Encyclopedia of Evolutionary Psychological Science

Living Edition
| Editors: Todd K. Shackelford, Viviana A. Weekes-Shackelford

John Bowlby: Pioneer of Attachment Theory

  • Tommie ForslundEmail author
  • Pehr GranqvistEmail author
Living reference work entry
DOI: https://doi.org/10.1007/978-3-319-16999-6_3593-1


Attachment Theory Child Psychiatrist Attachment Figure Attachment Behavior Maternal Deprivation 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Edward John Mostyn Bowlby (1907–1990): A British child psychiatrist and psychoanalyst who founded attachment theory, drawing on evolutionary theory and ethology, cybernetics, and cognitive theory.


John Bowlby, a psychoanalyst and child psychiatrist, sought to reform and modernize psychoanalysis to give it a scientific basis, as he was unsatisfied with parts of its metatheory. Bowlby was particularly concerned with the psychoanalytic explanation of why children develop strong emotional bonds – attachments – to their caregivers, monitoring proximity to the caregiver(s) and showing distress upon separations. Attachment was at the time considered secondary to other processes, such as special forms of psychical energies, which Bowlby argued was unscientific. As outlined below, Bowlby came to draw on ethology, cybernetics, and cognitive psychology and argued that humans and other primates over the course of evolution have developed an attachment behavioral control system with a goal of maintaining proximity to caregivers. This attachment behavioral control system was proposed to have served an evolutionary function of protection from predators and natural dangers, thus increasing survival and reproduction. Bowlby, who himself had experienced an influential separation from a caregiver (i.e., his nanny), also emphasized the importance of actual experiences with caregivers (i.e., quality of caregiving) for children’s psychological development and the need for testable psychological theories and empirical methodology using prospective designs.

Childhood and Education: Development of an Interest in Close Relationships with Caregivers and Effects of Early Separations

In keeping with the proposals of attachment theory – that early experiences with caregivers affect the development of the person and how we come to view ourselves and others – roots to Bowlby’s lifelong interest in the importance of close and enduring relationships between children and their caregivers may partly stem from his own early experiences. Bowlby was born as the fourth of six children in an upper middle class family in London, his father a doctor and surgeon to the royal family. The family situation was typical for its social class of the time, with nanny’s taking care of the household and the children. Bowlby’s interactions with his mother were therefore limited and, apart from holidays, typically restricted to her reading aloud to the children during an hour in the afternoon. It is thus not surprisingly that Bowlby formed his first strong emotional bond to one of his nannies, Minnie. However, Minnie left the family when Bowlby was only 4 years old. Bowlby admitted that this was a painful loss to him, and he likened it to losing a mother figure (Van Dijken 1998). During the First World War, when he was 10 years old, Bowlby was sent to boarding school together with his brother so that they would escape the presumed bombings of London. Bowlby however thought this was just an excuse, and that he would have been sent to the boarding school anyway, as this was the traditional first step towards becoming a gentleman (Holmes 1993). Thus, these early experiences may have affected his interest in the potential negative effects of early separations from caregivers and the importance of children’s experiences with their caregivers.

After boarding school Bowlby went to the navy cadet school to embark on a career as an officer. However, Bowlby soon got tired of this life, partly as he thought it would not give him opportunities to change the world for the better (Van Dijken 1998). Bowlby then studied medicine at Cambridge University, following in the footsteps of his father. He then came in contact with developmental psychology, gave up medicine, and chose to work as a teacher for a year. This experience, which included 6 months at a school for maladjusted children, came to have a marked influence on him, as he observed that many of the maladjusted and antisocial children he was working with had experienced a difficult childhood, including separations from their caregivers and being cared for by multiple substitute caregivers. Through a colleague Bowlby also got in contact with psychoanalysis. Indeed, Bowlby regarded this as the most important 6 months of his career, realizing that psychological problems should be understood at a developmental level and being influenced by experiences with caregivers. After his year as a teacher Bowlby studied psychology at Trinity College and worked with maladjusted children before eventually enrolling at University College Hospital in London to qualify in medicine and pursue a career as a child psychiatrist.

Bowlby and Psychoanalysis: A Complicated and Conflicted Relationship

To become a child psychiatrist he also enrolled himself at the institute for psychoanalysis, as psychoanalysis was a good basis for working with children. However, he first had to train in adult psychiatry, which included going through psychoanalysis himself, which he did with a coworker of Melanie Klein (his supervisor), Joan Riviere, as his analyst. Bowlby agreed with Melanie Klein and her colleagues that children’s first year are important for their subsequent development but disagreed heavily with the psychoanalytical explanations of the processes responsible for children’s development and functioning (see “Psychoanalytic Roots to Attachment Theory”). When discussing attachment theory in relation to psychoanalysis, Bowlby (1980) referenced Rapaport and Gill (1959), who described the psychoanalytical metatheory in terms of postulates regarding five principles (see “Psychodynamic Foundations”). Bowlby specifically questioned the “dynamic” and “economic” principles of the metatheory, which he termed the “psychical energy model.” According to this model humans are endowed with distinct types of psychical energies that builds up and needs discharge in order for homeostasis and that children’s development is affected by the channeling of these energies and infantile fantasies about their caregivers. Bowlby felt that this postulation was unscientific and that it did not sufficiently acknowledge the importance of children’s actual experiences with their caregivers. Indeed, according to these theories attachment was secondary to processes such as feeding and psychical energies. Bowlby thus termed these theories “secondary drive theories.” After being blocked several times Bowlby eventually got accepted into the British Psychoanalytical Association, stubbornly outlining his own ideas in his thesis “The influence of early environment for the development of neurosis and neurotic character” (1940). Interestingly, Bowlby called Klein his anti-thesis and felt that he learned more from the social workers he was working with at a child psychiatrist clinic than he did from his psychiatrist colleagues.

Bowlby was part of a “third-wave” psychoanalytic school, often referred to as the British school of objects relations, and which also included for example Donald Winnicott, Michael Balint, Harry Guntrip, and Ronald Fairbairn. Bowlby was particularly influenced by Donald Winnicott, who shared his view that children come into the world sensitive to social interactions and in need of healthy interactions to develop well. Although the object relation theorists agreed on some things there were also notable differences. A major difference, according to Bowlby, was that the other object relation theories, in comparison to his attachment theory, lacked an alternative to the instincts/drives proposed by Freud. Indeed, Bowlby felt strongly that in order for change to occur there had to be alternatives to Freud’s theory. Of course, the alternative that Bowlby came to suggest was attachment theory. Bowlby was however careful to state that attachment theory is a much more narrow theory than psychoanalysis, pertaining to one behavioral system, and not a theory that could exchange psychoanalysis as a whole.

Given that Bowlby contested some of the core tenets of psychoanalytic theory, that psychoanalysis at the time was rather territorial, and that the scientific methodology Bowlby argued for was largely alien to the psychoanalytical field, it is maybe understandable that Bowlby’s ideas were largely rejected by the psychoanalytic community and that Bowlby was deemed a reductionist that was willing to sacrifice central psychoanalytic concepts for what was observable. Bowlby was vice president in the British Psychoanalytical Association 1956–1961, under Donald Winnicott, but got increasingly ostracized, and after his position as vice president ended he gradually decreased his involvement and settled for passive membership. Bowlby was saddened, and puzzled, that his work in trying to modernize the psychoanalytic theory was rejected, particularly as he thought that attachment theory was in many ways in accordance with some of Freud’s ideas, although formulated in modern scientific terms (see “Psychoanalytical Roots to Attachment Theory”). For example, Bowlby agreed with Freud that we have psychological defenses against anxiety, although he formulated the defenses based on information processing and memory in terms of cognitive-affective strategies for dealing with threatening information (i.e., defensive exclusion and shifting of attention).

Attachment theory, together with the empirical research it has generated, has however become an important part of several modern psychoanalytical approaches and is now often seen as an important empirical foundation of psychoanalysis (Fonagy et al. 2008). An important factor in this rapprochement was the advent of the Adult Attachment Interview and other representational and linguistically based methods used for examining attachment in older children, adolescents, and adult (Main et al. 1985), which were developed based on Bowlby’s concept of internal working models of self and others (see “Offspring-Parent Attachment” and “Psychodynamic Roots to Attachment Theory”).

Scientific Work and Collaborations: Scientific Evidence for the Importance of Close Relationships with Caregivers, Negative Effects of Early Separations, and the Development of Attachment

After the Second World War, Bowlby became head of the Tavistock clinic’s child department, where he remained until his retirement in 1972. Tellingly, he immediately renamed the department “the department for children and parents,” to acknowledge a family perspective on children’s psychological problems. In 1944 Bowlby published a collection of clinical case reports, in which he compared the background of 44 children treated for antisocial behaviors with an equal number of children treated for other behavior problems. Bowlby found that the antisocial children had experienced markedly more losses and separations from their primary caregivers. Negative effects of separations from caregivers were also reported in children who were evacuated from their families in London (i.e., separated) so as not to have to suffer the bombings (Freud and Burlingham 1943). Reinforced by these findings, Bowlby went on to seek more evidence for the untoward effects of separation and the development of attachment.

Bowlby started up various scientific collaborations, most notably with James and Joyce Robertson and Mary Ainsworth. The Robertson couple, inspired by an ethological approach, conducted extensive, videotaped observations of children’s reactions prior to, during, and after being separated from their caregivers in the context of hospital or orphanage care. These now classic observations demonstrated with brutal clarity how the children suffered (e.g., Robertson and Bowlby 1952). Bowlby and the Robertson couple attributed the ill effects to the (temporary) loss of the caregiver(s): “from empirical observation we suggested that the young child’s hunger for his mother’s love and presence is as great as his hunger for food” (Bowlby, p. xiii). The validity of the observations was initially questioned, and the reactions attributed to factors other than the separation from the caregiver(s) per se, such as the strange environment. However, the observations eventually initiated a revolution in child health care in the beginning of the 1950s. In this regard Bowlby was particularly thankful to Winnicott, whose clinical work helped influence policy change.

Mary Ainsworth initially conducted naturalistic observations of children’s attachment behaviors in Uganda where she followed a number of children and their parents during the children’s first year (Ainsworth 1967). This provided Bowlby with invaluable information about the early development of attachment (see “Offspring-Parent Attachment”) and how individual variations in caregiver behavior are related to systematic variations in children’s organization of attachment (see “Individual Variations in Attachment”). As regards the development of attachment, Ainsworth found that most children seemed to develop an attachment during the second half of the first year, whereby they started to direct their attention and behaviors selectively towards their caregivers and monitor their whereabouts, used the caregivers as a reference point for exploration of their environment (i.e., as a secure base), retreating specifically to the caregivers in times of distress (i.e., as a safe haven), and seeking to remain in proximity with their caregivers and protesting separations from them.

The WHO Report and the Seminar Group: Developing a Theory of Attachment Based on Evolutionary Theory and Ethology, Cybernetics, and Cognitive Psychology

In 1950, Bowlby was asked by the World Health Organization to make a report on the available knowledge concerning factors that may affect children’s psychological health, particularly as regards the effects of separations. The report “Maternal Care and Mental health” was published in 1951, and through his work with the report Bowlby learned that other researchers and clinicians shared his ideas about the effects of early separations from caregivers. However, as discussed by Bowlby the WHO report had one major limitation, “whereas it had much to say about the many kinds of ill effects that that evidence shows can be attributed to maternal deprivation… it said very little indeed about the processes whereby these ill effects are brought into being… how does it come about that one or another of the events included under the general heading of maternal deprivation produces this or that form of psychiatric disturbance” (Bowlby 1980, p. xii)? Bowlby thus started turning his attention more towards understanding the nature of the special relationship – “tie” – between children and their caregivers.

After the WHO report Bowlby received funding that he used to start a seminar group, to which he invited participants from various academic disciplines including biology, psychiatry, psychology, and social anthropology. If attachment is not a secondary drive but a primary motivation in its own right, what is it then, and how to explain it in scientific terms? The contact with ethology, through the pioneering control system ethologist Robert Hinde, got Bowlby in contact with the ethological work by, for example, Konrad Lorenz (1963) and Niko Tinbergen (1951). Their work included the phenomenon of imprinting seen in some species, such as birds, which shows some similarities with behaviors in mammals (see “Evolutionary Foundations of the Attachment System and Its Function”). Bowlby realized the potential of explaining human children’s propensities for developing strong and enduring bonds to their caregivers using ethological principles. Birds, although able to find food for themselves nonetheless developed bonds to their parents and followed them wherever they went. Thus, the bond must have served some other evolutionary function than feeding.

Bowlby also got in contact with Harlow’s (e.g., 1958) research with infant rhesus monkeys and surrogate mothers, extending the findings on imprinting in a species more closely related to humans. This became particularly important for Bowlby’s subsequent formulation of the attachment system as a primary motivational system (see “Evolutionary Foundations of the Attachment System and Its Function”). Bowlby visited Harlow, and although he explicitly questioned the methods Harlow used in his experiments with rhesus monkeys on ethical grounds, Harlow’s experiments gave Bowlby invaluable information. Indeed, they clearly indicated that the need for a close and enduring relationship with a caregiver – or attachment figure – is a primary motivation (cf. instinct) and not a secondary result of feeding. Bowlby, drawing from Darwin’s evolutionary theory as well as ethological concepts and research, argued that this “instinct” is a behavioral control system that has evolved and been retained because it has served a particular function in our evolutionary history. This function, according to Bowlby, is protection from natural dangers such as predators, which follows from the predictable outcome of activation of the attachment system, proximity to caregivers. The similarities between humans and many other species regarding attachment bonds could, according to Bowlby, be explained with the notion of convergent evolution, as protection from predators have been common to many species of animals.

Bowlby also drew from cognitive theory and cybernetics in his formulation of the attachment system and its development. Bowlby was for example inspired by Piaget’s (1955) theory about children’s cognitive development, which came to influence Bowlby as regards the importance of cognitive development for the development of attachment during the early years. For example, older children are more able to endure separations due to their more advanced cognitive ability and perspective taking. Young’s (1964) and Craik’s (1943) theories, postulating that animals construct cognitive representations of the world, also influenced Bowlby in his proposal that the child’s experiences with his/hers attachment figures give rise to internal working models (IWMs; see “Offspring-Parent Attachment”) or affective-cognitive representations of self and others. Bowlby used the terms organismic and environmental models for these IWMs, which he argued were complementary. If a child is responded to in such a way that (s)he comes to perceive others as accessible and loving, (s)he will perceive him-/herself as lovable. Thus, when the IWMs are formed they are believed to affect how the child sees him-/herself, and how (s)he behaves in relation to close others, particularly in situations when the attachment system is activated. These models are hence important for predicting future situations with caregivers and important others and for efficient goal-directed behavior whereby the individual organizes his/her behavior based on the IWMs. This also provided Bowlby with an explanation of why children showed individual differences in attachment behavior and why children who had suffered from separations from their caregivers, and/or dysfunctional caregiving (i.e., had become insecurely attached), many years later could come to show difficulties in developing close relationships, be described as affectionless and cold, or display antisocial behavior problems (see “Individual Variations in Attachment” and “Outcomes of Individual Variations in Attachment”).

Bowlby published the main tenets of attachment theory in three articles, “The nature of a child’s ties to his mother” (1958), “separation anxiety” (1960a), and “Grief and morning in infancy an early childhood” (1960b). These articles, although initially largely ignored within psychoanalysis, were well received by researchers within developmental psychology, and Bowlby became a member of the medical research council in England. Bowlby thus got more time to think and write and subsequently published his classic book trilogy – attachment and loss – (1969, 1973, 1980), where he outlined the tenets of attachment theory in more detail, expanding on the three articles he had previously published.

Research Immediately Following in the Footsteps of Bowlby: Individual Differences in Attachment

Ainsworth (1967) had noted that the Ugandan children seemed to develop different strategies for maintaining proximity to their caregivers and that this seemed to depend on how the caregivers responded to the children’s signals. Seeking to replicate her Uganda findings in America, Ainsworth later devised a semi-structured laboratory observation, for which she methodologically drew on the ethologically inspired separation and reunion observations that the Robertson couple and Bowlby had used (e.g., Ainsworth et al. 1978). This laboratory observation, which induces slightly increasing levels of stress by means of a strange environment, a stranger, and two short separations from the caregiver, was deemed necessary to be able to observe children’s varying strategies for organizing their attachment behaviors in relation to their caregiver. Indeed, whereas the mere presence of Ainsworth in Uganda, where white women were uncommon, had triggered the attachment system of Ugandan children, this was not the case in American children. The laboratory observation, which is called the strange situation procedure (SSP), is still today the most widely used method for observing children’s organization of attachment to their caregivers.

Briefly (see “Individual Variations in Attachment” for a thorough discussion), Ainsworth and her colleagues’ research indicated that children whose caregivers noticed the infants signals and responded timely and sensitively (particularly during distress, i.e., when the attachment system was activated) became securely attached to their caregivers, as indicated by a balance between the attachment system and an exploratory system. That is, secure children are able to use the caregiver(s) as a secure base from which to explore the environment when there are no signals of danger and to use their caregivers as a safe haven to return to for comfort and protection when distressed. Ainsworth and colleagues also identified two patterns of insecure attachment, insecure-avoidant and insecure-ambivalent/resistant, who do not show the same ability to use the caregiver as a safe haven and secure base, and which respective organization of behavior in relation to their caregiver could be predicted by specific patterns of caregiver insensitivity. Although Bowlby had argued from the start that the quality of caregiving is crucial to variations in attachment, he lacked an empirical method for examining caregiving in relation to attachment and hence focused on the effects of separation which were readily examined during the aftermath of the Second World War. He was therefore hugely grateful to Ainsworth for her contribution, and to acknowledge her Bowlby named his last book about attachment after one of her concepts, “A secure base” (Bowlby 1988).

Mary Main, a PhD student of Mary Ainsworth, later identified a forth quality of attachment, disorganized/disoriented attachment, when seeking to understand the inexplicable and contradictory behaviors shown by some children who were therefore unclassifiable using Ainsworth’s established patterns (Main and Solomon 1990; see “Individual Variations in Attachment” and “Disorganized Attachment and Reactive Attachment Disorder” for a more thorough discussion). These children, she found, had one thing in common: a (momentary) breakdown in the organization of attachment behaviors in relation to the caregiver. As opposed to secure and organized insecure children (i.e., avoidant and ambivalent), these children could not maintain an organized strategy in relation to their caregivers in the strange situation. Main argued that a cause of this breakdown in behavior is that the attachment figure (i.e., source of safety) is simultaneously represented as a cause of alarm. She termed this “fright without solution,” an unsolvable behavioral paradox of simultaneously seeking to approach and avoid or escape the caregiver. Bowlby’s (1980) notion of “segregated systems” provided important guiding principles for Main’s theory of the development of disorganized attachment, which was in turn received favorably by Bowlby (Duschinsky 2015). Main and colleagues argued that disorganized-disoriented attachment is most consistently predicted by frightening/frightened behaviors on behalf of the attachment figure (e.g., Main and Hesse 1990).


Bowlby sought to develop a scientifically based theory for the phenomenon of attachment between children and their caregivers, to reform the parts of the psychoanalytical metatheory that he could not accept. In so doing Bowlby drew on his own work as a child psychiatrist, his scientific collaborations and the information he gathered from the seminar group he started and which included participants from various scientific disciplines, most notably ethology and cognitive psychology. A notable feature of Bowlby’s work was thus an openness to draw on and integrate available knowledge from various scientific disciplines. Bowlby concluded that humans, as some other animals, are endowed with an attachment behavioral control system, which functions to seek and maintain proximity to (usually) protective caregivers and which thereby increases the chances of survival and reproduction.



  1. Ainsworth, M. D. S. (1967). Infancy in Uganda: Infant care and the growth of love. Baltimore: The Johns Hopkins University Press.Google Scholar
  2. Ainsworth, M. D. S., Blehar, M. C., Waters, E., & Wall, S. (1978). Patterns of attachment: A psychological study of the strange situation. Hillsdale: Erlbaum.Google Scholar
  3. Bowlby, J. (1940). The influence of early environment in the development of neuroses and neurotic character. International Journal of Psychoanalysis, 21, 154–178.Google Scholar
  4. Bowlby, J. (1944). Forty-four juvenile thieves: Their characters and home life (I &II). International Journal of Psychoanalysis, 25, 154–178 & 107–127.Google Scholar
  5. Bowlby, J. (1951). Maternal care and mental health. Genéve: WHO.Google Scholar
  6. Bowlby, J. (1958). The nature of a child’s tie to his mother. The International Journal of Psycho-Analysis, 39, 350–373.PubMedGoogle Scholar
  7. Bowlby, J. (1960a). Separation anxiety. The International Journal of Psycho-Analysis, 41, 89–113.PubMedGoogle Scholar
  8. Bowlby, J. (1960b). Grief and mourning in infancy and early childhood. The Psychoanalytical Study of the Child, 15, 9–52.Google Scholar
  9. Bowlby, J. (1969). Attachment and loss: vol 1: Attachment. New York: Basic Books.Google Scholar
  10. Bowlby, J. (1973). Attachment and loss: vol 2: Separation. New York: Basic Books.Google Scholar
  11. Bowlby, J. (1980). Attachment and loss: vol 3: Loss, sadness & depression. New York: Basic Books.Google Scholar
  12. Bowlby, J. (1988). A secure base: Clinical applications of attachment theory. London: Routledge.Google Scholar
  13. Craik, K. (1943). The nature of explanation. Cambridge: Cambridge University Press.Google Scholar
  14. Darwin, C. (1859). The origin of species by means of natural selection, or, The preservation of favored races in the struggle for life. London: John MurrayGoogle Scholar
  15. Duschinsky, R. (2015). The emergence of the disorganized/disoriented (D) attachment classification, 1979–1982. History of Psychology, 18(1), 32–46.CrossRefPubMedPubMedCentralGoogle Scholar
  16. Fonagy, P., Gergely, G., & Target, M. (2008). Psychoanalytical constructs and attachment theory and research. In J. Cassidy & P. Shaver (Eds.), Handbook of attachment: Theory, research and clinical applications (pp. 783–810). New York: Guilford Press.Google Scholar
  17. Freud, A., & Burlingham, D. (1943). War and children. New York: Medical War Books.Google Scholar
  18. Harlow, H. (1958). The nature of love. American Psychologist, 13, 673–685.CrossRefGoogle Scholar
  19. Holmes, J. (1993). John Bowlby & attachment theory. London: Routledge.Google Scholar
  20. Lorenz, K. (1963). King Solomon’s ring: New light on animal ways. London: Routledge.Google Scholar
  21. Main, M., & Hesse, E. (1990). Parents’ unresolved traumatic experiences are related to infant disorganized attachment status: Is frightened and/or frightening parental behavior the linking mechanism? In M. T. Greenberg, D. Cicchetti, & E. M. Cummings (Eds.), Attachment in the preschool years: Theory, research, and intervention. The John D. and Catherine T. MacArthur Foundation series on mental health and development (pp. 161–182). Chicago: University of Chicago Press. xix, 507 pp.Google Scholar
  22. Main, M., & Solomon, J. (1990). Procedures for identifying infants as disorganized/disoriented during the Ainsworth strange situation. In M. T. Greenberg, D. Cicchetti, & E. M. Cummings (Eds.), Attachment in the preschool years: Theory, research, and intervention (pp. 121–160). Chicago: The University of Chicago Press.Google Scholar
  23. Main, M., Kaplan, N., & Cassidy, J. (1985). Security in infancy, childhood and adulthood: A move to the level of representation. In I Bretherton & E. Waters (Eds.), Growing points of attachment theory and research Chicago: (Monographs of the Society for Research in Child Development, Vol. 50 (1/2), pp. 66–104).Google Scholar
  24. Paulhus, D. L., Trapnell, P. D., & Chen, D. (1999). Birth order effects on personality and achievement within families. Psychological Science, 10(6), 482–488.Google Scholar
  25. Piaget, J. (1955). The child’s construction of reality. London: Routledge & Kagan Paul.Google Scholar
  26. Rapaport, D., & Gill, M. M. (1959). The points of view and assumptions of meta-psychology. The International Journal of Psycho-Analysis, 40, 153–162.PubMedGoogle Scholar
  27. Robertson, J., & Bowlby, J. (1952). Responses of young children to separation from their mothers. In J. Bowlby (1997). Attachment and loss: Vol. 1, attachment. London: Pimlico.Google Scholar
  28. Tinbergen, N. (1951). The study of instincts. Oxford: Clarendon.Google Scholar
  29. Van Dijken, S. (1998). John Bowlby – His early life: A biographical journey into the roots of attachment theory. London: Free Association.Google Scholar
  30. Young, J. Z. (1964). A model of the Brain. London: Oxford University Press.Google Scholar

Copyright information

© Springer International Publishing Switzerland 2016

Authors and Affiliations

  1. 1.University of UppsalaUppsalaSweden
  2. 2.Stockholm UniversityStockholmSweden