ERIC EJ750819: A Preliminary Analysis of Self-Control with Aversive Events: The Effects of Task Magnitude and Delay on the Choices of Children with Autism

2006·0.06 MB·English

by  ERIC

JOURNALOFAPPLIEDBEHAVIORANALYSIS 2006, 39, 227–232 NUMBER2 (SUMMER2006)
A PRELIMINARY ANALYSIS OF SELF-CONTROL WITH AVERSIVE
EVENTS: THE EFFECTS OF TASK MAGNITUDE AND DELAY ON THE
CHOICES OF CHILDREN WITH AUTISM
DOROTHEA C. LERMAN
UNIVERSITYOFHOUSTON–CLEARLAKE
AND
LAURA R. ADDISON AND TIFFANY KODAK
LOUISIANASTATEUNIVERSITY
Whenfacedwithachoicebetweentwoaversiveevents,apersonexhibitsself-controlbychoosing
asmaller,moreimmediateaversiveeventoveralarger,delayedaversiveevent.Taskdemandsare
oftenaversivetochildrenwithautismandotherdevelopmentaldisabilities.Thepurposeofthis
studywastoevaluatebehavioralsensitivitytodifferencesintheamountanddelayoftasksaspart
ofapreliminarystudyonself-control.Participantswere2childrenwithautismwhoengagedin
problembehaviormaintainedbyescape.Resultsindicatedalackofself-controlwithrespectto
choosingbetweentwoaversivetasksandsuggestedpotentialstrategiesforincreasingself-control
(i.e.,choosing asmallimmediate task over alarge delayedtask).
DESCRIPTORS: aversive events, self-control, task magnitude, taskdelay
_______________________________________________________________________________
Research findings indicate that problem (Iwata, Pace, Kalsher, Cowdery, & Cataldo,
behavior exhibited by children with develop- 1990).
mental disabilities is commonly shaped and A number of effective treatments have been
maintainedbyescapefromoravoidanceoftasks developed for escape-maintained behavior, in-
(e.g., Iwata et al., 1994). A choice may be cluding extinction, differential negative rein-
established when a child is instructed to forcement of alternative behavior, and demand
complete an aversive task. If the child exhibits fading. Research findings also suggest that
problem behavior, the child may receive giving students opportunities to choose among
negative reinforcement in the form of escape several tasks can decrease the likelihood of
from the task and possibly delay of further escape-motivated behavior (Romaniuk et al.,
instruction. However, if the child chooses to 2002). Further research is needed on factors
work on the task instead, the child may receive that influence choice, especially when choice is
punishment in the form of exposure to the provided within the context of aversive events
aversivetask(and possiblyadditionalworkafter such as task demands.
its completion), along with a small amount of The concept of self-control is relevant to
positive reinforcement in the form of praise. choicesituationsthatinvolveaversiveevents.An
Under such an arrangement, it is not surprising individual is said to exhibit self-control when
thatthechildwillfrequentlyengageinproblem behavior is more sensitive to delayed conse-
behavior rather than complete aversive tasks quencesthantoimmediateconsequences.When
the consequence is reinforcement, self-control is
defined as selecting the response that produces
Address correspondence to Dorothea C. Lerman, a large delayed reinforcer (e.g., saving money
UniversityofHouston–ClearLake,2700BayAreaBlvd., now to purchase a home next year) over the
Campus Box 245, Houston, Texas 77058 (e-mail:
response that produces a small immediate
[email protected]).
doi:10.1901/jaba.2006.90-05 reinforcer (e.g., spending money now to attend
227
228 DOROTHEA C. LERMAN et al.
a concert). When the consequence involves Results of a pretreatment functional analysis
aversiveevents,self-controlisdefinedasselecting indicated that their problem behavior was
the response that produces a small immediate maintained by escape from demands (data are
aversive event (e.g., going to the dentistnow for available from the first author). John’s sessions
aroutinecleaning)overaresponsethatproduces were conducted in an unused room at his
a large delayed aversive event (e.g., going to the school. Archie’s sessions were conducted in
dentist only when a root canal is needed). a therapy room at a university-based early
When a child must choose between two intervention program. Both rooms contained
aversive tasks—one that is large but can be tables, chairs, and the materials needed to
completed later and one that is small but must conduct the sessions.
be completed immediately—the child exhibits
self-control by choosing the immediate but Response Measurement and Reliability
smaller task. Although no applied research has ForJohn,choiceofTask1versusTask2was
been conducted in this area, basic studies defined as sitting in a chair that was placed
indicate that children often lack self-control directly in front of the designated task. Archie’s
when choices involve aversive events (Grusec, choice response was defined as touching
1968). Basic research also has identified several a picture card associated with the designated
strategiesthatmightincreaseself-control.These task. Previously trained undergraduate and
includeincreasingthedelaytobothtasksbythe graduate students collected frequency data on
same amount, increasing the amount of the the choice response using laptop computers,
largetask,ordecreasingtheamountofthesmall and the data were expressed as a percentage of
task (DeLuty, 1978; Grusec). opportunities. A second observer independently
Although it would be beneficial to establish collected data during 36% of the sessions for
a self-control repertoire as part of treatment for John and 74% of the sessions for Archie. Mean
negatively reinforced behavior, no studies have occurrence agreement for the choice response
examined the sensitivity of choices of children was 99% (range, 80% to 100%) for John and
with autism or other developmental disabilities 95% (range, 80% to 100%) for Archie.
to parameters of aversive events that are specific
Procedure
to self-control situations. Thus, the first step is
In all conditions, participants were asked to
toexaminethesensitivityofchoicestothedelay
choose between two tasks that differed only
andmagnitudeofaversivetasks.Thepurposeof
with respect to the amount of work that was
the study was to evaluate behavioral sensitivity
needed to complete the task or the amount of
to differences in the amount and delay of tasks.
time that passed between the choice response
We also conducted a preliminary analysis of
and the start of the task. Tasks used during the
factors that might influence self-control after
demand condition of the functional analysis
determining that the participants failed to
were selected for the study. John’s task was
demonstrate self-control.
placing lettersin a puzzle,and Archie’stask was
identifyinglettersoundsspokenbythetherapist
METHOD
by pointing to the correct letter card from
Participants and Settings among two concurrently presented cards. For
Participants were 2 children who had been John, the puzzles were placed on opposite ends
diagnosed with autism and who engaged in of a table. The therapist physically guided John
aggression and disruption. Both were 4 years to a location that was equidistant from the two
old, communicated vocally using complete tasks at the beginning of each choice trial and
sentences, and followed three-step instructions. instructed him to pick a task. For Archie,
A PRELIMINARY ANALYSIS OF SELF-CONTROL 229
picture cards associated with each task were the delay was placed on the side of the table
placed on a table. The therapist physically that was opposite to the chair, whereas the
guidedArchietoalocationthatwasequidistant task or card associated with the immediate
from the two cards and instructed him to pick task was placed directly in front of the chair.
one. Participants were verbally prompted to If the participant selected the immediate task,
choose if they failed to make a choice response he was prompted to make another choice
within10 softheinstruction.Reinforcementin 60 s after the completion of the task. If the
the form of a highly preferred food item was participant selected the delayed task, he
delivered for engaging in the choice response to was prompted to make another choice 10 s
increase the likelihood of a choice. Once the after completion of the task (to give the
task began, instructions were delivered contin- therapist an opportunity to prepare the task
uously using a three-step prompting sequence materials). To control for the immediacy and
(verbal, model, physical guidance), and praise density of attention across immediate versus
was delivered on a fixed-time (FT) 15-s delayed tasks, attention was delivered on an FT
schedule. All sessions consisted of five choice 15-s schedule during the delay or intertrial
trials. The left–right positions of the tasks or interval as well as during the task. The
choice cards were alternated each session. Two participant could move freely around the room
forced-choice trials were conducted prior to during the delay but did not have access to any
each session so that the participants would be leisure materials.
exposed to the contingencies associated with Self-control analysis. In the first phase, John
each choice. was instructed to choose between a small
Magnitude analysis. Participants chose be- immediate task and a large delayed task.
tween different amounts of the task. The Procedures were similar to those in the other
task amount for John was defined by the analyses. Because John showed self-control, the
number of puzzle pieces that had to be inserted amount of the immediate task was increased
into the puzzle (3 or 24). To assist with from small to medium (from 3 pieces to 12
discrimination, the designated number of pieces) to determine if the amount of the
puzzle pieces was placed next to each puzzle. immediate task might influence self-control.
The number of letter sounds that Archie was After he failed to show self-control, the effects
required to identify (5 or 10) defined the ofincreasingthedelaytobothtasksby10 sand
amount of the task. To assist with discrimina- 20 s were evaluated.
tion, pictures illustrating the different amounts Archie’s self-control analysis was slightly
of work were drawn on the choice cards. different. In the first phase, Archie chose
Immediately after the choice, the therapist between a small immediate task and a medium
initiatedthepromptingsequenceandcontinued delayed task. Procedures were similar to those
until the task was completed. The participant in the other analyses. Next, the effect of
waspromptedtomakeanotherchoice60 safter increasing the delay to both tasks by an
the completion of the task. equivalent amount was evaluated (20 s were
Delay analysis. Participants chose between addedtoeachinitially,followedby30 s).When
a task that had to be completed immediately this failed to alter self-control, we examined the
versusonethathadtobecompletedaftera60-s effects of increasing the size of the delayed task
delay. A medium amount of the task was used (from 10 letter sounds to 20 letter sounds) as
during this analysis (12 puzzle pieces for John wellasdecreasingthesizeoftheimmediatetask
and 10 letter sounds for Archie). To assist with (from five sounds to one sound) and the delay
discrimination, the task or card associated with to both tasks.
230 DOROTHEA C. LERMAN et al.
Figure 1. Percentage of trials in which John (top two panels) and Archie (bottom two panels) chose the large or
mediumtaskoverthesmalltaskduringthemagnitudeanalysis(topandthirdleftpanels);theimmediatetaskoverthe
delayedtaskduringthedelayanalysis(topandthirdrightpanels);andthesmallimmediatetaskoverthelargedelayed
task during the self-control analysis (second and fourth panels). Numbers indicate the delay values (in seconds) and
amountoftask associatedwitheach choice during the self-control analysis.
A PRELIMINARY ANALYSIS OF SELF-CONTROL 231
RESULTS AND DISCUSSION phase and a 30-s delay was introduced in the
sixth phase.
During the initial analyses of magnitude and
Results of this preliminary investigation
delay, John consistently chose the small task
showed that the choices of 2 children with
overthelargetask andthedelayedtask overthe
autism who engaged in problem behavior
immediate task (top two panels of Figure 1).
maintained by escape from demands were
Theseresultsindicatedthatthetasksfunctioned
sensitive to differences in the amount and delay
as aversive events. In the first phase of the self-
of aversive tasks. For John, the amount of the
control analysis (second panel), John demon-
immediate task and the length of the delays
strated self-control by consistently selecting the
associated with the tasks seemed to influence
small immediate task over the large delayed
self-control. However, these factors had less
task. This suggested that his behavior was more
influence on the self-control responses for
sensitive to the amount of the task than to the
Archie. It may be beneficial to establish a self-
delay. Thus, in the next phase, the amount of
control repertoire as part of treatment for
theimmediatetaskwasincreased(fromsmallto
negatively reinforced behavior. That is, com-
medium) to determine if John would continue
monly used treatments may be more effective if
to show self-control. He immediately began to
a child’s behavior is less sensitive to the
choose the delayed task more than the imme-
immediate (aversive) consequence that is asso-
diate task. Choice became more variable when
ciated with engaging in a requested task. Thus,
the delay to both tasks was increased by 10 s.
more research is needed to evaluate strategies to
He showed a somewhat clearer preference for
increase self-control when choices involve
the medium immediate task when the delay to
aversive events such as task demands.
both was increased by 20 s. This finding was
Future research will need to address several
replicated following a reversal to the 0-s versus
methodological issues that arose in the current
60-s delay values.
investigation. First, differences in the amount
Archie also consistently chose the delayed
of task during the magnitude analysis and
task over the immediate task and the small task
differences in the delay to the task were
overthemediumtaskduringtheinitialanalyses
necessarily associated with differences in
of magnitude and delay (third panel of
the delay to negative reinforcement (removal
Figure 1). These results indicated that the tasks
of the task). Second, the overall density of
functioned as aversive events. Unlike John,
attention was not equivalent for the two
Archie did not show self-control in the first
responseoptionsduringtheself-controlanalysis
phase of the self-control analysis (bottom
because the large delayed task took longer
panel). He consistently selected the medium
to complete than the small immediate task.
delayed task over the small immediate task.
In future studies, the interval between the start
Choice became more variable when the delay
of a choice trial and the next choice trial could
to both tasks was increased by 20 s and
be held constant. However, the density of
30 s, but he continued to choose the delayed
negative reinforcement (i.e., break from the
task on more trials than the immediate task.
task) would then be disparate across the
In lieu of increasing the delay further, we
response options.
manipulated the magnitude of the tasks.
Beginning with the fourth phase, the size
of the delayed task was doubled. This manip- REFERENCES
ulation failed to influence self-control, even
Deluty, M. Z. (1978). Self-control and impulsiveness
when the size of the immediate task was
involving aversive events. Journal of Experimental
decreased (to a single letter sound) in the fifth Psychology: Animal BehaviorProcesses,4,250–266.
232 DOROTHEA C. LERMAN et al.
Grusec, J. E. (1968). Waiting for rewards and punish- Romaniuk,C.,Miltenberger,R.,Conyers,C.,Jenner,N.,
ments: Effects of reinforcement value on choice. Jurgens,M.,&Ringenberg,C.(2002).Theinfluence
JournalofPersonalityandSocialPsychology,9,85–89. of activity choice on problem behaviors maintained
Iwata,B.A.,Pace,G.M.,Dorsey,M.F.,Zarcone,J.R., byescapeversusattention.JournalofAppliedBehavior
Vollmer, T. R., & Smith, R. G., et al. (1994). The Analysis, 35,349–362.
functionsofself-injuriousbehavior:Anexperimental-
epidemiological analysis. Journal of Applied Behavior
Analysis,27,215–240.
Iwata,B.A.,Pace,G.M.,Kalsher,M.J.,Cowdery,G.E.,
& Cataldo, M. F. (1990). Experimental analysis and Received July 12,2005
extinctionofself-injuriousescapebehavior.Journalof Final acceptance January20, 2006
AppliedBehavior Analysis, 23,11–27. Action Editor,John Northup