| |
Abstract:
Abstract: High-density ERPs were recorded from 15 healthy
volunteers during a three-phase object-memory experiment. In a
study phase, subjects viewed familiar object drawings and decided
whether they were natural or manufactured (incidental learning). In
a free-recall phase, they were asked to write down as many as
possible of the objects' names. In a test phase, they decided
whether the objects were old or new. Study-phase ERPs. ERPs to
objects during encoding were averaged based on whether they were
later recognized or not, and recalled or not. Recognition-based ERP
encoding effects were absent, possibly due to automatic encoding of
pictorial information. In contrast, the ERP to later-recalled
objects was significantly more positive than the ERP to
later-non-recalled objects, particularly over left
superior-posterior temporal scalp (peak:540 msec) and left inferior
frontal scalp (peak:660 msec), corresponding well to previously
reported PET activations in semantic encoding and categorization
tasks. This study thus appears to provide location and timing of
object-memory encoding based on semantic processing. Test-phase
ERPs. ERPs to objects during recognition showed two effects. An
early effect (500-600ms) was a significantly
left-greater-than-right dorsal-parietal positivity for all objects
categorized as "old" relative to "new". A later effect (800-900ms)
was a significantly right-greater-than-left dorsal-frontal
positivity which discriminated true from false recognitions,
suggesting a slower, but more accurate, verification process at
work during episodic retrieval of familiar objects.
|
