ERIC EJ931964: Replications and Extensions in Arousal Assessment for Sex Offenders with Developmental Disabilities

2011·0.32 MB·English

by  ERIC

JOURNALOFAPPLIEDBEHAVIORANALYSIS 2011, 44, 369–373 NUMBER2 (SUMMER2011)
REPLICATIONS AND EXTENSIONS IN AROUSAL ASSESSMENT FOR
SEX OFFENDERS WITH DEVELOPMENTAL DISABILITIES
JORGE R. REYES
WESTFIELDSTATEUNIVERSITY
TIMOTHY R. VOLLMER
UNIVERSITYOFFLORIDA
AND
ASTRID HALL
THESEGUINUNIT
Three adult male sex offenders with developmental disabilities participated in phallometric
assessmentsthatinvolvedrepeatedmeasuresofarousalwhenexposedtovariousstimuli.Arousal
assessment outcomes were similar to those obtained by Reyes et al. (2006). Additional data-
analysis methods provided further information about sexual preferences, thus replicating and
extending previous research. The results provide preliminary data for establishing a preference
gradientbyage.Implicationsfortheuseofrepeatedmeasuresandpreferencegradientsinarousal
assessments are discussed.
Keywords: arousal assessment, developmental disabilities, repeated measurement
_______________________________________________________________________________
Previous research has shown that within- Hanson & Morton-Bourgon, 2004; Quinsey,
subject phallometric assessments (i.e., repeated Chaplin, & Carrigan, 1980) and provide
measures of arousal to various stimuli using the direction for the development of individualized
penile plethysmograph) can produce differenti- treatments.
ated outcomes for sex offenders with develop- The majority of previous sex offender
mental disabilities (Reyes et al., 2006). Phallo- research contained limitations that are not
metric assessments are commonly used in sex relevant in typical repeated measures research.
offender evaluations to evaluate arousal to For example, Marshall and Fernandez (2003)
various forms of stimuli (e.g., auditory or conducted group comparisons across offenders
visual) or various categories of stimuli (e.g., and nonoffenders or among different types of
child vs. adult). These assessments may allow offenders. Other studies included a limited
the identification of the relevant conditions number of exposures to the stimuli used in the
under which an individual may be more likely assessments (e.g., Lalumiere & Harris, 1998;
to commit additional offenses (e.g., Barbaree & Laws & Osborn, 1983). The use of repeated
Marshall, 1988; Hanson & Bussiere, 1998; measures may attenuate the unknown deleteri-
ous effects of extraneous variables (e.g., illness,
lack of sleep, recent ejaculation, etc.) that may
This research was supported by a contract with the
Florida Agency for Persons with Disabilities. However, confound experimental results. In contrast to
opinions do not necessarily reflect those of the funding much of the previous research, Reyes et al.
agency. We thank Kimberly Sloman for helping with
(2006) conducted within-subject analyses of
variousaspects ofdata collectionandanalysis.
Correspondence concerning this article should be conditions that may produce arousal. Their
addressed to Jorge R. Reyes, Department of Psychology, results suggested that repeated phallometric
Westfield State University, 577 Western Ave., Westfield,
assessments for sex offenders with developmen-
Massachusetts 01086(e-mail: [email protected]).
doi:10.1901/jaba.2011.44-369 tal disabilities are useful for establishing the
369
370 JORGE R. REYES et al.
conditions under which arousal is likely. The and are designed specifically for use in arousal
purpose of the present study was to replicate assessments.Eachclipdepictedamaleorfemale
and extend the assessment methodology of of a particular age (kindergarten, 6 to 7 years, 8
Reyes et al. by evaluating the utility of to 9 years, teen, and adult) wearing a bathing
additional data-analysis techniques to provide suit and engaging in the same sequence of
more detailed information about the arousal behavior. The stimuli also included neutral
assessment outcomes. scenes (e.g., boating and fishing).
Allsessionswereconductedinaroom(2.1m
by 2.3 m) that contained a recliner, a 27-in.
METHOD
television used to present the video clips, a
Participants
camera that provided a live video feed of the
The first three individuals referred for
participant from above the shoulders, and a
assessment (including Participant 37 from
metal lap tray (to prevent the participant from
Reyesetal.,2006)bytheirresidentialtreatment
receiving visual feedback of arousal). The
facility participated in this study. Staff at the
technicians were located in an adjacent room
treatment facility were interested in conducting
that contained a computer, a video monitor
a new arousal assessment for Participant 37
showing the live feed, and a video cassette
given that his initial assessment had been
recorder used to present the assessment stimuli.
conducted approximately 8 years before the
Prior to beginning the assessment, a techni-
current assessment. All of the individuals had
cian conducted a baseline measure of penile
been placed in the residential treatment facility
circumference (toensure proper selection of the
and ordered by the court to participate in
straingaugeandtoestablishabaselinemeasure)
standard assessment procedures (charges within
and instructed each participant to attach the
the judicial system against all three participants
gauge properly. Before each daily session, a
had been dismissed due to incompetence to
technician calibrated the strain gauge with a
stand trial, despite accusations of multiple
system of computer readings, placed an absor-
offenses against young children). Approval by
bent disposable pad on the recliner, and placed
both the facility and university institutional
the calibrated gauge on the pad. The technician
review boards was obtained to evaluate and
provided each participant with the opportunity
report on the clinical process.
to use the bathroom immediately prior to the
Arousal Assessments sessions. Next, the technician instructed the
All of the procedures for the arousal participant to enter the session room, privately
assessments were identical to those used in pull down his pants and underwear to his
Reyes et al. (2006). Assessments consisted of ankles, sit in the chair, and attach the gauge as
measuring arousal levels, via a penile plethys- previously instructed. Once the gauge was
mograph, during exposure to deviant (typically attached, the participant placed the metal tray
defined as stimuli that include individuals onhislapandplacedhishandsonthetray.The
below the legal age limit for sexual activity technician turned on the video camera to start
[18 years]), nondeviant, and neutral stimuli. the video feed of the participant after checking
Arousal was measured with a circumferential thedatastreamforanyanomaliesthatmayhave
mercury-in-rubber strain gauge connected to a indicated a problem with the gauge.
computerizedinterface,whichallowedmeasure- The technician waited until detumescence
ment of real-time changes in penile circumfer- (i.e., decrease in arousal levels) was obtained (if
ence. The stimuli, which consisted of 11 video applicable) and stability criteria (no changes
clips (2.5 min each), are commercially available greater than 5 mm for a period of 1 min) were
AROUSAL ASSESSMENT 371
met before presenting the video clips in one of outcomes for the female stimuli show that
three predetermined orders. The technician arousal levels were lowest to the nondeviant age
compared the penile circumference (in milli- category (female adult) and highest to the three
meters) at the beginning of the video clip to youngest age categories, with the highest levels
the highest point reached during the video clip of arousal occurring to the female kindergarten
to determine the change in penile circumfer- stimulus.
ence in the presence of each stimulus. This Figure 1 also shows the arousal outcomes for
process was repeated until all 11 stimuli had Participant 5. High levels of arousal were
been presented. At the conclusion of the obtained for several of the age categories for
session, the participant removed the gauge, both the male and female stimuli. In general,
washed his hands, dressed, and exited the thearousallevelswerehighandundifferentiated
session room. One session was conducted per for the male stimuli (kindergarten through 8 to
day, and each session involved one presentation 9 years) and low for the male teen and adult
of each of the 11 stimuli. Sessions were stimuli. For the female stimuli, arousal levels
typically conducted three to five times per were differentially higher to the female kinder-
week. Sessions were continued until clear or garten stimulus, with high levels of arousal also
informative outcomes were obtained.
occurring for the female age 6 to 7 years, age 8
to 9 years, and teen on some sessions. There
Data Analysis
were two distinct patterns of responding for the
Arousal outcomes, measured by change in
male and female stimuli. Arousal levels for the
penile circumference, were expressed (a) across
male stimuli were higher as age categories
successive presentations of stimuli (presented to
increased until it peaked to the male age 8 to
the participants in a multielement design) as in
9 year stimulus and then dropped off for the
Reyes et al. (2006) and (b) as average level of
male teen and adult stimuli. The outcomes for
arousal obtained for each stimulus across the
the female stimuli showed a dissimilar pattern
entire assessment.
of responding, in that the arousal levels
decreased as the age categories increased, with
RESULTS AND DISCUSSION the highest levels occurring to the female
kindergarten stimulus and the lowest levels
Theresultsforallparticipantsaredepictedin
occurring to the female adult stimulus.
Figure 1. For Participant 43, arousal levels to
the male stimuli were lower than arousal levels The results for Participant 37 are also shown
to several of the female stimuli (left). Further- in Figure 1. Arousal levels to the male stimuli
more,arousallevelstothedeviantagecategories were uniformly low and undifferentiated. The
(i.e., kindergarten through 8 to 9 years) were arousal levels for the female stimuli were also
higher than to the nondeviant age categories. low, even with the exception of slightly higher
The right panelshows a summary of the overall levels obtained for the female adult stimulus.
levels of arousal obtained to the male and Arousallevelsforthefemalestimuliwerehigher
female stimuli and by age category across the than for the male stimuli but represent an
entire assessment, plotted as the average overall low level of arousal, especially compared
millimeter change obtained in each age and to previouslyreportedarousal outcomes forthis
gender category. Although the overall arousal participant (Reyes et al., 2006).
levels to the male stimuli were low, the lowest Overall, results of this study replicate and
arousal occurred to the nondeviant adult male extend those of Reyes et al. (2006). Three
stimulus. The arousal levels to the other male general patterns of arousal outcomes were
agecategorieswerelargelyundifferentiated.The obtained: differentiated deviant arousal (Partic-
372 JORGE R. REYES et al.
Figure 1. Assessment results for Participant 43 (top), Participant 5 (middle), and Participant 37 (bottom).
Millimeterchangeisdepictedintheleftpanels.Averagemillimeterchangeisshownintherightpanels;barsrepresent
standarderror.
ipants 43 and 5 for female stimuli), undiffer- evidence of habituation across repeated expo-
entiated deviant arousal (Participant 5 for male sures to the stimuli, and therefore provide
stimuli), and no deviant arousal (Participant additional support for the utility of repeated
37). None of the arousal outcomes suggested measurement assessments for this population.
AROUSAL ASSESSMENT 373
TheoutcomesforParticipant37arethemost because the rate of sexual offending does not
difficult to interpret, given that little or no appeartobedecreasingtoanysignificantdegree
arousal occurred to the deviant or nondeviant (U.S. Department of Health and Human
stimuli, and the outcomes obtained in the Services, 2008).
current study differed from previous assess-
ments for the same participant. In the Reyes et
REFERENCES
al.(2006)study,Participant37showedpatterns
Barbaree,H.E.,&Marshall,W.L.(1988).Deviantsexual
of both differentiated deviant arousal towards
arousal,offensehistory,anddemographicvariablesas
females ranging in age from 6 to 9 years and
predictors of reoffense among child molesters.
very little arousal to the female adult stimulus. Behavioral Sciences & theLaw,6, 267–280.
The current assessment results suggested no Hanson, R. K., & Bussiere, M. T. (1998). Predicting
relapse:Ameta-analysis ofsexualoffenderrecidivism
differentiated deviant arousal to the male or
studies. Journal of Consulting and Clinical Psychology,
female stimuli, but did show slightly elevated 66,348–362.
levels of arousal to the female adult stimulus. Hanson,R.K.,&Morton-Bourgon,K.(2004).Predictors
of sexual recidivism: An updated meta-analysis. (Cat.
Although the overall levels of arousal to the
No. PS3-1/2004-2E-PDF). Public Works and Gov-
female adult stimulus were modest, this still
ernment Services Canada.
potentially represents an important change in Lalumiere, M. L., & Harris, G. T. (1998). Common
the sexual preferences for this participant that questions regarding the use of phallometric testing
with sexual offenders. Sexual Abuse: Journal of
may have been undetected without using a
Research and Treatment, 10,227–237.
repeated measures approach to assessment. Laws,D.R.,&Osborn,C.A.(1983).Howtobuildand
Wealsoanalyzedanddepictedthedataasthe operate a behavioral laboratory to evaluate and treat
sexualdeviance.InJ.G.Greer&I.R.Stuart(Eds.),
average millimeter change per category across
The sexual aggressor: Current perspectives on treatment
theentire assessment. Toour knowledge, this is
(pp. 293–335). New York:Van NostrandReinhold.
the first reported example of arousal assessment Marshall,W.L.,&Fernandez,Y.M.(2003).Phallometric
databeinganalyzedasagradient(i.e.,arousalas testing with sexual offenders: Theory, research, and
practice.Brandon, VT:Safer Society Press.
a function of age categories). Analyzing the
Quinsey, V. L., Chaplin, T. C., & Carrigan, W. F.
results in this way provided additional infor- (1980). Biofeedback and signaled punishment in the
mation about individual sexual preferences and modification of inappropriate sexual age preferences.
Behavior Therapy,11,567–576.
helped to clarify the overall patterns. They also
Reyes, J. R., Vollmer, T. R., Sloman, K. N., Hall, A.,
generated what could be considered a rank
Reed, R., Jansen, G., et al. (2006). Assessment of
order of sexual preferences by showing how deviant arousal in adult male sex offenders with
arousal levels compare across stimulus classes developmentaldisabilities.JournalofAppliedBehavior
Analysis,39,173–188.
(i.e., age and gender). The repeated measures
U.S. Department of Health and Human Services. (2008).
approach and arousal gradient analysis may Child maltreatment. Retrieved from http://www.acf.
provide researchers with the necessary informa- hhs.gov/programs/cb/stats_research/index.htmcan.
tion to begin designing and evaluating effective
Received September 2,2009
treatments for this population. These assess-
Final acceptanceSeptember 16, 2010
mentsandtreatmentsseemespeciallyimportant Action Editor,Michael Kelley