Extinction
deficits in male rhesus macaques with a history
of self-injurious behavior
Lutz C, Tiefenbacher S, Meyer J, Novak
M.
New England Primate Research
Center, Southborough, Massachusetts,
USA.
Self-injurious behavior (SIB) occurs in both
human and nonhuman primate populations. Despite
the potential for harm, SIB may persist in part
because of an inability to inhibit behavior that
results in wounding. A lever-pressing task was
used to test the prediction that monkeys with
SIB would show greater persistence in
lever-pressing on extinction trials than monkeys
without the disorder.
The subjects were 15 individually-housed
adult male rhesus macaques, 10 of which (the SIB
group) had a veterinary record of self-inflicted
wounding. All of the monkeys were trained to
lever-press for food rewards to a criterion of
400 total responses. The test procedures
consisted of five daily 30-min sessions divided
into six 5-min intervals. On day 1, the subjects
received continuous reinforcement. On days 2-4,
testing consisted of alternating
reinforced/unreinforced 5-min intervals,
beginning with reinforcement. Reinforced
intervals were cued with a buzzer. On day 5, the
subjects received no reinforcement. The number
of lever-presses and behavioral responses were
recorded during each session. Saliva samples
were collected for cortisol measurement before
and after test sessions on days 1, 2, and
5.
As predicted, monkeys with SIB lever-pressed
more than controls during extinction intervals
on days 2-4. There was no difference on day 1 or
day 5. The frequency of scratching,
yawning, and abnormal behavior increased
when reinforcement was intermittent (days 2-4)
or absent (day 5). Cortisol levels were highest
with continuous reinforcement (day 1), and may
reflect differential levels of food intake
rather than stress. The presence of extinction
deficits suggests that SIB may persist in some
monkeys because they lack the ability to
regulate the intensity of their biting
behavior.
INTRODUCTION
Self-injurious behavior (SIB) is a
potentially destructive behavior that occurs in
both human and nonhuman primate populations. SIB
in humans often takes the form of self-cutting,
self-biting, and head-banging, without suicidal
intent [Carr, 1977; Favazza & Rosenthal,
19931. The rate of SIB tends to be higher in
subjects with mental retardation (10.-14%)
[Griffin et at, 1986; Maisto et al., 1978;
Schroeder et al., 1978] than in the general
population (4%) [Briere & Gil, 1998;
Klonsky et al., 2003]. However, individuals
with genetic disorders, such as Lesch-Nyhan or
Rett syndromes, are also known to commit
self-mutilating acts [Anderson et al., 1978;
Coleman et al., 1988; Nyhan, 1976].
Environmental risk factors associated with SIB
include restrictive housing conditions, such as
institutional rearing of children IjBeckett et
al., 2002] and adult internment in
correctional facilities [Boiko & Lester,
2000].
Similarly to humans, some macaque monkeys
may also develop an SIB syndrome [Bayne et
al., 1995; Bellanca & Crockett, 20021, which
can take the form of head-banging, excessive
hair-pulling, and self-directed biting,
occasionally resulting in self-inflicted
wounding [Bayne & Novak, 1998; Novak,
2003]. Risk factors for SIB in these animals
include variables such as individual housing at
an early age or for an extended period of time
tiBellanca & Crockett, 2002; Lutz et al.,
2003a]. Although environmental factors may
predispose an animal to develop SIB, specific
triggers or experiences may be needed to
initiate the actual selfinjurious act. For
example, stressful situations such as being
moved to a novel room [Lutz et al., 2003b)
or subjected to repeated blood draws [Lutz
et al., 2003a] have been associated with an
increase in self-directed biting or wounding.
With repetition of stressful triggers, SIB may
become ritualized, i.e., less variable, more
automatic, and more easily activated
[Dantzer, 1986].
Even though SIB may cause pain and/or
injury, it persists in captive populations of
macaque monkeys. Recent surveys of
individually-housed macaques found that
approximately 25% of the animals exhibited
self-biting behavior [Lutz et al.,
2003a]. However, a smaller portion of this
population (5-11%) produced wounds that required
veterinary care [Bayne et al., 1995;
Bellanca & Crockett 2002; Lutz et al.,
2003a1.
It is unclear why some animals bite
themselves without wounding, while others bite
and wound. One possible explanation is that
wounders find it more difficult to inhibit their
injurious behavior than other monkeys. Lending
some credence to this view is an early study by
Gluck and Sackett [19761, in which a
lever-pressing task was used to assess
extinction deficits (i.e., continuing to
leverpress in the absence of rewards) in monkeys
reared in isolation. Isolation rearing is a
major predictor of SIB in rhesus monkeys, and
the isolate-reared monkeys took longer than the
controls to extinguish their lever-pressing
behavior. In the present study, we used a
lever-pressing procedure to examine the
relationship between SIB and extinction deficits
in socially reared monkeys. Thus, SIB in this
population was not the result of impoverished
early rearing. If SIB is a manifestation of an
inability to inhibit certain kinds of behavior,
then it should take monkeys with an SIB syndrome
longer to extinguish their lever-pressing
behavior compared to monkeys with no wounding
history.
DISCUSSION
Monkeys with a veterinary record of
self-inflicted wounding showed more persistence
in lever-pressing during extinction intervals on
days 2-4 than those without such a record. This
result suggests that monkeys with a history of
SIB may be unable to inhibit the behaviors that
lead to wounding. At the present time, it is
unclear whether the problem is one of
suppressing biting behavior or of regulating the
intensity of the biting response. The latter
view is supported by the fact that some monkeys
habitually bite themselves but do not wound
themselves.
Differences in response to extinction cannot
be explained by differences in motivation
between monkeys with and without SIB. The groups
did not differ in the number of training
sessions required to meet the criterion, and
there were no group differences in
lever-pressing when the monkeys were exposed to
Continuous Reinforcement (day 1 of testing).
Both groups also showed a rapid cessation of
lever-pressing when exposed to the No
Reinforcement condition (day 5 of testing). This
latter point is interesting because it suggests
that the extinction deficit was present only
when reward alternated with nonreward, and did
not carry over from the intermittent condition
(days 2-4) to the nonrewarded condition on day
5.
All of the monkeys responded somewhat
negatively to the changing reinforcement
contingencies. Abnormal behavior, scratching,
and yawning increased when reinforcement
was intermittent (days 2-4) or absent (day 5).
Thus, exposure to extinction appeared to produce
some behavioral tension, regardless of group.
However, the cortisol results did not parallel
the behavioral reactions: the condition that
elicited the fewest behavioral responses
(Continuous Reinforcement) resulted in the
highest level of cortisol. Food consumption has
been shown to increase cortisol levels in humans
tHansen et al., 1997; Korbonits et al., 1996;
Rosmond et al., 2000]. In the present study,
both food intake and salivary cortisol levels
were greatest on day 1, which suggests that the
difference in cortisol across days may be due
more to food intake than to stress.
Taken together, these data suggest that
monkeys with SIB may have difficulty regulating
or controlling their responses under certain
conditions. The presence of extinction deficits
in some captive macaques may help to explain why
some animals persist in biting to the point of
injury.
Self-injurious behavior in male rhesus
macaques does not reflect externally directed
aggression.
Physiol Behav. 2003;78(1):33-39.
Lutz C, Marinus L, Chase W, Meyer J, Novak
M.
Self-injurious behaviors (SIB), such as
self-biting and self-wounding, have been
observed in a small percentage of captive
nonhuman primates. Because rhesus monkeys that
exhibit SIB also tend to be more aggressive, it
was hypothesized that SIB is related to
externally directed aggression and is associated
with contexts in which physical contact between
participants is prevented. The purpose of this
study was to test the hypothesized relationship
between SIB and outward aggression. Subjects
were first presented with videotapes of
conspecifics, scenery and a blank screen, and
their behavior was recorded. Levels of salivary
cortisol, an indicator of stress, were also
measured before and after presentation of the
videos. Although aggression increased when
subjects viewed tapes containing conspecifics,
neither cortisol levels nor self-biting behavior
varied as a function of tape content. The
subjects were then placed in two additional test
situations: an empty room and the same room
containing an unfamiliar conspecific. Aggression
was significantly higher in the stranger
condition compared to the empty room condition.
The two situations yielded parallel increases in
cortisol, suggesting that being alone was just
as stressful as being paired with an unfamiliar
conspecific. Self-biting rates were also similar
in these two conditions. Thus, contrary to our
prediction, increases in aggression did not
correlate with increases in SIB. These results
suggest that under similarly stressful
conditions, SIB and externally directed
aggression are unrelated.
Stereotypic and self-injurious behavior in
rhesus macaques: a survey and retrospective
analysis of environment and early
experience.
Am J Primatol. 2003;60(1):1-15.
Lutz C, Well A, Novak M.
Abnormal behavior in captive rhesus monkeys
can range from active whole-body and
self-directed stereotypies to self-injurious
behavior (SIB). Although abnormal behaviors are
common in singly-housed rhesus monkeys, the type
and frequency of these behaviors are highly
variable across individual animals, and the
factors influencing them are equally varied. The
purpose of this investigation was to survey
abnormal behavior in a large population of
rhesus macaques, to characterize the
relationship between stereotypies and
self-injury, and to identify potential risk
factors for these aberrant behaviors. Behavioral
assessments of 362 individually housed rhesus
monkeys were collected at the New England
Regional Primate Research Center (NERPRC) and
combined with colony records. Of the 362 animals
surveyed, 321 exhibited at least one abnormal
behavior (mean: 2.3, range: 1-8). The most
common behavior was pacing. Sex differences were
apparent, with males showing more abnormal
behavior than females. SIB was also associated
with stereotypies. Animals with a veterinary
record of self-injury exhibited a greater number
of self-directed stereotypies than those that
did not self-injure. Housing and protocol
conditions, such as individual housing at an
early age, longer time housed individually,
greater number of blood draws, and nursery
rearing, were shown to be risk factors for
abnormal behavior. Thus, many factors may
influence the development and maintenance of
abnormal behavior in captive primates. Some of
these factors are intrinsic to the individual
(e.g., sex effects), whereas others are related
to colony management practices, rearing
conditions, and research protocols.
The physiology and neurochemistry of
self-injurious behavior: a nonhuman primate
model.
Front Biosci. 2005;10:1-11.
Tiefenbacher S, Novak MA, Lutz CK, Meyer
JS.
Self-injurious behavior (SIB) is a serious
behavioral condition that afflicts millions of
individuals in the United States alone. The
underlying factors contributing to the
development of self-injury in people are poorly
understood, and existing treatment strategies
for this condition are limited. A low but
persistent percentage of socially reared
individually housed rhesus monkeys also
spontaneously develop SIB. Data obtained from
colony records suggest that the risk of
developing SIB in socially reared rhesus monkeys
is heightened by adverse early experience and
subsequent stress exposure. The present review
summarizes the physiological and neurochemical
findings obtained in this nonhuman primate model
of SIB, focusing on monoamine neurotransmitters,
neuropeptides, and neuroendocrine systems. The
results indicate that monkeys with SIB exhibit
long-lasting disturbances in central and
peripheral opioid and stress response systems,
which lead to increased levels of anxiety. Based
on these findings, we propose an integrated
developmental-neurochemical hypothesis in which
SIB arises from adverse life events in a subset
of vulnerable monkeys, is maintained by a
persisting dysregulation of several
neurochemical and physiological systems, and
functions to periodically reduce anxiety when
the levels of anxiety become excessive.
Implications of this hypothesis for
understanding self-injury in patients with
borderline personality disorder and members of
the general population are discussed.
