The treatment of self-injurious escape behavior using instructional fading with and without extinction


Material Information

The treatment of self-injurious escape behavior using instructional fading with and without extinction
Physical Description:
vi, 33 leaves : ill. ; 29 cm.
Zarcone, Jennifer R
Publication Date:


Subjects / Keywords:
Self-injurious behavior   ( lcsh )
Psychology thesis Ph. D
Dissertations, Academic -- Psychology -- UF
bibliography   ( marcgt )
non-fiction   ( marcgt )


Thesis (Ph. D.)--University of Florida, 1993.
Includes bibliographical references (leaves 29-32).
Statement of Responsibility:
by Jennifer R. Zarcone.
General Note:
General Note:

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 030387993
oclc - 31054348
System ID:

Full Text









I wish to thank those individuals who helped make this research possible, especially

Dr. Brian Iwata, my committee chair and major professor, for all his guidance throughout

this project and throughout graduate school. Appreciation is extended to the other members

of my dissertation committee, Dr. Marc Branch, Dr. Henry Pennypacker, Dr. William

Wolking, and Dr. Merle Meyer, for their assistance in varying capacities as this project


Additional thanks are extended to my fellow graduate students and friends, Beth

Duncan, Han Goh, Dorothea Lerman, Jodi Mazaleski, Bridget Shore, Rick Smith, and Tim

Vollmer, whose assistance in the conceptualization and implementation of this project was

greatly appreciated.

I also want to thank my family, including my mother and father for all their support.

with very special thanks to my husband Troy Zarcone, whose love and support was

limitless throughout graduate school. Finally, thank you Nicholas for all the joy you bring

to my life every day.


ACKNOW LEDGMENTS......................................................................... H

ABSTRACT......................................................................................... v


1 INTRODUCTION......................................................................... 1

2 METHOD................................................................................... 9
Subjects and Setting................................................................... 9
Response Measurement.............................................................11
Experimental Sequence and Design................................................ 11
Functional Analysis Conditions.................................................... 13
Treatment Conditions................................................................ 14
3 RESULTS................................................................................. 16
Functional Analysis Assessment................................................ 16
Treatment Evaluation................................................................20

4 DISCUSSION............................................................................ 24
REFERENCES.................................................................................... 30

BIOGRAPHICAL SKETCH................................................................ 34


Figure PE

1 Responses per minute of SIB across assessment conditions for Kate ............. 17

2 Responses per minute of SIB across assessment conditions for Jack............ 18

3 Responses per minute of SIB across assessment conditions for Jesse........ 19
4 Responses per minute of SIB (left axis) and instructional trials per
minute (right axis) during baseline and treatment. Arrows indicate
when the extinction contingency was added to treatment.......................... 21

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy



Jennifer R. Zarcone

December, 1993

Chairman: Brian A. Iwata
Major Department: Department of Psychology

The self-injurious behavior (SIB) of three developmentally disabled individuals

increased when a brief timeout was provided contingent on the behavior, indicating that

their SIB was maintained by negative reinforcement in the form of escape from

instructions. Based on these results, instructional fading was implemented as treatment for

all three subjects. The procedure involved initially eliminating all instructions from

sessions and subsequently increasing the rate of instructions gradually over time. During

baseline, instructions were presented at the rate of 2 per minute, and escape was provided

contingent on SIB. Treatment was implemented in a multiple baseline across subjects

design and began with the rate of instructions at 0 per minute. If SIB remained below 0.5

per minute for two consecutive sessions, another instruction was added the following

session. This continued until the baseline rate of instructions was reached. If SIB

remained high for 10 consecutive sessions, indicating a lack of success with the

instructional fading procedure, escape extinction was implemented. The extinction

procedure involved preventing escape by providing physical guidance through the task.

Treatment was completed when the rate of instructions was the same as in baseline, and

SIB remained below 0.5 per minute for two consecutive sessions.

Results showed that instructional fading without extinction was effective in initially

reducing SIB, but decreases in SIB were not maintained. When the extinction component

was added to the fading procedure, SIB was quickly suppressed for one subject but not for

the other two. All three subjects required over 150 sessions to meet the treatment

termination criteria. Subjects' history of intermittent escape may have been responsible for

the prolonged course of treatment.


Self-injurious behavior (SIB) has been defined as any behavior producing physical

injury to an individual's own body (Tate & Baroff, 1966) and has been estimated to occur

in 10% to 15% of the population residing in state facilities for the developmentally disabled

(Griffin, Williams, Stark, Altmeyer, & Mason, 1984; Schroeder, Schroeder, Smith, &

Dalldorf, 1978). Behavior problems such as SIB have often been cited as a factor

contributing to institutionalization (Eyman, Borthwick, & Miller, 1981), and appear to be

more prevalent among younger, more severely retarded residents who have been

institutionalized longer (Schroeder et al., 1978). Common topographies of SIB include

biting, head hitting, head banging, and scratching; however, eye gouging, hand mouthing,

ruminating, hair pulling, and pica have also been observed in this population (Griffin et al.,


Although some forms of SIB may be associated with medical conditions or

physiological disorders (Cataldo & Harris, 1982), there is evidence that SIB is often a

learned response maintained by specific reinforcement contingencies (Carr, 1977), which

may be identified using a variety of assessment methods (see Iwata, Vollmer, & Zarcone,

1990, for a review). The most accurate method is to directly observe the occurrence of the

behavior under experimental conditions. For example, Iwata, Dorsey, Siffer, Bauman, and

Richman (1982) exposed subjects to a series of conditions designed to assess the effects of

positive reinforcement, negative reinforcement, and automatic reinforcement on the

frequency of SIB. Results showed that for six of nine subjects, SIB was noticeably higher

in one specific condition, which differed across individuals. These results indicated that

there was a functional relationship between the reinforcement contingency manipulated and

SIB. Once the behavioral function has been identified, treatment procedures can be

developed based on directly manipulating the reinforcement contingency.

Several reinforcement contingencies can be identified through a functional analysis.

For example, SIB may be maintained by automatic or sensory consequences directly

produced by the behavior (Lovaas, Newsom, & Hickman, 1987; Skinner, 1969).

Although there is ample evidence that sensory and perceptual events can function as

reinforcement, there is very little empirical evidence showing the acquisition and

maintenance of problem behavior by sensory consequences. This is primarily due to the

difficulty in achieving external experimental control over the delivery of perceptual

reinforcers (Lovaas et al., 1987). Self-injury also can be maintained by the social

environment through positive or negative reinforcement. For many individuals, positive

reinforcement in the form of social interaction or attention can serve as reinforcement for

SIB. For example, Lovaas, Freitag, Gold, and Kassorla (1965) found that providing

attention (e.g., saying "I don't think you are bad") contingent upon head and arm banging

increased the frequency of a child's SIB. Finally, SIB can be maintained by negative

reinforcement in the form of contingent removal, reduction, or postponement of stimulation

(Hineline, 1977). Several studies have shown that SIB can be maintained by negative

reinforcement in the form of escape from or avoidance of academic or work tasks (e.g.,

Iwata et al., 1982; Steege, Wacker, Berg, Cigrand, & Cooper, 1989). For example,

Steege et al. (1989) presented academic tasks to a subject and prompted him to comply. If

the subject engaged in SIB, a 20-s timeout from the task was provided. This procedure

resulted in an increase in SIB, indicating that a break from instructions served as negative


Behavior that has been acquired and maintained by escape requires the development of

treatment approaches based on modifying the negative reinforcement contingency.

Behavior maintained by escape differs significantly from behavior maintained by positive

reinforcement because usually there is a discrete, antecedent event preceding the occurrence

of the escape behavior (i.e., the presentation of aversive stimulation), as well as an

observable consequence following the behavior (i.e., the termination of the stimulation).

As a result, several treatment approaches have been developed that are based on modifying

either the antecedents or consequences of escape.

In the first approach, antecedent events can be manipulated to decrease the likelihood

that the escape behavior will occur. These events often function as establishing operations,

which create the conditions that make escape reinforcing (Michael, 1982). One way the

establishing operations of escape can be modified is through a process known as stimulus

fading. Procedurally, stimulus fading involves decreasing escape behavior by modifying

or eliminating the aversive events that evoke the behavior. Once the behavior has been

eliminated or reduced, the aversive event can be gradually faded back to its initial level.

Stimulus fading has been used in a few studies with individuals who exhibit behavior

maintained by escape from instructions. In a study by Weeks and Gaylord-Ross (1981),

tasks were faded from easy to difficult across sessions and resulted in a decrease in SIB.

However, decreasing trends across both baseline and treatment may indicate that an

additional component to fading was in effect. Heidom and Jensen (1984) also used a

fading procedure that was based on modifying the number of tasks presented per session.

If SIB decreased and remained low, the number of instructions presented per session was

gradually increased. However, there were several additional components used during

treatment; thus, reductions in SIB may have been attributed to any one of several


In both of the above studies, modifying events antecedent to SIB resulted in immediate

decreases in SIB. This effect can be particularly advantageous with individuals who

exhibit severe forms of inappropriate behavior. It is important, however, that the

antecedent events occasioning SIB are not simply eliminated from the individual's daily

activity schedule. For individuals participating in habilitative programs, eliminating or

greatly reducing instructional activities may result in a lack of progress toward long-term

educational goals. By using a procedure such as fading, however, instructional activities

can be eventually returned to their former level.

The second treatment approach to reduce escape behavior involves modifying its

consequences. Research has shown that the treatment of behavior problems can be

facilitated when the reinforcement contingency maintaining the behavior is eliminated

through the process of extinction (e.g., Lovaas & Simmons, 1969; Mace, Page, Ivancic, &

O'Brien, 1986). For behavior maintained by negative reinforcement, "escape" extinction

involves two parts. First, aversive stimulation must be present to establish escape as

reinforcement. Second, escape from the aversive stimulation must be prevented (Iwata,


Several studies have shown that escape extinction is effective in eliminating SIB

maintained by escape from academic or physical tasks. For example, Repp, Felce, and

Barton (1988) showed that presenting tasks at the rate of one per minute and continuing to

verbally and physically prompt subjects to comply with the task resulted in a reduction in

SIB and stereotypy maintained by negative reinforcement. In studies by Homer, Day,

Sprague, O'Brien, and Heathfield (1991) and Steege et al. (1989), extinction in the form of

the continued presentation or redirection to a task was used to reduce escape-maintained

SIB. Other studies have implemented extinction by providing physical guidance through a

task contingent on SIB. For example, Iwata, Pace, Cowdery, Kalsher, and Cataldo (1990)

presented seven subjects with academic or medically related tasks. If SIB occurred during

a task, the subject was guided to complete the task. Finally, Sailor, Guess, Rutherford,

and Baer (1968), developed a hierarchy of tasks based on difficulty and presented difficult

tasks after each occurrence of tantrum behavior. In both the Iwata et al. and the Sailor et al.

studies, however, extinction may not have been the only process involved. For some

individuals, the contingent presentation of physical guidance or difficult tasks may have

functioned as punishment of the target behavior.

Although escape extinction has been shown to be an effective means of eliminating

escape behavior, several disadvantages are associated with its use. First, the procedure

may occasion an increase or "burst" in responding when treatment is initially implemented

(Iwata, Pace, Cowdery, Kalsher, & Cataldo, 1990; Repp et al. 1988). This can make

extinction procedures for SIB or aggression difficult from a practical standpoint because an

increase in the behavior can be dangerous to either the individual or to those in his or her

environment. Second, extinction may result in an increase in the intensity or severity of the

escape behavior. Third, other inappropriate escape behaviors may increase (e.g.,

aggression, destruction) when SIB is extinguished. For example, Goh and Iwata (in

press) showed that when extinction was implemented with SIB, not only did SIB initially

increase, but the frequency of aggression increased also. Finally, from a practical

standpoint, extinction may be difficult to implement with some individuals. For example, a

therapist may not be able to prevent escape when the subject is stronger than the therapist.


Thus, the therapeutic use of escape extinction may be practical only under certain


Recent studies have examined the separate and combined effects of stimulus fading and

extinction as treatment for escape behavior. Pace, Iwata, Cowdery, Andree, and Mclntyre

(1993) faded the frequency of task presentation using instructional fading. The procedure

involved initially removing all instructions from treatment sessions, and if SIB remained

low, increasing the rate of instructions over sessions until the original baseline instructional

rate was reached. In addition to the fading procedure, escape extinction was implemented

when SIB occurred during an instructional trial (i.e., the subject was physically guided

through the task). Results showed immediate reductions in SIB (i.e., there was no

extinction burst) that were maintained throughout treatment. It is not clear, however,

which component of treatment reduced behavior because the separate effects of extinction

and fading were not completely assessed. Although extinction alone was briefly examined

with one subject, the condition was discontinued after 16 sessions because SIB did not

decrease. It is possible, however, that if the extinction condition was continued, SIB may

have decreased within fewer sessions than when extinction was combined with fading.

Zarcone et al. (in press) extended the Pace et al. study by comparing the therapeutic

effects of escape extinction with and without instructional fading. Results showed that for

two subjects the end of treatment criterion was reached in fewer sessions when extinction

was implemented alone than when extinction was combined with fading. Specifically,

treatment was completed for one subject within 9 sessions with extinction alone; treatment

was completed in the extinction plus fading condition within 34 sessions, even when the

rate of fading was accelerated. The second subject completed treatment with extinction

alone in 21 sessions and extinction plus fading in 54 sessions. Although treatment was

completed within fewer sessions with extinction, there was a burst in responding at the

outset of treatment with extinction, which was not observed when extinction was combined

with fading.

Although previous research has shown that both extinction and extinction plus fading

are effective treatment procedures for escape-maintained SIB (e.g., Iwata, Pace, Cowdery,

Kalsher, & Cataldo, 1990; Pace et al., 1993; Zarcone et al., in press), the effects of fading

without extinction have yet to be clearly assessed. Presumably, this antecedent-only

approach would initially eliminate SIB at the outset of treatment, but without extinction in

effect any behavior that occurs would necessarily be reinforced with escape. Thus, there

are three possible therapeutic outcomes if fading were implemented alone. First, the

procedure may be effective in reducing SIB quickly and maintaining it at a low level.

Second, the procedure may be ineffective in reducing SIB because the behavior does not

decrease. Finally, fading may initially be effective in decreasing SIB, but as the rate of

instructions increases, the probability that SIB will occur will also increase due to

continued negative reinforcement.

The purpose of the present study was to identify individuals who exhibited SIB

maintained by escape from or avoidance of instructions. This was accomplished by

conducting a functional analysis of each subject's SIB. Those individuals identified with

escape-maintained SIB participated in an assessment of the effects of instructional fading as

the sole intervention in the treatment of SIB.


Subjects and Setting

Three individuals residing in a state facility for the developmentally disabled

participated. All three individuals were diagnosed with profound mental retardation and

were chosen to participate in this study because they exhibited SIB at a high frequency. In

addition, their SIB was maintained by escape from instructions (see functional analysis

conditions below).

Kate was an ambulatory 27-year-old woman whose SIB consisted of hitting her face

with her palm or fist. She also exhibited a number of self-stimulatory behaviors such as

body rocking, turning in circles, and pushing chairs in circles. Her sensory impairments

included blindness from cataracts and profound hearing loss. She had limited self-care and

feeding skills and was not toilet trained. Kate could follow a few simple instructions, but

had no expressive language skills. Her SIB significantly interfered with the training she

received at her residence and worksite. During the course of the study she did not receive

any psychotropic medication.

Jack was an ambulatory 38-year-old male whose SIB consisted of banging his head

against hard surfaces and biting his hand. He did not have any sensory impairments. Jack

wore a helmet throughout the day due to seizures and poor balance, which placed him at

risk for falling. His self-care skills were limited due to lack of motor skills; he required

assistance in bathing and dressing, but he could feed himself and signal when he needed to

be assisted to the bathroom. Although Jack could follow a number of one-step

instructions, his expressive language consisted of a single word ("No"), which he would

utter when demands were placed on him. Staff on Jack's residence noted that his SIB was

most severe when he was required to ambulate and that it interfered with daily training and

leisure activities. Throughout the study, Jack received low doses of Dilantin to control


Jesse was an ambulatory 40-year-old male whose SIB consisted of slapping the side of

his face, banging his head against hard surfaces, punching his sides, and hitting his hands

and arms against hard surfaces. Jesse exhibited additional behavior problems such as

aggression and spitting. He was toilet trained, able to feed himself, and perform most self-

care skills, except toothbrushing and bathing, without physical assistance. Jesse had no

sensory impairments. Although his expressive language skills consisted of only a few

echolaliac words, his receptive language skills were very good; he could follow many

complex instructions. Jesse often assisted staff on his home cottage in activities such as

mopping the floor, setting the table, and cleaning food trays; however, staff reported that if

he was working on a task and became "upset," they would give him a break from the

activity. During the course of the study, he received several medications: Dilantin for

seizures, and Sinequan (an antidepressant) and Mellaril (an antipsychotic) for behavior

problems. The dosages for each medication remained constant throughout the study.

The study was conducted on the grounds of the facility at a day program for the

assessment and treatment of SIB. Three to five daily sessions were conducted with each

subject, four to five days a week. Sessions lasted for 15 minutes, separated by 10- to 15-

minute breaks. Therapy rooms were either 3.7 by 5.7 m or 6.8 by 12.5 m and contained

chairs and appropriate materials for each session. Jack did not wear his protective helmet

during assessment or treatment sessions; however, the wall directly behind his chair was

padded. Sessions were terminated and all responses blocked if subjects engaged in any

behavior producing an open injury. An experimenter, and when necessary an assistant,

and one or two data collectors were present during all sessions.

Response Measurement and Reliability

Self-injurious responses were defined as follows: hitting (Kate and Jesse)-contact of

any part of the hand or arm with any part of the head, body, or any hard surface such as a

wall or table; head banging (Jack and Jesse)-contact of any part of the head with a hard

surface; hand biting (Jack)--closure of the teeth on any part of the skin from fingertips to

wrist. Data were also collected on the number of instructions presented by the

experimenter. During each session, an observer recorded data on a hand-held computer

(Assistant, Model AST 102) during continuous 10-s intervals. Session data were

converted to responses per minute for SIB and instructions.

Interobserver agreement was assessed by having a second observer simultaneously but

independently collect data during 59.9% of all assessment sessions, 57.4% of all baseline

sessions, and 43.1% of all treatment sessions. Agreement percentages were calculated

based on an interval-by-interval comparison of observers' records, in which the smaller

number of responses in each interval was divided by the larger number of responses.

These fractions were then summed across all intervals, divided by the total number of

intervals in the session, and multiplied by 100 to get the percentage agreement between the

two observers. Mean agreement percentages for SIB during assessment, baseline, and

treatment for all three subjects are listed in Table 1.

Experimental Sequence and Designs

The first phase of the study consisted of a functional analysis baseline, in which a

series of conditions was presented in a multielement format (i.e., conditions were presented

Table 1
Percentage of Assessment, Baseline, and Treatment Sessions with Interobserver
Agreement and Mean Agreement Percentages for SIB

Percentage of Sessions with Interobserver Agreement

Assessment Baseline Treatment

Kate 66.7 50.0 35.8

Jack 66.7 55.5 62.5

Jesse 46.2 66.7 31.1

Mean Agreement Percentages (range)

Kate 99.3 (97.7-100) 93.2 (92.6-93.7) 96.6 (83.9-100)

Jack 97.5 (94.0-100) 96.5 (89.8-99.4) 99.2 (95.0-100)

Jesse 98.2 (88.9-100) 96.6 (91.3-98.8) 97.5 (88.0-100)

each day in random order) (Sidman, 1960; Ulman & Sulzer-Azaroff, 1975). The purpose

of this analysis was to identify the variables maintaining each subject's SIB. Baseline

conditions were then initiated with all three subjects. Treatment was introduced in a

multiple baseline across subjects design.

Functional Analysis Conditions

Subjects were exposed to four assessment conditions based on those described by

Iwata et al. (1982).


This condition was designed to assess the effects of positive reinforcement (i.e., adult

attention) on the rate of SIB. The subject was placed in a therapy room with a variety of

toys and materials available. At the beginning of each session, the experimenter entered

the room and told the subject: "I will be here if you need me." The experimenter then sat

away from the subject and did not make any eye contact. Contingent on SIB, the

experimenter immediately approached the subject and delivered attention in the form of

concern and disapproval of the behavior (e.g., "Stop that, you'll hurt yourself."), and

physical contact (e.g., touching the subject's shoulder). All other behaviors were ignored

by the experimenter.


This condition was designed to assess the effects of negative reinforcement (i.e.,

escape from or avoidance of instructions) on the rate of SIB. The experimenter either sat

across the table from or stood in front of the subject and presented tasks on a fixed time 30-

s schedule (FT 30-s). The tasks used in this and all subsequent instructional conditions

were similar to those found in the subject's educational or habilitation plan. If the subject

did not comply with the instruction, the experimenter demonstrated the correct response

and physically guided the response if necessary. Praise, clapping, and pats on the back

were consequences for compliance. If the subject exhibited SIB during any part of the

instructional sequence, the experimenter terminated the task (i.e., removed the materials

and turned/walked away). The next instruction was presented according to schedule.


This condition was designed to assess the effects of automatic or sensory reinforcement

on the rate of SIB. The therapy room was arranged to approximate a "barren" environment

without any external forms of reinforcemetit, such as toys, materials, or attending adults.

The subject was placed in the therapy room alone, with only an observer present to collect

data. There was no interaction with the subject during these sessions.

This condition served as a control for the other three conditions. In this condition, the

therapy room was made to simulate an "enriched" environment. During sessions, the

experimenter provided attention to the subject, and play materials were made available. The

experimenter did not give any instructions during this condition. At the beginning of each

session, the experimenter entered the room and directed the subject to the materials. The

experimenter attended to the subject on a FT 30-s schedule. The attention consisted of

either presenting an item to the subject or providing verbal interaction (e.g., "you look nice

today"). Any SIB that occurred during this session was ignored by the experimenter.

Treatment Conditions

Procedures in effect during baseline were identical to those in the Demand condition:

the experimenter presented tasks, provided social praise contingent on compliance, and

terminated the trial contingent on the occurrence of SIB.

Tasks chosen to be used during baseline and treatment were associated with high rates

of SIB during the functional analysis. Specific tasks varied from subject to subject, but

generally they included gross motor and self-care tasks, such as "stand up," "sweep the

floor," and "comb your hair."

Instructional Fading

This procedure consisted of eliminating all instructions from the initial session of

treatment and gradually fading the instructions back into sessions over time. The number

of instructions per session was increased by one if the subject's SIB was at or below 0.5

responses per minute for two consecutive sessions. Consequences for SIB were the same

as during baseline; that is, SIB produced termination of the trial. Treatment was completed

when the rate of instructions was the same as in baseline (i.e., 2 instructions per minute or

30 instructions per session) and the subject's rate of SIB remained below 0.5 per minute

for two consecutive sessions.

Instructional Fading plus Escape Extinction

If the subject's rate of SIB did not decrease in the fading condition (i.e., after ten

consecutive sessions without meeting the criterion of 2 consecutive sessions with SIB

below 0.5 responses per minute) an escape extinction component was added to treatment.

Thus, contingent on the occurrence of SIB during a trial, the subject was not allowed to

escape (i.e., he or she was physically guided through the task). The extinction component

was withdrawn when subjects had two consecutive sessions with SIB at a rate below 0.5

responses per minute.


Functional Analysis Assessment

Assessment results are shown in Figures 1, 2, and 3. Figure 1 shows the results of

Kate's functional analysis. Highest rates of SIB occurred in the Demand condition (M =

3.3 responses per minute). Kate exhibited very little SIB in the other three conditions

(mean rate for Attention, Alone, and Play conditions combined = 0.1 responses per

minute). Kate's functional analysis was the briefest of all three subjects; it lasted 9

sessions. Results indicated that Kate's SIB was clearly maintained by negative

reinforcement in the form of escape from instructions.

Figure 2 shows the results of Jack's functional analysis, which were similar to those

obtained with Kate. The highest rate of SIB occurred in the Demand condition (M = 3.1

responses per minute) and very little SIB occurred in the other three conditions (M = 0.1

responses per minute). Jeff's functional analysis also indicated that his SIB was

maintained by escape from instructions.

Figure 3 shows the results of Jesse's functional analysis. His analysis was less clear

than Kate's and Jack's for two reasons. First, some SIB occurred in at least one session of

each condition, increasing the overall mean in the Attention, Alone, and Play conditions

(combined mean = 0.92 responses per minute). Second, there was a low rate of SIB in

one Demand session. Overall, however, the highest rates of SIB occurred in the Demand

condition (M = 2.2 responses per minute), indicating that his SIB was primarily maintained

by escape from instructions.


6- Kate a Alone

--- Demand
Z 4- Play

w \
0 2

0 r
24 6 8 10


Figure 1. Responses per minute of SIB across assessment conditions
for Kate.

lac Alone

.,, Demand
g Attn
0 Play

0 1 G 1i
2 4 6 8 10 12


Figure 2. Responses per minute of SIB across assessment conditions
for Jack.

-- Demand
--- Attn
--o- Play

2 4 6 8 10 12 14


Figure 3. Responses per minute of SIB across assessment conditions
for Jesse.

Treatment Evaluation

Figure 4 shows the results of baseline and treatment for all three subjects. Kate's data

(upper panel) show a high rate of SIB during baseline (M = 6.4 responses per minute).

When instructional fading was introduced, Kate's SIB decreased immediately to zero. As

the rate of instructions increased, her rate of SIB increased and became more variable. At

the rate of 0.5 instructions per minute (7 per session) there were 10 consecutive sessions

with a high rate if SIB. Extinction was then added to the fading procedure. After 17

sessions with fading plus extinction, Kate had two consecutive sessions with SIB below

0.5 responses per minute; thus meeting the criterion to withdraw extinction and increase the

instructional rate. Her rate of SIB under the fading procedure increased again at 11, 17,

and 20 instructions per session, resulting in the addition of extinction a total of four times.

Each time extinction was introduced, SIB decreased, and the extinction component was

withdrawn. Kate completed treatment after 194 sessions.

Jack's data (middle panel) show a variable rate of SIB during baseline. His mean rate

of SIB during baseline was 2.3 responses per minute, which nearly matched the rate of

instructions (2 per minute). When treatment was implemented, Jack's SIB immediately

decreased to zero for several sessions, and the rate of instructions was increased rapidly.

At the rate of 1.33 instructions per minute (20 per session), he had a high rate of SIB for

10 consecutive sessions, and extinction was added to the fading procedure. Jack's SIB

decreased below 0.5 per minute immediately; thus, after two sessions with extinction in

effect, the component was withdrawn and the instructional rate increased by one.

Extinction was introduced again at 1.8 instructions per minute (27 per session) and at

Figure 4. Responses per minute of SIB (left axis) and instructional trials per minute (right
axis) during baseline and treatment for Kate, Jack, and Jesse. Arrows indicate when the
extinction component was added to treatment.


I fig


" SIB (Fading)
" SIB (Fading+Extinction)
* Instructions

0.0 W
2.0 D

1.0 U
0.5 I'


2.0 -



50 100 150 200


2.0 instructions per minute (30 per session). Each time extinction was introduced, his SIB

quickly decreased. Treatment was completed after 129 sessions.

Jesse's data (lower panel) during baseline showed a fairly stable rate of SIB across 11

sessions. His mean rate of SIB during baseline was 2.6 responses per minute. The fading

procedure appeared to be very effective in reducing Jesse's SIB at the outset of treatment;

his rate of SIB remained very low, and instructions increased rapidly. At the instruction

rate of 1.53 per minute (23 per session), Jesse's SIB increased noticeably and extinction

was added to the fading procedure. His SIB continued to remain high with extinction in

effect for 10 sessions, at which point it decreased for two consecutive sessions. When

extinction was withdrawn and the instruction rate again increased, Jesse's SIB increased

immediately. In fact, from this point on during treatment, each time the instruction rate

increased, his SIB increased, requiring the reintroduction of extinction. Thus, it appeared

that fading alone was not effective. When extinction was implemented the fourth time, it

remained in effect until Jesse completed treatment. Even with extinction in effect,

however, there were 56 additional sessions before treatment termination criteria were met.

Treatment was finally completed for Jesse after 190 sessions.


Results of the present study are consistent with previous research indicating that

reducing the frequency of instructions produces rapid reductions in SIB at the outset of

treatment (Pace et al., 1993; Zarcone et al., in press). However, unlike the results of

previous studies, reductions in SIB were not maintained as treatment progressed. In the

present study, the increase in instructions increased the probability that SIB would occur,

and without extinction in effect, SIB continued to be reinforced. These results show that

without extinction, fading alone was ineffective in maintaining low rates of SIB.

Results of the present study also differed from previous research in terms of the length

of treatment. In the Pace et al. (1992) study, treatment lasted approximately 100, 80, and

60 sessions when fading was used with extinction. However, no specific criteria were

used to fade the frequency of instructions or to terminate treatment; thus, it is possible that

treatment could have been completed within even fewer sessions. For the two subjects

who completed the fading plus extinction condition in the Zarcone et al. (in press) study,

the treatment termination criterion was met within approximately 30 50 sessions. In the

present study, it took many more sessions for therapeutic effects to be obtained; over 140

sessions for Jack, and almost 200 sessions for Kate and Jesse. Although the fading criteria

were somewhat more conservative in the present study than in Zarcone et al., the length of

treatment doubled or even tripled that of the other studies. The primary difference in the

present study that may have prolonged the course of treatment was the absence of a

continuous extinction component.

When extinction was added to treatment, the behavior reducing effects varied across

subjects. Jack's SIB decreased immediately (i.e., within two sessions) the first two times

extinction was implemented and within 8 sessions the third time it was implemented,

indicating that extinction was effective in reducing his behavior quickly. For Kate, fewer

sessions with extinction were required across successive applications. Specifically, SIB

decreased within 17 sessions during the first application, and within 12, 7, and 5 sessions

during the second, third, and fourth applications, respectively. These results indicate that

for Kate, extinction may have become more effective in suppressing SIB each time it was

introduced even though the instructional rate had increased between applications. These

data are consistent with other research indicating that with each exposure to extinction

following reinforcement, behavior may decrease more quickly (Clark, 1964; Clark &

Taylor, 1960). Finally, the rapidity of reduction in Jesse's SIB when extinction was

implemented varied across successive applications. When extinction was added to the

fading procedure, Jesse's SIB decreased within 12 sessions, 4 sessions, 7 sessions, 22

sessions, 23 sessions, and 11 sessions. Although previous research has indicated that

extinction has fairly rapid behavior reducing effects (e.g., Iwata, Pace, Cowdery, Kalsher,

& Cataldo, 1990; Zarcone et al., in press), extinction was in effect throughout treatment in

both studies. In the present study, the effects of an intermittent history of escape available

and unavailable may have resulted in the variation in treatment effects each time Jesse was

exposed to extinction.

Although it appears that instructional fading alone may not be a desirable form of

treatment, there are several advantages to implementing fading in conjunction with

extinction. First, behavior is reduced immediately at the outset of treatment. For people

who engage in severe topographies of SIB, a rapid reduction in behavior may be necessary

to reduce the risk of injury to the individual. In addition, side effects associated with the

use of extinction alone (e.g., an extinction burst) may be eliminated.

There is one significant disadvantage when using fading or escape extinction: the

procedures reduce SIB by eliminating reinforcement or the events that establish escape as

reinforcement without providing reinforcement (positive or negative) for alternative

behaviors. Research on establishing alternative escape responses using procedures such as

"functional communication training" has indicated that teaching individuals with SIB to

request assistance or a break from training is an effective means of reducing SIB (e.g.,

Carr & Durand, 1985; Steege et al., 1990; Wacker et al., 1990). In addition, positive

reinforcement (e.g., praise) has been used to establish or maintain alternative behaviors

such as compliance. Recent studies manipulating the sequence of instructions have

indicated that compliant behavior can be increased by presenting a series of instructions for

which there is a high-probability of compliance immediately before the presentation of an

instruction for which there is a low probability of compliance (Davis, Brady, Williams, &

Hamilton, 1992; Mace et al., 1988; Homer et al., 1991). This procedure has been shown

to increase compliance, and when used in conjunction with extinction may also decrease

inappropriate escape behavior (e.g., Zarcone, Iwata, Hughes, & Vollmer, 1993).

Future research using fading procedures may focus on developing methods for

identifying the aversive events) establishing escape as reinforcement and developing

treatment approaches based on fading some dimension of the event. In addition to the

frequency of task presentation, several other dimensions of an instructional context can be

used as the basis for fading. For example, several studies have described assessment

conditions comparing easy versus difficult tasks and found that SIB was higher when

difficult tasks were presented (e.g., Mace et al., 1986; Weeks & Gaylord-Ross, 1981).

The difficulty of an instruction could be manipulated in several ways such as the duration

or the complexity of the task. In addition, tasks can be identified that have either a high-

probability or a low-probability of compliance (e.g., Mace et al., 1988) and fading could

occur along the dimension of probability of compliance. Finally, tasks could be faded

across multiple dimensions (e.g., difficulty and frequency). Instructional contexts have

several additional components that may occasion escape behavior including the work

materials, therapist, prompting sequence, or physical guidance. Although fading along

these dimensions may be somewhat more complex, it is possible to fade in the presence of

the work materials or therapist, or the frequency of prompts or physical guidance.

Finally, other sources of aversive stimulation beside instructional situations may be

amenable to fading, such as crowded or noisy conditions. McAfee (1987) showed that

increasing the density of children in a classroom resulted in an increase in aggressive

behavior with two subjects. Although environmental manipulations can limit the frequency

with which these events occur (e.g., minimizing ambient noise and reducing the size of

client groups), it may be beneficial to implement fading procedures to aid in eliminating

inappropriate behavior under these circumstances. Establishing appropriate forms of

escape behavior when the above sources of stimulation are present would also be


Stimulus fading procedures have been used as a basis for treating clinical phobias.

However, treatment usually involves the reduction of unobservable responses (e.g.,

fearfulness, anxiety) that are elicited by an aversive stimulus. These behaviors are often

measured using verbal reports of levels of anxiety or fear (e.g., Kirsch & Henry, 1977;

Ramirez & Kratochwill, 1990; Wolpe & Lang, 1964). For example, procedures such as

systematic desensitization involve "imagining" the aversive stimulus or event while

engaging in a behavior incompatible with the escape response (Paul, 1967; Wolpe, 1958).

Although desensitization procedures have reportedly been one of the most effective means

of treating phobias (Mathews, 1978), there is no observable behavior available to directly

measure the effects of therapy. A similar procedure often used in the treatment of phobias

is graduated exposure therapy, which involves brief, in vivo exposures to the aversive

stimulus (Weidner, 1970). This procedure is most like stimulus fading because a hierarchy

of aversive stimuli is usually identified and exposure to these stimuli is faded from least to

most aversive over time. The effects of therapy are also often assessed using unobservable

changes in the client's reported level of anxiety or fear (Mathews, 1978).

In the present study, the effects of fading were directly assessed using a clearly defined

response (i.e., SIB) as the dependent variable. If a directly observable response was

established to indicate anxiety or level of aversiveness for individuals with phobias, the

effects of therapy could be assessed more reliably. A few studies have attempted to use

behavioral measures of the effects of in vivo exposure therapy. For example, D'Zurilla,

Wilson, and Nelson (1973) measured the effects of several treatment procedures on

subjects' fear of dead rats using an avoidance test among other measures before and after

treatment. This test involved presenting several different dead rats to the subjects and

giving points every time the subjects looked at, touched, or held a rat. Although these

responses were not specifically escape responses terminating the presentation of the rats,

approach behavior is functionally the opposite of avoidance or escape behavior, and

measuring approach may allow a more empirical assessment of the effects of treatment.

This approach could be used in a manner similar to communication training. For example,

the subject can only escape from the testing area or end the presentation of the event by

engaging in an alternative escape response (e.g., touching the stimulus). The amount of


time the individual must engage in the alternative behavior could be faded across sessions.

Other empirical measures of the effects of exposure therapy could involve the use of

negative reinforcement (e.g., a break) when a subject tolerates exposure to the aversive

stimulus. The criterion for escape could then be faded across sessions.

In conclusion, although there are still many dimensions of stimulus fading that merit

further research, this study has shown that fading without extinction will not maintain

reductions in behavior. Further research on manipulating other dimensions of the

instructional sequence or applying fading procedures to other behavior disorders, however,

is clearly warranted.


Carr, E.G. (1977). The motivation of self-injurious behavior: A review of some
hypotheses. Psychological Bulletin, 84, 800-816.

Carr, E.G., & Durand, V.M. (1985). Reducing behavior problems through functional
communication training. Journal Applied Behavior Analysis, 18, 111-126.

Cataldo, M.F., & Harris, J. (1982). The biological basis for self-injury in the mentally
retarded. Analysis and Intervention in Developmental Disabilities, 2, 21-39.

Clark, F. C. (1964). Effects of repeated VI reinforcement and extinction upon operant
behavior. Psychological Reports, 15, 943-955.

Clark, F. C. & Taylor, B. W. (1960). Effects of repeated extinction of an operant on
characteristics of extinction curves. Psychological Reports, 6, 226.

Davis, C.A., Brady, M.P., Williams, R.E., & Hamilton, R. (1992). Effects of high-
probability requests on the acquisition and generalization of responses to requests in
young children with behavior disorders. Journal Applied Behavior Analysis, 25, 905-

D'Zurilla, T.J., Wilson, G.T., & Nelson, R. 91973). A preliminary study of the
effectiveness of graduated prolonged exposure in the treatment of irrational fear.
Behavior Therapy, 4, 672-685.

Eyman, R.K., Borthwick, S.A., & Miller, C. (1981). Trends in maladaptive behavior of
mentally retarded persons placed in community and institutional settings. American
Journal of Mental Deficiency, 5, 473-477.

Griffin, J.C., Williams, D.E., Stark, M.T., Altmeyer, B.K., & Mason, M. (1984). Self-
injurious behavior: A state-wide assessment of severe cases. In J.C. Griffin, M.T.
Stark, D.E. Williams, B.K. Altmeyer, & H.K. Griffin (Eds.), Advances in the
treatment of self-injurious behavior (pp. 1-25). Richmond, TX: Texas Planning
Council for Developmental Disabilities.

Goh, H. L., & Iwata, B.A. (in press). Behavioral persistence and variability during
extinction of self-injurious escape behavior. Journal Applied Behavior Analysis.

Heidomrn, S.D., & Jensen, C.C. (1984). Generalization and maintenance of the reduction of
self-injurious behavior maintained by two types of reinforcement. Behaviour Research
and Therapy, 22, 581-586.

Hineline, P.N. (1977). Negative reinforcement and avoidance. In W.K. Honig & J.E.R.
Staddon (Eds.), Handbook of operant behavior (pp. 364-414). Englewood Cliffs, NJ:

Homer, R.H., Day, H.M., Sprague, J., O'Brien, M., & Heathfield, L.T. (1991).
Interspersed requests: A nonaversive procedure for reducing aggression and self-injury
during instruction. Journal Applied Behavior Analysis, 24, 265-278.

Iwata, B.A. (1987). Negative reinforcement in applied behavior analysis: An emerging
technology. Journal Applied Behavior Analysis, 20, 361-378.

Iwata, B. A., Dorsey, M. F., Slifer, K. J., Bauman, K. E., & Richman, G. S. (1982).
Toward a functional analysis of self-injury. Analysis and Intervention in Developmental
Disabilities, 3, 1-20.

Iwata, B.A., Pace, G.M., Kalsher, M.J., Cowdery, G.E., & Cataldo, M.F. (1990).
Experimental analysis and extinction of self-injurious escape behavior. Journal of
Applied Behavior Analysis, 23, 11-27.

Iwata, B.A., Vollmer, T.R., & Zarcone, J.R. (1990). The experimental (functional)
analysis of behavior disorders: Methodology, applications, and limitations. In A.C.
Repp & N.N. Singh (Eds.), Perspectives on the use of nonaversive and aversive
interventions for persons with developmental disabilities (pp. 301-330). Sycamore, IL:
Sycamore Publishing Co.

Kirsch, I., & Henry, D. (1977). Extinction versus credibility in the desensitization of
speech anxiety. Journal of Consulting and Clinical Psychology, 45, 1052-1059.

Lovaas, O.I., Freitag, G., Gold, V.J., & Kassorla, I.C. (1965). Experimental studies in
childhood schizophrenia: Analysis of self-destructive behavior. Journal of Experimental
Child Psychology, 2, 67-84.

Lovaas, O.I., Newsom, C.D., & Hickman, C. (1987). Self-stimulatory behavior and
perceptual reinforcement. Journal of Applied Behavior Analysis, 20, 45-68.

Lovaas, O.I., & Simmons, J.Q. (1969). Manipulation of self-destruction in three retarded
children. Journal of Applied Behavior Analysis, 2, 143-157.

Mace, F.C., Hock, M.L., Lalli, J.S., Belfiore, P., Pinter, E., Brown, D.K, & West, B.
(1988). Behavioral momentum in the treatment of noncompliance. Journal of Applied
Behavior Analysis, 21, 123-141.

Mace, F.C., Page, T.J., Ivancic, M.T., & O'Brien, S. (1986). Analysis of environmental
determinants of aggression and disruption in mentally retarded children. Applied
Research in Mental Retardation, 2, 203-221.

Mathews, A. (1978). Fear-reduction research and clinical phobias. Psychological Bulletin,
85, 390-404.

McAfee, J.K. (1987). Classroom density and the aggressive behavior of handicapped
children. Education and Treatment of Children, .i, 134-145.

Michael, J. L. (1982) Distinguishing between discriminative and motivational functions of
stimuli. Journal of the Experimental Analysis of Behavior, 37, 149-155.

Pace, G.M., Iwata, B.A., Cowdery, G.E., Andree, P.J., & Mclntyre, T. (1993).
Stimulus (instructional) fading during extinction of self-injurious escape behavior.
Journal of Applied Behavior Analysis, 26, 205-212.

Paul, G.L. (1967). Insight versus desensitization in psychotherapy two years after
termination. Journal of Consulting and Clinical Psychology, 31, 333-348.

Ramirez, S.Z., & Kratochwill, T.R. (1990). Development of the fear survey schedule for
children with and without mental retardation. Behavioral Assessment, 12, 457-470.

Repp, A. C., Felce, D., & Barton, L. E. (1988). Basing the treatment of stereotypic and
self-injurious behaviors on hypothesis of their causes. Journal of Applied Behavior
Analysis, 21, 281-289.

Sailor, W., Guess, D., Rutherford, G., & Baer, D.M. (1968). Control of tantrum
behavior by operant techniques during experimental verbal training. Journal of Applied
Behavior Analysis, 1, 237-243.

Schroeder, S.R., Schroeder, C.S., Smith, B., & Dalldorf, J. (1978). Prevalence of self-
injurious behaviors in a large state facility for the retarded: A three year follow-up
study. Journal of Autism and Childhood Schizophrenia, 8, 261-269.

Sidman, M. (1960). Tactics of scientific research. New York: Basic Books.

Skinner, B.F. (1969). Contingencies of reinforcement: A theoretical analysis. New York:

Steege, M.W., Wacker, D.P., Berg, W.K., Cigrand, K.K., & Cooper, L.J. (1989). The
use of behavioral assessment to prescribe and evaluate treatments for severely
handicapped children. Journal of Applied Behavior Analysis, 22, 23-33.

Steege, M.W., Wacker, D.P., Cigrand, K.C., Berg, W.K., Novak, C.G., Reimers,
T.M., Sasso, G.M., & DeRaad, A. (1990). Use of negative reinforcement in the
treatment of self-injurious behavior. Journal of Applied Behavior Analysis, 23., 459-

Tate, B.G. & Baroff, G.S. (1966). Aversive control of self-injurious behavior in a
psychotic boy. Behavior Research and Therapy. 4, 281-287.

Ulman, J. D., & Sulzer-Azaroff, B. (1975). Multielement baseline design in educational
research. In E. Ramp & G. Semb (Eds.), Behavior analysis: Areas of research and
application (pp. 359-376). Englewood Cliffs, NJ: Prentice-Hall.

Wacker, D.P., Steege, M.W., Northup, J., Sasso, G., Berg, W., Reimers, T., Cooper,
L., Cigrand, K., & Donn, L. (1990). A component analysis of functional
communication training across three topographies of severe behavior problems.
Journal of Applied Behavior Analysis, 23, 417-429.

Weidner, F. (1970). In vivo desensitization of a paranoid schizophrenic. Journal of
Behavior Therapy and Experimental Psychology, 1, 79-81.

Weeks, M., & Gaylord-Ross, R. (1981). Task difficulty and aberrant behavior in
severely handicapped students. Journal of Applied Behavior Analysis, 14, 449-463.

Wolpe, J. (1958). Psychotherapy by reciprocal inhibition. Stanford, CA: Stanford
University Press.

Wolpe, J., & Lang, P.J. (1964). A fear survey schedule for use in behavior therapy.
Behaviour Research and Therapy, 2, 27-30.

Zarcone, J.R., Iwata, B.A., Hughes, C.E., & Vollmer, T.R. (1993). Momentum versus
extinction effects in the treatment of self-injurious escape behavior. Journal of Applied
Behavior Analysis, 26, 135-136.

Zarcone, J.R., Iwata, B.A., Vollmer, T.R, Jagtiani, S., Smith, R.S. & Mazaleski, J.L. (in
press). Extinction of self-injurious escape behavior with and without instructional
fading. Journal of Applied Behavior Analysis.


Jennifer R. Zarcone attended Florida State University and the University of Florida as

an undergraduate. As a major in psychology, she was required to take a number of courses

in all areas of psychology. After taking several courses in behavior analysis, she decided

that she wanted to do graduate work in the field of applied behavior analysis. She applied

to the graduate program in the experimental analysis of behavior and was accepted into the

program for fall of 1986.

As a graduate student, she has conducted research with developmentally disabled

individuals who exhibit self-injurious behavior. Throughout graduate school she has been

a research assistant at the Center for Self-Injury under the direction of her advisor, Brian

Iwata. Currently, she is assisting Dr. Iwata in developing an outpatient clinic to provide

assistance to families in the community with self-injurious children. In addition to the

assessment and treatment of self-injury, she is also interested in conducting behavioral

research in educational settings.

I certify that I have read this study and that in my opinion it conforms to acceptable
standards of scholarly presentation and is fully adequate, in scope and quality, as a
dissertation for the degree of Doctor of Philosophy.

B an A. Iwata, air
Professor of Psychology

I certify that I have read this study and that in my opinion it conforms to acceptable
standards of scholarly presentation and is fully adequate, in scope and quality, as a
dissertation for the degree of Doctor of Philosophy.

Marc N. Branch
Professor of Psychology

I certify that I have read this study and that in my opinion it conforms to acceptable
standards of scholarly presentation and is fully adequate, in scope a
dissertation for the degree of Doctor of Philosophy.

Henry S. Penny er
/ Professor of Psychology

I certify that I have read this study and that in my opinion it conforms to acceptable
standards of scholarly presentation and is fully adequate, in scope and quality, as a
dissertation for the degree of Doctor of Philosophy. s

Merle Meyer /
Professor of Psychology

I certify that I have read this study and that in my opinion it conforms to acceptable
standards of scholarly presentation and is fully adequate, in scope and quality, as a
dissertation for the degree of Doctor of Philosophy. /

William D. Working /
Professor of Special Educatio

This dissertation was submitted to the Graduate Faculty of the Department of
Psychology in the College of Liberal Arts and Sciences and to the Graduate School and
was accepted as partial fulfillment of the requirements for the degree of Doctor of

December, 1993 _________
Dean, Graduate School

Illl11111111 l 11111 11111
3 1262 08556 9449