Since the type I-type II alcoholism classification was developed, many researchers have confirmed the findings of the original studies and have further investigated differences between the two subtypes. In these studies, the age at onset and the type of alcohol-related problems emerged as the characteristics that most readily distinguished between the two subtypes2 (see table 2) (Babor et al. 1992; Gilligan et al. 1988; von Knorring et al. 1987a). Type I alcoholism developed during adulthood and generally was characterized by binge drinking (i.e., prolonged drinking bouts with default of responsibilities), interspersed with prolonged periods of abstinence; loss of control over drinking; excessive guilt about drinking; and rapid progression from mild to severe alcohol abuse, often accompanied by the development of alcoholic liver disease. Conversely, type II alcoholism generally commenced during adolescence or early adulthood, and alcohol consumption frequently was accompanied by fighting and arrests. In addition, alcohol abuse was moderately severe and frequently required treatment, although the severity of abuse did not change over time as it did in type I alcoholism.

Other researchers detected differences between type I and type II alcoholics not only in the age at onset and the type of alcohol-related problems, but also in certain neurobiological markers. For example, several studies found that compared with type I alcoholics, type II alcoholics exhibited lower activity levels of the enzyme monoamine oxidase (MAO) (von Knorring et al. 1987b; Sullivan et al. 1990). MAO is involved in metabolizing certain brain chemicals (i.e., neurotransmitters) that mediate signal transmission among nerve cells. One neurotransmitter metabolized by MAO is serotonin. Accordingly, reduced MAO activity could indicate a reduced turnover of serotonin in the central nervous system (CNS) (Oreland and Shaskan 1983).

Virkkunen and Linnoila (1990) also found that type I and type II alcoholics differed in their serotonin activities in the CNS. The serotonin levels were reduced in the brains of men with earlyonset alcoholism accompanied by violent behavior (i.e., type II alcoholics).

Type I and type II alcoholics also differed in their patterns of electrical brain waves as measured by an electroencephalogram (EEG). These differences existed when the subjects were resting (Cloninger 1987a), but also when they were exposed to certain stimuli. Uncommon stimuli interspersed among common stimuli (e.g., a rare green light among a sequence of red and yellow lights) elicit brain waves, called eventrelated potentials (ERP’s), that are made up of several components. A commonly studied ERP component is called P300, because it occurs about 300 milliseconds after the uncommon stimulus. Branchey and colleagues (1988) found that the P300 height (i.e., amplitude) was lower in violent alcoholics (i.e., likely corresponding to type II alcoholics) than in nonviolent alcoholics (i.e., likely corresponding to type I alcoholics). The P300 amplitude also was reduced in people at risk for type II alcoholism, such as sons of type II alcoholics who were not alcohol dependent themselves (Begleiter et al. 1987).

These neurobiological markers previously had been associated with certain stable personality traits. For example, low MAO activity was related to impulsiveness, desire to avoid monotonous tasks, extroversion, and sensationseeking behavior (von Knorring et al. 1987a). Moreover, reduced P300 amplitudes appeared to reflect the subjects’ inability to distinguish between common and uncommon stimuli (Cloninger 1987a). Accordingly, researchers investigated whether reproducible differences in personality traits existed between type I and type II alcoholics.

These analyses identified three heritable personality traits that could describe the prototypical characteristics of both alcoholism subtypes: harm avoidance, novelty seeking, and reward dependence (Cloninger 1987a). The term “harm avoidance” describes whether a person is cautious, apprehensive, pessimistic, and inhibited (i.e., high harm avoidance) or confident, relaxed, optimistic, and uninhibited (i.e., low harm avoidance). People with high novelty-seeking traits are impulsive, exploratory, and distractible, whereas people with low novelty-seeking traits are rigid, reflective, and attentive to detail. Finally, a high reward dependence describes subjects who are eager to help others, emotionally dependent, sentimental, and sensitive to social cues, whereas subjects with low reward dependence are socially detached, emotionally cool, practical, and tough minded.

In initial studies, type I alcoholics frequently exhibited high harm avoidance, low novelty seeking, and high reward dependence (Cloninger 1987a), personality characteristics indicating high levels of anxiety. Type II alcoholics often exhibited a reverse personality profile, with low harm avoidance, high novelty seeking, and low reward dependence. This combination of traits also describes people with antisocial personality disorder (ASPD) (Cloninger 1987b) and is consistent with findings that type II alcoholics frequently suffer from ASPD (Gilligan et al. 1988).

The differences in personality traits between type I and type II alcoholics led to a hypothesis about the underlying motivation for alcohol consumption in the two subtypes (Cloninger 1987a). According to this theory, type I alcoholics experience a late onset of alcoholism because their high harm avoidance trait initially inhibits the initiation and frequency of drinking. After an extended period of socially encouraged drinking (e.g., drinking with coworkers at lunch), the risk of alcoholism increases, because the drinkers experience relief of their anxieties after alcohol consumption. For type II alcoholics, who primarily are characterized by high novelty seeking, alcohol use is motivated by the desire to induce euphoria. This desire, which also may lead to other drug abuse, generally begins during adolescence or early adulthood.

The personality traits of harm avoidance, novelty seeking, and reward dependence likely are inherited independently of each other and are influenced by three brain systems that differ in the neurotransmitters they use (Cloninger 1987a). For example, the brain system for novelty seeking is predominantly influenced by the neurotransmitter dopamine (Cloninger 1987a). Accordingly, people who have a high novelty seeking trait are expected to react strongly to the stimulation of dopamineusing (i.e., dopaminergic) nerve cells. Researchers recently confirmed this hypothesis after finding the predicted correlations between novelty seeking, plasma prolactin levels, and heritable variants of cellular components mediating dopamine’s effects (Cloninger 1996).3 Likewise, the serotonin-using (i.e., serotonergic) nerve cells have complex effects on behavior, including facilitating harm avoidance and social cooperation (a measure of high reward dependence) (Cloninger 1995).

These observations suggest certain patterns of neurotransmitter activity in different alcoholic subtypes. For example, people with antisocial personality traits or type II alcoholism are expected to be uncooperative and to have low serotonergic activity in the CNS. Moreover, these individuals are expected to be high in novelty seeking and, therefore, low in dopaminergic CNS activity. In contrast, type I alcoholics, who typically are high in harm avoidance and reward dependence, are likely to be high in both dopaminergic and serotonergic CNS activity (Cloninger 1995). These predictions reflect the empirical findings that type II alcoholics consistently exhibit high novelty-seeking traits and low cooperativeness; however, their levels of harm avoidance may vary.

Because the personality characteristics are inherited independently of each other, traits such as high harm avoidance and high novelty seeking are not mutually exclusive and can occur in the same person. Accordingly, type I and type II alcoholism are not discrete diseases or separate entities; instead, alcoholism in each person is the manifestation of his or her individual combination of personality traits (Sigvardsson et al. in press). Thus, the type I and type II prototypes only represent the two extremes of a continuous spectrum of manifestations of alcohol abuse.

###

Cloninger, C Robert
Sigvardsson, Soren
Bohman, Michael

Alcohol Health and Research World
Washington
1996