Content uploaded by Shoshana Shiloh
Author content
All content in this area was uploaded by Shoshana Shiloh on May 09, 2015
Content may be subject to copyright.
REDUCTION OF STATE-ANXIETY BY PETTING
ANIMALS IN A CONTROLLED LABORATORY
EXPERIMENT
SHOSHANA SHILOH
a,
*, GAL SOREK
a$
and JOSEPH TERKEL
b
a
Department of Psychology and
b
Department of Zoology, Tel AvivUniversity, Ramat Aviv, 69978
Tel Aviv, Israel
The effect on anxiety of petting an animal and the underlying mechanisms of such an effect were examined by
a repeated-measures, within-session experiment with 58 non-clinical participants. Participants were exposed to
a stressful situation in the laboratory /the presence of a Tarantula spider, which they were told they might be
asked to hold /and then randomly assigned to one of five groups: petting a rabbit, a turtle, a toy rabbit, a toy
turtle or to a control group. Participants’ attitudes towards animals were measured as potential moderators.
State-anxiety was assessed at baseline, after the stress manipulation, and after the experimental manipulation.
The main findings showed that petting an animal reduced state-anxiety. This effect could not be attributed to
the petting per se, since it was observed only with animals and not with matched toys. The anxiety-reducing
effect of petting an animal applied to both the soft cuddly animals and the hard-shelled ones. The anxiety-
reducing effect applied to people with different attitudes towards animals and was not restricted to animal
lovers. The discussion addresses possible emotional and cognitive foundations of the observed effects and their
implications.
Keywords: State-anxiety; Petting animals
State-anxiety has been defined as a transitory emotional response involving unpleasant
feelings of tension and apprehensive thoughts (Spielberger, 1966). It is often elevated in
the presence of fear-arousing cues, and its control has been the target of many
interventions. The present study focused on pet-assisted intervention for reducing
anxiety, a method that has gained enthusiastic fans over the years but has suffered from
insufficient scientific investigation.
Reviews of the literature point to many potential psychological, social and health
benefits of the human-animal bond (e.g., Brasic, 1998; Cusak, 1988; Edney, 1995). Pets
have been suggested to provide an unconditional source of affection, enhance self-
esteem and emotional stability, reduce feelings of loneliness and isolation and help
people socialize, provide pleasurable activity and assistance, are something to care for
and a source of consistency and a sense of security (see e.g., Edney, 1995; Katcher and
Friedmann, 1980; McCulloch, 1984). Pets have also been suggested to serve a
supportive function that buffers people against stress and illness (Allen, 1985). Among
$
Deceased.
*Corresponding author. Fax: 972-3-6423422.
E-mail: shoshi@freud.tau.ac.il
ISSN 1061-5806 print: ISSN 1477-2205 online #2003 Taylor & Francis Ltd
DOI: 10.1080/1061580031000091582
Anxiety, Stress, and Coping,
December 2003, Vol. 16, No. 4, pp. 387 /395
pet owners experiencing high levels of stress, interaction with pets was identified as an
important stress management practice (Gage and Anderson, 1985). Owners of dogs, in
particular, were buffered against the impact of stressful life events on physician
utilization, as reflected in fewer doctor contacts during stressful life periods (Siegel,
1990). Other studies, however, failed to support the relationship between pet ownership
and improved psychological health (Watson and Weinstein, 1993).
These findings and assumptions were the background for the development of pet-
facilitated psychotherapy, in which a pet is used as a co-therapist and becomes an
integral part of the treatment process (Barba, 1995; Levinson, 1965). Pet-assisted
therapy has been adapted by practitioners in social work, marriage and family
counselling, psychology, and psychiatry, and evaluated by them as an effective technique
(Mason and Hagan, 1999). Animal-assisted programmes for therapeutic interventions
have been developed for special populations such as children and adolescents with
various emotional, behavioural or mental problems (Mallon, 1992), and posttraumatic
stress disorder patients (Altschuler, 1999); specific guidelines were developed for this
purpose (e.g., Society for Companion Animal Studies, 1990).
Among the benefits reported by therapists, we were especially interested in anxiety
reduction. Barker and Dawson (1998) reported that interaction with a trained animal
during a session of therapy reduced state-anxiety in psychiatric inpatients. When healthy
children were examined by a doctor in a within-subject, time-series designed experiment,
with and without a dog present, significantly greater reductions in behavioural distress
and physiological parameters of stress were found when the dog was present (Nagengast
et al., 1997). When the effects of interacting with a dog, reading aloud or reading quietly
were assessed on measures of anxiety among college students, interacting with a dog or
reading quietly lowered both physiological and psychological indicators of anxiety
(Wilson, 1991). Taken together, these findings indicate that pet animals can reduce
anxiety, especially in stressful situations.
Several recent reviews have raised serious criticisms of the abundance of descriptive
studies and the paucity of adequate quantitative methodology in this area. According to
Brasic (1998), much of the published literature consists of anecdotal case reports, while
studies utilizing experimental designs and statistical analysis are rare. Brodie and Biley
(1999) found methodological difficulties stemming from the complexity of the subject
area, including poor design, small sample sizes, and failure to randomize. Much of the
research has been conducted in the field rather than in controlled laboratory settings,
with unclear control over extraneous variables. The fact that studies have been
financially supported by special interest groups raises doubts about the objectivity of
their conclusions (Brasic, 1998). The extent to which participants were stressed is also
questionable in most of the studies, since standard experimental stressors were not used.
A further complicating factor is that not all individuals have positive attitudes
towards animals (Kidd and Kidd, 1989), and not all animals raise the same positive
responses from people. Soft, hairy animals are known to be preferred and better liked
than cold animals (Margadant-van Arcken, 1989). It is not known whether different
types of animals and different attitudes of individuals moderate the benefits of
interacting with animals.
The present study was designed to extend our knowledge by adding: (1) a controlled
experiment that would examine the specific effects of an animal on state anxiety among
stressed, non-clinical individuals; (2) an examination of the significance of the nature
388 S. SHILOH et al
.
and features of the animal in determining its stress-reducing effects; and (3) an
evaluation of the possible moderating effects of interpersonal variance in attitudes
towards animals on the effects of an animal.
Three questions were asked: (1) Is the effect on anxiety due to the object being a live
animal? This was studied by comparing the effect of holding and petting live animals to
that of holding and petting matching toys, utilizing a randomized experimental design.
(2) What types of pets are most effective? Are all pets equally effective, or is the potential
effect specific to soft cuddly animals? This was studied by comparing petting a rabbit
with petting a turtle, in a randomized experimental design. (3) Is the effect of petting
animals moderated by attitudes towards animals? We measured attitudes toward
animals among all participants in our experiment, and looked for statistical interactions
between experimental conditions and this variable.
METHODS
Overview
The study was designed as a repeated-measures, within-session experiment. Non-clinical
participants were exposed to a stressful situation in the laboratory /the presence of a
Tarantula spider, which they were told they might be asked to hold. They were then
randomly assigned to one of five groups: petting either a rabbit, a turtle, a toy rabbit, or
a toy turtle, or to a control group that got neither an animal or a toy. The animals were
trained to be tolerant to petting. The toys matched the size and texture of the real
animal. State-anxiety was measured three times: at baseline, after the stress manipula-
tion, and after the experimental manipulation. Attitudes towards animals were assessed
at the beginning of the session.
Participants
Fifty-eight individuals (35 women and 23 men) took part, mean age 26.16 years (SD/
6.83, range from 17 to 58). 45 were students at Tel Aviv University (34 undergraduates
and 11 graduate students), and 13 were university employees.
Measures
Two standard questionnaires were used to measure state-anxiety and attitudes towards
animals.
State-Trait Anxiety Inventory (STAI) STAI (Spielberger et al., 1970) measured the
dependent variable, using the state-anxiety part that asks respondents to indicate how
they feel right now. Twenty items describing feelings of tension, nervousness, worry and
apprehension are rated on a 4-point scale. This measure was found reliable and valid in
numerous studies, including its Hebrew translation (Teichman, 1978). Its internal
reliability in our sample ranged from Cronbach’s alpha/0.86 at baseline to 0.93 after
the stress manipulation.
The Companion Animal Semantic Differential The Companion Animal Semantic
Differential (Poresky et al., 1988) was used to measure our moderator variable. This 18-
PET EFFECTS ON STATE-ANXIETY 389
item scale measures attitudes toward pet animals using a 6-point semantic-differential
format (hot/cold; good /bad; friendly/unfriendly, etc.). It was reported reliable and
valid by its developers (Poresky, et al., 1988). In our sample its internal reliability was
Cronbach’s alpha /0.88, and its distribution was comparable to that reported in the
American sample (means 89.649
/12.07 and 94.379/10.80 respectively).
Procedure Participants were recruited by notices posted around the university campus
inviting participation in a study on attitudes towards animals for course credit or a small
fee. After administering the Companion Animal Semantic Differential and the STAI to
participants individually, the stress manipulation was conducted. We used a spider as a
stress-arousing stimulus because of its face validity in the context of the study, and its
high fear arousing potential among non-clinical populations (Arrindell, 2000). The
experimenter (a female graduate student) uncovered a glass jar containing a Tarantula
spider, and said: ‘‘This is a spider. The experiment has two groups: one will watch the
spider, and the other will be asked to hold it. You will be told your assignment shortly.
Now, I have to ask you to wait a minute while I get something from the next room’’. The
experimenter left the room for 2 minutes, returned and administered the STAI again. At
this point the participant, who had been randomly assigned to one of the five groups,
was handed one of the following: a rabbit (n/13), a turtle (n/11), a toy rabbit (n/
11), or a toy turtle (n/12), which had been covered until then. The subject was
instructed ‘‘to hold and pet it for a while’’. The control group (n/11) was asked to wait
a little while longer while the experimenter left the room again. The uncovered glass jar
containing the spider remained in the room the whole time. The petting/waiting
period lasted for 2 minutes. The pet/toy was taken away by the experimenter, who
administered the STAI for the third time. All participants were then told they would not
have to hold the spider. At the end of the session a short interview was conducted on
what they had done, felt and thought during the experiment. They were then debriefed
and thanked.
RESULTS
Manipulation Check
The mean state-anxiety score at baseline was 30.07 (SD /7.69) and rose to 36.07 (SD/
11.09) after the stress manipulation. A repeated-measure analysis of this difference
yielded significant results (t(1,57) /4.11, pB/0.001). The stress manipulation, therefore,
effectively increased state-anxiety among the participants.
Experimental Effects
Distributions of age, gender and attitudes towards animals (Table I) did not differ
among the groups (F(4,53)/0.65, chi-square/1.80, F(4,53)/0.50, n.s., respectively).
Means and standard deviations of the three state-anxiety measures for each of the five
groups are presented in Table II. At baseline, there were no significant differences
among the 5 groups in state-anxiety mean scores (F(4,53) /1.42, n.s.). The differences
among the groups in state-anxiety scores after the experimental manipulation were
found significant using ANCOVA with baseline and post-stress anxiety scores as
covariates, (F(4,51) /3.30, pB/0.05). The effect size was medium (Cohen’s d/0.58).
390 S. SHILOH et al
.
Thus, the experimental manipulation was shown to have affected the participants’ state-
anxiety.
In order to determine the sources of the experimental effect, two contrasts were first
compared: animals versus toys, and soft versus hard-shelled groups. Means and SDs of
post manipulation state-anxiety of these contrasts are presented in Table III.
Results of contrast analyses adjusted for baseline and post-stress manipulation
anxiety show the animals-toys contrast to be significant (F(1,55)/4.51, pB/0.05), with
a low effect-size (Cohen’s d/0.31), while the soft versus hard-shelled contrast is not
significant (F(1,55)/2.39, n.s.). In addition, the control group versus animals contrast
was significant (F(1,55) /9.49, pB/0.01), while the contrast of control group versus
toys was not (F(1,55) /1.87, n.s.). Thus, petting animals resulted in lower state-anxiety
scores compared to petting toys, while the texture of the petted object had no effect. Nor
was there a significant difference in state-anxiety between petting a real rabbit or a real
turtle (F(1,55)/2.38, n.s.).
Interactions with Attitudes Towards Animals
Pearson correlations between attitudes and state-anxiety scores were non-significant
(rs/0.01, 0.12, and -0.03, at baseline, after stress manipulation and after experimental
manipulation respectively). Consequently, entering attitudes as a covariant in the overall
analysis did not change the manipulation effect (F(4,50)/3.28, pB/0.05). This was also
demonstrated by two-way between-subjects analysis of variance using a median split (at
score 90.50) of the participants with high versus low attitudes towards animals. No
TABLE II Means and standard deviations of three measures of state-anxiety according to
experimental groups
State anxiety measure: Rabbit Turtle Toy-rabbit Toy-turtle Control
n/13 n/11 n/11 n/12 n/11
Baseline 29.699
/5.31 32.189/9.49 32.919/8.26 29.089/7.74 26.649/7.16
Post stress manipulation 36.549
/6.28 33.279/10.76 37.279/14.06 37.179/14.86 35.919/9.28
Post experimental manipulation 28.549
/7.33 28.919/7.93 32.099/10.82 34.339/9.39 34.369/8.02
TABLE I Demographic variables and attitudes toward animals according to experimental groups
Rabbit Turtle Toy-rabbit Toy-turtle Control
n/13 n/11 n/11 n/12 n/11
Age 27.009
/7.00 23.009/4.00 26.009/4.00 28.009/6.00 27.009/11.00
Gender (F/M) 8/5 8/3 7/4 7/5 5/6
Attitude toward animals 86.089
/18.40 90.369/8.69 92.649/10.24 90.339/9.72 85.829/21.21
TABLE III Means and standard deviations of state-anxiety according to animals-toys and soft-
shelled-hard-shelled contrasts
nMean Standard Deviation
Animals 23 28.71 7.45
Toys 23 33.82 9.44
Soft 24 30.70 8.75
Hard-shelled 23 31.73 8.96
PET EFFECTS ON STATE-ANXIETY 391
significant interaction with experimental groups on post-manipulation anxiety was
found (F(4,57)/1.39, n.s.). These findings do not support the moderating hypothesis of
attitudes towards animals on the stress-reducing effects of petting animals.
DISCUSSION
Our study demonstrated that a short period of petting an animal resulted in reduced
state-anxiety among non-clinical individuals in a stressful situation. The experimental
design ruled out the effect being due to the petting per se, since it was observed only
with animals and not with matched toys. The anxiety-reducing effect of petting animals
was also found to hold for hard-shelled animals like a turtle and was not limited only to
soft cuddly animals, indicating that it was the quality of being alive rather than the
texture of the object that produced the effect. Finally, our findings showed that the
anxiety-reducing effect applies to people with different attitudes towards animals, and is
not restricted to animal lovers.
These findings correspond with those of a few previous experimental studies
reporting positive effects on anxiety and distress associated with interacting with
animals (Barker and Dawson, 1998; Nagengast et al., 1997; Wilson, 1991). Our findings
refined and strengthened the previous ones by experimentally manipulating the stress
and anxiety in the situation, and by adding groups engaged with similar inanimate
objects. A comparable control used in another study (not on anxiety) found that
children with Down’s syndrome interacting with a real dog showed more sustained focus
for positive and co-operative interactions than did children interacting with an imitation
dog in controlled conditions under the direction of an adult (Limond et al., 1997).
How can this effect be explained? Psychoanalytic approaches suggest that the innate
drive to associate with animals satisfies people’s emotional need for affiliation
(Levinson, 1972). Animals are believed to relate to symbolic thought according to
Jungian theory on archetypes and the unconscious (Henderson, 1999). Additional
emotional gratification comes from animals’ non-evaluative support, which can reduce
threat to the ego (Allen et al., 1991).
In our experiment, participants were instructed to hold and pet the animals/objects.
There is evidence that touching another living thing (pets and humans) engenders
positive feelings and reduces stress, pain and anxiety (e.g., Lafreniere et al., 1999;
Montagu, 1978; Spence and Olson, 1997). The relaxing and comforting emotion
induced by touching and petting an animal can, therefore, account for our findings.
Although some studies show that touch per se is not necessary for achieving the positive
effects of animals, and that similar effects can be obtained from the mere presence of
animals, like watching an aquarium filled with fish (e.g., Cole and Gawlinski, 1995), we
did not differentiate between the effects of touching and the mere presence of an animal.
Such a distinction is recommended in future experiments.
Another potential mechanism underlying our findings may be cognitive. Brickel
(1982) suggested an ‘attentional shift hypothesis’, according to which pets divert
attention from an anxiety-generating stimulus, helping alleviate anxiety. Evidence that
distraction is effective in diminishing anxiety supports this hypothesis (Wilkins, 1971).
Pet animals are ideally suited for a distraction role because of their appealing
characteristics. They are complex, unpredictable, interactive, and operate on tactile,
392 S. SHILOH et al
.
auditory, visual, and probably other levels. The mediational role of attention in anxiety
has been supported by some researchers (Clark, 1999; Penfold and Page, 1999), and
rejected by others (Allen et al., 1991; Harris, and Menzies, 1998). Our findings, while
compatible with the ‘attentional shift hypothesis’, do not preclude other underlying
mechanisms such as the aforementioned emotional processes. Studies directly examin-
ing the hypothesis that the petting effect is mediated by distraction are strongly
recommended.
In conclusion, it might be appropriate at this stage to assume that both emotional and
cognitive mechanisms, separately or in combination, can potentially explain our results.
Further research applying carefully controlled methodology is necessary to reveal and
separate the contributions of various causal factors.
Limitations of the Study
Generalizations should be drawn from our results with caution. One limitation concerns
the specific animals considered. Different animals have different characteristics, and
compatibility between people and their pets on physical, behavioural and psychological
dimensions was found to be related to the owners’ mental and physical health (Budge et
al., 1998). Despite our failure to find interactions between attitudes towards animals
and the effects of petting, we should bear in mind that we tested only two, small,
friendly pets. An unsuitable, unhealthy, poorly behaved, nervous animal can be
hazardous both physically and psychologically to the individual (Duncan, 1998; Edney,
1995).
The absence of an interaction between the anxiety reducing effect and attitudes
towards animals may be due to a restriction of the range of attitudes in our sample. The
distribution of scores in CASD was skewed, and may indicate that only those favourably
predisposed towards animals participated in the study. A more balanced sample may be
needed in order to test the interaction hypothesis. Other interpersonal differences in
traits that were not investigated in the present research may also moderate the effect of
petting. Replications of our results with different pets, individuals and moderating
variables, and including behavioural/avoidance and physiological measures of anxiety
are also desirable.
Also, the design of our experiment, in which participants were left alone in the room,
did not allow a direct check that participants actually followed the experimenter’s
instruction ‘‘to hold and pet’’ the animal. Consequently, it may be more accurate to
conclude that holding and/or petting an animal produced the observed effects.
Another issue that should be addressed is the type of stress alleviated by pets in our
experiment. In another laboratory study, researchers failed to find reduced state-anxiety
and arousal among male students interacting with a dog during a stressful speech task
compared to a control group (Straatman et al., 1997). They concluded that the stress of
the speech task and the laboratory setting overrode the influence of the pet. More
research is required to map the types and intensities of stress situations that are suitable
for pet-assisted interventions.
Finally, it has been argued that the high degree of experimental control characteristic
of laboratory-based research often fails to yield a source of clinically relevant
information that can be extrapolated to natural conditions (Chorpita, 1997). General-
izing from a non-clinical sample has also been questioned. In our study, even after the
anxiety arousing manipulation, levels of anxiety did not reach clinical levels. However,
PET EFFECTS ON STATE-ANXIETY 393
in a summary of a mini-series of papers on laboratory research on anxiety, Eifert et al.
(1999) concluded that experimental research is both relevant and indispensable for the
continued advancement of our understanding and treatment of anxiety. They also stated
that creating clinically relevant phenomena in populations without known pathology
permits a ‘‘cleaner’’ examination of the variables and processes involved. We adopt these
views, and encourage the development of more interdisciplinary non-conventional co-
operations, like ours, between professionals and researchers from different perspectives,
from nursing through psychotherapy to zoology and veterinary medicine, attracted by
similar questions. Such an endeavour presents a great challenge, and can be both
exciting and fruitful to all involved.
Acknowledgements
This article is devoted to the memory of Gal Sorek. This study was done in partial
fulfilment of the Master’s Degree thesis of the second author. We want to thank Shani
Doron for her excellent assistance in data collection, and our colleagues at the
Zoological park at Tel Aviv University for their co-operation. We would also like to
acknowledge Keren Hadzdakah in name of Bracha and Motti Blisser, and Yad Hanadiv
Foundation for partially supporting the present research.
References
Allen, K.M. (1985). The Human-Animal Bond . Scarecrow Press, Metuchen, NJ.
Allen, K.M., Blascovich, J., Tomaka, J. and Kelsey, R.M. (1991). Presence of human friends and pet dogs as
moderators of autonomic responses to stress in women. Journal of Personality and Social Psychology,61,
582 /589.
Altschuler, E.L. (1999). Pet-facilitated therapy for posttraumatic stress disorder. Annals of Clinical Psychiatry,
11,29/30.
Arrindell, W.A. (2000). Phobic dimensions: IV. The structure of animal fears. Behaviour Research and Therapy,
38, 509 /530.
Barba, B. (1995). A critical review of research on the human/companion animal relationship 1988 /93.
Anthrozoo
¨s,8,9/15.
Barker, S.B. and Dawson, K.S. (1998). The effects of animal-assisted therapy on anxiety ratings of hospitalized
psychiatric patients. Psychiatric Services,49, 797 /801.
Brasic, J.R. (1998). Pets and health. Psychological Reports,83, 1011 /1024.
Brickel, C.M. (1982). Pet-facilitated psychotherapy: a theoretical explanation via attention shifts. Psychological
Reports,50,71/74.
Brodie, S.J. and Biley, F.C. (1999). An exploration of the potential benefits of pet-facilitated therapy. Journal of
Clinical Nursing,8, 329 /337.
Budge, R.C., Spicer, J., Jones, B. and George, R.S. (1998). Health correlates of compatibility and attachment in
human-companion animal relationships. Society and Animals ,6, 219 /234.
Chorpita, B.F. (1997). Since the operant chamber: are we still thinking in Skinner boxes? Behavior Therapy,28,
577 /583.
Clark, D.M. (1999). Anxiety disorders: why they persist and how to treat them. Behaviour Research and
Therapy,37,S5/S27.
Cole, K. and Gawlinski, A. (1995). Animal assisted therapy in the intensive care unit. Research Utilisation ,30,
529 /536.
Cusak, O. (1988). Pets and Mental Health . The Haworth Press Inc, New York.
Davis, J.H. and Juhasz, A. McC. (1984). The human/companion animal bond: how nurses can use this
therapeutic resource. Nursing and Health Care ,5, 496 /501.
Duncan, S.L. (1998). The importance of training standards for service animals. In: Wilson, C.C. and Turner,
D.C. (Eds.), Companion Animals in Human Health , pp. 251/266. Sage Publications, Thousand Oaks, CA.
Edney, A.T. (1995). Companion animals and human health: an overview. Journal of the Royal Society of
Medicine,88, 704 /708.
394 S. SHILOH et al
.
Eifert, G.H., Forsyth, J.P., Zvolensky, M.J. and Lejuez, C.W. (1999). Moving from the laboratory to the real
world and back again: increasing the relevance of laboratory examinations of anxiety sensitivity. Behavior
Therapy,30, 273 /283.
Gage, M.G. and Anderson, R.K. (1985). Pet ownership, social support, and stress. Journal of the Delta
Society,2,64/71.
Harris, L.M. and Menzies, R.G. (1998). Changing attentional bias: can it effect self-reported anxiety? Anxiety
Stress and Coping ,11, 167 /179.
Henderson, S.J. (1999). The use of animal imagery in counseling. American Journal of Art Therapy,38,20/26.
Jorgenson, J. (1997). Therapeutic use of companion animals in health care. Image: Journal of Nursing
Scholarship,29, 249 /254.
Katcher, A.H. and Friedmann, E. (1980). Potential health value of pet ownership. Compendium of Continuing
Education Practice Vet,2, 117 /121.
Kidd, A.H. and Kidd, R.M. (1989). Factors in adults attitudes toward pets. Psychological Reports,65, 903 /
910.
Lafreniere, K.D., Mutus, B., Cameron, S., Tannous, M., Giannotti, M., Abu-Zahra, H. and Laukkanen, E.
(1999). Effects of therapeutic touch on biochemical and mood indicators in women. Journal of
Alternative and Complementary Medicine,5, 367 /370.
Levinson, B.M. (1965). Pet psychotherapy: use of household pets in the treatment of behavior disorder in
childhood. Psychological Reports ,17, 695 /698.
Levinson, B.M. (1972). Pets and Human Development . Thomas, Springfield, IL.
Limond, J.A., Bradshaw, J.W.S. and Cormack, K.F.M. (1997). Behaviour of children with learning disabilities
interacting with a therapy dog. Anthrozoo
¨s,10,84/89.
Lynch, J.J. (1985). The Language of the Heart . Basic Books, New York.
Mallon, G.P. (1992). Utilization of animals as therapeutic adjuncts with children and youth /a review of the
literature. Child and Youth Care Forum,21,53/67.
Margadant-van Arcken, M. (1989). Environmental education, children, and animals. Anthropozoo
¨s,3,14/19.
Mason, M.S. and Hagan, C.B. (1999). Pet-assisted psychotherapy. Psychological Reports,84, 1235 /1245.
McCulloch, W. (1984). An overview of the human-animal bond: present and future. In: Anderson, R.K., Hart,
B.L. and Hart, L.A. (Eds.), The pet connection: Its influence on our health and quality of life,pp.30/35.
Globe, St. Paul, MN.
Montagu, A. (1978). Touching. Harper & Row, New York.
Nagengast, S.L., Baun, M.M., Megel, M. and Leibowitz, J.M. (1997). The effects of the presence of a
companion animal on physiological arousal and behavioral distress in children during a physical
examination. Journal of Pediatric Nursing ,12, 323 /330.
Penfold, K. and Page, A.C. (1999). The effect of distraction on within-session anxiety reduction during brief in
vivo exposure for mild blood-injection fears. Behavior Therapy,30, 607 /621.
Poresky, R.H., Hendrix, C., Mosier, J.E. and Samuelson, M.L. (1988). The companion animal semantic
differential /long and short form reliability and validity. Educational and Psychological Measurement,
48, 255 /260.
Siegel, J.M. (1990). Stressful life events and use of physician services among the elderly: the moderating role of
pet ownership. Journal of Personality and Social Psychology,58, 1081 /1086.
Society for Companion Animal Studies. (1990). Guidelines for the Introduction of Pets in Nursing Homes and
other Institutions. Straight Line, Glasgow.
Spence, J.E. and Olson, M.A. (1997). Quantitative research on therapeutic touch /an integrative review of the
literature 1985 /1995. Scandinavian Journal of Caring Sciences,11, 183 /190.
Spielberger, C.D. (1966). Anxiety and Behavior. Academic Press, New York.
Spielberger, C.D., Gorsuch, R.L. and Lushene, R.L. (1970). State-trait anxiety manual. Consulting
Psychological Press, Palo Alto, CA.
Straatman, I., Hanson, E.K.S., Endenburg, N. and Mol, J.A. (1997). The influence of a dog on male students
during a stressor. Anthrozoo
¨s,10, 191/197.
Teichman, Y. (1978). Affiliative reaction in different kinds of threat situations. In: Spielberger, C.D. and
Sarason, I.G. (Eds.), Stress and Anxiety, Vol. 5, pp. 131 /144. Halsted Press, Washington, DC.
Watson, N.L. and Weinstein, M. (1993). Pet ownership in relation to depression, anxiety, and anger in working
women. Anthrozoo
¨s,6, 135 /138.
Wilkins, W. (1971). Desensitization: social and cognitive factors underlying the effectiveness of Wolpe’s
procedure. Psychological Bulletin ,76, 311 /317.
Wilson, C.C. (1991). The pet as an anxiolytic intervention. Journal of Nervous and Mental Diseases,179, 482 /
489.
PET EFFECTS ON STATE-ANXIETY 395
