Poster No.: 829
VENTRAL SIMULTAGNOSIA AND ACTIVATION OF THE DORSAL STREAM: A CASE STUDY WITH NEURO-IMAGING TECHNIQUES
S. Lê, K. Boulanouar, D. Cardebat, M. Hénaff, F. Michel, M. Puel, J-F. Démonet

INSERM U455, Toulouse, France INSERM U280, Lyon, France

Introduction
A single case study of a 28 year-old young man (SB), who suffered a meningoencephalitis at the age of 3 years, is presented. SB has lesions in the ventral stream bilaterally and in the right dorsal stream. The primary visual cortex is partially spared. Neuropsychological investigations have indicated that the patient has a severe visual object recognition deficit (colour, faces, words), although movement and space perception, along with visual acuity, tactile recognition, and global intelligence, are largely preserved. Analysis of visual function demonstrated that SB uses a "feature-by-feature" recognition strategy, but is unable to integrate local features into a global shape. In that way SB seems to show the syndrome of "ventral simultagnosia" described by Farah (1) in 1990. On the other hand, strikingly, SB is able in some extent to recognise simple forms such as geometric drawings.

Methods
In normal subjects, form and contrast perception is performed mostly by the ventral stream. Since this processing is not functional in SB, we wondered what are the spared cerebral structures in SB allowing the perception of forms and contrast. We have studied this point by the mean of functional Magnetic Resonance Imaging (fMRI). The activation paradigm is inspired from studies by Treisman (1984) and consisted of a target detection amongst 2, 4, or 9 items. The distinction between the items was more or less salient for form (X or C versus O) and contrast (dark or light grey versus white) perception. The mean results of 3 volunteer Controls have been compared to SB’s.

Results and Discussion
The behavioural results show that SB’s performances are above chance level for the salient conditions (75%), whereas they are at chance level for the non-salient conditions (46%). The Controls performances were perfect. fMRI data (processed with SPM96 (2)) show no significant differences between contrast and form perception. But strong activations are found in the primary visual cortex, that seem to depend on the apparent size and/or of the luminance of the stimuli (3). Strikingly, in the global comparison "Task versus Rest", there is an intense activation of SB’s left parietal cortex that is not present in the Controls (Z-score = 7.27, Talairach’s co-ordinates x,y,z=-24,-53,40) – see figure below.
The present functional imaging results complement the cognitive analysis of SB’s performances and permit a better comprehension of somewhat paradoxical findings. Indeed, we may hypothesise that, from deficits that have occurred in his childhood, SB has developed a system of functional compensation, such as an activation of a particular region in the left parietal cortex, thus offsetting the quasi-absence of the ventral stream. An other explanation is that, as the paradigm was relatively trivial for the Controls, this particular activation in SB’s dorsal stream may be due to attentional processes rather than cerebral plasticity. Furthermore, comparisons between the spatial configuration of the items may also be taken into account for the analysis of the results.

References
(1) Farah, M.J., Cambridge, MA: The MIT Press, 1990.
(2) Friston et al., Human Brain Mapping, 1995.
(3) Goodyear, B.G and Menon, R.S, J Neurophysiol, Apr 79, 2204-7, 1998.