Elsevier

NeuroImage

Volume 18, Issue 4, April 2003, Pages 813-826
NeuroImage

Regular article
Comparison of fMRI activation at 3 and 1.5 T during perceptual, cognitive, and affective processing

https://doi.org/10.1016/S1053-8119(03)00002-8Get rights and content

Abstract

Previous studies comparing fMRI data acquired at 1.5 T and higher field strengths have focused on examining signal increases in the visual and motor cortices. No information is, however, available on the relative gain, or the comparability of data, obtained at higher field strengths for other brain regions such as the prefrontal and other association cortices. In the present study, we investigated fMRI activation at 1.5 and 3 T during visual perception, visuospatial working memory, and affect-processing tasks. A 23% increase in striate and extrastriate activation volume was observed at 3 T compared with that for 1.5 T during the visual perception task. During the working memory task significant increases in activation volume were observed in frontal and parietal association cortices as well as subcortical structures, including the caudate, globus pallidus, putamen, and thalamus. Increases in working memory-related activation volume of 82, 73, 83, and 36% were observed in the left frontal, right frontal, left parietal, and right parietal lobes, respectively, for 3 T compared with 1.5 T. These increases were characterized by increased activation at 3 T in several prefrontal and parietal cortex regions that showed activation at 1.5 T. More importantly, at 3 T, activation was detected in several regions, such as the ventral aspects of the inferior frontal gyrus, orbitofrontal gyrus, and lingual gyrus, which did not show significant activation at 1.5 T. No difference in height or extent of activation was detected between the two scanners in the amygdala during affect processing. Signal dropout in the amygdala from susceptibility artifact was greater at 3 T, with a 12% dropout at 3 T compared with a 9% dropout at 1.5 T. The spatial smoothness of T2* images was greater at 3 T by less than 1 mm, suggesting that the greater extent of activation at 3 T beyond these spatial scales was not due primarily to increased intrinsic spatial correlations at 3 T. Rather, the increase in percentage of voxels activated reflects increased sensitivity for detection of brain activation at higher field strength. In summary, our findings suggest that functional imaging of prefrontal and other association cortices can benefit significantly from higher magnetic field strength.

Introduction

Functional magnetic resonance imaging (fMRI) is now widely used to study brain function and dysfunction. Due to their wide availability, 1.5 T systems are currently used in a majority of fMRI studies. Increasingly, however, fMRI studies are being conducted at field strengths of 3 T or higher. Recent research has demonstrated that exquisite spatial resolution can be obtained with functional imaging at higher field strength in a manner that was not possible at 1.5 T. For example, researchers have demonstrated high-resolution maps of ocular dominance columns Cheng et al 2001, Menon et al 1997 and retinotopy within the lateral geniculate nucleus (Ugurbil et al., 1999). Many of the early studies comparing brain activation at 1.5 T with higher fields focused on examining signal increases in the visual Gati et al 1997, Turner et al 1993 and motor (Yang et al., 1999) cortices. No information, however, is available on the relative gain, or the comparability of data, obtained at higher field strengths for regions other than the visual and motor cortices. In this study we examine qualitative and quantitative differences in the patterns of activation observed at 3 and 1.5 T during sensory, cognitive, and affect-processing tasks.

For fMRI, a major advantage of using higher fields derives from the fact that the rate of transverse relaxation, R2* = 1/T2*, scales as the square of the external magnetic field for small blood vessels and capillaries, whereas the change is linear for large blood vessels (Gati et al., 1997). Since the blood oxygen level-dependent (BOLD) contrast originates from the intravoxel magnetic field inhomogeneity induced by paramagnetic deoxyhemoglobin, higher fields should result in improved sensitivity related primarily to BOLD changes in capillary beds in response to neural activity Kruger et al 2001, Ugurbil et al 1999. Thus, theoretical considerations indicate that higher fields should result in improved sensitivity and spatial specificity for detection of task-related brain activation. Furthermore, since BOLD signal changes at 1.5 T are rather small (on the order of a few percent), it is thought that higher field strengths of 3 T or more should significantly enhance the ability to reliably detect signals of interest.

Four published studies to date have directly compared changes in task-related fMRI activation at 1.5 T and higher field strengths Gati et al 1997, Kruger et al 2001, Turner et al 1993, Yang et al 1999. Two of these studies focused on the visual cortex, one on the motor cortex, and a fourth on the visual and motor cortices. Turner et al. (1993) compared activation in the visual cortex at 1.5 and 4 T during photic stimulation and reported an increase of approximately 300% in an eight-voxel (5 mm2) region at 4 T. More recent studies have taken scanner noise and physiological fluctuations into account and have reported more modest increase in signal. Gati et al. (1997) found a 70% increase in average percentage signal change in cortical gray matter activation during photic stimulation. Yang et al. (1999) examined activation in the motor cortex during a finger-tapping task and found that at 4 T, compared with 1.5 T, there was a 70% increase in the number of voxels activated and a 20% higher average t score for the activated voxels. Kruger et al. (2001) examined motor and visual cortex activation at 3 T, compared with 1.5 T. They found a 44% increase in the number of voxels activated in the primary motor cortex and 36% more voxels were activated in the visual cortex compared with activation during a checkerboard reversal task (Kruger et al., 2001). Taken together, these studies suggest that higher field strength provides an advantage for functional imaging of primary sensory and motor cortices. The extent to which signal gains of the type found in primary sensory and motor cortices might extend to association cortices, e.g., the prefrontal and parietal cortex, is not known. This is an important question to examine since fMRI is now extensively used to study cognitive function and dysfunction.

A second major issue that has not been addressed in studies to date is the comparative extent of susceptibility artifacts at higher field strengths. Susceptibility artifacts result from abrupt changes in magnetic susceptibility that occur across tissue interfaces such as the border between air-filled sinuses and brain parenchyma or between bone and brain parenchyma (Ojemann et al., 1997). Brain regions closest to such borders are especially prone to BOLD signal loss due to this artifact. Brain regions that are affected by these artifacts include the orbitofrontal cortex, hippocampus, amygdala, and anterior, inferolateral temporal pole Devlin et al 2000, Lipschutz et al 2001, Ojemann et al 1997. No study to date has examined differences in activation at different magnetic field strengths using tasks that are known to involve brain regions that are prone to susceptibility artifact.

In this study, we examined differences in activation at 1.5 and 3 T using three different tasks that are known to activate distinct brain regions. We used a visual perception task known to reliably activate striate and extrastriate cortices DeYoe et al 1994, Watson et al 1993, a working memory task known to reliably activate the prefrontal and parietal association cortices Smith and Jonides 1997, Smith and Jonides 1998, and an affect-processing task known to reliably activate the amygdala Breiter et al 1996, Yang et al 2002. A random effects model was used to examine differences in activation obtained at the two field strengths. The results of this analysis were used to isolate the precise voxels that showed statistically significant differences in activation between the two field strengths. For each task, we examined the increased sensitivity for detection of activation at 3 T; by “activation” we mean voxels for which the z scores exceed a specified threshold. Finally, we also examined the effect of susceptibility artifacts on amygdala activation at the two field strengths.

Section snippets

Subjects

Fourteen right-handed subjects (aged 17–25, mean age = 21.21 years, SD = 2.16; 7 male) participated in the study after giving written informed consent. fMRI data were acquired for each subject as he or she performed three different tasks involving visual perception, working memory, and affect-processing in scanners with 1.5- and 3-T magnetic field strengths. Both task order and scanner order were randomized and counterbalanced in this within-subjects design. Each subject was presented with the

Activation

Significant activation was observed in the striate, extrastriate, and posterior parietal cortex at both 3 and 1.5 T, with more extensive activation at 3 T (Fig. 1). A direct comparison of activations using random effects analysis revealed a greater number of activated voxels at 3 T in the striate cortex (VI) and extrastriate regions (V2 and V3) (Fig. 2). Significant between-scanner differences were detected in 896 voxels; 3811 voxels were activated at 1.5 T, implying an increase in activation

Discussion

During both the visual perception and visuospatial working memory tasks significantly greater cortical activation was observed at 3 T, compared with 1.5 T. Increased activation at 3 T was most dramatic in frontal and parietal cortices during the working memory task. Compared with an increase in visual cortex activation of about 23% during the visual perception task, increases in activated voxels of 78% in the prefrontal cortex and 59% in the parietal cortex were observed during the working

Acknowledgements

This research was supported by NIH Grants HD40761, MH62430, RR09784, and MH19908, and grants from the Norris Foundation, and the Lucas Imaging Center.

References (52)

  • W.R. Kates et al.

    Reliability and validity of MRI measurement of the amygdala and hippocampus in children with fragile X syndrome

    Psychiatry Resv75

    (1997)
  • S.J. Kiebel et al.

    Robust smoothness estimation in statistical parametric maps using standardized residuals from the general linear model

    NeuroImage

    (1999)
  • B. Lipschutz et al.

    Assessing study-specific regional variations in fMRI signal

    NeuroImage

    (2001)
  • V. Menon et al.

    Functional brain activation during cognition is related to FMR1 gene expression

    Brain Res.

    (2000)
  • K.D. Merboldt et al.

    Functional MRI of the human amygdala?

    NeuroImage

    (2001)
  • E.K. Miller

    The prefrontal cortexno simple matter

    NeuroImage

    (2000)
  • L.E. Nystrom et al.

    Working memory for letters, shapes, and locationsfMRI evidence against stimulus-based regional organization in human prefrontal cortex

    NeuroImage

    (2000)
  • J.G. Ojemann et al.

    Anatomic localization and quantitative analysis of gradient refocused echo-planar fMRI susceptibility artifacts

    NeuroImage

    (1997)
  • J.B. Poline et al.

    Combining spatial extent and peak intensity to test for activations in functional imaging

    NeuroImage

    (1997)
  • B.R. Postle et al.

    An fMRI investigation of cortical contributions to spatial and nonspatial visual working memory

    NeuroImage

    (2000)
  • F. Schneider et al.

    Functional MRI reveals left amygdala activation during emotion

    Psychiatry Res.

    (1997)
  • E.E. Smith et al.

    Working memorya view from neuroimaging

    Cogn. Psychol.

    (1997)
  • N. Tzourio-Mazoyer et al.

    Automated anatomical labeling of activations in SPM using a macroscopic anatomical parcellation of the MNI MRI single-subject brain

    NeuroImage

    (2002)
  • Y. Yang et al.

    Comparison of 3D BOLD functional MRI with spiral acquisition at 1.5 and 4.0 T

    NeuroImage

    (1999)
  • M.G. Baxter et al.

    The amygdala and reward

    Nat. Rev. Neurosci.

    (2002)
  • M.V. Chafee et al.

    Matching patterns of activity in primate prefrontal area 8a and parietal area 7ip neurons during a spatial working memory task

    J. Neurophysiol.

    (1998)
  • Cited by (111)

    • Design science and neuroscience: A systematic review of the emergent field of Design Neurocognition

      2023, Design Studies
      Citation Excerpt :

      The type of brain imaging tool and the tool's technical specifications, such as its data resolution, varied across studies. For fMRI, better spatial resolution can be obtained with functional imaging at higher field strengths of 3T compared to 1.5T (Krasnow et al., 2003). The majority of the Design Science studies were executed in a 3T MRI scanner (Fu et al., 2019; Goucher-Lambert et al., 2018a, 2019; Goucher-Lambert & McComb, 2019; Hay et al., 2019), while two studies were conducted with a 1.5T instrument (Alexiou et al., 2009, 2011).

    View all citing articles on Scopus
    View full text