Clinical and Psychometric Properties of the new WMS-IV
Design Memory Subtest
Holdnack, J.A. PhD1; Drozdick, L.W. PhD1
1Pearson
Clinical Groups Table 1: Immediate Designs by Clinical Group
Introduction Methods The clinical samples were collected as part of the 16
WMS-IV standardization. These groups included but Total
The Wechsler Memory Scale 4th edition (Wechsler, Procedures 14
were not limited to Moderate to Severe Traumatic Content
2009) is the most recent revision of the Wechsler The examinee is shown a page in the stimulus
12
Brain Injury (n=30), Right Temporal Lobectomy Spatial
Memory Scales. One of the primary goals of the book containing a 4 x 4 grid with designs placed
10
(n=15), Math Disorder (n=22), Schizophrenia (n=55), 8
revision from the WMS-III (Wechlser, 1997) was to in different blocks on the grid. There are 4 items
and Autism (n=22). Diagnostic criteria for each 6
improve the assessment of visual memory. Visual having 4, 6, 6, and 8 designs for the examinee to
groups are presented in the WMS-IV Technical and 4
memory functions in the third edition were remember, respectively. The examinee is told to
Interpretive Manual (Wechsler, 2009). An age, 2
assessed by the Faces and Family Pictures remember the designs and the location of the
education, and ethnicity matched normal control 0
subtests. designs. After seeing the stimulus page for 10
sample was derived for each clinical group from the
m
E
I
DX
a
TB
The Faces subtest measures immediate and seconds, the examinee is given a 4 X 4 puzzle grid
ni
TL
tis
standardization sample.
re
h
Au
ht
delayed recognition of faces. While previous and cards with designs on them. The examinee
at
ph
Results
zo
Ri
research has identified face memory as must select the cards with the correct designs
hi
Sc
differentiating right from left temporal lobe and place them in the puzzle grid in their correct
functioning (Milner, 1968), these findings are not location. Reliability Table 2: Delayed Designs by Clinical Group
always replicated (Glogau, Ellring, Elger, & After a 20-30 minute delay, the examinee is given In the normative sample, obtained internal
Helmstaedter, 2004). The Faces subtest exhibited 16
the cards to place in the grid. Following the consistency measures were: Immediate Total Total
14
some psychometric issues including floor delayed recall task, a delayed recognition Content
(.83-.90), Immediate Content (.66-.88), Immediate 12
problems (Holdnack & Delis, 2004), a high guess condition is administered. Scores are computed Spatial
Spatial (.70-.83), Delayed Total (.80-.90), Delayed 10
rate (Levy, 2006), and low communality with other for Total Immediate, Immediate Content, 8
visual memory measures (Millis, Malina, Bowers, Content (.70-.84) and Delayed Spatial (.67-.82).
Immediate Spatial, Total Delayed, Delayed 6
& Ricker, 1999). The Faces subtest appears to Test-Retest correlations for designs were: 4
Content, and Delayed Spatial.
have a very specific application in Immediate Total (.73), Immediate Content (.64), 2
neuropsychological assessment but may not Immediate Spatial (.50), Delayed Total (.72), 0
m
Example of a Designs Item
LE
DX
I
a
TB
Delayed Content (.64) and Delayed Spatial (.50).
ni
work optimally as a general indicator of visual
t is
tT
re
h
Au
at
ph
gh
memory functioning.
M
Correlations with Other Memory
zo
Ri
hi
Few studies have utilized the Family Pictures
Sc
subtest of the WMS-III. This subtest requires the Measures In comparison to normal controls, all the scores in
examinee to respond verbally to visually the clinical samples were significantly lower
presented images. The test measures visual- WMS-III except for Designs Immediate Content for
verbal associative memory (e.g., picture-name, The WMS-IV Designs Immediate Total correlated Schizophrenia patients and Delayed Content for
object-activity) as well as spatial memory (e.g., Schizophrenia, Right TLE and Math Disorder.
r=.35 with WMS-III Immediate Faces. Designs
location of characters in the picture). In order to Effect sizes for TBI ranged from .72 to 1.36 and for
Total Delayed correlated r=.38 with Delayed
purify the visual memory domain, efforts were the Right TLE .16 to 1.66 with most above 1.0 for
Faces. Designs Immediate correlated r=.41 with
made to reduce the impact of verbalization on both samples.
Immediate Family Pictures and Designs Delayed
visual memory tasks in the WMS-IV. Additionally, performance in patients with TBI on
Participants correlated r=.43 with Delayed Family Pictures. Delayed Content was significantly correlated with
The Designs subtest was developed to limit
Controls RBANS caretaker ratings of General Adaptive Functioning
confounding cognitive processes (e.g.,
The sample was comprised of 900 examinees (.63), Community Use (.63), Health and Safety (.63)
verbalization, motor control, and visual-spatial The Designs Total Immediate correlated (r=.38)
ages 16-69 years of age. Exclusionary criteria and Self-Care (.57) as measured on the ABAS-II
processing); however, these factors can never be with RBANS Immediate Memory and Designs
included any history of neurological, psychiatric, (Harrison & Oakland, 2003).
fully eliminated (Heilbronner, 1992). The Designs
Delayed (r=.44) with RBANS Delayed Memory.
subtest assesses memory for visual images developmental or medical condition affecting
within a grid, requiring the examinee to recall both cognitive functioning. Subjects were screened for
Clinical Data Conclusion
visual and spatial information. However, visual general cognitive impairment and poor effort. The
The WMS-IV Designs subtest provides a reliable
objects and location are processed though demographic characteristics of the sample were Performance on the Designs subtests (expressed in assessment of memory for visual details and
different visual systems (Ungerleider, Courtney & matched to 2005 census data for ethnicity and scaled score units) by various clinical groups is spatial location. It correlates low to moderate with
Haxby, 1998) and may be processed differentially education level. Examinees above the age of 70 presented in Tables 1 and 2. Patients diagnosed other visual memory measures. Designs scores
in clinical disorders such as temporal lobe were excluded based on research with previous with Autism, Schizophrenia, and TBI performed in are sensitive to brain injury and right temporal
epilepsy (Chiaravalloti & Glosser, 2004; Hermann, editions of this subtest that showed declining the low average range with scores between 6 and 8 lobe damage with large effect sizes between
Seidenberg, Wyler, & Haltiner, 1993), object and visual and spatial discrimination skills interfered on immediate and delayed Designs scores. The controls and clinical groups. In patients with
spatial memory are both measured and separate with older examinees performance on this Right TLE and Math Disorder groups performed brain injury, Design memory performance is
scores are provided for each visual memory type. subtest. mostly in the low average range but some scores, associated with deficits in adaptive functioning
Finally, recognition trials were added to each especially on delayed recall, were in the average providing useful treatment planning information.
visual memory subtest. range.
Clinical and Psychometric Properties of the New WMS-IV
Visual Working Memory Subtests
Holdnack, J.A. PhD1; Drozdick, L.W. PhD1
1Pearson
The examinee is shown a 4x4 grid with blue and/or The examinee is shown a series of designs of
Introduction red dots on it for 5 seconds. They are told to increasing length for 5 seconds and is then shown a
Results
remember the location of the blue dots and ignore
page with correct designs and foils. They must
The Wechsler Memory Scale 4th edition (Wechsler, any red dots that appear on the page. The Reliability
select the correct designs in the correct order. The
2009) is the most recent revision of the Wechsler examinee is then shown a second page with blue
and/or red dots on it for 5 seconds. The examinee examinee is awarded 2 points for getting the correct Across the normative sample age groups, Spatial
Memory Scale. One of the primary goals of the designs in the proper order and 1 point if they get Addition internal consistencies ranged from .89 to
then adds the two visual images together. The
revision from the WMS-III (Wechsler, 1997) was to examinee is given cards with blue, white, and red .93 and Symbol Span ranged from .76 to .92. Test-
the correct designs but in an incorrect order. There
improve the assessment of working memory. In dots and a 4 X 4 puzzle grid. The examinee must is only a forward condition as previous research retest correlations were .77 for Spatial Addition
the WMS-III, working memory was assessed place a blue dot on the grid in the location where and .72 for Symbol Span.
indicated that examinees would study the stimuli
primarily with Spatial Span and Letter-Number they saw the blue dots on either page and a white
from right to left during a backward condition which
Sequencing. The WMS-III Working Memory Index dot in any location that blue dots appear on both Concurrent Validity
pages, thus subtracting the images. does not require mental reversal of order.
was composed of these two tests, one auditory Spatial Addition and Symbol Span correlated .58
and one visual. Additionally, Letter-Number Example: symbols to be remembered
and .52 with WMS-III Spatial Span, .46 and .47 with
Sequencing appeared in both the WMS-III and Example: Page 1 WAIS-IV Digit Span, and .51 and .45 with WAIS-IV
WAIS-III Working Memory Indexes, a redundant
Arithmetic, respectively. Spatial Addition and
use of this subtest. For the revision, it was
Symbol Span correlated .70 and .47 with WIAT-II
decided to have no shared subtests between
Example: response page Numerical Operations and .65 and .62 with WIAT-II
WAIS-IV/WMS-IV. The WAIS-IV has auditory
Math Reasoning, respectively.
working memory subtests and the WMS-IV has
visual working memory tests. The co-norming and Clinical Studies
separation allows the auditory and visual working Clinical studies found large effect sizes for Spatial
memory indexes to be statistically compared in a Addition in the TBI (d=1.23), RTLE (d=1.00),
common normative sample. Autism (d=1.13) and Math Disorder (d=.80)
Another goal for the WMS-IV working memory groups, while, Symbol Span showed large effects
subtests was to increase the amount of mental Example: Page 2 in Autism (d=1.22) and Schizophrenia (d=1.01). All
manipulation required to complete the tasks. The differences between the controls and patients
Spatial Span subtest of the WMS-III is primarily an were statistically significant and had moderate to
indicator of visual storage, subsequently Spatial Participants large effect sizes. Table 1 provides level of
Addition was developed for the WMS-IV. A second Controls performance on working memory subtests by
visual working memory subtest, Symbol Span, The sample was comprised of 1400 examinees ages clinical group in scaled score units. In TBI
was developed as a visual analog to WAIS-IV Digit 16-90 years for Symbol Span and 900 examinees patients, performance on Spatial Addition related
Span. Symbol Span uses novel visual stimuli ages 16-69 for Spatial Addition. Exclusionary criteria to daily functioning skills including Community
which are difficult to verbalize and taps storage included any history of neurological, psychiatric, Use (r=.53), Self-Direction (r=.49) and Social
and manipulation functions. developmental or medical condition affecting Functioning (r=.52).
The WMS-IV visual working memory subtests cognitive functioning. Subjects were screened for
Example: Answer Table 1: Working Memory Performance by
were designed based on the work of Baddeley general cognitive impairment and poor effort. The Clinical Group
(2000) and Gathercole (2208). These tests demographic characteristics of the sample were
measure storage components of the visual-spatial 16
matched to 2005 census data for ethnicity and Spatial Addition
14
sketchpad. Both tests require mental education level. Examinees above the age of 70 Symbol Span
12
manipulation of visually-presented information were excluded for Spatial Addition based on 10
relating to the functions of the central executive. research with previous editions of this subtest that 8
Additionally, Spatial Addition requires the showed declining spatial discrimination skills 6
examinee to ignore irrelevant information, another interfered with older examinees ability to perform 4
function of the Central Executive (Baddeley, 2000; the task. 2
0
Gathercole, 2008). Clinical Groups
Symbol Span
sm
LE
I
DX
a
TB
ni
tT
re
Methods
t
The clinical samples were collected as part of the
Au
at
ph
gh
Symbol Span assesses storage and manipulation
zo
Ri
WMS-IV standardization. These groups included but
hi
of visual-details in working memory. The subtest
Sc
were not limited to Moderate to Severe Traumatic
Procedures was developed as a visual analog to Digit Span. Conclusions
Brain Injury (n=30), Right Temporal Lobectomy
Symbols were used instead of digits to limit the The new WMS-IV visual working memory
Spatial Addition (n=15), Math Disorder (n=22), Schizophrenia (n=55),
degree that the auditory working memory system subtests, Spatial Addition and Symbol Span
and Autism (n=22). Diagnostic criteria for each
The Spatial Addition subtest assess visual-spatial is invoked during the task. Earlier versions used have good reliability and concurrent validity.
group are presented in the WMS-IV Technical and
storage and manipulation in working memory. The a visual digit span but digits are easily read from The subtests are clinically sensitive yielding
Interpretive Manual (Wechsler, 2009). An age,
subtest is based on a modified n-back paradigm. the visual representation and rehearsed verbally; large effect sizes in patients with known brain
education, and ethnicity matched normal control
while, symbols are difficult to verbalize and injury. The tests relate to academic and daily
sample was derived for each clinical group from the
auditorily rehearse. living skills, indicating ecological validity.
standardization sample.
