Re: Ampakines

Anders Sandberg (asa@nada.kth.se)
10 Oct 1997 12:47:16 +0200


"Eliezer S. Yudkowsky" <sentience@pobox.com> writes:

> Strictly offhand, just extrapolating from those instances of Algernon's Law
> I've been able to find, I would guess that ampakines disenhance
> linear/sequential event memory and temporospatial solutions (catching a ball).
> Did they try testing for that?

They tested a digit cancellation task (cross out as many instances of
'1', '4' and '8' from a sheet of random digits, as accurately and quickly
as possible) and a finger tapping test (tap with the right index finger,
tap with the left, alternate index and middle fingers, alternate
left and right index fingers; the speed of tapping and correctness was
measured). Neither of these had any significant deviations from the
untreated subjects. There was a *tiny* (non-significant) decrease in
the finger tapping task, but not enough to support your above idea.
Of course, these were just two rather simple tasks.

However, my guess is that ampakines have some kind of drawback. They
seem to make the postsynaptic signals stronger in many neurons, which
may lead to problems with overstimulations (I think they are bad for
epileptics). Also, an increase in neural plasticity is not 100% positive.
If you learn easily, you will also easily erase old information; your
memory range may get shorter. I'm not sure yet if the ampakines mainly
act in the median temporal lobe (in that case they might make your
learning more selective) or in the cortex (in that case they might make
you forget old memories faster while you learn faster). We'll see...

> A 10-20% improvement may be indicative of cognitive resources being entirely
> focused on a single class of problem. This may or may not be worth the likely
> 30-60% disimprovements in cognitive areas outside the narrow ones studied.

So far we have not seen any evidence for disimprovements; my guess is
that the effects are a bit more subtle than a simple decrease in some
other ability; I would say the expected result would be a shift in the
balance between different cognitive abilities (such as learning/forgetting
and stable storage).

-- 
-----------------------------------------------------------------------
Anders Sandberg                                      Towards Ascension!
asa@nada.kth.se                            http://www.nada.kth.se/~asa/
GCS/M/S/O d++ -p+ c++++ !l u+ e++ m++ s+/+ n--- h+/* f+ g+ w++ t+ r+ !y