Children's Understanding of What is 'Real' on Television
A Review of the Literature
University of Wales, Aberystwyth
Without being taught to do so children make their own assessments of the reality status of television programmes. Based upon their growing knowledge of both the medium and the everyday world they make increasingly sophisticated judgements about what is 'real' on television using multiple criteria. My primary concern in this paper is to summarise and integrate key findings from the most widely-cited research studies which have investigated children's understanding of what is 'real' on television, in particular concerning developmental patterns in young viewers' use of various criteria for assessing the reality status of television programmes.
Referring to children in their study whose ages ranged from 6- to 12-years-old, Bob Hodge and David Tripp reported that 'calibrating television against reality is a major concern for children throughout this age group' (Hodge & Tripp, 1986, p. 126), and other studies (e.g. Flavell et al., 1990) suggest that this may well apply to even younger viewers. Hodge and Tripp have argued that watching television may play an important part in helping children to develop concepts of reality and fantasy. Cartoons, they suggest, may have a special function for young viewers. This was the favourite television genre of the 6- to 8-year-old children they studied in Australia, whilst most of the 9- to 12-year-olds in their study preferred TV dramas, so that the popularity of programmes amongst these children was 'directly the opposite of the order of reality, going from most unrealistic (cartoons) to most realistic (real-life characters)' (ibid., p. 119). After a detailed semiotic study of how children made sense of a television programme, these researchers argued that 'far from the fantastic nature of cartoons causing confusion between fantasy and reality, the largeness of the gap is helpful to young children in building up precisely this capacity to discriminate' (ibid., p. 9). Offering some explanation as to why children might be particularly concerned with making judgements about the reality status of television programmes, the psychologist Howard Gardner and his colleague Patricia Morison have plausibly suggested that 'the frightening status of certain fantasy figures may motivate children early on to master their reality status' (Morison & Gardner, 1978, p. 648). Learning to remind themselves of the constructedness of a television programme may help viewers to distance themselves from emotional responses to disturbing scenes.
In the research literature on this topic, children's understanding of what is 'real' on television tends to be discussed either under the heading of 'perceived reality' or under that of 'modality judgements'. Indeed the use of one of these terms rather than the other signals differing ideological stances amongst commentators. An objectivist leaning towards epistemological/ontological realism is flagged by the use of the term 'perceived reality', whilst a subjectivist leaning towards idealism - or at least a socially-inflected constructivist stance - is signalled by the term 'modality judgements'. In the interests of declaring my own biases, I should inform readers that my personal slant is constructivist.
Bob Hodge and David Tripp have been closely associated with the study of children's 'modality judgements'. In a semiotic approach to studying children's understanding of television in Australia (Hodge & Tripp, 1986) adopt the linguistic term 'modality' to refer to the reality status attributed to television programmes by viewers. Where there seems to be a great distance between a programme and everyday reality, television has 'weak modality'; where television seems like a 'window on the world' it has 'strong modality'. The point is that the modality of television varies, a dimension hardly allowed for in the approaches adopted by some researchers. Hodge and Tripp note that 'judgements about "reality" are complex, fluid and subjective' (ibid., p. 130), and that the modality judgements of young children 'tend to be polarized, contradictory and unstable' (ibid.).
Robert Hawkins (1977), in a very influential paper employing the more traditional term, nevertheless questioned its adequacy. There has often been a tendency to refer to perceived reality as if it were homogeneous, whilst at the same time researchers have sought to measure it by asking quite different arrays of questions. Hawkins stressed that it was misleading to regard 'perceived reality' as a unitary concept, arguing that it was more usefully seen as multidimensional. He applied factor analysis to 153 children's questionnaire responses, and he discerned several apparent subdivisions within the concept. Relating this to developmental patterns, Hawkins noted, 'given multiple perceived reality dimensions, developmental changes may take place along some dimensions but not others, or changes may occur at different rates or times on different dimensions. Second, to make things even more complex, it is quite possible that children's dimensional structures themselves differ with age' (Hawkins, 1977, pp. 305-6). Byron Reeves (1978) added that such dimensions 'may differentially influence how television affects children' (Reeves, 1978, p. 689).
Many commentators have subsequently adopted Hawkins's references to 'Magic Window' and 'Social Expectations' dimensions, although often in misleading references to the factors which Hawkins had actually identified in his data. It is perhaps worth noting at this point that whilst the abstract of his paper refers only to the Magic Window dimension (defined as 'the degree to which children believe they are viewing either ongoing life or drama') and the Social Expectations dimension (defined as 'the degree to which they believe television characters and events do or do not match their expectations about the world') (Hawkins, 1977, p. 299), his subsequent analysis was by no means so clearcut, referring also to factors such as the perceived 'usefulness' to young viewers of particular programme events or characters (to which I will allude in due course).
Although theorists may differ slightly in defining the various criteria which they identify in children's judgements about the reality status of television, all serious researchers in the field now treat 'perceived reality' as multidimensional. Researchers have referred to various criteria which seem to be involved in viewers' judgements about whether an object, character, event or setting on television is 'real' and I will shortly discuss each of these criteria in turn.
Developmental perspectives build upon the reasonable assumption that children's understanding of television programmes improves with age, and bears some relationship to general cognitive development. Twenty years ago, in his book Children in Front of the Small Screen, Grant Noble (1975) related children's understanding of television to a basic Piagetian framework. More specifically in relation to children's understanding what is 'real' on television, Mac Brown and colleagues (Brown et al., 1979), in a correlational study of sixty-four 6- and 7-year-olds, found a relationship between children's perception of TV reality (comparing an episode of the action drama StarTrek with a cartoon version) and their stage of cognitive development (assessed on standard Piagetian conservation tasks). Conservers performed better than non-conservers on perceived reality measures (using the researchers' own scale). One might also reasonably expect that children's understanding of what is real on television would bear some relationship to other aspects of general cognitive development, such as perspective-taking.
To track developmental patterns in the framing of television reality, Aimée Dorr (1983) conducted a series of interviews with 54 children (aged 5- to 6-years-old, 7- to 9-years-old and 11- to 12-years-old). She employed the promising strategy of asking children what they would tell a younger child who was puzzled about what to believe on television. A great leap forward in their sophistication at this task seemed to occur between the ages of 5- and 9-years-old. However, even before the age of 5-years-old, there are major developments in children's understanding of television reality which will be considered here in relation to the recognition of absence.
Children's understanding of what is 'real' on television clearly needs to be related more generally to the development of their understanding of what is real in the everyday world. In the preschool years, children's concepts of reality involve discriminating between the way objects appear and the way they really are (Taylor & Flavell, 1984; Flavell, 1986). Many young preschoolers (3- or 4-years-old) seem to have little grasp of a distinction between appearance and reality (e.g. when a toy car of one colour is screened by a transparent filter of another colour). This skill is highly correlated with visual perspective-taking tasks (Flavell, 1986). In contrast, 6- and 7-year-olds easily manage simple appearance-reality tasks but have difficulty reflecting on and talking about related notions such as 'looks like' and 'really and truly is'. By 11- or 12-years-old, children demonstrate considerable skill in making rich distinctions between appearance and reality (ibid.).
In relation to very young children's assessment of the reality of television, one must first consider the ontological status which they grant to identifiable objects appearing on the screen. We might refer to this as an issue of substance. In three carefully-designed experimental studies Flavell et al. (1990) investigated whether 3- and 4-year-old children interpreted television images as solid, physically-present objects or simply as insubstantial images of them. 3-year-olds seemed to assume the former; 4-year-olds clearly believed the latter. For instance, the younger children tended to agree that a bowl of popcorn shown on television would spill if the television set were turned upside down. However, the researchers argue that the 3-year-olds did not really believe that a television set contained physical objects but rather had difficulty in distinguishing conceptually between television images and their referents. They hypothesize an early developmental process: under-3s probably begin by assuming that what they see on TV are real, tangible objects inside the set; around the age of 3 years, children gradually learn that TV images do not behave like ordinary objects; and around the age of 4 years, children realize that TV images represent an absent reality and, when asked, are capable of distinguishing TV images from their referents.
A related issue reflected in very young children's judgements about the reality status of television is the viewer's understanding of the independent, uncontrollable nature of objects and events depicted on television. In a small but intensive longtitudinal study, Leona Jaglom and Howard Gardner (1981a, 1981b) noted that by the age of 3-years-old the children realized that they could not influence events on television and had generally realized that events on television could not directly involve them (Jaglom & Gardner, 1981a, p. 39). 'Between the ages of 3 and 4 years, children recognize the fact that the television world is in fact separate from their own. Its events do not actually exist in reality; they cannot be acted on directly' (ibid., 45). However, even at 3- or 4-years-old, children had difficulty in accepting that the timing and availability of broadcast programmes could not be controlled in the home (ibid., p. 44).
An important criterion involved in viewers' assessments of the reality status of specific programme content is variously referred to as 'the Magic Window' (Hawkins, 1977), 'fabrication' (Dorr, 1983), and 'factuality' (Fitch et al., 1993). As I have briefly noted, Robert Hawkins's widely-quoted reference to a Magic Window dimension referred to the degree to which TV programmes were regarded by viewers as either a window onto actual on-going life in the real world or as dramatic fiction. The criterion of fabrication, as framed by Aimée Dorr (1983), relates to whether a television programme is perceived by the viewer as 'made up' or alternatively as depicting events as they actually happen in real life. Hawkins's original paper referred specifically to the evaluation of dramatic fiction, and of course not all programmes fall into this category. However, all television programmes - even news broadcasts - do involve construction, and it is useful to think of this criterion as relating to an awareness of the constructedness of programmes.
It is widely noted that very young viewers start at the high end of the Magic Window dimension, attributing equal reality to everything on television. In a questionnaire study of 153 children of 4- to 12-years-old (with varying degrees of experimenter support), Hawkins confirmed previous research findings that children tend to perceive fictional television as increasingly less real as they grow older. Hawkins's data reflected a dramatic increase in children's knowledge in this regard around the age of 8 years; children over 8-years-old rarely thought of television as a magic window on the world, and understood that programmes were made up. This has been a general finding. For the various age-groups of children she studied, Aimée Dorr noted that judgements which could be ascribed to the criterion of fabrication represented the following percentages of all criteria employed: 5- and 6-year-olds, 15%; 7- to 9-year-olds, 23%; 11- and 12-year-olds, 19% (Dorr, 1983, p. 209).
According to Marguerite Fitch and colleagues (Fitch et al., 1993), by around the age of 10, children's judgements about what these researchers term 'factuality' are about as accurate as those of adults. They also note (citing Morison et al., 1981), that this particular criterion appears to be primarily dependent on a child's stage of cognitive development (rather than on such factors as experience with television), and they argue that a concern with factuality (and in particular a reliance on formal features as cues) is developmentally prior to a reliance on other criteria (Morison et al., 1981, p. 48). Hodge and Tripp found that methods of media production (actual or hypothetical) were the key criterion of reality for 8- and 9-year-olds (Hodge & Tripp, 1986, p. 126). Children's use of formal features of the medium as cues to reality status will be dealt with in some detail in the next section.
Relating to the issue of the constructedness of television is the ontological status of participants in programmes. A study by Joanne Quarforth (1979) sought to determine how far children with mean ages ranging from 6.0 to 10.0 would spontaneously sort pictures of television characters into groups reflecting the attributes human, animated and puppet. The various percentages doing so in each age group were as follows: 48% at a mean age of 6.0; 57% at 7.0; 75% at 8.1; 83% at 8.9; 95% at 10.0 (Quarforth, 1979, p. 213). A similar study by Patricia Morison and Howard Gardner (1978) also showed a steady progression in children's spontaneous classification of pictures as fantasy or real. In interviews, children in the Quarforth study from the age-groups with mean ages of 8.9 and over were significantly more accurate than those from the age-group with a mean age of 6.0 in attributing the quality of being alive to human characters and not to puppets or animated characters. The 6-year-olds were significantly less able to pick out the characters that could walk and talk by themselves than were those of around 7.0 and older. 18% of the 6-year-olds attributed only to human characters both the qualities of being alive and of autonomous movement, whereas 70% of the 10-year-olds did (Quarforth, 1979, p. 214). Whilst in this study only 15% of the 6-year-olds were able to fully and consistently differentiate human, puppet and cartoon characters, 85% of the 10-year-olds were able to do so (ibid., p. 216). One should note that the extent to which children ordinarily employ the real/fantasy distinction has been questioned (e.g. Reeves & Greenberg, 1977).
Hope Klapper (1981), in an interview study of eighty-eight 7- and 8-year-olds and eighty-five 10- and 11-year-olds, found that the spontaneous responses of at least a third of the 7- and 8-year-olds and at least two-thirds of the 10- and 11-year-olds showed that they knew about actors, scripts and plots. 'These children were... well aware that what they were judging was fiction' (Klapper, 1981, p. 80). Hope Kelly (1981), in a interview study of fifty-four children from 7- to 12-years-old, found that 9- and 10-year-old children were more aware that programmes have authors than 7- and 8-year-olds. They referred to the fact that programmes are scripted, acted, rehearsed, costumed etc. They were also aware of 'fiction based on fact'. David Fernie (1981) found age-related changes in children's knowledge that television characters were played by actors. Among 5- and 6-year-olds, 58% did not understand this; among 8-year-olds, 45% completely understood this, 26% partially understood it and 29% did not; among 11- and 12-year-olds, 65% completely understood this (Fernie, 1981, p. 54). Fitch et al. (1993) noted that most 11-year-olds know that an actor playing a police officer in a television drama is not a police officer in real life.
Dorr noted that for the children of 5-years-old and older whom she studied, criteria other than fabrication were more important in judging the reality status of television programmes. The other criteria she specified (possibility and probability, discussed below) assumed that programmes were fabricated, but required additional judgements (see also Dorr et al., 1990). In Hawkins's (1977) framework, other criteria formed what he called the 'Social Expectations' dimension and involved comparisons with the viewer's experience of the world.
The criterion of physical actuality involves assessing television reality in terms of whether a person or event shown on television is known to exist or happen in the real world. Hope Kelly (1981) found that children of 7- and 8-years-old seemed to assess television reality initially in terms of this criterion: if they considered that a person or event on TV existed or happened in the real world, then it was regarded as real. In contrast to the focus of the 7- and 8-year-olds on the criterion of physical actuality, the 9- and 10-year-olds in her study were more often asking themselves 'Does something like this exist?' or 'Is it about something that does (or did) exist?' (ibid., p. 67).
A study of fifty-four children from 7-years-old to 12-years-old by Morison et al. (1981) showed that actuality was the most frequent criterion cited, accounting for around half of the references to criteria offered, with no major fluctuations across these age-ranges (Morison et al., 1981, p. 236). Various researchers employ differing categorizations of criteria, of course: Dorr (1983), for example, does not separate a criterion of actuality from one of 'possibility', and such a distinction may depend on such subtleties as the directness of viewers' experience of the phenomena depicted or the degree of certainty which they express.
Hope Kelly (1981) reported that, in addition to actuality, 7- and 8-year-olds assessed possibility or impossibility (whether something could happen in real life) - especially physical impossibility - 'people can't fly unless they go in an aeroplane', etc. And Dorr (1983) found that as they grew older, children (from 5- to 12-years-old) became increasingly concerned with whether, on the basis of their direct or indirect knowledge of the world, a phenomenon on television seemed possible (however uncommon) in real life. A child might argue that 'the bionic man' appearing at the time in an action adventure programme (who had been completely 'rebuilt' by scientists), could be real because prosthetic devices are sometimes used in medicine (Dorr, 1983, p. 202).
Morison et al. (1981) found that amongst children from 7- to 10-years-old possibility accounted for around one-third of their references to criteria for assessing the reality status of television programmes (second only to actuality), but that amongst children of 11- and 12-years-old, this dropped to only 13% (Morison et al., 1981, p. 236). They added that whilst the negative criterion of impossibility was used by equal numbers of children from 7- to 12-years-old to explain why television programmes were not real, twice as many 9- and 10-year-olds as 7- and 8-year-olds used the positive criterion of possibility to confirm that programmes were real (ibid., p. 238). In her study, Dorr found that for the various age-groups, judgements which could be ascribed to possibility as a percentage of all the criteria she noted were as follows: 5- and 6-years-old, 17%; 7- and 9-years-old, 28%; 11- and 12-years-old, 47% (Dorr, 1983, p. 204). Indeed, she added that possibility remained the most common criterion amongst the adolescents and adults she studied. Clearly the findings regarding the criterion of possibility reported in these studies differ markedly for 11- and 12-year-olds (though this may be related to the fact that Dorr does not make use of a separate 'actuality' criterion).
With reference to the realism of social and psychological events on television commentators usually refer to 'plausibility' rather than possibility. The criterion of probability or plausibility relates to whether the phenomenon on television seems to the viewer to be 'true to life' or likely to happen in the real world in a similar manner (on the basis of their own experience or knowledge or that of personal acquaintances). Children may note that 'things like that do happen' or that 'people like that do exist'. Some writers refer to 'representativeness' - where people and/or events in a television programme, though accepted as fictional, are nevertheless regarded as representative of everyday reality (e.g. Dorr, 1983, Howard, 1993). As Dorr (1983) notes, such a criterion is less inclusive than possibility.
In a study by Morison et al. of children from 7- to 12-years-old, plausibility was the least common criterion used by children of 7- to 10-years-old (although as a percentage of their references to criteria it increased from 2% amongst 7- and 8-year-olds to 9% amongst 9- and 10-years-olds). A significant increase in the use of this criterion occurred amongst 11- and 12-year-olds, amongst whom this represented nearly a quarter of their references to criteria (making plausibility second only to actuality) (Morison et al., 1981, p. 236). Morison and her colleagues noted that in their study a concern with plausibility was what differentiated 11- and 12-year-olds from younger children (ibid., p. 240). Similarly, in her study, Hope Kelly (1981) noted that whilst 9- and 10-year-olds did not usually assess the plausibility of characters and events, for children of 11- or 12-years-old this was the main criterion for assessing television reality (they were hardly concerned with actuality or possibility). Amongst all those over 12-years-old, Dorr found that judgements based on the criterion of probability were far more common than amongst the children of 12-years-old and under. However, in a small-scale qualitative study of 9- and 10-year-old children in Australia and England, Susan Howard reported that even amongst these children, the criterion of 'representativeness' was 'one of the most frequently used in justifying a realistic/true-to-life classification' of a programme (Howard, 1993, p. 44). She also underlines the importance of the negative form of this criterion - and indeed classes implausibility as a separate criterion (ibid., pp. 46-7). Howard notes two very different ways in which implausibility was judged: some soap operas were judged unrealistic because of the sheer number of significant events which they packed into an episode; whilst some programmes were judged unrealistic because the behaviour of characters did not reflect the viewer's own cultural assumptions and practices (ibid. , p. 46).
The general pattern reflected in these studies remains one of a steady increase with age in the extent to which viewers draw upon the plausibility/probability criterion in evaluating the reality status of television programmes. However, referring to 'Social Expectations' factors (and in contrast to the 'Magic Window' dimension), Hawkins (1977) found in his statistical data no linear decrease with age (from 4- to 12-years-old) in the perceived reality of television drama, but rather a curvilinear trend regarding a factor relating to the social 'usefulness' of programmes to child viewers. In his study, children's scepticism about the usefulness of television for themselves showed up as significantly higher amongst both the youngest and the oldest children than amongst the middle age-group - who saw programmes as useful sources of information about the everyday world. Several commentators have found this paradoxical insofar as it seemed to suggest unusual scepticism about the reality status of programmes amongst the youngest children. A contributory factor may have been that the pre-schoolers may not have understood all of the researcher's questions. However, any such pattern is not in conflict with the findings of other researchers about the general increase with age in children's references to the plausibility criterion since it can be seen as reflecting patterns of motivation amongst young viewers.
Hawkins also found (though many commentators alluding to his paper omit this) yet another pattern regarding a more specific Social Expectations factor - the representation of families in television drama. According to his data, the children's perceptions of the realism of such depictions increased with age (Hawkins, 1977, p. 312). It is hardly surprising that Fitch et al. (1993) considered that there was little clear evidence of a developmental pattern for what they called social realism, partly because of the importance of specific programme content, but also because such judgements seem to be more a function of motives for viewing and of television experience than of cognitive development (Fitch et al., 1993, p. 43).
Children's judgements of the reality status of television programmes are not based solely on comparing specific programme content with their knowledge of the world. They also need to draw on their knowledge of the medium of television. Without the use of both kinds of knowledge, a documentary about an exotic country might seem as fantastic as a science-fiction adventure. Progressive sophistication with age and experience is evident in the development of children's use of what are normally referred to as the 'formal features' of the television medium as cues to the reality status of programmes (Hodge & Tripp, 1986, Fitch et al., 1993). These range from production and editing techniques and conventions to TV genres. Hodge and Tripp refer to such medium-specific cues to reality status as 'internal' criteria, in contrast to 'external' criteria, which involve comparisons with the viewer's knowledge and experience of the world (which is close to Hawkins's distinction between the 'Magic Window' and 'Social Expectations' dimensions).
Kelly (1981) notes that 7- and 8-year-old children 'unanimously chose Superman as more real than Charlie Brown on the basis of the former's superficial verisimilitude to life - that is filmed rather than animated. Thus, even though these youngsters can chronicle the many tricks underlying Superman's feats, format overrides content when children are forced to make a comparison. At this age, the answer to the question, Which is more real? is, quite simply, whichever looks more real' (Kelly, 1981, p. 66). 9- and 10-year-olds were much less likely to mention formal features as cues to reality; they were more concerned with content. Susan Howard refers to 9- and 10-year-olds often classifying the animated cartoon The Simpsons as realistic 'because it was judged to depict characters and situations that were representative of those in real life' (Howard, 1993, p. 50).
Genre is an important framework within which viewers make sense of particular programmes. In an intensive longtitudinal study of twelve children from 2- to 5-years-old, Jaglom and Gardner (1981a, 1981b) noted the development of genre distinctions. 2-year-olds did not recognize the beginnings and endings of programmes (Jaglom & Gardner, 1981b). The researchers found that for the 2-year-olds the disappearance of characters was a source of consternation: 'children become very upset and sometimes even cry when their favourite television personalities leave the screen' (Jaglom & Gardner, 1981a, p. 42): they suggested that this feature might assist their eventual identification of the advertisement genre. The researchers report the order of acquisition of the principal genre distinctions: advertisements (3.0-3.6); cartoons (3.7-3.11, early in interval); Sesame Street (3.7-3.11, late in interval); news (4.0-4.6); children's shows (4.0-4.6, late in interval); adult shows (4.0-4.6) (ibid., p. 41). They argue that 'in the first few years of attempting to sort out the confusing elements of the television world, children are concentrating on making distinctions between shows' (ibid., p. 42).
Dorr (1983) noted that children spontaneously referred to particular television genres and to specific programmes within them as a way of judging the reality status of programmes: news, sports, documentaries and crime dramas were realistic but cartoons were not. All of the 5- and 6-year-olds alluded to the cartoon form as 'pretend', whilst the percentages of children referring to the 'news' form as indicating reality were as follows: 5- and 6-year-olds, 10%; 7- to 9-year-olds, 37%; 11- and 12-year-olds, 57% (Dorr, 1983, p. 210).
Morison et al. (1981), in their interview study of fifty-four children of 7- to 12-years-old, noted references to TV-specific cues to reality status. The most common of these cues amongst 7- and 8-year-olds were what Morison and her colleagues referred to as 'physical features', including the presence or absence of stunts, camera tricks, costumes, props and sets; these constituted about half of the TV-specific references of 7- and 8-year-olds, whereas such references dropped dramatically to 18% of those of 9- and 10-year-olds and to 15% of 11- and 12-year-olds (Morison et al., 1981, p. 236). For 9- and 10-year-olds, the most common of these TV-specific cues were 'performance features', such as whether the programme was acted, scripted, rehearsed, live or filmed; this feature accounted for 59% of the references to medium-specific criteria cited by this age-group - and for much the same percentage (54%) amongst 11- and 12-year-olds, whereas they constituted only 25% of these references amongst 7- and 8-year-olds. There was a steady growth of references to the purpose of programmes (1% of TV-specific cues amongst 7- and 8-year-olds, rising to 4% amongst 9- and 10-year-olds and 10% amongst 11- and 12-year-olds). References to the programmes as authored rose from 5% amongst 7- and 10-year-olds to double this amongst 11- and 12-year-olds. References to animation dropped gradually from 12% amongst 7- and 8-year-olds to 9% amongst 9- and 10-year-olds and 7% amongst 11- and 12-year-olds (ibid.).
Sketching a schematic outline of the development of children's understanding of what is real on television, Fitch et al. suggest that 'initially, children probably believe all TV is real' (Fitch et al., 1993, p. 48). Around the age of 3- or 4-years-old they use formal features such as animation to identify what is not real. By 5- or 6-years-old, 'they begin to identify co-occurring features of form and content' (ibid.) which characterize a television genre or a particular TV series or serial. Morison et al. (1981) note that: 'children's reality-fantasy judgments about television shift, with age, from a focus on physical features and a rigid assessment of actuality, to a sensitivity to the plausibility of characters and plotline and an appreciation of authorial intent' (Morison et al., 1981, p. 229).
Hodge and Tripp (1986) found, in their study of 6- to 12-year-olds, that media-external features were more important to older children, who were more reliant on applying their knowledge of everyday reality in making modality judgements about television programmes. They also identified as a key developmental feature that older children used more features in judging programme reality than younger children did (Hodge & Tripp, 1986, p. 126). Hope Kelly (1981) noted that, like adults, 11- and 12-year-olds were likely to ask investigators, 'What do you mean, "real"? Real in what way?', underlining the observation that at this age multiple criteria are available for such judgements.
Stages of cognitive development clearly play an important part in children's understanding of what is real on television. But various other factors are also discussed in the literature, such as: motives for viewing; familiarity with television; relative amount of viewing (often referred to as 'exposure to TV'); and real-world experience (a useful discussion of various factors can be found in Potter, 1988). James Potter (1988) notes the importance of the viewer's particular motives for watching television. Some motives have been shown to be related to levels of perceived reality, in particular the motive of watching television in order to learn or to seek information. He observes that 'it is not surprising that people who find television more like real life expose themselves to it to seek information and instruction' (Potter, 1988, p. 33). Susan Howard noted that for the primary school children she studied one (relatively minor) criterion involved in judging a programme was that it was regarded as realistic 'if it taught them something' about the world or about 'how to live your life' (Howard, 1993, pp. 44, 49).
Although we may reasonably assume the importance of some degree of familiarity with television, Morison et al. (1979), in a study of 36 children from 6- to 12-years-old, could find no relationship between their ability to distinguish reality from fantasy and their degree of familiarity with television. Regarding 'exposure', there is some evidence that those who are 'heavy' viewers (who watch significantly more television than the 'average viewer') tend to regard television generally as more realistic (i.e. as accurately reflecting real life) than lighter viewers (e.g. Elliot & Slater, 1980, Greenberg & Reeves, 1976). Such findings are in accord with cultivation theory. However, the direction of causality has not been indisputably established.
Explorations of the role of the viewer's personal experience can be found in Greenberg and Reeves (1976), Elliot and Slater (1980) and Dorr et al. (1990). Surprisingly, we currently have little hard evidence of young viewers' use of personal real-life experience as a keystone in judging the reality status of television programmes, though we are surely safe to assume that this reflects the limitations of our current research strategies and that drawing upon relevant personal experience must play an important part when viewers judge the reality of television programmes. Bradley Greenberg and Byron Reeves (1976), in a correlational study of 201 (white) children aged from 8- to 12-years-old and living in suburban Michigan, found no evidence of the use of personal experience in judging portrayals of particular groups (families, policemen, blacks) as realistic. Younger children, less able children and heavy viewers all regarded television as more realistic than others did. Again, this is very much in accord with cultivation theory. Despite acknowledging that personal experience must play some part, Greenberg and Reeves suggested that the plausibility of fictional television programmes may tend to be judged against other programmes of related genres (and presumably second-hand knowledge) rather than against direct personal experience (Greenberg & Reeves, 1976, pp. 94-5).
James Potter (1988) noted that other than the Greenberg and Reeves study (1976) there was very little evidence of any association of perceived reality with factors such as IQ, gender, race or socio-economic status. However, socio-economic factors cannot be discounted. Peter Nikken and Allerd Peeters (1988) found that in the Netherlands young children from lower socio-economic environments showed a stronger tendency than other children to believe that a place called 'Sesame Street' really exists and that the characters resided inside the television set (Nikken & Peeters, 1988, pp. 448-50). David Buckingham (1993) reported that, in his small-scale study, middle-class children in Britain made more modality judgements and were more concerned with external than internal criteria. But as he notes, middle-class children may be more adept at giving interviewers what they seem to want (Buckingham, 1993, p. 233). In the United States, William Donohue and Thomas Donohue (1977) found that black working-class children tended to regard some role stereotypes and social scenarios on television as 'significantly more real' than white upper- and middle-class children did.
Susan Howard makes a more general allusion to the importance of differences between the cultural practices or assumptions of child viewers and the characters on screen in children's judgements of the plausibility of television programmes (Howard 1993, p. 46). Regarding gender and modality judgements, subtle differences between young viewers have sometimes been noted. For instance, Howard notes that in her study of primary-school children, 'boys seemed to be much more concerned than girls with how the illusion of realism was constructed' (ibid., p. 44).
Fitch et al. (1993) assert that children may be less able to differentiate factual from fictional television 'under casual circumstances that require little involvement, like home viewing' (Fitch et al., 1993, p. 48). Such a generalization makes no allowance for the fact that home viewing often involves parental mediation, or co-viewing with siblings or peers, both situations offering more potential for active interpretation than many laboratory experiments. One would expect more exploration of the role of parental mediation in the determination of the reality status of television; comparatively little has appeared in the research literature focusing on this specific topic, though some evidence links parents' opinions and children's beliefs about the reality status of television programmes (Greenberg & Reeves, 1976; see also Messaris & Kerr, 1983, p. 177). A study by Paul Messaris and Dennis Kerr (1983) noted the significance of family viewing styles. They found that families favouring free inquiry and self-expression for the whole family tended to be more sceptical of TV reality than those giving priority to family harmony and obedience to authority (Messaris & Kerr, 1983, pp. 180, 191).
Potter (1988) suggests that researchers should consider not only degree of contact with others, but also other variables such as locus of control, authoritarianism and anomie (Potter, 1988, p. 38). He adds that there may be individual differences in viewers' competence or strategies in interpreting reality cues in programmes.
A great deal of psychological research into children's understanding of TV portrays 'perceived reality' as an 'intervening variable' mediating 'the effects of television' on viewers (e.g. Feshbach, 1972, Hawkins, 1977, Reeves, 1978): according to this model, the more 'real' viewers perceive programmes to be, the greater the influence of these programmes is likely to be on their behaviour and/or attitudes. This model springs from a 'media effects' tradition which presents viewers as passively absorbing messages from programmes in which fixed meanings are embedded. The model has been criticised by more recent commentators who have stressed the active and dynamic interpretation of programmes by viewers, and the openness of programmes to such interpretation (within the scope of this review, such commentators include Hodge & Tripp, 1986, Buckingham, 1993, and Howard, 1993). Nevertheless, even semiotic theorists tend to argue that in making modality judgements, 'the more reality you attribute to a message, the more likely you will be affected by it in some way' (Howard, 1993, p. 43).
There is certainly empirical research supporting this stance. Seymour Feshbach is the main exponent of catharsis theory, according to which watching violence on television or film can reduce aggression in the viewer. In a widely-cited experiment (Feshbach, 1972, pp. 333ff; the 'Second Experiment'), forty children of 9- to 11-years-old were shown a 6-minute sequence on television depicting campus violence. It was in fact an edited compilation of news and movie clips, but prior to viewing one group of twenty children was told that it was from a news broadcast, whilst another group was told that it was fictional. Subsequently, the group which had been told that the clip was from the news showed significantly more aggression than a control group which had not seen it at all, whilst the group which had been told that the clip was fictional showed significantly less aggressive behaviour than the control group. In passing one should note the observation by Sawin (1981) that a better control would have been a group which saw the film but without its reality status being indicated. Aggression was measured with an adjective checklist, a behavioural measure and a measure of aggressive values (a short attitude scale). The behavioural measure of aggression involved individual participants punishing another (accomplice) player's errors in a guessing game by pressing buttons to generate various uncomfortable levels of noise. The only test administered both before and after viewing was the adjective checklist and this did not show any conventionally significant difference. Feshbach's experiment is frequently cited as evidence that film or television programmes perceived as 'real' may have more influence on viewers (or more specifically, children) than film perceived as fictional.
Investigating children's understanding of the reality status of television programmes is far from easy. A major problem for researchers is that young children may not always be able to explain what they mean by saying that events on television are 'real'. Aimée Dorr (1983) found that children were only consistently able to do so by the 6th grade (around 11- or 12-years-old). Frustratingly, the most dramatic advances in children's understanding of television occur before this age. Children's systems of classification do not always match those of researchers. Some commentators (pace Hodge and Tripp) have noted that distinctions between fantasy and reality may not always be prominent in a child's way of interpreting television (Morison & Gardner, 1978). Susan Howard notes that in her study of primary school children, children judged some programmes as realistic simply because they liked them (or unrealistic because they didn't), whilst for others the funnier the programmes, the less realistic they were regarded as being (Howard, 1993, pp. 44, 49 -50).
Regarding questions of methodology, many researchers have made use of standardized questionnaires and attitude-scales, generating data to which some have subsequently applied such statistical techniques as factor-analysis (e.g. Hawkins, 1977, Nikken & Peeters, 1988). Others would agree with Hope Klapper (1981) in emphasizing the limitations of closed-ended questions in investigating this topic with children.
Many researchers in this field also employ interviews with children and discussions between them (see Buckingham, 1991) but, of course, problems remain with interpreting and generalising from such data. Several researchers have developed picture-sorting tasks in which children group or rank pictures of television characters according to how 'realistic' or 'unrealistic' these characters (or the programmes in which they appear) are judged to be by individual children or groups. Sometimes experimenters require individual children to make 'forced-choice' decisions regarding which of the paired pictures offered is more or less 'real' (e.g. Morison et al., 1981); other techniques seek small-group consensus (e.g. Howard, 1993). It should be noted that what can be most enlightening about such techniques is not so much the categories to which particular programmes are assigned, but the justifications offered in the process of doing so. Nevertheless the classification criteria involved are not always easy to determine. Indeed, referring to 'criteria' as such may tend to suggest a rather more rational and detached process of 'judging the reality status of programmes' than seems likely for any viewer under normal conditions.
Many researchers make use of video-clips which are chosen to highlight reality issues, but the obvious objection is that watching such decontextualised clips is quite unlike the ways in which children watch at home. Similar criticisms apply, of course, to all experimental studies of this topic (and to laboratory-based research in particular). In a case-study of a 6-year-old child, David Tripp (1992) has offered a valuable cautionary tale which highlights the limitations of the approaches of adult experimenters, and which new researchers in the field would do well to read. Susan Howard argues that 'because of the subjective nature of modality judgements, the clear, unambiguous identification of concepts which could form the basis of an empirical test of children's perceptions and understandings is not possible' (Howard, 1993, p. 43). But such reservations should not dissuade researchers from continuing to seek provisional and situated understandings of this important topic.
David Buckingham (1993) has argued that 'modality is something we do rather than just something we "know"': stressing that we need to investigate children's modality judgements as a social act (Buckingham, 1993, p. 234). Whilst this is undoudtedly a pressing need, we have as yet few directly imitable and adequately-theorized examples of how this can be effectively accomplished in a way that allows us to build upon the achievements of other researchers and thus to acknowledge the social nature of academic research as well as of children's acts of making meaning. My own interpretive framework in writing this review paper is as disputable as that of anyone else, but I hope that this exploration of the criteria which various researchers have suggested as being involved in children's judgements of the reality status of television programmes will at least encourage others to relate their own frameworks to those of other researchers. Without such attempts at cross-referencing our findings, our collective understanding of this field stands little chance of becoming significantly deeper.
The preferred form of citation for the online version of this paper is as follows:
This page has been accessed times since 23rd October 1995.