Journal of Clinical and Experimental Neuropsychology 1998, Vol. 20, No. 4, pp. 496-502
1380-3395/98/2004-496$12.00 © Swets & Zeitlinger
Comparison Between a ‘‘Traditional’’ Memory Test and a ‘‘Behavioral’’ Memory Battery in Spanish Patients* Miguel Pérez and Juan Godoy Department of Clinical Psychology, University of Granada, Spain
ABSTRACT The main objective of this study was to compare a ‘‘traditional’’ test of memory, the Wechsler Memory Scale-Revised (WMS-R; Wechsler, 1987) with a behavioral evaluation test, the Rivermead Behavioural Memory Test (RBMT; Wilson, Cockburn, & Baddeley, 1985). Eighty volunteers participated in the study and were divided into four groups: 20 older adults with memory problems, 20 patients with temporal lobe epilepsy, 20 patients with Alzheimer’s disease, and 20 controls. A Spanish translation and adaptation of the WMS-R and a Spanish version of the RBMT were used. The results indicated that the RBMT was as robust as the WMS-R in differential discrimination of patients.
Investigation of everyday memory (EM) is a relatively new area of research which has produced developments at the basic and cognitive levels (Ceci & Bronfenner, 1991; Neisser, 1991) as well as at the clinical level of memory research. At the clinical level, research in EM has been motivated by the development of assessment with more ecologically valid tests as well as an interest in designing programs of memory rehabilitation (Heinrichs, 1990; Sbordone & Long, 1996; Wilson, 1989, 1991, 1993) and treatment outcomes (Crook & Youngjohn, 1993; Larrabee & Crook, 1988). Presently, there are two different approaches to assessing memory problems. One, often called ‘‘traditional’’, focuses on differential diagnosis (Kane, 1991; Lezak, 1995). Of the various instruments for this purpose, the best example is the Wechsler Memory Scale-III (WMS-III; Wechsler, 1997). The second type of assessment focuses on the design of rehabilitation programs. Nevertheless, new
instruments for this type of evaluation are being developed with increasing regularity, such as the Rivermead Behavioural Memory Test (RBMT; Wilson, Cockburn, & Baddeley, 1985) and the Computerized Everyday Memory Battery (CEMB; Larrabee & Crook, 1988). The difference between these two forms of evaluation is relatively small in some regards. First, instruments such as the RBMT and the CEMB have been used in differential diagnosis, indicating an ability to discriminate that is as good as that of the traditional tests (Beardsall & Huppert, 1991; Ivnik et al., 1993). Second, more traditional tests have been used to predict the success of the patient’s return to his community (Heaton, Chelune, & Lehman, 1978; Millis, Rosenthal, & Lourie, 1994; Ryan, Sautter, Capps, Meneese, & Barth, 1992); thus, it is possible and, sometimes suggested, that tests be combined during the evaluation, depending on the interests of the evaluator (Wilson, 1993). Nevertheless, these
* We wish to thank all the individuals who participated and collaborated in this article, including patients and experimenters, as well as Drs. Manuel Henz and Angel Gómez Camello, the neurologists who assisted us in the selection of patients. Also, we would like to thank Gabriel Salazar and Michelle Robinson for their assisstance in the translation of this article as well as Antonio E. Puente, for his commentaries. Address correspondence to: Miguel Pérez, Dpto. de Personalidad, Evaluación y Tratamiento Psicológico, Facultad de Psicología, Universidad de Granada, (18011) Granada, Spain. E-mail: [email protected]
Accepted for publication: April 30, 1998.
new instruments must be shown to be better or similar in evaluating EM than traditional instruments (Kane, 1991). Several studies have compared the two types of batteries, and we will review some of them. Beardsall and Huppert (1991) studied the utility of diverse measures of memory in early differential diagnosis of Alzheimer’s disease (AD). The results indicated that incidental recall of six designs, free recall of the words from the 16word list, immediate and delayed recall of a route, and delayed recall of names and faces were the tasks that best discriminated between clinical groups. This study demonstrated the value of the RBMT as a discriminative tool in the case of Alzheimer-type dementia. In another study, Goldstein, McCue, Rogers, and Nussbaum (1992) demonstrated the validity of the RBMT as a measure of daily life abilities when different measures were compared. They studied the capability of the WMS, the memory scale of the Luria-Nebraska Neuropsychological Battery (M-LNNB; Golden, Hammeke, & Purisch, 1980) and the RBMT to predict abilities of daily functioning as measured by the Performance Assessment of Self-Care Skill (PASS). At three and six months, the pattern of correlations was similar, with the RBMT being the only test in the AD group that correlated in the expected direction, though not significantly (r = –0.65). With regards to the CEMB, Youngjohn, Larrabee, and Crook (1992a) studied test-retest reliability and practice effects for some tasks of the CEMB battery and five traditional neuropsychological tests. Results showed that everyday memory tasks had higher test-retest reliability than the traditional tests (r = 0.64 to 0.83 vs. r = 0.53 to 0.72). In the comparison between tasks with similar structure, all everyday memory tests were equal to or better than the traditional tests, except in attention and concentration (for example, WAIS/ Digit Span). With respect to practice effects, approximately half of the everyday tests yielded a significant practice effect, but all of the traditional tests displayed it. This study showed that everyday tests are equal to or more reliable than traditional tests but with minor practice effects. In another study, Young-
john, Larrabee, and Crook (1992b) compared the CEMB with the subtest of Logical Memory and Learning of Paired Associates of the WMS and the Benton Revised Visual Retention Test (BVRT-R; Benton, 1974) in their ability to discriminate patients with AD from older adults diagnosed with Age-Associated Memory Impairment (AAMI). The results indicated that the CEMB was similar to the traditional tests in its ability to discriminate and classify clinical patients. On the other hand, Ivnik et al. (1993) compared traditional and everyday tests regarding their ability to detect memory impairments after temporal lobectomy of epileptic patients. Results showed that traditional and everyday tests were similarly effective while the last served to generalize daily activities. These promising results of everyday memory tests in comparison with traditional tests are counteracted by others which are negative. Goldstein and Polkey (1992) found that the RBMT failed to differentiate in impairment after two types of surgery in epileptic patients (unilateral temporal lobectomy or amygdalo-hipocampectomy) when 78.3% of these patients had shown memory deterioration when evaluated with the RBMT. In the same study, one traditional subtest (Wechsler Logical Memory) detected differences between type and side (right or left) of operation. In another study, Corcoran and Thompson (1993) reported similar results. In this case, the main objective of the study was to investigate the major causes of memory complaints in epileptic patients. The results suggested that only three traditional tests significantly discriminated the groups. Few studies have compared traditional tests of memory with tests of everyday memory. The results of these studies are generally inconclusive. However, it is critical to note to that these investigations use subtests and not complete batteries. The aim of this study was to compare a traditional test of memory (WMS-R) with a behavioral evaluation test (RBMT) with regards to discriminative abilities using the complete tests and batteries and well-defined clinical samples.
M. PÉREZ AND J. GODOY
METHOD Participants Eighty volunteers participated in the study (age: M = 57.36 years, SD = 16.73, Range = 21–82; years of education: M = 6.88, SD = 2.99, Range = 3–15). They were divided into four groups, with equal numbers of men and woman in each group. The first group consisted of 20 patients diagnosed with AD (age: M = 69.95 years, SD = 7.60, Range = 50–81; years of education: M = 5.30, SD = 3.22, Range = 3–15) who were in the beginning stage of this disorder. All patients were diagnosed with probable AD according to NINCDS-ADRDA criteria (McKhann et al., 1984) after receiving neurological evaluation in the Department of Neurology, Clinical Hospital and Hospital of Traumatology, Granada, and from the Department of Neurology, Clinical Hospital of Jaen. Other causes of dementia were excluded through use of medical history, neurologic examination including brain imaging, and appropiate laboratory tests. The second group consisted of 20 patients diagnosed with Partial Complex Epilepsy (age: M = 34.55 years, SD = 11.803, Range = 21–63; years of education: M = 8.05, SD = 2.85, Range = 3–13). Ten patients had left temporal foci, 8 had right temporal foci, and 2 had bilateral foci. The mean of duration of epilepsy was 24.65 years (SD = 12.61, Range = 7–53), and seizures were pharmacologically controlled. In all cases, it had been 1 week from the last seizure prior to neuropsychological assessment. These patients were obtained from the same hospitals mentioned above. The third group was comprised of 20 older adults (age: M = 71.00 years, SD = 5.24, Range = 64–83; years of education: M = 6.15, SD = 2.77, Range = 3–12) who were residents in the old–age home ‘‘Buenos Aires’’ of Granada. Participants were recruited by placing announcements on the bulletin boards in the retirement facility, where it was stated that if a person had memory problems, they could participate in a study whose objective was to evaluate one’s memory. Finally a group of 20 controls with no memory problems was used (age: M = 53.95 years, SD = 4.05, Range = 47–60; years of education: M = 8.05, SD = 2.21, Range = 6–15). Materials A translation and adaptation of the Wechsler Memory Scale-Revised (WMS-R; Wechsler, 1987) and the Spanish version of the Rivermead Behavioural Memory Test (RBMT; Mozaz, 1991) were used for this study.
Procedures The tests were administered together with two other memory tests, which are not the aim of this study, by four graduate-level testers who had been previously trained. Before the start of this study, interobserver reliability was evaluated and found to be r = 0.905, varying between r = 0.819 and r = 0.996. The order of administration was counterbalanced, in each sample, in such a way that in all the groups all the possible sequences occurred the same number of times. At the same time, each tester evaluated 5 participants from each group, using a distinct counterbalanced sequence with each of the participants. Each tester administered each sequence the same number of times. Assignment of a participant to an observer and to a particular sequence was done randomly. Participants with AD and those with epilepsy were scheduled by telephone. They were also informed of the duration of the tests and transportation arrangements were made. The tests were given in the Department of Psychology at University of Granada, for the participants from Granada, and in the Neurology Clinic of the Clinical Hospital of Jaen for the volunteers from Jaen. In the case of the older adults, the tests were administered in the old age residence ‘‘Buenos Aires’’. The controls were tested in the laboratories of the research group of Human Psychophysiology and Health, University of Granada. After the volunteer arrived, the evaluator explained to him/her the nature of the tests and the number of tests that were going to be performed. These were administered in two parts, corresponding to the first two and the last two in the sequence, in order to reduce the possibility of fatigue. Test duration varied between 1 hr to 1½ hr. The time between administration of the two parts ranged from 1 hr to a few days, depending on the needs of the volunteers.
RESULTS First, two analyses of variance (ANOVAs) were conducted using a unifactorial between-groups design with the variable group as factor over the levels of age and years of education to examine if the groups were equal on these variables. The results showed that there were significant differences regarding age (F(3,76) = 96.94; MCe = 60.27; p < .000) as well as years of education (F(3,76) = 4.93; MCe = 7.79; p < .003) (see Table 1).
Table 1. Demographic Data. Group AD M Age (years) Education (years)
Older Adults (SD)
M 71.00 6.15
(SD) (5.24) (2.77)
Epileptic M 34.55 8.05
(SD) (11.80) (2.85)
Control M 53.95 8.05
Note. AD = Alzheimer’s disease. * p < .01.
Given these differences in age and education, the next step was to obtain the residual score for age and education predicting each score by age, education, and age by education interaction. The residual not predicted by these variables was saved and two separate discriminant function analyses were performed for every battery. With respect to the WMS-R, the results showed that only the first discriminant function was significant (P2(39) = 89.835; p = .000), explaining 81.6% of the variance. The subtests and their standardized coefficients are shown in Table 2. With this discriminant function, the WMS-R correctly classified 70% of AD patients, 65% of older adults, 55% of epileptic patients, 75% of the controls, which results in a
mean of 66.25% (SD = 7.39; see Table 3). Subtests that contributed most to group discrimination were Information and Orientation, Visual Reproduction I, Digit Span, Verbal Paired Associates II, and Visual Reproduction II. Regarding the RBMT, results showed that only the first discriminant function was significant (P2(39) = 97.345; p = .000), explaining 90.1% of the variance. The subtests and their standardized coefficients are shown in Table 4. With this discriminant function, the RBMT correctly classified 80% of AD patients, 75% of older adults, 75% of epileptic patients, 60% of the controls resulting in a mean of 72.5% (SD = 7.5) (see Table 3). Subtests that contributed most to group discrimination were Recall of a
Table 2. Standardized Coefficients for the Significant Discriminant Function of the WMS-R. Subtests
Information and Orientation Mental Control Figural Memory Logical Memory I Visual Pair Associates I Verbal Pair Associates I Visual Reproduction I Digit Span Visual Memory Span Logical Memory II Visual Pair Associates II Verbal Pair Associates II Visual Reproduction II
0.334 -0.248 -0.198 0.252 -0.287 0.002 0.834 0.321 0.161 -0.070 -0.110 0.463 -0.508
Eigenvalue Explained Variance
Note. WMS-R = Wechsler Memory Scale – Revised.
M. PÉREZ AND J. GODOY
Table 3. Correct Classification for the WMS-R and the RBMT. Group Total
WMS – R RBMT
Older Adults %
Note. AD = Alzheimer’s Disease; WMS-R = Wechsler Memory Scale – Revised; RBMT = Rivermead Behavior Memory Test.
Belonging, Immediate Recall of a Story, Delayed Recall of a Story, Face Recognition, and Delayed Recall of a Message. Finally, two cross tables were conducted to test if the differences on the percent of correct classification was significant and to test if there were significant differences of correct classification between the samples of this study for each battery. The results showed that there were no significant differences between the WMS-R versus the RBMT.
DISCUSSION The main objective of this study was to compare a ‘‘traditional’’ test of memory (WMS-R) with a behavioral test (RBMT) with regard to discriminative abilities. The results showed that there were no significant differences in the discriminative ability of the WMS-R and the RBMT. These results support other studies which have found that the behavioral evaluation tests
Table 4. Standardized Coefficients for the Significant Discriminant Function of the RBMT. Subtests
1st Name 2nd Name Belonging Appointment Pictures Immediate Story Delayed Story Faces Immediate Route Delayed Route Immediate Message Delayed Message Orientation
0.119 -0.297 -0.354 -0.217 0.055 -0.586 0.548 -0.521 -0.238 0.029 -0.025 -0.375 0.061
Eigenvalue Explained Variance
Note. RBMT = Rivermead Behavioral Memory Test.
of memory have, at the least, equal ability of discrimination as the traditional evaluation tests of memory (Beardsall & Huppert, 1991; Goldstein et al., 1992; Ivnik et al., 1993; Youngjohn et al., 1992a, 1992b). Additionally, as suggested by Youngjohn et al. (1992b), everyday memory tests were designed to develop programs of memory rehabilitation (Heinrichs, 1990; Wilson, 1989, 1991, 1993), treatment outcome (Crook & Youngjohn, 1993; Larrabee & Crook, 1988), but not differential diagnosis (discriminant validity). However, traditional tests of memory were developed for the express purpose of clinical differentation. In clinical practice, the choice of one test vesus the other depends on the questions asked in the evaluation (Wilson, 1993). This study suggests that, when a differential diagnosis is necessary, either the WMS-R or the RBMT can be used adequately. It is important to note that each one of them performs better in some aspects than the other. For example, the WMS-R mainly supplies information on the nature of the memory deficit, whereas the RBMT supplies information on the functional repercussion of the deficits in the patient’s life. Hence, a combined use may provide a more comprehensive analysis of the memory deficit and its rehabilitation. It is not possible to obtain differential conclusions on the discriminative ability of these batteries in relation to the samples used in this study, given that this study was designed to compare batteries and not populations. In spite of this, the ad hoc analysis to test significant differences in classifying subjects with respect to each sample group showed no significant differences. In conclusion, the present study indicates that there are no differences between the WMS-R and the RBMT in their ability to discriminate subjects in various diagnostic groups, and, thus, both can be used for this purpose. This conclusion is tempered by the specific tests and samples employed as well as a relatively modest participant-to-variable ratio.
REFERENCES Beardsall, L. & Huppert, F.A. (1991). A comparison of clinical psychometric and behavioural memory tests: Findings from a community study of the early detection of dementia. International Journal of Geriatric Psychiatry, 6, 295-306. Benton, A.L. (1974). The Revised Visual Retention Test (4th ed.). New York: Psychological Corporation. Ceci, S., & Bronfenbrenner, U. (1991). On the demise of everyday memory: ‘‘The rumors of my death are much exaggerated’’ (Mark Twain). American Psychologist, 46, 27-31. Corcoran, R., & Thompson, P. (1993). Epilepsy and poor memory: Who complains and what do they mean?. British Journal of Clinical Psychology, 32, 199-208. Crook, T.H., & Youngjohn, J.R. (1993). Development of treatments for memory disorders: The necessary meeting of basic and everyday memory research. Applied Cognitive Psychology, 7, 619-630. Golden, C.J., Hammeke, T.A., & Purisch, A.D. (1980). Diagnostic validity of a standarized neuropsychological battery derived from Luria’s neuropsychological test. Journal of Consulting and Clinical Psychology, 46, 1258-1268. Goldstein, L.H., & Polkey, Ch.E. (1992). Behavioural memory after temporal lobectomy or amygdalohippocampectomy. British Journal of Clinical Psychology, 31, 75-81. Goldstein, G., McCue, M., Rogers, J., & Nussbaum, P.D. (1992). Diagnostic differences in memory test based prediction of functional capacity in the ederly. Neuropsychological Rehabilitation, 2, 307317. Heinrichs, R.W. (1990). Current and emergent applications of neuropsychological assessment: Problems of validity and utility. Professional Psychology: Research and Practice, 21, 171-176. Heaton, R.K., Chelune, G.J., & Lehman, R.A.W. (1978). Using neuropsychological and personality tests to assess the likelihood of patient employment. Journal of Nervous and Mental Disease, 166, 408-416. Ivnik, R.J., Malec, J.F., Sharbrough, F.W., Cascino, G.D., Hirschorn, K.A., Crook, T.H., & Larrabee, G.J. (1993). Traditional and computerized assessment procedures applied to the evaluation of memory change after temporal lobectomy. Archives of Clinical Neuropsychology, 8, 69-81. Kane, R.L. (1991). Standarized and flexible batteries in neuropsychology: An assessment update. Neuropsychological Review, 2, 281-339.
M. PÉREZ AND J. GODOY
Larrabee, G.J., & Crook, T.H. (1988). A computerized everyday memory battery for assessing treatment effects. Psychopharmacology Bulletin, 24, 695697. Lezak, M.D. (1995). Neuropsychological assessment (3rd ed). London: Oxford University Press. Millis, S.R., Rosenthal, M., & Lourie, I.F. (1994). Predicting comunity integration after traumatic brain injury with neuropsychological measures. International Journal of Neuroscience, 79, 165167. McKhann, G., Drachman, D., Folstein, M., Katzman, M., Price, R., & Stadlan, E.M. (1984). Clinical diagnosis of Alzheimer’s disease: Report of the NINCDS-ADRDA work group. Neurology, 34, 939-944. Mozaz, T. (1991). Test Conductual de Memoria Rivermead. Madrid: TEA. Neisser, U. (1991). A case of misplaced nostalgia. American Psychologist, 46, 34-36. Ryan, T.V., Sautter, S.W., Capps, C.F., Meneese, W., & Barth, J.T. (1992). Utilizing neuropsychological measures to predict vocational outcome in a head trauma population. Brain Injury, 6, 175-182. Sbordone, R.J. & Long, Ch.J. (1996). Ecological validity of neuropsychological testing. Delray Beach, FL: GR Press/ St. Lucie Press. Wechsler, D. (1987). Manual for the Wechsler Memory Scale-Revised. San Antonio, TX: The Psychological Corporation.
Wechsler, D. (1997). Manual for the Wechsler Memory Scale-III. San Antonio, TX: The Psychological Corporation. Wilson, B.A. (1989). Designing memory-therapy programs. In L. Poon, D.C. Rubin, & B.A. Wilson (Eds.), Everyday cognition in adulthood and late life (pp. 615-638). Cambridge, UK: Cambridge University Press. Wilson, B.A. (1991). Long-term prognosis of patients with severe memory disorders. Neuropsychological Rehabilitation, 1, 117-134. Wilson, B.A. (1993). Ecological validity of neuropsychological assessment: Do neuropsychological indexes predict performance in everyday activities? Applied and Preventive Psychology, 2, 209-215. Wilson, B.A., Cockburn, J., & Baddeley, A.D. (1985). The Rivermead Behavioural Memory Test. Titchfield, UK: Thames Valley Test Company. Youngjohn, J.R., Larrabee, G.J., & Crook, T.H. (1992a). Test-retest reliability of computerized, everyday memory measures and traditional memory tests. The Clinical Neuropsychologist, 6, 276286. Youngjohn, J.R., Larrabee, G.J., & Crook, T.H. (1992b). Discriminating Age-Associated Memory Impairment from Alzheimer’s Disease. Psychological Assessment, 4, 54-49.