See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/239055837

The Behavioral Inattention Test for Unilateral Visual Neglect

Article  in  Physical & Occupational Therapy in Geriatrics · July 1989

DOI: 10.1300/J148v07n03_04

CITATIONS READS

5 9,316

2 authors, including:

Sharon Cermak
University of Southern California
215 PUBLICATIONS   8,125 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Sensory Processing in ASD View project

All content following this page was uploaded by Sharon Cermak on 11 September 2014.

The user has requested enhancement of the downloaded file.

 The Behavioral Inattention Test

for Unilateral Visual Neglect:

A Critical Review

Sharon A. Cermak, EdD, OTR

Joann Hausser, OTR

ABSTRACT. The Behavioral Inattention Test (BIT) which was de­

signed to measure unilateral visual neglect (UVN) consists of two
parts: six conventional subtests which have commonly been used to
detect UVN and nine behavioral subtests which were developed to
simulate activities of daily living. The test was standardized on 50
normal subjects and was administered to 80 eVA patients in Great
Britain. The test appears to have high inter-rater (.99), test-retest
(.99), and parallel from reliability (.91). Validity of the conven­
tional subtests has been examined, but the validity of the behavioral
subtests has not. The test appears to be useful to clinicians because it
is relatively simple to admmister, is standardized, and is intended to
identify problems that patients will demonstrate in daily life. How­
ever to improve its usefulness, further studies should address a nor­
mative sample in the U.S. with greater attention to age and sex dis­
tribution, LQ., and educational level. The relationship between
performance on the behavioral subtests and functional task perfor­
mance needs to be examined.

Occupational and physical therapists frequently treat patients

with unilateral neglect (UN) secondary to evA. The presence of

Sharon A. Cermak is Associate Professor of Occupational Therapy at Sargent

College, Boston University, 1 University Rd., Boston, MA 02215. Joann Hausser
is an Advanced Master's degree student in Occupational Therapy at Sargent Col­
lege, Boston University, and a clinical specialist in the occupational therapy de­
partment at New England Rehabilitation Hospital, 1 Rehabilitation Way, Wo­
burn, MA 01208. .
Test source: Thames Valley Test Company, 34/36 High Street, Titchfield,
Fareham, Hants. P0144AF, England.
Physical & Occupational Therapy in Geriatrics. Vol. 7(3) 1989
© 1989 by The Haworth Press, Inc. All rights reserved. 43
44 PHYSICAL & OCCUPATIONAL THERAPY IN GERIATRICS
.I:
U ...
.. ,.;
UN and related disorders has been reported to negatively impact .•.
L
..
• 11
.... -..
081»10- •
oc
0
c-...
.I:
0'11 0
_.....
L

L _C . . . r.
U U0
•
•L
•C
U
functional outcomes in rehabilitation (Denes, Semenza, Stoppa, & C ... L C._III)(

., .. ·"., ­.
.11'1
• ...ooccn
.. "
:>0 II II
o ­ 11.­.
Lis, 1982; Kinsella & Ford, 1980), therefore it is important that
clinicians assess and report this condition. A wide variety of testing
U

':C
o "'.

.. 0
L
E

..
c-
GJ-
> II 0
-&&.UI >OlD
01
Q. • ...
EM­

. ­ ...
• L L C"
z
.
c
__
_N
o
NO.ll

........... ­
r.-L
L
0
U
II
LX.
0::1.
...
•
.•
E
o
L

"ao 0 0 ­ U .... 11 C Q I I.
procedures have been developed including drawing and copying, -...
.1:.
OIL
.I: U
01"
.I:
01
L.,
-;Oi'~ci • C...
.-c~x
r. ......... 10
• C
>­
...
L

..
-.
L. Ut .... )( "')lE CDO -:I
co "'"
x>u
block designs, line bisection, letter cancellation, computerized key­ ...N' III Cf:) L ... UrJZ OCG

..
0­ L
II OX - L >011"­
0-· . . ..•
"I':-~ .: '
O'OC::E Ill. ... :1'''' C · ."
board tapping, and adaptations of reading, writing, and arithmetic
- C. N
. ·L
11...
It ·.,faL .." "u....r.~ .. ..• D
.. 0 ...... 01­

.z. It"" .. u C ......... C C.

..••
eeoc; " U
tests. (For reviews see Lezak, 1983; Siev, Freishtat, & Zoltan, ~LU ~UI
It_ .11 0­ - ... It
Q.,JI::J'ca.
-UIIJ:I­
0«»­
0.'1) r.
-­
00.
1986; Strub & Black, 1985; and Van Deusen, 1988.) The limita­
CI'
o
U
V1
-...
CD It
-U:::I
C)(
e
-
"C
0 c
Nil
....."
N::E
-EUII-.
oCOISLOJ:
- U - ' O Q.VI
L"
- L 0....
M.oO ... :I
o.E
-U
0
L

tions of these tests are that most are not standardized and a relation­ II
•o
L
ship between performance on the tests and functional performance • U

.· -.
01 • II> II
c ...
.
..•• .... ."•.
has not been clearly established. C
-OL
'0 .I:
II L

The purpose of this paper is to review and critique a new test, the '"0­ )0,11 .. It C 0. ... U .I:
...'"0­ 0'00 :J
':2 .. 01

-" ... ....

• "E

.... .. -.
L ... 0.0"
.I:
Behavioral Inattention Test (BIT) which was developed to measure lD LII 'C
'0 .. 0 I'J""LL
01 U a+' ~ II"
E
"
::l "E C" L
~ i..
-... -..••• . -....
"'0 OC
unilateral visual neglect (UVN) in CVA patients. It consists of a >o L -"'-II>' II>

-.
V1
:aeo Q. .. II
uoo I! ~
..
11-
..J
~~ cu "
battery of six conventional subtests that have traditionally been used ~
.. U L 'H
to assess this condition and nine behavioral subtests which simulate
o
0­
C'o

IIU
- .I:
U
E :>
"L
'0"
" ... -...
'Oc.-
OoCco
c>o..
__ c ... ,,·
....
"c
-c -
UII
:>
'0
-I!
U
C

-..
..
...>
Z • :> IIC LL

.. .
CL --oelt'­ - "0 E
activities of daily living. (Tables 1 and 2 describe these tests.) 1lI­
co ­ ."
The test manual (Wilson, Cockburn, & Halligan, 1987a) reports
z
o
u
-"
.1:­
C
-.I:
11.01
.. :>
uO
C
II"
L­
.)0,.'"
L
.. 0 ....
OU.CUUQ.
c-u
- ... II It c.. Q.
~II -u
L
..
-.
CC
E
II
II>
L
o
..
L

.
U L C L " O+' II II
CII .. .I: -Ill II"" C -II II II.
that the behavioral subtests were selected by psychologists and oc­
cupational therapists because of their testability and their relevance •
,
'"
0"
" . -"
I! '0 . C'CO"
L
OGUC".
1I>:>'o.CO
.. 11. - ..
.-It
0 ...
NO
­
LL
'" .
...•
'0
•U
..
eLU

.-
OLa 'U
oC . )I"
u:n LII _II
D
0­II II
O·
L
.
'00"
C)I" COli) r.
Olt
r. .. L

. - .• .
to everyday life. The authors believe that this assessment has value 001
.,11
"'llIe
C
L'O­
C­
It-a ...
QL L
-aL"o.·.
'" " C
MGI
u
II.
1I.x
ll.U
II. 0
U
0111. ~ o L L
because of its ecological validity, because it has been standardized, C" C.x
011111
C I!
Q.CG IIJ
>-u;~~ E!
..C.;:]-
"..
II 0
CII."
-.I:. -II OU

·
CE •• c ..
and because it addresses the heterogeneous nature of UVN by pro­
....
C"C
II-L·r.Q. "II

viding a battery of subtests. They also suggest that this assessment

.. C­
.. 0
110
- L.I:
II.U ".".'1).
.... II
II"
• 'CIJQ II "L • ..
-II •C
>
51
... .. ... •..
L CIIU .1:" L
C .I:
11.-.
..
C ...
C
!
.­ O.
OllU 0'0 0.1]- 11­ 01
UC .. U. U-.I: C LC U
allows the clinician to identify the types of problems the patient will O. eu • .:>
01 -
.. ..."
_
..... 0
-OlllI!C 01
o "C
co
•
I! L
01:>
-"
.... ..
Z a+'J: 01110
be experiencing in everyday life and to construct a profile of the o .110 "L "LL "."'.00
"11 U J:I 0>­

.... .. "
.... Il."=' 11. .. 11.11.1: ".0 II II
0­ CO
" .. ;:] . - 011 C­ ..
patient's strengths and weaknesses which can form a basis for inter­ Q.
....
a:
....
C-J:
It L'"
L "C
C­ ""
C • ~U&"e2' "u
C:>
11.-."-­ OIL
.. -...
C:I'L
GI
C
-
0
enL"
o ~ 11.'5.2"= ""
U 110 0.
vention. V1
uJ
.. •
L .. )C
OI­
L
•OIL
01'
LIIC :1' ..... - ­
..... OL.'O
ell
LC L
11)'0"
L :>
o.MJ:I
..
•
C Q. 0­ D.'" 0.­ .. 0.­

2' 2'
STANDARDIZATION SAMPLE )0, )I
II
C ~ L

The scores of 50 control subjects between the ages of 19 and 82 ... ..

o C
o
u
~
C

..
.
.....
C C
2
were used to establish the limits of normal performance on the con­
ventional and behavioral subtests. This group of 50 individuals was
01
C
•
.. •
U '"'0 ..
o
U
o

..
C
" U C •"
~
C
composed of hospital employees, community volunteers, and mem­ V1
0­0-1
...
L
U
U
L
C
II
U
•:> m •"
bers of a university subject panel. The BIT was also administered to :m "C . ..
2 L
II
L
•
C
L•
II.
:l
'"
" '""
V1 -
80 patients with initial diagnoses of unilateral CVA admitted to the ..J ..J V1 lL ..J

Rivermead Rehabilitation Centre, Oxford, England, over an 18 45

 Sharon A. Cennak and Joann Hausser 47
II
II ..
.c:~ ~ ~ ~ month period. Of an original pool of 125 patients, 45 were excluded
:; ~: ~.. -: for reasons such as bilateral motor weakness, apraxia, severe men­
=.. ~: : ~ " . ~ '0 : tal impairment, general cognitive deterioration, severe language
~ iii ~ ~~~ ~ ~ ~~ ~ ~ comprehension deficits, or non-CVA etiology. Of the remaining 80
i~ ~
-u
2 ~~
~~~
:
~
§
~.
H 5 ~
~
patients, 26 had left hemisphere lesions and 54 had right hemi­
.. Q ~
II 0
! ~ .. ~
ux~
~ :!c 'D
c~
~:
~
'D
~
sphere lesions.
~J f ~ ~~~ : i : ~; ~ Demographic information on the control group and the right and
~~ i ~ H! ~ ~ § 2' ~ ~ left CVA patient groups is given in the test manual and includes
o.: -~ 0.
l ! . c.. <II 0C ~C ~.c
.. ..
(;C .,~ mean age, age ranges, mean IQ , sex, an d vlsua
' I f'Ie Id delects.
& AI -
Hi; ~H ~ ::~;; ~~ thoughthemeanag~forthecontrolgroup(58years)wassimila.rto
.. 0 0 0 ~~u .. ..: ~~ 'D : D that for the two patIent groups (left CVA = 54.6 years and nght
0­ ... "-E"D"'~.c. 10 It C:J • ... • ...
:til..
• L
~
:J
d ..~
IO~O
., ~ ~ u
..~~.....:J ~~.c.
0" CVA = 57.7 years), the dlstnbutlOn of subtects J
withIn the age
<II ,>:.8e '0 .. HI C II ~ :J 2 ~Vl "" ..' ~" ranges was not reported. The standard deviation for mean age of the
~ ~ U IU L ..... 0­ II
E ~'D ~ : ~g B 8. ~: ~~ .: ~ ~~ normal subjects was larger than the two patient groups, especially
ii:III .. C
.. §II ~.
L
US"0
0 - II
lL C0 .c:;; u
~ "- ~ '0 ..'D the ng'h t CVA group. I n add"" Itlon, there was one rna Ie per two fe-
~ ;~ .c0.
c ..
'D ~ .xe.,
g&c ~
'D
gc ... ~ <II
"0 0."
iii ~;:.. rna es In t e norma contro s, but two rna Ies per one f
I "h I I emale'In the
.. ...c. ~ -u ...00 r S" h h h d"f&
patient groups. Ince researc as s own sex I lerences In ' a num­
:
U .. 0 0­ .. C 0­ •
!:l 2c ~: 2 i ~ 'D.c O!.. ~ :: ~~
~
w
ii =~
"DlJ .....
E
L
~ ~~
-lJ·
~
.....
~:~:
... II
8.
C
~ ~~
E.,
ber of different tasks of visual perception (Caplan, MacPherson, &
..
III i2 ~e 'D 0. 2'~-g .. -~ li'; ~.!' x:; TobIn, 1985; McGlone, 1980), data should have been reported and
~ ue 1 0 . 'nOD ..... CO • LX ~ IOE ..
.... . . . ."i. "2.,~ ~ ii .@ii
'­ <II 0 U!{j
"Doa..c
"W
0>-10
.. E~
.....
analyzed for sex. No data were provided on educational level of
i ; ~ .. e e i '0 B~ ~ ~~:: ~~ i :.. these individuals. Finally, the mean IQ of the patient groups (102),
... ~ II i..
0.:;c 4> ~ ~
H, II
nL"D
~
D
-;: ~
nu
...
~
~ ~ -~ .c ~
Q~ ~
although clearly within the average range, was lower than that of
e- i- U ~
0.. _.. ~ .....
5- f § ::;~.c~.,e
~ .. e.."0. ~.. §., the norma I contro Is (115) •
;­ ~8 ; ~ ~to "D ~~ »0 -­ ~ ~;
aUt L C"'" Oil... II "C ..... "C_' II.£: .""
I! .. i .... - = a.:"D ~x lj" ~~~ ~~ ~ l:e
.. 0
o ..... .c II..
u .. ..... .... • L .. ·
0. ...
.. W
L .t:.
- 0"
D ... 0
.. "
u .. uo
-.,........ I SCQ,orlUG
ft.I.l""
o eo C~. U L~ ~~ .~ ~ L W~
Ei ~ 8~_ i;~
~a.
~_~I~"
• •D 'i. &-oc
• L>'. ~., ~ .. ~
:l e 2 &..... ~_ =~ =~
.:J 0­ -II '" ClL • • •
z
o
un
-G-­ .. • ~ c,,-~ OG.~
&~!!D
"CI
~~
- ......
E
L
The scores of the control subjects were used to establIsh the lIm­
~ :: :: : :~ :2!~ ~ ~ :.. :: iii ~.!! B ~~ its of normal performance and determine cut-off scores for individ­
~;.: i: ~ i: .t"<11~ i­ ~';; i;';~:::
W - ... ualsubtests.Itappearsthatthecut-offscoreforeachsubtestisone
~ :1 :L : :L -I,,¥ :fi
W L _ L ~ L L 0 ... - e 0. L"
:. :i
L eL'D >­
a-~
- 0.
:; ~~ 0 U
. below the lower lImit
pomt . . of the range . . the normal sub­
In which
o ~> 0.­ 0. o.~ _~.~ 0.11 0.0 o.-D O~ • z~

jects performed. For example, the range of scores of the control

a ~ subjects on the line crossing subtest was 35-36, and the cut-off
2' 2' ~ ~ score is reported as 34. Scores were based on performance of the
c ~ 2' ~ ~ § overall control group. No distinction was made as a function of a
~ 0 2': 1? ~ 1? ~ 2' patient's age even though Wilson, Cockburn, and Halligan (1987b)
:;;1; ~ l ~ ;! ~ ;2' ~ ~ reported that all normal c~ntrols under age 65 perfo~med without
~ ~ ! 2; =: ;: H: ~ ~ any errors on all tests while two of the control subjects over 65
~ a: :'!. ~ ~ :'!. ~ 8 ~ 8:' ~ failed two test items.
46 The test manual (Wilson et aI., 1987a) suggests that a diagnosis
48 . rtfYSICAL &: OCCUPATIONAL THERAPY IN GERIATRICS Sharon A. Cermak and Joann Hausser 49

of UVN should be based on the total score ascertained from the two versions was .91. Finally, test-retest reliability was computed
conventional subtests. The total maximum score for these tests is by assessing ten subjects on two separate occasions, approximately
146, and based on the results of the 50 control subjects, a cut-off 15 days apart, yielding a correlation of .99. Although not specifi­
score of 129 was established. Thus an individual score below 129 cally stated in the manual, one assumes that inter-rater and test­
would be considered indicative of UVN. Using this method, of the retest correlations reflect overall scores on the entire test. Reliabili­
80 patients tested, a total of 37.5% of the patients were classified as ties for the individual subtests were not reported. Moreover, it is not
showing neglect. Of the 54 patients with right CVA, 48% were so clear whether the "subjects" on whom reliability data were com­
classified, and of the 26 patients with left CVA, 15% were so clas­ puted were patients (in which case, data were not reported on age,
sified. side of lesion, presence or absence of neglect) or normal controls.
One potential problem with use of this total score is that although Since similar data are reported in Wilson, Cockburn, and Halligan
six conventional subtests contribute to the score, the maximum (1987b), for patients, the reliability data here is assumed to be for
score for each subtest ranges from three to 54. Thus, the weighting patients. If so, reliability for the whole test can be considered quite
of subtests in terms of their contributions to the total score is highly good. The high inter-rater agreement would be expected since there
variable with some subtests having the potential to contribute 15 + are few subjective ratings required on the test. Parallel form and
times more heavily than another. Although specific cut-off scores test-retest reliability is high especially given that variability in per­
are given for the individual subtests, it is the total score for which formance frequently characterizes patients with central nervous sys­
reliability and validity data are reported, and it is the total score tem damage.
which is recommended to diagnose UVN. (It is interesting to note
that each of the behavioral subtests has a maximum score of nine,
thus the weighting of each to the total behavioral score is much VALIDITY
more comparable.)
The authors recommend that the behavioral subtests be used to Very limited data on validity are reported in the test manual
indicate the type of everyday problems the patient is likely to expe­ (Wilson et aI., 1987a). According to the authors of the test, validity
rience and to guide therapists in the selection of tasks for remedia­ was established two ways. First, the relationship between total
tion. Because Wilson et al. (1987a) do not endorse the use of the scores on the nine subtests in the behavioral battery and total scores
behavioral subtests to diagnose UVN, they give no cut-off scores on the conventional subtests was examined for all 80 patients (r =
for this part of the test. However, the manual indicates the range of .92). Because the conventional subtests were selected from previ­
normal scores, so that one may infer that a score which falls below ously published studies and are considered valid measures of UVN
this is abnormal or questionable. (Wilson, Cockburn, & Halligan, 1987b), the strong correlation be­
tween them and the behavioral subtests help establish concurrent,
criterion-related validity. The second method of validation identi­
RELIABILITY fied by the test's authors compared the behavioral scores for each
patient with the responses to a short questionnaire completed by the
According to the test manual (Wilson et aI., 1987a) inter-rater relevant therapist at the time of the assessment (r = .67). No infor­
reliability was established by having 13 subjects scored indepen­ mation describing or validating this questionnaire is reported.
dently and simultaneously by two raters, yielding a correlation of Additional data on validity of the test can be found in a recently
.99. Parallel form reliability for the behavioral subtests was calcu­ published article by Wilson, Cockburn, and Halligan (1987b). In
lated by administering two alternative versions of the test to ten that article, the test is referred to as the Rivermead Bchavioral Inat­
subjects, in a counterbalanced manner. The correlation betwcen the tention Test (RBlT) and appears to be an earlier version of the be­

~---
 50 • "YSICAL & OCCUPATIONAL THERAPY IN GERIATRICS Sharon A. Cermak and Joann Hausser 51

havioral components of the BIT. Although this test (RBIT) uses whether this is a test of unilateral neglect or a test of inattention.
behavioral subtests very similar to those on the BIT, it employs a The test manual describes this test as "an instrument for measuring
simpler pass/fail scoring system. The authors report that when a unilateral visual neglect which is characterized by a failure to re­
sample of 28 eVA (20 right and eight left) patients was examined, spond to objects or situations in the space contralateral to a cerebral
55% of the right eVA sample was classified as having inattention, lesion" (Wilson et aI., 1987a, p. 4). However, the test is called the
and 37% of the left eVA sample was classified as having inatten­ "Behavioral Inattention Test" and, although the examiner is en­
tion. Wilson et al. (1987b) state that this figure is consistent with couraged to record the location of the errors, the scoring does not
incidence of neglect for right eVA but is high for left eVA and take into account the side on which the error was made, viz.
believe it may reflect the heavy reliance on language in the RBIT. whether or not it was contralateral to the side of the lesion. Errors or
This finding of consistency in the percentage of the right eVA pop­ omissions from the right, left, or center are not differentiated in the
ulation diagnosed as having UVN by the RBIT and the known inci­ scoring, and all are counted equally.
dence of neglect among this population provides a type of criterion­ A final concern about validity deals with the fact that the authors
related validity, the criterion being the known incidence of neglect suggest that "the behavioral subtests be used (a) to indicate the type
in right eVA. of everyday problems likely to occur because of UVN, and (b) to
An intercorrelation matrix between the behavioral subtests (the guide therapists in the selection of tasks to work on in rehabilita­
RBIT) and six conventional tests of inattention (similar to but not tion" (Wilson et aI., 1987a, p. 9). Although validation has been
exactly the same as the conventional tests in the present version of done against scores on other conventional tests of neglect and
the test) showed that positive and significant correlations (r = .58 against chart records, it has not been done against actual perfor­
to .87) were found between the RBIT and five of the six conven­ mance. Thus, one still does not know if the test does indeed relate
tional tests of inattention. (One test had a lower correlation of r = to actual functional performance in everyday situations. Acker
.34.) Again, these correlations between accepted measures of ne­ (1986) emphasizes the importance of examining this relationship.
glect and the behavioral subtests serve to establish concurrent valid­
ity for the behavioral measures. Patients were also grouped into
categories of inattention and noninattention according to their SUMMARY AND CONCLUSIONS
scores on the RBIT. Independent t-tests indicated significant differ­
ences between the RBIT inattention and noninattention groups on Although we have critiqued this test, we want to also commend
the authors for developing a test of this nature and to state that we
four of the six conventional tests. In fact, the use of multiple t-tests
within a single research investigation is a misapplication of this feel it holds exceptional promise for use by occupational therapists.
As with any new test, validity data is scant and additional validity
statistical procedure. It produces an increased probability of a Type studies need to be performed. Anastasi (1988) points out that estab­
I error, i.e., rejecting a null hypothesis when it should have been
supported (Ottenbacher, 1983). The test's authors also examined lishing the validity of a test is an ongoing process. The authors of
. patient records to see whether those patients diagnosed as having the BIT have begun to address this area. It is now up to occupa­
neglect on the RBIT were the same as those diagnosed as having tional therapists and others using this test to continue the effort. One
neglect by their doctors and/or therapists. There was an 82% agree­ critical area to investigate is the relationship between performance
ment between the records and the RBIT. on the behavioral subtests of the BIT and actual performance on
similar tasks in everyday life. An additional area to consider is that
The major issue in any examination of validity is the question­ of the normative sample. A sample of 50 subjects between the ages
Does the test assess what it purports to assess? With regard to the of 19 and 82 is quite meager. Moreover, the sample is British, and
BIT, it is not clear from the manual exactly what the authors intend educational data are not reported. We need to examine normal indi­
the test to measure. Specifically, there is not a clear distinction as to viduals in this country, considering age, race, sex, socioeconomic
52 ,dtSICAL &: OCCUPATIONAL THERAPY IN GERIATRICS Sharon A. Cennak and Joann Hausser 53

level, and educational level and determine which variables are criti~ spatial abilities exist? A multilevel critique with new data. American Psycholo.
cal in establishing cut-off scores. This process will enable us to gist, 40, 786-799.
Denes, G., Semenza, C., Stoppa, E., & Lis, A. (1982). Unilateral spatial neglect
either support the use of the norms from Wilson et at. or to recom­ and recovery from hemiplegia. Brain, 105, 543-552.
mend new cut-off scores for an American sample. Additionally, the Kinsella, G., & Ford, B. (1980). Acute recovery patterns in stroke. The Medical
patients in the Wilson et al. (1987b) study were assessed when they Journal ofAustralia, 2, 663-666.
entered the rehabilitation center, approximately two months post­ Lezak, M. (1983). Neuropsychological assessment (2nd ed.). New York: Oxford
stroke. Many of our patients enter a rehabilitation program two to University Press.
four weeks postonset. Since neglect is shown to clear rapidly within McGlone, J. (1980). Sex differences in human brain asymmetry: A critical sur­
vey. The Behavioral and Brain Sciences, 3, 215-263.
the first six months postonset, it is important to examine the vari­ Ottenbacher, K. (1983). A "tempest" over t-tests. The American Journal of Oc­
able of time post-CVA on performance on the tests and in terms of cupational Therapy, 37,700-702.
prediction of functional performance outcome. Siev, E., Freishtat, B., & Zollan, B. (1986). Perceptual and cognitive dysfunc­
Clinicians will find the BIT relatively easy to administer. Both tion in the adult stroke patient. New Jersey: Slack.
parts can usually be completed in one hour. The behavioral subtests Strub, R., & Black, F. (1985). The mental status examination in neurology (2nd
ed.). Philadelphia: F.A. Davis.
have a British cultural bias and may present some difficulties to Van Deusen, J. (1988). Unilateral neglect: Suggestions for research byoccupa­
patients from other cultures. Administration of the test to language­ tional therapists. The American Journal of Occupational Therapy, 42, 441­
impaired patients will require adaptations which are suggested in 448.
the manual and may not be altogether satisfactory. Wilson, B., Cockburn, J., & Halligan, P. (1987a). Behavioral inattention test
In summary, this is a unique test of neglect which deserves con­ manual. Fareham, Hants, England: Thames Valley Test Co.
Wilson, B., Cockburn, J., & Halligan, P. (1987b). Devetopment of a behavioral
sideration by occupational therapists. Although limitations have test of visuospatial neglect. Archives of Physical Medicine and Rehabilitation,
been noted, this is the only published test of neglect which attempts 68,98-102.
to examine the importance of functional task performance. Thus,
rather than discarding the test because of its limitations, we encour­
age occupational therapists to utilize and supplement the psychome­
tric data of the test in order to more fully explore its diagnostic and
remedial utility.
A strength of the BIT is its intent to provide assistance in under­
standing the specific problems that patients will encounter in every­
day life. With this stated goal, the test uses what the authors con­
sider ecologically valid tasks which simulate everyday activities.
This test has been validated against traditional measures of neglect;
however, validation is needed using daily performance as the crite­
rion measure in order to truly examine its ecological validity.

REFERENCES
Acker, M.B. (1986). Relationships between test scores and everyday life func­
tioning. In B.P. Uzzell and Y. Gross (Eds.). Clinical neuropsychology of in­
tervention (pp. 85-118). Boston: Martinus Nijhoff Publ ishing.
Anastasi, A. (1988). Psychological testing (6th ed.). New York: MacMillan.
Caplan, P., MacPherson, G., & Tobin, P. (1985). Do sex-related differences in

View publication stats