At 09:30 AM 3/24/00 +0000, Paul Fletcher wrote:
>
>>The difficulty with this is that it is difficult to interpret a drug x task
>>interaction in a region of the brain that is not activated by the task in
>>the 'no-drug' or placebo condition and therefore this cannot circumvent the
>>problem of differential gain response in different regions on drug. However,
>>it is important to bear in mind that this should not invalidate the
>>technique, particularly if your effects are predicted and meaningful.
>>
>
>Hello,
>I've been following this discussion with much interest. This last point has
>confused me somewhat. I'm not sure why the appearance of a task-related
>effect in the presence of the drug (when there is no effect in the drug's
>absence) should be difficult to interpret - or rather any more difficult to
>interpret than the attenuation, augmentation or disappearance of a regional
>task-related effect in response to the drug. Perhaps you could expand Mitul.
>
>One other point implicit in Jenny Coull's first message but worth
>considering more explicitly is that a regional drug by task interaction may
>not be due to a non-specific vascular change, a regionally specific
>vascular change or a directly pharmacologically mediated effect on regional
>function. It could be a downstream effect of a drug-related change in the
>way that subjects carry out the cognitive task. That is, if the effect of
>the drug is to make the task in any way more demanding (or less demanding)
>then task by drug interaction effects may reflect this change. Simply put,
>subjects may deal with the activation task differently depending upon
>whether or not they have the drug on board. If the drug makes it difficult
>for them, they might engage new strategies to optimise performance. If the
>drug makes it easier for them, they may dispense with strategies.
>
>There is an important difference between this effect and what seems to be
>the more common interpretation of drug by task interactions. The
>difference between saying 'the drug changes the brain - different brain
>equals different activation' and saying 'the drug changes the task -
>different task equals different activation'
>
>This extra ingredient in the pot is certainly not a reason for being put
>off such studies though. Personally, I find it even more interesting.
>
>Paul Fletcher
>
>
>
>Dear Paul, Mit, Jenny, and all,
I think that there are a number of important issues raised in the last two
mails from Mit and Paul, and I think some answers may be gleaned from
computational studies of neuromodulation.
I think Mit's point is that neuromodulatory agents will alter signal
properties, and so one would expect changes in those regions where there is
increased (or decreased) signal during performance of some cognitive task.
However, I think Paul raises an important issue, which may relate to
"noise" rather than signal.
First, I think Paul is right that it is not necessarily any more difficult
to interpret findings where one sees task * region interaction effects in
regions that are not activated in the task condition alone. It is well
established that neuromodulatory drugs such as dopamine, noradrenaline, and
acetylcholine alter signal to noise properties of neurons. Thus, it is
quite possible that, say an antagonist drug would lead to increased "noise"
and decreased "signal", which could lead to increased activity in regions
of the brain not activated in the main effect of task (or indeed, not
"deactivated" in any main effects).
So I think Mit and Paul's points are two sides of the same coin. Altered
signal processing should lead to changes in those regions acitivated by
task. However, increased noise in the system may lead to activations in
neural regions not activated in the main effect of task.
Paul makes another interesting point about potential changes in "strategy"
by subjects, which could instead be interpreted in a signal / noise model.
I would imagine that if subjects are changing strategy in some way (for
example if the task appears more "effortful") one would see some impact
upon performance measures. However, if regions are activated because of
increase "noise" one would not necessarily expect to see alterations in
performance (as long as the noise was not too great).
Paul's idea would also suggest that changes in say rCBF on drug caused by
increased "effort" should be similar to changes induced by increasing task
difficulty / demand, which is certainly testable. However, thinking in
terms of altered signal / noise ratios enables a link with computational
models of neuromodulation, whcih may be useful in generating interfield
theories.
However,I don't think any of these issues, or the ones raised by Paul, Mit
and Jenny, are necessarily helpful in distinguishing between drug effects
on neuronal and vascular function. I think we will have to wait for monkey
fMRI to develop so that we can combine fMRI, single unit
physiology,microdialysis and iontophoretic application of neuromodulators
to help answer some of these difficult questions (although I don't think we
should not do the experiments in humans!)
With best wishes,
Andrew
___________________________________________________
Andrew Lawrence PhD,
MRC Cognition and Brain Sciences Unit,
15 Chaucer Rd,
Cambridge CB2 2EF,
UK
>
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|