temerarment autism

8/13/2019 Temerarment Autism

1/85

SSStttooonnnyyyBBBrrrooooookkkUUUnnniiivvveeerrrsssiiitttyyy

The official electronic file of this thesis or dissertation is maintained by the UniversityLibraries on behalf of The Graduate School at Stony Brook University.

AAAlll lll RRRiiiggghhhtttsssRRReeessseeerrrvvveeedddbbbyyyAAAuuuttthhhooorrr...


2/85

Temperament Intervention for Problem Behavior in Children with Autism Spectrum

Disorders

A Dissertation Presented

by

Lauren Adamek

in Partial ulfillment of the

!e"uirements

for the De#ree of

Doctor of Philosophy

in

Clinical Psycholo#y

Au#ust $%&&


3/85

Stony Brook University

The 'raduate School

Lauren Adamek

(e) the dissertation committee for the above candidate for the

Doctor of Philosophy de#ree) hereby recommend acceptance of this dissertation*

Dr. Daniel Klein Dissertation Advisor

Professor, Department of Psychology

Dr. Joanne Davila hairperson of Defense


Dr. !verett "aters


Dr. Shana #ichols

linical Director and $esearcher, %ay J. &indner enter for A'tism and

Developmental Disa(ilities

This dissertation is accepted by the 'raduate School

Lawrence +artin Dean of the 'raduate School

ii


4/85

Abstract of the Dissertation

Temperament Intervention for Problem Behavior in Children with Autism Spectrum

Disorders

by

Lauren Adamek

Doctor of Philosophy

Clinical Psycholo#y

Stony Brook ,niversity

$%&&

Problem behavior is a ma-or barrier to #ood "uality of life for families who have

children with Autism Spectrum Disorders .ASD/* The concept of modifyin# the

environment to produce a better match with a child0s temperament is commonly used to

inform interventions in the child development field* 1owever) temperament has not yet

been inte#rated into problem behavior interventions for children with ASD) nor have

temperament2based strate#ies been evaluated in a systematic way* The purpose of the

present study was to employ temperament2based interventions to modify problematic

environmental conte3ts so that they are a better fit for the temperament styles of hi#hly

e3traverted children and to evaluate these interventions to determine whether they result

in a reduction of problem behavior and an increase in "uality of life* Si3 hi#hly

e3traverted children with ASD who display problem behavior participated* Assessments

of problem behavior) e3traversion) and "uality of life were conducted and parents were

tau#ht to miti#ate challen#in# situations to make them a better fit for their child0s

temperament* A multiple baseline e3perimental desi#n was used to evaluate intervention

iii


5/85

effects for specific hi#h priority conte3ts* !esults indicated that modifyin# the

environment to better fit a child0s temperament was associated with decreased problem

behavior and increased percenta#e of task steps completed correctly* Subse"uent to the

e3perimental demonstration) a clinical e3tension of the intervention methodolo#y was

applied for each child to an additional problem conte3t in order to further enhance

intervention benefits* T2test results of ancillary pre and post intervention measures

indicate that intervention was associated with a decrease in overall problem behavior*

This research hi#hli#hts the importance of understandin# temperament when assessin#

and treatin# problem behavior in children with ASD*

iv


6/85

Dedication

This dissertation is dedicated to Dr. Edward (Ted) Carr. Ted was a brilliant

thinker and researcher, whose work forever changed the lives of those with autism and

their families. Ted always thought ahead of the field by questioning the status quo and

diligently seeking to understand fields outside of autism, in order to more fully solve

roblems related directly to autism. !t was with Ted"s guidance that ! sought to

understand temerament and its relationshi to roblem behavior in children with #$D.

Ted taught me to think broadly, write clearly, and do work that will make a ositive

imact on others. ! hoe that this dissertation, and work to come, will be a tribute to the

life of Ted Carr.


7/85

Table of Contents

List of Tables....................................................................................................................viii

List of Figures.....................................................................................................................ix

Acknowledgements..............................................................................................................x

I. Introduction..............................................................................................................1

Overview of the Temerament Construct................................................................!

"xtraversion#$urgenc% and "xternali&ing 'roblem (ehavior in )eurot%ical

*outh........................................................................................................................+"xtraversion#$urgenc% in *outh with A$,.............................................................-

"xtraversion#$urgenc% and "xternali&ing (ehavior in *outh with A$,................

Intervention for 'roblem (ehavior/ Temerament 'ersectives.............................0

$ecific Comonents of Temerament(ased Intervention....................................2The 'resent $tud%..................................................................................................13

II. 4ethod...................................................................................................................11

'articiants.............................................................................................................11

'rocedure...............................................................................................................1+Comletion of 'reIntervention 4easures.................................................1+

Identification of Intervention Contexts.....................................................1-

(aseline for "xerimental Contexts..........................................................15's%choeducation about Temerament.......................................................16

Overview of Intervention for "xerimental Contexts...............................16

Clinical "xtensions....................................................................................!Comletion of 'ostIntervention 4easures...............................................!2

III. 7esults...................................................................................................................!6

"xerimental Contexts...........................................................................................!6

Clinical Contexts....................................................................................................+3

Interrater 7eliabilit%...............................................................................................+1Ancillar% 4easures................................................................................................+1

I8. ,iscussion..............................................................................................................++

Intervention "fficac%9.........................................................................................++

Concetual Issues..................................................................................................+"cological 8alidit%................................................................................................+0

Future ,irections...................................................................................................+5

Tables.....................................................................................................................+2

vi


8/85

Figures....................................................................................................................--

7eferences..............................................................................................................-6

Aendix ...............................................................................................................2

vii


9/85

List of Tables

Table 1 'articiant Characteristics at (aseline.......................................................+2

Table ! Task Anal%ses of "xerimental Contexts. ................................................+6

Table + Intervention Fidelit% Checklists for "xerimental Contexts.....................-1

Table - ,ifferences between 'reIntervention and 'ostIntervention 4easures...-+

viii


10/85

List of Figures

Figure 1 The ercent of activit% stes comleted and latenc% to session terminationfor the first three articiants :grou 1; during baseline and

intervention................................................................................................-

Figure ! The ercent of activit% stes comleted and latenc% to session termination

for the second three articiants :grou !; during baseline and

intervention................................................................................................-0

Figure + The ercent of activit% stes comleted for all six articiants :grou 1

and !;.........................................................................................................-5

Figure - The latenc% to session termination for all six articiants :grou 1 and !;

during clinical contexts..............................................................................-2

ix


11/85

Acknowledgements

I wish to acknowledge the following people, without whom, the successfulcompletion of this project would not have been possible.

A mentor and friend, Ted Carrtaught me how to be a scientist, and inspired me to

use science to improve the lives of those with disabilities and their families. Countlesshours of excited conversation about temperament-based intervention led to this project.

Shana Nichols. When I needed guidance with this project, hana stepped in

without hesitation. hana is trul! rare in that she is an expert researcher and clinician, andtherefore, her feedback throughout the stud! was expansive and invaluable. "urther,

hana was not just a mentor for this project, but provided me with career development

advice, and was a supportive figure both professionall! and personall!.

Dan Klein. I have alwa!s admired #an for his expansive knowledge, well-articulated feedback, and kindness. It was m! pleasure to work more closel! with him on

this dissertation, and his comments on this project, and previous temperament projects,

made them better.

Joanne Davila. $oanne has supported m! growth and development throughcoursework, clinical supervision, and now b! providing feedback on m! research. %er

insight was extremel! beneficial.Everett Waters. &verett has provided feedback on several temperament projects,

and his comments reflect immense thought and prompt me to think more deepl! about the

constructs.

Caitlin Walsh. Caitlin provided intervention to a famil! in this stud!. he was an

excellent therapist, worked diligentl!, and alwa!s strived to provide top 'ualit! care and

research integrit!. Caitlin provided several intervention ideas for other families during

our length!, weekl! talks.Samara Tetenbaum. amara and I became a (dissertation team) and both

motivated one another to continue to strive for excellence. amara has incredible creative

abilit! when designing high-intensit! interventions or visual aids. *hese contributionsmade the intervention strategies more desirable to the children in the stud!.

Lauren Moskowit. +! consultations with auren when first planning this

intervention stud! were imperative to maintaining research integrit! while also achievingefficienc!. "urther, auren is more familiar with the past and current autism literature

that an!one I know and she alerted me to relevant studies.

Sarah !arber"#uest$ Sara !u%%erd$ Care& Dowlin'$ and Jen Tomlinson.*hese

women were with me ever! step of the wa!, encouraging me to keep persevering fromstats exams to first clients to committee meetings, and so much more. *he unconditional

love and support given to me b! these great friends has allowed me to become a more

confident researcher, a better advocate for m! clients, and a more balanced individual.Sam Cook. "or this project, m! fianc, am, pla!ed the role of statistics

consultant and patientl! discussed statistics implications with me. +ore importantl!, am

considers m! clinical and research goals e'ual to his own career goals, and has providedthe utmost support for an! new endeavors.

Mitch and Linda (damek. +! parents have taught me the importance of

education, hard work, and kindness/ three ideals that were necessar! for m! graduate

career.


12/85

The )amiles.ix special families welcomed me in their homes and inspired me withtheir dedication and love for their children. *hese families, and the families in the two

previous temperament and A# studies, have contributed to the bod! of knowledge thatwill hopefull! help to improve the 'ualit! of life of man! children with autism and their

families.


13/85

I. Introduction

Problem behaviors, such as aggression, tantrums, self-injury, and property

destruction are commonly displayed by people with developmental disabilities (Emerson

et al., 2!". #uch behavior can negatively impact education, sociali$ation, community

inclusion and employment (%ruinin&s, 'ill, )orreau, !*++ oegel, oegel,

unlap, !**/". %ecause of its deleterious effects on 0uality of life, problem behavior has

been a major focus of research and intervention. 1he reduction of problem behavior is

essential for individuals with developmental disabilities to achieve valued outcomes and

a high 0uality of life.

n emerging model of intervention for problem behavior in the disabilities field

focuses on enhancing conte3tual fit. 4onte3tual fit refers to the degree of match between

an individual5s competency and the performance re0uirements of the environment (4arr,

4arlson, 6angdon, )agito )c6aughlin, 7arbrough, !**+". Poor conte3tual fit results

when an individual5s s&ills do not match the demands of the environment. s a result, the

individual may be more li&ely to display problem behavior. 1his model of conte3tual fit

is useful because it suggests that one can intervene either with respect to the individual

(competency building" or with respect to the environment (environmental modification".

concept similar to conte3tual fit, referred to as goodness-of-fit, is discussed in

the developmental literature. 8oodness-of-fit is defined by the degree of match between a

child5s temperament and the performance re0uirements of the environment (4hess

1homas, !**!". 9hen the child5s temperament is a poor fit with the environment,

problem behavior is more li&ely to occur. #ince temperament is moderately stable

throughout childhood (:ovosad 1homan, !*** ;oberts el


14/85

0uite plausible that temperament may be relatively resistant to change. 1hus, intervention

efforts might focus more profitably on environmental modification that creates a better

match between the child5s temperament and his=her environment. 1o date, temperament-

based interventions have not been systematically applied to children who have #. 1his

study attempts to address this gap by investigating the impact of temperament-based

intervention strategies on problem behavior and 0uality of life of children who have

#.

Overview of the Temperament Construct

1emperament is a major focus of the literature on child development. 4urrently,

theorists view temperament as individual differences in a child5s response to various

situations in his or her self-regulation of attention, emotion, and activity (;othbart

%ates, 2/". central tenet in the study of temperament is that genetic influences

greatly impact the e3pression of temperament traits. In the field of behavior genetics,

heritability estimates are used to e3press the proportion of the observed temperament trait

that is due to genetic factors. >ne large study of twins found the overall heritability of

temperament to be about .?, with individual temperament traits ranging from a

heritability of .!+ to .@@ (>nis$c$en&o, Aawad$&i, #trelau, ;iemann, ngleitner,

#pinath, 22". nother crucial finding in the study of temperament is the moderate

stability of temperament throughout early childhood. recent meta-analysis yielded

cross-time correlations of .?2 from B to ?.* years of age (;oberts el


15/85

to situations with negative emotional states. Effortful control is defined as the efficiency

of e3ecutive attention, including the ability to inhibit a dominant response and=or to

activate a subdominant response, to plan, and to detect errorsF (;othbart %ates, 2/,

p. !B/". E3traversion=surgency refers to a given individual5s tendency to be more active,

enjoy higher intensity activities, be more impulsive, and act less shy than his=her peers

(;othbart, et al., 2!". Efficacy of temperament based-interventions has not yet been

evaluated with children who have # therefore, it is important and clinically

meaningful to investigate the efficacy of such interventions with respect to one or more

temperament dimensions. 1his study focuses on the temperament dimension of

e3traversion=surgency.

Extraversion/Surgency and Externalizing Problem ehavior in !eurotypical "outh

1he factor of e3traversion=surgency, as well as individual subscales that

contribute to this factor, have been found to predict problem behavior (e.g., aggression,

delin0uent behavior" in neurotypical youth. In one study, aggression was positively

related to high e3traversion in children ages B-@ years old (;othbart et. al., 2!". In a

group of male and female college students, greater e3traversion predicted a greater

li&elihood of assaulting others (Edmunds, !*@@". In addition, female college students who

were more e3traverted were more li&ely to engage in verbal and indirect aggression, and

display irritability. #everal specific constructs associated with e3traversion=surgency are

predictive of e3ternali$ing problem behavior. In one sample of children, high impulsivity

predicted e3ternali$ing problem behavior (Eisenberg et al., 2?". In another sample,

children who preferred higher intensity activities were more li&ely to engage in

e3ternali$ing problem behavior (>rmel et al., 2?". dditionally, in families with

B


16/85

parental psychopathology, children with a high activity level were more li&ely to show

behavior problems ()un, Dit$gerald, ther research has

shown that children with # have higher e3traversion=surgency, and in particular,

higher activity level than neurotypical children ('epburn #tone, 2/ 8aron, %ryson,

Awaigenbaum, #mith, %rian, ;oberts, et al., 2*".

G


17/85

Extraversion/Surgency and Externalizing ehavior in "outh with #S$

8iven the breadth of literature on temperament in developmental psychology, and the

growing literature on group temperament differences in #, one might presume that the

lin& between temperament and problem behavior has also been e3amined in the

disabilities field, however, little research on these potential lin&s has been conducted.

>ne recent study e3amining the relationship between temperament and problem behavior

found that effortful control and negative affectivity, but not e3traversion=surgency

predicted e3ternali$ing problem behavior in children between + and !/ years of age with

high-functioning autism, as measured by the %ehavior ssessment #cale for 4hildren

(#chwart$ et al., 2*". In another study, greater e3traversion=surgency was found to

predict problem behavior in children between B and @ years of age who had #

(dame&, :ichols, 1etenbaum, %regman, Pon$io, 4arr, 2!". In addition, children

with a higher activity level, as well as children who preferred high intensity pleasure

activities, were more li&ely to show problem behavior. n observational study also found

that children ages B to @ years showed problem behavior when as&ed to participate in low

intensity situations (dame& 4arr, 2!". #pecifically, two children low in

e3traversion and two children high in e3traversion were observed when as&ed to

participate in low intensity activities (e.g., boo&, pu$$le, board game" or high intensity

activities (e.g., bas&etball, bi&e riding, trampoline". 1he children with low e3traversion

temperament styles did not show problem behavior in either situation. 1he children with

high e3traversion temperament styles did not engage in problem behavior in the high

intensity situations but displayed outbursts of problem behavior within ? minutes of being

as&ed to engage in the low intensity activity. 1he differing results of these studies

?


18/85

indicate that lin&s between e3traversion=surgency and e3ternali$ing behavior need to be

e3amined further.

%ntervention for Problem ehavior& Temperament Perspectives

In the developmental literature, there are three critical concepts central to

understanding children5s temperament and applying this &nowledge to assessment and

intervention. #pecifically, the concepts of set point, goodness-of-fit, and niche pic&ing

create a framewor& for analy$ing the interplay between individual temperament

characteristics and the conte3ts in which children must function.

#et point is a personal baseline that remains constant over timeF (Dujita

iener, 2?". pplied to temperament, set point can be understood as a child5s average

way of behaving based on his=her genetic predisposition to various temperament

characteristics. 1o illustrate, in the study of body weight regulation, various interventions

(e.g., restricting food inta&e or consuming particular nutrients" can alter weight within a

small range around the set point but, over time, body weight tends to be stable around

that set point ('arris, !**". #imilarly, in the study of children5s temperament, various

medical, behavioral, and educational interventions might plausibly alter undesirable

temperament characteristics within a small range around the set point but would not

drastically shift temperament characteristics far from the set point. Dor e3ample, consider

a child whom we will call ;oger, who is impulsive, full of energy and prefers activities

such as shooting a bas&etball or playing tag. ;oger has a high set point for

surgency=e3traversion. %ehavioral procedures and=or medication may somewhat decrease

;oger5s e3traversion, but, compared to other children, ;oger5s level of e3traversion will

still be higher than average.

/


19/85

s previously noted, goodness-of-fit refers to how well a child5s temperament

matches the situation in which he=she has to cope (4hess 1homas, !**!". If activities

or situations are a poor fit with a child5s temperament, problem behavior will li&ely

occur. 4onsider our e3ample of a highly e3traverted child, ;oger. 1o teach ;oger to

identify his colors, his parents and teachers as& him to sit at a table, prompt him to name

the correct colors (red, blue, green", and reward him when he chooses correctly. In

response to this sedentary activity, ;oger throws tantrums and hits his parents and

teachers when they sit him down to learn colors. 1he situation does not allow ;oger to be

e3traverted, and therefore, constitutes a poor fit with ;oger5s temperament. s e3pected,

;oger engages in high levels of problem behavior in the situation that is incongruent with

his temperament.

:iche pic&ing is defined as choosing a situation that best fits an individual5s

temperament (#uper 'ar&ness, !**G".1ypical children often niche pic& to choose

situations that are the best fit for their temperament. Dor e3ample, a child could choose to

join a football team or a chess team, based on his=her temperament traits. 4hildren with

autism, however, are often unable to select their own niche due to demands placed on

them by parents and teachers and because of difficulties communicating their wants and

needs. 1hus, niche-pic&ing interventions for children with autism re0uire parents and

staff to assess a child5s temperament and then choose niches that are li&ely to be a good

fit for the child and structure academic and home demands to be congruent with this

niche. 1o illustrate niche pic&ing, consider ;oger again. In order to niche pic&, a parent

or teacher must alter the environment in order to allow ;oger to be more e3traverted

within a given situation. 1his strategy wor&s with ;oger5s temperament, rather than

@


20/85

against it. 1hus, a parent or teacher could, for e3ample, tape red, blue, and green s0uares

to a wall and as& ;oger to run to red,F run to blue,F etc. 1his situation would represent

a good niche because ;oger could engage in high intensity activity while learning colors

at the same time.

Specific Components of Temperament'ased %ntervention

>ne clinical trial has tested the effectiveness of temperament-based intervention.

In this clinical trial, groups of mothers of B-? year old neurotypical children with difficult

temperament characteristics (e.g. negative mood, high-intensity emotionality, or high

activity level" received a temperament parent training program. 1his included psycho-

education about temperament, identification of their child5s temperament profile,

e3planations of how to ma&e their demands more similar to their child5s temperament

characteristics, and behavior management techni0ues (#heeber Hohnson, !**G". 1he

mothers who received this intervention reported fewer child behavior problems, increased

satisfaction in their relationship with their child, and greater perceived parental

competence as compared to the mothers in the waitlist control group. 1here are also

0ualitative reports of the outcomes of temperament-based interventions. Dor instance, in

the :ew 7or& 6ongitudinal #tudy (:76#", 4hess and 1homas met with parents for an

average of 2-B sessions to provide them with goodness-of-fit suggestions based on their

children5s temperament (!*+/". ppro3imately ? of cases were considered successful

based on clinical judgment of parental change and improvement in the child5s behavior.

In another temperament intervention program, temperament was assessed, strategies for

strengthening the parent-child relationship were introduced, and specific parenting advice

using goodness-of-fit principles were delivered (#mith, !**G". fter completing this

+


21/85

program, @* of parents indicated that they were helped much or very much by the

programF.

)ost temperament-based intervention programs contain three elementsC (!"

general educational discussions with parents to increase their awareness and

understanding of the concept of temperament, (2" identification of the particular child5s

temperament profile to provide a more organi$ed and objective picture, (B" interventions

that influence the temperament J environment interaction by improving goodness-of-fit

(4arey, !**G".

(eneral educational discussionsabout temperament should include defining

individual differences, e3plaining various types of temperament differences, giving

parents an understanding of why it is better to wor& withrather than againsta child5s

temperament, and advising parents that different strategies wor& with different

temperament styles (4hess 1homas, !*+/". 1hese discussion points can alleviate

parents5 guilt about their children5s behavior and can shift parents5 cognitive or

motivational e3planations to a temperament-based e3planation (eogh, !**G".

1o identify and teach parents about a child5s temperament profile, one can

employ various assessment methods. >ne commonly used assessment method is parent

0uestionnaires (;othbart et al., 2!". Parents=caregivers spend a large amount of time

with their children, and therefore, have a broad &nowledge of their child5s temperament

that can be reported through 0uestionnaires. In addition, it can be helpful to observe the

child, either informally or with a temperament rating system, in order to assess

temperament ()ajadand$ic van den %oom, 2/". ;egardless of which method is

used, it is important to assess the child5s behavior in a wide range of situations in order to

*


22/85

identify their temperament style (4hess 1homas, !*+/". 1he child5s temperament

profile should then be e3plained thoroughly to the parents.

9hen improving the goodness'of'fitbetween the child5s temperament and his=her

environment, temperament intervention researchers suggest that clinicians begin by

choosing only one or two situations to improve (4hess 1homas, !*+/". 1hen,

clinicians should counsel parents on the specific details of the environment that will be

modified, as well as what outcome to e3pect as a result of the intervention. Dor e3ample,

for a child with a high activity level, temperament interventionists suggest see&ing after

school programs that emphasi$e active play, not e3pecting the child to sit during an entire

dinner or a lengthy car ride, and giving the child errands to run during school to get a

reprieve from sitting at a des& (4hess 1homas, !**!".

The Present Study

1he present study investigated the effectiveness of temperament-based

intervention for highly e3traverted=surgent children with # in reducing problem

behavior and improving family 0uality of life.

!


23/85

II. )ethod

Participants

Participants in this study were si3 children diagnosed with an autism spectrum

disorder. 1he inclusion criteria for this study were as followsC (!" individuals must have

been diagnosed with # (2" individuals must have been between the ages of B-@ (B"

individuals must have had a history of problem behavior (score K!/.? on the Irritability

#cale of the berrant %ehavior 4hec&list see below for discussion" and (G" individuals

must have had highly e3troverted temperament styles (score K?.B! on the

e3traversion=surgency factor of the 4hild %ehavior Luestionnaire see below for

discussion". iagnosis of # was verified through home or school records and was

based on either a prior evaluation by a psychologist or psychiatrist through use of the

utism iagnostic >bservation #chedule (># 6ord, ;utter, i6avore, ;isi, 2!"

or based on #)-I< criteria (merican Psychiatric ssociation, !**G".

'istory of problem behavior was verified through parent report on the Irritability

#cale of the berrant %ehavior 4hec&list (%4 man #ingh, !*+/ ppendi3 ".

1he %4 is fre0uently used to assess problem behavior in people with developmental

disabilities. 1he irritability subscale is a !? item measure that assesses severe problem

behaviors for e3ample, items include injures self on purposeF and aggressive to other

children or adultsF. Each item is scored on a -B 6i&ert scale ranging from not at all a

problemF to the problem is severe in degree.F 1he scale has good internal consistency

demonstrated by a coefficient alpha of .*2 (man, #ingh, #tewart, Dield, !*+?".

Interrater reliability of the subscale, using #pearman correlation coefficients, ranges from

.B* to .@, dependent on rater pairings. 1est-retest reliability of the irritability subscale

!!


24/85

using a #pearman correlation has been found to be .*+ (man, #ingh, #tewart, Dield,

!*+?". 1he researchers previously gathered scores on the %4 from a sample of !!B

children with #, ages 2 to * years of age (dame& et al., 2!". Drom this previous

sample, a @?thpercentile cutoff score for problem behavior of !/.? was determined.

E3traversion was verified through parent reports on the ctivity 6evel,

Impulsivity, 'igh Intensity Pleasure, and #hyness subscales that contribute to the

e3traversion=surgency factor of the 4hild %ehavior Luestionnaire (4%L-short form

Putnam ;othbart, 2/, see ppendi3 %". 1his 6i&ert-type scale for assessing

temperament in children ages B-@ years has 2? items. 1he subscales of the

e3traversion=surgency factor of the 4%L-short have good internal consistency, with

coefficient alphas over .@G for the four subscales (;othbart et al., 2!". Interrater

reliability coefficients of mothers and fathers range from .?! to .@* for the G subscales.

1he researchers previously gathered scores on the 4%L from a sample of !!B children

with # (dame& et al., 2!". Drom this previous sample, a @?thpercentile cutoff score

for e3traversion of ?.B! was determined.

Participant characteristics are reported in 1able !. 1hese characteristics includeC

name of child (all names have been changed for confidentiality", age, gender, diagnosis,

full scale IL, score on the berrant %ehavior 4hec&list, and score on the e3traversion

scales of the 4hild %ehavior Luestionnaire. Participants ranged in age from B.!to @.2

years.Dive children were male, and one child was female. ll children had an autism

spectrum diagnosis.Dull scale IL scores ranged from G to !!2.>ne family was unable

to provide IL score information because their child was unable to engage in any test

items during two separate evaluations attempts. ll si3 children5s scores on the berrant

!2


25/85

%ehavior 4hec&list were above the @?thpercentile cutoff of !/.? and ranged from !+ to

B?. ll si3 children5s scores on the E3traversion subscales of the 4hild %ehavior

Luestionnaire were above the @?thpercentile cutoff of ?.B! and ranged from ?.BB to /.2+.

Procedure

Completion of Pre'%ntervention )easures

Damilies were recruited for this study by letters distributed to local agencies and

schools for children with #, and through advertisements on autism listserves. uring

the initial meeting, parents signed consent forms and completed a battery of

0uestionnaires. Parents completed the Irritability #cale of the berrant %ehavior

4hec&list (%4" as well as the ctivity 6evel, Impulsivity, 'igh Intensity Pleasure, and

#hyness subcales of the 4hild %ehavior Luestionnaire (4%L". In addition, families

completed a battery of pre-intervention ancillary 0uality of life measures. 1he Parenting

#tress Inventory J #hort Dorm (P#I=#D bidin, !**@ ppendi3 4" measures the amount

of stress parents encounter on a daily basis. 1he scale contains B/ parent-report items that

produce three factorsC Parental istress (P", Parent-4hild ysfunctional Interaction

(P4I", and ifficult 4hild (4". 1est-retest reliability has been computed between

scores at ! year apart (r M .@?,p N .! 'as&ett, hern, 9ard, llaire, 2/". 1he

'ome #ituations Luestionnaire ('#L %ar&ley, !*+! ppendi3 " measures how much

the child5s problem behavior disrupts home situations such as mealtime or bathtime. It

contains !/ parent-report items that are scored on a -* li&ert scale ranging from absentF

to severeF. 1he scale has good internal consistency demonstrated by a coefficient alpha

of .*B, and good test-retest reliability (#pearman 4orrelation *! uPaul %ar&ley,

!**2". 1he Parental 6ocus of 4ontrol #cale (P64# 4ampis, 6yman, Prentice-unn,

!B


26/85

!*+/ ppendi3 E" measures how much parents feel in control of their children. It

contains items that are scored on a !-? 6i&ert scale ranging from strongly disagreeF to

strongly agreeF. In previous wor&, items were shown to be internally consistent, with an

alpha coefficient of .+! and have good test-retest reliability, with a reliability coefficient

of .+B (;oberts, Hoe, ;owe-'allbert, !**2".

%dentification of %ntervention Contexts

Each family participated in a follow up assessment to identify intervention

conte3ts. 1he researcher (a Ph student in clinical psychology" gave e3amples of what

constitutes a low intensity situation, such as those found in ppendi3 D. 1hrough a

discussion with parents, low intensity activities that were typically problematic for their

child and typically yielded high instances of problem behavior were identified. Parents

were then as&ed to identify one top priority conte3t as the e3perimental conte3t, while the

other conte3ts were considered for later clinical e3tensions. 1he intervention in the

e3perimental conte3t employed temperament intervention strategies and included

e3tensive data collection.

%dentification of %ntervention Contexts& )ichael*)ichael was a ?-year-old boy

diagnosed with utistic isorder (" who attended a special education class and lived

at home with his mother, father, twin brother, and two older brothers. 'e communicated

through the use of single words and gestures as well as limited use of a Picture E3change

4ommunication #ystem (PE4# %ondy Drost, !**G". 9hen presented with a list of low

intensity activities, his parents identified playing a game as a conte3t they were

concerned about therefore, it was selected as the e3perimental conte3t. #pecifically,

)ichael5s parents reported that )ichael engaged in problem behavior (e.g., tantrums,

!G


27/85

aggression, self-injury, and noncompliance" when playing board games. 1ypically,

)ichael would throw the game pieces, bang his head, or run away when as&ed to play.

%dentification of %ntervention Contexts& +obbie*;obbie was a @-year-old boy

diagnosed with who attended a special education class and lived at home with his

mother, father and younger brother. 'e was nonverbal and communicated through the use

of gestures as well as limited use of PE4#. 9hen discussing low intensity conte3ts with

the researcher, his mother identified coloring as a primary conte3t she was concerned

about. #pecifically, ;obbie5s mother reported that ;obbie engaged in problem behavior

(e.g., tantrums, aggression, self-injury, and noncompliance" when instructed to color.

1ypically, ;obbie would refuse to color at all, and when prompted to color, he would

tantrum, bang his head, or hit his mother or brother.

%dentification of %ntervention Contexts& $anny* anny was a ?-year-old boy

diagnosed with Pervasive evelopmental isorder-:ot >therwise #pecified (P-:>#"

who attended a special education class and lived at home with his mother, father, younger

brother, and younger sister. 'e communicated through the use of complete, short

sentences. 'is parents identified dinnertime as the most problematic low intensity conte3t

for their family. #pecifically, anny5s parents reported that he engaged in problem

behavior (e.g., tantrums, noncompliance" when instructed to eat his dinner at the table.

anny5s parents would try to redirect him to the table, but most often anny would

continue to tantrum or run away.

%dentification of %ntervention Contexts& Eli,ah*Elijah was a B-year-old boy

diagnosed with P-:># who attended a typical preschool and received after school

early intervention services through a local agency for children with developmental

!?


28/85

disabilities. 'e lived at home with his mother and father and communicated through the

use of simple words and gestures. 9hen presented with a list of low intensity activities,

his parents identified coloring as an important conte3t that was difficult for Elijah.

#pecifically, Elijah5s parents reported that Elijah engaged in problem behavior (e.g.,

tantrums and noncompliance" when as&ed to color. 1ypically, Elijah would refuse to

begin coloring, run away from the coloring materials, or throw a tantrum.

%dentification of %ntervention Contexts& -elicia*Delicia was a /-year-old girl

diagnosed with who attended a special education class and lived at home with her

mother, father, and older sister. #he communicated through the use of complete, short

sentences. uring discussions with the researcher, Delicia5s mother identified reading a

boo& as a problematic, low intensity conte3t that was important to their family.

#pecifically, Delicia5s mother reported that she engaged in problem behavior (e.g.,

tantrums and noncompliance" when instructed to listen to a boo& being read to her.

1ypically, Delicia would refuse to even begin listening to a story being read, and if her

mother further prompted her to sit and listen, Delicia would throw a long, intense tantrum.

%dentification of %ntervention Contexts& Connor*4onnor was a G-year-old boy

diagnosed with P-:># who received early intervention services through a local

agency for children with developmental disabilities and lived at home with his mother,

father, and younger sister. 'e was fully verbal and communicated through the use of

complete, comple3 language. 9hen presented with a list of low intensity activities, his

mother identified playing a board game with others as an important, problematic conte3t.

#pecifically, 4onnor5s mother reported that 4onnor engaged in problem behavior (e.g.,

tantrums, aggression, and noncompliance" when playing games. 1ypically, 4onnor would

!/


29/85

successfully begin a game, but in the middle of the game would refuse to play, throw a

tantrum, or hit his mother or sister.

aseline for Experimental Contexts

multiple baseline design was used across participants ('ersen %arlow, !*@/".

1he si3 participants were randomly assigned to be in group ! or group 2, and were

assigned to have the researcher complete B, G, or ? baseline observations. In single

subject research, replication is often achieved through the use of a reversal design by

alternating baseline and treatment periods however, due to ethical concerns about

withdrawing the treatment as well as worries about carryover effects, a multiple baseline

design was employed ()organ )organ, 2!". 1he staggered introduction of

intervention strategies aims to improve internal validity by showing that behavior change

occurs at various time-points (a$din, 2B". Past researchers have established that a

minimum of two baseline sessions per participant is necessary, therefore we chose B-?

baseline sessions ('ersen %arlow, !*@/". )ultiple baseline designs are commonly

implemented across three participants. 1wo groups of three participants were used to

more 0uic&ly deliver intervention to families and to replicate the findings across groups.

)ultiple baseline observations were conducted in the top priority, e3perimental

conte3t identified by the researcher and parents. Parents were as&ed to complete two !-@

6i&ert scale items assessing the fre0uency and intensity of their child5s problem behavior

during this conte3t in the past month. 1he scale for the first item ranges from

infre0uentF to e3tremely fre0uentF and the second item ranges from not intenseF to

e3tremely intenseF. 1hese items can be found in ppendi3 8. %aseline observations in

the low intensity conte3t assessed latency to problem behavior and percent of the tas&

!@


30/85

completed. Dor latency to problem behavior, observers recorded the amount of time from

the beginning of the tas& to the onset of problem behavior (4arr 4arlson, !**B".

1ypical problem behaviors are tantrums, noncompliance, or disruptive behavior. If these

occurred during baseline, the low intensity activity was stopped and parents were as&ed

to carry out the natural procedures that they typically use when their child engages in

problem behavior.

If no problem behavior occurred, the researcher recorded the amount of time from

the beginning of the tas& to the completion of the tas&. Dor tas& completion, the

researcher developed a tas& analysis for each e3perimental tas& that bro&e down the tas&

into its component steps. ll tas& analyses are outlined in 1able 2. Dor some tas&s, such

as coloring, there was no clear end point. 1herefore, tas& components spanning a certain

amount of time (e.g., color for ! minute" were created. 1hese time increments were

chosen with parent input so that the total duration of time was appro3imately twice as

long as the child had ever been able to spend doing that particular activity. uring

baseline, the researcher recorded the number of steps the participant successfully

completed without problem behavior. t the conclusion of the tas&, the researcher

divided the number of steps successfully completed by the total number of steps in the

tas& to derive the percent of tas& completed.1he investigator and a parent collected

reliability data on the relevant study variables during one baseline session. binary

reliability inde3 was used to assess agreement on percentage of tas& steps completed and

latency to session termination. 1hus, for each session, reliability was scored as either

perfect agreement or no agreement. greement was defined as both observers recording

!+


31/85

the same number of tas& steps completed and latency measures that were within ?

seconds of one another.

Psychoeducation about Temperament

irectly following baseline, the researcher met with the parents to teach them

about temperament, and temperament-based intervention. temperament curriculum

(ppendi3 '" was used as a guide during this session. 1his curriculum was supplemented

by a variety of e3amples presented to parents, as well as detailed discussion to ensure

parents5 understanding.

Overview of %ntervention for Experimental Contexts

#trategies to improve the fit between the e3perimental conte3t and the child5s

e3troverted temperament were developed with the help of the parents during a problem

solving session (5Aurilla :e$u, 2! #tiebel, !***". uring this session, parents and

the researcher wor&ed their way, systematically, through a problem-solving template

(ppendi3 I" applied to the conte3t of interest. Parents were &nowledgeable about which

strategies were most feasible within their family, which can be an important asset in

developing a treatment plan (6ucyshyn, lbin, :i3on, !**@


32/85

intervention components for each participant. 1he investigator and parent recorded

chec&mar&s whenever a specific component of the intervention was implemented by the

parent. Hust as in baseline observations, both the parent=s and researcher assessed latency

to problem behavior and percent of the tas& completed each time the intervention was

used.lso similar to baseline, the investigator and a parent collected reliability data on

the relevant study variables during 2? (B of !2 intervention sessions". binary

reliability inde3 was used to assess agreement on intervention fidelity, percentage of tas&

steps completed, and latency to session termination. fter the intervention in the

e3perimental conte3t was complete, parents rated the overall fre0uency and intensity of

their child5s problem behavior during the past month using the 6i&ert scale item

previously used during baseline. summary of the intervention fidelity chec&lists can be

found in 1able B and more detailed descriptions of the intervention strategies can be

found below.

%ntervention Strategies for a (ame& )ichael* )ichael5s parents chose the game

#nails, Pace, ;aceF as the e3perimental conte3t. 1his game is typically played by rolling

a small, colored, dice and moving a snail with the corresponding color one space. 1his is

a poor match for a child with a surgent temperament because the game is sedentary,

typically played sitting at a table, does not involve physical activity, and re0uires fine

motor s&ills rather than gross motor movements. >ne intervention strategy was to replace

the small dice provided with the game with a / inch colored dice that could be tossed in

the air. 1his encouraged standing, throwing and gross motor play. In addition, )ichael5s

parents helped him hold a snail after he threw the dice, and e3claim, (color" snail, -

OOOOOO snail, OOOOOO snail, bounceF while jumping with )ichael. 1his made the game

2


33/85

more interactive, and allowed )ichael to move around. )ichael5s parents provided him

with surgent praise (e.g., hugs, high fives, pats on the bac&, tic&les". 1his gave )ichael

positive, physically active interactions with his parents. 6astly, intervention included

singing a song about snails a few times throughout the game. )ichael5s parents sang this

song to him while swinging him bac& and forth. 1his engaged )ichael in energetic

activity at various times throughout the game.

%ntervention Strategies for Coloring& +obbie* ;obbie5s mother chose coloring as

the e3perimental conte3t. 4oloring, as it is typically presented, is a poor match for a child

with a surgent temperament because it is done while sitting and re0uires little gross

movements or high-intensity actions. 1he intervention was established so that ;obbie

alternated between two coloring stations. 'is mother chased him between these stations.

1his allowed ;obbie to ta&e short brea&s from the low-intensity activity of coloring

during the time he was being chased. >ne coloring station included a rubber disc filled

with air to sit on and a vibrating pen to color with. ;obbie bounced on the disc in his seat

while coloring with the vibrating pen. 1he other station included large white paper taped

to the wall and large, triangular crayons. 9ith these materials ;obbie could run the

length of the wall while coloring with crayons that were easier to manipulate than typical

crayons.

%ntervention Strategies for $innertime& $anny*anny5s parents chose eating

dinner at the table as the e3perimental conte3t. Eating dinner is a poor match for a child

with a surgent temperament because it is typically done while sitting and does not involve

any high-intensity activity. 1he intervention included the use of a surgent visual schedule.

It should be noted that during baseline, anny5s parents used a visual schedule, therefore,

2!


34/85

this behavioral strategy was enhanced within a temperament framewor& by ma&ing the

visual schedule a better fit for a surgent child. 1he visual schedule consisted of pictures

of one-fourth, one-half, three-fourths, and all of the food on a plate eaten. :e3t to these

pictures were musical greeting cards. 9hen anny ate a fourth of his food, he was

allowed to stand up from his seat, wal& to the visual schedule, open a greeting card, and

dance to the !-2 second clip of music. 'e then was re0uired to return to his seat and eat

the ne3t fourth of food before getting up again. 1hese scheduled, controlled bursts of

surgent activity helped anny to be able to sit while eating his dinner. In addition,

anny5s parents provided him with surgent praise (e.g., hugs, high fives, pats on the

bac&, tic&les" when he returned to his seat. 1his gave anny positive, physically active

interactions with his parents. 6astly, the intervention strategies included replacing the

edible reinforcer that anny5s parents previously gave him with a more active reinforcer.

9e chose activities such as playing with a toy steering wheel that made racing noises,

and jumping on a trampoline as positive reinforcement for anny eating his entire dinner

without any problem behavior.

%ntervention Strategies for Coloring& Eli,ah* Elijah5s parents chose coloring as the

e3perimental conte3t. 4oloring, as it is typically presented, is a poor match for a child

with a surgent temperament because it is done while sitting and re0uires little gross

movements or high-intensity actions. >ne of the main elements of the intervention was

the use of a color me a songF board. 1his toy has buttons that ma&e sounds when

pressed, and when colored on, it plays music to the pace of the coloring. Elijah was as&ed

to push a colored button and color with the same colored crayon as the button. #mall

crayons were replaced with large, triangular crayons. Elijah was encouraged to color as

22


35/85

0uic&ly as possible. 1he use of this board allowed coloring to be more active than when

done with typical paper and crayons. uring baseline, Elijah5s parents often used a

monotone tone of voice when giving instructions or praise. uring intervention, they

were taught to use an enthusiastic tone of voice to be more engaging and intense for

Elijah. 6astly, Elijah5s parents provided him with surgent praise (e.g., hugs, high fives,

pats on the bac&, tic&les" when he was coloring. 1his gave Elijah positive, physically

active interactions with his parents.

%ntervention Strategies for +eading& -elicia* Delicia5s mother chose listening to a

boo& being read as the e3perimental conte3t. 6istening to a boo& is a poor match for a

child with a surgent temperament because it is typically a sedentary activity done while

sitting and re0uires very little action or input from the child. uring intervention Delicia

was instructed to choose from three interactive boo&s. 1hese boo&s were about cars,

trains, and princesses, and included sound buttons to push, flaps to move, or pu$$les to

complete. %y using boo&s that re0uired Delicia to do something, rather than to only sit

and listen, it became a more surgent activity. Delicia5s mother learned to as& Delicia

0uestions about characters in the boo&s, to as& her to ma&e noises or movements about

characters in the boo&, and to as& her to push buttons, move flaps, or do pu$$les. In

addition, Delicia5s mother prompted Delicia to get a toy that went along with the boo&

(e.g., car, train, princess wand". #he then as&ed Delicia to play with the toy when

appropriate in the conte3t of the story (e.g., )ove 1homas the train 0uic&ly across the

table 'e has to get to Percy the trainF". 6astly, Delicia5s mother would say, !-2-B, turn

the pageF and prompt Delicia to turn the page. Encouraging Delicia to turn the pages,

2B


36/85

play with a toy, and interact with the boo& made listening to a boo& a higher energy

activity.

%ntervention Strategies for a (ame& Connor* 4onnor5s mother chose board games

as the e3perimental conte3t. 9e decided to use #piderman 4hutes and 6addersF and

ora the E3plorer 4andy 6andF as the board games for intervention. 1hese games are a

poor match for a child with a surgent temperament because they are sedentary, typically

played sitting at a table, do not involve physical activity, and re0uire fine motor s&ills.

>ne intervention strategy was to encourage 4onnor to wear a &ing hat and pretend to be

the &ing of the gameF and e3plain the rules to everyone playing. 1his made the

beginning of the game more interactive. 9hen playing 4hutes and 6adders, 4onnor5s

mother encouraged him to tell and act out stories about the super hero characters in the

game. 1he spinner in the game was replaced with a / inch foam dice that could be thrown

in the air. 9hen 4onnor landed on a ladder, he climbed on a stepstool that we pretended

was a ladder. 9hen 4onnor landed on a chute, he rolled on the ground to pretend that he

was going down the chute. 9hen playing 4andy 6and, 4onnor5s mother encouraged him

to tell and act out stories about ora the E3plorer characters. 1he cards in the game were

put into a large fishbowl, rather than a small pile. 6aminated color s0uares were placed

around the room and when 4onnor drew a colored card he would move his character and

go jump on the corresponding colored s0uare himself. 9hen playing either game, 4onnor

was prompted to say, B-2-!, blast offF (one of his favorite sayings", and jump in the air

when ta&ing a turn. ll of these intervention strategies helped to include rowdy, active,

and imaginative play in the conte3t of the board games.

2G


37/85

Clinical Extensions

fter parents demonstrated intervention mastery for the e3perimental conte3t

(based on e3perimenter observation and intervention fidelity chec&list", the researcher

met with the parents to discuss additional low intensity conte3ts where intervention was

needed. 1he researcher developed temperament-based intervention strategies for these

clinical e3tensions. 4linical e3tensions are important because they help create

multicomponent interventions that promote the use of intervention strategies across a

variety of conte3ts (4arr 4arlson, !**B". Interventions that address multiple conte3ts

are more li&ely to produce general decreases in problem behavior as well as

improvements in overall family 0uality of life. Dre0uency and intensity of child problem

behavior in each conte3t was assessed before and after the implementation of intervention

in the clinical conte3t. Intervention fidelity chec&lists, based on the intervention

components for the additional conte3ts, were developed to evaluate intervention integrity

for these conte3ts. 1he researcher observed ! session of baseline and 2 sessions of

intervention for each clinical e3tension conte3t and completed the intervention fidelity

chec&list as well as assessed latency to problem behavior and percent of tas& completed.

Clinical Extension& )ichael* )ichael5s parents had also identified going to a

restaurant as a low-intensity, problematic conte3t for )ichael. 9e first employed

intervention strategies in a fast food restaurant. )ichael5s parents were instructed to use a

surgent visual schedule that consisted of a magnetic board with pictures of restaurant

activities (e.g. wait in line, ta&e our food to a table, sit in a chair, eat", and empty spots for

)ichael to place magnetic letters on the board when he completed a restaurant activity.

9hile in line, )ichael5s parents played #imon saysF with him and prompted )ichael to

2?


38/85

do active movements such as, jump up and downF or touch your toesF. 9hen in his

seat, )ichael sat on an inflatable, rubber dis& that he could bounce on. If )ichael ate his

dinner and did not e3hibit any problem behavior, his parents played a surgent activity

with him when they returned home (e.g., tag, spin in circles". Eventually, the intervention

was e3tended further to include going to a sit-down family restaurant in addition to fast

food restaurants.

Clinical Extension& +obbie* ;obbie5s mother identified eating a snac& at the table

as a second problematic conte3t for ;obbie. >ne intervention strategy was to use food

that was shaped li&e animals, or other characters. ;obbie5s mother was taught to prompt

;obbie to play games related to the food, such as ma&ing a cat noise after eating a

crac&er shaped li&e a cat. counting coo&ie jar toy was also used to ma&e eating the

snac& more energetic. fter ;obbie ate a bite of food, he placed a toy coo&ie into the

coo&ie jar. In addition, ;obbie sat on a vibrating cushion on his chair to provide him with

physical input. 6astly, ;obbie5s mother was taught to use surgent praise (e.g., hugs, high

fives, pats on the bac&, tic&les" when he was sitting in his chair and eating his snac&

without any problem behavior in order to give him a positive, physically active

interaction.

Clinical Extension& $anny* anny5s parents identified playing 4hutes and

6addersF as another difficult, low intensity conte3t for anny. 1his intervention included

the same concept of climbing a stepstool when landing on a ladder on the board and

rolling on the ground when landing on a chute on the board as was used with 4onnor.

Durther, the spinner in the game was replaced with a / inch foam dice that could be

thrown in the air. In addition, a large turn board was used that was adhered to the wall.

2/


39/85

anny would get up and place a picture of himself, his mother, his father, or his brother

on the turn board to show that it was that person5s turn. 1hese brea&s which included

getting up from the table and changing the picture on the turn board allowed anny to be

more lively during the game. 6astly, anny5s parents were encouraged to provide him

with surgent praise (e.g., hugs, high fives, pats on the bac&, tic&les".

Clinical Extension& Eli,ah* Elijah5s parents chose tracing letters as another

problematic conte3t. >ne intervention strategy was to allow Elijah to stand, rather than

re0uiring him to sit while tracing letters. In addition, wor&sheets and small crayons were

replaced with large, colorful posterboards and large, triangular crayons. 1he original

wor&sheets included letters to trace and animals or objects that corresponded with the

letter being traced. 1he poster boards also included these elements, and Elijah was

prompted to imitate the animal or object by using noises or movement. 6astly, Elijah5s

parents provided him with surgent praise (e.g., hugs, high fives, pats on the bac&, tic&les"

when he was tracing letters. 1his gave Elijah positive, physically active interactions with

his parents.

Clinical Extension& -elicia* Delicia5s mother chose going to the grocery store as

another difficult, low intensity conte3t for Delicia. >ne intervention strategy used during

the car ride was for Delicia5s mother to play games with her. 1his included ma&ing voices

of isney characters and singing some of Delicia5s favorite songs. >nce at the store

Delicia5s mother gave her a boo& of pictures of grocery store items and had her find items

in a particular aisle. 1his gave Delicia an active tas& to be engaged with rather than

wal&ing through the store with nothing to do. lso, when wal&ing down the aisles,

Delicia was encouraged to s&ip or hop, rather than wal&. 9hen Delicia successfully

2@


40/85

completed a trip to the grocery store, she was provided a surgent reward (e.g., going to a

play space, driving on a bumpy road".

Clinical Extension& Connor* 4onnor5s mother chose completing academic

wor&sheets as a problematic, low intensity conte3t for 4onnor. uring baseline,

wor&sheets were used that re0uired 4onnor to circle the smallest or largest picture.

uring intervention, the wor&sheets targeting smallest were placed on the &itchen table

and the wor&sheets targeting largest were placed on 4onnor5s des& in the living room.

9hen completing the wor&sheets 4onnor was as&ed to act out the pictures on the

wor&sheets (e.g., bu$$ li&e a bee, fly li&e a jet". fter completing a wor&sheet targeting

smallest at the &itchen table, 4onnor was told to race to the middle of the living room and

choose the smallest of three balls. >nce he correctly pic&ed the smallest ball, he ran to the

des& in the living room and completed a wor&sheet targeting largest. 1hen, 4onnor raced

to the middle of the living room and chose the largest of three balls. 1his continued until

four of each wor&sheet was completed. %y encouraging 4onnor to race, move around the

room, manipulate objects (balls", and act out pictures from the wor&sheets, the concepts

of smallest and largest were practiced while allowing 4onnor to be more active.

Completion of Post'%ntervention )easures

fter the interventions for the e3perimental conte3t were employed for !2 wee&s

and the clinical e3tensions were conducted for at least 2 wee&s, a post intervention

assessment battery of 0uestionnaires was administered to the parents. 1hese included the

#urgency items of the 4hild %ehavior Luestionnaire, the Irritability #cale of the berrant

%ehavior 4hec&list, the Parenting #tress Inventory, the 'ome #ituations Luestionnaire,

and the Parental 6ocus of 4ontrol #cale.

2+


41/85

III. ;esults

Experimental Contexts

Percent Tas. Steps Completed* 1he percentage of activity=routine steps completed

for the three participants in group ! are shown in Digure !, and for the three participants

in group 2 in Digure 2. )ichael completed an average of BB.B of the steps that

constituted playing a board game during baseline. 'owever, during intervention, he

completed an average of *?.? of the steps. ;obbie completed a mean of 2@.+ of the

steps associated with coloring during baseline. 'owever, during intervention, he

completed ! of the steps. uring baseline, anny completed a mean of B? of the

steps involved in eating dinner. 'owever, during intervention, he completed ! of the

steps. Elijah completed !* of the steps that constituted coloring during baseline.

'owever, he completed *+of steps during intervention. Delicia completed 2? of the

steps involved in being read a boo& during baseline. 'owever, she completed ! of

the steps during intervention. 4onnor completed GB of steps associated with playing a

board game during baseline. 'owever, he completed ** of the steps during

intervention.

atency to Session Termination* Digure ! and Digure 2 present data on the amount

of time that elapsed before the session was terminated (due to problem behavior or

successful completion of the activity" for the si3 participants.Dor )ichael, the mean

latency to problem behavior during baseline was 2 minutes, ! seconds. uring

intervention sessions when )ichael engaged in problem behavior (B out of !2 sessions",

the mean latency to problem behavior was !! minutes, ?B seconds. uring intervention

sessions with no problem behavior (* out of !2 sessions", successful completion of

2*


42/85

playing the board game occurred at !/ minutes and * seconds. Dor ;obbie, the mean

latency to problem behavior for coloring at baseline was 2G seconds. uring intervention,

there was no problem behavior and latency to successful completion of coloring was B

minutes, 2@ seconds. Dor anny, during baseline, the mean latency to problem behavior

during dinnertime was G minutes, 2 seconds. uring intervention, anny did not engage

in problem behavior and latency to successful completion of eating dinner was !@

minutes, B* seconds. Dor Elijah, during baseline, the mean latency to problem behavior

for coloring was 2 seconds. uring intervention, Elijah did not engage in problem

behavior and latency to successful completion of coloring was @ minutes, GG seconds. Dor

Delicia, during baseline, the mean latency to problem behavior for reading was G minutes,

2 seconds. uring intervention, Delicia did not engage in problem behavior and latency to

successful completion of being read a boo& was * minutes, ?G seconds. Dor 4onnor, the

mean latency to problem behavior at baseline was 2 minutes, ! seconds. uring

intervention sessions when 4onnor engaged in problem behavior (! out of !2 sessions",

the latency to problem behavior was !G minutes, * seconds. uring intervention sessions

with no problem behavior (!! out of !2 sessions", successful completion of playing the

board game occurred at a mean of 2 minutes and G+ seconds.

Clinical Contexts

Percent Tas. Steps Completed* 1he percent of activity=routine steps completed

during clinical conte3ts for all si3 participants (group ! and 2" is shown in Digure B.

uring the baseline session, participants completed between 2-B of tas& steps.

'owever, during intervention sessions, participants completed ! of tas& steps.

B


43/85

atency to Session Termination* 6atency to session termination during clinical

conte3ts for all si3 participants (group ! and 2" is shown in Digure G. uring the baseline

session, latency was short, and all si3 participants engaged in problem behavior.

'owever, during intervention sessions, participants did not engage in any problem

behavior, and latency to session termination increased at least G across all

participants.

%ntervention -idelity* uring baseline, a mean of of the intervention

components were implemented by each respective parent for all conte3ts. uring

intervention, for the e3perimental conte3ts, a mean of ! of the intervention

components were implemented by each respective parent. uring intervention for the

clinical e3tension conte3ts, a mean of ! of the intervention components were

implemented by each respective parent.

%nterrater +eliability

1he investigator and a parent collected reliability data on the relevant study

variables during one baseline session and three intervention sessions for the e3perimental

conte3ts and during one baseline session and one intervention session for the clinical

conte3ts. greement on intervention fidelity, percentage of tas& steps completed, and

latency to session termination was noted for ! of baseline and intervention sessions

across all participants.

#ncillary )easures

In addition to outcome measures related to latency to problem

behavior=successful completion of routines, data were collected to measure the

B!


44/85

differences between pre and post measures of conte3t specific problem behavior, global

problem behavior, temperament, parental stress, the child5s disruption of home situations,

and perceived parental control. Paired sample t-tests were performed on all ancillary

measures to compare scores during baseline with those following intervention. In

addition, effect si$es using 4ohen5s d were calculated incorporating the correlation

between measures for each set of variables (unlap, 4ortina,


45/85

I


46/85

Improvements in problem behavior and tas& completion were observed across

participants in the current sample. 1hese participants were diverse in gender, age,

cognitive ability, and communication s&ills. 1o summari$e, participants included ? boys

and ! girl, who were B to @ years of age, with IL5s ranging from G to !!2, and verbal

abilities ranging from non-verbal to conversationally verbal. 1emperament-based

strategies that target the fit between temperament characteristics and the environment can

be fle3ible in regard to other individual characteristics of the child. #o, strategies for an

older child with a higher IL and more advanced verbal abilities can be more comple3

than those for a younger child with a lower IL and little verbal s&ills. #ince temperament-

based strategies can be applied uni0uely for particular children these strategies can be

successful in decreasing problem behavior regardless of gender, age, cognitive ability,

and verbal s&ills.

1emperament, as measured by the 4%L, did not change following intervention.

1his is consistent with previous findings that temperament is moderately stable over time

(;oberts el


47/85


48/85

problem behavior. #ome conte3ts cannot be avoided, and for those conte3ts, interventions

can be designed that increase the goodness-of-fit between the child5s temperament and

the situations in which he=she has to cope. Drom the present study, general themes and

more specific strategies were identified for increasing goodness-of-fit for children high in

surgency=e3traversion. ctivities for children high in this temperament dimension can be

engineered to be more energetic and active, include gross motor play and decrease fine

motor play, and increase physical interaction (e.g. hugs, high-fives, pats on the bac&"

between children and their parents.

1emperament-based intervention strategies are not intended to replace standard

educational or behavioral interventions. Instead, understanding the role of temperament,

and increasing goodness-of-fit between a child5s temperament and a particular

environment or activity may increase the efficacy of educational and behavioral

interventions during that activity.

Ecological 0alidity

1he interventions implemented in this study were employed in natural settings

(e.g., family homes, restaurants, grocery stores", using natural intervention agents (e.g.

parents and grandparents", to modify natural activities and routines (e.g., reading, playing

board games, dinner". 1his stands in direct contrast to many intervention studies that

target problem behavior in # by utili$ing discrete trial methodology in a controlled or

laboratory setting (Iwata, orsey, #lifer, %auman, ;ichman, !*+2". 1hese studies fail

to address 0uestions about generali$ation to natural environments (4arr et al., 22". %y

utili$ing natural settings, intervention agents, and activities, and through the completion

of clinical e3tensions, the results of the current study demonstrate that temperament-

B/


49/85

based strategies are ecologically valid and can be generali$ed to many real-world

situations.

-uture $irections

1his is the first empirical study of temperament-based intervention for children

with #. 1herefore, additional studies that e3amine temperament-based interventions

are necessary to further investigate the efficacy of this approach. dditional intervention

research should include a greater number of participants and children with a greater

variability in their characteristics (e.g., age, gender, temperament profile". Dor e3ample, a

randomi$ed clinical trial comparing outcomes of children who receive temperament-

based intervention to children on a waitlist would be useful. mong neurotypical

children, lin&s have been found between specific temperament profiles and internali$ing

problems, social difficulties, and poor academic functioning ()uris, )eesters,

%lijlevens, 2@ #pinrad, et al., 2@


50/85

1able !

Participant Characteristics at aseline

:ame ge 8ender iagnosis Dull #cale IL

()M!,#M!?"

%4

()M!,#M!"

4%L

()M G.@,#M .@*,

from an

#

sample"

)ichael ?. )ale utistic

isorder

Qnable to

>btain

2@. ?./G

;obbie @.2 )ale utistic isorder #tanford-

%inetC G

B. /.!B

anny ?.? )ale Pervasiveevelopmental

isorder, :>#

#tanford-%inetC @

2+. /.2+

Elijah B.! )ale Pervasive

evelopmental

isorder, :>#

%ayleyC * !+. ?.BB

Delicia /./ Demale utistic

isorder

#tanford-

%inetC?*

2G. ?.GG

4onnor G./ )ale Pervasiveevelopmental

isorder, :>#

9PP#I-IIIC !!2

B?. ?.@2

1Stanford inet 2 Stanford'inet %ntelligence Scale& -ifth Edition13PPS%'%%% 2 3echsler Preschool and Primary Scale of %ntelligence 4 Third Edition

1ayley 2 ayley Scales of %nfant and Toddler $evelopment 4 Third Edition

B+


51/85

1able 2

Tas. #nalyses of Experimental Contexts

:ame 1as& 1as& 4omponents

)ichael Play the game#nails, Pace,

;aceF

!. 4ome to the table in the living room.

2. Put the snails on the starting spots.B. ;oll the dice.

G. )ove a snail.

?. 9ait while )om ta&es a turn./. ;epeat steps B-?.

@. ;epeat steps B-?.

+. ;epeat steps B-?.

*. ;epeat steps B-?.!. ;epeat steps B-? until a snail wins.

!!. Put the game pieces bac& into the bo3.

;obbie 4oloring

!. 4ome to the materials at the &itchen table.

2. Pic& up a crayon=pen=mar&er.

B. 4olor on the paper for ! seconds.G. 4olor on the paper for another ! seconds.

?. 4olor on the paper for another ! seconds.

/. 4olor on the paper for another ! seconds.

@. 4olor on the paper for another ! seconds.+. 4olor on the paper for another ! seconds.

*. 8ive the materials to )om or sign all doneF.

anny Eating inner

!. Put plate on the dining room table.

2. #it down at the table.

B. Eat R of the food on his plate.G. Eat another R of food on his plate.

?. Eat another R of food on his plate.

/. Eat last R of food on his plate.@. 8et a reward.

+. Put plate in the &itchen.

Elijah 4oloring!. 4ome to the table.2. Pic& up crayon.

B. 4olor on paper for ! minute.

G. 4olor on paper for another minute.?. 4olor on paper for another minute.

/. 4olor on paper for another minute.

@. 4olor on paper for another minute or until hesays, ll doneF.

Delicia 6istening to a

boo& read by her

!. 4ome to the couch=bed=chair.2. 4hoose a boo&.

B. #it with )om while she reads 2 pages.

B*


52/85

)om G. 1urn the page.

?. ;epeat steps B-G./. ;epeat steps B-G.

@. ;epeat steps B-G.

+. ;epeat steps B-G until the boo& is finished.

*. Put the boo& away.

4onnor Play a board

game with )om

!. 4ome to the game on the living room floor.2. Place the game pieces on starting spots.

B. 4hoose a card, spin the spinner, or roll the dice.

G. )ove a game piece.

?. 9ait while )om ta&es a turn./. ;epeat steps B-?.

@. ;epeat steps B-?.

+. ;epeat steps B-?.*. ;epeat steps B-?.

!. ;epeat steps B-?.

!!. ;epeat steps B-?.!2. ;epeat steps B-G until someone wins the game.

!B. Put the game pieces bac& in the bo3.

G


53/85

1able B

%ntervention -idelity Chec.lists for Experimental Contexts

:ame Intervention 4omponents 4ompletedS

)ichael!. Qse the giant dice.

2. #ay, (color" snail, OOOOO snail, OOOOOO snail,bounceF and help )ichael bounce.

B. Qse surgent praise (hugs, high fives, pats on the bac&,

tic&les".

G. #ing the snail song while playing.

;obbie

!. Qse the bouncy seat.

2. Qse triangular crayons.

B. Qse the wiggly pen.

G. Put large white paper on the wall.

?. 4hase ;obbie from one station to another.

anny!. #how anny the surgent visual schedule.

2. fter anny finishes each fourth of his food, prompthim to open a music card on the surgent visual schedule.

B. Qse surgent praise (hugs, high fives, pats on the bac&,tic&les".

G. 9hen he finishes dinner, present anny with theracing wheel, trampoline, or other active reinforcer.

Elijah

!. Qse an enthusiastic tone when giving Elijah

instructions.

2. Qse the 4olor me a #ongF board.

B. Instruct Elijah to push the color buttons in between

coloring.

G. Qse large, triangular crayons.

B. Qse surgent praise (hugs, high fives, pats on the bac&,

tic&les".

G!


54/85

Delicia

!. Qse an interactive boo&.

2. Prompt Delicia to push the sounds=do the pu$$le.

B. s& Delicia 0uestions about the boo&.

G. s& Delicia to ma&e noises=movements to go along

with the boo&.

?. Prompt Delicia to pic& a prop=doll=toy to go along with

the boo&.

/. #ay !-2-B, 1urn the pageF in an e3cited voice and

then have Delicia turn the page.

4onnor

!. 'ave 4onnor dress up as &ing of the gameF.

2. Prompt 4onnor to tell stories about the characters in

the game.

B. )a&e materials larger (large dice, fish bowl".

G. Encourage gross motor play (color road, step stool".

?. Prompt 4onnor to say B-2-!-%last offF before histurn.

G2


55/85

1able G

$ifferences between Pre'%ntervention and Post'%ntervention )easures

OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO

)easure Pre-Intervention Post-Intervention t- value 4ohen5s d

)ean )ean

OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO

Dre0uency of Problem %ehavior /.BB !.+B +.TT B.!B

(E3perimental 4onte3t"

Intensity of Problem %ehavior ?.BB !.BB !.*?TT ?.@*

(E3perimental 4onte3t"

Dre0uency of Problem %ehavior ?.+B 2.+B !!./2TT B.2!

(4linical 4onte3t"

Intensity of Problem %ehavior ?.BB 2.!@ /./GTT 2.B

(4linical 4onte3t"

8lobal Problem %ehavior 2@. !?. *.*!TT 2.!G

(%4 irritability subscale"

#urgency=E3traversion ?.@/ ?.@+ .@ .G

(4%L"

Parenting #tress [email protected] *!. !./? ./?

(P#I"

'ome #ituations ?*.BB G/.+B !.+ .@/

('#4"

Parental 4ontrol /*.? @@.? !.!? .//

(P64#"

OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO

TpN.?

TTpN.!

GB


56/85

Digure 4aptions

-igure 5. 1he percent of activity steps completed and latency to session termination for

the first three participants (group !" during baseline and intervention. 1he solid blac& bars

denote sessions that included problem behavior. 1he grey bars denote sessions in which

the activity was successfully completed without any problem behavior.

-igure 6. 1he percent of activity steps completed and latency to session termination for

the second three participants (group 2" during baseline and intervention. 1he solid blac&

bars denote sessions that included problem behavior. 1he grey bars denote sessions in

which the activity was successfully completed without any problem behavior.

-igure 7. 1he percent of activity steps completed for all si3 participants (group ! and 2"

during clinical conte3ts. F is the percent of activity steps completed in baseline and

%F is the mean percent of activity steps completed during intervention.

-igure 8* 1he latency to session termination for all si3 participants (group ! and 2" during

clinical conte3ts. 1he solid blac& bars denote baseline sessions that included problem

behavior. 1he grey bars denote intervention sessions in which the activity was

successfully completed without any problem behavior. It should be noted that )ichael

and Delicia5s clinical conte3ts (restaurant and grocery store" ta&e considerable more time

than the other clinical conte3ts (snac&, board game, and academic tas&s" when

successfully completed.

GG


57/85

Digure !

Percent

StepsCompleted

0

20

40

60

80

100

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 170

5

10

15

20

Michael

Percent

StepsCompleted

0

20

40

60

80

100

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

Laten

cyMin!"

0

1

2

3

4

5

6#o$$ie

Percent

StepsCompleted

0

20

40

60

80

100

Sessions

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

LatencyMin!"

0

2

4

6

8

1012

14

16%anny

LatencyMin!"

&aseline 'nter(ention

G?


58/85

Digure 2

Percent

StepsCompleted

0

20

40

60

temerarment autism

Documents