Transcript Increasing the Manding Repertoire of Children with Autism

Increasing the Manding Repertoire
of Children with Autism
Using the Transitive Establishing Operation
Danielle Draper
Vincent J. Carbone
Margaret Hagerty
Zach Ikkanda
Emily J. Sweeney
Carbone Clinic
Valley Cottage, NY
www.drcarbone.net
To Be Presented at the :Association for Behavior Analysis Conference
Atlanta, Georgia, May 2006.
1
• Keller and Schoenfeld (1950) were the first to identify the
establishing operation (EO) as an independent variable.
• Through a series of writings, Michael (1982, 1988, 1993, 2000)
refined the concept and principle of the establishing operation
(EO) and defined it as an “environmental event, operation, or
stimulus condition that affects an organism by momentarily
altering a) the reinforcing effectiveness of other events and b)
the frequency of occurrences of that part of the organism’s
repertoire relevant to those events as consequences.”
• Michael (1993) described two (2) types of EOs: unconditioned
and conditioned. Unconditioned EOs (UEOs) are “events or
operations or stimulus conditions whose value altering effects
are unlearned,” Conditioned EOs (CEOs) “value altering
effects have been learned during the individual organism’s
learning history.”
• Michael identified and more clearly described the three (3)
different types of CEOs. He named them the surrogate, the
reflexive and the transitive.
2
• All three (3) have been implicated as behavioral variables
within the research related to the treatment of persons with
autism and developmental disabilities (Smith & Iwata, 1997,
Wilder & Carr, 1998; McGill, 1999).
• The transitive CEO appears to be most relevant to the
conditioning of stimuli as reinforcers (Sundberg, 2005) and
may play an important role in teaching language to children
with autism who fail to acquire verbal behavior through typical
means.
• Michael originally referred to this conditioning effect by another
name as a blocked access or interrupted chain effect. In other
words, if some item is highly valuable at this moment but some
additional action, item or even information is necessary to
produce the valued item then any of the additional stimuli
would now act as reinforcers for any behavior that would
produce them.
• In other words, transitive establishing operations convert
neutral stimuli to conditional conditioned reinforcers and
therefore evoke all responses that have in past been
strengthened by their delivery.
3
• By contriving this type of situation a teacher could condition items or
activities in the environment as reinforcers and use these now
established reinforcers to teach language disordered persons to mand
for them.
• Since the mand or request response plays an important role in the
language training of persons with autism (Sundberg & Michael, 2001) a
teacher’s knowledge of how to make use of the transitive establishing
operation as an independent variable may be critical.
• Shafer (1994) points out that the transitive EO can be used to establish
the reinforcing potential of many previously neutral items without the
disadvantages associated with unconditioned establishing operations.
• Moreover, by using the transitive EO to condition items as reinforcers in
a chain of responses, language disordered children can be taught to
mand for items primarily under the control of the EO and therefore free
their responses from the additional control exerted by the presence of
the item.
• A few studies have used the transitive establishing operation to
increase the spontaneous manding repertoire of persons with
developmental disabilities. (Carroll & Hesse, 1987; Hall & Sundberg,
1987; Sigafoos, Doss & Reichle, 1989; Sundberg, Loeb, Hail &
Eigenheer, 2002)
4
• Hall and Sundberg (1987) were among the first experimenters to
successfully demonstrate the value of using an interrupted chain to
teach persons with developmental disabilities to mand for missing
items necessary to complete chains related to making coffee, soup
and purchasing items from vending machines.
• The purpose of this study was to replicate the findings of previous
studies that suggest that transitive EO is an independent variable
relevant to the teaching of the mand and to verify the effects of the
repertoire with children with autism.
• In addition, multiple studies have demonstrated the functional
independence of the mand and tact relations. ( Lamarre & Holland,
1985, Hall & Sundberg, 1987, Twyman, 1996, Sigafoos et al. 1999,
Gomez & Greer 2004, Peursdottir, Carr, & Michael, 2005)
• A second purpose of this study was to test for transfer of the mand
relation to the tact relation.
• Participants in these studies were trained across multiple conditions
consisting of mand training, tact training and a combination of the two.
5
METHOD
Learner and Setting
• There were three (3) participants in this study. All participants were
enrolled in a private educational setting offering one-on-one intensive
teaching in the form of discrete trial training, interspersed with teaching in
the natural environment facilitated through play based activities. Vincent
and Cadia attended the educational setting twice per week and Naryan
attended four (4) times per week. All participants attended the
educational setting for three (3) hour sessions.
• All three (3) participants manded vocally for various items and had some
mands solely under the control of the motivation. However, the majority
of the mands were multiply controlled by both the items presence and the
motivation.
• Vincent was a five (5) year old boy with autism with moderate disabilities.
He has limited but developing tact and intraverbal repertoires.
• Naryan was an eight (8) year old boy with autism and a secondary
diagnosis of seizure disorders with moderate to severe disabilities. He
also had limited but developing tact and intraverbal repertoires.
• Cadia was a five (5) year old girl with PDD/autism spectrum disorder with
moderate disabilities. Cadia had a well developed tact and intraverbal
repertoire and acquired new responses with few presentations of the
stimulus.
6
Response Definition
 The dependent variables were the occurrence of a
spontaneous or EO controlled mand response, a
prompted mand response, or the non-occurrence of a
mand response for the missing item.
 Spontaneous or EO controlled mands were defined as
any mand that occurred in the absence of any echoic or
non-verbal stimulus when the motivation for the item was
established.
 Prompted mands (P) were defined as any mand that
occurred as the result of the teacher’s presentation of a
vocal (echoic) stimulus when the item delivered would
act as a reinforcer.
 Non-occurrence (NR) or no response was defined as
failure to emit any response when a mand would result
in reinforcement.
7
Data Recording and Reliability
 The participants’ instructors served as the primary data collector.
For Vincent, each session was video-taped and later coded in
order to ensure the accuracy of both prompting and data recording.
For Participants 2 and 3 additional instructors were trained to
record data simultaneously but independently for the purposes of
inter-observer agreement (IOA).
 The participants’ instructor and the observer were standing next to
the participant with their data sheet on individual clipboards.
 The targeted items necessary to complete the response chain were
laid out in front of the participant in the setting in which the
response chain was typically conducted.
 The data sheet included prompt level, and a column to record
comments regarding the participant’s reaction to the absence of the
targeted item such as looking or searching when it became
necessary in completing the response chain.
8
_ Agreements and disagreements regarding the occurrence of responses were coded
on a data sheet. An agreement was recorded when both recorders coded a
response identically as an occurrence without a prompt, with a prompt or no
response for the same trial.
– A disagreement occurred when one observer recorded the occurrence of a
behavior that the other observer did not record or the observers did not record the
same prompt level needed to evoke the mand.
– Inter-observer agreement was calculated by dividing agreements by
agreements plus disagreements and multiplying by 100. Inter-observer
agreement was recorded across all conditions.
– The average IOA for Vincent was 96% with a range from 72% to 100%.
The average IOA for Naryan was 92% with a range from 25% to 100%,
and the average IOA for Cadia was 98% with a range from 85% to 100%
for all sessions for which reliability was assessed.
•
Inter-observer agreement was assessed for about 25% of the all sessions.
Experimental Design
•
A multiple baseline across items was used to verify the effectiveness of the
experimental procedures (Baer, Wolf and Risley, 1968).
9
Experimental Conditions
Pre-baseline
• All three (3) participants were exposed to and taught to complete
three (3) response chains each ending in a terminal activity or
stimulus that was identified as having reinforcing value.
• Each response chain included multiple steps leading to the terminal
reinforcer.
• The participants were guided through the steps in the chain once per
session until all the responses in the chain were acquired with very
few prompts.
• Criterion for mastery of a chain was established as 80% of the
responses performed independently over three (3) consecutive
sessions.
10
The response chains for Vincent included:
• Painting a picture - Objects included a smock, paper, clip,
paintbrush, water, paint and an easel. The actions consisted of
putting on the smock, handing the clip to the instructor, picking up the
paintbrush, putting the paintbrush in the water, putting the paintbrush
in the paint and applying the paintbrush to the paper and then
repeating some of these actions several times during the activity.
• Making an art project - Objects included paper cut into shapes, glue
and glitter. The actions consisted of picking up the paper shapes,
putting glue on each shape, arranging the shapes into a picture,
putting glue on the arranged shapes and sprinkling glitter onto the
shapes.

Making a sandwich - Objects in this response chain included bread,
a toaster, a plate, peanut butter and a knife. The actions consisted of
opening the bag of bread, putting the bread into the toaster, pushing
the toaster down, taking the bread out of the toaster, putting the
bread on the plate, opening the peanut butter, putting the peanut
butter on the knife, spreading the peanut butter on the bread and
eating the sandwich.
11
The response chains for Naryan included:

Listening to music - Objects included a portable CD player, a CD,
and headphones. The actions consisted of opening the CD player,
putting the CD in the CD player, putting the headphones on, and
pressing the play button.

Science project - Objects included a plastic container, bottle of
water, 2 bottles of food coloring, and a spoon. The actions consisted
of pouring the water into the container, dropping the food coloring
into the the container, picking up the spoon, mixing the liquid with a
spoon.

Painting a picture - Objects included a smock, paper, paintbrush,
paint, water, paint and an easel. The actions consisted of putting on
the smock, picking up the piece of paper, picking up the paintbrush,
dipping the paintbrush in the water, putting the paintbrush in the paint
and applying the paintbrush to the paper and then repeating some of
these actions several times during the activity.
12
The response chains for Cadia include:
• Making juice - Objects included a cup, powder to make juice, spoon,
2 ice cubes and a measuring cup. The actions consisted of scooping
powder into cup, pouring water in the cup, and putting the ice cubes
into the cup.
• Making an art project - Objects included paper, 3 crayons, glue
stick, and glitter. The actions consisted of coloring the picture,
rubbing the glue on the paper, and sprinkling glitter on the paper.
• Painting a picture - Objects included smock, paper, clip, paintbrush,
water, paint and an easel. The actions consisted of putting on the
smock, picking up the piece of paper, handing the clip to the
instructor, picking up the paintbrush, dipping the paintbrush in the
water, putting the paintbrush in the paint and applying the paintbrush
to the paper and then repeating some of these actions several times
during the activity.
13
Baseline
• During baseline the participants were given the opportunity
to complete the response chains and receive the terminal
reinforcer.
• For each chain the participants were presented with a
specific command to begin the chain, e.g. “paint a picture.”
• The participants were presented with all items necessary to
complete the response chains except for the item selected
as the missing target.
• Within the context of the “paint a picture” chain for Vincent
the easel was missing and targeted for mand training. The
glitter was targeted for training within the art project and the
toaster was the missing item in the “make a sandwich”
chain.
14
•
For Naryan within the “listen to music” chain the CD player was missing and
targeted for training. The food coloring was targeted for training within the
“science project” chain and easel for “painting a picture”.
•
The missing items for Cadia were as follows, the measuring cup within the
context of “making juice”, glitter for “making an art project”, and the clip for
“painting a picture”.
•
During each session the learner was exposed to all three (3) chains.
•
Within the context of completing each chain, the participants paused at the
point at which the missing item was needed to complete the chain and the
instructor began timing a 10 second interval.
•
If the missing item was vocally manded during this interval, the item was
delivered by the experimenter and the response chain was completed.
•
If the missing item was not vocally manded or an incorrect mand occurred
during the 10 second delay, the response chain was discontinued and all
items relevant to the chain were removed.
•
The baseline condition was maintained until learner responding stabilized.
15
Prompt and Prompt Fade Procedures
• During the experimental condition, the participants were again
presented with the response chains and given 10 seconds to mand for
the missing items under the control of the EO.
• If a mand occurred during the 10 second delay, the mand was
immediately reinforced by the delivery of the missing item.
• If a mand did not occur at the end of the 10 second delay the participant
was then given an echoic (vocal) prompt in the form of the missing
item’s name.
• If the participant then manded under the multiple control of both the
vocal prompt (echoic) and the EO, the missing item was then delivered.
• When responding during the experimental met the criterion of three (3)
consecutive session with mands for the missing item occurring solely
under the control of EO, a second chain and missing item was
subjected to the experimental variables.
• The procedures described above were continued until all three (3)
response chains were subjected to the experimental condition for all
three (3) subjects.
16
Transfer Across Operants
• Once EO controlled manding was occurring for the
missing items in each of the response chains for at least
all three consecutive sessions a probe was conducted to
test for transfer of stimulus control from the mand to the
tact repertoire.
• Stimulus control transfer was said to occur if the
participant correctly tacted the missing item previously
taught in the mand condition.
• Probes for stimulus control transfer were conducted in
an unfamiliar environment.
• Probes were conducted by presenting each participant
with the missing item from each response chain and
presenting the discriminative stimulus “What is it?”
17
• Following the presentation of the missing item and the
discriminative stimulus each participant was given five
(5) seconds to respond and data were recorded on the
accuracy of the response.
• A correct response was defined as the participant
tacting the missing item using the same response form
which was taught in the mand condition.
• An incorrect response was defined as the participant not
responding within the five (5) second interval or
responding incorrectly to the item presented.
18
Generalization
• Generalization probes were conducted for all three (3)
participants and all of the chains.
• Novel chains were developed that contained slightly
different stimuli, settings, and instructors, however the
terminal reinforcer or missing item for each chain remained
the same.
• For Vincent, six (6) response chains were developed to
probe for generalization. Each response chain included
different stimuli, settings, and instructors.
• For the “make a sandwich” chain an English muffin and a
bagel were substituted for the bread. For the “paint a
picture” chain finger paint and dry erase markers were
substituted for watercolor paints. For the “art project” chain
a ladybug and Thomas the train picture were substituted
for a Santa Clause picture.
19
•
Probes for generalization were conducted in different settings than the one
in which the original chain was taught.
•
In addition, probes for generalization across instructors were conducted by
having an additional instructor, trained in guiding the participant through
each response chain with minimal prompting, record the occurrence or non
occurrence of the targeted mand response.
•
For Naryan and Cadia, nine (9) response chains were developed to probe
for generalization. However, for these participants’ one (1) independent
variable such as stimuli, setting, or instructor was changed at a time.
•
For Naryan, within the “listen to music” chain different stimuli included
different headphones, a bowl was substituted for a plastic container within
the “science project” chain, and markers were substituted for watercolor
paints within the “paint a picture” chain.
•
For Cadia, lemonade was substituted for kool-aid within the “make a juice”
chain, a ladybug picture was substituted for the flower picture within the “art
project” chain and dry erase markers were substituted for watercolor paints
within the “paint a picture chain.”
•
As with Vincent, probes for generalization were conducted in different
settings than the one in which the original chain was taught as well as
having an additional instructor trained in the prompting and recording
procedures.
20
RESULTS
• As shown in Figure 1 baseline responding for Vincent was
stable at zero (0) spontaneous mands per session for two
(2) of the response chains (“paint a picture” and “do an art
project”).
• Spontaneous manding during the “make a sandwich”
response chain occurred three (3) times for the missing
item, toaster, during the baseline condition.
• Following the implementation of prompt and prompt fade
procedures, spontaneous manding for all missing items
increased at various frequencies per session.
• The effectiveness of the treatment methods in producing an
increase in EO controlled mands was verified by this
response pattern within the context of a multiple baseline
design across items.
21
• Spontaneous manding for the first missing item, glitter, began
occurring following three (3) experimental sessions.
• Glitter met criteria for mastery after six (6) teaching sessions.
• While new response chains underwent teaching, spontaneous
manding for glitter was maintained for nine (9) sessions.
• Spontaneous manding for the second missing item, easel, began
occurring following two (2) teaching sessions.
• Easel met criteria for mastery after five (5) teaching sessions.
• While the last response chain underwent teaching, spontaneous
manding for easel was maintained for seven (7) sessions.
• Spontaneous manding for the third missing item, toaster, began
occurring reliably following two (2) teaching sessions.
• Toaster met criteria for mastery after five (5) teaching sessions.
22
• Figure 2 shows data for Naryan. Baseline responding was
stable at zero (0) spontaneous mands per session for all
three (3) response chains (“listen to music”, “science
project”, and “paint a picture”).
• Spontaneous manding for the first missing item, CD
player began to occur following four (4) experimental
sessions.
• CD player met criteria for mastery after thirteen (13)
teaching sessions.
• Spontaneous manding for CD player maintained for
fourteen (14) sessions, while the next response chain was
subjected to the experimental conditions.
• Spontaneous manding for the second missing item, food
coloring began occurring following three (3) teaching
sessions.
23
• Food coloring met criteria for mastery after nine (9) teaching
sessions.
• While the last response chain underwent teaching, spontaneous
manding for food coloring was maintained for five (5) sessions.
• Spontaneous manding for the third missing item, easel began
occurring after two (2) teaching sessions. Easel met criteria for
mastery after five (5) teaching sessions.
• Figure 3 shows data for Cadia. Baseline responding was stable
at zero (0) spontaneous mands per sessions for all three (3)
response chains.
• Spontaneous manding for the first missing item, clip began
occurring following one (1) experimental session.
• Clip met criteria for mastery after four (4) teaching sessions.
• While the next chain underwent treatment, spontaneous manding
for clip was maintained for eleven (11) sessions.
24
•
Spontaneous manding for the second missing item, glitter, began occurring
following five (5) teaching sessions.
•
While the last chain underwent teaching, spontaneous manding for glitter
was maintained for four (4) sessions.
•
Spontaneous manding for the third missing item, measuring cup, began
occurring following one (1) teaching session.
•
Measuring cup met criteria after three (3) teaching sessions.
•
For all three (3) participants stimulus control transferred from the mand
condition to the tact condition without any teaching.
•
The results for generalization probes for Vincent (Figure 1) are as follows,
the participant did not mand for the easel in either of the two (2) novel chains
for “paint a picture”, which included different stimuli, different instructor, and
different setting.
•
Generalized responding did occur for the four (4) chains that contained
different stimuli, settings, and instructor for both “make an art project” for
glitter, and “making a sandwich” including the toaster.
•
For Participants 2 (Figure 2) and 3 (Figure 3), generalized responding
occurred for each of the nine (9) novel chains that were developed including
stimuli, setting and instructor.
25
DISCUSSION
• The results of this study demonstrate that previously neutral items can
be engendered with reinforcing value through the manipulation of
transitive establishing operations with children with autism.
• These findings replicate the work of Hall and Sundberg (1987) and
extend the research literature on motivative operations to children with
autism.
• Although, some spontaneous manding did occur for Vincent during the
baseline for one (1) item, reliable spontaneous manding only occurred
following the implementation of the experimental procedures.
• It is interesting to note that during both baseline and early in the
experimental conditions Vincent's responses to the missing items
indicated their newly acquired value although he had no formal method
of requesting.
• He was frequently observed to emit responses indicating the value of
the missing item at the point in the chain when the target item was
inaccessible ,e.g. looking around, searching, etc.
26
• These types of responses confirmed the momentary value altering
effects the transitive establishing operation had on the previously
neutral item.
• Once an appropriate vocal mand was taught, the relevant
establishing operation reliably evoked all three (3) responses for
missing items for each participant .
• The transitive EO had established new stimuli as reinforcers and had
therefore made it possible to teach novel responses to each
participant.
• Moreover, the participants’ acquired mands primarily under the
control of the motivation without requiring the supplemental control of
the presence of the item or a vocal prompt.
• This increased the control that the participants’ now had over the
social environment.
27
•
Using this teaching procedure can effectively condition previously neutral
stimuli as reinforcers and therefore increase the variety of stimuli that
function as reinforcers.
•
In addition, by conditioning other reinforcers the frequency and variety of
mands can be increased for children with autism.
•
Furthermore, all three (3) participants were able to tact the targeted item's
that were originally taught as mands without intensive teaching.
•
The transfer of stimulus control from the mand to the tact was also
demonstrated in Peursdottir et al. (2005), where mand training resulted in
tact acquisition for all participants.
•
Peursdottir et al. (2005) attributed the transfer from the mand to the tact to
the existence of the conditions of the controlling variables of the tact being
present during the mand training condition.
28
• Other research by Hall and Sundberg (1987), Gomez and Greer
(2004), Twyman (1996), and Sigafoos et al. (1999) have also
tested for functional independence. These studies have produced
inconsistent results concerning functional independence.
• The results of this study verify Hall and Sundberg (1987) and
Sigafoos et al. (1999) finding in that the untaught verbal operant
(tact) had begun to occur after all the mands underwent the
experimental conditions and had been occurring solely under the
control of the MO.
• Additional research on the development of a more advanced
manding repertoire will increase the understanding of whether
transfer to different verbal operants occur more readily as Skinner
suggests. In addition, it would be interesting to determine if
transfer from tacts to mands would occur.
29
References
Baer, D., Wolf, M., & Risley, T. (1968). Some current dimensions of applied behavioral
analysis. Journal of Applied Behavioral Analysis, 1, 91-97.
Carroll, R. J., & Hesse, B. E. (1987). The effects of alternating mand and tact training on the
acquisition of tacts. The Analysis of Verbal Behavior, 5, 55-65.
Hall, G. A., & Sundberg, M. L. (1987). Teaching mands by manipulating conditioned
establishing operations. The Analysis of Verbal Behavior, 5, 41-53.
Keller, F. S., & Schoenfeld, W. N. (1950). Principles of psychology. New York: AppletonCentury-Crofts.
McGill, P. (1999). Establishing operations: Implications for the assessment, treatment, and
prevention of problem behavior. Journal of Applied Behavioral Analysis, 32, 393-418.
Michael, J. (1982). Distinguishing between discriminative and motivational functions of
stimuli. Journal of the Experimental Analysis of Behavior, 37, 149-155.
Michael, J. (1982). Distinguishing between discriminative and motivational functions of
stimuli. Journal of the Experimental Analysis of Behavior, 37, 149-155.
Michael, J. (1988). Establishing operations and the mand. The Analysis of Verbal Behavior,
6,3-9.
Michael, J. (1993). Establishing operations. The Behavior Analyst, 16, 191-206.
Michael, J. (2000). Implications and refinements of the establishing operation concept.
Journal of Applied Behavior Analysis, 33, 401-410.
Shafer, E. (1994). A review of interventions to teach a mand repertoire. The Analysis of Verbal
Behavior,12,53-66.
30
Sigafoos, J., Doss, S., & Reichle, J. (1989). Developing mand and tact repertoires
in persons with sever developmental disabilities using graphic symbols.
Research in Developmental Disabilities, 10, 183-200.
Sundberg, M. L. (2005). A behavioral analysis of motivation and its relation to mand
training. In L. W. Williams (Ed.). Development disabilities: Etiology, assessment,
intervention, and integration (pp. 1-22). Reno, NV: Context Press.
Sundberg, M. L., Loeb, M., Hail, L., & Eigenheer, P. (2002). Contriving establishing
operations to teach mands for information. The Analysis of Verbal Behavior, 18,
14-28.
Smith, R. G., & Iwata, B. A. (1997). Antecedent influences on behavior disorders.
Journal of Applied Behavior Analysis, 30, 343-375.
Wilder, D. A., & Carr, J. E. (1998). Recent advances in the modification of
establishing operations to reduce aberrant behavior. Behavioral Interventions,
13, 43-59.
31
“Make a sandwich”
Occurrence of Prompt
Level
“Paint a picture”
Occurrence of Prompt
Level
“Do an art project”
Occurrence of Prompt
Level
EO = EO Controlled
P = Vocally Prompted
NR = No Response
Baseline
Prompt Fade
EO
P
NR
Glitter
EO
P
NR
Easel
EO
P
NR
Toaster
1
3
5
7
9
11
13
15
17
19
Sessions
Vincent
Figure 1. The occurrence of mands for missing items recorded by controlling variable (MO, prompted, no response)
across baseline and treatment conditions.
Bas eline
EO = EO Controlled
P = Vocally Prompted
NR = No Response
Prom pt Fade
“Listen to Music”
Occurrence of
Prompt Level
EO
P
NR
CD Player
“Do a Science
Project”
Occurrence of
Prompt Level
EO
P
NR
“Paint a Picture”
Occurrence of Prompt
Level
Food Coloring
EO
P
NR
Easel
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Sessions
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
Naryan
Figure 2. The occurrence of mands for missing items recorded by controlling variable (MO, prompted, no response)
across baseline and treatment conditions.
Baseline
EO = EO Controlled
P = Vocally Prompted
NR = No Response
Prompt Fade
“Paint a Picture”
Occurrence of
Prompt Level
EO
P
NR
“Do an Art Project”
Occurrence of
Prompt Level
Clip
EO
P
NR
“Let’s Make Some
Juice”
Occurrence of
Prompt Level
Glitter
EO
P
NR
Meas uring Cups
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Figure 3. The occurrence of mands for missing items recorded by controlling variable (MO, prompted, no
response) across baseline and treatment conditions.
Cadia