Children’s perceptions of their imaginary companions and the purposes they serve: An exploratory study in the United Kingdom

Abstract

Imaginary companions of school age children are a relatively unexplored phenomenon. This article reports on a qualitative study carried out in the UK investigating British children’s perceptions of their imaginary companions. Eight children aged between 5 and 11 years were interviewed. Interpretative Phenomenological Analysis was used to explore individual and cross-case themes. It was found that imaginary companions provided friendship, playmates and entertainment, enabling children to overcome times of boredom and loneliness. They appeared to provide a level of wish fulfilment for some children. Children’s interactions with their imaginary companions provided support when there were problem situations in the children’s lives.

Keywords

Imaginary companions imaginary friends Interpretative Phenomenological Analysis qualitative research

Introduction

Parents and child practitioners sometimes express concern about whether the imaginary companion is beneficial, or a detrimental phenomenon to be discouraged (see, for example, parent information websites: Brott, 2004; Gurain, 2004; Heins, 2004). In the few studies that have investigated parent perspectives of imaginary companions, there have been varied findings. Manosevitz et al. (1973) found from a parent questionnaire analysis that 62% of parents (63 parents, mainly mothers) said that the imaginary companion was beneficial for their child. Brooks and Knowles (1982), in an analysis of questionnaires of mothers and fathers, report less positive views and found that a ‘substantial proportion’ of parents would discourage play with imaginary companions. Gleason (2005) found differences between mothers and fathers in the perceived benefits of their child’s imaginary companions and how positively or not they viewed pretend play. It could be argued that these differences in perspective may reflect the lack of knowledge generally about the phenomenon of imaginary companions.

Early imaginary companion research drew on clinical case studies (e.g. Bender and Vogel, 1941; Nagera, 1969). Whilst children with emotional and/or psychological problems were not usually referred to clinicians because they had imaginary companions, it was the curiosity of clinicians such as Bender and Vogel and Nagera which led to them researching their role and purpose when they emerged as phenomena during clinical casework.

Developmental research on imaginary companions has utilised primarily quantitative or mixed methods and has been undertaken with children typically aged 3–5 years in early childhood (e.g. Carlson and Taylor, 2005). Gleason (2002), Gleason et al. (2000), Taylor (1999), Pearson et al. (2001) and Mauro (1991) have researched imaginary companions mostly of young children from normative samples and have found them to be a common occurrence. The need for the exploration of the nature and purposes of imaginary companions in normative samples has been identified by developmental psychologists (Gleason et al., 2000).

Recent studies show that imaginary companions are more common in middle and/or late childhood (6–9 years, 10–12 years respectively) than previously thought, albeit in a more private form (e.g. Hoff, 2004–2005, 2005; Pearson et al., 2001; Taylor et al., 2010). This discovery has opened up new possibilities for researching the phenomenon and developing our understanding. It could be argued that school aged children are more able to tell us about their experiences of their imaginary companions and to express their views about them. This was a primary objective of the research.

Definitions and reported incidence of imaginary companions and related phenomena

Early research defined the imaginary companion as an invisible friend that the child talks to or plays with and refers to when communicating with others. The imaginary companion appears real to the child and the child interacts with the imaginary companion for an extended period – at least several months (Svendsen, 1934).

Subsequently, children having a similar relationship with a special toy began to be included in the research. This category of imaginary companion is referred to in the research as a ‘personified object’ and the process has been termed ‘personification’ (Harris, 2000). The current study uses this inclusive definition of imaginary companion in line with most contemporary research.

In the United States, 152 children aged 3–4 years and their parents were interviewed about imaginary companions. In a follow-up study three years later of 100 of these children and their parents, Taylor et al. (2004) found that by the age of 7 years, 65% of these children had had an imaginary companion. Having an imaginary companion at 6–7 years of age was at least as likely as for the 3–4 year age range. In the UK, Pearson et al. (2001) asked a random sample of approximately 1800 children aged between 5 and 12 years whether they had current or past imaginary companions. They found that 46.2% reported having, or having had an imaginary companion. Of children aged 5–9 years, 33 –43% reported current imaginary companions. There was a noticeable decline in reported incidence with age: 19% and 9% at ages 10 and 12 years respectively. As Pearson et al. (2001) comment, this may be an underestimate. They noted (as does Hoff, 2004–2005) that some of the older children were reluctant to answer the questions about their imaginary companion, with some letting the researcher know only at a later point that they did have an imaginary companion. These findings suggest that parental and cultural expectations may have an influence on the reporting and prevalence of imaginary companions in older children.

Exploration of purposes served by imaginary companions

Harris (2000) is a British developmental psychologist whose pioneering work on imagination challenged prevailing theories. Harris acknowledges Piaget’s work on pretend (symbolic) play (e.g. Piaget, 1962 [1923]), though disagrees with Piaget’s theory that pretend play is an immature period, lacking adaptation and is ‘outgrown’ as the child develops. In contrast, Harris theorises that through sustained role play, children imagine different possibilities which ultimately lead to a developed concept of reality. Harris comments that in the process of role play, the child is enacting how characters might behave in novel situations that the child might not have previously encountered. According to this theory, the child constructs a ‘suitable plan’ drawing from but not confined to knowledge of the real world. He points out that it is when children are not able to engage in sustained role play such as some children on the autistic spectrum, that children have difficulties in social understanding, communication and cognitive development. Harris conceptualises ‘sustained role play’ as a high level form of imaginary activity in childhood, influencing cognitive, social and emotional development. He categorises imaginary companions, alongside personification and impersonation (where a child takes on the identity of another person or character for an extended period) as evidence of ‘sustained role play’.

Clinician researchers have mostly drawn upon psychodynamic theory in their studies of children from clinical and sometimes normative samples (e.g. Benson and Pryor, 1973). Analysing clinical cases, Nagera (1969) maintained that imaginary companions served a range of positive purposes relating to ego development and conflict resolution for the child which can prevent the development of clinical mental health problems. Thus a case is described where a girl undergoing a stressful family situation used an imaginary companion to ‘avoid regression and symptom formation’ (Nagera, 1969: 189). Sugarman and Jaffe (1989) and Seiffge-Krenke (1993, 1997), drawing on Winnicott’s (1971) work, have conceptualised the imaginary companion as a form of transitional phenomenon serving important developmental and psychological purposes. According to this theory, transitional phenomena enable the child to develop a sense of self, cope with frustration when needs are not immediately met, and cope with anxiety in an environment not controlled by the child.

In Germany, Seiffge-Krenke (1993, 1997) investigated both a ‘deficit hypothesis’ and a ‘coping hypothesis’ in two studies of 94 adolescents (80 females and 14 males) who wrote to their imaginary companions in their diaries. Diaries underwent content analysis, and participants’ relationships with real friends were compared to their relationships with their imaginary companions. Participants completed questionnaires which included measures of self-concept and coping behaviour. Seiffge-Krenke concluded that both real friends and imaginary companions provided support and validation influencing personal growth and coping in a positive way. Adolescents in this study had imaginary companions in addition to, rather than instead of, real friendships. Seiffge-Krenke uses this as evidence against a deficit hypothesis.

Taylor et al. (2010) carried out a six-year longitudinal study of 12-year-olds identified as being at high risk of developing problem behaviour. Of the 152 children, 13 reported current imaginary companions. These children were reported as having developed better coping strategies, but had fewer positive peer nominations and more externalizing behaviour. When followed up at age 18, adolescents from the imaginary companion group were more likely to have achieved a positive outcome as evaluated by diagnostic interview, school and court records, drug use, psychiatric diagnosis and school graduation. The authors consider that having an imaginary companion may directly enhance resilience, being a coping strategy for the young person.

Hoff (2004–2005) carried out interviews with 26 10-year-olds from a normative sample in Sweden. Twelve reported previous imaginary companions and 14 current imaginary companions. Hoff asserted that overall imaginary companions served the purpose of identity formation by becoming ‘inner mentors’. Interview analysis revealed a wide range of purposes served, which were grouped into five main categories: comfort or substitute for company, motivation and self-regulation, self-esteem enhancement, extended personality and life quality enhancement. Hoff gives examples of extended personality: children might have an imaginary companion of the opposite sex which enabled the child to explore aspects of personality transcending gender stereotyped roles. Another example is a child had imaginary companions who had opposite characteristics. Hoff suggests that this enabled the child to experiment with possible selves. Hoff also comments here that imaginary companions can serve the purpose of a transitional object, and that the contrasting imaginary companions can lead a child to being able to integrate positive and negative aspects of themselves. Hoff describes life quality enhancement as where children go on to develop imaginary worlds (termed paracosms by Cohen and Mackeith, 1991) which enabled them to express creativity and ‘provides life with some extra color’ (Hoff, 2004–2005: 171).

There is consensus that imaginary companions serve a range of purposes, whether these are developmental or reactive. Commonly, theoretical orientations conceptualise imaginary companions alongside other imaginary phenomena as facilitating an understanding of reality, supporting cognitive, emotional and social development, enabling children to respond to events in their lives, facilitating a developing sense of self and providing friendship and company. It was anticipated that an in-depth exploration of children’s accounts of their imaginary companions and their interactions with them would further inform our understanding of the phenomenon.

Method

This research aimed to explore children’s perceptions of their imaginary companions in order to investigate purposes as the children themselves reveal them in semi-structured interviews. Interpretative Phenomenological Analysis (IPA) was applied. IPA aims to interpret the individual subjective experience (Smith et al., 1999) – though does not exclude noting commonalities. It has been common practice of researchers that if the child has more than one imaginary companion, to ask about the first or most important. In contrast, this study set out to explore imaginary companions that a child has had over time.

The procedure for obtaining consent and protecting the rights and well-being of participants was informed by ethical guidelines (British Psychological Society, 2004). Ethical approval for the research was obtained from the Institute of Education, University of London.

A pilot study was completed. Four of the six children interviewed had previous imaginary companions and demonstrated only partial recall of them. As a result, child participants in the main study were included only if they had one or more current imaginary companions. One child with current imaginary companions was included in the main study.

Main study

Research participants were recruited through two sources. First, following an interview about imaginary companions on British national radio, parents were invited to contact me. From the 50 parent responses, four children with current imaginary companions were identified who agreed to participate. The children’s ages ranged from 5 to 10 years. They were homogeneous in terms of age and having current imaginary companions for several years known to parents. Second, a short questionnaire was given to children in two classes of 10- to 11-year-olds (except where parental consent had not been given). This asked children about activities they liked to do at school and at home, previous imaginary friends and whether they had current imaginary friends. Children were asked to give a description of their imaginary friends and to indicate whether they were willing to be interviewed. Three interviews from girls aged 11 years were included in the study. The cumulative sample consisted of eight children, five girls and three boys aged between 5 and 11 years.

All children were given explanations about the research at the interview. They were reminded that they could stop the interview at any time and that they could pass over any question they did not wish to respond to. Children were given assurances about confidentiality. Substitute names have been used in this article.

Data collection

The semi-structured interview is the primary research method used in IPA (Smith and Eatough, 2006). Each child was asked to describe current and previous imaginary friends, in terms of name, whether they were completely invisible or based on a toy, physical characteristics, personality, and where they lived. The children were then asked about what they did or talked about with their imaginary friends and whether they had any likes and dislikes about them. They were also asked whether and why their imaginary friends were special and who knew about them. Parents of the younger children were asked to provide brief information, e.g. about siblings, events in their child’s life and how long their child had had their imaginary companions.

Analysis

Each interview was taped and transcribed. Procedures for IPA as described by Smith et al. (1999) were followed. Interviews were analysed individually, and a series of steps taken to reread each transcript, and identify and refine themes. A summary table of main themes and subthemes was derived from each interview transcript, then integrated to form a main summary table. A colleague conversant with IPA provided feedback on interview transcripts and analyses to check that the categorisation of themes was plausible. This strategy was used to provide a ‘credibility check’, a guideline recommended by Elliott et al. (1999) for evaluating qualitative research.

This article presents the main themes from the study: characteristics of imaginary companions, independence of imaginary companions, public and private dimensions, children’s feelings about their imaginary companions and their perceptions of purposes served. The last of these themes is given prominence here, with a more detailed reporting of subthemes, as it was the primary research question. For a full report of all themes see Majors (2009).

Characteristics of the children’s imaginary companions

An overview of these characteristics is given in Table 1 in the Appendix.

The imaginary companions were diverse in terms of being animal or human, had a broad range of characteristics and varied in terms of gender, age and temperament. Seven of the children interviewed had had more than one imaginary companion. All children had imaginary companions from the age of 3 or 4 years, a substantial part of their lives.

Antonio had three imaginary companions who were brothers. Harry’s Barnaby Bear and Manager were based on television characters. Lisa’s Minty was partly based on a toy, but was also an imaginary companion. Invisible Amy, Katie and Megan were smaller versions of her real friends. John had also created an imaginary world known as ‘Pinwave’. Carmel had a family of imaginary companions. Holly had her previous imaginary companions when she was 4 years old. Tara’s horses were based on horses she met on holiday, the imaginary companions she had when younger were based on real people around her at the time. Ella’s Polly was described as the spirit of her grandmother, who emerged when Ella’s grandmother died when Ella was 4 years old.

Independent imaginary companions

Imaginary companions were reported as having lives away from the child and showed independence of will, in one case an imaginary companion had their own imaginary companion – this may serve to increase the feeling that the imaginary companion is real, despite the child knowing that they are imaginary.

Some imaginary companions were not always compliant and some showed unfriendly behaviour. For two girls, their imaginary horses’ non-compliance seemed to increase their interest in them and made games played with the imaginary companions more enjoyable. Holly:

The thing is – I don’t want a perfect horse – it just makes it annoying kind of … and when I’m riding I don’t want her to do exactly what I say …

Public and private dimensions

Imaginary companions of older children were more private, and sometimes a secret with a close friend. None of the current imaginary companions were known to parents of the three eldest children, Ella, Holly and Tara. Holly, when asked how her parents might respond if they knew, said:

Ummm, I don’t know. I think they’d be supportive, but I don’t know.

In one case the imaginary friend was not known to anybody else. Ella was uncertain of friends’ responses and acknowledged that others might perceive her as strange:

So I haven’t told anyone just in case.

Older children mentioned seeking privacy to interact with their imaginary companions. Carmel:

I would wait until I am alone and at home.

Concealment appeared to reflect actual and perceived parent and peer responses. Carmel let me know that classmates had teased her at an earlier age when her imaginary companions had been revealed. Social and cultural factors may influence the existence of imaginary companions, particularly in older childhood, and how far they are public or private.

Child’s feelings about imaginary companions

Imaginary friend is important/needed

Children spoke of enjoyable interactions with their imaginary companions with all mentioning play as well as talking to their companions. Children reported that time spent with their imaginary companions was entertaining and pleasurable. Interactions with imaginary companions sometimes related to events in the child’s life, for example riding horses, fairground attractions, family holidays and trips, moving house, illness and finding something difficult. All children reported that their imaginary companions were important. Carmel:

Yes, they’re like family. I don’t know what I could do without them if they weren’t there.

Carmel, Antonio and Harry spoke of their ‘need’ for their imaginary companions. Carmel said that when she was feeling lonely, finding it difficult to get to sleep or had no one to play with, she called upon her imaginary friends and that ‘when I need them, they’re always there’. I asked Antonio how he would feel if Britten was not there. Antonio:

I would feel sad … I need Britten … it’s lucky I’ve got Britten here.

Imaginary companion is similar to child

Seven children talked about things they had in common with some of their imaginary companions. John commented that imaginary friend Tom liked game-boy games ‘same as me and my brother’. Harry reported that:

I and Ducky have got flat feet and that’s something in common.

Ella shares several similar characteristics with Polly:

Well she’s the same height as me and erm she is the same hair colour as me – because I’m very different, ’cos in my family they all have blond hair and black brown hair – so she has the same coloured hair as me and her personality is that she’s very nice and kind.

Having things in common with the imaginary companion may make it easier to relate to, and identify with the imaginary companion. The child may imagine that as they have things in common, their imaginary friend does understand what they are talking about.

Child perceptions of purposes served

Overcome boredom/loneliness/provide entertainment

Six children spoke of how their imaginary companions prevented them from boredom and loneliness. Ella:

Sometimes she makes me laugh and if I get bored at home – ’cos I’ve only got little siblings.

Imaginary companion is good friend/playmate/there to talk to

All children expressed how important their imaginary companions were in providing companionship/friendship. Lisa commented that Minty ‘always’ played with her and kept her company. It appeared that Carmel valued being able to confide in her imaginary companions, having companions that were always there for her, played with her and who would take her away from her problems:

… I would tell them if something had gone wrong but then all my emotions would just slip away and I would focus on that thing that I was talking to them about or playing with them.

In contrast to some of her real life friendships, the most liked characteristic of Tara’s current imaginary horse Fantazia is that she was ‘really, really friendly’ and this is mentioned on several occasions. This suggests that the imaginary horses served the purpose of supplementary friends that could be trusted and were constant:

I can trust them … Like if I tell them a secret it’s not like they can tell anyone … ’Cos they’re horses – so like they can keep secrets …

These features are not always fully available from friendships in children’s lives. Friends cannot always be there when needed. Also, however much a real best friend is trusted there may well be a need to censure what is divulged, and in the real world there is always a risk that some secrets may be divulged to others.

At other times the imaginary companions’ occasional unfriendly behaviour appeared to serve the purpose of enabling the child to express and deal with angry or upset feelings.

To express and release angry/upset feelings

Three of the older children commented that contact with their imaginary companions helped defuse angry or upset feelings. John gave an example of how Tom met his needs:

John:

I’ll go upstairs, shut the door and say pretend Tom’s a pillow and sometimes I do this at Daddy’s when people are around, when I’m angry I’ll make the pillow jump …

Interviewer:

How does that make you feel when you do that?

John:

It makes letting my anger off but also I’m also playing a bit of a game.

John related another similar example commenting that:

It takes my anger out without actually hurting someone.

John let me know that despite these interactions, Tom was still a friend:

Tom’s not an imaginary human punch bag. He’s friends.

John identified himself with Tom – they both have the same likes and dislikes, e.g. both Tom and John disliked ‘Brussels sprouts and tea’ and both liked coffee. Tom has a fiery temper (as well as being a friend) – perhaps a projection of John’s hidden angry feelings. John uses his imaginary companion to release anger. Significantly, this anger and upset is not apparent to others. Nina, John’s mother, described him as friendly and easy going.

Support for problem situations in child’s life

Six children gave examples of how they gained support from their imaginary companions when there were difficulties in their lives. Harry’s Ducky seems to have been consciously created when Harry needed support in swimming sessions:

I was swimming a width and then I thought I would do another one with my imaginary friend and I did and I needed it to be a swimming one so I chose a duck.

Harry (and his mother) reported that swimming was initially problematic for him. Harry reported that Ducky ‘helps me swim’ and asked to explain how, he said:

When he tells me lots of things that have nothing to do with me even, and then that’s the thing that helps me do it.

Harry reports that Ducky is still around but mostly working at another pool. This suggests that imaginary companions can help children overcome problems in their lives.

Ella told me that imaginary friend Polly was very important, serving the purpose of having someone to talk to when things went wrong:

… maybe I’m not in very comfortable position talking to my maybe my parents; I can talk to her. And then I’m very comfortable with talking with her. Maybe I can’t talk to my friends … If I want to say something to my friends I can say it to her, and then maybe she would understand how I am feeling and that’s what’s good about her.

Talking through her problems in this relaxed state with Polly seemed to help Ella to gain a better understanding of the problem and what to do about it. In some ways here, Polly appeared to act like an ‘inner mentor’ (Hoff, 2004–2005):

Well she always talks to me and if I’ve done something wrong … she kind of tells me ‘oh you’ve done this wrong because or this is how you can sort it out, or don’t feel bad about it’ – so she’s really quite special to me.

Ella finds reassurance from the interaction and is able to rid herself of ‘bad feelings inside of me’.

Ella and Holly both mentioned friendship difficulties at school and found contact with their imaginary companions at these times supportive. Holly:

I have a lot of problems sometimes in school because my friend can be really mean to me a lot of the time (sighs) and sometimes I talk to my mum because she’s a psychologist and sometimes I talk to her. But you can’t always count on mum so that’s why I talk to Dream sometimes.

The preference for talking to the imaginary companion rather than the parent when there are problems may reflect a developmental shift. Whereas younger children might indeed involve their parent(s) when there are difficulties, older children are becoming more selective about topics they feel comfortable talking to their parents about. This may be due to a developing social/emotional maturity and a consequent need for privacy. Another indication that imaginary companions were a source of support at times of difficulty was when I asked Tara how she would feel if she didn’t have her imaginary companions. She commented:

I don’t think I’d really mind, because I wouldn’t know that they ever really existed. So it wouldn’t make much difference to me. I mean obviously when I was like, I mean that if the house was pitch black and I was going out every night or something, then obviously I might get a bit more scared. But um but if I didn’t have these imaginary horses then I’d probably have um other ones.

When I asked Tara how her imaginary horses helped her when she was scared at night she replied:

’Cos it’s someone to be with you in the dark.

There is an element of contradiction here, with Tara appearing to claim that her current imaginary companions are replaceable, suggesting that they were less important and also suggesting the power and control she has over them. Simultaneously, Tara’s vulnerabilities are also exposed; her child-like fears of the dark provides the driving force and need to conjure up other imaginary companions as a coping strategy to reduce anxiety.

In terms of other purposes served, Ella articulated how Polly helped her overcome shyness and developed her confidence:

I’d probably feel like very shy, ’cos before when I was like 3 years old, I wouldn’t talk to anyone and when I got my imaginary friend, I got, I built up my confidence and if she wasn’t there I’d probably be quite shy now.

Wish fulfilment

Some imaginary friends appeared to served the purpose of wish fulfilment. Thus, Holly comments on her previous imaginary companion: twin, Lily:

… I already had my imaginary twin because I didn’t [have] an imaginary friend – I mean I thought – I already have friends – why do I need an imaginary friend, so I thought imaginary twin – I always wanted a twin … do everything with them … be exactly like you and so it was kind of something I could have.

Tara, who had imaginary horses Fantazia and Tom, is emphatic about how much she would like to have her own animal:

But if you like um don’t have an animal then it’s like you can see why and you want an animal. Like I’m desperate to get an animal.

She suggests in the quote above, that her imaginary horses are a substitute for the real horses or animals that she would love to have. Antonio’s imaginary companions are three brothers. Antonio’s parents commented that it was very much his desire to have brothers.

Discussion

Some imaginary companions were independent characters with their own sense of will and individual personality characteristics which differed to those of the child. It appears that these differences enabled the child to have experiences (albeit in imaginary form) that contrasted to their real lives. This was most evident with regard to wish fulfilment for a sibling or pet. It also allowed exploration of different aspects of personality. (Thus John who was characteristically friendly and even-tempered, in his interactions with imaginary friend Tom was able to experience and release angry feelings in a way that was beneficial to him and I would argue preferable to bottling up his anger.) Hoff (2004–2005) also found similar examples of personality exploration.

Seven children in this research mentioned having similar characteristics to at least one of their imaginary companions. Gleason (2004) reported likewise on research carried out by Mauro (1991). Identifying with the imaginary companion in this way might serve different though related purposes. In western culture and society, common experiences and perceived similarities can facilitate trust and friendships (Dunn, 2004; Roffey et al., 1994). Therefore, a child’s perceived similarities with their imaginary companion may be a way of developing a connection and friendship with them and help them to feel that their imaginary companion understands them. Most children valued their imaginary companions and said that they were important to them. Identifying similarities with their valued imaginary companion may serve to enhance their own self-esteem, particularly if the imaginary companion has admired qualities to which a child might aspire. It may help them to consider possible selves (Hoff, 2004–2005). All children identified friend and playmate as a main purpose being served. This has been found in other research (e.g. Gleason et al., 2000; Kalyan-Masih, 1986; Manosevitz et al., 1973). All children highlighted the friendship quality of their imaginary companions. Mostly, they were good friends, liked playing with the child, and were, in some cases, someone in whom to confide and share secrets.

My analysis revealed that children’s interactions with their imaginary companions sometimes related to events or problems in their lives, and appeared to help them to express feelings and explore personal issues which seemed to bring relief. In some cases, imaginary companions provided a distraction from upsets with others. Hoff (2004–2005) reports that she found it ‘remarkable’ that despite children being aware that the imaginary companions were not real they still perceived that they provided support to enable them to meet various goals. She questions whether the presence of imaginary companions indicates that young children have been given too much responsibility, or are they, alternatively, part of normal development, where they can be seen as transitional objects, promoting growth and development. Findings of the current study in the light of the work of Taylor et al. (2004) and Pearson et al. (2001) indicate that imaginary companions are often a normal developmental phenomenon and can be considered beneficial (Harris, 2000). The children in the study spoke positively about their imaginary companions. They valued the entertainment, game playing and friendship elements of interactions with the imaginary companions and not just the supportive elements. Both Harris (2000) and Winnicott (1971) note that it is when children are not able to engage in imaginative play that there should be concern. In my view, the emphasis given to play by all the children including the older ones in this research is because play is a developmental phenomenon with ‘adaptive functions’ in terms of learning, cooperation and socialisation (Bjorklund, 2007). It is possible that children’s play with imaginary companions helped them to achieve some of the developmental tasks of middle childhood (Majors, 2012). These include the ability for sustained play and games with other children, learning to initiate and sustain friendships with others and coping with the making and breaking of friendships that can be typical in middle childhood. Children in this study enjoyed play and games with their imaginary companions, sometimes involving others, and used interactions with imaginary companions to cope with real life friendship difficulties, which in turn supported the maintenance of friendships in the real world.

It is acknowledged that children with imaginary companions are not a homogeneous group and that children with psychological difficulties who have faced trauma and have imaginary companions may have developed these in response to extreme circumstances as a coping strategy where needed support was not available.

Whilst acknowledging the small sample size, a range of purposes served by children’s imaginary companions were identified through this analysis. Imaginary companions may serve different purposes for different children and I have argued that children with imaginary companions are a heterogeneous group in terms of age, cognitive functioning, communication and social relationships (Spender et al., 2011). It is possible that a larger sample might have enabled a wider variety of purposes to be identified. The contribution of this study is that it investigated the range and characteristics of imaginary companions the children had had over time, how they interacted with them and how these interactions sometimes related to events in their lives. It enabled the exploration of children’s perceptions and feelings towards their various imaginary companions. The children were able to articulate which imaginary companions were special and why. These features have supported identification of some of the purposes served and shed light on how the children’s interactions with their imaginary companions seemed to meet their various needs. This study found that some children had a number of imaginary friends with different characteristics linked with the different purpose they appeared to serve for the child. The detailed exploration of a small sample (employing appropriate methodology; in this case, IPA) has enabled the systematic analysis of both the similarities and distinctiveness of imaginary companions. This investigation might have been compromised by the search for general themes in a larger population. Further research may well reveal other purposes. Nevertheless, it is anticipated that this analysis will be of use both to child practitioners and to researchers in coming to an understanding of some of the purposes served by a child’s imaginary friends.

Interviews with a larger sample of girls and boys would enable comparison of the characteristics of imaginary companions and whether there are similarities and differences in purposes served.

Most children in this study had more than one imaginary companion over time. Further research could establish whether different purposes were served by the various imaginary companions. The traditional research focus on a single primary imaginary companion is likely to reveal only some of the purposes served for the child when they have more than one.

There is a need to compare the characteristics and features of imaginary companions in normative and particular clinical samples (McLewin and Muller, 2006). More qualitative research of clinical and normative samples could lead to identifying psychopathology at an earlier stage.

Conclusion

This study explored how imaginary companions featured in the lives of a small sample of school aged children. The purposes identified were varied and this research emphasises the positive contributions imaginary companions bring to the lives of children.

Footnotes

Appendix

Table 1.

Characteristics of children’s imaginary companions.

Child/Age of child	Animal/Name of imaginary companion	Human/Name of imaginary companion	Gender/Age of imaginary companion	Imaginary companion shows independent will	Purposes served by imaginary companion^a	Imaginary companion is known to family	Imaginary companion is known to select others
Antonio 5 yrs		Britten	Male 12 yrs	Yes	1,2,4,5	Yes	No
		Ridey	Male 8 yrs	Yes	1,2,3,4	Yes	No
		Buzzie	Male 6 yrs	Yes	2,3,4	Yes	No
Harry 6 yrs	Barnaby Bear		Male adult	Yes	1,2,5	Yes	Yes
		Manager	Male adult	Yes	1,2,5	Yes	No
	Ducky		Male 15 yrs	Yes	4	Yes	No
Lisa 6 yrs	Pony Minty		Female	Yes	1,2,5	Yes	Yes
		Amy, Katie, Megan	Females 5 yrs	No	1,2	Yes	Yes
John 8 yrs		Tom	Male 17 yrs	Yes	1,2,3,4,5	Yes	No
Carmel 10 yrs		Tinton and Dubbish	Males 7 yrs	Yes	1,2,3,4	Yes	Yes
		Betty and Sinjon	Parents	Yes	1	Yes	Yes
		Susan and Eileen	Baby sisters	No	1	Yes	Yes
Holly 11 yrs	Pony Dream		Female	Yes	1,2,3,4,5	No	No C^b
	Puppy Jasper^c		Male	No	1	Yes	No
	Puppy Dora		Female	No	1’	Yes	No
		Twin Lily	Female 4 yrs	Yes	1,2,5	Yes	No
Tara 11 yrs	Horse Fantazia		Female 18 yrs	Yes	1,2,3,4,5	No	Yes C^b
	Horse Tom		Male 17 yrs	Yes	1,2,3,4,5	No	Yes C^b
		Ricky	Female/male	Yes	3	No	No
		Sophie	Female 11 yrs	Yes	3	No	No
		Annie	Female older	Yes	3	No	No
Ella 11 yrs		Polly	Female 11 yrs	Yes	1,2,3,4	No	No

Purposes code: 1. overcome boredom/loneliness/entertainment, 2. friend/playmate, 3. express/release upset feelings, 4. Support for problem situations, 5. wish fulfilment.

C Concealed in games.

Previous ICs are shown in italics.

Acknowledgements

I am very grateful to the children that took part in this research. I would very much like to thank Dr Virginia Eatough, Birkbeck, University of London, for constructive feedback on the analysis, and Brian Apter for continued support and encouragement.

Funding

This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.

References

Bender

Vogel

(1941) Imaginary companions of children. American Journal of Orthopsychiatry 11(1): 56–65.

Benson

Pryor

(1973) When friends fall out: Developmental interference with the function of some imaginary companions. Journal of the American Psychoanalytic Association 21(3): 457–473.

Bjorklund

(2007) Why Youth is not Wasted on the Young. Oxford: Blackwell.

British Psychological Society (2004) Ethical Guidelines: Guidelines for Minimum Standards of Ethical Approval in Psychological Research. Leicester: British Psychological Society.

Brooks

Knowles

(1982) Parents’ views of children’s imaginary companions. Child Welfare LXl (1): 25–33.

Brott

(2004) Imaginary friends: Should you be concerned. Available at: www.familyresource.com/parenting/6/551 (accessed August 2004).

Carlson

Taylor

(2005) Imaginary companions and impersonated characters: Sex differences in children’s fantasy play. Merrill-Palmer Quarterly 51(1): 93–118.

Cohen

Mackeith

(1991) The Development of Imagination: The Private Worlds of Childhood. London: Routledge.

Dunn

(2004) Children’s Friendships: The Beginnings of Intimacy. Oxford: Blackwell.

10.

Elliott

Fischer

Rennie

(1999) Evolving guidelines for publication of qualitative research studies in psychology and related fields. British Journal of Clinical Psychology 38(3): 215–229.

11.

Gleason

(2002) Social provisions of real and imaginary relationships in early childhood. Developmental Psychology 38(6): 979–992.

12.

Gleason

(2004) Imaginary companions: An evaluation of parents as reporters. Infant and Child Development 13(3): 199–215.

13.

Gleason

(2005) Mothers’ and fathers’ attitudes regarding pretend play in the context of imaginary companions and of child gender. Merrill-Palmer Quarterly 51(4): 412–436.

14.

Gleason

Sebanc

Hartup

(2000) Imaginary companions of preschool children. Developmental Psychology 36(4): 419–428.

15.

Gurain

(2004) When your child’s new friend is imaginary. Available at: www.aboutourkids.org/aboutour/articles/imaginary.html (accessed August 2004).

16.

Harris

(2000) The Work of the Imagination. Oxford: Blackwell.

17.

Heins

(2004) Imaginary playmates. Parent Kids Right. Available at: www.parentkidsright.com/pt-imaginary.html (accessed August 2004).

18.

Hoff

(2004–2005) A friend living inside me: The forms and functions of imaginary companions. Imagination, Cognition and Personality 24(2): 151–189.

19.

Hoff

(2005) Imaginary companions, creativity, and self-image in middle childhood. Creativity Research Journal 17(2–3): 167–180.

20.

Kalyan-Masih

(1986) Imaginary play companions: Characteristics and functions. International Journal of Early Childhood 18(1): 30–40.

21.

McLewin

Muller

(2006) Childhood trauma, imaginary companions, and the development of pathological dissociation. Aggression and Violent Behaviour 11(5): 531–545.

22.

Majors

(2009) Children’s imaginary companions and the purposes they serve. Unpublished doctoral thesis, Institute of Education, University of London.

23.

Majors

(2012) Friendships: The power of positive alliance. In: Roffey

(ed.) Positive Relationships: Evidence Based Practice Across the World. London: Springer.

24.

Manosevitz

Prentice

Wilson

(1973) Individual and family correlates of imaginary companions in pre-school children. Developmental Psychology 8(1): 72–79.

25.

Mauro

(1991) The friend that only I can see: A longitudinal investigation of children’s Imaginary companions. Unpublished doctoral dissertation, University of Oregon.

26.

Nagera

(1969) The imaginary companion: Its significance for ego development and conflict resolution. The Psychoanalytic Study of the Child 24: 165–196.

27.

Pearson

Rouse

Doswell

. (2001) Prevalence of imaginary companions in a normal child population. Child: Care, Health and Development 27(1): 13–22.

28.

Piaget

(1962 [1923]) Play, Dreams and Imitation. London: Routledge and Kegan Paul.

29.

Roffey

Tarrant

Majors

(1994) Young Friends: Schools and Friendship. London: Cassell.

30.

Seiffge-Krenke

(1993) Close friendship and imaginary companions in adolescence. New Directions for Child Development 60(1): 73–87.

31.

Seiffge-Krenke

(1997) Imaginary companions in adolescence: A sign of a deficient or positive development? Journal of Adolescence 20(2): 137–154.

32.

Smith

Eatough

(2006) Interpretative phenomenological analysis. In: Breakwell

Hammond

Fife-Schaw

Smith

(eds) Research Methods in Psychology. London: Sage.

33.

Smith

Jarman

Osborn

(1999) Doing interpretative phenomenological analysis. In: Murray

Chamberlain

(eds) Qualitative Health Psychology: Theories and Methods. London: Sage.

34.

Spender

Barnsley

Davies

Murphy

(2011) Primary Child and Adolescent Mental Health: A Practical Guide, Vol. III. London: Radcliffe Publishing.

35.

Sugarman

Jaffe

(1989) A developmental line of transitional phenomena. In: Fromm

Smith

(eds) The Facilitating Environment: Clinical Applications of Winnicott’s Theory. Madison, CT: International Universities Press.

36.

Svendsen

(1934) Children’s imaginary companions. Archives of Neurology and Psychiatry 2: 985–999.

37.

Taylor

(1999) Imaginary Companions and the Children Who Create Them. Oxford: Oxford University Press.

38.

Taylor

Carlson

Maring

. (2004) The characteristics and correlates of fantasy in school-age children: Imaginary companions, impersonation, and social understanding. Developmental Psychology 40(6): 1173–1187.

39.

Taylor

Hulette

Dishion

(2010) Longitudinal outcomes of young high-risk adolescents with imaginary companions. Developmental Psychology 46(6): 1632–1636.

40.

Winnicott

(1971) Playing and Reality. London: Tavistock Publications.