The compensatory dynamic of inter-hemispheric interactions in visuospatial attention revealed using rTMS and fMRI

Abstract

A balance of mutual tonic inhibition between bi-hemispheric posterior parietal cortices is believed to play an important role in bilateral visual attention. However, experimental support for this notion has been mainly drawn from clinical models of unilateral damage. We have previously shown that low-frequency repetitive TMS (rTMS) over the intraparietal sulcus (IPS) generates a contralateral attentional deficit in bilateral visual tracking. Here, we used functional magnetic resonance imaging (fMRI) to study whether rTMS temporarily disrupts the inter-hemispheric balance between bilateral IPS in visual attention. Following application of 1 Hz rTMS over the left IPS, subjects performed a bilateral visual tracking task while their brain activity was recorded using fMRI. Behaviorally, tracking accuracy was reduced immediately following rTMS. Areas ventro-lateral to left IPS, including inferior parietal lobule (IPL), lateral IPS (LIPS), and middle occipital gyrus (MoG), showed decreased activity following rTMS, while dorsomedial areas, such as Superior Parietal Lobule (SPL), Superior occipital gyrus (SoG), and lingual gyrus, as well as middle temporal areas (MT+), showed higher activity. The brain activity of the homologues of these regions in the un-stimulated, right hemisphere was reversed. Interestingly, the evolution of network-wide activation related to attentional behavior following rTMS showed that activation of most occipital synergists adaptively compensated for contralateral and ipsilateral decrement after rTMS, while activation of parietal synergists, and SoG remained competing. This pattern of ipsilateral and contralateral activations empirically supports the hypothesized loss of inter-hemispheric balance that underlies clinical manifestation of visual attentional extinction.

Keywords: TMS; fMRI; inter-hemispheric interaction; visual extinction; visuospatial attention.

Figure 1

Behavioral Task - Design and Results. (A) Visual tracking task. Stimuli were high-contrast pairs of pinwheels displayed on either side of a central fixation cross. (B) At the start, the targets (a randomly selected spoke on each pinwheel) were cued briefly. Following the cues disappearance, the pinwheels rotated at a fixed rate, determined by individual subjects' threshold for 85% correct performance, while subjects tracked the targets. After 3 s, both pinwheels stopped and were aligned so that all probes on the target pinwheel appeared as a cross. Subjects responded using a four-alternative forced-choice paradigm (“up,” “down,” “left,” or “right” keys) to report which of the probes represented the originally-cued target. (C) Results of tracking accuracy: contra, contralateral (right hemifield); ipsi, ipsilateral (left hemifield); y-axis, tracking accuracy following rTMS as a proportion of that following sham; x-axis, experimental runs (1 through 3). Values below 1 represent a decrement and above 1 show improvement following rTMS vs. sham.

Figure 2

fMRI-Regions of interest analysis (ROIs) identified from Multi-Subject Fixed Effects GLM analysis. FMRI Activation Maps displaying results of Multi-subject fixed effects GLM with comparisons between pooled runs (1, 2, and 3) of rTMS vs. sham. Locus of rTMS targeting is shown as filled purple circle over the left IPS. Red-to-yellow: activation following rTMS > sham; blue-to-green: activation following rTMS < sham. Abbreviations are expanded in Table 1.

Figure 3

Analysis of Homologous ROIs. Quantitative comparison of t-values of intensities of homologous ROIs. For homologous ROIs of comparable size (number of voxels noted in parentheses), t-values of intensity on left (targeted hemisphere) are compared with those of their homologues on right. (A) Comparison of homologous voxels pairs in parietal lobes. (B) Comparison of homologous voxels pairs in occipital and temporal regions. While IPS, IPL and MoG show lower intensities in targeted hemisphere, SPL, SoG, Lingual and MT+ show higher intensities in targeted than right hemisphere. These results confirm findings in Figure 2.

Figure 4

Activation-accuracy relationships for IPS and Parietal regions. Figure shows relationship between accuracy in contralateral and ipsilateral fields in runs 1 and 2 and volumes of activation of parietal ROIs. Note that accuracy in each hemifield is represented as accuracy following rTMS normalized to that following sham. Similarly, volume of activation of an ROI or its intensity is computed as that following rTMS vs. that following sham. Pearson's r values are listed in Table 2. Relationships are denoted between (A) contralateral (Right Field) accuracy and activation of Right IPS in Run 1, (B) ipsilateral (Left Field) accuracy and activation of Right IPS in Run 1, (C) contralateral accuracy and activation of Left IPS in Run 2, (D) ipsilateral accuracy and activation of Right IPS in Run 2, and (E) ipsilateral accuracy and activation of Left IPL in Run 2.

Figure 5

Activation-accuracy relationships for other ROIs. Figure shows relationship between accuracy in contralateral and ipsilateral fields in run 2 and volumes of activation of occipito-temporal ROIs. Note that accuracy in each hemifield is represented as accuracy following rTMS normalized to that following sham. Similarly, volume of activation of an ROI or its intensity is computed as that following rTMS vs. that following sham. Pearson's r values are listed in Table 2. Relationships are denoted between (A) contralateral (Right Field) accuracy and activation of Right Lingual in Run 2, (B) ipsilateral (Left Field) accuracy and activation of Right Lingual in Run 2, (C) contralateral accuracy and activation of Left MoG in Run 2, (D) ipsilateral accuracy and activation of Left Lingual in Run 2, (E) contralateral accuracy and activation of Right SoG in Run 2, and (F) ipsilateral accuracy and activation of Right MT in Run 2.