Neocortical system abnormalities in autism: an fMRI study of spatial working memory

OBJECTIVE

To test the hypothesis that deficits in spatial working memory in autism are due to abnormalities in prefrontal circuitry.

METHODS

Functional MRI (fMRI) at 3 T was performed in 11 rigorously diagnosed non-mentally retarded autistic and six healthy volunteers while they performed an oculomotor spatial working memory task and a visually guided saccade task.

RESULTS

Autistic subjects demonstrated significantly less task-related activation in dorsolateral prefrontal cortex (Brodmann area [BA] 9/46) and posterior cingulate cortex (BA 23) in comparison with healthy subjects during a spatial working memory task. In contrast, activation of autistic individuals was not reduced in other regions comprising the neural circuitry for spatial working memory including the cortical eye fields, anterior cingulate cortex, insula, basal ganglia, thalamus, and lateral cerebellum. Autistic subjects also did not demonstrate reduced activation in any brain regions while performing visually guided saccades.

CONCLUSION

Impairments in executive cognitive processes in autism may be subserved by abnormalities in neocortical circuitry as evidenced by decreased activation in prefrontal and posterior cingulate circuitry during a spatial working memory task.

Visual analysis of variance: a tool for quantitative assessment of fMRI data processing and analysis

Analysis of variance (ANOVA) is widely used for the study of experimental data. Here, the reach of this tool is extended to cover the preprocessing of functional magnetic resonance imaging (fMRI) data. This technique, termed visual ANOVA (VANOVA), provides both numerical and pictorial information to aid the user in understanding the effects of various parts of the data analysis. Unlike a formal ANOVA, this method does not depend on the mathematics of orthogonal projections or strictly additive decompositions. An illustrative example is presented and the application of the method to a large number of fMRI experiments is discussed.

Maturation of widely distributed brain function subserves cognitive development

Cognitive and brain maturational changes continue throughout late childhood and adolescence. During this time, increasing cognitive control over behavior enhances the voluntary suppression of reflexive/impulsive response tendencies. Recently, with the advent of functional MRI, it has become possible to characterize changes in brain activity during cognitive development. In order to investigate the cognitive and brain maturation subserving the ability to voluntarily suppress context-inappropriate behavior, we tested 8-30 year olds in an oculomotor response-suppression task. Behavioral results indicated that adult-like ability to inhibit prepotent responses matured gradually through childhood and adolescence. Functional MRI results indicated that brain activation in frontal, parietal, striatal, and thalamic regions increased progressively from childhood to adulthood. Prefrontal cortex was more active in adolescents than in children or adults; adults demonstrated greater activation in the lateral cerebellum than younger subjects. These results suggest that efficient top-down modulation of reflexive acts may not be fully developed until adulthood and provide evidence that maturation of function across widely distributed brain regions lays the groundwork for enhanced voluntary control of behavior during cognitive development.

Time course of fMRI-activation in language and spatial networks during sentence comprehension

Functional neuroimaging previously has been considered to provide inadequate temporal resolution to study changes of brain states as a function of cognitive computations; however, we have obtained evidence of differential amounts of brain activity related to high-level cognition (sentence processing) within 1.5 s of stimulus onset. The study used an event-related paradigm with high-speed echoplanar functional magnetic resonance imaging (fMRI) to trace the time course of the brain activation in the temporal and parietal regions as participants comprehended single sentences describing a spatial configuration. Within the first set of images, on average 1 s from when the participant begins to read a sentence, there was significant activation in a key cortical area involved in language comprehension (the left posterior temporal gyrus) and visuospatial processing (the left and right parietal regions). In all three areas, the amount of activation during sentence comprehension was higher for negative sentences than for their affirmative counterparts, which are linguistically less complex. The effect of negation indicates that the activation in these areas is modulated by the difficulty of the linguistic processing. These results suggest a relatively rapid coactivation in both linguistic and spatial cortical regions to support the integration of information from multiple processing streams.

Evaluation of respiratory artifact correction techniques in multishot spiral functional MRI using receiver operator characteristic analyses

Navigator corrections and low-spatial frequency (LSF) oversampling are investigated as methods for reducing respiration-related effects in multishot functional MRI. Both techniques take advantage of the smoothly varying or nearly constant phase variations linked to the respiration cycle. These techniques were tested in functional MRI studies with spiral k-space acquisitions. Receiver operator characteristic (ROC) analyses and the temporal variance averaged across the brain were used to evaluate their effectiveness. Both methods were found to increase the area under the ROC curve and to reduce the standard deviation, with the LSF oversampling method being more effective.

Estimating test-retest reliability in functional MR imaging. I: Statistical methodology

A common problem in the analysis of functional magnetic resonance imaging (fMRI) data is quantifying the statistical reliability of an estimated activation map. While visual comparison of the classified active regions across replications of an experiment can sometimes by informative, it is typically difficult to draw firm conclusions by inspection; noise and complex patterns in the estimated map make it easy to be misled. Here, several statistical models, of increasing complexity, are developed, under which "test-retest" reliability can be meaningfully defined and quantified. The method yields global measures of reliability that apply uniformly to a specified set of brain voxels. The estimates of these reliability measures and their associated uncertainties under these models can be used to compare statistical methods, to set thresholds for detecting activation, and to optimize the number of images that need to be acquired during an experiment.

Estimating test-retest reliability in functional MR imaging. II: Application to motor and cognitive activation studies

Functional magnetic resonance imaging (fMRI) using blood oxygenation contrast has rapidly spread into many application areas. In this paper, a new statistical model is used to evaluate the reliability of fMRI activation in a finger opposition motor paradigm for both within-session and between-session data and in a working memory paradigm for between-session data. A slice prescription procedure for between-session reproducibility is introduced. Estimates are made for the probabilities of correctly and falsely classifying voxels as active or inactive and receiver operator characteristic curves are generated. In the motor paradigm, estimated between-session reliability was found to be somewhat reduced relative to within-session reliability; however, this includes additional sources of variation and may not reflect intrinsically lower reliability. After matching false-positive classification probabilities, between-session reliability was found to be nearly identical for both motor and cognitive activation paradigms.

A spectral approach to analyzing slice selection in planar imaging: optimization for through-plane interpolation

Interpolation between slices is necessary whenever reslicing a volume of data into a different coordinate frame. This may be done to view the data from different perspectives, to align data from different sessions, or to remove the effects of head movement in functional imaging studies. In this paper, issues surrounding slice-selection in two-dimensional imaging are examined in the context of through-plane interpolation and a spectral framework is introduced to describe the sources of error when interpolating between slices. This framework suggests that there is a trade-off between precision in localization, which requires high spatial frequencies, and interpolation, which requires a narrow spectrum of spatial frequencies. An analysis of the sources of error has lead to several approaches to reducing interpolation error including elimination of interslice gaps or making slices overlap, use of a slice profile with a narrower spatial frequency bandwidth such as a Gaussian profile, and use of high-order interpolation. The simulation and experimental data demonstrate significant reductions in interpolation error for these approaches.

Improved image registration by using Fourier interpolation

This paper presents a technique for performing two-dimensional rigid-body image registration for functional magnetic resonance images (fMRI). The method provides accurate motion correction without local distortion. The approach is to perform the translation and rotation in the Fourier domain. For images sampled on a grid, such as in echo-planar imaging (EPI), one potential stumbling block to this approach is the computational burden of reconstruction, since the rotated image will no longer be on the Cartesian grid. A method of approximating rotations via local translations (shearing) is presented, which keeps the data on the Cartesian grid. This can provide quite accurate approximations with only a moderate amount of computation. A mean squared error (MSE) criterion is used for determining the registration parameters. This method is tested on several sets of simulated images and shown to have an accuracy ranging from 0.02 to 0.3 pixels for images with SNRs ranging from 100 to 10, respectively. These techniques have been tested on several sets of images. They are shown to work well on real subjects, for both echo-planar and spiral data acquisition schemes. The techniques are used in an activation study in which the subject moved his head during image collection. After use of this registration technique, the activation is easily detected.

Brain activation modulated by sentence comprehension

The comprehension of visually presented sentences produces brain activation that increases with the linguistic complexity of the sentence. The volume of neural tissue activated (number of voxels) during sentence comprehension was measured with echo-planar functional magnetic resonance imaging. The modulation of the volume of activation by sentence complexity was observed in a network of four areas: the classical left-hemisphere language areas (the left laterosuperior temporal cortex, or Wernicke's area, and the left inferior frontal gyrus, or Broca's area) and their homologous right-hemisphere areas, although the right areas had much smaller volumes of activation than did the left areas. These findings generally indicate that the amount of neural activity that a given cognitive process engenders is dependent on the computational demand that the task imposes.

Improved assessment of significant activation in functional magnetic resonance imaging (fMRI): use of a cluster-size threshold

The typical functional magnetic resonance (fMRI) study presents a formidable problem of multiple statistical comparisons (i.e., > 10,000 in a 128 x 128 image). To protect against false positives, investigators have typically relied on decreasing the per pixel false positive probability. This approach incurs an inevitable loss of power to detect statistically significant activity. An alternative approach, which relies on the assumption that areas of true neural activity will tend to stimulate signal changes over contiguous pixels, is presented. If one knows the probability distribution of such cluster sizes as a function of per pixel false positive probability, one can use cluster-size thresholds independently to reject false positives. Both Monte Carlo simulations and fMRI studies of human subjects have been used to verify that this approach can improve statistical power by as much as fivefold over techniques that rely solely on adjusting per pixel false positive probabilities.