Bx. intervention article 2004

krystalized

BehavioralInterventions-2004-Ellis-Modificationstobasicfunctionalanalysisproceduresinschoolsettingsa.pdf

Home >Applied Sciences homework help >Bx. intervention article 2004

Behavioral Interventions

Behav. Intervent. 19: 205–228 (2004)

Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/bin.161

MODIFICATIONS TOBASIC FUNCTIONAL ANALYSIS PROCEDURES IN SCHOOL SETTINGS: A SELECTIVE REVIEW

Janet Ellis* and Sandy Magee

University of North Texas, Denton, TX, USA

This review describes applied behavioral research involving functional analyses conducted in public

school settings. Functional analyses in public school settings often require added conditions. The

modified conditions described herein include changes to experimental designs, antecedent changes that

include task variation, tasks included, idiosyncratic variables, physiological conditions, and modified

escape conditions. Finally, consequent modifications cover peer attention, tangibles, varied attention,

and altered escape. Copyright # 2004 John Wiley & Sons, Ltd.

INTRODUCTION

The primary body of functional analysis (FA) literature has historically focused on

persons with developmental disabilities in institutional/residential settings who

engaged in severe self-injurious behavior (SIB). Mace and Lalli (1991) noted that

interventions based on FAs conducted in experimental settings under highly

controlled analog conditions may be effective only to the extent that those analog

conditions match the subject’s natural environment. Johnston (1993) recommended

that, once a procedure has been experimentally developed, its value and applicability

should be assessed under practical/natural conditions. Further, passage of Public Law

105-17, Individuals with Disabilities Education Act (IDEA), in 1997 mandated that a

‘functional behavioral assessment’ be conducted on students who exhibit significant

behavior and adjustment problems. For at least these reasons, FA research has moved

beyond the tightly controlled laboratory setting and into more natural environments

involving more diverse populations. Development of behavioral assessments of

problem behavior in school settings had empirical roots—for example, 36 years ago

Thomas, Becker, and Armstrong (1968) noted that classroom teacher’s disapproval

increased rates of student’s disruptive behavior. These assessments allowed effective

*Correspondence to: Janet Ellis, Department of Behavior Analysis, University of North Texas, P.O. Box 310919, Denton, TX 76203-0919, USA. E-mail: [email protected]

1099078x, 2004, 3, D ow

nloaded from https://onlinelibrary.w

iley.com /doi/10.1002/bin.161 by B

ehavior A nalyst C

ertification, W iley O

nline L ibrary on [14/11/2022]. See the T

erm s and C

onditions (https://onlinelibrary.w iley.com

/term s-and-conditions) on W

iley O nline L

ibrary for rules of use; O A

articles are governed by the applicable C reative C

om m

ons L icense

behavior change procedures to be implemented in the classroom across a wide range

of student behavior (Thomas et al., 1968; Zimmerman & Zimmerman, 1962). Not

until the early 1990s did FA procedures begin to proliferate in various school

environments (Northup et al., 1994; Sasso et al., 1992).

Iwata, Dorsey, Slifer, Bauman, and Richman (1982/1994) proposed a general

model for analysis of the determinants for problem behavior, which concurrently

assessed the target responses across four conditions (three tests and one control)

using the multielement experimental design (Hanley, Iwata, & McCord, 2003). While

there are two frequently reported types of FA, the antecedent–behavior (AB) model

(Carr & Durand, 1985) and the antecedent–behavior–consequence (ABC) model

(Iwata et al., 1982/1994), the preponderance of data reported (87%) implemented/

involved the ABC model (Hanley et al., 2003). Furthermore, because potential

reinforcers are not manipulated in the AB FA, and consequences maintaining

problem behavior must be inferred, the ABC model is said to provide a more rigorous

demonstration of causation (Hanley et al., 2003). Finally, the ABC FA model has

evolved through changes to the four conditions initially described in the Iwata et al.

(1982/1994) seminal study, and implemented in a majority of the FAs reported (e.g.

alone/ignore condition—59.6%; attention—82.6%; play—91.7%; demand/

escape—89.2%).

Ellis and Magee (1999) noted that the lack of experimental control in the

classroom setting, the presence of uncontrolled sources of reinforcement for aberrant

behavior, and the possible resistance of school personnel to the remediation process

may necessitate modifications to the analog conditions described by Iwata et al.

(1982/1994). Sugai, Horner, and Sprague (1999) suggested that functional analysis

research continue so that the technology will meet the diverse situations occurring in

our schools, homes, and communities. These authors recommended increasing ‘the

efficiency of the process within a variety of environments, especially public school

settings . . . ’ (p. 256). However, Axelrod (1987) warned that without modification

school personnel may view functional analysis methodology as complex, time-

consuming or contrived, and thus reject the FA out of hand. ‘Adoption of functional

analysis by teachers almost certainly will be governed by the complexity of the

analysis and the time it requires’, according to Scruggs and Mastropieri (1996, p. 70).

Walker and Sprague (1999) point out additional problems with the functional

behavior analysis approach—i.e. lack of generalizability in other settings or

situations; the labor intensity involved in establishing such generalizability via

multiple FAs conducted for individual students; co-occurring behavioral determi-

nants of problem behavior; restricting participant characteristics; and artificiality of

many of the settings in which FAs have been conducted. By modifying the specifics

of various FA conditions or adding new conditions, each of these issues could be

directly addressed.

206 J. Ellis and S. Magee

1099078x, 2004, 3, D ow

nloaded from https://onlinelibrary.w

iley.com /doi/10.1002/bin.161 by B

ehavior A nalyst C

ertification, W iley O

nline L ibrary on [14/11/2022]. See the T

erm s and C

onditions (https://onlinelibrary.w iley.com

/term s-and-conditions) on W

iley O nline L

ibrary for rules of use; O A

articles are governed by the applicable C reative C

om m

ons L icense

WHY MODIFICATIONS ARE NECESSARY

While the degree of control afforded in laboratory settings may explain why a

substantial proportion of FAs are not conducted in the client’s natural setting, specific

topographies of problem behavior may occur exclusively in particular environments.

Furthermore, interventions based on FAs conducted in those settings may be more

effective in controlling the problem behavior in those settings (Mace & Lalli, 1991).

Although this review focuses on FAs conducted in school environments, the need for

modifiying FA conditions is not limited to academic settings. FAs conducted in non-

laboratory settings such as hospitals, private homes, group homes, outpatient clinics,

and supervised employment settings also have necessitated revisions to the four

standard experimental conditions.

According to Broussard and Northup (1995), since 1985 school psychologists have

been prevailed upon to emphasize assessment activities that will directly lead to

effective interventions. Technology for treating behavior problems has moved from

the traditional approach of selecting interventions from a textbook list (e.g. time-out,

overcorrection, differential reinforcement of other behavior) to one that is based

directly on the FA of the problem behavior itself. However, those attempting to use

FA to assess in-school problem behavior have been met with several challenges

requiring modifications to standard analog FA procedures. These challenges include

unwillingness on the part of school administrators and classroom personnel to allow

experimental analyses that explicitly set the occasion for high rates of problem

behavior. Likewise, Iwata (1996) noted that, ‘Some clients have unusual histories that

may require modifications to the [general set of] conditions or the addition of new

conditions’. Finally, systematic alteration of the analysis conditions may be needed

‘if initial assessment data are unclear, consistency of implementation has been

verified, and conditions have been attempted using the reversal design . . . ’ (p. 2).

Conducting an FA of problem behaviors may require consideration of a wider array

of antecedent and consequent variables, interacting antecedent events, and complex

classes of behavior. The goal of modifying the standard methods is to increase the

confidence that results will generalize outside the analysis setting and generate a

successful intervention. ‘Evolution of functional analysis methodologies is inevitable

and is to be encouraged’ (Mace, 1994, p. 385).

Instead of modifying FA procedures in order to perfect the ‘true’ FA procedure,

Horner (1994) encourages the development, improvement, and systematic compar-

ison of multiple procedures as the technology continues to be refined. Existing

procedures that focus on a four-part taxonomy of variables that maintain problem

behaviors (attention, tangible, escape, automatic) may require modification to

improve the precision with which we identify the variables controlling problem

behaviors in a specific context.

Modifications to basic functional analysis procedures 207

1099078x, 2004, 3, D ow

nloaded from https://onlinelibrary.w

iley.com /doi/10.1002/bin.161 by B

ehavior A nalyst C

ertification, W iley O

nline L ibrary on [14/11/2022]. See the T

erm s and C

onditions (https://onlinelibrary.w iley.com

/term s-and-conditions) on W

iley O nline L

ibrary for rules of use; O A

articles are governed by the applicable C reative C

om m

ons L icense

Iwata and his associates (1982/1994) developed a single comprehensive analysis

procedure that tested all three of Carr’s (1977) hypotheses for the controlling

conditions of SIB (positive reinforcement, negative reinforcement, and independent

variables). These three hypotheses seemed to account for most environmentally

based problem behaviors. Repp (1994) noted that, because Iwata’s analysis procedure

passed tests of systematic replication, his FA model came to be considered a

standardized approach, which can identify operant functions of aberrant behavior and

lead to the development of effective interventions. The FA procedure involves the

actual manipulation of environmental conditions to determine their influences on

problem behavior. Analog conditions designed to simulate the contingencies

suspected to maintain problem behaviors in the natural environment are system-

atically presented and withdrawn while recording problem behavior occurrences. The

standard analog conditions include alone, attention, play, and demand conditions.

Conditions are presented in the order listed to limit unwanted yet maximize desirable

carryover effects and are rapidly alternated in an arrangement known as a

multielement format. The four standard analysis condition procedures are described

below.

The basic assessment design of a functional analysis includes four specific

conditions: alone, which is a condition that tests for automatic reinforcement in that

there is no programmed source of social reinforcement—such as attention—and if

the targeted behavior occurs at a high rate there is reason to speculate that the

behavior itself is producing its own source of reinforcement (as in self-stimulation);

attention, which is a condition that assesses the role of positive social reinforcement

in maintaining high rates of targeted behavior; if the target behavior is higher in this

condition than in others, the indications are that the behavior is differentially

sensitive to attention as a consequence; play, which is the control condition for the

other three test conditions because the student is not alone, attention is available, and

no demands are placed on the student; demand, which tests for social negative

reinforcement—such as consistently making requests for student compliance with

the task at hand—in the form of escape from demands; if the target behavior occurs

most frequently in this condition, it is possible that the behavior is differentially

sensitive to escape as a reinforcer.

The purpose of this review is to examine the FA as it has been modified for use in

analyzing variables accounting for problem behavior in school settings. Based on the

Hanley et al. (2003) best practice recommendation, only experimental studies that

manipulated potential maintaining consequences were included in this review. Next,

FAs conducted with significant changes in the aforementioned standard procedures

were considered modified. Articles that met the preceding criteria are categorized

according to the specific variable(s) that were altered. In reviewing this literature FA

modifications were found to fall into three general areas: changes to experimental

208 J. Ellis and S. Magee

1099078x, 2004, 3, D ow

nloaded from https://onlinelibrary.w

iley.com /doi/10.1002/bin.161 by B

ehavior A nalyst C

ertification, W iley O

nline L ibrary on [14/11/2022]. See the T

erm s and C

onditions (https://onlinelibrary.w iley.com

/term s-and-conditions) on W

iley O nline L

ibrary for rules of use; O A

articles are governed by the applicable C reative C

om m

ons L icense

design, changes to antecedent components and changes to consequent components

(Table 1). Specific research studies exemplifying each of the aforementioned

modifications are discussed below.

EXPERIMENTAL DESIGN MODIFICATIONS

The modifications that will be described herein include FAs using experimental

designs other than multielement/alternating treatment design or those that conducted

conditions in a non-standard order (other than random or the alone, attention, play,

demand sequence).

Berg et al. (2000) hypothesized that the sequence of FA conditions could have a

substantial impact on outcomes—specifically, skewing the effect of attention as a

reinforcer. These authors modified the order of condition presentation to examine

sequence effects. They conducted a standard attention condition following either a

play or a demand condition. They found higher target rates in attention conditions

that followed demand conditions, versus those following the play condition,

demonstrating that sequence of conditions can alter FA outcomes.

Carr et al. (1997) conducted FAs using the Iwata et al. (1982/1994) protocol, except

that the alone and play conditions were omitted. Because both interview and natural

observations suggested the problem behavior was socially motivated, only theattention

and demand conditions were run. Authors specifically state that a reversal design, rather

than multielement, was used to allow comparison between these two conditions.

In the work of Magee and Ellis (2000) initial standard initial FAs resulted in the

occurrence of only one of several targeted problem behaviors. The analysis condition

with the highest rates of that one target behavior was extended and extinction applied

in a multiple baseline across target behavior design. This design modification was

used because multiple behaviors were to be assessed and the reinforcement of one of

those behaviors appeared to preclude the occurrence and subsequent analysis of other

target behaviors apparently maintained by the same reinforcer.

Moore, Mueller, Dubard, Roberts, and Sterling-Turner (2002) omitted the alone

condition from their initial FA at the request of the school. Because this FA indicated

multiple control and authors suspected a false positive outcome, the two conditions

with the highest frequency of problem behavior were continued using a reversal

design. Subsequently, outcomes indicated attention-maintained problem behavior,

not multiply controlled behavior.

One FA described by Northup et al. (1994) included no play condition, and the

alone, attention, and escape conditions were conducted only once each for 10 min.

These design changes (lack of condition replication) were made due to the severity of

the participant’s self-injury, which often caused both bruising and lacerations, and

Modifications to basic functional analysis procedures 209

1099078x, 2004, 3, D ow

nloaded from https://onlinelibrary.w

iley.com /doi/10.1002/bin.161 by B

ehavior A nalyst C

ertification, W iley O

nline L ibrary on [14/11/2022]. See the T

erm s and C

onditions (https://onlinelibrary.w iley.com

/term s-and-conditions) on W

iley O nline L

ibrary for rules of use; O A

articles are governed by the applicable C reative C

om m

ons L icense

T ab

le 1

. S

u m

m ar

y o

f F A

re v

ie w

ar ti

cl es

ca te

g o

ri ze

d b

y ty

p e

o f

m o

d ifi

ca ti

o n

A n te ce d en t C h a n g es

C o n se q u en t ch a n g es

E xp er im en ta l

T a sk

Id io sy n cr a ti c

P h ys io lo g ic a l

E sc a p e

P ee r

Ta n g ib le

V a ri ed

A lt er ed

d es ig n

va ri a ti o n

in cl u d ed

va ri a b le

co n d it io n s

so ci a l

a tt en ti o n

es ca p e

B er

g et

al .

(2 0

0 0

) X

B ro

u ss

ar d

& X

X X

N o

rt h

u p

(1 9

9 5 )

B ro

u ss

ar d

& X

X N

o rt

h u

p (1

9 9

7 )

C ar

r (1

9 9

7 )

X X

D er

b y

et al

. (1

9 9

4 )

X F

re a

& H

u g

h es

(1 9

9 7

) X

G u

n te

r et

al .

(1 9

9 3

) X

Jo n

es et

al .

(2 0

0 0

) X

L o

h rm

an n

-O ’R

o u

rk e

& X

Y u

rm an

(2 0

0 1

) M

ac e

et al

. (1

9 8

8 /8

9 )

X M

ag ee

& E

ll is

(2 0

0 0

) X

X M

ag ee

& E

ll is

(2 0

0 1

) X

M cC

o m

as et

al .

(2 0

0 0

) X

X M

o o

re et

al .

(2 0

0 2

) X

X M

u el

le r

et al

. (2

0 0

1 )

X X

N o

rt h

u p

et al

. (1

9 9

4 )

X X

N o

rt h

u p

et al

. (1

9 9

5 )

X X

N o

rt h

u p

et al

. (1

9 9

7 )

X X

O ’R

ei ll

y X

& C

ar ey

(1 9

9 6

) R

o m

ia n

u k

et al

. (2

0 0

2 )

X S

as so

et al

. (1

9 9

2 )

X X

T ay

lo r

& X

R o

m an

cz y

k (1

9 9

4 )

V an

C am

p et

al .

(2 0

0 0

a) X

V an

C am

p et

al .

(2 0

0 0

b )

X V

o ll

m er

et al

. (1

9 9

5 )

X W

al la

ce &

K n

ig h

ts (2

0 0

3 )

210 J. Ellis and S. Magee

1099078x, 2004, 3, D ow

nloaded from https://onlinelibrary.w

iley.com /doi/10.1002/bin.161 by B

ehavior A nalyst C

ertification, W iley O

nline L ibrary on [14/11/2022]. See the T

erm s and C

onditions (https://onlinelibrary.w iley.com

/term s-and-conditions) on W

iley O nline L

ibrary for rules of use; O A

articles are governed by the applicable C reative C

om m

ons L icense

because during the course of this analysis corrective surgery was being completed.

Despite these limitations, the FA findings led to an effective intervention.

Van Camp et al. (2000b) conducted initial FAs using standard procedures with

participant 1, but few occurrences of the target behavior were observed. Anecdotal

observations indicated the target behavior occurred when the participant interacted

with a specific toy. Authors evaluated the effect of toy presence by conducting a toy/

no-toy reversal design comparison. Results of this follow-up reversal suggested that

the target behavior was occasioned by the presence of the toy and maintained

independent of social consequences.

With a second participant initial FA showed the highest levels of target behavior

occurring during the play condition. Because the play (control) condition is expected

to produce differentially low rates of problem behavior, this atypical finding led

authors to hypothesize the problem behavior was maintained either automatically or

by escape from stimuli included in this condition. Thus, a reversal design was used to

compare an escape condition with a no-escape condition. Results indicated that the

problem behavior was occasioned by the continuous presentation of attention and

toys; however, the behavior did not appear sensitive to contingent escape.

To compare brief and extended FAs of disruptive behavior, Wallace and Knights

(2003) implemented a pairwise design followed by a multielement design. The initial

pairwise design brief FAwas modified so that the conditions were alternated as follows:

attention–control, demand–control, ignore–control. This brief pairwise FA was

followed by an extended FA in which these same conditions were alternated in the

standard order within a multielement design. This comparison demonstrated that brief

FAs (i.e. 36 min total) ‘ . . . can be effective [when compared to extended analyses—i.e.

310 min total] in identifying maintaining variables of disruptive behavior . . . ’ (p. 126).

ANTECEDENT COMPONENT MODIFICATIONS

Modifications to antecedent components of FA conditions described herein will

include (i) varying difficulty level of instructions in demand conditions, (ii) adding

tasks to conditions other than the demand condition, (iii) encompassing idiosyncratic

objects or events in various FA conditions, (iv) conducting sessions under differing

physiological conditions, and (v) antecedent conditions hypothesized to motivate

escape behavior other than demands.

Varying Difficulty Level of Tasks and Task Areas

Broussard and Northup (1995) observed higher rates of disruptive behavior in the

classroom when difficult tasks versus easy tasks were assigned. Based on their

Modifications to basic functional analysis procedures 211

1099078x, 2004, 3, D ow

nloaded from https://onlinelibrary.w

iley.com /doi/10.1002/bin.161 by B

ehavior A nalyst C

ertification, W iley O

nline L ibrary on [14/11/2022]. See the T

erm s and C

onditions (https://onlinelibrary.w iley.com

/term s-and-conditions) on W

iley O nline L

ibrary for rules of use; O A

articles are governed by the applicable C reative C

om m

ons L icense

observations, the authors conducted two modified demand conditions, labeled Escape

from Academic Tasks—difficult and easy. Non-preferred, difficult academic

worksheets were provided in one of these conditions, in the other, preferred, easy

academic worksheets. Escape was delivered contingent upon target occurrences, and

the two conditions differed only in the difficulty level of the task. When these two

conditions were compared, target occurrence was higher in the difficult condition

(70%) than in the easy condition (3%). Furthermore, during the easy condition, all

tasks were 100% accurate and 100% completed. The authors concluded that their

initial hypothesis (i.e. that target behavior was maintained by escape from difficult

academic tasks) was supported by the data.

After initial standard FAs indicated that escape from demands was the controlling

variable for target behavior of three participants, McComas, Hoch, Paone, and El-

Roy (2000) continued the analysis focusing on establishing operations (EOs). The

authors modified the conditions to identify those antecedent events associated with

occurrence of the aforementioned escape-maintained targets. These modified

conditions, which altered the task aversiveness parameter, were compared with a

standard escape condition, and although academic demands set the occasion for the

targets, adding task-related manipulatives, choice of tasks, and novel non-repetitive

tasks reduced severity and rate for all three participants.

Taylor and Romanczyk (1994) conducted a brief FA that included a standard play

condition and four modified attention conditions the authors labeled ‘low-demand,

therapist ignore (LDI); low-demand, therapist attention (LDA); high-demand,

therapist ignore (HDI); high-demand, therapist attention (HDA)’ (original italics, pp.

257–258). Because escape was not delivered in any of these conditions, and attention

was delivered contingently in each of these conditions, unlike the authors, we do not

refer to these as demand conditions but rather as attention conditions modified to

include tasks.

The LDI condition included an easy task and verbal redirection contingent on

problem behavior, whereas the LDA condition included an easy task but attention (in

the form of task assistance) and praise was delivered contingent on appropriate

behavior. In the HDI condition difficult tasks were presented and verbal redirection

followed problem behavior; however, in the HDA condition the difficult task was

accompanied by assistance and praise contingent on appropriate behavior. Following

this analysis, these authors suggested that the problem behavior occurring at highest

rates in LDI and HDI conditions was maintained by attention in the form of

redirection. They also suggested the problem behavior occurring at the highest

percentages during HDI and HDA was maintained by escape from demands. One

participant’s problem behavior occurred at the highest rates during conditions with

low demands (LDI and LDA), which the authors stated could be indicative of

boredom or an undifferentiated behavior pattern. Although authors suggested that

212 J. Ellis and S. Magee

1099078x, 2004, 3, D ow

nloaded from https://onlinelibrary.w

iley.com /doi/10.1002/bin.161 by B

ehavior A nalyst C

ertification, W iley O

nline L ibrary on [14/11/2022]. See the T

erm s and C

onditions (https://onlinelibrary.w iley.com

/term s-and-conditions) on W

iley O nline L

ibrary for rules of use; O A

articles are governed by the applicable C reative C

om m

ons L icense

problem behavior occurring at the highest percentages in conditions involving low

rates of therapist attention could be attention maintained, it is likely that this outcome

could also be attributed to the verbal attention (redirection) made contingent on

occurrences of problem behavior. Next, the authors’ conclusion that problem

behavior exhibited at highest percentages during high demand conditions was

maintained by escape is problematic, because escape as a contingent consequence

was not described as having been delivered. Although the antecedent modifications

made to the conditions could have generated new information, the lack of differential

consequences for the targeted problem behavior(s) makes interpretation difficult.

O’Reilly and Carey (1996) hypothesized that the effects of operations prior to the

analog FA could influence results/outcomes of analog conditions. In this study two

classroom conditions were conducted prior to the FA: classroom demand and

classroom attention. During the classroom demand pre-analysis condition, the teacher

instructed the participant to engage in self-care and hygiene tasks that were difficult

for her to complete. A 10 s escape period was allowed contingent on participant’s

aggression, and instructions reinstated following the 10 s escape period, or when

aggression ceased. A peer included in this condition was given toys but no instructions

or attention. Next, the pre-FA classroom attention condition was conducted. This

condition was conducted as described above except now instructions were delivered to

the peer, while the participant played with toys. Additionally, the teacher delivered

reprimands for 10 s and redirected the participant to toys contingent on aggression.

The experimental conditions were conducted over a 10 day period with classroom

conditions randomly assigned as follows: on five days (days 1, 3, 6, 9 and 10) an

analog analysis was preceded by the classroom demand condition and on five days

(days 2, 4, 5, 7 and 8) by the classroom attention condition. Each analog analysis

included two demand, play, and attention conditions. These were standard Iwata et al.

(1982/1994) procedures except that during the demand condition two different tasks

were presented in random order by the teacher. This change to the demand condition

procedures is another antecedent component modification—varying task areas.

Overall, these modifications resulted in higher levels of aggression in the attention

condition than in the demand or play conditions. During the analog analysis

aggression was substantially higher in demand conditions that followed classroom

attention than those following classroom demand. The authors concluded that ‘This

study demonstrated a functional relationship between prior classroom conditions and

performance under analog analysis conditions’ (p. 584).

Adding Tasks to Conditions Other Than the Demand Condition

Broussard and Northup (1995) conducted FAs for three students referred

for disruptive behavior in regular education classrooms. Following descriptive

Modifications to basic functional analysis procedures 213

1099078x, 2004, 3, D ow

nloaded from https://onlinelibrary.w

iley.com /doi/10.1002/bin.161 by B

ehavior A nalyst C

ertification, W iley O

nline L ibrary on [14/11/2022]. See the T

erm s and C

onditions (https://onlinelibrary.w iley.com

/term s-and-conditions) on W

iley O nline L

ibrary for rules of use; O A

articles are governed by the applicable C reative C

om m

ons L icense

assessment-based hypotheses, each student was exposed to one of three experimental

conditions: teacher attention, peer attention, and escape from demands. All FA

conditions included academic tasks, and in addition to each student’s problem

behavior, data were collected on each student’s academic responding. Although

academic tasks are typically included only in the standard demand condition, these

authors sought to demonstrate the possibility of extending FA procedures to the

regular education classroom and, therefore, included academic activities in all

conditions. FA conditions were conducted using a reversal design, including a

contingency reversal condition to evaluate differential reinforcement and extinction

as potential intervention components. These modifications allowed experimenters to

demonstrate contingencies maintaining disruptive behavior in FA conditions. ‘For all

3 students, contingency reversals resulted in near zero occurrences of target

behaviors . . . [and] . . . a corresponding increase in academic work completion and

accuracy . . . ’ (p. 161).

Broussard and Northup (1997) conducted FAs in regular education classrooms

with four students. Because classroom observations indicated that teacher attention,

peer attention, and escape from demands were likely to follow disruptive behavior,

these three conditions were implemented with each student. Tasks were included in

all three conditions as follows: teacher and peer attention conditions included

assignments previously completed with at least 90% accuracy, whereas the escape

condition included assignments previously completed with less than 50% accuracy.

As this study sought to evaluate the effectiveness of intervention techniques aimed at

decreasing disruptive behavior and increasing on-task behavior, data were collected

on both behavioral categories—disruptive and on task. For each student data

indicated that peer attention maintained disruptive behavior. Subsequent contingency

reversals resulted in reductions in disruptive behavior to zero, and ‘substantial

increases in on-task behavior for all participants’ (p. 73). By modifying the FA so that

all conditions included academic tasks, these researchers were able to demonstrate

both behavioral topographies (disruptive and on task) were members of the same

response class and, were, therefore, remediated using a single contingency reversal.

Because the behavior of interest was hand-raising and a hand-raise/disruptive

behavior sequence, Gunter, Jack, Shores, Carrell, and Flowers (1993) conducted all

classroom analysis conditions with the participant engaged in independent work

activities (language, math, or social studies). Pre-assessment lag sequential analysis

indicated that the most common antecedent for the 12-year-old special education

student’s disruptive behavior was hand-raising, while the most common consequence

for disruptive behavior was teacher attention. Based on data from the lag sequential

analysis, authors hypothesized that disruptive behavior was maintained by attention.

Therefore, two conditions were conducted: correction (an attention condition) and

planned ignoring. During the correction condition teachers delivered reminders

214 J. Ellis and S. Magee

1099078x, 2004, 3, D ow

nloaded from https://onlinelibrary.w

iley.com /doi/10.1002/bin.161 by B

ehavior A nalyst C

ertification, W iley O

nline L ibrary on [14/11/2022]. See the T

erm s and C

onditions (https://onlinelibrary.w iley.com

/term s-and-conditions) on W

iley O nline L

ibrary for rules of use; O A

articles are governed by the applicable C reative C

om m

ons L icense

contingent on the hand-raise/disruptive sequence and delivered praise with assis-

tance contingent on hand raises without disruptive behavior. During the planned

ignoring condition teachers continued the hand-raise-without-disruption contin-

gency. Although the form of attention could be considered negative in one instance

and positive in the other, because attention was delivered contingent on both hand-

raise-with-disruption and hand-raise-without-disruption, there was no definitive

demonstration of the variable(s) maintaining the target behavior. However, because

the disruptive behavior decreased in the planned ignoring condition, it is still possible

to infer that the target behavior was maintained, at least in part, by attention.

Northup et al. (1995) analyzed out-of-seat behavior and inappropriate vocaliza-

tions of three students with ADHD. These authors conducted two analysis conditions:

teacher attention and peer attention. In both conditions students were given easy math

worksheets and instructed to remain seated and work quietly. Although these

attention conditions were modified to include tasks normally assigned only in the

demand condition, these authors do not state a specific rationale for task inclusion.

However, ‘Contingent peer attention resulted in a substantially higher percentage of

target behaviors than did teacher attention’ (p. 227).

Northup et al. (1997) examined inappropriate vocalizations, out-of-seat, and

playing-with-objects behavior of one student diagnosed with ADHD. These authors

also included tasks in all conditions, explaining that easier tasks were included in the

attention condition to decrease the probability of escape-maintained responses.

However, again, the authors do not explain specifically why tasks were included in all

conditions.

In an experimental analysis conducted by the teacher in an elementary school self-

contained classroom for students with autism, Sasso et al. (1992) modified their

attention condition to include a self-help task and their tangible condition to include

individual class work. Authors explained that these two conditions were modified

to include tasks so that the teacher could conduct the experimental analysis during

the ongoing classroom routine. Despite the addition of tasks to two test conditions,

target behavior occurred at highest rates during the escape condition for both

students.

Romaniuk et al. (2002) conducted an FA with seven students with various

diagnoses and problem behaviors. Because teachers suggested that certain academic

assignments appeared to evoke the problem behavior of these students, the attention

condition was modified to include tasks. More specifically, during the attention

condition the therapist worked with each student on one of the teacher-nominated

tasks. Contingent on target occurrences, the therapist delivered mild reprimands (e.g.

‘Don’t do that’, ‘That’s not appropriate’, ‘Keep on working’). In addition to this

change to standard attention condition procedures, experimenters included the

typical (i.e. demand condition) three-prompt sequence following noncompliance.

Modifications to basic functional analysis procedures 215

1099078x, 2004, 3, D ow

nloaded from https://onlinelibrary.w

iley.com /doi/10.1002/bin.161 by B

ehavior A nalyst C

ertification, W iley O

nline L ibrary on [14/11/2022]. See the T

erm s and C

onditions (https://onlinelibrary.w iley.com

/term s-and-conditions) on W

iley O nline L

ibrary for rules of use; O A

articles are governed by the applicable C reative C

om m

ons L icense

Results indicated that the three participants’ problem behavior was maintained by

attention under demand antecedent conditions.

Encompassing Idiosyncratic Objects or Events in Various Functional Analysis Conditions

When initial FA for one student did not reflect the high rates of problem behavior

reported by the teacher, and naturalistic observations revealed that absence of certain

idiosyncratic variables (small objects) evoked high rates of target behavior, Carr,

Yarbrough, and Langdon (1997) conducted a second series of FAs. Although initial

attention and demand conditions, conducted with small objects included across all

sessions, did not clearly identify the maintaining variable(s), replicating these

conditions without small objects present resulted in high rates of targets in the

demand condition. Despite this outcome, authors hypothesized the target behavior

was motivated by the absence of small objects rather than escape from demands.

Given this possibility, a third set of modified conditions was run in which demands

were eliminated and the attention condition conducted with either small or large balls

present. Outcomes supported the hypothesis in that problem behavior occurred at low

frequencies in the small-balls sessions and at ‘substantially higher frequencies’

(p. 680) in the large-balls sessions despite the elimination of demands.

Derby et al. (1994) suggested the analyzing the outcomes of FAs conducted to

assess multiple aberrant responses simultaneously could be problematic. Although it

is generally assumed that separate topographies of a problem behavior are members

of the same functional response class, this is not always the case. At least

occasionally, occurrence of multiple topographies of a target behavior maintained by

different reinforcers may have misrepresented FA outcomes. To test this possibility,

authors conducted an aggregate analysis [multiple topographies assessed simulta-

neously] of aberrant behavior, which indicated that both stereotypy and self-injury

were maintained by sensory reinforcement. However, subsequent separate analyses

of self-injury and stereotypy indicated these targets were a function of two different

variables: escape maintained self-injury, whereas sensory reinforcement accounted

for stereotypy.

FAs included these modified conditions: high sensory, in which students were

exposed to non-contingent loud and constant noise while target behavior was

ignored; and escape from high sensory condition, the same as high sensory except

that all sources of stimulation were terminated for at least 15 s contingent upon target

behavior. Including noise as an antecedent, hypothesized to be aversive and,

therefore, to evoke escape behavior, is a modification of the standard demand

condition. This modification resulted in high rates of target behavior for some

participants in this study.

216 J. Ellis and S. Magee

1099078x, 2004, 3, D ow

nloaded from https://onlinelibrary.w

iley.com /doi/10.1002/bin.161 by B

ehavior A nalyst C

ertification, W iley O

nline L ibrary on [14/11/2022]. See the T

erm s and C

onditions (https://onlinelibrary.w iley.com

/term s-and-conditions) on W

iley O nline L

ibrary for rules of use; O A

articles are governed by the applicable C reative C

om m

ons L icense

Mace, Yankanich, and West (1988/1989) assessed potential environmental

components of aberrant classroom behavior, stereotypy. The authors state that rather

than ‘employing a set of standard conditions for most subjects, conditions [were]

developed which are idiosyncratic to a given student and his or her educational

setting’ (p. 74). Student records, medical records, anecdotal reports, and direct

observations led authors to predict antecedent conditions under which high rates of

stereotypy would occur. In testing their predictions, four modified FA conditions

were conducted.

In the playtime–no music condition target rates were measured during unstructured

play in the classroom, an antecedent situation predicted to evoke high rates. A second

condition, playtime–music, was identical except that easy listening music played at a

moderate volume was introduced. This antecedent variable was added after direct

observation indicated that it was associated with reduced rates of stereotypy. A third

condition, playtime–music with headphones, was implemented to determine whether

decreasing background noise while continuing to play music would result in further

reductions in stereotypy. Finally, the playtime–quiet room condition was executed to

assess whether the music or the reduction in background noise was the critical

variable. Although this analysis identified relationships between naturally occurring

antecedent events in the classroom setting and the problem behavior of interest, the

maintaining consequence was not identified in this article.

After anecdotal observations indicated that the student engaged in target

behavior when interacting with a specific toy, a Bumble Ball1, Van Camp et al.

(2000b) made changes to the FA to evaluate whether the presence of the ball

differentially affected levels of target behavior. Authors conducted an alone

condition in which the student had continuous access to the ball and, for

comparison, an alone condition with no ball. When the target behavior occurred

only in those conditions with the ball present a second FA was conducted. Three

conditions—contingent removal (therapist removed ball for 20 s contingent on

target occurrences), no interaction (ball present), and contingent access (student

received 20 s access to ball contingent on target occurrence)—were conducted to

evaluate potential functional relationships.

When undifferentiated levels of target behavior occurred in this analysis, authors

conducted a microanalysis involving three additional modified conditions to isolate

the precise component(s) of the ball (vibration, noise, and/or plastic protrusions) that

actually evoked the target behavior. In a reversal design, three conditions—intact, no

vibration and no protrusions—were compared. Results of this component analysis

revealed that vibration was the stimulus feature that was associated with target

occurrence. ‘Although results of the analysis clearly identified the functional

antecedents of problem behavior, the precise sources of reinforcement that

maintained the behavior were not determined’ (p. 219).

Modifications to basic functional analysis procedures 217

1099078x, 2004, 3, D ow

nloaded from https://onlinelibrary.w

iley.com /doi/10.1002/bin.161 by B

ehavior A nalyst C

ertification, W iley O

nline L ibrary on [14/11/2022]. See the T

erm s and C

onditions (https://onlinelibrary.w iley.com

/term s-and-conditions) on W

iley O nline L

ibrary for rules of use; O A

articles are governed by the applicable C reative C

om m

ons L icense

Conducting Sessions Under Differing Physiological Conditions

Lohrmann-O’Rourke and Yurman (2001) analyzed the variables maintaining hand

and object mouthing of a 6-year-old boy in a self-contained classroom. When the FA

revealed high rates in all three test conditions compared with the low rates in the two

control conditions, researchers concluded the behavior was multiply controlled.

However, because of the boy’s chronic sinus infections authors suspected presence or

absence of this physical condition might be functionally related to target behavior

occurrences. Because the first set of conditions was conducted when symptoms were

present and antibiotics were being administered, an absence-of-symptoms condition

was carried out. Comparing target rates under these two antecedent physiological

conditions revealed that mouthing was 21% more likely to occur when infection

was present.

To assess potential drug–behavior interaction effects Northup et al. (1997)

performed each of their conditions, teacher reprimand, peer prompts, and time-out

with both medication (methylphenidate) and placebo. Outcomes indicated that

‘ . . . (a) disruptive behavior was maintained by positive reinforcement in the form of

peer attention and (b) methylphenidate functioned to alter [reduce] either the saliency

of peers as [evocative] antecedent stimuli or the reinforcing value of peer attention’

(p. 123).

Other Antecedents Evoking Target Behavior

While the standard FA demand condition involves repeated task presentations,

other antecedent events may also set the occasion for escape. In fact, Frea and

Hughes (1997) implemented an escape-social condition in their FAs of inappropriate

social communications of two high-school students. This condition, in which the

teacher maintained conversation throughout the session but discontinued conversa-

tion contingent on target occurrences, tested for sensitivity to escape from social

interaction. Next, the attention condition was modified to include a peer to whom the

teacher addressed conversation until a target occurred. These modified conditions

showed that social attention was a reinforcer for student 1’s perseverative speech;

while for student 2 escape from social attention maintained inappropriate laughter

and displays of disgust.

ALTERATIONS TO CONSEQUENT COMPONENTS

These modifications describe analyses that included alternate forms of escape,

attention, or the delivery of tangible items contingent on target behavior. Broussard

and Northup (1995) state that ‘Current literature suggests three variables as most

218 J. Ellis and S. Magee

1099078x, 2004, 3, D ow

nloaded from https://onlinelibrary.w

iley.com /doi/10.1002/bin.161 by B

ehavior A nalyst C

ertification, W iley O

nline L ibrary on [14/11/2022]. See the T

erm s and C

onditions (https://onlinelibrary.w iley.com

/term s-and-conditions) on W

iley O nline L

ibrary for rules of use; O A

articles are governed by the applicable C reative C

om m

ons L icense

often related to classroom disruptive behavior: teacher attention, peer attention, and

escape from academic demands’ (p. 152). Therefore, when analyzing disruptive

classroom behavior of a student in regular education classes, these authors added a

contingent peer attention condition to their analysis set. In this condition two selected

peers joined the student participant in another classroom where all three were

instructed to ‘Work quietly and complete these [academic] worksheets’ (p. 157).

Students considered likely to respond to target behavior occurrences were included,

but were not given instructions on consequence delivery. This condition produced the

highest percentage of target occurrences and lowest rates of work completion across

all conditions conducted, indicating that the student participant’s disruptive behavior

was occasioned by peer attention.

Prior to FA, Broussard and Northup (1997) conducted descriptive observations in

four student participants’ classrooms. Observations indicated that teacher attention,

peer attention, and termination or delay of tasks were likely to follow target behavior

occurrences—off task and out of seat. Based on this, a peer attention condition was

included in each participant’s FA. In this condition a peer ‘confederate’ was

instructed to provide consequences for the behavior of interest by speaking to the

student participant if out-of-seat behavior or talking during work time occurred.

Although no control condition was included, for each of the four participants the

peer attention condition was associated with highest percentage of intervals of

target behavior when compared with the other conditions: teacher attention and

escape.

Jones et al. (2000) noted that direct observation prior to FA indicated that

disruptive behavior occurred during academic tasks when peers were present and

interacting with the student participant. A peer attention condition was included in

their FA in which a peer made statements of disapproval, casual conversation,

laughed or imitated following target occurrences. The highest levels of disruptive

behavior were observed during this peer attention condition, ranging from 60 to

100% of intervals. Although no control condition was included, these high rates

demonstrated that the problem behavior was maintained by peer attention when

compared with the escape and teacher attention condition outcomes. The authors

discussed limitations in drawing conclusions when the peer attention condition is

the only test condition in which a peer is present, implying that some disruptive

topographies may be more likely to occur merely as a function of peer presence.

Northup et al. (1995) included only two test conditions in their initial FA of

classroom disruptive behavior. Out of seat and inappropriate vocalizations were

assessed in both teacher and peer attention conditions. During the peer attention

condition a peer ‘confederate’ was instructed to ‘Say something’ to the student

participant if that person was out of seat or talked. Participants were given easy math

worksheets and instructed to remain in their seats and to work quietly. ‘Contingent

Modifications to basic functional analysis procedures 219

1099078x, 2004, 3, D ow

nloaded from https://onlinelibrary.w

iley.com /doi/10.1002/bin.161 by B

ehavior A nalyst C

ertification, W iley O

nline L ibrary on [14/11/2022]. See the T

erm s and C

onditions (https://onlinelibrary.w iley.com

/term s-and-conditions) on W

iley O nline L

ibrary for rules of use; O A

articles are governed by the applicable C reative C

om m

ons L icense

peer attention resulted in a substantially higher percentage of target behaviors than

did teacher attention’ (p. 227), indicating that peer attention was the primary

reinforcer for target occurrences. Authors suggest that, ‘ . . . peer and teacher attention

may not be functionally equivalent [and] that peer attention can function as a unique

form of positive reinforcement’ (p. 228).

While assessing drug–behavior interaction effects, Northup et al. (1997) included

what they referred to as a ‘peer prompts condition’ (p. 123). Similar to previous peer

attention conditions described above, a peer ‘confederate’ was instructed to speak to

the student participant if he left his seat or talked. The students were each given

worksheets and instructed to stay seated and work quietly. The disruptive behavior

occurred primarily during the peer attention sessions when the student participant

received placebo. ‘These results are consistent with the hypothesis that . . . disruptive

behavior was maintained by positive reinforcement in the form of peer attention . . . ’ (p. 123).

FAs that included a peer attention condition for comparison with a teacher (adult)

attention condition have demonstrated that (i) peer attention is a distinctive form of

social positive reinforcement, which (ii) would not have been identified as a

maintaining variable if the standard FA format had been used. FA conditions that

have been modified to include nonstandard consequence delivery are not limited to

peer attention, but may also include tangible access contingent on occurrence of the

behavior of interest. Although this condition is not typically considered part of the

standard FA (Iwata et al., 1982/1994), Iwata was the first to include this condition in

an analysis set.

While the authors do not explain specifically why it was included, McComas et al.

(2000) conducted a tangible condition with one of their three participants. Access to

tangible items was restricted and only delivered contingent on occurrences of

destructive behavior. This condition varied from the Iwata tangible condition in that

the participant had unlimited access to teacher attention throughout. However,

destructive behavior occurred almost exclusively during demand sessions, with little

or no destructive behavior occurring in tangible sessions.

Moore et al. (2002) set out to analyze the variables maintaining the SIB of a 6-year-

old preschool student. Although the alone condition was omitted at the request of the

public school elementary administrators, attention, play, and demand conditions were

supplemented with a tangible condition. Authors describe this tangible condition as

being preceded by 1 min access to a highly preferred stimulus, juice. Then juice was

removed, non-contingent access to toys provided, and further access to juice was

made contingent on SIB. Specifically, following each instance of SIB the therapist

provided 30 s access to juice, saying ‘You must want your juice’. Higher rates of SIB

occurred ‘ . . . during attention and tangible conditions than in the other functional

analysis conditions’. (p. 284). Authors state that this outcome ‘ . . . appear[s] to

220 J. Ellis and S. Magee

1099078x, 2004, 3, D ow

nloaded from https://onlinelibrary.w

iley.com /doi/10.1002/bin.161 by B

ehavior A nalyst C

ertification, W iley O

nline L ibrary on [14/11/2022]. See the T

erm s and C

onditions (https://onlinelibrary.w iley.com

/term s-and-conditions) on W

iley O nline L

ibrary for rules of use; O A

articles are governed by the applicable C reative C

om m

ons L icense

demonstrate multiple functions for . . . SIB’ (p. 284). However, because verbal

attention was delivered with the juice in tangible sessions, a follow-up analysis with

conditions further modified compared tangible delivery paired with verbal attention

to tangible delivery with no verbal attention. Results illustrated that when verbal

attention accompanied the juice SIB occurred at a much higher frequency than when

juice alone was delivered, indicating a false positive outcome in the previous tangible

condition.

In the FA by Mueller et al. (2001), a tangible condition was implemented instead of

an alone condition. The student participant had 1 min access to a highly preferred

book that was removed at the onset of the session, when the experimenter offered

stuffed animals as alternatives. Although these alternatives were available for the

entire session, each instance of aggression resulted in 30 s access to the book. Rates

of aggression were highest in the tangible condition in comparison with the other

conditions, attention, play, and demand, indicating aggression was maintained by

access to preferred items. Following these results, authors then systematically

investigated the impact of changes to antecedent variables—restricting access to

high-, medium- or low-preference items—on FA outcomes.

A stimulus preference assessment confirmed previous teacher report, placing three

books in the high-preference group (HPG); three spinning objects in the medium-

preference group (MPG); and three stuffed animals in the low-preference group

(LPG). This information formed the basis for four subsequent antecedent

manipulation phases: phase 1—restricted access to items from the HPG while

offering an alternative item from either the HPG, MPG, or LPG; phase 2—restricted

access to items from the MPG while offering an alternative item from the HPG,

MPG, or LPG; phase 3—restricted items from LPG and offered alternatives from the

HPG, MPG, or LPG. Phase 4 was a replication of phase 1, and throughout each phase

all possible pairings of restricted and alternative items were presented at least five

times. Outcomes of this antecedent analysis indicated that the highest rates of

aggression occurred when HPG items were restricted, regardless of the alternatives

offered. However, ‘A substantial amount of aggression occurred even when items

from the LPG were restricted and items from the HPG were available as alternatives’

(pp. 239–240). The authors suggest three possible explanations for these results, yet

conclude by saying ‘Further research is needed to delineate these processes . . . ’

(p. 240). Additional description of the antecedent analysis including specific

restriction procedures and any consequences delivered contingent on aggressive

behavior might allow more definitive interpretations of these outcomes.

In evaluating the feasibility of school personnel conducting FAs in actual

classroom settings, Northup et al. (1994) examined the severe problematic behavior

exhibited by five student participants. For one student the authors included a tangible

condition in the analog set, yet did not specifically describe those procedures.

Modifications to basic functional analysis procedures 221

1099078x, 2004, 3, D ow

nloaded from https://onlinelibrary.w

iley.com /doi/10.1002/bin.161 by B

ehavior A nalyst C

ertification, W iley O

nline L ibrary on [14/11/2022]. See the T

erm s and C

onditions (https://onlinelibrary.w iley.com

/term s-and-conditions) on W

iley O nline L

ibrary for rules of use; O A

articles are governed by the applicable C reative C

om m

ons L icense

Furthermore, pre-assessment observations of this one student participant led to their

hypothesis that his SIB was maintained by negative reinforcement, and actual

outcomes supported this hypothesis. Therefore, the rationale for modifying the

standard analysis procedures to include this tangible condition is unclear.

Sasso et al. (1992) included a tangible condition in their analyses of the aggression

and inappropriate language of two student participants. This condition was described

as ‘ . . . similar to the attention condition except that a favorite toy, activity, or edible

was provided contingent upon each emission of the target behavior’ (p. 812).

Although the authors concluded that escape maintained target behavior for both

participants, the tangible condition ‘consistently yielded moderate levels of

aggressive behavior’ (p. 815). Their rationale for including this condition was not

stated; however, it may have been included because the authors sought to compare

findings from teacher-conducted classroom analyses with prior experimenter-

conducted analyses.

Van Camp et al. (2000a) conducted FAs with two student participants and included

a tangible condition in their analysis of aggression and self-injury. In this tangible

condition ‘ . . . 20 s of access to leisure materials was delivered contingent upon each

occurrence of aggression or self-injury, and the participant had noncontingent

continuous access to attention’ (p. 549). This attention component differs from prior

descriptions of tangible conditions; however, the authors do not explicitly state the

reason for its inclusion in their analysis. Results for both participants indicated that

their problem behavior was maintained by access to leisure materials.

Vollmer et al. (1995) also employed a tangible test condition in their FAs of

multiple target behaviors for 20 student participants. In this condition, ‘The

participant was provided with access to tangible stimuli just before beginning the

session. When the session began, the tangible stimuli were removed from reach, but

were in view of the participant. Contingent on occurrences of target behavior, access

to the tangible stimuli was presented for about 20 s’ (pp. 566–567). The authors

explained that the tangible condition was conducted only with those students whose

parents or teachers reported problems with tangible stimuli or whose aberrant

behavior was observed to be associated with tangible items. Of the 20 student

participants, six displayed behavioral sensitivity to reinforcement by tangible stimuli

during the analysis procedures.

Altered Forms of Teacher Attention

When results of naturalistic observations indicated that teachers used physical

restraint as a consequence for the problem behavior of two male student participants,

Magee and Ellis (2001) extended the standard analysis set to include a condition

designed to test the possible reinforcing effects of restraint. For student participant 1

222 J. Ellis and S. Magee

1099078x, 2004, 3, D ow

nloaded from https://onlinelibrary.w

iley.com /doi/10.1002/bin.161 by B

ehavior A nalyst C

ertification, W iley O

nline L ibrary on [14/11/2022]. See the T

erm s and C

onditions (https://onlinelibrary.w iley.com

/term s-and-conditions) on W

iley O nline L

ibrary for rules of use; O A

articles are governed by the applicable C reative C

om m

ons L icense

this condition was identical to the attention condition except that instead of making

statements of disapproval, the therapist placed him face down on the floor, holding

his arms for 10 s contingent on target occurrence(s). For student participant 2

procedures were identical to the demand condition except that instead of removing

the task, the therapist folded his arms across his chest and held his wrists for 10 s

contingent on target occurrences. The problem behavior of both students was found

to be sensitive to physical restraint as a consequence, indicating a specific topography

of social positive reinforcement was the maintaining variable.

Altered Forms of Escape

During initial multi-element FAs with two elementary school students (each at a

different school), Magee and Ellis (2000) implemented standard FA conditions:

alone, attention, and play. The demand condition, however, was nonstandard in that

the therapist delivered requests every 10 s, omitted the three-step prompt procedure,

and instead of removing task materials left the room for 30 s contingent on target

occurrences. These changes were made following classroom observations that

indicated that both teachers frequently made repeated verbal demands for which no

guidance/prompting was offered. Intermittently following the occurrence of problem

behavior the teachers would leave the room to obtain assistance. Results indicated

that this form of contingent escape maintained the target behavior of one of the two

students.

Northup et al. (1997) conducted a time-out (demand) condition to test for behavior

maintained by escape. The student was asked to complete difficult math sheets but,

unlike the standard demand condition, was told he would be placed in time out unless

he remained seated and worked quietly. The specific form of the escape consequence

involved turning the student’s chair to face away from his desk, other students, and

ongoing classroom activities. Additionally, the teacher moved away for 30 s

contingent on the occurrence of disruptive target responses. A three-stage prompt

procedure was included beginning with verbal prompts, followed by gestural prompts

and then physical prompts. This prompt sequence, however, was delivered contingent

on out of seat—a target behavior. Although these modifications were likely made so

that the condition would be more analogous to classroom procedures, disruptive

behavior occurred at near zero rates throughout these sessions.

Magee and Ellis (2001) altered the demand condition so that the escape

consequence in the analog condition replicated that which occurred in the classroom.

One of the participants was wheelchair bound and following each target occurrence

(i.e., yelling, self-injury, and aggression) was wheeled into a small time-out room for

30 s. Frequency of target responses increased across consecutive demand sessions

while remaining at or near zero in alone, attention and play conditions.

Modifications to basic functional analysis procedures 223

1099078x, 2004, 3, D ow

nloaded from https://onlinelibrary.w

iley.com /doi/10.1002/bin.161 by B

ehavior A nalyst C

ertification, W iley O

nline L ibrary on [14/11/2022]. See the T

erm s and C

onditions (https://onlinelibrary.w iley.com

/term s-and-conditions) on W

iley O nline L

ibrary for rules of use; O A

articles are governed by the applicable C reative C

om m

ons L icense

DISCUSSION

This review has discussed wide-ranging modifications to standard FA proce-

dures—one major impetus for these modifications has been the setting in which the

analyses have been implemented—public and private schools. Alterations have

included experimental design changes, antecedent variable modifications, and

delivery of various nonstandard consequences described in the 26 published studies

reviewed herein. This research represents the emerging ‘technology of functional

assessment that meets the full range of situations faced in our schools, homes, and

communities’ (suggested by Sugai et al., 1999, p. 253).

Future research extending and altering the procedures may include specific

variables that set the occasion for problem behavior in the classroom and other school

settings. For example, escape-maintained problem behavior may occur only during

particular activities (e.g. physical education class) or in the presence of certain school

staff who have come to function as conditioned aversive stimuli. Likewise, escape

from the repeated delivery of demands may function as a reinforcer when escape

from the task, per se, does not. Antecedent conditions (e.g. academic assignments)

may increase the probability of/set the occasion for attention-maintained problem

behavior, particularly when the attention involves academic assistance. Such

consequences as task reduction, in-school suspension, or at-home suspension may

be powerful reinforcers for targeted problematic behavior. Challenges such as these

may be unique to school settings, highly idiosyncratic, and difficult to reproduce

unless FA conditions are modified. Because these modifications test the roles of

antecedent and consequent events tailored to the student, they may provide more

practical, applicable options when designing the subsequent behavior management

intervention. Furthermore, data indicate that attention extinction for minor problem

behavior may result in the occurrence of more problematic responses (Magee & Ellis,

2000). Replication research, involving systematic implementation of extinction

across target responses, that identifies problem behavior response class hierarchies

could preclude (at least) some of the problematic side-effects associated with

intervention strategies subsequently implemented.

Involving school personnel in assessment procedures while making those procedures

more efficient and directly relevant for designing interventions may promote

acceptance of FA procedures and behavioral technology. Additional examination of

brief FA procedures may be modified for the purpose of simplifying and reducing the

time required to implement them. Doing so may not only encourage wider acceptance

among school personnel, but may also enable other school professional staff (e.g.

school psychologists, school diagnosticians, etc.) to make use of these procedures.

Heeding Axelrod’s (1987) warning, and acknowledging that in some cases school

personnel have neither the time nor the training to carry out FAs, behavior analysts

224 J. Ellis and S. Magee

1099078x, 2004, 3, D ow

nloaded from https://onlinelibrary.w

iley.com /doi/10.1002/bin.161 by B

ehavior A nalyst C

ertification, W iley O

nline L ibrary on [14/11/2022]. See the T

erm s and C

onditions (https://onlinelibrary.w iley.com

/term s-and-conditions) on W

iley O nline L

ibrary for rules of use; O A

articles are governed by the applicable C reative C

om m

ons L icense

can and should function as a consultative resource. Our role as behavior analysts will

include conducting the FA (including any necessary modifications), subsequently

providing an intervention based on the FA outcomes, training the classroom staff to

implement the intervention, and gradually fading out as school personnel become

proficient.

While the need exists for individualized FAs conducted in naturalistic settings,

modified based on direct observation and initial standard FA outcomes, changing a

well established experimental procedure is not without pitfalls. Moving beyond

tightly controlled laboratory settings into more chaotic natural environments exposes

the research to the greater possibility of encountering uncontrolled/confounding

variables in those settings. What at first may appear to be a questionable undertaking

inviting criticism, even skepticism, is, in fact, what occurs in the process of proving

the efficacy, effectiveness, and validity of any new extension of a recognized

technology.

As the populations we serve expand, so too has our methodology evolved to

encompass the greater number and complexity of critical environmental variables.

While applied behavior analysis has as its foundation basic laboratory research, in all

likelihood these modified FA procedures will someday be integrated into educational

environments, including special education, alternative education, and regular

education classrooms. Because of its dynamic properties it is essential that behavior

analysts retain responsibility for the implementation of this highly specialized

analysis model.

ACKNOWLEDGEMENT

The authors wish to thank the two un-named reviewers of this manuscript for their

helpful input.

REFERENCES

Axelrod, S. (1987). Functional and structural analyses of behavior: Approaches leading to reduced use

of punishment procedures. Research in Developmental Disabilities, 8, 165–178.

Berg, W. K., Peck, S., Wacker, D. P., Harding, J., McCommas, J., Richman, D., & Brown, K. (2000). The

effects of presession exposure to attention on the results of assessments of attention as a reinforcer.

Journal of Applied Behavior Analysis, 33, 463–477.

Broussard, C. D., & Northup, J. (1995). An approach to functional assessment and analysis of

disruptive behavior in regular education classrooms. School Psychology Quarterly, 10,

151–164.

Modifications to basic functional analysis procedures 225

1099078x, 2004, 3, D ow

nloaded from https://onlinelibrary.w

iley.com /doi/10.1002/bin.161 by B

ehavior A nalyst C

ertification, W iley O

nline L ibrary on [14/11/2022]. See the T

erm s and C

onditions (https://onlinelibrary.w iley.com

/term s-and-conditions) on W

iley O nline L

ibrary for rules of use; O A

articles are governed by the applicable C reative C

om m

ons L icense

Broussard, C., & Northup, J. (1997). The use of functional analysis to develop peer interventions for

disruptive classroom behavior. School Psychology Quarterly, 12, 65–76.

Carr, E. G. (1977). The motivation of self-injurious behavior: A review of some hypotheses.

Psychological Bulletin, 34, 800–816.

Carr, E. G., & Durand, V. M. (1985). Reducing behavior problems through functional communication

training. Journal of Applied Behavior Analysis, 18, 111–126.

Carr, E. G., Yarbrough, S. C., & Langdon, N. A. (1997). Effects of idiosyncratic stimulus variables on

functional analysis outcomes. Journal of Applied Behavior Analysis, 30, 673–686.

Derby, M., Wacker, D. P., Peck, S., Sasso, G., DeRaad, A., Berg, W., Asmus, J., & Ulrich, S. (1994).

Functional analysis of separate topographies of aberrant behavior. Journal of Applied Behavior

Analysis, 27, 267–278.

Ellis, J., & Magee, S. K. (1999). Determination of environmental correlates of disruptive classroom

behavior: Integration of functional analyses into public school assessment process. Education and

Treatment of Children, 22, 291–316.

Frea, W. D., & Hughes, C. (1997). Functional analysis and treatment of social-communicative behavior

of adolescents with developmental disabilities. Journal of Applied Behavior Analysis, 30,

701–704.

Gunter, P. L., Jack, S. L., Shores, R. E., Carrell, D. E., & Flowers, J. (1993). Lag sequential analysis as a

tool for functional analysis of student disruptive behavior in classrooms. Journal of Emotional and

Behavioral Disorders, 1, 138–148.

Hanley, G. P., Iwata, B. A., & McCord, B. E. (2003). Functional analysis of problem behavior: A review.

Journal of Applied Behavior Analysis, 36, 147–185.

Horner, R. H. (1994). Functional assessment: Contributions and future directions. Journal of Applied

Behavior Analysis, 27, 401–404.

Iwata, B. A. (1996). Functional analysis assessment of self-injurious behavior (SIB) workshop

literature. Florida Center on Self-Injury, University of Florida, Gainesville, FL.

Iwata, B. A., Dorsey, M. F., Slifer, K. J., Bauman, K. E., & Richman, G. S. (1982/1994). Toward a

functional analysis of self-injury. Journal of Applied Behavior Analysis, 27, 197–209.

Johnston, J. M. (1993). A model for developing and evaluating behavioral technology. In R. Van

Houten & S. Axelrod (Eds.), Behavior analysis and treatment (pp. 323–343). New York: Plenum

Press.

Jones, K. M., Drew, H. A., & Weber, N. L. (2000). Noncontingent peer attention as a treatment for

disruptive classroom behavior. Journal of Applied Behavior Analysis, 33, 343–346.

Lohrmann-O’Rourke, S., & Yurman, B. (2001). Naturalistic assessment of and intervention for

mouthing behaviors influenced by establishing operations. Journal of Positive Behavioral

Interventions, 3, 19–27.

Mace, F. C. (1994). The significance and future of functional analysis methodologies. Journal of Applied

Behavior Analysis, 27, 385–392.

Mace, F. C., & Lalli, J. S. (1991). Linking descriptive and experimental analyses in the treatment of

bizarre speech. Journal of Applied Behavior Analysis, 24, 553–562.

Mace, F. C., Yankanich, M. A., & West, B. J. (1988/1989). Toward a methodology of experimental

analysis and treatment of aberrant classroom behaviors. Special Services in the Schools, 4,

71–88.

Magee, S. K., & Ellis, J. (2000). Extinction effects during the assessment of multiple problem behaviors.

Journal of Applied Behavior Analysis, 33, 313–316.

Magee, S. K., & Ellis, J. (2001). The detrimental effects of physical restraint as a consequence for

inappropriate classroom behavior. Journal of Applied Behavior Analysis, 34, 501–504.

226 J. Ellis and S. Magee

1099078x, 2004, 3, D ow

nloaded from https://onlinelibrary.w

iley.com /doi/10.1002/bin.161 by B

ehavior A nalyst C

ertification, W iley O

nline L ibrary on [14/11/2022]. See the T

erm s and C

onditions (https://onlinelibrary.w iley.com

/term s-and-conditions) on W

iley O nline L

ibrary for rules of use; O A

articles are governed by the applicable C reative C

om m

ons L icense

McComas, J., Hoch, H., Paone, D., & El-Roy, D. (2000). Escape behavior during academic tasks: A

preliminary analysis of idiosyncratic establishing operations. Journal of Applied Behavior Analysis,

33, 479–493.

Moore, J. W., Mueller, M. M., Dubard, M., Roberts, D. S., & Sterling-Turner, H. E. (2002). The

influence of therapist attention on self-injury during a tangible condition. Journal of Applied Behavior

Analysis, 35, 283–286.

Mueller, M. M., Wilczynski, S. M., Moore, J. W., Fusilier, I., & Trahant, D. (2001). Antecedent

manipulations in a tangible condition: Effects of stimulus preference on aggression. Journal of

Applied Behavior Analysis, 34, 237–240.

Northup, J., Broussard, C., Jones, K., George, T., Vollmer, T. R., & Herring, M. (1995). The differential

effects of teacher and peer attention on the disruptive classroom behavior of three children with a

diagnosis of attention deficit hyperactivity disorder. Journal of Applied Behavior Analysis, 28,

227–228.

Northup, J., Jones, K., Broussard, C., DiGiovanni, G., Herring, M., Fusilier, I., & Hanchey, A. (1997). A

preliminary analysis of interactive effects between common classroom contingencies and

methylphenidate. Journal of Applied Behavior Analysis, 30, 121–125.

Northup, J., Wacker, D. P., Berg, W. K., Kelly, L., Sasso, G., & DeRaad, A. (1994). The treatment of

severe behavior problems in school settings using a technical assistance model. Journal of Applied

Behavior Analysis, 27, 33–48.

O’Reilly, M. F., & Carey, Y. (1996). A preliminary analysis of the effects of prior classroom conditions

on performance under analogue analysis conditions. Journal of Applied Behavior Analysis, 29,

581–584.

Repp, A. (1994). Comments on FA procedures for school-based behavior problems. Journal of Applied

Behavior Analysis, 27, 409–411.

Romaniuk, C., Miltenberger, R., Conyers, C., Jenner, N., Jurgens, M., & Ringenberg, C. (2002). The

influence of activity choice on problem behaviors maintained by escape versus attention. Journal of

Applied Behavior Analysis, 35, 349–362.

Sasso, G. M., Reimers, T. M., Cooper, L. J., Wacker, D., Berg, W., Steege, M., Kelly, L., & Allaire, A.

(1992). Use of descriptive and experimental analysis to identify the functional properties of aberrant

behavior in school settings. Journal of Applied Behavior Analysis, 25, 809–821.

Scruggs, T. E., & Mastropieri, M. A. (1996). Teacher perceptions of mainstreaming/ inclusion, 1958–

1995: A research synthesis. Exceptional Children, 63, 59–74.

Sugai, G., Horner, R. H., & Sprague, J. R. (1999). Functional-assessment-based behavioral support

planning: Research to practice to research. Behavioral Disorders, 24, 253–257.

Taylor, J. C., & Romanczyk, R. G. (1994). Generating hypotheses about the function of student

problem behavior by observing teacher behavior. Journal of Applied Behavior Analysis, 27,

251–265.

Thomas, D. R., Becker, W. C., & Armstrong, M. (1968). Production and elimination of disruptive

classroom behavior by systematically varying teacher’s behavior. Journal of Applied Behavior

Analysis, 1, 35–45.

Van Camp, C. M., Lerman, D. C., Kelley, M. E., Contrucci, S. A., & Vorndran, C. M. (2000a). Variable-

time reinforcement schedules in the treatment of socially maintained problem behavior. Journal of

Applied Behavior Analysis, 33, 545–557.

Van Camp, C. M., Lerman, D. C., Kelley, M. E., Roane, H. S., Contrucci, S. A., & Vorndran, C. M.

(2000b). Further analysis of idiosyncratic antecedent influences during the assessment and treatment

of problem behavior. Journal of Applied Behavior Analysis, 33, 207–221.

Modifications to basic functional analysis procedures 227

1099078x, 2004, 3, D ow

nloaded from https://onlinelibrary.w

iley.com /doi/10.1002/bin.161 by B

ehavior A nalyst C

ertification, W iley O

nline L ibrary on [14/11/2022]. See the T

erm s and C

onditions (https://onlinelibrary.w iley.com

/term s-and-conditions) on W

iley O nline L

ibrary for rules of use; O A

articles are governed by the applicable C reative C

om m

ons L icense

Vollmer, T. R., Marcus, B. A., Ringdahl, J. E., & Roane, H. S. (1995). Progressing from brief

assessments to extended experimental analyses in the evaluation of aberrant behavior. Journal of

Applied Behavior Analysis, 28, 561–576.

Wallace, M. D., & Knights, D. J. (2003). An evaluation of a brief functional analysis format within a

vocational setting. Journal of Applied Behavior Analysis, 36, 125–128.

Walker, M. H., & Sprague, J. R. (1999). Longitudinal research and functional analysis issues.

Behavioral Disorders, 24, 335–337.

Zimmerman, E. H., & Zimmerman, J. (1962). The alternation of behavior in a special classroom

situation. Journal of the Experimental Analysis of Behavior, 5, 59–60.

228 J. Ellis and S. Magee

1099078x, 2004, 3, D ow

nloaded from https://onlinelibrary.w

iley.com /doi/10.1002/bin.161 by B

ehavior A nalyst C

ertification, W iley O

nline L ibrary on [14/11/2022]. See the T

erm s and C

onditions (https://onlinelibrary.w iley.com

/term s-and-conditions) on W

iley O nline L

ibrary for rules of use; O A

articles are governed by the applicable C reative C

om m

ons L icense