ERIC EJ789960: Functional Analyses and Treatment of Precursor Behavior

2008·0.31 MB·English

by  ERIC

JOURNALOFAPPLIEDBEHAVIORANALYSIS 2008, 41, 97–105 NUMBER1 (SPRING2008)
FUNCTIONAL ANALYSES AND TREATMENT OF
PRECURSOR BEHAVIOR
ADEL C. NAJDOWSKI, MICHELE D. WALLACE, CARRIE L. ELLSWORTH,
ALICIA N. MACALEESE, AND JACKIE M. CLEVELAND
UNIVERSITYOFNEVADA,RENO
Functionalanalysishasbeendemonstratedtobeaneffectivemethodtoidentifyenvironmental
variablesthatmaintainproblembehavior.However,therearecaseswhenconductingfunctional
analyses of severe problem behavior may be contraindicated. The current study applied
functional analysis procedures to a class of behavior that preceded severe problem behavior
(precursorbehavior)andevaluatedtreatmentsbasedontheoutcomesofthefunctionalanalyses
of precursor behavior. Responding for all participants was differentiated during the functional
analyses,andindividualizedtreatmentseliminatedprecursorbehavior.Theseresultssuggestthat
functionalanalysisofprecursorbehaviormayofferanalternative,indirectmethodtoassessthe
operantfunction of severeproblem behavior.
DESCRIPTORS: functional analysis, functional communication training, precursor
behavior,response class, severe problem behavior
_______________________________________________________________________________
Experimental functional analysis of behavior aggressionorinappropriatesexualbehavior(e.g.,
disorders involves the direct manipulation of grabbing staff’s or peers’ genitalia) must be
antecedents and consequences hypothesized to weighedagainsttheimmediateeffectsofevoking
occasion and maintain problem behavior such behavior and the more lasting effects of
(Iwata, Dorsey, Slifer, Bauman, & Richman, reinforcing it. In addition, schools, day pro-
1982/1994). Effects of these manipulations are grams, parents, and other consumers of behav-
evaluated in highly controlled conditions in ioral services may not permit severe problem
which problem behavior is evoked and, pre- behavior to be reinforced during assessment.
sumably, reinforced; thus, functional analyses In light of these concerns, a number of
are likely to produce temporary increases in strategies have been designed to reduce the
problembehaviorinoneormoreconditions.In possible risks associated with conducting exper-
fact, elevated measures in one or more exper- imentalfunctionalanalysesofproblembehavior.
imental conditions are taken as evidence of the One strategy has been to limit the length of the
operant functions of problem behavior. assessment (e.g., Derby et al., 1992; Kahng &
Therearereasonableconcernsassociatedwith
Iwata, 1999; Vollmer, Marcus, Ringdahl, &
assessment procedures that systematically evoke
Roane,1995).However,briefassessmentsdonot
andprovidepotentiallyreinforcingconsequences
identify maintaining variables consistently; as a
for severe problem behavior. For example, the
result, extended analyses requiring a longer
relative benefits of conducting experimental
assessment period may be necessary (Vollmer
analyses of severe problem behavior such as
et al.). Another strategy has been to use pro-
tective equipment during functional analyses
We thank Charna Mintz and Becky Penrod for their
assistanceinconductingthisproject.AdelC.Najdowskiis (e.g.,Le&Smith,2002).However,researchhas
now affiliated with the Center for Autism and Related demonstrated that the addition of such equip-
Disorders,Inc.,andMicheleD.Wallace isnowaffiliated
mentcanaltertheresultsofafunctionalanalysis
withCalifornia StateUniversity, LosAngeles.
RequestsforreprintsshouldbeaddressedtoMicheleD. (Borrero, Vollmer, Wright, Lerman, & Kelley,
Wallace,CaliforniaStateUniversity,LosAngeles,Charter 2002). Moreover, it may be difficult or impos-
College of Education, Division of Special Education and
sible to add protective equipment for some
Counseling, KH C1064, LosAngeles, California 90032.
doi:10.1901/jaba.2008.41-97 behavior (e.g., inappropriate sexual behavior).
97
98 ADEL C. NAJDOWSKI et al.
Another strategy for reducing risk associated treatments based on the results of the analysis
with an experimental functional analysis is to ofprecursorbehaviorwouldeliminateprecursor
infer the variables that maintain severe problem behaviorwhilemaintainingloworzerolevelsof
behavior indirectly by conducting a functional severe problem behavior.
analysis of precursor behavior (Smith &
Churchill, 2002). Precursor behavior is any METHOD
response that tends to occur immediately prior Participants and Settings
to severe problem behavior. For example, an Three individuals had been referred for
individual may yell or threaten to engage in assessment and treatment of severe problem
aggression before engaging in aggression. behavior. All participants were able to follow
Although topographies of precursor behavior simple instructions and communicated using
differ from severe problem behavior, precursor speech. Brent was an 8-year-old boy who had
and severe problem behavior may be related beendiagnosedwithanemotionalhandicapand
functionally if they produce the same conse- traumaticbraininjury.Petewasa5-year-oldboy
quence (Grow, Kelley, Roane, & Shillingsburg, who exhibited developmental delays. Brent and
2008). When different topographies are sensi- Pete had been referred for aggression towards
tivetothesameconsequences,theymay besaid teachers (Brent) and peers (Brent and Pete) at
to be members of a common response class school. Tom was a 45-year-old man who had
(Skinner, 1969). Thus, functional analysis of beendiagnosedwithmentalretardation.Hehad
precursor behavior is based on the notion that been referred for engaging in masturbation and
contingencies applied to a member of a grabbing or attempting to grab the genitalia of
response class have similar effects on other staff and peers at his day program.
responses in that class (Harding et al., 2001; Sessionswereconducted3to5 daysperweek
Lalli, Mace, Wohn, & Livezey, 1995; Parrish, in a classroom for children with behavior
Cataldo, Kolko, Neef, & Egel, 1986; Sprague problems (Brent), a treatment room attached to
& Horner, 1992). This strategy presumes that the kindergarten (Pete), and either a treatment
precursorbehaviorandsevereproblembehavior room or in the natural environment at the day
are members of the same functional class and program (Tom). All settings resembled a
thatit should be possible to reduce or eliminate classroom, with academic materials, tables,
severe problem behavior using treatments chairs, and leisure items available (depending
corresponding to results of functional analyses on the condition).
of precursor behavior.
SmithandChurchill(2002)comparedresults Response Measurement and
offunctionalanalysesofsevereproblembehavior Interobserver Agreement
with functional analyses of precursor behavior. The dependent variables were occurrences of
Results demonstrated that functional analyses of severe problem behavior, precursor behavior,
precursorbehavioridentifiedthesamemaintain- and an alternative mand (during treatment).
ing contingencies as those identified during Occurrences of vocal precursor behavior (e.g.,
functional analyses of severe problem behavior. whining, crying) were scored if they were
Treatments based on the results of a precursor separated by 3 s or more from a previous
analysis were not evaluated. occurrence. We identified precursors to severe
The current study attempted to replicate and problem behavior by asking participants’ care-
extendproceduresusedbySmithandChurchill giverstoidentifyresponsesthatwerelikelytobe
(2002) by conducting functional analyses of observedimmediatelypriortotheoccurrenceof
precursor behavior and by testing whether severe problem behavior and by observing each
FUNCTIONAL ANALYSIS OF PRECURSORS 99
participant to confirm that the precursors responses per minute. Interobserver agreement
identifiedbycaregiversnotonlyprecededsevere was collected during 39% of functional analysis
problem behavior but also occurred in close sessions and 37% of treatment sessions. Agree-
temporal proximity to it. ment was calculated by dividing session time
Aggression (Brent and Pete) was defined as into continuous 10-s intervals and comparing
hitting (contact between a participant’s hand observers’scoresonaninterval-by-intervalbasis.
and another person’s body), kicking (contact The smaller of the two observers’ scores was
between a participant’s foot and another divided by the larger per 10-s interval. The
person’s body), biting (contact between a resulting quotients then were summed, divided
participant’s teeth and another person’s body), by the number of observation intervals, and
scratching (contact between a participant’s multiplied by 100%. During the functional
fingernail and another person’s skin), throwing analyses, mean interobserver agreements for
objects at others, and attempts to make contact precursor and severe problem behavior, respec-
that were dodged or blocked by others. tively, were 99% (range, 93% to 100%) and
Masturbation (Tom) was defined as contact 99.7% (range, 97% to 100%). During baseline
between a participant’s hand and his genitals and treatment sessions, mean agreements for
(either on top of or underneath clothing). precursor behavior, severe problem behavior,
Inappropriate grabbing (Tom) was defined as andalternativemandswere97%(range,85%to
contact or attempted contact between the 100%), 99.8% (range, 97% to 100%), and
participant’s hand and another person’s genita- 96% (range, 83% to 100%), respectively.
lia. Brent’s precursor behaviors were whining,
crying, complaining (e.g., ‘‘I don’t like this,’’ or Functional Analysis
‘‘Why do I have to do this?’’), and threatening A functional analysis of precursor behavior
to hurt himself or others (e.g., ‘‘I am going to wasconductedwitheachparticipant(Iwataetal.,
killmyself,’’or‘‘Iamgoingtohityou.’’).Pete’s 1982/1994). Alone (Tom only), attention, play,
precursor behaviors were whining, crying, demand, and tangible (Pete only) conditions
statements of toy possession (e.g., ‘‘That’s my were presentedin a multielement design (proce-
toy!’’ or ‘‘Why does he get that toy?’’), and duraldetailsareavailablefromthecorresponding
reaching or grabbing for toys possessed by author).Experimentalcontingencieswereplaced
others. Tom’s precursor behavior was uttering onprecursorbehaviorratherthansevereproblem
sexual statements about a person’s genitalia. behavior. All conditions were 5 min in length
During functional communication training and presented until variables maintaining pre-
(FCT),Brent’salternativemandwasrequestinga cursor behavior were apparent.
teacher’s attention by raising an attention card During all conditions for Brent, the experi-
above his head. Pete’s alternative mand was menters, Brent, other students, Brent’s teacher,
asking for toys appropriately (e.g., ‘‘May I have and aides were present; however, only experi-
that, please?’’). Tom’s alternative mand was menters interacted with him. During all
pressing a button on a necklace to page staff to conditions for Pete, the experimenters, Pete,
providehimwithattention.Mandingwasscored andaconfederatestudentfromPete’sclassroom
(for Tom only) as independent if it occurred were present. The confederate student was
without vocal instruction from an experimenter present because Pete’s aggression was directed
and as prompted if the experimenter vocally towards classmates. Any attempts to harm the
instructed him to press the button. confederate student were blocked to keep the
Datawererecordedindependentlybytrained student safe. The experimenterwas in the room
observers on Palm Pilots and were calculated as with Tom in all conditions except alone.
100 ADEL C. NAJDOWSKI et al.
Figure1. Precursorbehaviors perminute during Brent’s, Pete’s, andTom’s functional analyses.
Results. Results of the functional analyses are highest during thetangiblecondition(M5 2.4
depicted in Figure 1. Only precursors per responses per minute), indicating a tangible
minute are depicted, because none of the items function.
participantsengagedinsevereproblembehavior
during their functional analyses. All of the Treatment
participants showed differentiated responding Individualized interventions involving FCT
in one condition. Precursor behavior for Brent weredesignedbasedonthevariablesidentifiedto
and Tom was highest during the attention maintain precursor behavior during each func-
condition (M 5 1.5 and M 5 1.17 responses tionalanalysis.Effectsoftreatmentsonprecursor
per minute, respectively), indicating an atten- and severe problem behavior were evaluated
tion function. Precursor behavior for Pete was using reversal designs. Baseline and treatment
FUNCTIONAL ANALYSIS OF PRECURSORS 101
sessions were 5 min in length for Pete and toy to Pete for 30 s if Pete manded for the toy
10 min in length for Brent and Tom. Sessions appropriately (i.e., ‘‘May I have that, please?’’).
for Tom eventually were extended to 30 min. In addition, the experimenterpromptedPete to
We also reviewed archived records kept by the mand if he engaged in precursor behavior.
school (for Brent) or asked staff to complete a Reinforcementofthemandwasthinnedusinga
questionnaire regarding the occurrence of severe graduated multiple-schedule arrangement in
problem behavior for Pete and Tom. which periods of reinforcement and extinction
During baseline and the reversal, Brent’s were alternated and signaled by a card held up
teachers were instructed to interact with him as by an experimenter (Hanley, Iwata, & Thomp-
theyusuallydidbygivingBrentattentioninthe son, 2001), initially in 30-s and then in 1-min
form of either a verbal reprimand or redirection intervals. The toy was provided for appropriate
contingentoneitherprecursororsevereproblem manding when the card had Pete’s name on it
behavior.DuringFCT,Brentwasprovidedwith but not when the card had the confederate
a card that read TA (for teacher’s attention). At peer’s name on it.
thestartofeachsession,hewasinstructedtoraise At the start of baseline and reversal sessions,
thecardabovehisheadifhewantedtotalktothe theexperimenterprovidedTomwithanecklace
teacher. The teacher provided Brent with 5 to withabuttonthatpagedtheexperimenterwhen
10 s of attention (e.g., ‘‘Good TA, what do you pressed. The experimenter informed Tom that
need?’’) when he raised the card. The teacher she had work to do but that he could color or
prompted Brent to use his card by saying, do a puzzle. The experimenter provided Tom
‘‘Remember to use your TA card if you need withattentionintheformconcern(e.g.,‘‘Tom,
something,’’ when Brent engaged in precursor please don’t say that.’’) contingent on precursor
behavior. Because administrators of Brent’s behavior and ignored pages and instances of
school expressed the importance of not allowing severe problem behavior.
severeproblembehaviorstooccur,anagreement During FCT, Tom continued to wear the
was made that were severe problem behavior to necklace and was provided with coloring mate-
occur, teachers could take whatever action they rialsandpuzzleswhiletheexperimenterappeared
deemed necessary. to be busy. Occurrences of precursor and severe
During baseline and the reversal, Pete and a problem behavior were ignored, and at the start
confederate student played together with toys of the session, the experimenter prompted Tom
(e.g.,blocks).TheexperimentergavePete2 min topressthebuttononhisnecklaceifhewantedto
ofaccesstoatoyofhischoiceimmediatelyprior talk. Subsequently, the experimenter prompted
to sessions. An experimenter took the toy away Tominitiallytopagehimorherevery3 s,which
fromPeteand gaveittotheconfederatestudent increased contingent on three consecutive pro-
while saying, ‘‘It’s [confederate’s] turn,’’ at the mptedpagesduringwhichnoprecursorbehavior
start of sessions. If Pete attempted to engage in occurred (Worsdell, Iwata, Hanley, Thompson,
aggression towards an experimenter or the & Kahng, 2000). The interval was increased
confederate student, it was blocked by an within and between sessions from 3 s to 10 min
experimenter. If he engaged in precursor behav- (detailsavailablefromthecorrespondingauthor)
ior, the experimenter said, ‘‘Okay, you can have when Tom met this criterion. We increased
it’’ and provided 30-s access to the toy. The session length from 10 to 30 min in the
experimenter took the toy away again and treatment room when Tom reached the 3-min
repeatedthestepsattheendofthe30-sinterval. promptinginterval.Eventually,wealsoconduct-
FCT procedures were identical to baseline ed 30-min sessions in the natural environment
except that the confederate peer delivered the where other clients and staff were present.
102 ADEL C. NAJDOWSKI et al.
Figure2. Responsesperminuteforprecursorbehavior,severeproblembehavior(aggressionforBrentandPeteand
masturbation and grabbing for Tom), alternative mands (Brent and Pete), and prompted and independent alternative
mands(Tom) during baseline (BL),FCT, andextended30-min sessions (Tom) for Brent,Pete,andTom.
Results. Figure 2 depicts the effects of low levels (Ms 5 0.02 and 0.01 responses per
treatment on precursor behavior, severe prob- minute) and use of the card increased (Ms 5
lem behavior, and alternative mands for Brent, 0.3and0.29responsesperminute)duringFCT
Pete, and Tom. Precursor behavior occurred at phases. Brent never engaged in aggression
high levels (Ms 5 2.93 and 4.7 responses per throughout baseline or FCT sessions.
minute) and use of the attention card occurred Precursor behavior occurred at high levels
at low levels (Ms 5 0.13 and 0.1 responses per (Ms 5 3.98 and 1.4 responses per minute) for
minute) for Brent during baseline and reversal, Pete during baseline and the reversal. Aggres-
respectively. Precursor behavior decreased to sion and appropriate manding never occurred
FUNCTIONAL ANALYSIS OF PRECURSORS 103
during baseline, and aggression increased (M 5 (second respondent) during the month follow-
0.5 responses per minute) and appropriate ing the treatment evaluation.
manding decreased (M 5 0.7 responses per
minute) during the reversal. Precursor behavior
DISCUSSION
decreased to lower levels (Ms 5 0.77 and 0.36
responses per minute), aggression never oc- All participants in this study responded with
curred, and appropriate manding increased (Ms differentiation to particular reinforcement con-
5 1.99 and 1.65 responses per minute) during tingencies during functional analyses of precur-
FCT phases. sorbehavior.Thus,clearoperantfunctionswere
Precursor behavior for Tom occurred at high identified and used to design individualized
levels (Ms 5 3.83 and 7.1 responses per treatments. Overall, these treatments were
minute), and he never used the pager during found to eliminate precursor behavior and
baseline and the reversal. Precursor behavior possibly prevented occurrences of severe prob-
occurredatzeroorlowlevels(Ms50and0.03 lem behavior.
responses per minute), and prompted (Ms 5 Althoughthemechanisms responsible forthe
1.77 and 0.6 responses per minute) and nonoccurrence of severe problem behavior
unprompted (Ms 5 6.67 and 1.29 responses remain unclear, some potentially influential
per minute) paging increased during FCT variables can be identified. For example,
phases. Precursor behavior remained low (Ms potential sources of reinforcement were provid-
5 0.03 and 0.01 responses per minute), edforprecursorbehaviorduringtestconditions
prompted paging was low (Ms 5 0.01 and of the functional analyses and during baseline
0.05 responses per minute), and independent conditions of treatment; therefore, it may have
paging remained high (Ms 5 1.39 and 0.91 been unnecessary for participants to engage in
responses per minute) during extended FCT severe problem behavior to obtain reinforce-
sessions in the treatment room and in the ment. Thus, reinforcement for precursor be-
natural environment. Tom never engaged in havior may have functioned as differential
masturbation or inappropriate grabbing reinforcement of alternative behavior, whereby
throughout baseline or FCT sessions. the precursor behavior was strengthened and
severe problem behavior was eliminated (Smith
Pre- and Posttreatment Measures of Severe & Churchill, 2002). Furthermore, precursor
Problem Behavior behavior may have required less effort and
Data collected by Brent’s teachers and aides inflicted less pain (Lalli et al., 1995), thereby
revealed that mean number of instances of producing a bias in favor of precursor behavior
aggression per day was 1.75 during the 4 days over severe problem behavior.
prior to the treatment evaluation and zero The current results are consistent with an
during the 4 days following the treatment account that precursor and severe problem
evaluation.Datafrom aquestionnaireindicated behavior are hierarchically occurring members
that Pete engaged in aggression a few times per ofthesameresponseclass.Previousresearchhas
week during the month prior to intervention shown that reinforcement of an early response
and not at all during the last 4 days of school. in a hierarchy of responses in thesame response
Data from a questionnaire indicated that Tom class may suppress later responses in the
engaged in masturbation a few times per week hierarchy (Lalli et al., 1995); thus, a reasonable
and grabbing at least once per week during the accountofthenonoccurrenceofsevereproblem
month prior to intervention, and he engaged in behavior during precursor functional analyses
masturbation once and grabbing or attempting and treatment conditions is that responses that
to grab either never (one respondent) or once occurred earlier in the response-class hierarchy
104 ADEL C. NAJDOWSKI et al.
produced reinforcement. However, these results acquired for Brent and Pete because data were
are tentative and should be interpreted with not collected separately on independent and
caution because experimental analyses of severe prompted mands.
problem behavior were not conducted, and The current study extends previous research
there are several alternative accounts for the on precursor behavior by implementing an
nonoccurrence of severe problem behavior. intervention based on the outcomes of a
It is possible that the relevant antecedents functional analysis of precursor behavior that
and consequences were absent during the reduced precursor behavior and was associated
functional analysis and treatment conditions. withzerolevelsofsevereproblembehavior.The
Thus, although precursor behavior showed datafromthecurrentinvestigationhighlightthat
differentiation during functional analysis and an especially important area for future research
treatment conditions, it is possible that precur- will be the resolution of measurement issues
sor behavior was otherwise unrelated to severe related to inferring the effects of precursor-
problem behavior. Alternatively, it is possible based procedures on problem behavior.
that the precursors and alternative mands
competed with severe problem behavior.
REFERENCES
Given that we conducted only a preliminary
evaluationoftherelationbetweenprecursorand Borrero,J.C.,Vollmer,T.R.,Wright,C.S.,Lerman,D.
severe problem behavior in the current study, a C.,&Kelley,M.E.(2002).Furtherevaluationofthe
roleofprotectiveequipmentinthefunctionalanalysis
number of strategies for future research may be
ofself-injuriousbehavior.JournalofAppliedBehavior
suggested.First,effortsshouldbemadetocollect Analysis, 35,69–72.
as much data on severe problem behavior as Derby, K. M., Wacker, D. P., Sasso, G., Steege, M.,
Northup,J., Cigrand,K.,etal.(1992).Brieffunctional
possible prior to and following treatment.
assessmenttechniquestoevaluateaberrantbehaviorin
Another strategy would be to conduct probe anoutpatientsetting:Asummaryof79cases.Journalof
sessions in which reinforcement is temporarily AppliedBehaviorAnalysis,25,197–209.
withheld for all behavior, which may occasion Grow,L.L.,Kelley,M.E.,Roane,H.S.,&Shillingsburg,
M.A. (2008). Utility of extinction-induced response
the reappearance of behavior in the same
variability for the selection of mands. Journal of
responseclassasprecursorbehavior.Occurrences AppliedBehaviorAnalysis,41,15–24.
of severe problem behavior would suggest a Hanley,G.P.,Iwata,B.A.,&Thompson,R.H.(2001).
Reinforcementschedulethinningfollowingtreatment
common source of control between severe
with functional communication training. Journal of
problem behavior and precursor behavior. AppliedBehaviorAnalysis,34,17–38.
An additional limitation in the current Harding,J.W.,Wacker,D.P.,Berg,W.K.,Barretto,A.,
Winborn, L., & Gardner, A. (2001). Analysis of
procedures is that we interviewed teachers and
response class hierarchies with attention-maintained
staff to identify precursor behaviors, which may
problem behaviors. Journal of Applied Behavior
have resulted in the identification of responses Analysis, 34, 61–64.
that were correlated with, but did not reliably Iwata,B.A.,Dorsey,M.F.,Slifer,K.J.,Bauman,K.E.,&
Richman,G.S.(1994).Towardafunctionalanalysis
precede, severe problem behavior. More formal
ofself-injury.JournalofAppliedBehaviorAnalysis,27,
descriptive procedures to verify that a substan- 197–209.(ReprintedfromAnalysisandInterventionin
tial proportion of severe problem behavior is DevelopmentalDisabilities,2,3–20,1982)
Kahng, S., & Iwata, B. A. (1999). Correspondence
preceded by the forms of behavior identified as
between outcomes of brief and extended functional
precursors should be investigated. Another
analyses. Journal of Applied Behavior Analysis, 32,
limitation is that we conducted only one 149–159.
reversal session with Brent and two reversal Lalli,J.S.,Mace,F.C.,Wohn,T.,&Livezey,K.(1995).
Identification and modification of a response-class
sessions with Pete. Finally, it is impossible to
hierarchy. Journal of Applied Behavior Analysis, 28,
determine when independent manding was 551–559.
FUNCTIONAL ANALYSIS OF PRECURSORS 105
Le,D.D.,&Smith,R.G.(2002).Functionalanalysisof Sprague, J. R., & Horner, R. H. (1992). Covariation
self-injury with and without protective equipment. within functional response classes: Implications for
JournalofDevelopmentalandPhysicalDisabilities,14, treatment of severe problem behavior. Journal of
277–290. Applied BehaviorAnalysis,25, 735–745.
Neef, N. A., & Iwata, B. A. (1994). Current research in Vollmer,T.R.,Marcus,B.A.,Ringdahl,J.E.,&Roane,
functional analysis methodologies: An introduction. H. S. (1995). Progressing from brief functional
JournalofAppliedBehavior Analysis,27,211–214. assessments to extended experimental analyses in the
Parrish,J.M.,Cataldo,M.F.,Kolko,D.J.,Neef,N.A., evaluation of aberrant behavior. Journal of Applied
& Egel, A. L. (1986). Experimental analysis of Behavior Analysis,28,561–576.
response covariation among compliant and problem Worsdell, A. S., Iwata, B. A., Hanley, G. P., Thompson,
behaviors. Journal of Applied Behavior Analysis, 19, R.H.,&Kahng,S.W.(2000).Effectsofcontinuous
241–254. andintermittentreinforcement forproblembehavior
Skinner, B. F. (1969). Contingencies of reinforcement: A
duringfunctionalcommunicationtraining.Journalof
theoretical analysis. New York: Appleton-Century-
Applied BehaviorAnalysis,33, 167–179.
Crofts.
Smith,R.G.,&Churchill,R.M.(2002).Identificationof
environmental determinants of behavior disorders Received November 12,2004
through functional analysis of precursor behaviors. Final acceptanceJuly 12,2007
JournalofAppliedBehavior Analysis,35,125–136. Action Editor,Richard Smith