Functional Imaging of Craving

Alcohol Research & Health, Fall, 1999 by Daniel W. Hommer

To visualize brain activity associated with mental states, such as craving for alcohol and other drugs (AODs), researchers have begun to use functional imaging techniques. Three commonly used techniques are single photon emission computed tomography (SPECT), positron emission tomography (PET), and functional magnetic resonance imaging (fMRI). Studies using these three approaches have been reviewed in order to evaluate the validity of a proposed model of the brain regions involved in alcoholism and the craving for alcohol. This model suggests a central role for a connected group of brain regions that include the basal ganglia, thalamus, and orbital cortex. A study using SPECT technology in alcoholics, however, found altered brain activity in only some of those regions during craving. Additional studies in alcoholics, as well as cocaine users, identified several other brain regions whose activities appeared to change in response to craving. These studies have led to the development of a revised model of brain r egions involved in craving for AODs. Numerous questions remain, however, that must be answered before the brain areas involved in craving can be identified conclusively. KEY WORDS: AOD (alcohol and other drug) craving; single photon emission computed tomography; positron emission tomography; magnetic resonance imaging; brain imaging; blood flow measurement; cocaine; brain; basal ganglia; limbic system; brain function; neurotransmission; dopamine; glucose metabolism; literature review

Primates, including humans, are visual animals--that is, they like to see things--giving rise to the adage "seeing is believing." One cannot see thoughts, feelings, or mental states, however, whether one's own or those of other people. One can only perceive these states either directly, in one's own consciousness, or indirectly, through the reports and behavior of other people. "Craving" is a term derived from popular psychology that is used to describe one of these mental states--namely, the intense desire for a certain object or experience (e.g., alcohol or other drugs [AODs]).

Cognitive neuroscience postulates that mental states are the product of neurochemical processes in specific brain regions (i.e., brain states). Although one cannot see mental states, one can generate images that help visualize the intensity and location of physiological processes associated with brain states.

At the most basic level, the physiological processes underlying brain states involve changes in and the maintenance of differences in the concentrations of molecules that carry electrical charges (i.e., ions) across the membrane forming the boundary of each nerve cell, or neuron. Changes in the concentrations of various ions inside and outside of the cell serve as signals that can be transmitted from one neuron to another. Because differences in ion concentrations already exist between the inside and the outside of a resting or inactive neuron, the process of moving additional ions to generate a nerve signal requires energy. In the brain, this energy is supplied by the brain's metabolism of the sugar glucose, which is delivered to the brain through the bloodstream. Accordingly, when a certain brain region is actively involved in the generation of mental states, the energy requirements of that region increase and, consequently, so does the blood flow there. By measuring blood flow or glucose metabolism (a pro cess known as functional brain imaging), one can determine which brain areas are most active during a particular brain state.

This article reviews the small number of studies that have applied functional brain imaging to investigating brain states associated with craving for AODs, particularly alcohol and cocaine. It includes studies on cocaine craving because (1) to date, only two studies have attempted to image craving for alcohol, and (2) the brain states associated with craving may be similar for all AODs. The article first describes the three most commonly used functional imaging techniques. It then introduces a preliminary model of the functional anatomy of craving to provide a framework for understanding the studies on craving conducted among alcoholics and cocaine abusers. After comparing the results of studies that examine the preliminary model, the article discusses future directions in the functional imaging of craving.

METHODS OF FUNCTIONAL BRAIN IMAGING

Currently three major techniques are used to visualize the brain activity associated with various mental states: single photon emission computed tomography (SPECT), positron emission tomography (PET), and functional magnetic resonance imaging (fMRI). Each technique involves measuring local changes in cerebral [1] blood flow or metabolism. To that end, researchers usually obtain at least two separate images representing two different states of the brain regions of interest. In theory, one could just visually compare the two images to identify brain regions in which brain activity differs depending on mental state. In many cases, however, the differences between the two images are too subtle to be detected or quantified accurately by simple visual inspection. Therefore, a new image representing the differences between the two original images (i.e., a difference image) is created using computerized analysis. This can be done because each image is made up of small units, or pixels, each of which represents a smal l brain area. For each pixel the blood-flow or glucose-metabolism value in the first image is "subtracted" from the value in the corresponding pixel in the second image. For example, assume that the blood flow in a given pixel is low in the first brain state (e.g., no craving) and is assigned a value of 2 on a scale from 0 to 10. In the second brain state (e.g., craving), the blood flow in the same pixel is high (i.e., is assigned a value of 9). Thus, in the difference image, this pixel will have a blood flow value of 7, representing a relatively large difference between the two brain states. This subtraction process is performed for all pixels in the images, generating an image that represents the extent of changes in blood flow between the two brain states being assessed. Statistical analysis of several difference images then allows the identification of the brain regions whose activities vary most strongly between the two brain states.