Research reportRetinotopic organization of visual mental images as revealed by functional magnetic resonance imaging
Introduction
The role of the striate cortex (primary visual area, area V1) in visual mental imagery has long been debated. Many neuroimaging studies have revealed activation of early visual areas during visual mental imagery [1], [13], [16], [17], [18], [19] (for a review, see Ref. [15]), but many other studies have not revealed such activation (e.g., Refs. [3], [21]; for reviews, see also Refs. [11], [14], [20], [22], [24]). Nevertheless, the results from applying focal repetitive transcranial magnetic stimulation (rTMS) to the occipital cortex strongly suggest that the primary visual cortex is recruited at least in some forms of visual imagery. Specifically, performance was impaired in analogous imagery and perception tasks that required participants to compare gratings of different orientations and shapes after neural activity in visual occipital cortex was temporarily disrupted by focal TMS [18]. In addition, the existence of back-projections from high-order visual areas to low-level retinotopic areas provides neuroanatomical grounds for inferring that mental images may be produced via backward projections from visual memory areas down to V1 [8], [30].
If visual imagery and visual perception share anatomical substrates, they should be affected by the functional characteristics of those structures. In the striate cortex of primates, including humans, images from the external visual world are known to project onto the cortical ribbon along a retinotopic anatomical layout. Accordingly, the cortical ribbon along the visual cortex is functionally organized in a retinotopic manner [28], [31]. Mental images ‘projected’ onto the early visual cortex should thus exhibit some degree of retinotopy. Indeed, based on a study using positron emission tomography, a previous report showed that the spatial extent of activation in visual areas varied with the size of the mental images, coarsely following the retinotopic eccentricity along V1 [17]. However, no study has yet obtained retinotopic maps during mental imagery and compared those maps with retinotopic maps obtained during perception. Such a direct comparison would provide strong evidence that visual imagery and visual perception do in fact share low-level neuronal mechanisms.
In the present study, we used event-related (ER) functional magnetic resonance imaging (fMRI) to investigate whether visual mental images retinotopically activate early visual cortex. With ER-fMRI, it is possible to obtain on an individual basis high-resolution retinotopic maps from specific visual perception paradigms [4], [5], [26], [27], [29] and to monitor activation patterns from single imagery events [13]. Individual mapping is particularly important to overcome the great anatomical and functional variability of the visual cortex. Participants were instructed to visualize horizontally or vertically oriented flashing bow-tie shaped stimuli. These orientations were chosen because visual perception of similar stimuli is known to activate well-defined and well-separated retinotopic loci on the visual cortical ribbon, namely the horizontal meridian (along the calcarine fissure) and the vertical meridian (along the lateral V1/V2 border).
Section snippets
Participants
Six healthy participants (four male and two female, aged 20 to 25 years) participated in the study. All reported being in good health and free of any ophthalmologic disorders. All volunteers gave their informed written consent. The study was approved by the local Institutional Ethic Committee.
Materials and task design
The participants were scanned under three experimental conditions (visual perception, visual imagery, and visual retinotopic mapping).
In the perceptual condition (Fig. 1a), participants were supine and
Results
For each participant, reliable cortical representations of the horizontal and vertical meridians were obtained from the phase-encoded retinotopic paradigm. The spatial activation patterns associated with the actual (perceived) horizontal and vertical stimuli precisely matched the corresponding location of the horizontal and vertical meridians (Fig. 2). During visual imagery, we found significant activation in V1 in five out of the six participants (threshold p value=10−2, clusters of at least
Discussion
In this study, we clearly obtained V1 activation for all participants and all imagery events, although the degree and precise location of activation varied among participants. This finding in itself confirms previous research demonstrating that area V1 is involved in visual mental imagery. However, when we compared activation during imagery to the baseline, we did not find evidence that V1 activation was retinotopically organized. However, one has to keep in mind that with the current fMRI
Acknowledgements
The authors wish to thank Stanislas Dehaene and Yves Frégnac for their helpful comments.
References (31)
- et al.
A functional MRI study of mental image generation
Neuropsychologia
(1997) - et al.
Distributed neural systems for the generation of visual images
Neuron
(2000) - et al.
Increased activity in human visual cortex during directed attention in the absence of visual stimulation
Neuron
(1999) - et al.
Reopening the mental imagery debate: lessons from functional anatomy
NeuroImage
(1998) - et al.
Visual imagery and visual representation
Trends in Neurosciences
(1994) - et al.
New images from human visual cortex
Trends in Neurosciences
(1996) - et al.
From retinotopy to recognition: fMRI in human visual cortex
Trends in Neurosciences
(1998) - et al.
Human primary visual cortex lateral geniculate nucleus activation during visual imagery
NeuroReport
(1988) - et al.
Occipital cortex in man: organization of callosal connections, related myelo- and cytoarchitecture, and putative boundaries of functional visual areas
Journal of Comparative Neurology
(1990) - et al.
Mapping striate and extrastriate visual areas in human cerebral cortex
Proceedings of the National Academy of Sciences of the United States of America
(1996)