functioning in a rather intuitive fashion, without
much empirical evidence for their validity. How-
ever, all these methods are fallible.
METHODS OF
NEUROPSYCHOLOGICAL
ASSESSMENTAssessment is a complex affair that involves many
issues in addition to those already discussed. Let us
consider a few of the more important ones (Zillmer
et al., 2008).
Major ApproachesShould a standard test or test battery be adminis-
tered to all patients referred for a neuropsychologi-
cal workup, or should the test(s) be chosen on the
basis of clinical judgment, the nature of the referral,
or the clinician’s special skills and proclivities? The
first approach is sometimes termed thestandard bat-
tery approach or fixed battery approach. It has the
advantages of evaluating patients for all basic neuro-
psychological abilities, accumulating a standard
database for all patients over time, and allowing
for the identification of important patterns of scores.
Major disadvantages include the time and expense
involved, the potential for patient fatigue, and the
inflexibility of this approach (in that assessments are
not tailored to individual patients).
The second approach is called theprocess/flexible
approachor thehypothesis-testing approach. Here, each
assessment is tailored to the individual patient, with
the neuropsychologist choosing tests based on her
or his hypotheses about the case. In some cases, a
test may be altered in the way it is administered to
the patient so that additional hypotheses can be
tested. Some argue that the individualized approach
is a sensitive one that capitalizes on the clinician’s
best impressions. Others suggest that if a clinician
picks the wrong test(s), it may result in a poor
assessment. Also, the individualized approach ham-
pers the systematic collection of data from specific
tests on specific kinds of patients. Of course, some
clinicians combine these two strategies by using one
or more standard screening devices and then going
to other specific tests depending on the outcome of
the initial screening.Interpretation of Neuropsychological
Test Results
Zillmer and colleagues (2008) note a number of
ways in which neuropsychologists interpret test
data. First, a patient’s level of performance may be
interpreted in the context of normative data. For
example, does a patient’s score fall significantly
below the mean score for the appropriate reference
group, suggesting some impairment in this area of
functioning? Second, some calculatedifference scores
between two tests for a patient; certain levels of
difference suggest impairment. Third,pathognomonic
signs of brain damage (e.g., failing to draw the left
half of a picture) may be noted and interpreted.
Fourth, apattern analysisof scores may be under-
taken; certain patterns of scores on tests have been
reliably associated with specific neurological injuries
or impairments. Finally, a number of statistical for-
mulas that weight test scores differentially may be
available for certain diagnostic decisions.
In a test of the diagnostic validity of four dif-
ferent approaches to interpreting test data, Ivnik
et al. (2000) compared the absolute score/patho-
gnomonic sign, difference score, and pattern of
scores/profile variability approaches. Interestingly,
the authors found that the absolute score approach
resulted in the best diagnostic accuracy (predicting
cognitive impairment in older adults). This suggests
that cutoff scores on measures can be used instead of
more complex indices like difference scores or score
patterns.
A final point about interpretation has to do
with the desirability of making qualitative evalua-
tions of patients’responses. Should neuropsycholo-
gists depend on qualitative testing or quantitative
methods that reduce the results to numerical values?
Is the way a patient responds the important datum,
or is it the scored responses that are critical? Many
neuropsychologists probably combine the two
approaches, which need not be mutually exclusive.
Whether a patient scores zero points for barelyNEUROPSYCHOLOGY 523