The purpose of this study is :

• to look for changes induced by aging in a population of old subjects, during silent word generation tasks studied by fMRI.
• to look for a possible link between these changes and alterations in language

Introduction : Several factors may influence cerebral organization of language, age being one of the most important. We previously showed that activations were different in old subjects compared to young for a silent word generation task : In Old subjects, activations appeared less important and more parceled but involved the same regions of the left hemisphere than for young subjects. Several small clusters of activation were present on the right hemisphere of old subjects and more or less mirrored left activations in the same group, leading to a less lateralized pattern than in young subjects. Moreover, during the same task, activations decreased for about of 1/3 subject whereas is increased or was stable for young. These changes may be a preclinical indicator of cognitive decline.

Objective: In this study, we propose to rescan the same subjects 2 years after the first protocol using the same paradigm, and to look for possible changes in location, extend and time course of activations. These results will be compared to neuropsychological examination in order to find a possible correlation between possible changes and cognitive impairment.

Subjects : the 22 old right-handed subjects (age>76y, 11M, 11F) enrolled in the preliminary study will be included.

Method :

• fMRI : Data will be collected on a 1.5 T General Electric Signa MR scanner. Functional images will cover the whole brain and will be scanned using a single-shot echoplanar imaging sequence (FOV : 24 cm ¡thickness : 5 mm - 64 x 64 matrix - TE : 60 ms ¡ TR : 3s ¡ flip angle : 90°). The paradigm will follow a box car design and 3 periods of activation (30 s duration) will alternate with 3 periods of fixation (30s duration). During the active conditions subjects will have to mentally generate words corresponding to a visually presented semantic category (condition 1), or beginning by a visually presented letter (condition 2). During the control condition, they will be requested to concentrate on their breathing. Data will be analyzed using the FSFAST package. MR data will be motion corrected using AFNI and then smoothed with a FWHM of 5 mm. The voxel intensities will be rescaled such that the in-brain grand mean will be 1000 for all subjects. The hemodynamic response will be modeled as a gamma function (delay : 2.25 s - dispersion :1.25 s). The offset and linear drift in the BOLD signal will be removed by estimating those components simultaneously with the amplitude of the hemodynamic response. A t-test will then be performed on the amplitude of the hemodynamic response to determine which voxels has significant paradigm-related activity. Time course of the signal will be separately studied for each condition
• anatomical MR : Results will be displayed on the inflated cortical surface for each subject. A group analysis will be performed for each contrast using a surface coordinate system and random model effect and General linear model using age an neuropsychological scores as regressors.
• Neuropsychological examination : subjects will be interviewed by a psychologist 4-6 weeks after scanning session. Semantic and lexical verbal fluency will be evaluated, as will general cognitive function (MMS, lokk for Alzheimer's disease, IADL, BEC96), and episodic memory (Grober & Buschke).

Inclusion Criteria:

• preliminary included in preliminary study (PHRC-R 2001)
• delay of 18-24 months after previous MRI

Exclusion Criteria:

• contraindicate for MRI