| |
Abstract:
Often, subjects with amnesia from orbito-frontal or basal
forebrain damage produce memories that are not completely
veridical, with some memories tending to the bizarre. Explanations
for this phenomenon (confabulation) vary - one view is that they
are pure fabrications, but an alternative view is that the recall
contains elements of veridical memories which have been combined
incorrectly. Often, however, when dealing with memories involving a
subject's autobiography, evaluation of this is problematic since
verification of details is difficult. Using a task designed to
study long-term recall, we present here a case study of a patient,
RF, previously known to confabulate (Damasio, Tranel & Damasio,
1989) who suffered an anterior communicating artery aneurysm
damaging the left basal forebrain. RF and control subjects were
presented ten novel stories and then recall was tested at
systematic intervals (one hour to six months). Subjects' free
recall protocols were scored for accuracy and for whether errors
contained "intrinsic" confabulations (elements of one story being
recalled incorrectly within another story) or "extrinsic"
confabulations (elements that formed no part of the original
learning). Relative to controls, RF made a larger ratio of
intrinsic to extrinsic errors, suggesting that confabulations in
recall can involve mis-combinations of veridical elements. It is
suggested that this explanation could also be invoked to explain
confabulations found in autobiographical recall. This finding is
discussed in relation to current models of memory recall.
|
