MFB - medial forebrain bundle

Stephan Anagnostaras stephan at
Wed Aug 16 05:46:02 EST 1995

In article <40rbuo$37r at>, sethbh at (Seth BH) wrote:

> Regarding "the" reward system, first of all, there is no single location
> that can be pointed to as a reward center; however, stimulation of the
> septal nuclei, MFB and hypothalamic regions can elicit behaviors which
> seem as if the subject is being rewarded (lord how radical behaviorist I

There are at least fifty sites which have reinforcement properties (i.e.,
if stimulated they increase the probability of behaviors they follow),
and most of these also have reward characteristics (generally, 'rewards'
are defined in that they have both reinforcement properties and can
elicit approach behaviors; the earlier is assessed with operant
bar-press ICSS tasks while the latter is usually asssessed in
conditioned place preferecence paradigms - this is somewhat controversial;
but, things can be reinforcers but not be rewards, e.g., electrical
shocks; some people do not believe in any kind of reinforcement mechanisms
only the ability to elicit approach behaviors, tho': you might look at
reviews by Bindra).
Aside from the sites you mention (the popular sites) you can look
at almost any part of the brain (Amygdala, HPC, brainstem, cortex)
and find good ICSS sites; there are also lots of sites for aversive

> sound...) But remember that the MFB is not a nucleus of cells (and thus a
> potential "center") but rather a collection of fibers from various
> aminergic neurons in the brainstem which project through the lateral
> hypothalamus to the telencepthalon.  You might want to look into the
> nucleus accumbens as a major reinforcment "center."   

Although there is good general agreement that the rewarding properties
of both many types of ICSS as well as stimulant (and perhaps opiate)
drugs is due to dopamine release in the nucleus accumbens there is
fairly poor evidence that this is a central reward mechanism. Although
natural rewards (e.g., food) do cause stimulation of this system,
and DA-ergic blockade is often sufficient to block the behaviors,
electrolytic or excitotoxic lesions of the nucleus accumbens do
surprisingly little to the animal's normal behavior.  Some people
have reported larger effects with 6-OHDA lesions of the acc, but those
also blow out most of the VTA. 

 In any case
the argument is generally made by ICSS and drug-addiction people that
the system is the vta-accumbens dopamine pathway. The cells of origin for
this "reward" system are in the vta, while the target is in the acc;
this pathway courses thru the MFB, but MFB stimulation "reward" does
not directly cause DA release in the acc. The complicated circuit
was worked out, but it amounts to the electrical stimulation causing
backward release of NE at the VTA which results in DA release in
the acc (-- note however, that it is the DA release that is responsible
apparently for the rewarding properties of the stimulation).  Many
years were spent trying to solve this problem of how exactly ICSS
worked physically that the theory of "reward" wasn't well developed,. Noone
really believes this is a central reinforcing mechanism anymore;
the argument has been refined now in that this is variously considered
a mechanism for modulating incentive strength or salience and that
ICSS or DA manipulations alter these values (for you Rescorla-Wagner
types these would be like alpha and beta or v; the original argument
was that ICSS had some kind of lambda type value).  The evidence for
that is inconsistent, and the effects of stimulation of the vta-acc
pathway cannot be well-characterized theoretically as either
reinforcing, hedonic, rewarding, or incentively reinforcing effects or
as affecting salience or rate in any of those mechanisms.

> Anyway, a classic in the field is Olds & Milner (1954) Positive
> reinforcement produced by electrical stimulation of the septal area and
> other regions of the rat brain. J. Comp. Physiol. Psychol.  47: 419-427.
> For a more recent review, check out 
> Toates (1986).  Motivational systems.  Cambridge, England: Cambridge
> University Press

The leading theorist on the "reward" theories of these systems is
probably Roy Wise. You can check several of his reviews. There is
an excellent book called "Neuropharmacology of Reward" which contains
articles by most leading people in the field.  You might also check
out the drug-addiction (esp amph) literature which parallels this interestingly:
e.g., Koob (1992) Proc NY acad sci or Robinson & Berridge (1993) Brain
Res Rev.

One problem in all of this literature is that the concept of reward is
not well-defined in learning theory and the ICSS and drug people in
the past have generally been poorly informed on what little definition
exists (it is just posited apriori as some kind of hedonic state). Recently
however, there is ample evidence that the mechanisms which underlie
hedonic states and reinforcement are not the same (e.g., liking vs.
wanting arguments) even tho' most people assumed this was the case.
So your chase would first be to find a good definition of reward.
Even if you find someone who knows what they are talking about,
you will find they have largely ignored the aversive conditioning
literature and their argument only works for behaviors that are
supported thru appetitive mechanisms; so even at this point you
get lost even before you start to look at the brain.

> There is a lot of literature on this, do a MedLine, Bio Abstracts or
> PsychLit search and you'll probably spend most of your day just chasing
> downt the refs, let alone reading them.

Truly Seth is right, this literature is vast and ugly.
> Good luck.
> Seth Boatright-Horowitz
> Dept. of Neuroscience
> Brown University


More information about the Neur-sci mailing list