It is stated in the middle of page 2 of the handout that neuroscientific evidence tends to support separate brain mechanisms for Pavlovian and instrumental conditioning.

The basic material for this can be found on pp 111-112/p 118 at the end of Chapter 3 (Motivation) in Gleitman et al. (1999/2004), which covers the finding of self-stimulation or "pleasure" centres in the rat brain, and the assumption that addictions to nicotine, cocaine, alcohol and other substance of abuse come about because these pleasure centers and other associated brain circuits, become "pathologically subverted" (Everitt, Dickenson and Robbins, 2001).

Little detail about these circuits is needed for present purposes, but plenty of relevant material is available via the internet. The Society for Neuroscience has a good pdf file called "brainfacts" which has clear diagrams of the nucleus accumbens and other parts of the brain reward circuits on page 37. However it is 58 pages long and would be difficult to download over a telephone modem.

An alternative is the web pages at and [reproduced in part on p. 6 of the paper handout: may not work with Netscape 4.x]. is the page about the amygdala and its allies.


More detail can be found in the online lecture on “The Biology of Reward Mechanisms” by Dr Eliot L. Gardner, Director, Basic Psychiatry, Albert Einstein School of Medicine, Bronx, NY 10461-1602.

It is far too detailed for the present purposes, but includes at the beginning the basic point that "Addictive drugs have in common that they are voluntarily self-administered by laboratory animals (often avidly) and that they facilitate the functioning of the pleasure/reward/reinforcement circuitry of the brain."

The rat brain diagram on this site can be viewed separately by clicking here

The sketch of the rat obtaining intracranial self-stimulation is here


A general review is Wise, R. A. (1998). Drug-activation of brain reward pathways. Drug and Alcohol Dependence, 51(1-2), 13-22. which is available online via ScienceDirect (log on first for for access outside the College)

A broadly similar treatment, which includes diagrams of the human brain sites, is given here at the USA's National Institute on Drug Abuse.

Volkow et al., (2004) reviewed imaging studies of addicts — also ScienceDirect

A particular theory, “incentive sensitization” is put forward by Robinson and Berridge (2003) in the Annual Review of Psychology [available online from Birkbeck]  

One of the key sites in the vertebrate reward circuit is called the "nucleus accumbens", which is linked to the use of several drugs such as cocaine and amphetamine. It is speculatively identified as a crucial part of the "goal seeking" mechanism (Ikemoto and Panksepp, 1999). The paper by Knutsen et al. (2001) referred to in the lecture claimed that in human subjects anticipation of increasing monetary rewards increased both self-reported happiness and activtiy in the nucleus accumens, whereas anticipation of losing money elicited neither. This paper is only available online, at

The picture of the brain scans with activity in the n. accumbens only when reward increases are anticipated (top left) can be viewed separately here

This supports a widely held view that the midbrain and limbic reward circuits (using the neurotransmitter dopamine) can in someways be separated from the limbic circuits which response to unpleasant or aversive stimuli, in which the amygdala plays an important role.

On the other hand there is evidence that the nucleus accumbens is involved in both the "craving" part of drug dependence and the "rush" or euphoric state which follows drug ingestion (Myrick et al., 2004)  

The nucleus accumens is of course only one of many areas of the brain involved in reward mechanisms. Important areas in the human brain include parts of the frontal cortex called "orbitofrontal cortex".

Small et al (2001) suggested that "there may be a functional segregation of the neural representation of reward and punishment within this region." This was based on repeated brain scanning while volunteers ate chocolate upto and beyond satiety. Therefore the stimulus input was very similar, but subjective effects very different. The authors conclude that the results " support the hypothesis that there are two separate motivational systems: one orchestrating approach and another avoidance behaviours"
abstract | full article in Brain (if at BK, otherwise ask in the library for the Brain password

There are several similar papers in which the rewards and punishments for human volunteers were real or imagined gains or losses of money. Elliot et al. (2003) say that the frontal cortex brain areas which react to these gains and losses are very similar to those which react to wanted or unwanted chocolate. (Elliot et al, 2003 abstract)

O'Doherty et al. (2001) which have a figure showing brain areas reacting to losing money at the outer sides (lateral) of the orbitofrontal cortex, in contract to areas activated when monetary gains are anticipated, which are more towards the middle (medial).
This article online

Rolls,E.T., O'Doherty,J., Kringelbach,M.L., Francis,S., Bowtell,R. and McGlone,F. (2003) Representations of pleasant and painful touch in the human orbitofrontal and cingulate cortices. Cerebral Cortex 13:308-317. This is a recent article from the same laboratory also available online.

The same circuits are involved, in a more exagerrated way, in addictions.