A Problem-Solving
Version of the Allen
Cognitive Level Test
Naomi ]osman, Noomi Katz
Key Words: adolescence. cognitive
development (assessment) • psychiatry. tests,
by title, Allen Cognitive Level Test
The purpose of this study was to construct and validate a problem-solving version of the existing ALLen
Cognitive Level Test (ACL) (ALLen, 1985). The new problem-solving version of the ACL (ACL-PS) foLLows the
theoretical developments of the cognitive disability theory and the information processing approach. It was
constructed to proVide a more accurate assessment of
the problem-solving process as weLL as task performance, especial~y at the higher cognitive levels. Both
tests were administered to a psychiatric adolescent
group (n = 49) who were subdivided according to diagnosis and to a matched nondysfunctional control
group (n = 29).
The results showed that both the ACL and the ACLPS dzfferentiated significantly between the patients
and the control subjects and among the patient
groups. At Level 6 of the ACL, none of the subjects needed any demonstration, with aLL scores distributed between independent peJionnance or perfonnance follOWing verbal instructions only, that is,
problem-solving phases that were added with the ACLPS The scoring of the ACL-PS is provided in addition
to detailed scoring of tbe cognitive levels. It is suggested tbat the ACL-PS be implemented as a clinical evaluative tool with adolescents.
Naomi Josman, MA, OTR, is a donoral srudem in the Occupational Therapy Department, SEHNAP Education Building, 11th
Floor, New York University, 35 West Fourth Street, New York,
New York 10003. At the time of this study, she was an Instructor in the Occupational Therapy Department, Tel Aviv UniverSity, Israel.
Noomi Katz, PhD, OTR. is Associate Professor, School of Occupational Therapy, Hebrew University, Jerusalem, Israel.
This article was accepted JOT publication July 8, 1990.
T
he purpose of this study was to construct and validate a problem-solVing version of the existing Allen
Cognitive Level Test (ACL) (Allen, 1985). The new
version was developed to enable more accurate evaluation of planning, problem solving, and deduction abilities
with the ACL.
Problem solVing is a process by which one (a) recognizes a problem, (b) defines the problem, (c) identifies
alternative plans, (d) selects a plan, (e) organizes steps in
a plan, and (f) implements a plan and evaluates the outcome (American Occupational Therapy Association,
1989). Problem solving is defined as the transformation of
an initial state into a goal state by a set of operations. To
accomplish this, problem-solving behavior involves the
interaction between (a) the information processing system (the problem solver), (b) the task environment, and
(c) the problem space.
In approaching a task, the problem solver (the information processing system) represents the situation in
terms of a problem space, which is his or her perception
of the task environment (Greeno, 1978). The term problem refers to a situation in which a person is called on to
perform a task not previously encountered. In addition,
externally prOVided instructions do not comprehensively
specify the mode of solution. Thus, the particular task is
considered novel for the person, although available processes or knowledge may be summoned for a solution
(Resnick & Glaser, 1976). Cohen (1983) categorized
these processes into four stages of information processing: (a) problem representation, (b) selection of operators, (c) implementation of selected operators, and (d)
evaluation of current state.
Problem solving is thus conceived as "a complex
form of cognitive activity that is engaged in to achieve a
goal that cannot be reached directly or automatically"
(Szekeres, Ylvisaker, & Cohen, 1987, p. 99).
Information processing encompasses all of the mental (i.e., cognitive) processes that underlie human action
(Mayer & Stone, 1988). It explains "the way man collects,
stores, modifies and interprets environmental information or information already stored internally ... how he
adds information to his permanent knowledge of the
world, how he accesses it again, and how he uses his
knowledge in every facet of human activity" (Lachman,
Lachman, & Butterfield, 1979, p. 6). Cognition has been
defined as the "processing of information for particular
purposes and within certain mental structures" (Dodd &
White, 1980, p. 8), including components of structure,
systems or processes, and functional integrative performance (Szekeres et aI., 1987).
Allen (1987) used an information-processing-system
model to explain how this process gUides voluntary motor actions in human activities. According to this model,
sensory cues are the input to the system; this input is
influenced by the disability. Sensorimotor associations
are the throughput or the internal process of the system;
The American Journal oj Occupational Therapy
Downloaded From: http://ajot.aota.org/ on 07/31/2017 Terms of Use: http://AOTA.org/terms
331
this throughput is influenced also by the purpose and
meaning the person attributes to the information processed. The output observed is a voluntary motor action
that is guided by the specific activity performed. This
cence, and formal operational thinking is the major concern and interest of adolescence. Because problem-solving ability influences treatment, this ability should be
assessed at the initial evaluation.
functional output provides feedback, which enters as new
In a previous study, when a psychiatric adolescent
sensory cues into 'the person's information processing
system. In contrast to most models, Allen's model portrays an information processing system that does no.t use
language as the main mediator of cognitive processes.
Mayer and Stone (1988) defined Allen's Cognitive
Disability Model as a behavioral hierarchy in which a relationship among mental diseases; functional capabilities,
and cognitive capacities is presented. According to these
researchers, the quality of a person's information processing system (cognition) in the brain can be inferred
.from observing that person's task behavior. The degree
of thought used to gUide different functional states is
arranged into 6Ieve]s, ranging from profoundly impaired
(Levell) to normative behavior (Level 6), as assessed
with the ACL
Mayer (1988b) analyzed the cognitive disability theory as representative of the information processing approach. She used this approach to conceptualize planning and problem-solving abilities, incorporating Cohen's
(1983) analysis to expand the theoretical and empirical
database of the Cognitive Disability Model. Mayer also
established congruent validity of the ACL-Expanded (Allen, 1988) as a measure of cognition, by correlating it with
the Wechsler Adult Intelligence Scale-Revised (WAIS-R)
(Wechsler, 1981). She found that the fluid information
processing abilities defined as the perceptual organization factor in the WAIS-R were significantly correlated
with the ACL, as was the distractibility factor. Furthermore, in an analysis of variance, these two factors made
the strongest contribution to ACL performance, which
suggests that fluid or perceptual integrative abilities and
not necessarily verbal information processes are central
to this task performance and, by extrapolation, are central
to routine task performance. This is in keeping with AlIen's (1985) theoretical approach.
Mayer (1988a) further compared information processing with the cognitive levels. In her analysis, she included the attributes used by Allen (1985) to describe the
levels, adding social consequences elaborated on by Allen
and Allen (1987) with treatment indicators in the form of
assistance needed.
Our major concern in this study was problem-solving
abilities. As part of higher cognitive functioning, wefocused our attention on Allen's cognitive Levels 5 and 6
and centered the investigation on adolescents'
performance.
. In the treatment of adolescent mental health patients, clinicians' major emphasis is on improving problem-solving abilities. The rationale underlying this approach is taken from Piagetian theory, which states that
problem-solving capability is fully attained only in adoles-
group and a matched nondysfunctional control group
were tested with the ACL, significant differences between
the two groups were found (t = 2.71, P < .01) (Katz,
]osman, & Steinmetz, 1988). The patients' ACL scores
ranged from 2 to 6: half of them scored 5, but the other
half were divided between low levels and the highest
level, 6. All subjects in the control group scored 5 or 6.
These results are also supported by the present study and
suggest that both nondysfunctional and mentally ill adolescents score higher on the ACL than do many nondysfunctional and mentally ill adults (Allen, 1985; Katz et aI.,
1988; Katz & Heimann, in press). Further investigation of
the use of the ACL for this population, therefore, is
needed.
Moreover, it is questionable whether the ACL accurately evaluates cognitive Level 6. Allen termed Level 5
exploratory actions and Level 6 planned actions, which
implies problem-solving ability. As stated by Katz (1988),
the requirement of the ACL is lower than the normative
performance described at Level 6, due to several discrepant aspects in the validity of the measurement of this
level:
332
1. Level 6 tasks do not require attention to symbolic
cues.
2. Level 6 tasks do not require independent
planning.
3. The task demand is to imitate, which is unnecessary at Level 6.
4. Deductive reasoning is not evaluated in Level 6
tasks.
5. There are patients who perform at Level 6 on the
ACL but who seem to have cognitive difficulties
in planning, problem solving, and paying attention to such symbolic cues as verbal or written
instructions.
Tolchinsky-Landsmann and KatZ (1988) argued that
the analysis of cognitive levels should be extended to
include higher-order cognitive structures. Their findings
suggested that when tested on classic Piagetian cognitive
tasks, the adult psychiatric outpatient population retained some higher level abstraction and deductive abilities but lost lower functions. This validates Allen's (1985)
focus on lower-level structures but also suggests that the
ACL may have a ceiling effect. (Namely, it shows a Type II
error, in which we decide that a patient has no disability
[Level 6], while in fact the patient has undetected cognitive deficits.)
To exemplify the points raised above, the following
case of a 16-year-old girl with anorexia nervosa is reported. The patient described her functioning for the 6
Downloaded From: http://ajot.aota.org/ on 07/31/2017 Terms of Use: http://AOTA.org/terms
April 1991, Volume 45, Number 4
months before admission as intact. Social isolation started when she began losing weight. Her ACL score at admission was Level 6, and her score on the new problemsolving version of the ACL (ACL-PS) was 16 (see the
Appendix). On the basis of this result, the patient was
asked in treatment to choose a task by herself from available suggestions in a book and to perform that task according to the written instructions. She chose a patchwork and said that she understood the instructions.
When she started working, however, she could not figure
out how to plan and put together the various cloth pieces.
The therapist provided verbal explanations, which were
not helpful. Only when a demonstration was given could
the patient grasp the idea and continue the task by herself. The score of 6 on the ACL was not accurate in predicting this patient's functional ability, whereas the score
of 16 on the ACL-PS indicated that she was able to use
covert processes of problem solving, but only after receiving an initial demonstration of the process. Using this
scoring procedure, the therapist would not ask the patient to start the task without showing her at least once
how to do it. Clinical problems like this are seen frequently in patients who are still unable to plan or identify solutions to more complex or unfamiliar tasks, even though
they score at Level 6.
Our objective in the present study was to investigate
the validity of the ACL-PS in differentiating among adolescent patients' performance according to diagnoses, as
well as to differentiate their performance from that of a
nondysfunctional control group. A further aim was to
obtain information concerning adolescent psychiatric patients' planning and problem-solving abilities
The ACL-PS: Rationale
We used a modified version of the ACL, the ACL-PS,
which was designed to examine the process whereby a
patient reaches his or her test result. The ACL-PS does
not change the test materials or the task itself; instead, it
influences the sequence of instructions given, the observation of performance, and the ultimate scoring of the
test. It does not excessively increase test duration. In this
way the test became more process oriented, but still did
not require verbalization of the process and was therefore
not used as a comprehensive dynamic assessment, as
suggested by Lidz-Schneider (1987).
The following sequence of performance, arranged
hierarchically according to level of solution, was examined (the performance may be with or without a problemsolVing method): (a) independent problem solving (highest level), which is defined as the ability to perform the
task by oneself without any instruction or additional cues
except for the existing end result, which has to be repeated; (b) problem solving follOWing verbal instructions; and
(c) performance by imitation (lowest level).
At each level of the problem-solving method, the
therapist is able to differentiate between types of per-
formance on the basis of overt processes of trial and error
or on covert processes in which mental images are used
to solve the problem. An overt tria/-and-error process is
defined as "an empirical method of establishing a satisfactory solution to a problem for which there is no existing
or conveniently applicable theory, consisting of repeating
experimental trials of various hypotheses until error is
sufficiently reduced or eliminated" (Morris, 1976, p.
1369). In this process, "the relationship between an exploratory action and its effect on material objects is established after the action has been performed" (Allen, 1985,
p. 38) and "errors are understood after the action has
been performed" (p. 56). Overt trial and error, therefore,
is related also to an inductive reasoning process that goes
from the specific to the generaL Inductive reasoning is "a
principle of reasoning to a conclusion about all the members of a class from examination of only a few members of
the class" (Morris, 1976, p. 671). It consists of reasoning
from the particular to the general.
Allen (1985) conceptualized the process of overt trial
and error as part of cognitive Level 5 abiJity. In the modified ACL-PS version, we suggest that at Levels 3, 4,5, and
6. this process can be seen and differentiated from covert
processes. At Level 5, however, trial and error expands to
incorporate the ability for new learning that does not
occur at the lower levels.
Covert processes are defined as "bodily events which
are not readily observable without the use of eqUipment
or apparatus to extend the scope of our senses" (McGuigan, 1978, p. 463). Allen (1985), however, defined in the
same manner covert trial-and-error problem solving, a
concept which is inherently contradictory because, as defined above, an empirical action is essential in trial and
error. "The relationship between anticipated action and
its effect on matenal objects is established before the
action is performed. Several possibilities can be considered, and many errors can be avoided" (Allen, 1985, p.
38) On the basis of this definition, we assume that Allen
refers also to covert processes, which use images to tcst
solutions to problems.
Covert processes together with deductive reasoning
are used in new learning and higher cognitive levels. Deductive reasoning is defined as going from the general to
the specific. "The general classes of motor actions are
used to generate hypothetical possibilities. One selects
the best possibility for an unknown situation. The possibility selected is a planned or designed mode of action"
(Allen, 1985, p. 38). Deduction is "the process of reasoning in which a conclusion follows necessarily from the
stated premises: inference by reasoning from the general
to the specific" (Morris, 1976, p. 344).
According to Allen (1985), cognitive Level 6 is exemplified by the abilities of coven processes and deductive
reasoning to solve problems; we therefore suggest a way
to test these processes with the ACL-PS that the ACL does
not allow. Moreover, the evaluation of trial-and-error ver-
The American journal of Occupational Therapl'
Downloaded From: http://ajot.aota.org/ on 07/31/2017 Terms of Use: http://AOTA.org/terms
333
sus covert processes in relation to familiar or new tasks
has significant implications for treatment planning in occupational therapy. Thus, the use of the ACL-PS contributes additional essential information for the evaluation of
the cognitive levels, which further reflect functional
performance.
Method
Subjects
Seventy-eight subjects between the ages of 12 and 18
years participated in the study. The sample consisted of
two groups: (a) 49 adolescents with mental illness who
were admitted to a psychiatric hospital during a 6-month
period and (b) a nondysfunctional control group comprising 29 adolescents proportionally matched with the
research group for age, sex, education, and place of residence. From the original sample of 80 subjects, 1 subject
from each group was excluded because of insufficient
demographic data.
The sample was almost evenly divided between male
(n = 36) and female (n = 42) subjects. Of the patients,
24 were male; 25, female. Of the control subjects, 12 were
male; 17, female. All were high-school students. Seventysix subjects were born in Israel, 1 in Europe, and 1 in
Africa. Approximately 65% of the subjects were from the
older group (16-18 years), 21% from the middle group
(14-15 years), and 13% from the youngest group (12-13
years) .
The patient group was divided into two subgroups
according to diagnosis, with use of the Diagnostic and
Statistical Manual ofMental Disorders (3rd ed.) (American Psychiatric Association, 1980), as follows: (a) schizophrenia or brief reactive psychosis (cognitive disabilities
are assumed to be part of these diagnoses) (n = 19) and
(b) conduct disorders, personality disorders, anorexia
nervosa, affective disorders, and those under observation
(cognitive disabilities were not assumed to be necessarily
part of the patients' personality or behavioral disorders)
(n = 30). Twenty-five patients received antipsychotic
medications, 5 received tranquilizers, and 19 received no
medication.
Instrument
A problem-solVing version of the ACL, the ACL-PS, was
developed for this study. The testing and scoring procedures of the ACL were altered. Each subject initially received the leather lacing with the three different completed stitches-the running stitch (PSI), the whip stitch
(PS2), and the single cordovan stitch (PS3). In the original
ACL, for each stitch, verbal instruction with demonstration is provided. In the ACL-PS, for each stitch, the subject was first asked by the tester to independently repeat
the stitch twice. Inability to perform the task independently elicited from the tester Allen's (1985) standardized
verbal instructions as to how the stitch was to be made.
Further inability to complete the task led the tester to
demonstrate the stitch, and the subject was then asked to
imitate it (the imitation task is that of the original ACL).
This three-step process was repeated, if necessary, for
each stitch.
For each stitch, therefore, there are three different
input modes. The highest level of problem solving is independent performance of the task, in which the input
given is the end result (i.e., the three sample completed
stitches), with no further instructions necessary as to the
process. The subject is thus required to represent the
problem internaJJy and to process the information to
achieve the desired outcome, which is stitches identical
to the input samples.
In the next level of problem solving, the subject must
perform the task after being given two forms of input:
verbal instructions of the process and the end result. The
subject is thus required to process the additional auditory
verbal input to reach the desired output. At the lowest
level of performance, the subject is given verbal and visual
input of both the process and the end result. He or she is
thus required to imitate the demonstrated process of the
task performance.
It is assumed that when a subject is unable to solve
the task independently, each additional process input
provided adds to the subject's ability to perform the task.
Performance on the ACL-PS was scored according to the
hierarchical sequence of solution. Scores range between
2 and 20. The subject's final score would thus reflect not
only his or her ability or inability to complete the task, but
also the process of problem solVing (see the research
version in Table 1 and the analogous refined cognitive
level).
Procedure
The subjects with mental illness were evaluated within
the first week of admission with a battery of five instruments, including the ACL-PS. The study was presented to
the patient as part of the procedure of acceptance to the
occupational therapy department. The same investigator
(the first author) evaluated all subjects. The ACL-PS version took approximately 15 minutes to administer. Each
subject received scores on the ACL-PS and the equivalent
score on the original ACL. Demographic data such as age,
sex, years of education, country of origin, diagnosis, and
medications were collected from the patient's chart. The
control group (n = 29) was tested in the same manner by
the same investigator.
Data Analysis
A Spearman correlation coefficient for each score was
performed between the groups, and a Mann Whitney U
test was used to compare the control group and the two
patient subgroups. A Kendall tau correlation was used to
334
Downloaded From: http://ajot.aota.org/ on 07/31/2017 Terms of Use: http://AOTA.org/terms
Aprit 1991, votume 45, Number 4
Table 1
Frequencies, Means, and Standard Deviations of the Allen Cognitive Level Testa (ACL) and the Allen Cognitive Level
Test-Problem Solving (PS) for the Patient and Control Groups
Patients (n = 49)
Variable
ACL
PSI (Running stitch)
PS2 (Whip stitch)
PS3 (Single cordovan stitch)
Value
Frequency
%
3
4
5
6
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
1
13
12
23
1
1
3
2
22
20
2
0
4
1
20
22
12
0
7
3
10
3
2.0
26.5
24.5
469
2.0
2.0
6.1
4.1
449
409
4.1
00
82
20
40.8
44.9
24.5
0.0
143
61
20.4
6.1
Control Group (n = 29)
M
SD
Frequency
%
114
SD
5.2
0.89
000
1.20
7.9
0.58
131
1.20
14.0
018
17.2
2.50
00
0.0
00
100
0.0
00
33
00
33
93.4
00
0.0
00
0.0
3.4
965
0.0
0.0
267
24.5
41.0
7.7
6.0
7.1
0
0
0
29
0
0
1
0
1
27
0
0
0
0
1
28
0
0
8
18.3
095
7
12
2
"(Allen, 1985).
scudy the relationship of performance and the demographic variables.
P
ul
No significant correlations were found between the ACL
or ACL-PS scores and demographic variables of age and
sex with the use of the Kendall tau correlation.
Frequencies, means, and standard deviations on the
ACL and ACL-PS for all subjects are presented in Table 1.
All subjects in the control group scored 6 on the ACL
About half the subjects in the patient group scored at
Level 6 and the other half scored either at Level 4 or Level
5. On the ACL-PS, the performance of the control group
ranged from 17 to 20, indicating either independent perfonnance or performance with verbal instructions.
Among the patient group, 3'5 (71%) reached the PS3 levels, namely 5 or 6. Twelve of these subjects (24.5%) performed at Level '5, but the remaining 23 (47%) performed
the stitch independently or with verbal instruction. This
distribution is similar to that of the control group, thus
suggesting that all subjects who scored at Level 6, whether patients or control subjects, were able to problemsolve without demonstration.
A mean comparison shows higher scores for the
control group on all pans of the test. The Mann Whitney
U test between the patients and control group on the ACL
and ACL-PS significantly differentiated berween the
groups at the level of .0001 for all four scores (ACL, PSI,
PS2, PS3).
Frequencies, means, and Standard deviations for the
twO patient subgroups are shown in Table 2. Only 4 pa-
tients (21%) of Subgrou p 1 scored 6 on the ACL, compared with 19 (63%) of Subgroup 2. Three subjects from
Suhgroup 1 scored 5, compared with 9 subjects in Subgroup 2, and 12 subjects in Subgroup 1 scored 4, compared with only 1 subject in Subgroup 2. The pattern on
the ACL-PS at PS3 indicates that even the few patients
from Subgroup 1 who performed at Level 6 were able to
problem-solve without demonstration. A mean comparison between the patient subgroups shows higher scores
for Subgroup 2 on all pans of the test. A Mann Whitney U
test significantly differentiated between the subgroups at
p < .0003 for ACL,p < .018 for PSI, andp < .0006 for PS3.
Only PS2 failed to reach significance in differentiating
between the patient groups.
With the use of Kendall tau correlations berween the
ACL and ACL-PS and medications, significant relationships were obtained (ACL, r = .57,p < .0001; PSI, r =
A3,P < .0001; PS2, r = .39,p < .0001; PS3, r = Al,p <
.0001). These results confirm the relationship found between the diagnoses and the test result, that is, an adolescent diagnosed as schizophrenic usually received antipsychotic medications. Patients who received a higher
dosage of antipsychotic medication did worse on the test.
The higher correlation of medications with the ACL stems
perhaps from the larger sample, because it includes all
subjects (n = 49), compared with only a partial group
who reached PS3 (n = 35).
Intercorrelations between the four different test
scores with the Kendall tau correlation were calculated
for the combined groups. A strong correlation was found
between the ACL and the PS3 scores (r = .78,p < ,0001)
The American Journal of Occupalional Therapy
Downloaded From: http://ajot.aota.org/ on 07/31/2017 Terms of Use: http://AOTA.org/terms
335
Table 2
Frequencies, Means, and Standard Deviations of the Allen Cognitive Level Testa (ACL) and the Allen Cognitive Level
Test-Problem Solving (PS) for the Two SUbgroups of Patients
Subgroup 1 (n
V;lri;lble
ACL
PSI (Running stitch)
PS2 (Whip stitch)
PS3 (Single cmdovan stitch)
= 19)
Subgroup 2 (n
m
V;llue
Frequency
%
M
3
4
5
6
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
0
12
3
4
1
00
6:31
ISH
21.1
53
53
10.5
105
42.1
265
53
0.0
15.R
0.0
,)68
42.1
15H
0.0
5.3
0,0
105
53
4.5
08:3
65
150
I
2
2
H
5
1
0
3
0
7
8
.'3
0
1
0
2
1
Frequency
9£
AI
SD
1
1
9
19
0
0
1
:3'3
3:3
300
6:3.4
0.0
0.0
33
0.0
46.7
50.0
33
0.0
3.:3
33
433
46.7
300
00
200
100
267
6.7
55
073
7.4
067
13.2
120
17.1
2.00
a
lAO
128
14
15
1
0
1
157
2.40
= 30)
1
13
14
9
a
6
3
8
2
Nu/e. Diagnoses for Subgroup 1 were schizophrenia and brief reaClive p,ychosi" for Subgroup 2, conduct disorders, personality disorclers, anorexia
m:rvosa, affective disorclers, and observation.
"(Allen, 1985).
and between the PSI and PS2 scores (r = .65,p < .0001).
Lower correlations are seen for the ACL and the PSI score
(r = .43,p < .0001) and the ACL and the PS2 score (r =
.48,p < .0001). As expected, overall significant intercorrelations between the scores exist, because the scores are
interdependent.
Discussion
Both the ACL and ACL-PS significantly differentiated between the adolescents with mental illness and the nondysfunctional control group as well as between the two
subgroups of patients. A~ in previous studies, Subgroup
1, which consisted mainly of persons with schizophrenia,
performed significantly lower, mostly at Level 4, whereas
Subgroup 2 performed at Levels 5 and 6. Thus, the ACL~
PS was found to be valid, similar to the original ACL, in
differentiating hetween two known groups. The ACL-PS,
however, provides additional information. Inspecting the
frequencies of the patient group, one can see that in PSI
and PS2, 85% of subjects performed the running and whip
stitch independently, which suggests that there is no reason to start with a demonstration in each case, as in the
original ACL's instructions. Still, the patients that performed independently were divided into twO groups:
those who performed the stitch by trial and error and
those who used covert processes to perform the task.
This suggests two ways of processing the information
with the same input, even at Levels 3 and 4 and also for
patients who function at Level 6 (see Table 2). This find-
ing has direct clinical implications for the treatment process. The demands the therapist sets for the patient and
the kind of input provided should be planned
accordingly.
All of the nondysfunctional adolescent subjects and
47% of the patients performed at Level 6 on the original
ACL, but their performance on the ACL-PS indicates a
difference in the process of performing the single cordovan stitch. None of these subjects, however, needed any
demonstration.
These findings support the postulate that the ACL
requirement is too low for the attributes described in
cognitive Level 6 (Allen, 1985), but show that among the
subjects who performed at this level, differences existed
in their ability to solve a problem. Performing the task
independently may indicate more precisely the attributes
comprising Level 6, such as attention to symbolic cues,
independent planning, and deductive reasoning.
Here too, implications for treatment are direct and
important, because the evaluation provides the therapist
with information on the kind of problem-solVing processes the patient is using, thus enabling a better match
of the task demands. Referring back to the case illustration in the introduction, the patient was performing at
Level 5.2, according to the suggested refinement of the
levels, which may indicate more accurately her performance at the time of testing. Interestingly, none of the
subjects in this study received a score of 16 (see Tables 1
and 2), suggesting that scores 15 and 17 indicate Level 5
336
Downloaded From: http://ajot.aota.org/ on 07/31/2017 Terms of Use: http://AOTA.org/terms
April 1991. Volume 45, Number 4
performance more accurately with the use of trial-anderror processes with at least one process input in addition to the end result.
Hence, the major additions of the ACL-PS are in
providing three modes of input hierarchically organized
first, a visual end result of the task without any instruction
of the process; second, additional verbal instruction of
the process; and third, verbal instruction with demonstration of the process. In this way, the task difficulty was
generally increased, and the level of demands appears to
more closely match the attributes of the higher cognitive
levels without losing the evaluation of the lower levels.
Furthermore, the evaluation of processes of problem
solving in all parts of the test in addition to a measure of
task performance allows refined inferences about the person's competence and performance.
On the basis of the results of this study, we recommend use of the ACL-PS with the scoring suggested in the
Appendix. In this way, the appropriate cognitive level is
clearer. including structure and process components, and
a hierarchy of problem solving is proVided. This breakdown of the levels differs fmm the modification in the
ACL-Expanded (Allen, 1988), which aims to refine the
precision of the original scoring based on an analysis of
errors occurring during the task performance. The relationship of both modifications should be studied.
We also recommend that the validity of the ACL-PS
be studied in various ropul:nions, especially concerning
functional performance and the interpretation of the relationship between the distribution of scores at Levels ')
and 6.
Although the ACL-PS follows the theoretical development of the cognitive disability frame of reference anel
the information processing approach, it is theoretically
based on Piagetian cognitive development theory. Information processing is not a developmental or a stage model, as is Piaget's, and unlike Piaget's cognitive theolY, it
focuses on the study of adult cognitive processes. However, taken together, information processing and Piagetian
theory describe cognition over the life span. Continued
work on the theoretical knowledge base and its clinical
implications is warranted. A
AppendLx
Stitches and Their Operational Definitions on the Allen
Cognitive Level Testa-Problem SolVing
Score Level
Running stitch
Unable to imitate the running stitch.
Ahle to imitate using trial and error, tWO demonstratiuns,
two stitches'
Able to imitate using covert processes, one
demonstration, twO stitches'
Able to perform with verhal instructions, using trial and
e'Tor, two stitches.
Able to perform with verbal instructions, using covert
processes, two stitches
2.0
3
3.1
4
3.2
5
3.3
6
3.4
Able to perform independently, using trial and error, two
stitches.
Able to perform independently, using covert processes,
two stitches.
Whip stitch
Able to imitate using trial and error, twO demonstrations,
two stitches.
Able to imitate using covert processes, one
demonstration, [WO stitches'
Able to perform with verbal instructions, using trial and
error, tWO stitches.
Able to perform with verbal instructions, using covert
processes, two stitches.
Able to perform independently, using trial and error,
two stitches.
Able to perform independently, using covert processes,
two stitches.
Single cordovan stitch
Able to imitate using trial and error, twO demonstrations,
two stitches.
Able to imitate using covert processes, one
demonstratiun, two stitches'
Able to perfurm wilh verbal instructions, using trial and
t:rror. twO slilches'
Able to perform with verbal instructions, using covert
processes, twO stitches.
Able to perform independently. using trial and error,
two stitches.
Ahle to perfurm independently, using covert processes,
t\\O stitches.
7
35
B
36
9
41
10
4.2
11
43
12
4,4
13
45
14
4.6
15
5.1
16
52
17
53
18
61
19
6.2
20
6.3
No/e. Scurers are instructed to circle the appropriate score for each
stitch and LO indicate the highest level.
<'(Allen, 1985).
Ik(erenl.'l:~
Allen. C. K. (19H'j). Occupalional Iherapyfor psychialnc
diseases: .V/easurellleni and management oFcognilive disabililies. Boston Little. Bl'Own
AJlen, C. K. (1987). 1987 Eleanor Clarke Slagle I.ectureActivitv Occupational therapy's treatment method. American
Journal o( Occupalional 7herapy, 41, '563-575.
Allen. C. K. (1988). Cognitive disabilities. In S. C. RoberLSon
(Ed.), ,\lemal heallh FOCUS· Skills for assessmenl and treatmenl (pp. 3-18-5-33). Rockville, MD: American Occupational
Therapv Association.
Allen. C K, & Allen, R. E. (198 7 ). Cognitive disabilities:
Measuring the social consequences of mental disorders. }our7!ol ol Cliuleal P\ychialr)', 48, 181-19l.
American Occupational Therapy Association (1989). Uniform terminology for occupational therapy- second edition.
Alllerican .Ioumal of Occupalional Therapv, 43, 808-81 '5.
American Psychiatric Association. (1980). Diagnosllc and
Slallslical manual of menial disorders (3rd ed.). Washington,
DC Author
Cohen, G (1983). Tbe pSl'cholopy olcognltion New York:
Ac«clemic P,-ess
Dodd. D.. & White, R. M. Jr. (1980). Cognition Menial
Siruciures aud processes. Boston: Allyn & Bacon.
Greeno, J. O. (1978). Natures of problem solving abilities.
[n W. K. bres (Ed.), Handhook olleaming and cognitive
pl'ocesses Vol. .5. Human inFormation processing (pp. 239267). Hillsdale, NJ. EribauI11.
Katz, N. (1988). ACL-expanded: Test results Paper presenred at the 68th Annual Conference of the Anlerican Occupational
Therapy Association, Phoen ix.
KatZ, N., & Heimann, N. (in press). Review on research
conducted in Israel on cognitive disability instrumentation. Occupational Therapy in ivlental Health.
Katz, N., Josman, N., & Steinmetz, N. (1988). Relationship
Tbe American journal 0,/ OccupCllional TiJerapJ'
Downloaded From: http://ajot.aota.org/ on 07/31/2017 Terms of Use: http://AOTA.org/terms
337
between cognitive disability theory and the human occupation
model in the assessment of psychiatric and nonpsychiatric adolescents. Occupational Therapy in Mental Health, 8, 31-43.
Lachman, R., Lachman, J.. & Butterfield, E. (1979). Cogni-
tive psychology and information processing: An introduction.
Schwartzberg (Eds.), Case simulations in psychosocial occupational therapy (pr. 83-116). Philadelphia: F. A. Davis.
McGuigan, F. J. (1978). Cognitive psychophysiology, prinCiples of covert hehavior. Englewood Cliffs, NJ: Prentice-Hall.
Morris, W. (Ed.). (1976). American heritage dictionary.
Hillsdale, NJ: Erlbaum.
Lidz-Schneider, C. (Ed.). (1987). Dynamic assessment. An
international approach to evaluating learning potential. New
York: Guilford Press.
Mayer, M. A. (1988a). Analysis and synthesis of infonnation processing and cognitive disabili~y theory. Paper presented at the 68th Annual Conference of the American Occupational
Therapy Association, Phoenix.
Mayer, M A. (l988b). Analysis of information processing
and cognitive disability theory. American journal of Occupational Therapy, 42, 176-183.
Mayer, M. A., & Stone, E. L. (1988). Aprlication of a cognitive disability model. In L. W. Duncombe, M. C. Howe, & S. L.
Boston: Houghton Mifflin.
Resnick, L. B., & Glaser, A. (1976). The nature of intelligence. Hillsdale, NJ: Erlbaum
Szekeres, S. F., Ylvisaker, M., & Cohen, S. B. (1987). A frame
work for cognitive rehabilitation therapy. In M. Ylvisaker &
E. M. R. Gobble (Eds.), Communi~y re-entry for head injured
adults (pp. 87-136). Boston: Little, Brown.
Tolchinsky-Landsmann, L., & Katz, N. (1988). Concrete to
formal thinking: Comparison of psychiatric outpatients and a
normal control group. Occupational Therapy in Mental
Health, 8, 73-94.
Wechsler, D. (1981). WechsLer AduLt Intelligence ScaleRevised [Manual]. Cleveland: Psychological Corporation.
Home Rehabilitation Exercises
4th International Multidisciplinary Conference
September 27-29, 1991 • University of Toronto
Especially for
therapists working
with clients to
regain full range
of motion . ..
Keynote Speakers:
Hamilton Hall MD, FRCS(C)
John McM. Mannell MD, Cert. PM
Scott Haldeman DC, MD, FRCPC
Other Speakers Include:
Billie Cussons RN; Patricia Dryden RMT; Gill Faclier
MD; Michael Feuerstein PhD; Chan Gunn MD; Anne
Hartley CAT(c); David Imrie MD; Elizabeth Kaegi MD;
Arthur Laxton DO; Stephen Levin MD; Karel Lewit MD;
Stuart McGill PhD; Jennifer Moffett PT; Howard Vernon D.C.
Registration Fees: Registration postmarked by:
(In US Dollars)
MD, DO, DC, DMD, NO
RPT, RMT, CAT(c)
Student, Intern, Resident
(*does
Mar. 15
Jul. 15
Au. 15
$300
$325
$350
$260
$285
$310
$175*
$200*
$225*
not include social functions)
cancellation and refund policy: before June 15, a $35
administration fee will be charged; before August 15, 50%
refund. No refunds will be available after August 15, 1991For registration and more information contact:
Marc. J. White, Executive Director
Physical Medicine Research Foundation
510-207 West Hastings Street
Vancouver, B.C., V6B 1H7 Canada
Phone (604) 684-4148 Fax (604) 684-6247
338
Downloaded From: http://ajot.aota.org/ on 07/31/2017 Terms of Use: http://AOTA.org/terms
An ideal teaching aid
for clients exercising to
regain full use of hands.
shoulders, elbows,
forearms, and wrists,
these two new booklets feature clear illustrations and
simple guidelines on improving strength and flexibility
Home Rehabilitation Exercises: Hand, focuses on
range of motion exercises for the hand.
Home Rebabilitation Exercises: Sboulder, Elbow,
Forearm, Wrist includes range of motion exercises for
the shoulder, elbow, forearm, and wrist.
Both booklets have space to note precautions based on
your clients' specific needs.
5:\.00 each or 10 for 52500 AOTA member
54.20 each or 10 for 53500 non-member
Order now with MasterCard or VISA' Call 1-800-THEAOTA (AOTA members), 1-800-654-5584 (MD
members), or (301) 948-9626 (non-members).
April 1991, Volume 45, Numbe1' 4