| |
Abstract:
Functional MRI (fMRI) data suggest that parahippocampal
responses during encoding predict subsequent memory (Brewer et al.,
1998; Wagner et al., 1998) whereas depth electrode recordings in
humans show differential encoding-related activity in perirhinal
cortex for subsequently remembered vs forgotten words (Fernandez et
al., 1999). Here we address whether the lack of anterior medial
temporal activation observed in subsequent memory fMRI experiments
is due to a true lack of differential haemodynamic responses to
subsequently remembered or forgotten items, or to decreased
sensitivity of fMRI in these anterior perirhinal regions. During
scanning, 14 subjects rote-encoded 12 words and, following a
distractor task, engaged in free recall. T2* images comprised 2mm
axial slices, with restricted in-plane field of view, positioned to
cover perirhinal, entorhinal and parahippocampal cortices and
hippocampus. Prior to the acquisition of each slice, a saturation
pulse was applied to a cor! onal section, positioned to cover the
eyes and frontal pole, to minimise frontal-occipital wrap-around
and Niquist ghosting of the eyes. An echo-time of 30 ms was used to
minimise temporal lobes signal drop-out. These parameters enabled
the different medial temporal structures to be easily discriminated
on T2* images. Initial single subject analyses demonstrated greater
anterior medial temporal activation during encoding for
subsequently remembered vs forgotten words. Hence, encoding-related
anterior medial temporal activation, predictive of subsequent
memory, is detectable with high-sensitivity fMRI.
|
