Functional Imaging of Craving : Part 5

by National Institute of Health

(Page 5 of 6)

The researchers took advantage of the fact that fMRI allows the collection of many images during a brief period of time and examined the relationship between subjective response and the time course of brain activation. They found that the activity of the brain regions that showed an early peak response followed by a rapid return to pre-cocaine levels correlated with euphoria. Conversely, the activity of the regions that showed a sustained increase in activity (the nucleus accumbens/subcallosal gyrus region and the basal forebrain) correlated positively with craving. Finally, the time course of the BOLD signal in the amygdala negatively correlated with craving.

Two studies used methylphenidate to elicit craving in cocaine abusers who had been hospitalized for treatment for at least 3 weeks. In one of those studies, changes in brain glucose metabolism were measured in 20 cocaine abusers before and after intravenous administration of methylphenidate. ROI analysis demonstrated that methylphenidate increased brain glucose metabolism in the anterior cingulate gyrus, right thalamus, and cerebellum. Furthermore, methlyphenidateinduced increases in brain glucose metabolism in the right orbital frontal cortex and right caudate nucleus correlated positively with craving for cocaine. Because the study included no control subjects, however, it is unclear whether this pattern of activation is unique to cocaine abusers or would also occur among non-drug users.

The other study involving methylphenidate examined changes in the release of the neurotransmitter dopamine using PET technology and an agent called 11C-raclopride, which can interact with a specific receptor for dopamine. Dopamine release may be associated with the pleasurable or rewarding effects of AODs and with the motivation to use AODs to experience those rewarding effects. Methylphenidate causes the release of dopamine in the brain, which then displaces 11C-raclopride bound to D2 receptors. Thus, changes in 11C-raclopride binding provide a measure of how much dopamine is released in response to methylphenidate administration. The study found that compared with control subjects, cocaine abusers exhibited decreased dopamine release in the striatum, specifically the caudate nucleus, but increased release in the thalamus. Furthermore, the magnitude of thalamic dopamine release among the cocaine abusers was positively correlated with craving for cocaine. However, the relevance of this correlation is unclear, because both the release of dopamine and the concentration of dopamine receptors are low in the thalamus.

Comparison of the Study Results

The six imaging studies conducted among cocaine abusers and the one successful study conducted in alcoholics identified numerous brain regions that appear to be involved in craving. The specific pattern of brain regions that is activated appears to depend to some extent on the technique used to elicit craving. One group of studies induced craving by administering small amounts of the abused drug itself or of an agent with similar effects. Those studies reported increased activity in the thalamus during craving. In contrast, studies that used visual cues to induce craving reported increased brain activity primarily in the dorsolateral prefrontal cortex.

The prefrontal cortex generally is activated during tasks involving both working memory (the short-term storage of information) and episodic memory (the recall of sequences of past events). Accordingly, activation of the dorsolateral prefrontal cortex during cue-induced craving may occur because craving based on visual cues requires the activation of certain memory circuits, whereas such memories may not be required for the occurrence of drug-induced craving. It is unclear why cue-induced craving fails to activate the thalamus, because both drug-induced and cue-induced craving are associated with activity changes in the anterior cingulate gyrus, caudate nucleus, amygdala, and orbital cortices. All these regions, along with the thalamus, are components of various basal gangliathalamocortical circuits; because of their anatomical connections, one would expect these regions to be functionally linked.

Although the studies discussed in this article indicate that the brain regions mentioned earlier are activated during craving, one cannot exclude the possibility that other regions are activated as well. All but two of the studies used ROI methods of image analysis and therefore examined only limited portions of the brain. In fact, most of the studies limited their analysis to the caudate nucleus; thalamus; amygdala; and orbital, dorsolateral frontal, and anterior cingulate cortices. The only two studies to analyze data from the entire brain detected activation in the nucleus accumbens and insular cortex in addition to the regions listed above. The nucleus accumbens and insular cortex, as well as the other activated regions, are components of striatal-thalamocortical circuits. Although the precise function of those circuits in humans is unknown, they are thought to be involved in emotion, motivation, and social behavior. In addition, the insular cortex is involved in visceral functions related to feeding and emotion.

A Revised Model of the Functional Neuroanatomy of Craving

The brain regions identified as being involved in craving for alcohol or cocaine all participate in various basal gangliathalamocortical circuits, which influence a person's sequential choice of behaviors (actions, thoughts, and emotions) based on the behavioral context. Three of those regions (the caudate nucleus, thalamus, and orbital frontal cortex) were included in the initial craving model developed by Modell and colleagues. In addition, the dorsolateral prefrontal cortex may be involved in craving, as discussed earlier in this article. The anterior cingulate cortex, which was activated in most studies of craving, is involved in the regulation of attention and emotion and may be particularly important in learning associations between certain stimuli and rewards. Finally, the amygdala is commonly associated with fear but also relays information about stimuli associated with reward to the orbital cortex.

The brain regions associated with craving also are generally associated with motivation, a state that can be described as inducing those behaviors that are related to satisfying an organism's needs. Furthermore, with the exception of the thalamus, all these brain regions receive information from dopamine-releasing neurons that are located in the midbrain but extend their axons to these regions and release dopamine there. Dopamine is often considered a reward neurotransmitter whose release provides the drive behind behavior undertaken in anticipation of a pleasurable outcome (appetitively motivated behavior). Conversely, reductions in dopamine activity may be involved in motivating behavior undertaken in order to avoid the loss of pleasure (aversively motivated behavior). This bivalent nature of dopamine's motivational function may mirror the mixture of motivations that appear to comprise craving. On the one hand, craving can result from an alcoholic's desire for the pleasurable effects of alcohol (an appetitive drive). On the other hand, craving also can result from an alcoholic's desire to avoid the unpleasant experiences of withdrawal or another negative emotional state (an aversive drive). Thus, craving may involve either aversive or appetitive motivation and, at times, perhaps both simultaneously.

About the Author

NIH is the nation's medical research agency - making important medical discoveries that improve health and save lives. The National Institutes of Health (NIH), a part of the U.S. Department of Health and Human Services, is the primary Federal agency for conducting and supporting medical research.