contrast. We studied the noise properties and statistical power of ASL contrast, with a focus on temporal autocorrelation and spatial coherence, at both 1.5- and 4.0-T field strengths. Perfusion fMRI time series were found to be roughly independent in time, and voxelwise statistical analysis assuming independence of observations yielded false-positive rates compatible with theoretical values using appropriate analysis methods. Unlike BOLD fMRI data, perfusion data were not found to have spatial coherence that varied across temporal frequency. This finding has implications for the application of spatial smoothing to perfusion data. It was also found that the spatial coherence of the ASL data is greater at high magnetic field than low field, and including the global signal as a covariate in the general linear model improves the central tendency of test statistic as well as reduces the noise level in perfusion fMRI, especially at high magnetic field.

Neural correlates of interindividual differences in the subjective experience of pain

Some individuals claim that they are very sensitive to pain, whereas others say that they tolerate pain well. Yet, it is difficult to determine whether such subjective reports reflect true interindividual experiential differences. Using psychophysical ratings to define pain sensitivity and functional magnetic resonance imaging to assess brain activity, we found that highly sensitive individuals exhibited more frequent and more robust pain-induced activation of the primary somatosensory cortex, anterior cingulate cortex, and prefrontal cortex than did insensitive individuals. By identifying objective neural correlates of subjective differences, these findings validate the utility of introspection and subjective reporting as a means of communicating a first-person experience.

The single-epoch fMRI design: validation of a simplified paradigm for the collection of subjective ratings

One of the goals of human functional imaging studies is to interpret brain activation in the context of an individual's subjective experience. However, functional magnetic resonance imaging (fMRI) studies usually employ a block design that consists of multiple epochs of stimulation; this strategy does not readily allow subjective responses to be assessed on a stimulus-by-stimulus basis. To address this issue, we developed a "single-epoch" design, consisting of a single stimulation period presented between two baseline periods. This allows subjective ratings to be acquired after each stimulus, while minimizing rating-induced confounds. To evaluate its sensitivity and utility, we obtained fMRI data using single-epoch and block designs (five stimuli between six baselines) and assessed regional brain activations evoked by both visual (a checkerboard pattern) and painful (noxious heat to right calf) stimuli. For both types of stimulation, data collected using the single-epoch design displayed activation patterns that were generally similar to those detected with the block design. Furthermore, only one single-epoch acquisition series was sufficient to detect bilateral activation in the visual cortex during visual stimulation and activation in the primary somatosensory cortex, the anterior cingulate cortex, and other regions during painful stimulation. In addition, analyses of a series of single-epoch data from a single individual revealed a stimulus-by-stimulus decrease in the activation in the anterior cingulate cortex that paralleled the decrease in the subject's psychophysical responses. These findings confirm that the single-epoch design is sensitive to regional signal changes and serves as a viable alternative to the block design when the collection of subjective responses is of critical importance.

Diagnosis and exploration of massively univariate neuroimaging models

The goal of this work is to establish the validity of neuroimaging models and inferences through diagnosis and exploratory data analysis. While model diagnosis and exploration are integral parts of any statistical modeling enterprise, these aspects have been mostly neglected in functional neuroimaging. We present methods that make diagnosis and exploration of neuroimaging data feasible. We use three- and one-dimensional summaries that characterize the model fit and the four-dimensional residuals. The statistical tools are diagnostic summary statistics with tractable null distributions and the dynamic graphical tools which allow the exploration of multiple summaries in both spatial and temporal/interscan aspects, with the ability to quickly jump to spatiotemporal detail. We apply our methods to a fMRI data set, demonstrating their ability to localize subtle artifacts and to discover systematic experimental variation not captured by the model.

Variational Bayesian inference for fMRI time series

We describe a Bayesian estimation and inference procedure for fMRI time series based on the use of General Linear Models with Autoregressive (AR) error processes. We make use of the Variational Bayesian (VB) framework which approximates the true posterior density with a factorised density. The fidelity of this approximation is verified via Gibbs sampling. The VB approach provides a natural extension to previous Bayesian analyses which have used Empirical Bayes. VB has the advantage of taking into account the variability of hyperparameter estimates with little additional computational effort. Further, VB allows for automatic selection of the order of the AR process. Results are shown on simulated data and on data from an event-related fMRI experiment.

Arterial spin labeling perfusion fMRI with very low task frequency

Functional magnetic resonance imaging (fMRI) has become the most widely used modality for visualizing regional brain activation in response to sensorimotor or cognitive tasks. While the majority of fMRI studies have used blood oxygenation level-dependent (BOLD) contrast as a marker for neural activation, baseline drift effects result in poor sensitivity for detecting slow variations in neural activity. By contrast, drift effects are minimized in arterial spin labeling (ASL) perfusion contrast, primarily as a result of successive pairwise subtraction between images acquired with and without labeling. Recent data suggest that ASL contrast shows stable noise characteristics over the entire frequency spectrum, which makes it suitable for studying low-frequency events in brain function. The present study investigates the relative sensitivities of ASL and BOLD contrast in detecting changes in motor cortex activation over a spectrum of frequencies of experimental design, where the alternating period between the resting state and activation is varied from 30 s up to 24 hr. The results demonstrate that 1) ASL contrast can detect differences in motor cortex activation over periods of minutes, hours, and even days; 2) the functional sensitivity of ASL contrast becomes superior to that of BOLD contrast when the alternating period between the resting state and activation is greater than a few minutes; and 3) task activation measured by ASL tends to have less intersubject variability than BOLD contrast. The improved sensitivity of the ASL contrast for low task frequency and longitudinal studies, along with its superior power in group analysis, is expected to extend the range of experimental designs that can be studied using fMRI.

Optimal spline smoothing of fMRI time series by generalized cross-validation

Linear parametric regression models of fMRI time series have correlated residuals. One approach to address this problem is to condition the autocorrelation structure by temporal smoothing. Smoothing splines with the degree of smoothing selected by generalized cross-validation (GCV-spline) provide a method to find an optimal smoother for an fMRI time series. The purpose of this study was to determine if GCV-spline of fMRI time series yields unbiased variance estimates of linear regression model parameters. GCV-spline was evaluated with a real fMRI data set and bias of the variance estimator was computed for simulated time series with autocorrelation structures derived from fMRI data. This study only considered fMRI experimental designs of boxcar type. The results from the real data suggest that GCV-spline determines appropriate amounts of smoothing. The simulations show that the variance estimates are, on average, unbiased. The unbiased variance estimates come at some cost to signal detection efficiency. This study demonstrates that GCV-spline is an appropriate method for smoothing fMRI time series.

Quantitative fMRI assessment of the differences in lateralization of language-related brain activation in patients with temporal lobe epilepsy

Defining language lateralization is important to minimize morbidity in patients treated surgically for temporal lobe epilepsy (TLE). Functional magnetic resonance imaging (fMRI) offers a promising, noninvasive, alternative strategy to the Wada test. Here we have used fMRI to study healthy controls and patients with TLE in order to (i) define language-related activation patterns and their reproducibility; (ii) compare lateralization determined by fMRI with those from of the Wada test; and (iii) contrast different methods of assessing fMRI lateralization. Twelve healthy right-handed controls and 19 right-handed preoperative patients with TLE (12 left- and seven right-TLE) were studied at 3T using fMRI and a verbal fluency paradigm. A Wada test also was performed on each of the patients. Greater activation was found in several areas in the right hemisphere for the left-TLE group relative to controls or right-TLE patients. Relative hemispheric activations calculated based on either the extent or the mean signal change gave consistent results showing a more bihemispheric language representation in the left-TLE patients. There was good agreement between the Wada and fMRI results, although the latter were more sensitive to involvement of the nondominant right hemisphere. The reproducibility of the fMRI values was lowest for the more bihemispherically represented left-TLE patients. Overall, our results further demonstrate that noninvasive fMRI measures of language-related lateralization may provide a practical and reliable alternative to invasive testing for presurgical language lateralization in patients with TLE. The high proportion (33%) of left-TLE patients showing bilateral or right hemispheric language-related lateralization suggests that there is considerable plasticity of language representation in the brains of patients with intractable TLE.

Reading speech from still and moving faces: the neural substrates of visible speech

Speech is perceived both by ear and by eye. Unlike heard speech, some seen speech gestures can be captured in stilled image sequences. Previous studies have shown that in hearing people, natural time-varying silent seen speech can access the auditory cortex (left superior temporal regions). Using functional magnetic resonance imaging (fMRI), the present study explored the extent to which this circuitry was activated when seen speech was deprived of its time-varying characteristics. In the scanner, hearing participants were instructed to look for a prespecified visible speech target sequence ("voo" or "ahv") among other monosyllables. In one condition, the image sequence comprised a series of stilled key frames showing apical gestures (e.g., separate frames for "v" and "oo" [from the target] or "ee" and "m" [i.e., from nontarget syllables]). In the other condition, natural speech movement of the same overall segment duration was seen. In contrast to a baseline condition in which the letter "V" was superimposed on a resting face, stilled speech face images generated activation in posterior cortical regions associated with the perception of biological movement, despite the lack of apparent movement in the speech image sequence. Activation was also detected in traditional speech-processing regions including the left inferior frontal (Broca's) area, left superior temporal sulcus (STS), and left supramarginal gyrus (the dorsal aspect of Wernicke's area). Stilled speech sequences also generated activation in the ventral premotor cortex and anterior inferior parietal sulcus bilaterally. Moving faces generated significantly greater cortical activation than stilled face sequences, and in similar regions. However, a number of differences between stilled and moving speech were also observed. In the visual cortex, stilled faces generated relatively more activation in primary visual regions (V1/V2), while visual movement areas (V5/MT+) were activated to a greater extent by moving faces. Cortical regions activated more by naturally moving speaking faces included the auditory cortex (Brodmann's Areas 41/42; lateral parts of Heschl's gyrus) and the left STS and inferior frontal gyrus. Seen speech with normal time-varying characteristics appears to have preferential access to "purely" auditory processing regions specialized for language, possibly via acquired dynamic audiovisual integration mechanisms in STS. When seen speech lacks natural time-varying characteristics, access to speech-processing systems in the left temporal lobe may be achieved predominantly via action-based speech representations, realized in the ventral premotor cortex.

On bias in the estimation of autocorrelations for fMRI voxel time-series analysis

For fMRI time-series analysis to be statistically valid, it is important to deal correctly with temporal autocorrelation in the noise. Most of the approaches in the literature adopt a two-stage approach in which the autocorrelation structure is estimated using the residuals of an initial model fit. This estimate is then used to "prewhiten" the data and the model before the model is refit to obtain final activation parameter estimates. An assumption implicit in this scheme is that the residuals from the initial model fit represent a realization of the "true" noise process. In general this assumption will not be correct as certain components of the noise will be removed by the model fit. In this paper we examine (i) the form of the bias induced by the initial model fit, (ii) methods of correcting for the bias, and (iii) the impact of bias correction on the model parameter estimates. We find that while bias correction does result in more accurate estimates of the correlation structure, this does not translate into improved estimates of the model parameters. In fact estimates of the model parameters and their standard errors are seen to be so accurate that we conclude that bias correction is unnecessary.

Correlation between motor improvements and altered fMRI activity after rehabilitative therapy

Motor rehabilitation therapy is commonly employed after strokes, but outcomes are variable and there is little specific information about the changes in brain activity that are associated with improved function. We performed serial functional MRI (fMRI) on a group of seven patients receiving a form of rehabilitation therapy after stroke in order to characterize functional changes in the brain that correlate with behavioural improvements. Patients were scanned while performing a hand flexion-extension movement twice before and twice after a two-week home-based therapy programme combining restraint of the unaffected limb with progressive exercises for the affected limb. As expected, the extent of improvement in hand function after therapy varied between patients. Therapy-related improvements in hand function correlated with increases in fMRI activity in the premotor cortex and secondary somatosensory cortex contralateral to the affected hand, and in superior posterior regions of the cerebellar hemispheres bilaterally (Crus I and lobule VI). fMRI offers a promising, objective approach for specifically identifying changes in brain activity potentially responsible for rehabilitation-mediated recovery of function after stroke. Our results suggest that activity changes in sensorimotor regions are associated with successful motor rehabilitation.

Sensitivity-encoded single-shot spiral imaging for reduced susceptibility artifacts in BOLD fMRI

Sensitivity encoding (SENSE) with iterative image reconstruction was used to shorten the readout duration in single-shot spiral imaging by a factor of 2. This enabled susceptibility-related blurring and signal loss artifacts to be reduced and spatial resolution to be improved. As a beneficial side effect, the gradient duty cycle was also reduced. The spiral SENSE technique was applied to functional MRI (fMRI) with blood oxygen level-dependent (BOLD) contrast and compared to a conventional spiral acquisition. Stimulation experiments were performed in seven volunteers using motor, visual, and taste paradigms. The signal-to-noise ratio (SNR) and signal-to-fluctuation-noise ratio (SFNR) of the SENSE acquisitions were reduced by 20% and 13%, respectively, with respect to the longer readout. The overall activation detected was comparable to that of the conventional spiral acquisition, even though difficulties in reproducing the stimulation response hampered the evaluation. In some cases, the application of SENSE enabled recovery of activation in regions affected by signal loss due to field inhomogeneity.

Effects of word form on brain processing of written Chinese

Both logographic characters and alphabetic pinyins can be used to write words in Chinese. Here we use fMRI to address the question of whether the written form affects brain processing of a word. Fifteen healthy, right-handed, native Chinese-reading volunteers participated in our study and were asked to read silently either Chinese characters (8 subjects) or pinyins (7 subjects). The stimulus presentation rate was varied for both tasks to allow us to identify brain regions with word-load-dependent activation. Rate effects (fast minus slow presentations) for Chinese character reading were observed in striate and extrastriate visual cortex, superior parietal lobule, left posterior middle temporal gyrus, bilateral inferior temporal gyri, and bilateral superior frontal gyri. Rate effects for pinyin reading were observed in bilateral fusiform, lingual, and middle occipital gyri, bilateral superior parietal lobule/precuneus, left inferior parietal lobule, bilateral inferior temporal gyrus, left middle temporal gyrus, and left superior temporal gyrus. These results demonstrate that common regions of the brain are involved in reading both Chinese characters and pinyins, activated apparently independently of the surface form of the word. There also appear to be brain regions in which activation is dependent on word form. However, it is unlikely that these are entirely specific for a given word form; their activation more likely reflects relative functional specializations within broader networks for processing written language.

The role of ipsilateral premotor cortex in hand movement after stroke

Movement of an affected hand after stroke is associated with increased activation of ipsilateral motor cortical areas, suggesting that these motor areas in the undamaged hemisphere may adaptively compensate for damaged or disconnected regions. However, this adaptive compensation has not yet been demonstrated directly. Here we used transcranial magnetic stimulation (TMS) to interfere transiently with processing in the ipsilateral primary motor or dorsal premotor cortex (PMd) during finger movements. TMS had a greater effect on patients than controls in a manner that depended on the site, hemisphere, and time of stimulation. In patients with right hemiparesis (but not in healthy controls), TMS applied to PMd early (100 ms) after the cue to move slowed simple reaction-time finger movements by 12% compared with controls. The relative slowing of movements with ipsilateral PMd stimulation in patients correlated with the degree of motor impairment, suggesting that functional recruitment of ipsilateral motor areas was greatest in the more impaired patients. We also used functional magnetic resonance imaging to monitor brain activity in these subjects as they performed the same movements. Slowing of reaction time after premotor cortex TMS in the patients correlated inversely with the relative hemispheric lateralization of functional magnetic resonance imaging activation in PMd. This inverse correlation suggests that the increased activation in ipsilateral cortical motor areas during movements of a paretic hand, shown in this and previous functional imaging studies, represents a functionally relevant, adaptive response to the associated brain injury.