Abstract
Objectives
To determine the age at which children with apparently normal development can recognise the television image of a cat, dog, or baby.
Design
Observational study.
Setting
District general hospital in north London.
Subjects
797 children with apparently normal development aged between 8 and 23 months and 26 children with Down's syndrome aged 18 months.
Outcome measures
Whether or not the child recognised the television image of a cat, dog, or baby by naming, imitating, or pointing at the image.
Results
By 18 months of age 96% (95% confidence interval 94% to 98%) of normal children recognised the television image of a cat, dog, or baby compared with 5 of 26 (19%) children with Down's syndrome.
Conclusion
Recognition of the television image of a cat, dog, or baby is a simple milestone, which may help in the developmental assessment of young children.
Introduction
A child who is slow to talk at 18-24 months of age is likely to be normal but can cause concern about learning disabilities or a severe language disorder. It is sometimes difficult to assess the development of such young children, and there are few relevant and validated milestones in this age group. A pilot investigation led us to hypothesise that determining how much interest children show in television images might provide helpful information about children's development.
Methods
One of us (BL) approached the parents of children who were either patients or siblings of patients at North Middlesex Hospital in north London. We excluded children who were preterm, who had known developmental problems, or whose parents did not speak English or own a television.
We first asked the parents “Does your child recognise the picture of a cat, dog, or baby on the television screen?” If the response was “yes,” we asked “How do you know?”
We concluded that the child recognised the television image if he or she named, imitated, or pointed at it. Becoming excited or patting the screen was not considered sufficient evidence.
The same questions were used by KB when interviewing the parents of 18 month old children with Down's syndrome by telephone. These families were identified by the Family Fund, a national organisation that supports the families of children with disabilities.
Results
We interviewed the parents of 797 children of apparently normal development aged between 8 and 23 months. At each month of age we interviewed the parents of between 36 and 61 children (mean 49.8). In addition, we interviewed 26 parents of children with Down's syndrome aged 18 months.
A statistical model describing the percentage of children at each age who recognised a television image was fitted using logistic regression (figure).
By 18 months of age 96% (95% confidence interval 94% to 98%) of normal children were reported to recognise the television image of a cat, dog, or baby (figure) compared with 5 of 26 (19%) of the 18 month old children with Down's syndrome.
Discussion
Findings and shortcomings
Nearly all (96%) of the normal children were reported to recognise the television image of a cat, dog, or baby by 18 months of age. Some parents may have exaggerated their child's abilities, but the pattern of responses (figure) supports the view that our cut off point of 18 months is meaningful. Our study population seems likely to be broadly representative of British children. We do not think our findings were distorted by the possible inclusion of a few children with unrecognised developmental problems.
We did not question the parents in enough detail to be absolutely sure that the child's pointing was designed to involve the parent (protodeclarative) rather than being a form of instruction (protoimperative). Assuming the pointing was protodeclarative, our milestone is a test of both understanding (does the child recognise the image?) and “joint attention behaviours” (does the child want to share their interest?). The first of these skills is impaired in a child with learning difficulties and the second of these is impaired in a child with autism.
Reynell described the development of the symbolic understanding necessary for the development of language.1 As a child becomes older he or she is able to recognise increasingly abstract representations of an object, including increasingly abstract pictures. The television image of an object is more abstract than pictures in books designed for very young children. Nevertheless, our milestone is not a test of symbolic understanding.
Previous work
Review of the published literature on pointing shows some disagreement about when children recognise images other than those seen on television. According to the most recent manual for use with Griffiths testing, a child “enjoys picture book” by 15 months of age and “likes adult to show book” from 17 months of age.2 Sheridan reported that at 12 months of age a child “points with index finger at objects of interest” and “shows interest in pictures.” At 15 months of age he or she “looks with interest at coloured pictures in book and pats page.” At 18 months of age he or she enjoys simple picture books “often recognising and putting index finger on boldly coloured items on page.”3 In contrast, Reynell reported that at about 20-24 months of age a child of normal development is capable of recognising “clear coloured pictures.”1 Baron-Cohen et al reported that children aged 18 months were likely to be diagnosed later as being autistic if they failed three items of a developmental assessment: protodeclarative pointing, gaze monitoring, and pretend play.4
Usefulness of milestone
When a child with normal vision is slow to pass our milestone, three main developmental disorders should be considered: learning disabilities, a pure language disorder, and autism. The range of normality is such that, as with any milestone, some children of normal development will fail to pass it on time. Equally, some children with learning or communication problems will not be detected by the use of our milestone, as shown by our findings in children with Down's syndrome.
Formal testing systems such as the Griffiths and Bayley tests require training and special equipment.1,5 The Denver developmental assessment test is simpler to use.6 In this test, however, the only test item that is relevant to language development, that does not involve an ability to talk, and that should reliably be achieved before 24 months of age, is the reported ability to help about the house.
We do not fear that asking about our milestone could encourage excessive television viewing. Indeed, we usually couple inquiry about this milestone with suggestions about limiting television viewing.
We consider that our milestone is a useful part of the developmental assessment of children aged 15-24 months. Unlike many milestones, ours is supported by information about a large number of children—roughly twice the number tested over the same age range during the recent revision of the Griffiths's scale.1
Our milestone needs neither special equipment nor special training. This makes it particularly useful for general practitioners and health visitors.
What is already known on this topic
Few simple and validated milestones exist to help assess the development of children under 2 years old who do not talk
What this study adds
By 18 months of age 96% of children of normal development were reported by their parents to show that they recognised the television image of a cat, dog, or baby by naming, imitating, or pointing at the image
Acknowledgments
We thank Mrs Dot Lawton (Social Policy Research Unit) who, acting on behalf of the Family Fund, put us in touch with the families of children with Down's syndrome, and Dr Hilary Cass and Dr Dominic Croft for helpful suggestions on an earlier draft of the manuscript.
Footnotes
Funding: None.
Competing interests: None declared.
References
- 1.Reynell J. Language development and assessment. Lancaster: MTP; 1980. [Google Scholar]
- 2.Huntley M. The Griffiths mental development scales. From birth to 2 years. High Wycombe: Association for Research in Infant and Child Development; 1996. [Google Scholar]
- 3.Sheridan MD. From birth to five years. Children's developmental progress. Slough, Middlesex: National Foundation for Educational Research; 1997. [Google Scholar]
- 4.Baron-Cohen S, Cox A, Baird G, Swettenham J, Nightingale N, Morgan K, et al. Psychological markers in the detection of autism in infancy in a large population. Br J Psychiatry. 1996;168:158–163. doi: 10.1192/bjp.168.2.158. [DOI] [PubMed] [Google Scholar]
- 5.Bayley N. Bayley scales of infant development. 2nd ed. San Antonio: Harcourt Brace; 1993. [Google Scholar]
- 6.Frankenburg WK, Dodds J, Archer P, Shapiro H, Bresnick B. The Denver II: a major revision and restandardisation of the Denver developmental screening test. Pediatrics. 1992;89:91–97. [PubMed] [Google Scholar]