Clinical neuropsychology is a specialized field of endeavor which seeks to apply the knowledge of human brain-behavior relationships to clinical problems. Human brain-behavior relationships refer to the study of research-derived associations between an individual's behavior, both normal and abnormal, and the functioning of his or her brain. The clinical neuropsychologist takes extensive measurements of a variety of kinds of human behavior, including receptive and expressive language, problem-solving skills, reasoning and conceptualization abilities, learning, memory, perceptual-motor skills, etc. From this complex and detailed set of behavioral measurements, a variety of inferences can be drawn relating directly to the functioning of an individual's brain. In clinical neuropsychology, the operation and condition of an individual's brain is assessed by taking measures of his or her intellectual, emotional and sensory-motor functioning.

In studying brain functioning by measuring behavior, the clinical neuropsychologist makes use of a specialized set of tools which is appropriately labeled the clinical neuropsychological evaluation. This instrument is generally composed of numerous psychological and neuropsychological procedures which measure various abilities and skills. Some of these procedures are drawn from psychology (WAIS-R, Form Board in TPT) and others have been developed specifically from neuropsychological research (Category Test, Speech Sounds Perception Test, etc.). These strictly neuropsychological procedures compose the greater part of the evaluation, especially since they were developed specifically to assess brain functioning by measuring higher mental abilities. Still other procedures in the evaluation were borrowed directly from neurology (certain items on Aphasia Screening; Sensory Perceptual Examination) and were standardized in their administration. Some of the procedures in the evaluation are rather homogeneous in that they depend on mainly one ability or skill for success or failure (Finger Oscillation Test primarily relies on motor tapping speed). Other procedures are more heterogeneous and depend on the organized and complex interaction of several distinct skills or abilities for success (Tactual Performance Test - tactile perceptual ability; appreciation of two-dimensional space; planning and sequencing ability; etc.). In all, the clinical neuropsychological evaluation gives the practitioner in this field a wealth of information about an individual's unique pattern of skills and abilities.

The clinical neuropsychological evaluation has essentially two main purposes: one involving diagnosis and the other involving behavioral description. The diagnostic power of a neuropsychological instrument, such as the Halstead-Reitan Battery, has been well documented and need not be discussed in detail (Vega and Parsons, 1967; Filskov and Goldstein, 1974; Reitan and Davison, 1974). In neuropsychological diagnosis, the presence or absence of impairments in brain functioning can be determined along with other important factors, such as lateralization, localization, severity, acuteness, chronicity or progressivity, and type of impairment suspected of being present (tumor, stroke, closed head injury, etc.). Four primary methods of inference are utilized in making these determinations, namely, level of performance, pathognomonic sign, comparison of the two sides of the body and specific patterns of test scores.

The level of performance approach primarily involves determining how well or how poorly an individual performs on a certain task, usually by means of a numerical score. Cut-off scores are generally developed for such a task, which allow the practitioner to classify an individual as either impaired or unimpaired with respect to brain functioning, depending upon whether his score falls above or below the cut-off value in use. The Halstead Category Test provides an example of this level of performance approach. On this procedure, a score of 51 errors or above places an individual in the impaired range. Likewise, a score of 50 errors or below places the individual in the normal range generally characteristic of individuals with unimpaired brain functioning. The primary danger of using level of performance measures alone to diagnose brain dysfunction is that of classification errors. In most cases, the cut-off score will not completely separate individuals with brain dysfunction from those without. Therefore, both false-positive and false-negative errors can be expected, depending upon the particular cut-off score established. Such a procedure in fact used in isolation is tantamount to employing single tests to diagnose "brain damage, and this approach has been justly criticized in previous work (Reitan and Davison, 1974). Additional methods of inference are used in neuropsychological assessment in order to sharpen diagnosis and minimize errors.

The pathognomonic sign approach essentially involves identifying certain signs (or specific types of deficient performance) which are always associated with brain dysfunction whenever they occur. An example of such a pathognomonic sign would be an instance of dysnomia on Aphasia Screening made by an individual with a college degree and normal IQ values. Such an individual would not be expected to say "spoon" when shown a picture of a fork and asked to name this object. The appearance of a true pathognomonic sign in a neuropsychological evaluation can always be associated with some sort of impairment in brain functioning. However, the converse is not true. That is, the absence of various pathognomonic signs in a particular individual's record does not mean that this individual is free of brain dysfunction. Thus, using, the pathognomonic sign approach alone, one runs a considerable risk of making a false-negative error or discounting the presence of brain dysfunction when it in fact does exist. If other methods of inference are employed with this approach, however, then the likelihood is increased that any brain dysfunction present will be identified even in the absence of pathognomonic signs. Therefore, one may again see the value of and necessity for multiple and complimentary methods of inference in clinical neuropsychology.

The third method of inference involves a comparison of the performances of the two sides of the body. This method was borrowed in principle almost directly from clinical neurology but involves measurement of a variety of sensory, motor and perceptual-motor performances on the two sides of the body and comparing these measures with respect to their relative efficiency. Since each cerebral hemisphere governs (more or less) the contralateral side of the body, some idea of the functional condition of each hemisphere relative to the other can be gleaned from measuring the performance efficiency of each side of the body. An example here is the Finger Oscillation Test. Here, tapping speed in the dominant hand is compared with tapping speed in the non-dominant hand. If certain expected relationships are not obtained, then inferences with respect to the functional efficiency of one hemisphere or the other can be made. This inferential approach provides important corroborative and complementary information, especially with respect to lateralization and localization of brain dysfunction.