The Dynamics of Interhemispheric Compensatory Processes in Mental Imagery

See allHide authors and affiliations

Science  29 Apr 2005:
Vol. 308, Issue 5722, pp. 702-704
DOI: 10.1126/science.1107784


The capacity to generate and analyze mental visual images is essential for many cognitive abilities. We combined triple-pulse transcranial magnetic stimulation (tpTMS) and repetitive TMS (rTMS) to determine which distinct aspect of mental imagery is carried out by the left and right parietal lobe and to reveal interhemispheric compensatory interactions. The left parietal lobe was predominant in generating mental images, whereas the right parietal lobe was specialized in the spatial comparison of the imagined content. Furthermore, in case of an rTMS-induced left parietal lesion, the right parietal cortex could immediately compensate such a left parietal disruption by taking over the specific function of the left hemisphere.

Mental imagery refers to the experience of a perception in the absence of a corresponding physical stimulus. In our everyday life, mental imagery represents a crucial element of numerous cognitive abilities, such as object recognition, reasoning, language comprehension, and memory. Because of its importance, the exact processes associated with imagery have long occupied cognitive psychologists and been a matter of debate and controversy (1).

Mental imagery is accompanied by the activation of frontoparietal networks (25), but the exact brain areas engaged in imagery depend on the specific features of the imagery task (6). When spatial comparisons between imagined objects are required, most functional imaging studies show bilateral parietal activation in homologous intraparietal sulcus areas of the left and right hemispheres (3). However, neuropsychological studies on patients with focal brain lesions generally support a dominant role of the left hemisphere in imagery [(7), but see (8)].

Time-resolved functional magnetic resonance imaging (fMRI) has been used to address this apparent contradiction between functional imaging studies and findings in focal brain injury patients (4). An earlier cluster of activation in both parietal cortices (with left predominance) can be separated from a late cluster confined to the right parietal cortex (Fig. 1). These results support the involvement of both parietal lobes in mental imagery but suggest that each parietal lobe has a distinct functional role at different moments in time. The sequential parietal activation might represent a transition from an earlier more distributed processing stage of image generation to a later right-hemispheric lateralized stage of spatial analysis of the images (4). In a combined fMRI and rTMS study, only rTMS to the right parietal lobe led to an impairment of spatial imagery performance during and immediately after rTMS (5). Hence, only the right parietal activity seems to be functionally relevant for spatial imagery, whereas the rTMS-induced disruption of the left parietal activity has no measurable impact on the ability to perform the imagery task.

Fig. 1.

Brain areas activated during spatial imagery. Temporal sequence of cognitive processes associated with the mental clock task (A) and the activated brain areas as revealed by time-resolved fMRI based on a different subject sample from a previous study. (B) View from the occipital pole of both inflated hemispheres. The color code from blue, to green, to yellow, to red indicates the sequence of activated brain areas associated with the four required cognitive processes. The results distinguish an earlier, bilateral parietal activation (green) from a later, strictly right parietal activation (yellow).

However, the behavioral consequences of focal disruption of brain activity (by injury or transiently by rTMS) show the ability of the rest of the brain to cope with the insult. Therefore, the apparently unaffected imagery performance after left parietal disruption could be due to compensatory processes within a distributed neural network that cannot be sustained after right parietal disruption. We hypothesized that the right parietal cortex is critical for the spatial comparison of mental images but is also able to compensate a disruption of left parietal activity during image generation. This assumes that in case of a left parietal lesion, the right parietal cortex will subserve both functions, image generation and spatial comparison.

We introduced a previously untested TMS procedure, which combines the respective advantages of creating a temporary virtual lesion by rTMS with the precise chronometric study offered by event-related tpTMS (9). We used tpTMS to chart the time points at which right parietal activity is critical for spatial imagery, depending on whether the homologous left parietal cortex had or had not been suppressed by preceding rTMS (Fig. 2).

Fig. 2.

TMS protocol. tpTMS (red coil) identifies the time points at which right parietal activity is critical for spatial imagery (A) after sham TMS has been applied to the left parietal cortex (gray coil) and (B) after inducing a virtuallesiontotheleft parietal cortex by preceding rTMS (blue coil). PPC, posterior parietal cortex.

Subjects were asked to imagine two analog clock faces based on acoustically presented times (e.g., two o'clock and five o'clock) and to judge at which of the two times the clock hands form the greater angle (4, 5, 10). This task requires the subjects to first generate visual mental images of two analog clocks and then mentally compare the angles formed by the clock hands (11). The different mental processes associated with the mental clock task include the encoding of the acoustic stimuli, the generation and maintenance of the mental images, the spatial comparison, and the decision and response by button press (Fig. 1). We hypothesized that tpTMS over the right parietal cortex would only impair performance when applied at a specific late time point during the task (representing the spatial comparison stage). Moreover, if the right parietal cortex can compensate a disruption of left parietal activity (representing image generation), a functional lesion of the left parietal cortex by means of preceding rTMS should result in an extension of this late critical time point of right parietal activity toward an earlier time point. This implies that in case of a left parietal lesion, the right parietal cortex is able to subserve image generation and spatial comparison (Fig. 2).

Every subject completed four right parietal tpTMS runs, two with and two without preceding left parietal rTMS (rTMS factor) (12). A within-subject repeated measures analysis of variance with the two-level rTMS factor and a four-level time factor of right parietal disruption revealed a significant interaction (F = 6.358; P = 0.05). More specifically, in the runs without preceding rTMS over the left parietal cortex (intact left parietal lobe), tpTMS over the right parietal cortex only significantly impaired the behavioral performance in the mental clock task when applied at a time interval of 4200 ms after trial onset (13), which fell within time bin 3 (spatial comparison; Student's t test = 4.099, df = 5, P < 0.05; Fig. 3; individual data available in fig. S1). This highly specific and rather late time point corresponds well to the highly lateralized right parietal activity cluster previously revealed by time-resolved fMRI (4). In contrast, in the runs with preceding rTMS (i.e., during the temporary disruption of left parietal cortical function), the temporal characteristics of the behavioral effects of tpTMS to right parietal cortex changed significantly and now not only included time bin 3 (spatial comparison; t = 4.511, df = 5, P < 0.05) but also several earlier time intervals (14) during time bin 2 (image generation; t = 6.217, df = 5, P < 0.05; Fig. 3; see also fig. S1 for individual data). These earlier time windows correspond well to the bilaterally distributed parietal activation cluster revealed by time-resolved fMRI (4). Analysis of the error rates revealed no significant differences between the different time points and/or conditions (fig. S2). In an additional control experiment, left parietal rTMS alone, without subsequent right parietal tpTMS, led to no significant impairments in the task performance of the mental clock task (t = 0.601; df = 5; P = 0.574).

Fig. 3.

Behavioral results. Mean percentage change of reaction time during tpTMS over the right parietal cortex at the different time intervals, independently for the two rTMS conditions: (A) without preceding left parietal rTMS and (B) with preceding left parietal rTMS.

Based on the temporal activation sequence from left to right parietal cortex during an imagery task that involves the generation as well as spatial comparison of mental images (4), we applied tpTMS to fractionate specialized processing components in the right parietal cortex. We were able to show that an rTMS-induced unilateral disruption can lead to a significant change in the critical functional time point, at which activations in the homologous cortical area of the contralateral hemisphere are crucial for the execution of a specific function. Whereas the mental process of spatial comparison is highly lateralized to the right hemisphere, image generation shows only a weak left-hemispheric lateralization with a more bilateral distribution. By charting and comparing the time courses of critical right parietal activity with and without preceding left parietal disruption, our data suggest that the right hemisphere is able to compensate for (virtual) lesions of the left hemisphere by taking over this specific mental process. Whereas an rTMS-induced left parietal disruption alone did not impair task performance, a subsequent right parietal tpTMS early in the task (during image generation) unmasked the behavioral deficit, most likely by blocking the right parietal compensation.

Discrepancies across studies concerning the hemispheric lateralization during mental imagery likely arise, because different aspects of imagery are carried out by different parts of a bihemispheric neural network. Our results are in accordance with neuropsychological models of spatial imagery, which propose that the generation of mental images relies primarily on structures in the posterior left hemisphere, whereas spatial operations on these images are subserved by the posterior right hemisphere (15). Moreover, our results reveal that in case of left parietal disruption, the generation of mental images might be carried out by the right hemisphere as part of an asymmetric interhemispheric compensatory mechanism. An isolated deficit of the ability to generate inner visual images after unilateral lesion is clinically hardly ever reported (8), which could be explained on the basis of the compensatory processes revealed in our study. This also parallels evidence from spatial hemineglect, which mostly occurs after right hemispheric lesions. This phenomenon has been explained on the basis of an asymmetrical distribution of spatial attention (16), in which a right parietal lesion would lead to hemineglect for the left visual field, whereas a respective left parietal lesion could be compensated for by the right hemisphere. However, also on the basis of such an asymmetrical distribution of interhemispheric attention, it cannot be assumed that the two hemispheres simply process information independently. Human cognition rather includes highly complex processes of interhemispheric competition, cooperation, and suppression. It remains speculative whether the revealed functional compensation represents a real cortical reorganization or whether the right hemisphere compensates the behavioral consequence due to the bilateral distribution of image generation; our results are in line with the latter. In this respect, the mental process of image generation might indeed be carried out by both hemispheres. However, under normal physiological circumstances, the functional necessity of the right hemisphere becomes redundant because of the left hemispheric dominance. Hence, the revealed compensatory processes might be based on a disruption of the suppressive influence of one hemisphere within an interhemispheric competition. The release of this suppressive influence by rTMS would then restore the functional relevance of the contralateral hemisphere, thereby compensating or even enhancing the cognitive functions it subserves (17).

Supporting Online Material

Materials and Methods

Figs. S1 and S2

References and Notes

View Abstract

Stay Connected to Science

Editor's Blog

Navigate This Article