>> The brain formation that recognizes new episodes and commits them to
>> memory is the right hippocampus (HC).
> That would be visual/spatial episodes. Linguisticly based episodes are
> normally processed by the Left Hippocampus (95% of right handers and
> about 75% of left handers). Since most episodes in life are a combination
> of linguistic and nonlinguistic components, both hippocampi are usually
Yes, and the two hippocampi are connected by a small bundle of neurons,
presumably so each HC can share information with the other HC in
situations that include both semantic and episodic components.
>> By removing redundancy and compressing the input pattern to essential
>> features, the DG discards information, but without details, sooner or
>> later a new pattern will match a prior pattern based on only the
>> features, but not match the details in the uncompressed memory. When
>> the details do not match, the prior memory does not become conscious, but
>> the HC has already signaled that the pattern is a familiar one because
>> the essential features do match. This is the deja vu feeling. We feel
>> that a new situation has occurred before but cannot recall when or where.
> My initial gut reaction is that this isn't quite what déjà-vu is about.
> It's not experienced so much as a memory but a reliving.
The reason it feels like a reliving is because the details in the present
episode are falsely tagged as "experienced before" but no details are
recalled from a past episode. If the two episodes are similar, e.g. when
revisiting the place of your childhood, past episodes become conscious
and past details can be compared to corresponding details in the present
episode. Differences are noticed - the height of the tree in the front
the color of the paint, a new garage that was not there when you were a
child, etc. But in déjà vu, the details are all different and recall fails
because mismatched past episodes are repressed.
> One can simply explain déjà-vu as an electrical blip that shouldn't
> have happened.
This is similar to what happens in a computer data base management
system when an incorrect record is retrieved. Let's suppose you request a
personnel record by social security number, but to speed up retrieval the
search program checks only the last 4 digits of the number. If the 4 digits
do not match, a "record not found" message is displayed. And if the
4 digits do match, it is usually the correct record. But in rare instances,
the 4 digits match but an incorrect record is retrieved.
That happened to me once when a pension record for another person was
sent to me by mistake (different name, different birth date, different
different salary, different gender, etc), because the last 4 digits of the
person's social security number were the same as mine. This is not déjà vu,
but I mention it as an example of how retrieval systems can make errors
when a search argument is kept short for high speed.
Neurons are so slow that rapid retrieval with a few rare errors has much
better survival value than slow accurate retrieval. Oh yes, now I remember,
the mushroom I just ate was the same kind of poisonous mushroom that
killed my sister.