HCT Overview

Overview of the Halstead Category Test

NewCat is a cross-platform (Win/Mac) computer version of the Halstead Category Test (HCT) that greatly simplifies the administration and scoring of this well-researched measure of cognitive impairment. In order to understand the design and advantages of NewCat as a clinical tool, it is important to know about the history of the test.

The HCT was developed by psychologist Ward Halstead in the 1940s as a measure of organic brain damage or dysfunction. Inasmuch as test performance may be compromised by any number of issues — for example, problems associated with attention, memory, visual-spatial perception, the ability to employ a flexible problem-solving strategy in response to feedback, as well as the client’s emotional state and level of cooperation — it is typically no longer employed to localize suspected brain damage. Still, it remains among the most sensitive overall measures of cognitive dysfunction.

By taking into account variables such as age, education, motivation, response pattern, personal history, and performance on multidimensional measures of intellectual ability such as the Wechsler intelligence tests, it can provide important clues to the nature of a client’s impairment. While a core element of the Halstead-Reitan Neuropsychological Test Battery, it may be employed independent of that full battery when evaluating possible cognitive impairment. However, owing to its lack of specificity, it is recommended for clinical use only as part of an evaluation of those factors that can affect performance.

The design of the test is relatively straightforward: The client is instructed that for each subtest the series of one or more geometric figures presented conform to a particular rule, which must be identified by pressing a switch between 1 and 4. (The correct response may be as simple as counting the number of geometric figures displayed on the screen, whereas later subtests require more discrimination.) Feedback is provided that indicates whether or not the chosen response was correct, giving the respondent the opportunity to try out a different rule when presented the next item. For each subtest, the score is the number of errors. Subsequent subtests demand more complex analytical reasoning. The sum of the subtest error scores represents the total test score.

When first introduced as a commercial product, the HCT required a large and rather cumbersome tabletop device. (See my sketch on the right.) A carousel slide projector, located behind the box and controlled by the examiner, displayed the test figures on a screen. The four numbered switches on the front console enabled the client to signal a response. An audible tone — a bell for correct and a buzzer for incorrect — provided feedback.

The design was very clever for its time, but the downside was that the device was not portable, limiting its availability to facilities that could both afford and house its installation.

Author's sketch of the original HCT projection device
Original HCT projection device

One advance in the early 1980s that made the HCT more widely available to clinicians was a booklet version that presented each figure on a separate printed page. Although differing in format from the projection screen and substituting the clinician’s verbal feedback for the bell/buzzer, it is generally accepted that booklet test scores are essentially equivalent to the original test.

The early 1990s saw the first computer versions of the HCT. Unlike the booklet test, these once again displayed the figures on a screen, with the advantage that the presentation of the items, the recording of the client’s keypress responses, and the test feedback (often emulating the original audio tones) were all now entirely automated. Moreover, it became possible to score and save to a computer file a record of the client’s performance.

Research has documented that both booklet and computer versions of the Category Test yield total scores essentially the same as those employed when using the original device. See, for example:

Choca, J., & Morris, J. (1992). Administering the Category Test by computer: Equivalence of results. Clinical Neuropsychologist 6: 9-15.

A detailed history of Halstead Category Test was published by my colleague James Choca around the 50-year anniversary of its introduction, which includes an extensive list of references up to that time:

Choca, J. P., Laatch, L., Wetzel, L., & Agresti, E. (1997). The Halstead Category Test: A Fifty Year Perspective. Neuropsychology Review v7 n2.

One potential drawback of the original Category Test is that it may require up to 40 minutes to administer when working with a cognitively impaired client. As a practical matter, the test probably should be terminated once it becomes evident that the client cannot identify the rules governing the more advanced subtests. However, when employed as a screening tool as part of an evaluation, several short forms are available that correlate highly with the full test result. I’ve included as an option in NewCat the short revised version (RCAT) developed by Russell & Levy, which employs a prorated score that compares very favorably with the full test:

Russell, E. W., & Levy, M. (1987). Revision of the Halstead Category Test. Journal of Consulting and Clinical Psychology 55: 89S-901.

Taylor, D. J., Hunt, C., & Glaser, B. (1990). A crossvalidation of the Revised Category Test. Psychological Assessment: A Journal of Consulting and Clinical Psychology, 2(4), 486–488.

As regards norms and cutoff scores, Halstead originally employed a score of 51 or above as suggestive of cognitive impairment. However, inasmuch as his original normative sample was a bit young—and because some decline is natural with age—one conservative rule of thumb is to increase the cutoff by one point for every year beyond age 51.

More detailed norms — stratified by age, education, etc. — are available in print, although some have been questioned on account of the small size of the subgroups included. Perhaps the best advice is to observe the client’s actual performance — which may provide important clues to the nature of any deficit — and incorporate this into an overall evaluation that takes into account the individual’s personal history and other relevant test results along with any cutoff score.