Reminiscence therapy is a direct outgrowth of the life review hypothesis (Butler 1980). This therapy consists of having the individual reflect on their life. Through this reflection or reminiscence individuals are postulated to resolve conflict, deal with past painful experiences, and thus be better able to deal with the present. A reorganization of the personality is thought to occur.

As with any new therapy, a large amount of the information concerning the effectiveness of reminiscence therapy is in the form of case reports or anecdotal data (for example, Kaminsky, 1984). Some experimental and quasi-experimental data has been collected, however. This will be presented below.

Havighurst and Glasser (1972) conducted a study examining the relationship of frequency or reminiscence to personal-social adjustment. Three hundred subjects (150 men; 150 women) were asked to complete a questionnaire concerning amount of reminiscence, quality of reminiscence, attitudes and life-style, and personality and perception characteristics of the respondent. The researchers found the following three characteristics to be highly correlated: high reminiscence, positive affect of reminiscence, and good personal-social adjustment. It should be remembered, however, that this is a correlational study. Therefore, no cause and effect statements can be made. It is uncertain whether good personal adjustment leads to high reminiscence, high reminiscence leads to good personal adjustment, or whether they are related by some third variable, such as health.

Boylin, Gordon, and Nehrke (1976) examined the relationship of reminiscing and ego integrity in 41 elderly institutionalized males. The amount of each subject's reminiscing was determined by having the subjects complete part of the Havighurst and Glasser (1972) questionnaire. Ego integrity was measured using a structured interview and was based upon Erikson's (1959) theory of adult development. Results demonstrated a high positive correlation between reminiscence and ego integrity. The authors concluded that the reminiscence in the sample was in the form of the life-review. This study represents another example of correlational data and thus, the results should be viewed as exploratory.

Lewis (1971) examined the relationship of reminiscence to stress adaptation. Twenty-four men were selected to participate in the study. These men were designated as either a high reminiscer or a low reminiscer. Group placement was based upon the number of references each man made to past events (5 years or more in the past) during a nondirective interview. Following group designation, each subject was placed in a socially stressful situation. The hypothesis was that high reminiscers would experience a greater consistency of self-concept than low reminiscers. This hypothesis was supported. The authors conclude that reminiscence may be a factor in successful aging. As this study is quasi-experimental the results should be viewed with caution.

Fallot (1980) conducted an experimental intervention study on the use of reminiscence as a therapeutic tool. Thirty-six women, of various ages, participated in two one-hour therapy sessions. These two sessions consisted of a reminiscence session and a nonreminiscence session. These were counterbalanced so that one-half of the women received the reminiscence session first and the other half-received the nonreminiscence session first. Self-rating of mood were made before and after each session. Results demonstrated increased positive mood following the reminiscence session. These results support the hypothesis that reminiscence may be adaptive. It should be noted that no differences were found between various ages. Therefore, the therapeutic nature of reminiscence may not be age-specific.

Using clinical populations, Liton and Olstern (1969) and Hellebrandt (1978) found that reminiscence therapy improved self-respect and self-awareness in individuals who had been diagnosed as suffering from senile dementia of the Alzheimer1s type. Also, McMahon and Rhudick (1967) examined the relationship of depression and reminiscence. The study found that nondepressed subjects reminisced with greater frequency than depressed subjects. These studies demonstrate that reminiscence may also be of value to a clinical population.

Kaminsky (1978) proposes that the use of reminiscence may serve functions other than those proposed by Butler (1963). He proposes four primary uses for reminiscence therapy. First, reminiscence may be used as a defense mechanism that allays anxiety and maintains self-esteem. The use of denial is important to this function of reminiscence. for example, an individual experiencing intellectual decline may reminisce upon times of intellectual integrity, thus reinforcing that individual1s sense of intellectual wholeness. Second, reminiscence may provide a means of communication between individuals and thus further interpersonal relationships. For example, reminiscence can function as storytelling or the passing on of the oral history as a legacy. Third, reminiscence can aid individuals in dealing with personal loss, grief, or depression. For example, an individual may not feel comfortable directly expressing anger towards the death of a significant other, but may be able to express such emotions through reminiscence. The final use that Kaminsky (1978) proposes for reminiscence is for the process of the life-review and the reintegration of the personality. Kaminsky (1978) demonstrates these four reminiscence functions by means of presented case studies.

The studies presented above all provide support for the use of reminiscence as a therapeutic tool. It should be noted, however, that the majority of the research is quasi-experimental. Therefore, causality can not be conclusively determined. Reminiscence may lead to positive functioning or positive functioning may lead to reminiscence or they could be related to some third variable such as health or cognitive functioning.

One study failed to find a relationship between reminiscence and improved self-esteem or decreased depression. Perrotta and Meacham (1981) randomly assigned subjects to one of three groups:

1. A treatment group that received the reminiscing intervention (n-7);

2. A control group that received a current life events treatment (n=7);

3. a no-treatment control group (n=7).

Subjects were given a pretest and a post test for self esteem and depression. All three groups met for a period of five weeks. The analysis consisted of a 3 (conditions) by 2 (pretest/post test) analysis of variance. No significant main effects or interactions were found. The authors conclude: "This study provides no support for claims that reminiscing can be effective short-term therapeutic intervention" (p. 29).

In short, the above study does not support the use of reminiscence as a therapeutic tool. This study is particularly important as it is experimentally designed and has relatively high internal validity. One problem with the study however is the small sample size; each group contained only seven subjects. Therefore, the study should be replicated with a larger sample.