Review of the research project with the autistic child and his family

The assesment of the presence of an assistance-trained dog on the autistic child and the family

M. Trudel 1* Professor, Psychoeducation Department, Faculty of Education Université de Sherbrooke, Sherbrooke (Quebec), Canada.

S. Fecteau 2 Doctorand, Faculty of Education Université de Sherbrooke, Sherbrooke (Quebec), Canada.

N. Champagne 3 Mira Foundation, Sainte-Madeleine (Quebec), Canada

1) Psychoeducation Department, Faculty of Education, 1111 Saint-Charles street West, suite 252-23 (East Tower), Longueuil (Quebec), J4K 5G4, Canada. Telephone : (450) 463-1835 (extension 61752)

2) Faculty of Education, 1111 Saint-Charles street West, Longueuil (Quebec), J4K 5G4, Canada.

3) Mira Foundation, 1820, Rang Nord-Ouest, Sainte-Madeleine (Quebec), J0H 1S0, Canada. Telephone : (450) 795-3725.

* Email : marcel.trudel@usherbrooke.ca

Summary

The study was completed through a partnership with an organization offering the services of assistance-trained dogs. The objective of the study was to evaluate the impact of the introduction of an assistance-trained dog on the autistic child and the child’s family. The article will first review the research literature from the perspective of the controversy around autism and the mother-child attachment relationship. From there, the authors will tie in the results of studies examining the impact of the assistance-trained dog on the child and the family, with a particular focus on autism and family relations. The choice of a mixed methodology in the interest of an ecological analysis of the family will be discussed, and the results will be presented from the perspective of the diversity of measures, issues of ecological validity and the constraints of assessing the atypical child in the everyday family environment.

Keywords : autistic child, assistance-trained dog, attachment, mix methods, family ecology.

Context of the Execution of the Research Project

This article aims to describe the methodological approach chosen to conduct the assessment of a group of autistic children at their home. This research project is part of a collaboration initiative with the Mira Foundation (see http://www.mira.ca). This organization from the Province of Quebec has developed a recognized expertise about the assistance-trained dogs for visually handicapped persons. More recently, the organization worked toward the implementation of a new support service aimed at the families with young children suffering of Autism Spectrum Disorders (ASD), to assist the parents in their role. 108 families of children with a confirmed ASD diagnosis (including autism, Asperger syndrome and unspecified ASD) (1) are expected to participate in this project. Our research group was initially approached to implement an assessment protocol enabling the evaluation of the impact of the presence of an assistance-trained dog on the child in his everyday family environment. During the first year, a pilot project was launched to develop a protocol that would enable the assessment of the changes induced by the dog. As of now, about thirty families have participated to the project, at the rate of three evaluation sessions: one before the introduction of the dog, and two once the dog has been introduced. On the scientific knowledge level, the project is likely to give relevant information pertaining to the understanding of the performance of an ASD child in his family environment, and of the effects of the assistance-trained dog – on an individual and on a relational level.

Summary Assessment of Research Studies in the Subject Area

The Autism Spectrum Disorders issue, and more specifically autism, has significantly increased in the last forty years. Thus, based on two reviews of about forty studies each, the authors conclude that the prevalence of the issue has substantially increased since the 60’s (2-4). This assessment does not reach consensus among the researchers; some claiming that the evolution would more likely be linked to changes applied to the typological classification of the issue (5). It is also noted that autism affects more boys than girls, with a 4 to 1 ratio. This would be attributable to a reduced sensibility of the boys’ brain to encode the information associated with the social indications (6,7). On this matter, the results of Naber and his contributor’s study (8) bring to light the great difficulty for the autistic children to maintain interactions implying mutuality at the social attention level. On the other hand, the link between genes and an autism diagnosis was often documented in the studies related to family and twins (9,10). It is estimated that the genes contributions would be of about 90%, although epigenetic factors and the environment could influence the expression of the different forms of autism (11). Moreover, the stress associated with the presence of a child with autism would mean imposing constraints on the quality of interactions within the family, and to accentuate the tendency of the children’s withdrawal and communication difficulties (12). All things considered, to this day, no consensus exists with regards to the origin of autism.

Globally, autism is considered as a neurodevelopmental disorder characterised by the presence of deficiencies in social functions, in verbal and non-verbal communication, and by the manifestation of stereotyped behaviour and interests (13). On the behavioural characteristics of the autistic child, problems are sometimes observed with cooperative play, social interactions and the expression of empathy towards peers (14). An issue with the modulation of social interactions and the adaptability to change in routine is also identified, thus restricting reciprocity in social exchanges. In 1943, Kanner was already describing (15), the main symptoms of autism as a social and affective deviance. Moreover, autistic children manifest, when compared to so-called normal children, a strong preference for non-social objects (16).

In such case, it is not surprising to find an increasing interest from researchers in the parent-child attachment relationship. These relationships are regulated by a system of behaviours emitted by the child towards his mother or significant persons in his circle of acquaintances (17). The main function of the attachment behaviour demonstrated by the child, when he feels threatened or in a state of distress, is the preservation of the closeness to the person who gives him a socio-emotional security (18). Beyond infancy, these attachment relationships are governed by internal working models built from past socio-emotional experiences with significant figures (19). When a secure attachment relationship exists, the child then feels at ease to explore his environment. Two conditions are mandatory to the existence of a secure attachment relationship: the mother’s receptiveness to the child’s signals and her responses to his needs; in addition to pleasant social interactions in a context of warmth and intimacy with the parental figure (20). As to the mother’s internal working model, it is built based on past experiences with her own parents and how she interprets them, the interactions with her child and of her representation of herself as a parent (21).

Considering the social and affective constraints found in an autistic child, is it reasonable to wonder about the possibility that they put at risk the establishment of attachment relationships? The results of some studies reveal than the autistic child seems apt to distinguish between the parent and the stranger by demonstrating a more secure behaviour towards his parent in situation of distress (22, 23). Moreover, following a separation from the parent, the child would demonstrate more behaviour of search for proximity to the parental figure (24, 25). However, in spite of the presence of attachment behaviour in the autistic child, it is reasonable to wonder if this profile is comparable to the profile of so-called normal children. The results of the meta-analysis recently published by Rutgers and his contributors (26), referring to 10 studies, highlight that globally, autistic children would establish a less secure relationship with their parent when compared to non-autistic children. In fact, autistic children are able to develop an attachment relationship based on safety, but the parent-child relationship type would be less flexible; thus affecting the synchrony on the interactive plan. Secondly, this study identifies that the presence of a mental deficiency would represent an important moderating factor qualifying the scale and the direction of the observed impacts. Thus, for the autistic children who do not present comorbidity implying a severe mental handicap, no difference on the indication of socio-emotional security is found in comparison to non-autistic children. Finally, the variations linked to the severity of the diagnosis of autism would also contribute to the quality of the parent-child attachment relationship. The summary of this meta-analysis (26) nevertheless give an empirical demonstration of the conditions having an unfavourable impact on the establishment of a secure attachment relationship for the autistic child. However, it is important to emphasize that the studies estimating the impacts of autism on attachment relationships were predominantly realized based on Ainsworth’s “Strange Situation” (18), thus in a laboratory context. Moreover, this procedure is only applicable to children younger than 2 years old. The use of this protocol implies that the autistic child experiences the inevitable stress deriving from the separation from his mother, but also a change of routine of such scale that it is hard to bear for children who do not have a great adaptability to change (27). The use of another assessment tool, like Water’s Q-Short methodology (1987) may present several advantages (28): a greater ecological validity since the autistic child is observed at his home, the possibility of obtaining iterated readings, contrary to “Strange Situation” protocol, and the opportunity to not only estimate the behaviour referring to the security in the attachment, but also those linked to the child’s dependency. Despite the results indicating that autism could have adverse effects on attachment relationships, controversy still persist among researchers. Moreover, few studies addressed the long term effects of autism on the child, and even more rarely on the family.

Furthermore, several studies seem to reveal that the presence of an assistance-trained dog may prompt autistic children to increase their social contacts and facilitate social interactions (30). Since one the symptoms of autism is linked to an impairment of social functioning and that the presence of a dog seem to contribute to the development of these skills, it is plausible to think that the integration of a well-trained dog could constitute a promising intervention path in the field of autism. Besides influencing positively the development of the child’s social skills, would it be reasonable to suppose that the permanent presence of the dog can contribute to strengthen the attachment relationship of the autistic child, and to affect the level of social cohesion in the family? Cain (31), as well as Albert and Bulcroft (32), demonstrate that an important proportion of people consider their domestic animal as a family member. These people report having developed new social contacts thanks to the presence of their domestic animal because it gives them attention and affection. Referring to Ainsworth’s work, Sable (33) suggest that domestic animals have the potential to procure a bond susceptible to contribute to the development of a feeling of well-being and security. Melson’s (34) work with regards to children’s attachment behaviour in the presence of a dog especially contributed to the development of this field of study. The author identifies four major dimensions to consider when studying the relationship between the dog and the child : 1- the type of activity and the time spent with the animal, 2- the child’s interest for the animal, 3- the child’s knowledge of the animal and it’s care, and 4- the reciprocity of the exchanges while interacting (35). If the presence of a domestic animal in the family unit seems favourable to the development of the communications between the members, is it possible to apply the same concept to the family of a child with ASD? More recently, Martin and Farnum (30) have assessed the impacts of the interactions with a dog in therapy for young children with ASD. They conclude that dog-assisted therapies are superior since their results show that the children were more joyful, attentive and aware of environmental changes when the dog was present, when compared to interventions where no animal was present. Redefer and Goodman (36) make the hypothesis that the dog could constitute a powerful multi-sensory stimulus allowing to counter the lack of integration of the emotional and sensory stimulations which is found in autistic children. Their results confirm that a dog-assisted therapy effectively lead to an increase of social interactions and a reduction of social isolation of the autistic child.

The present study pursues four additional objectives. First of all, to document and assess the quality of the attachment relationships of children with ASD by taking into account the severity of the diagnosis and the family stress levels. The second objective is oriented towards the evaluation of the impacts of the integration of an assistance-trained dog on the child’s attachment and social behaviours. The third objective is aimed towards the appraisal of the domesticated dog’s impact on the family’s cohesiveness. Lastly, the study of the relationship between the child and the assistance-trained dog will allow for the definition of the mediating or moderating role of the dog in the parent-child attachment relationship.

Methodological Choices

Participants and Research Specifications

In order to be selected to for this study, the family will receive an assistance-trained dog from the Mira Foundation, following a five days of training given by an experienced trainer. During this training, the parent will learn how to interact effectively with the dog and how to incite the animal and the child to interact together (1). The families accepted in this program must meet the criteria defined by the Mira Foundation, one of these being the presence of a child, aged between five and ten years old, having received an ASD diagnosis from a healthcare professional (autism, Asperger syndrome or unspecified ASD). A total of 108 families are targeted to participate in this project. The specifications plan for the participation of subgroups composed of 8 families every three months to allow for the preliminary training of dogs. On the assessment level, three home meetings are performed: three weeks before the introduction of the dog (period 1), a month after the insertion of the dog (period 2) and six months after the dog’s integration (period 3). The addition of a control group composed of about fifty families is planned. This sample is derived from the list of families of autistic children awaiting an assistance-trained dog. Two evaluation sessions will be carried out with these families. The visits are composed of four playing periods of 8 to 10 minutes each and one discussion period with the parent. The session starts with a free playing period between the child and the mother with a toy belonging to the child. Afterwards, the mother leaves to fill out the Family Environment Scale questionnaire (37) thus creating a non-availability period; during that time, the assessor plays with the child, with the same object. Once the mother has completed the questionnaire, a second playing period between her child an her is started, using a toy provided by the assessor. The last playing period between the mother and the child is a problem solving task using a puzzle also provided by the assessor. At the end of each visit, an interview with the parents takes place. Many open-ended questions with regards to various areas of the child’s development are addressed to draw up a complete assessment of the child and his family. These aspects are the pregnancy and childbirth conditions, the child’s characteristics (emotional, cognitive, psychomotor and social), the circumstances of the autism diagnosis and the services received, the child’s usual behaviour (at bedtime, at school or in daycare), the family members’ behaviour towards the child, the child-family interaction, the dog’s behaviour, the child and family.

To the families meeting the research criteria, the Mira Foundation offers a dog aged between 15 and 24 months old. The dog has already completed a 3-4 months training program with a trainer certified by the foundation. The dogs are Labrador, Bernese, Golden Retriever or Labernese (Labrador and Bernese hybrid). These dogs are of average or below average weight for their race in order to facilitate the child’s adaptation. In addition, these dogs are selected according to very strict physical and mental health criteria. Thus, they must show no anxiety, demonstrate a great tolerance to manipulation, be able to manage their insecurity, be calm, present no aggressiveness and adapt to various environments. Lastly, before they are introduced in their new home, these dogs must be house trained and have successfully passed their obedience training. This training procedure ensures a great homogeneity of the behavioral profile of the dogs; creating for this project a unique assessment context when compared to the studies usually carried out in the field.

Procedure Description and Main Instruments

Quantitative Measurements

After the initial home visit of approximately 90 minutes, the assessor completes two evaluation tools: the Attachment Q-Sort (AQS) (38) and the Coder Impressions Inventory (CII) (39). The AQS facilitates the evaluation of the quality of the child-parent attachment relationship. The third version of this tool was developed with the collaboration of experts in the field of attachment who formulated 90 statements describing the attachment behaviour of a child to a parent. Immediately after the home visit, the statements are distributed among 9 categories varying from "extremely typical" to "extremely atypical" when referring to the mother-child dyad. Beforehand, the observer was trained on the attachment theory and on the tool’s utilisation. The child’s description, given by the assessor, is then put in correlation with the normative reference profiles from experts. These profiles refer to two constituents linked to attachment issue: security and dependence (38). These theoretical criteria result from average results obtain by experts which completed the questionnaire according to what they consider as being characteristic of a child representing the extreme point of a dimension connected to attachment. The results correspond to the calculation of the correlations between the distribution of the 90 items made by the assessor and the experts. The validity of the instrument was repeatedly documented in studies about attachment (38).

As for the CII questionnaire, it was developed as part of the American Fast Track project on the prevention of behavioural issues and aims at assessing the educational practices of the parent, as perceived by an observer. Research studies results confirm the validity of this tool (40). The CII is composed of 81 statements measuring six elements: parental support, discipline and the strictness of the parent’s behaviour, the positive or negative affect of the child, and the child’s obedience.

The evaluation performed by the parent is based on an abridged electronic version of the Parenting Stress Index (PSI) which is sent by email. This instrument was adapted for and validated by mothers from the province of Quebec (41,42). The PSI enables the assessment of child rearing difficulties encountered by parents with the help of 36 statements distributed among two categories: the stressors linked to the child’s sphere and those linked to the parental sphere. The results of validation study directed by Lacharité and his collaborators (42) reveal that the internal consistency is established at 0,86 for the elements part of the child’s sphere and at 0,91 for the parent’s sphere. The internal consistency for the total stress score is equally satisfactory at 0,93.

Furthermore, the parent and the teacher each complete a version of the Pervasive Developmental Disorder Behaviour Inventory (PDDBI) (43). The parent’s version consists of 10 assessment scales divided in two dimensions: the inadequate behaviours (sensory behaviours and perceptions, fears, awakening difficulties, behavioural issues and aggressiveness) and the social or linguistic pragmatic issues, and the adapted behaviours (social approach, receptive communication, learning and memory, phonological skills, semantic skills). During the first home visit, the parent also completes a questionnaire about the social cohesion of the family, the Family Environment Scale (37). This instrument facilitates the assessment of the family environment and is composed of 90 statements evaluating 10 sub-scales with the aid of a dichotomous scoring. Three dimensions are given prominence: the support in the family relations, the context of the personal growth and the recognition linked to the family system’s organisation. Moos (44) reports indications of reliability varying from 0,60 to 0,80.

During the second home visit, a French adaptation of the protocol developed by Waters and Rogrigues-Doolabh (45) with regards to secure attachment gives access to the maternal representation of the attachment. This protocol consists of six lists composed of 12 words distributed between three columns. The parent gets familiarized with the words list and then must make up a story including each word. The audio part of the interview is recorded to facilitate the subsequent decoding. These stories are revealing of the base scripts that the parent has internalized. A score, varying from 1 to 7, is assigned to each story. A 7 indicates the presence of a secure base of attachment (46). This protocol is currently being validated with samples from various regions, including the province of Quebec. Articles reporting on the validity of the material have been published recently (47,48).

Finally, a questionnaire created for this study, the ICACE ("Inventaire des Comportements d’Attachement Chien-Enfant", the inventory of child-dog attachment behaviours), allows for the assessment of the behaviour of the child towards the dog and vice versa. The ICACE is composed of 23 questions. The manifested behaviours of the child towards the dog are assessed using 15 items while 8 items are used to assess the dog’s behaviour. These statements are assigned a score with the help of a four levels Likert type scale, varying from "rarely" to "always". This instrument is completed by the mother in the presence of the assessor. The decision to create this tool was essentially based on the fact that no tool evaluating the impact of a domestic dog on ASD children. The ICACE was developed following an exhaustive review of the measuring tools developed by Poresky (CABS and CASD) (49,50), Holcomb (PAS) (51), Melson (PASC) (52), Johnson (LAPS) (53) and Zasloff (CCAS) (54). It also takes into account notions from the attachment theory, autism and the reciprocal link between the child and his dog. However, the majority of the tools developed to this day assess the attachment of a person to its animal without consideration for the animal’s behaviour (53). Thus, the question series about the dog’s behaviour towards the child should remedy to the limit identified by Serpell (55) which is the exclusive consideration of the human being’s point of view in the study of the relation between the child and the animal.

The third and last visit is roughly identical as the first evaluation session. However, the data collection with regards to the dog’s behaviour using the ICACE is added to the other repeated measures.

Qualitative measurements

In less than 48 hours after the end of each home visit, a recorded interview of the assessor is conducted by the chief researcher order to write up the summary of the visit. Globally, this interview establishes the visit report by using open-ended questions linked to the child’s characteristics as observed during the home visit, the quality of the interactions between the child, the family and the dog, and the description of the performance during the implementation of the evaluation protocol.

Critical Assessment of the Evaluative Approach

On the whole, the procedure and instruments choice seems appropriate to the evaluation context of autistic children in their home environment. Moreover, few studies address the daily performance of the autistic child in his family. The selected evaluation modalities are diversified (interviews, observations and questionnaires) and report the abstract orientation of the project which considers that the study of the dog’s impact on the autistic child must be discussed from an ecosystem-based approach of the family. Incidentally, as stated by Odendaal (56), an ecological approach must be privileged in this field of study in spite of the constraints which it imposes on the progress of the research. The measures used in the present study try to report this holistic conception of the family.

By using a mixed evaluative approach, which is, for that matter, an increasingly favoured research method, the project’s orientation recognizes the importance and the utility of the quantitative and qualitative methods. Actually, this approach postulates that it is essential to privilege a more pragmatic conception of research which would rather report a continuum between these two epistemological poles to optimize the advantages and to minimize the weaknesses of each of the methods (57,58). Thus, the present study is inspired, to a certain extent, by this new orientation in the research field by recognizing the contribution of each of the approaches, while facilitating their complementarities.

Considering the diversity of the evaluation context, both the diagnosis characteristics and the variations related to the children’s age, it seems appropriate to ponder on the relevance of the utilisation of some instruments of measure. It is the case for the Q-Sort on the parent-child attachment since it must be recognized that the instrument was rarely used with autistic children. Thus, it seems reasonable to question the instrument’s adaptation in a study on atypical children. The analysis that will be realized based on the attachment database must take into account this methodological constraint, while contributing to further document this appraising issue. The AQS instrument was nevertheless favoured, comparatively to other attachment evaluation methods, because it offers an alternative to the issues raised by the "Strange Situation". The assessment of the mother-child attachment in the family environment gives a great advantage to this instrument since unfamiliar situations and places seem to generate much anxiety for the majority of the children living with an autism diagnosis.

Finally, it seems appropriate to underline an innovative aspect of the study which is the creation of a measuring instrument addressing the evaluation of the dog’s impact on the autistic child and the dog’s behaviour by itself. Conceptually, the content of the ICACE, which is based on the theory of attachment, should reflect the relational component involved in assessing the effects of an assistance-trained. Following this first experimentation, several adjustments will have to be made to the instrument, one of them being a modification of the scales and the addition of a few statements, as suggested by the parents. However, the ICACE reveals itself as a relevant tool in the study of the relationship between a child and an assistance-trained dog. It also enables the establishment of a link between the parent-child attachment and the attachment involving the dog. In this line of thought, the multi-respondents approach promoted in this study helps to nuance and balance the analysis of the perceptions of the parents and observers.

References

1. Viau R, Champagne N, Lupien S, Walker D, Sabourin G. Protocole d’expérimentation pour le projet des enfants qui présentent des troubles envahissants du développement notamment l’autisme. Ste-Madeleine, Québec : Fondation Mira; 2006.

2. Fombonne E. Epidemiology of autistic disorder and other pervasive developmental disorders. Journal of Clinical Psychiatry 2005; 66 10: 3–8.

3. Williams JG, Higgins JPT, Brayne CEG. Systematic review of prevalence studies of autism spectrum disorders. Archives of Disease in Childhood 2006: 91 1: 8–15.

4. Wing L, Potter D. The epidemiology of autistic spectrum disorders: is the prevalence rising? Mental Retardation and Developmental Disabilities Research Reviews 2002; 8 3: 151-61.

5. Blaxill MF. What’s going on? The question of time trends in autism. Public Health Reports 2004; 119 6: 536–51.

6. Baron-Cohen S. The extreme male brain theory of autism. Trends in Cognitive Sciences 2002; 6 6: 248–54.

7. Lutchmaya S, Baron-Cohen S. Human sex differences in social and non-social looking preferences, at 12 months of age. Infant Behavior and Development 2002; 25 3: 319–25.

8. Naber F, Swinkels S, Buitelaar JK, Dietz C, Van Daalen E, Bakermans-Kranenburg MJ, Van IJzendoorn MH. Joint attention and attachment in toddlers with autism. Journal of Abnormal Child Psychology 2007; 35 6: 899–911.

9. Bespalova IN, Buxbaum JD. Disease susceptibility genes for autism. Annals of Medicine 2003; 35 4: 274-81.

10. Ronald A, Happé F, Bolton P, Butcher LM, Price TS, Wheelwright S, et al. Genetic heterogeneity between the three components of the autism spectrum: A twin study. Journal of American Academy of Child and Adolescent Psychiatry 2006; 45 6: 691-9.

11. Muhle R, Trentacoste SV, Rapin I. The genetics of autism. Pediatrics 2004; 113 5: 472-86.

12. Schieve LA, Blumberg SJ, Rice C, Visser SN, Boyle C. The relationship between autism and parenting stress. Pediatrics 2007; 119 1: 114–21.

13. Volkmar FR, Chawarska K, Klin A. Autism in infancy and early childhood. Annual Review of Psychology 2005; 56 1: 315–36.

14. Rutter ML, Garmezy N. Developmental psychopathology. In: Hetherington EM, ed. Handbook of child psychology (Vol 4). New York: Wiley; 1983.

15. Kanner L. Autistic disturbance of affective contact. The Nervous Child 1943; 2: 217-50.

16. Swettenham J, Baron-Cohen S, Charman T, Cox A, Baird G, Drew A, Rees L, Wheelwright S. The frequency and distribution of spontaneous attention shifts between social and nonsocial stimuli in autistic, typically developing and nonautistic developmentally delayed infants. Journal of Child Psychology and Psychiatry 1998; 39 5: 747–53.

17. Bowlby J. Attachment and loss : Attachment (Vol 1). New York: Basic; 1969.

18. Ainsworth MDS, Blehar MC, Waters E, Wall S. Patterns of attachment. Hillsdale, NJ: Erlbaum; 1978.

19. Bretherton I, Munholland KA. Internal working models int attachment relationships: A construct revisited In: Cassidy J, Shaver PR, eds. Handbook of attachment : Theory, research and clinical applications. New York: Guilford; 1999.

20. Bowlby J. A secure base: Parent-child attachment and healthy human development. Londres: Basic; 1988.

21. Ammen S, York L. The many measures of attachment. Fresno, CA: California School for Professional Psychology; 1994.

22. Buitelaar J. Attachment and social withdrawal in autism: hypotheses and findings. Behaviour 1995; 132 5-6: 319-50.

23. Dissanayake C, Crossley SA. Autistic children’s responses to separation and reunion with their mothers. Journal of Autism and Developmental Disorders 1997; 27 3: 295-312.

24. Bernabei P, Camaigni L, Levi G. An evaluation of early development in children with autism and pervasive developmental disorders from home movies: Preliminary findings. Autism 1998; 2 3: 243-58.

25. Sigman M, Mundy P. Social attachments in autistic children. Journal of American Academy of Child and Adolescent Psychiatry 1989; 28 1: 74-81.

26. Rutgers AH, Bakermans-Kranenburg MJ, Van IJzendoorn MH, Van Berckelaer-Onnes IA. Autism and attachment: A meta-analytic review. Journal of Child Psychology and Psychiatry 2004; 45 6: 1123-34.

27. Van Berckelaer-Onnes IA. Early childhood autism - A child rearing problem. Lisse: Swets and Zeithlinger; 1983.

28. Rutgers AH, Van IJzendoorn MH, Bakermans-Kranenburg MJ, Swinkels SHN, Van Daalen E, Dietz C, et al. Autism, attachment and parenting: A comparison of children with autism spectrum disorder, mental retardation, language disorder, and non-clinical children. Journal of Abnormal Child Psychology 2007; 35 5: 859-70.

29. Waters E. Attachment behavior Q-set (version 3.0). Stony Brook, NY: State University of New York at Stony Brook; 1987.

30. Martin F, Farnum J. Animal-assisted therapy for children with pervasive developmental disorders. Western Journal of Nursing Research 2002; 24 6: 657-70.

31. Cain, A.O. A study of pets in the family system. In: Katcher AH, Beck AM, eds. New perspectives on our lives with companion animals. Pennsylvania: University of Pennsylvania Press; 1983.

32. Albert A, Bulcroft K. Pets, family and the life course. Journal of Marriage and the Family 1988; 50 2: 543-552.

33. Sable P. Pets, attachment and well-being across the life cycle. Social Work 1995; 40 3: 334-41.

34. Melson GF. Studying children’s attachment to their pets: A conceptual and methological review. Anthrozoös 1989; 4 2: 92-8.

35. Melson GF, Peet S, Sparks C. Children’s attachment to their pets: links to socio-emotional development. Children’s Environments Quarterly 1991; 8 2: 55-65.

36. Redefer LA, Goodman JF. Brief report: Pet-facilitated therapy with autistic children. Journal of Autism and Developmental Disorders 1989; 19 3: 461-7.

37. Moos RH, Moos BS. Family Environment scale manual. Palo Alto, CA: Consulting Psychologists Press; 1986.

38. Waters E, Vaughn BE, Posada G, Kondo-Ikemura K. Caregiving, cultural, and cognitive perspectives on secure-base behavior and working models: New growing points of attachment theory and research. Monographs of the Society for Research in Child Development 1995; 60 No 244): 234–46.

39. McMahon RJ, Lengua LJ. Scale construction for the Coder Impressions Inventory. Fast Track Project technical report. Seattle, Washington: University of Washington; 1996.

40. Dishion TJ, Hogansen J, Winter C, Jabson J. The Coder Impressions Inventory. Eugene, Oregon: Child and Family Center, University of Oregon; 2004.

41. Bigras M, Lafrenière PJ, Abidin RR. Indice de stress parental: manuel francophone en complément à l’édition américaine. Toronto, Canada: Multi Health Systems; 1996.

42. Lacharité C, Éthier L, Piché C. Le stress parental chez les mères d’enfant d’âge préscolaire : validation et normes québécoises pour l’Inventaire de Stress Parental. Santé mentale au Québec 1992 ; 17 2: 183-204.

43. Cohen IL, Schmidt-Lackner S, Romanczyk R, Sudhalter V. The PDD Behavior Inventory: A rating scale for assessing response to intervention in children with PDD. Journal of Autism and Abnormal Development 2003; 33 1: 31-45.

44. Moos RH. Conceptual and empirical approaches to developing family-based assessment procedures: resolving the case of the Family environment scale. Family Process 1990; 29 2: 199-208.

45. Waters HS, Rodrigues-Doolabh L. Manual for decoding secure base narratives. Unpublished manuscript. Stony Brook, NY: State University of New York at Stony Brook; 2004.

46. Waters HS, Waters E. The attachment working model concept: Among other things, we build script-like representation of secure base experiences. Attachment and Human Development 2007; 8 3: 185-97.

47. Vaughn BE, Verissimo M, Coppola G, Bost KK, Shin N, McBride B, Krzysik L, Korth B. Maternal attachment script representations : Longitudinal stability and associations with stylistic features of maternal narratives. Attachment and Human Development 2007; 8 3: 199-208.

48. Vaughn BE, Coppola, G, Verissimo, M, Monteiro, L, Santos, AJ, Posada, G, et al. The quality of maternal secure-base scripts predicts children’s secure-base behaviour at home in three sociocultural groups. International Journal of Behavioral Development 2007; 31 1: 65-76.

49. Poresky RH, Hendrix C, Mosier JE, Samuelson ML. The Companion Animal Bonding Scale: Internal reliability and construct validity. Psychological Reports 1987; 67: 743-6.

50. Poresky RH, Hendrix C, Mosier JE, Samuelson ML. The companion animal semantic differential - long and short form reliability and validity. Educational and Psychological Measurement 1988; 48 1: 255-60.

51. Templer DI, Salter CA, Dickey S, Baldwin R, Veleber DM. The construction of a Pet Attitude Scale. Psychological Record 1981; 31 3: 343-8.

52. Melson GF. Availability of and involvement with pets by children: Determinants and correlates. Anthrozoös 1988; 2 1: 45-52.

53. Johnson TP, Garrity TF, Stallones L. Psychometric evaluation of the Lexington attachment to pets scale (LAPS). Anthrozoos 1992; 5 3: 160-75.

54. Zasloff RL. Measuring attachment to companion animals: A dog is not a cat is not a bird. Animal Behaviour Science 1996; 47 1: 43-8.

55. Serpell, JA. Evidence for an association between pet behaviour and owner attachment levels. Applied Animal Behaviour Science 1996; 47 1-2: 49-60.

56. Odendaal JSJ. Animal-assisted therapy: magic or medicine? Journal of Psychosomatic Research 2000; 49 4: 275-80.

57. Johnson RB, Onwuegbuzie AJ. Mixed methods research: A research paradigm whose time has come. Educational Researcher 2004; 33 7: 14-26.

58. Morgan DL. Paradigms lost and pragmatism regained. Journal of Mixed Methods Research 2007; 1 1: 48-76.