Jennifer G. Wishart, Professor of Special Education
Moray House Institute of Education
The University of Edinburgh
Holyrood Road
Edinburgh EH8 8AQ
Scotland, UK
  Reprinted with the permission of the author and the Down Syndrome Research Foundation and Resource Centre
Originally published in Down Syndrome in the 21st Century: 1st Biennial Scientific Conference on Down Syndrome
© 1998 Down Syndrome Research Foundation and Resource Centre

Development in children with DS: maximising potential

Down syndrome (DS) is relatively well known as a genetic disorder to the general public and children with this syndrome form one of the most readily identifiable groups of children with moderate to severe learning difficulties. It is over 130 years since Langdon Down first described DS and 30 years since the presence of the defining extra copy of chromosome 21 was identified by Lejeune and his team of French geneticists. The adverse effects that DS has on children's development are widely assumed to be well understood by now and the belief that DS is invariably associated with very limited abilities is deeply entrenched in the public's mind. Many professionals who work with children with DS are as susceptible as to the man on the street to the stereotype of children with DS as loving, sociable but not very bright children, and indeed even parents of children with DS will often agree that such a stereotypical description broadly fits their own child - although paradoxically considering it inappropriate to describe other children with DS (Wishart & Johnston, 1990).

The widely held stereotype of children with DS reflects a belief that children with DS are all very similar in their ability levels and personalities. Little scientific evidence supports either of these generalizations and findings from a significant number of studies directly contradict them. This paper will focus mainly on learning ability in children with DS but studies of personality and temperament also support the view that there is considerable individual variability (for reviews, see e.g. Cicchetti & Beeghley, 1990; Stratford & Gunn, 1996). In the case of cognitive abilities in early childhood, evidence for the existence of a very wide range of ability levels is particularly long-standing. Some children with DS experience profound levels of mental handicap but the majority fall into the moderate to severe range, with a minority demonstrating borderline normal levels of intelligence. Reviews of longitudinal and cross-sectional psychometric studies of IQ levels (e.g. Carr, 1985, 1995) show that the IQ scores in children and adults with DS can vary over 50-60 IQ points. This is not very different to the spread of IQ in the rest of the population, albeit displaced to the lower end of the normal distribution.

This last statement should not in any way be taken as implying that most children with DS do not experience significant learning difficulties throughout childhood. Some children do indeed do remarkably well in school and go on to cope well with semi-independent living in adulthood, achieving all sorts of skills previously thought to be beyond the reach of those with DS. Access to early intervention services, better health care, family support services and, most importantly, access to proper schooling in recent decades have all contributed to lifting overall levels of achievement. Claims for what is potentially achievable by anyone with DS given appropriate educational and social support are sometimes immoderate, however, and it is important to recognise the severe biological limitations placed on development by the underlying genetic anomaly. The number of genes implicated in the syndrome is large and the presence of extra copies of these leads to harmful gene dosage effects which undermine many aspects of the developmental process from the very earliest stages onwards. However, the wide variation in achievement levels seen in those with DS at all ages also demonstrates that in itself DS does not necessarily and inevitably result in profound or severe levels of mental handicap. Clearly other factors must be playing a significant role in impeding or facilitating progress in individual children.

Identifying those factors and understanding how and when they impact on developmental processes is a major task for researchers interested in DS and especially for those with a specific interest in trying to help children who have DS to make the most of their abilities. Throughout life, the process of development reflects a complex interplay between genes and the environment. Chromosome 21 is one of the smallest human chromosomes and although present in triplicate rather than duplicate in DS (also known as Trisomy 21), all of the triplicated genes are 'normal'. They also represent only a tiny proportion of the genetic complement which children with DS inherit from their two biological parents. This notwithstanding, standard Trisomy 21 shows only a weak correlation between the degree of learning disability experienced and parental IQ levels, inheritance factors which normally strongly influence children's achievement levels. This suggests that other non-biological factors, including psychological factors, may play an important role in determining the progress made. Since family socioeconomic status also shows only a weak correlation with difference in developmental outcomes, clearly any environmental factors in play are more subtle than those which are usually associated with childhood achievement levels.

The learning environments experienced by children with DS, along with individual differences in response to these, would seem obvious targets for further investigation. Despite the very large number of children who have DS and the levels of difficulties they face in their learning, surprisingly little psychological research addresses these problems directly. In recent years in particular, children with DS have more often been used as controls in studies of normal cognitive development than studied in their own right. Implicit in this use is an assumption that DS development is basically a slowed-down version of normal development. This assumption is difficult to reconcile with the specific pattern of disruption to brain functioning and development now known to be associated with DS. This clearly underpins many of the children's cognitive difficulties but there is also now some evidence that they may add to their inherent disadvantage by becoming increasingly reluctant learners as they grow older. Psychological factors may therefore be playing an important role in the failure to maintain developmental rate which is characteristic of later years in DS. In contrast to biological factors, these psychological factors may well prove to be open to facilitative intervention.

Some basic facts about children with DS

DS may be one the best known of all of the learning disabilities and yet there are still probably more misconceptions about DS than about almost any other condition affecting children's mental development both amongst the general public and within many of the professions which come into contact with children with DS. Anyone working in the field who had a dollar for every person who said to them 'you work with children with DS - how nice - such lovely children - so affectionate, such a good sense of fun, great mimics, and so fond of music....' would be very rich by the end of their career. While children with DS might perhaps be seen as fortunate that the public stereotype of them is so positive, stereotyping by definition denies individual differences and it is this ironically that is one of the most interesting aspects of the syndrome - just how much variability there can be in the development of children who have DS.

DS is also still seen as a predominantly medical problem by many people, with those having the syndrome often described as 'sufferers' for whom it is assumed there is no 'treatment'. In reality, however, the majority of children with DS in this and future generations can safely be predicted to live much longer and healthier lives than most of their predecessors did and for many visits to the doctor are little more frequent than for anyone else their age. There are numerous reasons for this, not least of which is the unprecedented explosion over the last few decades in knowledge of how DS affects health and development. Huge advances have occurred in almost every scientific discipline with any bearing on DS and in most of the fields in medicine of importance to the health and well-being of those born with DS.

Advances in cardiology and in immunology have had perhaps the biggest impact. In the 1930s few children with DS survived into their teens but average life expectancy is now very little behind that of the rest of the population, 60+. There are also now several reports of people with DS who are still enjoying life well into their 70s and even 80s, with no greater diminution in their capacities than would be expected at such an advanced age. Although it has been shown that Alzheimer's disease is strongly associated with DS and there is clear evidence of distinctive degenerative changes in the brain at earlier ages in those with DS than in the rest of the population, this pathology does not seem inevitably to result in the expected clinical signs of dementia. This provides a sharp reminder of how little we still understand about the precise relationship between brain and behaviour. The sophistication of genetic technology and the powerful imaging techniques of the neurosciences have brought huge advances in our understanding of how DS impacts on mental development but what is not known still greatly outweighs what is known.

The new genetics has of course also brought with it the promise of less invasive and increasingly accurate methods of prenatal screening for DS. Before serum screening became available, maternal age was the only known indicator of risk of a DS pregnancy and expensive diagnostic procedures were therefore offered only to women over 35. 'Older' mothers account for only a minority of all births, however, and as a result two thirds of all children with DS were being born to unscreened, younger mothers. Today, local health authorities in many countries provide serum screening services to all pregnant women over a lover cut-off age and in some areas screening is available at all ages. Private screening services offering a postal service have also appeared on the scene, offering screening tests to anyone with a Visa card and a friendly GP. This increased availability of screening has undoubtedly had a big impact on DS incidence rates although exact statistics and patterns are difficult to establish. In the UK though, when only maternal age was used to estimate risk, around 1000 babies a year were born with DS; with serum screening, current figures suggest this has dropped to around 500-600 DS births per year.

In assessing the significance of these changes, however, there are two facts we should not lose sight of: there are a great many children who already have DS, and screening notwithstanding, significant numbers of children are still likely to be born with DS in the foreseeable future. Many developed countries do not yet provide routine screening and even where they do, this is not the only factor influencing incidence rates. In the UK, for example, demographic studies predict that incidence rate will rise rather than fall, to levels higher than at any point in the last 20 years; population prevalence is also expected to reach its highest ever level by the year 2000 (Nicolson & Alberman, 1992). There are many reasons for this, not least the shift in child bearing to later years in this generation. In Scotland last year, for example, first births to mothers over 30 increased by 150%.

It is crucial not to lose sight of any of these facts nor to allow acknowledgement of recent major inroads into understanding of how DS affects development to lull us into complacency that the ability to apply all of this new knowledge to the benefit of those who already have the disorder will necessarily follow. By its very nature, DS is an unlikely candidate for gene therapy, for instance. It is estimated that there are around 1000-1500 genes on 21, although only a section containing 650 or so genes is implicated in the characteristic DS phenotype and possibly only 50-100 of these in the mental retardation associated with the syndrome. We still have no idea what the majority of these genes do, either separately or in interaction with other gene products. Of those identified thus far, however, all have major adverse effects on CNS development when present in triplicate and it is likely that some compromise development from the moment of conception onwards. Unless these adverse effects can be reversed in later years or gene therapy can be implemented at a very early stage in pregnancy, this form of intervention is likely to be of limited utility in the case of DS, no matter how much we would wish otherwise.

The key problem for children with DS - learning

Many scientists interested in DS are interested in the genetic disorder in its own right and for many their research does not involve any direct work with those who have DS. To developmental psychologists, however, the core defining feature of DS is the effect it has on the ability of those who have it to learn. It is perhaps worth reminding ourselves just how little we have progressed in this particular area. Unpopular though it may be to say so, the evidence is that most programs of early intervention have not yet had the impact that we would have hoped for on children's learning skills and on achievement levels (see e.g. overviews by Gibson & Harris, 1988; Spiker and Hopmann, 1996). Although the supportive value to parents of participating in such programs cannot be overestimated, in many cases the effect is often simply to bring forward the age of acquisition of skills which are normally delayed but would appear in due course. Skills which are truly cognitive - and language skills - have so far proved much more resistant to early intervention, a fact which is also unfortunately largely true in respect of educational success in these areas in later years. Many children with DS do not progress beyond the intellectual capabilities of the average 6-8 year old and a significant number do not achieve even that. Children with DS may be leading healthier and happier lives and benefiting greatly from the growing acceptance of their right to take their place in the community but psychologists have so far made disappointing progress in identifying ways of helping them to compensate for the very real problems they encounter in learning basic childhood skills.

When working only with children with DS, it can be all too easy to see only the progress being made and to lose sight of what is not being achieved and indeed may never be achieved. It is therefore especially important in evaluating findings from empirical studies of children with DS to have access to data from concurrent control groups of typically-developing children. Many studies, however, simply evaluate outcomes by comparing them with developmental norms for children without learning disabilities or with other findings in the DS literature. For lots of reasons, even when control groups have been used in empirical studies of development in DS, studies have more often been cross-sectional than longitudinal in design - looking at a group or groups of children across a specific age range but testing each individual child usually only once, or sometimes twice. Many of these have focused on a very limited age range of children and on the 'ifs and whens' of learning rather than on the actual dynamics of development. Averaging snap-shot data from cross-sectional studies of children of differing age levels is a very crude way of investigating the developmental process and given the huge variability in rate known to characterise development in DS, it is a method which can at best provide very limited developmental information. Longitudinal research is time-consuming and expensive and with children with DS often scattered across a wide range of schools, large scale studies are often logistically problematic. This kind of research is nevertheless crucial to achieving any real understanding of developmental processes in DS.

In our own research, we have used both cross-sectional and longitudinal approaches, often in parallel in order to compare the accuracy of the developmental information provided by the two methods. Longitudinal work invariably provides less ambiguous data than cross-sectional studies but as some of our longitudinal data demonstrate, this approach can have its problems too, with much depending on the frequency with which development is sampled and on the level of detail used in analysis (see below). Unless qualitative as well as quantitative aspects of performance are taken into account, longitudinal data is equally capable of providing misleading information about the nature and course of development in DS. At the level of the individual child with DS, it can also provide a very inaccurate account of both current cognitive status and likely future potential.

Developmental differences in children with DS

This paper looks at how young children with DS approach the task of learning rather than at what children with DS can and cannot learn. In particular, it looks at the motivational deficits that they often show in learning contexts and examines to what degree they may effectively add to their pre-existing cognitive handicap by adopting counterproductive learning behaviours. There is little that can be done about the harmful effects the extra chromosome has on their development, but adverse learning styles should be much more open to intervention. Children with DS may in fact be making learning much harder for themselves than it really needs to be. Many of the children we see actually seem to go out of their way to avoid opportunities to learn: the more experience of learning they get the more they seem to rely on the help of other people - even when they do not need it - and the less willing they appear to be to take any of the initiative in problem-solving.

There is a long-standing debate over whether development in children with DS can best be understood in terms of a slowed-down version of normal development, i.e. with only the rate and endpoint distinguishing development in children with DS from normally-developing children, or whether development in DS differs in fundamental ways from normal developmental processes. The data to be presented below support the view that there are significant differences in how development unfolds in children with DS, and also suggest that some of these may stem as much from crucial differences in the psychological environment in which children with DS grow and learn as from the biological disadvantages they carry from birth. This 'difference' account may not be the favoured viewpoint, but it does have the considerable merit of being consistent with the growing body of data from the neurosciences showing significant differences in the structure of the DS brain and in how it works (see e.g. Wisniewski, Kida & Brown, 1996; Nadel, 1996; Wishart, 1996). We are still a very long way from being able to make direct links between specific brain pathology and specific behaviours, least of all cognitive behaviours, but the characteristic reduction in size of the cerebellum, for instance, is likely to underlie the poor muscle tone and motor sequencing deficiencies commonly found with DS, and the immaturity found in frontal and temporal lobes is almost certainly implicated in the significant memory deficits and impairment in spatial representation and temporal sequencing skills seen in those with DS. Abnormal development in these areas might also be implicated in the accelerated aging process found in DS.

This inherent biological disadvantage can only be compounded by the adverse psychological effects of the protracted nature of the children's learning experience in virtually all developmental domains. Their frequent exposure to failure in new learning contexts and the typically low expectations of many of their partners in learning cannot help either. Provided with a far-from-perfect set of tools for learning and with a quite different psychological environment in which to learn, it would surely be remarkable if development in DS children did not follow different pathways to those seen in normally-developing children from a very early stage onwards.

It is important to remember that this difference delay-debate is not simply of theoretical interest to academics: it has major implications for educational practice. If development is in fact different in children with DS, attempts to ameliorate the effects of DS on development will need to recognise these inherent and acquired differences and reflect them in the design of teaching methods. Without an understanding of the natural learning 'style' of children with DS, our attempts at facilitative intervention are severely handicapped. Intervening in processes that are not fully understood can at best hope to meet with limited success; at worst, it may run the risk of changing slow but willing learners into reluctant, avoidant learners.

Factors undermining developmental progress: the learning 'style' of children with DS

One of the more robust findings about cognitive development in DS is that of a decline in developmental rate as children get older. Declining developmental rate was a consistent finding in some of the very earliest studies of children with DS and is a finding which has continued to emerge in more recent studies, early intervention notwithstanding (see e.g. Carr, 1985, Duffy, 1990; Wishart, 1993a). One of the main aims of our own research program has been to try to pinpoint some of the factors underlying this decline in DS developmental rate. Space constraints allow only a very brief presentation of some illustrative data but full details of the relevant studies can be found in the original papers. The children with DS in our various studies have ranged in age from birth to 14 years and numbers have varied from 10 to 50 in individual studies. In longitudinal studies, data collection had extended from 1 to 5 years depending on the age of children at entry to the study and the nature of the study in question.

In all of the studies, control groups of children without DS have also been tested in order to allow direct evaluation of difference-versus-delay theories. Typically, these groups of normally-developing (ND) children were matched with the children with DS on either chronological age or stage in development depending on the focus of the particular study. As will be apparent from some of the findings, however, 'matching' is a somewhat misleading and sometimes meaningless concept when it comes to DS. Children with DS simply do not respond in test situations in the same way as ND children, even when they achieve similar 'scores'. Children with DS do not even provide satisfactory matches for themselves: often the content and the quality of their responses in two identical, closely-spaced testing sessions fluctuate greatly, with successes in the first session no longer demonstrated in the second and 'failures' miraculously turning into successes on repeat testing.

Findings from two sets of studies - on operant learning and on object concept development - are outlined below, both sets demonstrating quite well some of the key features which would appear to characterise early cognitive development in children with DS:

Settling for the easy option: performance on operant learning tasks

Our studies of operant learning provide the most direct evidence that very young children with DS can quickly learn to depend on the support of others in learning contexts, even when that support is not needed. For anyone not familiar with the experimental paradigm, operant learning studies investigate the ability of babies and young children to detect that their activity has caused something to happen. Understanding this kind of relationship is basic to most learning and is also essential to the formation of any belief in self efficacy - the belief that you can have some control over what happens around you and to you. 50 children with DS aged between birth and 2 years of age took part in the studies which were both cross-sectional and longitudinal in design (see Wishart 1991; 1993a). Because children with DS often have poor muscle tone in the early years, a task suitable for use with the youngest and least motorically-able children was designed. The children sat securely in a baby chair in such a position that if they kicked either foot through 60 degrees this would break a light beam, causing a 1-second rotation of a brightly-coloured mobile. In this sort of situation, when the contingent relationship is detected, a child's kicking rate typically rises, sometimes accompanied by smiling or excited vocalisations. The criterion we adopted for having detected the contingency between kicking and the turning of the mobile was a 1.5 increase in each infant's personal baseline rate (Rovee-Collier, 1987).

These studies are basically about control and the exercise of control. There were two experimental variations:

The children with DS did less well on these tasks than ND children of the same chronological age, as would be expected, but although generally slower to detect the relationship between their kicking and the mobile's turns, they were eventually able to solve this problem in all of its various versions. In developmental terms, however, of more interest was how they responded to the non-contingent, 'free' turns of the mobile. With increasing age, the children with DS would often simply sit happily watching the mobile's random turns, only occasionally bothering to kick for themselves. In the longitudinal studies, it might seem tempting to suggest that this was because they had simply lost interest, having been exposed to the task so often. The flurries of activity seen when either the 'free' turns were switched off or when the mobile was turned off completely suggested the opposite, however, while also confirming that they were still aware of their potential to control the mobile and still keen to see it go round. In the non-contingent part of testing sessions, it seemed rather that the children were effectively relinquished their own control over what was going on around them. Providing encouragement in the form of the occasional 'free' turn was leading to a decline in self-generated activity at older ages, the exact opposite of what we would seek to achieve.

The data from cross-sectional subjects provided additional evidence of a growing unwillingness to take much initiative in what was happening around them. The children with DS tested at 6,12,18 and 24 months in these studies also sat happily through most of the 18 minute testing sessions with very little complaint, alert, apparently interested and often showing signs of pleasure at the rotation of the mobile, irrespective of whether this was being generated by themselves or by the computer. In marked contrast, the ND children we tested (matched for either CA or MA) were far less tolerant and far less passive. By a developmental age of 6 months, as soon as they had detected the nature of the contingency between their actions and the turns of the mobile, they retained little interest in any version of the task and unambiguously expressed this. Sessions had to be quickly terminated, although protests typically subsided on being allowed out of the chair and the children would generally be happy to go on to work at other problem-solving tasks, such as object concept tasks.

Investing effort in searching for hidden objects

The second set of illustrative data is drawn from our long-running studies of object concept development in children with and without DS (see e.g. Wishart & Bower, 1984; Wishart & Duffy, 1990; Wishart, 1993a, b). For the purposes of this paper it is not necessary to go into any great detail about object concept development itself but it is important to note that the cognitive skills crucial to succeeding on object concept tasks are considered by many psychologists to be key ones. All of us have to learn at some early point, for example, that objects exist independently of our actions and continue to exist even when we cannot see them or act upon them. We also have to learn that objects cannot be in two places at the one time and that they all have unique identities - while two objects seen at different times may look identical, they are not necessarily one and the same object. Understanding these key physical properties of objects and the laws which govern their interactions is central to making sense of even the simplest of everyday events. This applies equally to understanding the actions of 'social' objects, for example if a person goes out of sight by leaving the room or hides a favourite toy by closing their hand over it.

A series of hiding tasks of increasing complexity are usually used to assess how much infants understand about objects and the physical laws that govern their movements. Most versions of these are based on the original tasks described by Piaget more than sixty years ago. The easiest tasks involve hiding a small, attractive toy fully or partially under a cloth or cup; in the most difficult tasks, the child must choose from two or more identical occluders, after a hiding sequence in which either the position of the toy or the occluders has been changed.

30 infants with DS aged between birth and 2 years 9 months took part in our longitudinal study. We used 4 levels of task and although there was already evidence to suggest that children with DS are considerably delayed in acquiring each of the stages in object concept development, we - fortunately - decided to try all levels of task with all ages of subject. Testing sessions were fortnightly, with the same tester being used on all occasions.

Findings have been reported in full in a number of papers (see above) and space limitations allow only key features of the data to be outlined here. The data gathered on what is widely known as the AAB or A-not-B task, the second easiest of the 4 tasks, however, give a good idea of the flavour of findings. This task is typically solved by the average child somewhere between 8 and 10 months. A toy is hidden three times, in one of two identical occluders positioned at A and B, in the sequence AAB. There is no sleight of hand and the infant sees the object being hidden each time. Success requires the infant to search at B on the B trial on 4 out of 4 trials; in longitudinal studies, success is required in two consecutive sessions. Prior to success, the characteristic error made by young children is to continue to search at A on the B trial, i.e. at the place where the toy has previously been found. Repeated experience with this sort of problem can lead to success at a younger age, but the characteristic error usually still reliably appears in testing sessions prior to this success, regardless of actual age at time of first success.

Age of first success to criterion - 4/4 correct on two consecutive testing sessions - on this AAB task was 7¾ months for the control children in our studies and 10½ months for the children with DS but the range of age of success in the children with DS is also worth noting: 7¼ - 14 months. In sum then, the children with DS may on average have taken longer to achieve their first success on this task but the mean age at which they did so was not far off cross-sectional norms for age of acquisition of this stage with a small number in fact succeeding at surprisingly early ages. This was true of performance on higher level tasks, too (although we shall have to ignore this aspect of the data here).

Examination of post-success performance profiles provided a less encouraging picture, however, with performance on this same task fluctuating from session to session. Subsequent failures were sometimes the result of a refusal to engage in sufficient trials to be credited with a pass (4/4 correct final searches at B) rather than because of clearly erroneous search on any trial and sometimes success could be restored by hiding chocolate or a rusk instead of a toy, although this strategy was not always successful. Quite a number of the young children with DS we tested, however, although watching the hiding carefully and clearly capable of precise search, would simply either sweep both cups to the floor or pick the same cup on each trial, a very low-level strategy which at best would give them a 50% return rate (and would, of course, count as a 'fail').

Counter-productive behaviours were also frequently produced in response to tasks which were still difficult for the children, 'difficult', that is, in terms of their current developmental level. In this case, engagement was more clearly withdrawn, with difficult tasks often actively avoided after only one or two trials, either with protests or by resorting to diversionary strategies, such as pretending to be very interested in something else, producing some sort of 'party-trick' to divert the tester into some other, off-task activity (such as blowing raspberries or clapping hands), or just turning on the charm. It was striking how well this behaviour fitted the DS stereotype - but it was only being turned on at specific times, when the going got hard. It stuck out because of its inappropriateness - not because it was typical of behaviour at other times.

These avoidance behaviours cannot simply be written off as perfectly normal responses to being repeatedly presented with a task which is now 'easy' or being faced with a task that is just too difficult. The control children typically worked hard at all levels of these tasks, whether they were above or below their current developmental level in terms of difficulty. There were minor lapses but it was usually immediately obvious interest was only temporarily lost, and attention - and success - could usually quickly be restored by hiding rusks or chocolate. Ordinary children typically enjoy showing off their abilities and we must ask ourselves why is it that even when success could be within their grasp, children with DS seem to prefer to respond in a way that avoids running the risk of making any error.

In relation to the developmental-delay debate, it is worth noting that these fluctuations in performance could have lead to a very different picture of development emerging if behaviour in the children with DS in these studies had only been sampled at six-month intervals rather than every 2 weeks. At 12 months, for example, it would have been easy to conclude mistakenly that many of them had not yet reached the AAB stage in development, a plausible enough interpretation given that the average child without a learning disability does not reach this stage until 8-10 months. Some of our current knowledge about cognitive abilities in children with DS may therefore be suspect if based only on findings from cross-sectional studies of development.

These diversionary and delaying tactics, the non-committal responses, the misuse of social skills - all of the behaviours we have seen at these earliest stages in development - have also re-appeared in various guises when we have worked with older children with DS. The children's responses at later ages to standardised 'IQ' tests have provided some particularly clear examples of what we have called 'cognitive avoidance' strategies. In some cases, when we have tested children on two separate but closely-spaced sessions, we have found IQ performance to vary on as many as 30% of the test items presented, with many children passing items previously failed and failing items previously passed. This has often meant that two very different 'IQ's resulted from these two testing sessions, neither presumably reflecting accurately the full range of skills available to the child at that point in his or her development. Sometimes of course the two effects cancel each other out, giving overall scores which are very similar on both occasions and a false impression that performance is stable and test scores are valid.

These sorts of findings should make us very wary about the confidence which can be placed in findings from studies in which DS and non-DS subjects were 'matched' for IQ on entry to the study. There are also implications at the level of the individual child. When a child with DS is uncooperative and refuses to complete test items, it is always tempting to classify such a child as 'untestable', a label which does nothing to increase our understanding of the child's developmental problems or of the nature of development in DS. Children with DS are seldom given the benefit of the doubt that when they do not perform on a given task that it may be a case of 'won't do' rather than 'can't do'; it is much easier to assume that the tasks is just too difficult for them yet.

How fluctuations in performance profiles on IQ tests relate to developmental progression is difficult to establish. In our studies, it was clear that new skills were indeed being gained but that the effects of these on test scores was often being cancelled out by a failure to reproduce skills gained at younger ages. There was then clearly the potential for constant progress - but only if each new skill had been consolidated into the repertoire and used to the full. For children with DS, there may well be fundamental problems in consolidating new skills; when finally acquired, new skills may simply not be as readily available to them as to children whose development is progressing normally. This may be because of an inherent instability in the developmental process itself (as Shapiro (1975) and others have suggested for some time now), or it could be because of a complex interaction between inadequate motivation and inefficient learning processes: that is, poor levels of motivation may be contributing directly to the growth of deficits in cognitive functioning. Skills after all need to be practised if they are to become firmly embedded in the repertoire. If the inefficient reproduction of skills seen in testing situations is typical of how children with DS use their skills in the real world, it is perhaps not surprising that they fail to realise their full developmental potential.


What then should we conclude from all of this? Firstly, it would seem that:

Psychological as well as biological factors clearly play a role in influencing what happens - or does not happen - in cognitive development in children with DS. The wide differences seen in ability levels are not likely to stem simply from 'brighter' children having had more opportunities to learn than others - early intervention, professional input, and full-time schooling are now the norm. We are still a very long way from understanding why it is that some children with DS succeed in mastering so much more than others but looking much more closely at the contexts in which the children learn and at the ways in which we support their learning might lead to at least part of the answer.

It will be important that any lessons to be learnt from research into cognitive development in children with DS should also be appraised for its potential relevance to older age groups. Many of the psychological obstacles to learning revealed in early childhood must also operate in later life. If anything it seems even more likely that at later ages motivational deficits and underperformance will depress both the acquisition and development of new skills. While acquiring cognitive skills is clearly a major obstacle for those with DS, it is important that all of our attempts at interventions should not focus too narrowly on promoting only the development of these skills. There are many other skills which are at least as important if our aim is to help people with DS to lead fulfilling lives and to be fully involved in the life and work of the community around them. Non-cognitive skills should not be neglected and may indeed prove to be more responsive to intervention than cognitive skills. By focusing too exclusively from a very early stage on the acquisition of only cognitive skills - and particularly academic skills - we may be reinforcing the children's feelings of failure - thereby increasing the chances that they may become reluctant learners - happier to give up than to risk failure when presented with unfamiliar problems.


Carr, J. (1985). The development of intelligence. In D. Lane & B. Stratford (Eds.). Current Approaches to Down's Syndrome (pp. 167-186). London: Holt, Rinehart & Winston.

Carr, J. (1995). Down's Syndrome: Children Growing Up. Cambridge: Cambridge University Press.

Cicchetti, D. & Beeghly, M. (Eds.). (1990). Children with Down Syndrome: A Developmental Perspective. New York: Cambridge University Press.

Duffy, L. (1990). The relationship between competence and performance in early development in children with Down's syndrome. Unpublished Ph.D. dissertation, University of Edinburgh.

Gibson, D. & Harris, A. (1988). Aggregated early intervention effects for Down Syndrome persons: Patterning and longevity of benefits. Journal of Mental Deficiency Research, 32, 1-17.

Nadel, L. (1996). Learning, memory and neural functioning in Down's syndrome. In J. A. Rondal, J. Perera, L. Nadel, & A. Comblain (Eds.): Down's Syndrome: Psychological, Psychobiological and Socio-educational Perspectives (pp. 21-42) London: Whurr.

Nicholson, A. & Alberman, E. (1992). Prediction of the number of Down's syndrome infants to be born in England and Wales up to the year 2000 and their likely survival rates. Journal of Intellectual Disability Research, 36, 505-518.

Rovee-Collier, C. K. (1987). Learning and memory in infancy. In Osofsky, J. D. (Ed.). Handbook of Infant Development (2nd ed.) (pp. 98-148). New York: Wiley.

Rovee-Collier, C. K. (Eds.), Advances in Infancy Research: Vol.3. (pp. 57-123). Norwood, New Jersey: Ablex.

Shapiro, B. L. (1975). Amplified developmental instability in Down's syndrome. Annals of Human Genetics, 38, 429-437.

Spiker, D. & Hopmann, M. R. (1996). The effectiveness of early intervention for children with Down syndrome. In M. J. Guralnick (Ed.), The effectiveness of early intervention: Directions for second generation research (pp. 271-306). Baltimore: Brookes.

Stratford, B. & Gunn, P. (Eds.), (1996). New Approaches to Down's Syndrome. London: Cassell.

Wishart, J. G. (1991). Taking the initiative in learning: a developmental investigation of infants with Down's Syndrome. International Journal of Disability, Development and Education, 38, 27-44.

Wishart, J. G. (1993a). The development of learning difficulties in children with Down's syndrome. Journal of Intellectual Disability Research, 37, 389-403.

Wishart, J. G. (1993b). Learning the hard way: avoidance strategies in young children with Down's syndrome. Down's Syndrome: Research & Practice, 1, 47-55.

Wishart, J. G. (1996). Avoidant learning styles and cognitive development in young children with Down's syndrome. In B. Stratford & P. Gunn (Eds.), New Approaches to Down's Syndrome (pp.173-205). London: Cassell.

Wishart, J. G. & Bower, T. G. R. (1984). Spatial relations and the object concept: a normative study. In: L. P. Lipsitt & Wishart, J. G. & Duffy, L. (1990). Instability of performance on cognitive tests in infants and young children with Down's Syndrome. British Journal of Educational Psychology, 59, 10-22.

Wishart, J. G. & Johnston, F. (1990). The effects of experience on attribution of a stereotyped personality to children with Down's syndrome. Journal of Mental Deficiency Research, 34, 409-420.

Wisniewski, K. E., Kida, E. & Brown, W. T. (1996). Consequences of genetic abnormalities in Down's syndrome on brain structure and function. In J. A. Rondal, J. Perera, L. Nadel & A. Comblain (Eds.): Down's Syndrome: Psychological, Psychobiological and Socio-educational Perspectives (pp. 3-20). London: Whurr.

Acknowledgements: The author's research was carried out with the support of project grants from the Medical Research Council, United Kingdom. Thanks are also extended to the all of the children and families who helped with the studies and to the Scottish Down's Syndrome Association and local health and education authorities for their collaboration