Pseudoscience and the Paranormal

response of many psychoanalysts to the refutation of the basis of Freud’s dream theory was to deny that the refutation mattered. One general approach (see LaBruzza 1978; Letters to the Editor 1978) was to claim that mind and brain are different and that knowledge about the function of the brain cannot be used to constrain theories about the nature of cognitive processes. In fact, theories of cognition (of mind, if you will) must be constrained by what is known about the function of the brain. A theory of mind that is inconsistent with what is known about brain function must be wrong. Similarly, a theory of brain function that is inconsistent with what is known about cognitive function is also in need of serious revision. Returning directly to the issue of the validity of psychoanalysis, it is clear that psychoanalytic theory rests on incorrect notions of how the neurons in the brain operate. On those grounds alone, the theory can be rejected.

CLINICAL APPLICATIONS OF PSYCHOANALYSIS

Psychoanalysis was developed to be more than a theory of the mind: It was also developed as a method of treatment for numerous psychological disorders. This section will consider three issues related to the clinical applications of psychoanalysis. The first is the use of psychoanalysts’ clinical experiences as evidence for the validity of psychoanalytic theory. The second is the use of projective tests, because the rationale for these tests rests largely in psychoanalytic theory. The third issue concerns the efficacy of psychoanalytic therapy.

Clinical Experiences As Proof for Psychoanalytic Theory

Psychoanalysts since Freud have argued that the strongest proof of the validity of psychoanalytic theory is to be found in the clinical experiences of actual psychoanalysts. This is the data on which their “science” is founded. These data consist of symbolic interpretations of dreams, free associations, and the like which seem to confirm psychoanalytic theory. The data are taken as strong evidence for the theory. Thus, an analyst might have a male patient whom he believes has unresolved Oedipal feelings. That is, he still has some incestuous sexual desire for his mother. One night the patient dreams that he entered a cozy, familiar home through a window. Further, he was wearing an absurdly large tie at the time. Obviously, such a dream can be interpreted symbolically as a wish to have sexual intercourse with the mother, symbolized by the familiar house. The large tie is a symbolic representation of the erect penis. The serious problems with symbolic interpretations have been discussed earlier in this chapter. The important point here is that interpreting a patient’s dream or free association in such a way that it appears to confirm one’s diagnosis will help convince both the psychoanalyst and the patient that psychoanalytic theory “works.” Numbers of such “correct” interpretations over the years of an analyst’s experience will, understandably, result in a powerful belief that psychoanalytic theory is valid. It is clear, however, that the psychological processes that yield this strong belief in the validity of psychoanalytic theory are the same as the processes that engender strong, but incorrect, beliefs in the validity of the predictions made by psychics, as discussed in chapter 2. The license to interpret symbolically ensures that it will be possible to interpret any dream or action of the patient in a way consistent with the psychoanalyst’s diagnosis. Thus, like the vague psychic predictions that “come true” after the fact, the perceived validity of psychoanalytic theory, at least as far as it is based on data from the clinical situation, is based on the fallacy of personal validation and the P. T. Barnum effect.

Projective Tests and Illusary Correlation

On most projective tests, test takers are presented with some ambiguous stimulus that they must make up a story about or describe. Responses are interpreted, frequently symbolically, and are said to reflect the basic, stable personality characteristics of individuals. The responses can also be used to reveal hidden psychological problems, desires, and anxieties, even if these are lodged in the unconscious. Thus the term projective: the test taker is assumed to project information about personality and any psychological disturbances onto the ambiguous stimulus. The classic example of a projective test is the Rorschach ink blot test (Rorschach 1942), in which test takers are presented with a series of ink blots and asked to tell what each reminds them of. Another type of projective test is one in which test takers produce some nonverbal response. This type of projective test is typified by the Draw-A-Person test in which individuals are simply asked to draw a person. The drawings are then interpreted to reveal personality traits, anxieties, and so forth.

Projective tests are widely used by psychiatrists and psychologists. The important question is whether they actually provide useful information. That is, do they “work” in any objective sense? For a test to work in this sense, it must possess two basic characteristics: it must be both reliable and valid. Briefly, a test is reliable if it gives the same individual close to the same score on two different test administrations. A test is valid if it can be shown to measure what it is claimed to measure. There are numerous ways of measuring reliability and validity that are beyond the scope of this section, but they can be found in any text on psychological testing. Projective tests lack both reliability and validity (Anastasi 1976). Specifically regarding the issue of validity, Anastasi (1988) says, “The accumulation of published studies that have failed to demonstrate any validity for such projective techniques as the Rorschach and the D-A-P (Draw A Person Test) is truly impressive. Yet after five decades of negative results, the status of projective techniques remains substantially unchanged” (p. 621; emphasis in original). Twelve years later Lilienfeld, Wood, and Garb (2000a) published a thorough review of the research on the most widely used projective tests—the Rorschach, the Thematic Apperception Test, and Draw-A- Person techniques. Their conclusion was that these tests had such low validity that they were essentially useless. For a less technical summary of this review, see Lilienfeld, Wood, and Garb (2000b).

If projective techniques are so poor when measured objectively, why are they still so widely used, especially by psychoanalysts? One reason is that in many cases the test results are interpreted symbolically. For example, in an unpublished report it was noted that one test taker “… provided a Rorschach response of cells dividing, in the process of pulling apart, reflecting a lack of separateness and differentiation. Another subject’s sense of incompletion and deficiency was illustrated in many anatomical responses, which indicate concerns over bodily integrity.” The projective test essentially becomes the “gimmick” in a cold reading (see chapter 2), replacing the astrological chart, tea leaves, or palm lines that the storefront cold reader uses. Thus, it will always be possible for projective test users to convince themselves that the test has revealed something about the test taker.

Another, somewhat more subtle, mechanism also works to increase the perceived validity of projective tests. Known as illusory correlation, this mechanism has been extensively studied by Chapman and Chapman (1967, 1969; Chapman 1967). Illusory correlation simply means that the individuals perceive certain variables as co-occurring more frequently than they actually do. Chapman and Chapman (1967) demonstrated this in a series of studies using the projective Draw-A-Person (DAP) test. In these studies, college undergraduates with no knowledge of the DAP were presented with DAP drawings. Each drawing was paired with two statements that, the students were told, described the psychological symptoms of the man who had drawn each drawing. For example, a drawing would be accompanied with a statement such as “The man who drew this (1) is suspicious of other people (2) is worried about how manly he is.” Of course, the statements that were presented with each drawing had nothing to do with the actual symptoms, if any (some of the drawings were done by presumably normal graduate students), of the man who had produced the drawing. The drawings varied on a number of characteristics, several of which clinicians with experience using the DAP believe are signs of particular symptoms. For example, a drawing that has broad shoulders or is muscular or manly is said to be characteristic of a man who is worried about his masculinity. A man worried about his intelligence is said to draw a large or emphasized head. The actual set of drawings and symptom statements used by Chapman and Chapman was specifically constructed such that there was no correlation between any symptoms and any drawing characteristic.