ERIC EJ910166: Functional Analysis and Treatment of Aggression Maintained by Preferred Conversational Topics

2010·0.25 MB·English

by  ERIC

JOURNALOFAPPLIEDBEHAVIORANALYSIS 2010, 43, 723–727 NUMBER4 (WINTER2010)
FUNCTIONAL ANALYSIS AND TREATMENT OF AGGRESSION
MAINTAINED BY PREFERRED CONVERSATIONAL TOPICS
EILEEN M. ROSCOE, ARIANNE E. KINDLE, AND SACHA T. PENCE
NEWENGLANDCENTERFORCHILDREN
NORTHEASTERNUNIVERSITY
After an initial functional analysis of a participant’s aggression showed unclear outcomes, we
conducted preference and reinforcer assessments to identify preferred forms of attention that
may maintain problem behavior. Next, we conducted an extended functional analysis that
included a modified attention condition. Results showed that the participant’s aggression was
maintainedbyaccesstopreferredconversationaltopics.Afunction-basedinterventiondecreased
aggression andincreasedan appropriatecommunicative response.
Keywords: attention,functionalanalysis,preferenceassessment,socialpositivereinforcement
_______________________________________________________________________________
A number of studies have documented the Kelley, 2007; Richman & Hagopian, 1999), a
occurrence of unclear or uninterpretable func- systematic method for identifying these atypical
tional analysis outcomes that required subse- forms of attention prior to testing them in a
quent modification and extension of the functional analysis has not been evaluated.
functional analysis before the data could be Systematic preference and reinforcer assess-
used to inform treatment (Bowman, Fisher, ments, similar to those used for identifying
Thompson,&Piazza,1997;Thompson,Fisher, edible- and leisure-item reinforcers (e.g., Fisher
Piazza, & Kuhn, 1998). One potential expla- et al., 1992), may be useful for identifying
nation for unclear functional analysis outcomes various forms of attention that may maintain
is that the specific form of attention that problembehavior.Inthecurrentstudy,afteran
maintains problem behavior is not delivered initial functional analysis resulted in inconclu-
during the attention condition. In the Iwata, sive outcomes, we evaluated the utility of
Dorsey, Slifer, Bauman, and Richman (1982/ preference and attention analyses for systemat-
1994) attention condition, the therapist deliv- ically identifying novel forms of attention that
ered a brief vocal reprimand contingent on may maintain problem behavior. Second, we
problem behavior. If other forms of attention conducted an extended functional analysis that
maintain problem behavior, low levels of includedthemodifiedformofattention.Third,
problem behavior across all conditions and a we conducted a treatment assessment to
false-negative outcome for attention as a evaluate the utility of the extended functional
maintaining reinforcer may result. analysis outcome.
Although previous research has shown that
various forms of attention may maintain
METHOD
problem behavior (Fisher, Ninness, Piazza, &
Owen-DeSchryver, 1996; Kodak, Northup, & Participant, Target Response, and Setting
Carriewasa13-year-oldgirlwhoresidedata
ThisresearchwasconductedbyArianneKindletowards
residential school for children with autism and
fulfillment of her master’s degree at Northeastern
relateddisabilities.Shehadbeendiagnosedwith
University.
Address correspondence to Eileen Roscoe, Assistant pervasive developmental disorder and commu-
Director of Research, New England Center for Children, nicated using modified signs, pictures, and a
33 Turnpike Road, Southborough, Massachusetts 01772
speech augmentation device. Her target behav-
(e-mail: [email protected]).
doi:10.1901/jaba.2010.43-723 ior was aggression, defined as any instance of
723
724 EILEEN M. ROSCOE et al.
actualorattemptedhittingabovetheshoulders, them. During the session, the therapist re-
biting, or hair pulling. We conducted sessions presented the items for 30 s contingent on
in a quiet area of the classroom. aggression. During the demand condition, the
therapist presented instructions using a three-
Response Measurement and step prompting hierarchy. Compliance resulted
Interobserver Agreement inpraise,andaggressionresultedintermination
Trained observers scored aggression using a of instructions. During the control condition,
latency measure (i.e., the duration of time from moderately preferred leisure items were contin-
the start of the session until the first instance of uously available. The therapist delivered atten-
aggression). During the treatment analysis, tion (praise paired with brief physical contact)
observers collected data on a communication on a fixed-time 30-s schedule and did not
response (touching a communication card) provide consequences for aggression. Figure 1
using response latency and frequency. Interob- shows the results of the standard functional
server agreement was calculated for 34% of analysis,theattentionanalysis,andtheextended
functional analysis and treatment sessions for functional analysis. Please note that aggression
aggression and for 40% of treatment sessions is plotted as occurring at 300 s when it did not
for the communication response. We calculated occur. During the standard functional analysis,
interobserver agreement by dividing the shorter aggression did not occur in several of the
latency (in seconds) by the longer latency and conditionsandoccurredonlysporadicallyinthe
converting the ratio to a percentage. Mean demand condition, preventing definitive con-
agreementwas96%(range,87%to100%),and clusions regarding the function of Carrie’s
100%, for aggression and the communication aggression. Because anecdotal observation indi-
response, respectively. cated that Carrie often asked to talk about
specific activities andexhibited aggression when
Functional Analysis
the therapist denied these requests, we conduct-
Thefunctionalanalysisprocedurewassimilar ed subsequent analyses to identify whether
to that described by Iwata et al. (1982/1994) conversational topics might function as a
and by Thomason-Sassi, Iwata, Neidert, and reinforcer for her aggression.
Roscoe (in press). Four conditions, including
attention, tangible, demand, and control, were Preference Assessment
conducted using a multielement design. Ses- We conducted a paired-stimulus preference
sions lasted a maximum of 5 min. Attention, assessment,similartothatofFisheretal.(1992),
tangible, and demand sessions ended immedi- to determine high-preference (HP) and low-
ately following the delivery of the consequence preference (LP) conversational topics for use
for the first instance of the target response or at duringtheattentionanalysis.Linedrawingswere
5 min, and control sessions always ended at used to represent the different conversational
5 min. The therapist wore different-colored topics included. During each trial, the therapist
shirts and presented photographs depicting the presented two line drawings simultaneously and
relevant motivating operation during sessions. asked Carrie to select the item about which she
During the attention condition, the therapist wantedtotalk.Contingentonselection(defined
presented moderately preferred leisure items aspointingtooneofthetwolinedrawings),the
and diverted her attention. Following an therapist withdrew the drawings and initiated
occurrence of aggression, the therapist delivered social questions and statements concerning the
brief vocal and physical attention. Prior to the selected conversational topic for 30 s. Each
startoftangiblesessions,thetherapistpresented drawing was paired with every other drawing in
preferred items for 1 min and then removed thearray.Resultsfromthepreferenceassessment
FUNCTIONAL ANALYSIS 725
Figure1. Latency(inseconds)toaggressionduringthestandardfunctionalanalysis,theattentionanalysis,andthe
extended functional analysis (top). Percentage of trials Carrie selected each of the conversational topics during the
preferenceassessment(middle).Latency(inseconds)toaggressionandtothecommunicationresponseduringtheFCT
treatmentassessment (bottom).
(Figure 1) indicated that Carrie selectedconver- al topics. Conditions included an HP topic, an
sational topics about zoos or dogs most often LPtopic,andacontrol.Sessions lasted5min if
(93%), and she never selected the weather. aggression did not occur. In both the HP and
Therefore, zoos and dogs were identified as HP LP topic conditions, the therapist terminated
topicsandweatherastheLPtopicforuseduring the session after delivering the first conse-
the attention analysis. quence.IntheHPtopiccondition,thetherapist
placed a picture of an HP form of conversation
Attention Analysis (zoos or dogs, alternated across sessions) in
We conducted an attention analysis to front of Carrie. Prior to the start of the session,
determine whether Carrie’s aggression was the therapist initiated conversation about the
maintainedbyaccesstoparticularconversation- topic for 30 s, then stated ‘‘We aren’t talking
726 EILEEN M. ROSCOE et al.
about [topic] anymore,’’ and diverted her differential consequences contingent on occur-
attention. Contingent on aggression, the ther- rences of the communication response. Because
apist reinitiated conversation about the same the communication response was novel, FCT
HP topic for 30 s. The LP topic condition was training sessions, consisting of prompts that
identical to the HP topic condition except that were successively faded, were conducted. Fol-
the LP topic was discussed. In the control lowing training, the FCT treatment phase was
condition, a picture of the HP topic was initiated. This phase was identical to baseline,
depicted, and the therapist continuously con- exceptthatthetherapistprovidednodifferential
versed with Carrie about that HP topic. If consequences for aggression (extinction was in
aggression occurred, the therapist ignored the effect) and 30-s access to HP topics contingent
behavior. Results from the attention analysis on occurrences of the communication response.
(Figure 1) indicated that differentially short Prior to the first session of the second baseline
latencies to aggression consistently occurred phase(Session12),thetherapistsaid,‘‘Useyour
during the HP topic condition, whereas long card if you want to talk.’’
latencies or no aggression occurred most often Treatment assessment results are depicted in
during both the LP topic and control condi- Figure 1. During baseline, Carrie exhibited
tions. These findings indicated that HP topics, short latencies to aggression and never emitted
identified during the preference assessment, the communication response. During the first
maintainedCarrie’s aggression,whereasconver- FCT intervention phase, she did not exhibit
sational topics about the weather did not. aggression and emitted the communication
response at mostly short latencies and at a high
Extended Functional Analysis frequency (M 5 5.8). During the second
During this phase, attention, tangible, de- baseline, she exhibited aggression at short
mand, and control conditions were conducted latencies and did not exhibit the communica-
as described above. However, the HP topic tion response for the last two sessions. During
condition (described above) was also included. the return to the FCT intervention, she did not
Results from this analysis (Figure 1) showed exhibit aggression for most of the sessions and
differentially short latencies to aggression dur- she exhibited the communication response at
ing the HP topic condition relative to control, short latencies and at a high frequency (M 5
attention, demand, and tangible conditions. 6.3).
These findings indicate that when a modified
attention condition (the HP topic condition)
DISCUSSION
was included in the functional analysis, a clear
function could be determined. In the current study, we replicated previous
research by showing that an initial functional
Treatment Assessment analysis did not yield conclusive outcomes and
Based on the results of the extended that subtle modifications to conditions were
functionalanalysis,afunctionalcommunication necessary for determining behavioral function.
training(FCT)interventionwasevaluatedusing Wealsoextendedpreviousresearchinanumber
a reversal design. During all sessions, the of ways. First, we conducted a preference
therapist placed a card depicting ‘‘I want to assessment of conversational topics to identify
talk’’ in front of Carrie. During baseline, the HP topics. Second, we evaluated the utility of
therapistpresentedHPconversationaltopicsfor the preference assessment by conducting an
30 s and then diverted her attention and began attention analysis to determine whether the HP
the session. The therapist provided 30-s access topic condition resulted in shorter response
to HP topics contingent on aggression and no latencies than did LP topic and control
FUNCTIONAL ANALYSIS 727
conditions. Third, we used the results of the Fisher,W.W.,Ninness,H.A.C.,Piazza,C.C.,&Owen-
preference assessment and attention analysis to DeSchryver,J.S.(1996).Onthereinforcingeffectsof
the content of verbal attention. Journal of Applied
inform a subsequent modified functional anal-
Behavior Analysis,29,235–238.
ysis condition. We then used the results of the Fisher, W., Piazza, C. C., Bowman, L. G., Hagopian, L.
extended functional analysis to develop an P., Owens, J. C., & Slevin, I. (1992). A comparison
of two approaches for identifying reinforcers for
effective function-based intervention.
personswithsevereandprofounddisabilities.Journal
There are some limitations of the current of AppliedBehaviorAnalysis,25,491–498.
study that deserve comment. First, we did not Iwata,B.A.,Dorsey,M.F.,Slifer,K.J.,Bauman,K.E.,&
Richman,G.S.(1994).Towardafunctionalanalysis
conduct an equal number of sessions within
ofself-injury.JournalofAppliedBehaviorAnalysis,27,
each condition of the attention analysis and the 197–209.(ReprintedfromAnalysisandInterventionin
extended functional analysis. Therefore, it is DevelopmentalDisabilities,2,3–20,1982)
Kodak, T., Northup, J., & Kelley, M. E. (2007). An
unclear whether this may have affected our
evaluation of the types of attention that maintain
results. In addition, we evaluated only one problem behavior. Journal of Applied Behavior
general class of attention—conversational top- Analysis,40,167–171.
Richman, D. M., & Hagopian, L. P. (1998). On the
ics—during the preference assessment and
effects of ‘‘quality’’ of attention in the functional
attention analysis. It may be helpful to evaluate
analysis of destructive behavior. Research in Develop-
different forms of attention other than conver- mental Disabilities,20,51–62.
sational topics (e.g., different types of physical Thomason-Sassi, J. L., Iwata, B. A., Neidert, P. L., &
Roscoe,E.M.(inpress).Responselatencyasanindex
attention) in a preference assessment prior to
of response strength during functional analyses of
testing whether they maintain problem behav- problembehavior.JournalofAppliedBehaviorAnalysis.
ior. Thompson,R.H.,Fisher,W.W.,Piazza,C.C.,&Kuhn,
D. E. (1998). The evaluation and treatment of
aggression maintained by attention and automatic
REFERENCES reinforcement. Journal of Applied Behavior Analysis,
31,103–116.
Bowman, L. G., Fisher, W. W., Thompson, R. H., &
Piazza, C. C. (1997). On the relation of mands and Received June 15,2009
the function of destructive behavior. Journal of Final acceptanceJanuary 11, 2010
AppliedBehavior Analysis, 30,251–265. Action Editor,Cynthia Anderson