Computer-Based Cognitive Testing

Even two decades ago, computers were seen as having advantages in testing far beyond data storage. ese were said to include data scoring and analysis, analysis of test pro les for diagnostic classi cation, increased reliability, enhanced capacity to generate complex stimuli, greater accuracy and superior time resolution, standardization of stimulus presentation, and ease of administration, which reduces the need for highly skilled personnel (Adams, 1986; Adams & Brown, 1986). Limitations were also recognized, and these were said to include alterations in the patient’s perception of and response to the test when a standardized paper and pencil test is adapted for computerized administration (Adams & Brown, 1986) and di, erential familiarity of patients with computer manipulanda used in some computerized test batteries (Kapur, 1988). Larrabee and Crook (1991) also emphasized the need to thoroughly validate computerized batteries and not assume that a validated paper and pencil test remains valid when administered via computer, the need for More than a decade ago, in the second edition of this book, we described the general merits and limitations of using computers in cognitive testing and went on to describe a number of speci c batteries available to clinicians and researchers (Larrabee & Crook, 1996). In looking back, we see that, like old soldiers, several of the test batteries we described have just faded away, while others have been further re ned and strengthened, and yet others, many others, have emerged and may provide useful new tools in neuropsychological evaluation. Aside from these new tools, though, the decade has illustrated the limitations of computerized cognitive testing. Principally, that neuropsychological diagnostic testing without a neuropsychologist is probably not a great idea. A number of Web sites and test vendors have purported to do just that and have, thankfully, failed to nd acceptance in the marketplace. at is not to say that computerized memory testing via the Internet is not appropriate for preliminary screening, and even repeated testing over time to detect change, but it is not now, and is not likely to become, a substitute for traditional clinical diagnostic evaluation. Moreover, any Internet-based testing should be under the supervision of a quali ed psychologist to minimize the chances of misattribution of results and related iatrogenic factors (Mittenberg et al., 1992; Suhr & Gunstad, 2002). ere are so many new computerized cognitive test batteries that a complete review could occupy this entire volume. ere is also a great 5