Biological setting events for self-injury

zy

zyxwvutsrq
zyxwvutsrq

MENTAL RETARDATION AND DEVELOPMENTAL DISABILITIES
RESEARCH REVIEWS 7:94-98 (7995)

BIOLOGICAL SETTING EVENTS
FOR SELF-INJURY
Edward G. Carr and Christopher E. Smith

State University of New York at Stony Brook (E.G.C.) and Developmental Disabilities Institute (C.E.S.), Stony Brook, New York

Self-injurious behavior typically occurs in response t o specific
antecedent stimuli that predict the occurrence of such desirable
consequences of the behavior as obtaining attention from others,
escaping aversive situations, and securing preferred foods, objects,
and activities. Setting events constitute an additional controlling
variable, altering the probability that a specific antecedent stimulus
will evoke self-injury. We discuss the influence of several biological
factors, including menses, otitis media, fatigue, allergies, and constipation, that may serve as setting events. We also present a model
which suggests that these events alter the functional properties of
the antecedent stimuli that control self-injurious behavior, thereby
strengthening the specific consequences that maintain the behavior.
One implication of this model is the likelihood that effective
intervention will have to have multiple components, addressing
both the biological and environmental variables that control the
behavior. Asecond implication is that greater collaboration between
medical and psychological or educational practitioners will be
necessary in order to address the multiple factors that affect selfinjury.

D 1995 Wiley-Liss, Inc.

zyxwvutsrqp

M R D D Research Reviews 1: 94-98
~~

much more frequent, but when they withhold attention, the
level of self-injury dramatically decreases [Lovaas et al., 1965;
Carr and McDowell, 19801. Similarly, the escape function has
been demonstrated in studies showing that when hstrating task
demands are withdrawn in response to self-injury, the behavior
increases, but that when such demands are not withdrawn after
display of the behavior, self-injury becomes much less likely to
occur [Iwata et al., 1982; Iwata et al., 19901. Finally, the
tangible-seeking function of self-injury has been demonstrated.
Studies have shown that presentation of desired tangibles after
self-injurious behavior causes an increase in the f?equency of
such behavior [Derby et al., 19921. Other functions ofself-injury
that have been demonstrated include, for example, avoidance of
social situations [Taylor and Carr, 19921 and seeking sensory
stimulation (e.g., auditory, visual, and gustatory) [Favell et al.,
1982; Rincover and Devaney, 19821. However, the bulk of the
relevant research has examined attention, escape, and tangbles as
sources of reinforcement for self-injury.
Because self-injury typically is purposeful, it does not
occur randomly. Rather, it is closely associated with antecedent
stimuli that predict whether the behavior is likely to be followed
by desirable consequences. Thus, a low level of attention can
function as an antecedent stimulus that triggers episodes of

self-injury [Carr and Durand, 19851. The process underlying this
outcome is likely to involve a history in which an individual has
repeatedly experienced the reinstatement of high levels of
attention following the display of self-injury. Over time, the
individual learns that exhibiting self-injurious behavior in the
presence of low levels of attention from others “pays off’ in the
sense that it is quickly and reliably followed by nurture, support,
and expressions of concern from others. It is likely that a similar
process is responsible for self-injury that serves the function of
escape or attaining tangibles. For example, in the case of escape,
task demands, including those involving academic or work
activities, sometimes function as antecedent stimuli that trigger
episodes ofself-injury [Can et al., 19761. This may be related to a
history in which an individual has repeatedly experienced the
removal of frustrating, boring, or effortful tasks following the
display of self-injury. Over time, the individual learns that such
behavior exhibited in the presence of task demands “pays off’ in
the sense that the task is quickly and reliably removed [Carr et al.,

~

~

-

~~

~

Key Words: self-injurious behavior, setting events, developmental

disabilities, functional analysis

rom a behavioral perspective, self-injury is a function of
three sets of variables: consequences, antecedents, and
setting events. Consequences refer to the effects that
self-injury has on others. For example, a young boy diagnosed as
retarded may engage in head banging. One consequence of this
behavior is that the boy’s anxiety-stricken mother quickly
approaches her child and offers him physical comfort, soothing
words, and other forms of attention. Alternatively, the boy may
start head banging when confronted with a frustrating academic
task. In consequence, his alarmed teacher withdraws the task,
thereby permitting the boy to escape from the unpleasant
situation. O n another occasion, the boy may exhibit head
banging in a toy store. The consequence is that his embarrassed
father offers him several toys in order to placate him. In each
case, self-injury serves a particular function or “purpose” for the
child. The specific consequence of each self-injurious act allows
one to infer what purposes the behavior serves-in the examples
gwen, attention-seeking, escape from aversive events, and
attempting to secure desired tangible items.
Research supports the idea that individuals generally
exhibit self-injury because it produces desirable consequences for
them. The attention-seeking function has been demonstrated in
studies showing that when adults offer children comforting
words or pleas to desist from self-injury, the behavior becomes

F

o 1995 Wiley-Liss, Inc.

zyxwvu
zyxwvuts
Address reprint requests to Edward Cam, Department of Psychology, State Unlvenity of
New York, Stony Brook, N Y 11794-2500.


19911. The research literature is replete
with descriptions of other stimuli that
trigger self-injury: being teased, receiving
negative feedback about task performance, desiring objects that are momentarily inaccessible, and being prompted
by a teacher to improve performance on a
given task [Carr et al., 1990; O’Neill et
al., 1990; Gardner and Sovner, 19941.
Although it is clear that self-injury
is a joint function of antecedents and
consequences, an emerging literature
now suggests that a third set of variables,
setting events, also has an important role.
Setting events are broad contextual influences that alter ongoing relationships
between antecedent stimuli and responses
such as self-injury [Bijou and Baer, 19611.
For example, we noted earlier that there
&equently is a relationship between
antecedent stimuli such as task demands
and self-injurious behavior. This relationship can be made more or less probable
through the influence of various setting
events. Some setting events are environmental. For example, Kennedy and
Itkonen [1993] reported that for one
woman diagnosed as being retarded, an

environmental setting event-use
of a
specific bus route to school-became
associated with an increased probability
of self-injurious behavior in the presence
of a variety of antecedent stimuli, including self-help and community-based demands. In contrast, the absence of the
setting event-use of an alternative bus
route-was
associated with negligible
levels of self-injury in the presence of the
same types of task demands.
Other setting events are biological.
Although these have been frequently
alluded to in the literature, they have not
yet been systematically analyzed. O u r
purpose here is to focus on the nature of
these biological setting events and propose a model for their operation that can
serve as a heuristic basis for subsequent
assessment and intervention research.

Menses
A number of studies involving
women with psychiatric disorders have
described an exacerbation of problems
associated with those disorders during the
premenstrual phase, the first few days of
menstrual flow, or both. These problems
include an increase in affective disorders
such as depression [Endicott et al., 19811,
and psychotic episodes [Felthous et al.,

19801. Several case studies involving
women diagnosed as retarded have also
noted an increase in depressive symptoms
[Dennerstein et al., 19831, as well as
fearfulness and irritability [Altschule and
Brem, 19631 associated with menses.
Most interestingly, a recent study of 24
women diagnosed as retarded reported a
positive correlation between menses and

occurred at this time. Although this
correlation is of considerable interest, the
lack of a more fine-grained analysis in this
study makes it impossible to determine
whether the increase in self-injury was
triggered by the menses, by a mensesrelated alteration in the response to
specific environmental cues, or by some
third factor spuriously related to menses.

Otitis Media
Some forms of self-injury, such as
head banging, are by no means limited to
people with developmental disabilities.
There have been detailed descriptive
accounts of head banging occurring as
early as the first year of life in children
without disabilities [deLissovoy, 19621.
In one study of 374 normal children
between the ages of 19 and 32 months,
deLissovoy [1961, 19641 observed head

banging by 57 children or 15.2% of the
sample. Interestingly, in the deLissovoy
[1962] study, the tendency toward head
banging increased in some children when
they had colds or other unspecified
illnesses. In another study, deLissovoy
[1963] found an association between
otitis media (middle ear infection) and the
onset of head banging by young children.
There is no evidence that children with
developmental disabilities are less susceptible to such infection than are other
children. Because otitis media is so
common in infancy and early childhood,
one would expect that many children
with disabilities develop the condition
from time to time and that, as in the
general pediatric population, a certain
percentage of these children will develop
associated head banging.
Several investigators in the disabilities area have suggested a causal link
between otitis media and self-injury
[Bailey and Pyles, 1989; Gardner and
Sovner, 19941. However, only one group
of investigators reported, and only anecdotally, a case documenting the association between head banging and middle
ear infection [Gunsett et al., 19891.
Moreover, that case involved an adult.
Given the data from the general pediatric
literature and the allusions made to otitis
media in the literature on disabilities, the
time appears to be right for a systematic

experimental analysis of the role this
biological factor plays in producing selfinjury.

zyxwv

zyx
zyxw
zyxwvutsrq
Although it is clear that
self-injury is a joint
function of antecedents
and consequences, an
emerging literature now
suggests that a third set of
variables, setting events,
also has an important
role. Setting events are
broad contextual
influences that alter
ongoing rela tionships
between antecedent
stimuli and responses
such as self-inju ry.

DESCRIPTION OF BIOLOGICAL
SETTING EVENTS
Self-injurious behavior has been
linked to biological setting events associated with every major physiological
system: central nervous, gastrointestinal,
immune, reproductive, and respiratory

pailey and Pyles, 1989; Gunsett et al.,
1989; Gardner and Sovner, 19941. Although these linkages have not yet been
demonstrated experimentally, the correlational and descriptive evidence is pervasive enough to warrant closer scrutiny by
scientists and clinicians.

MRDD RESEARCH
REVIEWS

BIOLOGICAL

self-injury [Ghaziuddin et al., 19931.
Clearly, a variety of problems
associated with menses has been identified. However, the studies cited were
based in large measure on anecdotal data.
Moreover, only one of these studies
involved self-injury. Recently, however,
a detailed, systematic report on 9 women
with mental retardation has described
more clearly the association between
menses and self-injury [Taylor et al.,
19931. The major finding was that the
frequency of self-injurious behavior was
typically elevated during the early phases
of the menstrual cycle (menses, early and
late follicular). The vast majority of selfinjurious episodes, which included selfbiting, head banging, and face slapping,

SETTINGEVENTS

CARRS~SMITH

Other Biological Variables
Although menses and otitis media
are not the only biological factors that
have been linked to the occurrence of
self-injurious behavior, they are the ones
described in greatest detail in the litera-

zyxw
95


zyxwvu

Fig. 1. A conceptual framework for understanding the effects of biological setting events on self-injurious behavior. Problem behavior is
influenced by context (setting events and trigger stimuli). Setting events alter the likelihood that antecedent stimuli will trigger self injury
by changing the value of the consequences (escape, attention, tangibles) of self-injury.
r

zyxwvutsrqponmlkjihgf

ture. Scattered throughout the literature
are references, almost always anecdotal,
to other factors. The scope and variety of
these factors strengthens the argument
that biological setting events are pervasive
and therefore worth looking at more
systematically than has been done to date.
Fatigue has been suggested as a
possible setting event for self-injury. A
number of variables can induce fatigue:

nocturnal sleep deficit, prolonged physical activity, temperature extremes, and
lack of sugar, salt, or water in the body.
For this reason, remediation of problem
behavior associated with fatigue should
plausibly be linked with the specific
variable underlying the fatigue. The
relationship between fatigue and selfinjury has been observed in populations
with developmental disabilities [Gardner
et al., 1986; Bailey and Pyles, 19891.
However, documentation has been purely
anecdotal, leaving unclear the precise
relationship between fatigue and selfinjury.
Allergies have been identified as a
factor influencing problem behavior in
people without developmental disabilities
[e.g., Field, 19811. Interestingly, Gardner
[1985] noted similar influences in the case
of a young man diagnosed as being
severely retarded. Careful observation of
this person over a long period suggested a
relationship between his severe selfinjury, which necessitated the use of
restraints, and seasonal allergies. Staff
were trained to monitor his medical
condition and administer antihistamines
when appropriate. The result was that
restraints were no longer necessary and
self-injury became infrequent. This case
study is suggestive, as are other anecdotal
accounts of the relationship between
allergies and self-injury [Gourash, 1986;

Bailey and Pyles, 19891.

Some reports indicate that constipation is more common among people with
mental retardation than among others
[Roy and Simon, 19871. In one study,
staff reported that individuals most often
engaged in self-injurious behaviors such
as hitting themselves in the abdomen
when they had distended abdomens and
were screaming, which led to the hypothesis that such behavior was a response to
pain associated with constipation [Lekkas
and Lentino, 19781. Several other case
reports have suggested that chronic constipation makes individuals more irritable
than is normal and may therefore be a
predisposing factor for increased levels of
self-injurious behavior such as head
banging and self-slapping [Gunsett et al.,
1989; Gardner and Sovner, 19941.

BIOLOGICAL SETTING
EVENTS: A CONCEPTUAL
FRAMEWORK
Figure 1 illustrates a model for
understanding the impact of biological
setting events on problem behavior such
as self-injury. This model brings together
the three sets of controlling variables we
have described thus far: antecedents
(trigger stimuli), consequences, and setting events.
Consider an antecedent (trigger)

stimulus such as an academic demand
(e.g., “Tommy, point to the picture of
the horses.”). The stimulus may occasionally trigger an episode of self-injurious
behavior because, in the past, it has
caused the teacher to stop teaching and
withdraw the task. That is, the consequence of self-injury in the presence of
the trigger stimulus is that the child is
allowed to escape from a putatively
aversive task. However, the trigger stimulus does not always evoke self-injury at
the same level. Instead, it appears that
when an individual with disabilities is
suffering from some medical condition or
particular physiologcal state, self-injury is
more likely to occur. Thus, in our
example, if Tommy is suffering from
otitis media, constipation, fatigue, or
allergies, he is much more likely to
engage in self-injurious behavior when
presented with demands. Why should
this be the case?
One current view is that setting
events, including the biologcal factors
we have described, can alter the aversiveness of antecedent trigger stimuli, thereby
increasing the motivation for escape
behavior such as self-injury [Michael,
1982; Bailey and Pyles, 1989; Gardner
and Sovner, 19941. Our model has
considerable face validity given that the

zyxwvutsr

96

Summary
Our overview of the literature
suggests that the relationship between
self-injurious behavior and biological
factors is pervasive, cutting across a
variety ofphysiological states and medical
conditions. Many medical conditions are
mentioned in passing in the literature.
Despite the suggestive nature of this
relationship, the role of biological setting
events has not been systematically explored. The best studies, and there have
been few, present detailed correlational
data rather than the results of systematic
experimental analyses. The substantial
majority of reports present only case
studies or anecdotes. The rudimentary
but suggestive data in this field argue in
favor of a more detailed and organized
approach. To that end, we propose a
model for exploring the relationship
between self-injury and b i o l o g d setting
events.

MRDD RESEARCH
REVIEWSBIOLOGICALSETTINGEVENTS

zy

CARR& SMITH


various biological factors we have discussed have in common pain, physical
discomfort, or both. Thus, the notion is
that a mildly aversive antecedent such as a
task demand becomes extremely aversive
when it occurs in the context of a painful
or uncomfortable biologcal condition.
An important point is that with this
process in play, escape behavior (selfinjury) becomes more likely. One could
plausibly extend this model to other
sources of motivation besides escape.
Thus, a person who is ill and in pain may,
given an appropriate learning history,
exhibit self-injury in the presence of a
trigger stimulus to secure comfort, nurturance, and support (attention-seeking
pattern) o r specific foods, toys, or activities (tangible-seeking pattern).
The model we have just described
is a plausible one that has been discussed
in the literature. However, because of the
paucity of research in this area, it is fair to
say that, in large measure, setting events is
a descriptive rather than an explanatory
term. Thus, the lawful behavior changes
that occur in response to the manipulation of setting events do not in themselves
explain the observed correlations or
identift. the responsible processes. Nonetheless, it is the observed lawfulness of
such behavior change that justifies making the exploration of setting events a

research imperative for the future.
Because the proposed model links
self-injurious behavior to specific consequences (escape, attention, and tangibles),
it is an example of operant conditioning
[Skinner, 19381-that is, the control of
behavior by its consequences. There is an
important alternative to this operant
setting event model. Specifically, behavior can also be controlled by its antecedents in a reflexive paradigm [Pavlov,
19271. In this case, the painful stimulus
functions as an unconditional stimulus
that elicits self-injurious behavior irrespective ofwhether, for example, a demand is
present or absent. Thus, the pain associated with otitis media or constipation
could set off a bout of self-injury even in
the absence of demands. A biological
reflexive model is especially plausible in
light of animal experiments showing that
the delivery of a painful stimulus (electric
shock) can elicit episodes of self-injury in
monkeys [Frank et al., 1977; Gluck et al.,
19851. It is likely that self-injurious
behavior is influenced by both operant
and reflexive biologcal factors. It is
useful, therefore, to consider how the
relative contribution of each could be
assessed empirically.

ASSESSMENT STRATEGY
W e can illustrate one approach to
the assessment issues just described by
referring to our ongoing work on menstrual discomfort as a biological setting

event for self-injurious behavior [Carr et
al., 19931. W e have worked with several
young women living in community
residences whom staff members described
as exhibiting greatly increased levels of
self-injury during or just prior to their
menses. Staff descriptions clearly noted
that the women showed many verbal and
nonverbal manifestations of pain and
discomfort at this time. Further, it
appeared that it was the combination of
menstrual pain and the demands of daily
living that exacerbated the problem
behavior. To test this notion, we pre-

pain alone, even in the absence of trigger
stimuli (the second assessment condition),
should have generated self-injury, but it
did not. Second, it is likely that an
operant setting-event model was operating because self-injury occurred almost
exclusively as a joint function of the
presence of both the trigger stimuli and
the biologcal setting event (the fourth
condition); this is the model outlined in
Fig. 1. Trigger stimuli alone (the third
condition), menses alone (the second
condition), and, of course, the absence of
both factors (the first condition) all failed
to produce significant levels of selfinjury. Thus, this assessment format can
be useful in disentangling the role of

reflexive and operant factors, thereby
permitting identification of true setting
events.

The likelihood that

INTERVENTION
IMPLICATIONS
The likelihood that successful intervention will be multicomponent in nature is the most important implication of
a setting-event model. That is, some
aspects of the intervention will need to
target the biological component (e.g.,
menses) and some aspects will need to
target the environmental component
(e.g., demands). Again, we can illustrate
these points with reference to our study
of menses.
W e constructed multicomponent
interventions that addressed the pain and
discomfort associated with menses. For
one woman, for example, we included
the following intervention components:
use of a hot water bottle to alleviate
cramping; restructuring her diet to avoid
acidic foods that exacerbated stomach
upset during menses; responding to her
request for panty liners rather than pads
because she found the latter more uncomfortable; providing comforting hugs, pats,
and the like by female staff during periods
of increased discomfort. W e also included

intervention components to address issues related to the trigger stimuli (demands): temporarily reducing demands
during periods of intense discomfort;
shifting the demands to times of day
when the woman was feeling better;
teaching her to request assistance from
staff in carrying out demands when she
was feeling badly; implementing positive
mood induction techniques as a way of
reducing the aversiveness of the demands.
The combination of biological and
environmental elements into a multicomponent intervention package is a logical
byproduct of a model that stresses the
dual control of self-injurious behavior by
environmental trigger stimuh and biologi-

successful intervention
will be multicomponent
in nature is the most
important implication of
a setting-event model.
That is, some aspects of
the intervention will need
to target the biological
component (e .g .,menses)
and some aspects will
need to target the
environmental component
(e.g., demands).

zyxw

zyxw
zyxw

sented various combinations of trigger
stimuli-eg.,
demands involving food
preparation, grooming, cleaning one’s
room, and the like-in
the presence or
absence of the biological setting event
(i.e., obvious discomfort or pain associated with menses). The assessment consisted of four conditions: trigger stimulus
absent and setting event absent; trigger
stimulus absent and setting event present;
trigger stimulus present and setting event
absent; trigger stimulus present and setting event present. Self-injury was negligible during the first three conditions and
severe during the final condition.
This method of assessment is useful
in addressing several issues raised earlier.
First, reflexive factors were not a major
influence. If they had been, menstrual

zyxwvutsrq
zyxwvutsrq

MRDD RESFARCH REVIEWS BIOLOGICALSETTINGEVENTS CARR& SMITH

zyxwvu
97

cal setting events. Because of the complexities involved in such dual control,
increased collaboration between medical
practitioners and psychologists or educators in designing intervention is another
logical consequence of the model we
have presented. H

teaching behavior of adults. J Appl Behav
Analysis 24523-535.
deLissovoy V (1961): Head banging in early
childhood. J Pediatr 58:803-805.
deLissovoy V (1962): Head banging in early
childhood. Child Dev 33:43-56.
deLissovoy V (1963): Head banging in early
childhood: A suggested cause. J Genet Psychol
102:109-1 14.
deLissovoy V (1964): Head bangmg in early
childhood: Review of empirical studies.
Pediatr Digest 649-55.
Dennerstein L, Judd F, Davies B (1983): Psychosis
and the menstrual cycle. Med J Aust 1524526.
Derby KM, Wacker DP, Sasso G, et al. (1992): Brief
functional assessment techniques to evaluate
aberrant behavior in an outpatient setting: A
summary of 79 cases. J Appl Behav Analysis
25:713-721.
Endicott J, Halbreich U, Schacht S, et al. (1981):
Premenstrual changes and affective disorders.
Psychosomat Med 43:519-529.
Favell JE, McGimsey JF, Schell R M (1982):

Treatment of self-injury by providing alternate sensory activities. Analysis Intervention
Dev Disabilities 2:83-104.
Felthous AR, Robinson DB, Conroy R W (1980):
Prevention of recurrent menstrual psychosis
by an oral contraceptive. Am J Psychiatry
137:245-246.
Field C R (1981): Ecological illness: A factor in your
caseload and perhaps in your life. J Rehabil
47:15-18.
Frank R G , Gluck JP, Strongin TS (1977): Response
suppression to a shock-predicting stimulus in
differentially reared monkeys (Macaca mulafa).
Dev Psychol 13:295-296.
Gardner JM (1985): Using microcomputers to help
staff reduce violent behavior. Comput Hum
Serv 1:53-61.
Gardner WI, Cole CL, Davidson DP et al. (1986):
Reducing aggression in individuals with
developmental disabilities:An expanded stimulus control, assessment, and intervention
model. Educ Training Ment Retard 21:3-12.
Gardner WI, Sovner R (1994): Self-Injurious
Behaviors. Willow Street, P A Vida Publishing.
Ghaziuddin M, Elkins TE, McNeeley SG, et al.
(1993): Premenstrual syndrome in women
with mental handicap: A pilot study. Br J Dev
Disabil39:104-107.
Gluck JP, Otto MW, Beauchamp AJ (1985):
Respondent conditioning of self-injurious

behavior in early socially deprived rhesus

monkeys. J Abnorm Psychol 94:222-226.
Gourash LF (1986): Assessing and managing medical factors. In Barrett R P (ed): Severe
Behavior Disorders in the Mentally Retarded.
New York: Plenum, pp 157-205.
Gunsett R P , Mulick JA, Fernald WB et al. (1989):
Brief report: Indications for medical screening
prior to behavioral programming for severely
and profoundly mentally retarded clients. J
Autism Dev Disord 19:167-172.
Iwata BA, Dorsey MF, Slifer KJ, et al. (1982):
Toward a functional analysis of self-injury.
Analysis Intervention Dev Disabil2:3-20.
Iwata BA, Pace GM, Kalsher MJ, et al. (1990):
Experimental analysis and extinction of selfinjurious escape behavior. J Appl Behav
Analysis 23:ll-27.
Kennedy CH, Itkonen T (1993): Effects of setting
events on the problem behavior of students
with severe disabilities. J Appl Behav Analysis
26:321-327.
Lekkas C N , Lentino W (1978): Symptomproducing interposition of the colon: Clinical
syndrome in mentally deficient adults. J Am
Med Assoc 240:747-750.
Lovaas 01, Freitag G, Gold VJ, et al. (1965):
Experimental studies in childhood schizophrenia: Analysis of self-destructive behavior. J
EXPChild Psycho1 2~67-84.
Michael J (1982): Distinguishing between discriminative and motivational functions of stimuli. J
Exp Analysis Behav 37:149-155.
O’Neill RE, Horner R H , Albin R W , et al. (1990):
Functional Analysis of Problem Behavior: A
Practical Assessment Guide. Sycamore, IL:

Sycamore Press.
Pavlov IP (1927): Conditioned Reflexes. New
York: Dover Publications.
Rincover A, Devaney J (1982): The application of
sensory extinction procedures to self-injury.
Analysis Intervention Dev Disabil2:67-81.
Roy A, Simon GB (1987): Intestinal obstruction as
a cause ofdeath in the mentally handicapped.J
Ment Defic Res 31:193-197.
Skinner BF (1938): The Behavior of Organisms.
New York Appleton-Century-Crofts.
Taylor JC, Carr EG (1992): Severe problem
behaviors related to social interaction. I:
Attention seeking and social avoidance. Behav
Modif 16:30S-335.
Taylor DV, Rush D, Hetrick WP, et al. (1993):
Self-injurious behavior within the menstrual
cycle ofwomen with mental retardation. Am J
Ment Retard 97559-664.

zyxwvutsrq
zyxwvutsr

ACKNOWLEDGMENT
Preparation of this manuscript was
supported in part by Grant H133G20098
and Cooperative Agreement H133B20004
from the U.S. Department of Education.
We thank Dr. Martin Hamburg, Executive Director, Developmental Disabilities
Institute, for his generous support.

References

Altschule MD, Brem J (1963): Periodic psychosis of
puberty. AmJ Psychiatry 119:1176-1178.
Bailey JS, Pyles DAM (1989): Behavioral diagnostics. In Cipani E (ed), The treatment of severe
behavior disorders, Monographs ofthe Amencan Association on Mental Retardation, 12.
Bijou SW, Baer DM (1961): Child Development I:
A Systematic and Empirical Theory. Englewood Cliffs, NJ: Prentice-Hall.
Carr EG, Durand VM (1985): Reducing behavior
problems through functional communication
training. J Appl Behav Analysis 18:111-126.
Carr EG, McDowell JJ (1980): Social control of
self-injurious behavior of organic etiology.
Behav Ther 11:402-409.
Carr EG, Newsom CD, Binkoff JA (1976):
Stimulus control of self-destructive behavior
in a psychotic child. J Abnorm Child Psychol
4:139-153.
Carr EG, Robinson S, TaylorJC, CarlsonJI (1990):
Positive approaches to the treatment of severe
behavior problems in persons with developmental disabilities: A review and analysis of
reinforcement and stimulus-based procedures.
Monograph of The Association for Persons
with Severe Handicaps 4.
Carr EG, Smith CE, Magito-McLaughlin D, Reeve
CE (1993, May): Mood, menses, and meaning: Complex determinants ofproblem behavior. Paper presented at the annual meeting of
the Association for Behavior Analysis, Chicago.
Carr EG, TaylorJC, Robinson S (1991): The effects
ofsevere behavior problems in children on the

98

zyxwvuts

MRDD RESEARCH
REVIEWS BIOLOGICALSETTINGEVENTS

zy

CARR& SMITH