The meaning of risk and protective factors in infancy

European Journal of Psychology of Education 1989, Vol. IV, n:' 2, 161-173 © 1989, I.S.P.A. The Meaning of Risk and Protective Factors in Infancy Hellgard Rauh Free University of Berlin, West Germany Recent psychological research on infants and newborns led to discoveries of early competencies in young infants hitherto unbelieved by most scientists and difficult to account for in traditional models of child development. On the other hand, accumulated empirical longitudinal evidence challenged long-held preconceptions of the enduring impacts of early experience on later development. The risk concept, adapted from pediatrics and psychiatries, was introduced with the expectation of identifying those subgroups of infants that are of increased risk of later behaviouralor developmental maladaptation. The risk concept, however, turned out to be imprecise, having too many different meanings, being too abstract, and applicable only to populations or groups, not to individuals. Combinations of the risk concept with recently developed more sophisticated models of early development, such as transactive and systemic models of development, include such concepts as vulnerability/resilience as stable personality characteristics, protection or risk as setting characteristics, and coping in the face of actual stress as situational process variables. Implications for prevention, intervention and early education are discussed with respect to some selected groups of infants considered at-risk. Infancy, its place in human development «Infancy - Its Place in Human Development» was the title of a hook published by Kagan and his co-workers in 1980 (Kagan, Kearsley & Zelazo, 1980). This book summarised not only the results of a project that the authors had started in 1968, but also signaled a change in research direction in the field of Child Development and Developmental Psychology. Specifically, the very young infant was attributed psychological characteristics similar to those previously reserved for the preschool child. Rather than being viewed as a somewhat passive recipient of attention, care, and stimulation, the newborn was portrayed as possessing an extensive repertoire of social and cognitive competencies (Appleton, Clifton, & Goldberg, 1975; Bruner, 1968; Stone, Murphy, & Smith, 1973). This shift in perspectiveis also reflected in the entries in the Child Development Abstracts. In 1966, the «newborn» and the «neonate» were separately listed and made up one third I would like to express my gratitude to the reviewers for their helpful comments and especially to Jacqui Smith, Max-Planck Institute of Education and Human Development, Berlin (West), and Brian Hopkins, Free University Amsterdam, 'for their careful English editting, 162 H. RAUH of the 60 publications. During the following decade these two index terms faded out and perinatal research became subsumed under the general label «infancy». Until 1969, medicaIIyrelated topics and terms (e.g., anatomy, reflexes, sensory functioning) dominated research papers on early infancy. Psychological topics and terms such as preference for novelty, temperament, social interaction, self concept, communication, imitation, socio-emotional behaviour, and ecological differences emerged after 1973 and quantitatively replaced the earlier contributions. Enthusiasm about the infant's competencies was dampened, however, when accumulated longitudinal evidence seriously challenged the idea that differences in competencies or experiences in infancy had appreciable prognostic value for individual differences at school age or even adulthood (Clarke & Clarke, 1976; Kagan, 1980; Brim & Kagan, 1980; Kagan, Kearsley & Zelazo, 1980). Even severe neglect over prolonged periods of infancy and early childhood could be largely compensated for later in childhood (Koluchova, 1976). Neither motor or cognitive milestones of infant development nor physical or psychical trauma in infancy, were found to predict later outcomes - with the exception of severe neurological damage in early infancy. Consequently, developmental models were re-evaluated under the perspectives of continuity and change or stability and plasticity of behaviour from infancy to adulthood (Brim & Kagan 1980). Concepts such as risk and protective factors, vulnerability and resilience, and prevention and compensation were introduced in an effort to explore in more detail the possible relationships between early and later development. This paper focuses on the risk concept and asks to what extent this concept has both furthered our understanding of the competencies of the young child and contributed to developmental models of continuity and change over the life span. After outlining the dimensions defining the use of the risk concept in infancy research, recent personality concepts, such as vulnerability and resilience, and setting characteristics are considered with respect to their applicability in the infancy period. They are discussed in the frameworks of contrasting models of early development. In a final section, the situation-specific concepts of stress and coping are introduced in their applicability to infants and with their implications for psychological and educational prevention and intervention. The risk concept in infancy research The risk concept entered developmental and educational psychology only recently, at least as an explicit concept organizing books or papers by title or as an index reference (Farran and Cooper, 1986). The subject index in Carmichael's Manual of Child Psychology (Mussen 1970, Vol I & II), for example, had only two entries associated with «risk», namely «risktaking behaviour». Mussen's 1983 Handbook, in contrast, included a full chapter in the volume on «Infancy and Developmental Psychobiology» devoted to «Risk factors in development» (Kopp, 1983). «Risk» as defined in Brockhaus (1978) means venture, danger, or threat, especially the danger of loss while aiming at a special profit or gain, as in a risky financial decision. The origin of the word goes back to ancient Italian and late Latin (Duden, 1981) where it meant either the cliff a boat has to sail around, or, for pedestrians, the root of a tree that one might tumble over. This word, risk, implies metaphorically that the traveller on sea, land or during one's development has some chance to reach his goal without encountering any problems, or to reach it with some extra effort and skill, but that he might also fail totally. Risk implies uncertainty either because of a lack of clear knowledge about the location, the time, and/or the kind of the danger, or because of insufficient skills, competencies, and equipment to deal with an expected danger adequately. As a purely statistical term, «risk» refers to the probability relationship between some defined predictor variable and some negatively valued outcome variable. Such a purely statistical concept of risk is in order as long as the nature of the relationship between the RISK AND PROTECTIVE FACTORS 163 variables is either unclear or unknown, or when they are mediated by events that cannot be controlled or are unknown. Risk, in this sense, usually refers to group or population data, not to individuals. Such a concept of risk is certainly insufficient if decisions have to be made with regard to expensive preventive or interventive measures for an individual case. In social and psychological research the statistical risk concept can only be considered as a first approximation to a problem field that eventually has to be replaced by more sophisticated models of the processes linking specific predictors with specific outcomes. The risk concept in infant development research has been used in several different senses: as a probable negative outcome, as a predictor variable, and as a descriptor of negative life conditions. These different meanings are not always spelled out clearly (Keogh, Wilcoxen & Bernheimer 1986). The term «risk» has been used to characterize a probable negative outcome variable, such as early death (e.g. SIDS). psychopathology, criminal behaviour, drug abuse, mental deficiency. and school drop-out (Parker & Asher. 1987; Meyer-Probst & Teichmann, 1984; Silbereisen & Kastner; 1987). Research in this context is aimed at finding early predictors of increased risk for such negative outcomes. Cumulative screening methods have been suggested as in the «risk-group approach» but also in combination with preventive intervention (Liden, Nichter & Murphy, 1982; Farran, 1986; Gordon & Jens, 1987). Statistical issues, design problems and ethical concerns with this method are discussed by Brandtstadter (1982b), Bell (1986), and Keogh, Wilcoxen and Bernheimer (1986). More recently, even mild kinds of problems, such as behavioural. learning, and attentional difficulties within the scope of normal development, have become a focus of risk research (Grossmann, in press; Rauh 1984). Often, the kind of expected peril is not clearly defined, or at best is described as any reduction of optimal development. This leaves, however, the researcher with the difficult task of defining what is optimal development (Brandtsatdter, 1982a,c; Kagan. Kearsley & Zelazo, 1980). In particular, the problem remains as to whether developmental outcome is to be evaluated relative (a) to social goals (normative, ethical, role-adequate behaviour) as is usually implied in indices of social adjustment, (b) to functional goals (such as adaptive competencies, problem-solving abilities) as might be captured by measures of mental development, or (c) to individual criteria of subjective well-being which have rarely been used in risk research with children. In infancy research the term «risk» is most often applied to specific early predictors of unfavourable later outcomes. Such «risk factors» include chromosome defects, injury to the CNS, perinatal health problems, premature birth, pregnancy and delivery complications, and social, economical, or health problems in the family, but also early behavioural signs in the child, such as irritability or developmental retardation (Kopp, 1983; Werner, 1986). Whereas in earlier research single variables have often been used as predictors of later behavioural or developmental problems, cumulative risk indices, sometimes roughly categorized as «biological» or «psychosocial» risk factors (Meyer-Probst & Teichmann 1984) or the lack of such factors, as in the optimality index (Prechtl, 1980; Michaelis & Haas, 1987), have proven to be more effective (Rauh, 1984). Again, the sensitivity, specificity. and efficacy of prediction of later problem behaviour are far from optimal (Parker & Asher, 1987; Meyer-Probst & Teichmann, 1984). A reduced prediction value could be due to the invalidity of the predictor variables, as well as to the stochastical nature of the relationship between predictor and outcome variables. High prediction values can imply two things: Either the early behaviour or the risk factors captured by the index are causally related to the later problem behaviour, or they are early manifestations or «lead behaviours» of the later problem behaviour (Parker & Asher, 1987). High prediction values also imply high stability and little plasticity of behaviour during early development. If there is, however, an established direct causal relation, for example between a specific chromosome aberration and mental retardation, then terms such as «risk factor» and «risk outcome», are inappropriate, «for it is a certainty» (Wilson & Crough, 1987, p. 267). By definition, one part of «risk» is stochastic - there should also be a positive chance - Whereas the other part- of the risk might be systematic (Wilson & Crough, 1987). 164 H. RAUH The term «risk» has also been used as a synonym for unfavourable life conditions and life events, such as poverty, family discord, radical separation from the family, and a problematic social/urban environment (Farran & Cooper, 1986). Risk, according to Ramey and MacPhee (1986) is a property of the environment within which the human organism develops. «Risk» in such a conception comprises stresses and strains the individual has to adapt to or to cope with. The active part of the individual is emphasized as well as a transactional, dynamic nature of the developmental process. This conception comes closest to the old metaphor of the traveller mentioned above. «Risk» as used with adults and their behaviour usually implies some more or less rational decision on behalf of the subject to gamble, e.g. to succumb to some immediate pleasures or ventures such as drugs, fast driving etc., and to take into account some possible but not inevitably fatal end. This kind of risk-behaviour can certainly be discarded with respect to infants. Still, Ulich (1988) when referring to Murphy and Moriarty (1976), contends that development implies from the perspective of the developing individual, some striving for novelty as expressed in motivational concepts such as curiosity and exploratory behaviour. We will see later to what degree this aspect of the risk concept may help to understand some of the developmental problems of infants at-risk. The later two, environment and behaviour-related conceptions of risk, seem to imply some cognitive ability, namely a more or less sensitive «risk perception», involving a perceived lack of control (Slovic, 1987). In a functional analysis (Ramey & MacPhee, 1986), the infant's parents are assumed to act in loco infantis, that is, as those responsible for perceiving and evaluating the risk for the child. In a transactional or a cognitive model of development, however, the cognitive abilities even of the very young child are considered and related to some kind of risk perception as a theoretical possibility. «The concept of risk is a difficult one and becomes more difficult as our understanding of developmental processes deepens», Farran & Cooper (1986, p, 187) characterised the scientific state of art. Psychological concepts, such as vulnerability and resilience, stress and coping, have been introduced into the discussion on risk development to account for these processes. In the next sections, I will discuss these concepts with respect to infants and recent models of infant development. Protective factors, vulnerability and resilience Research on infants at-risk has primarily evolved from epidemiological and longitudinal research. Several studies have shown that some children, contrary to a prognosis based on major perinatal problems or adverse social and economical life conditions, develop normally into adolescence and adulthood (Elder, 1974; Garmezy, 1981, 1983; Hetzer, 1937; Rutter, 1979; Werner & Smith, 1982). Closer inspection of the biographies of these «normal» developers revealed that these persons lived in environments or had personal characteristics that ameliorated or compensated for the negative effects of their early risk condition. In most research on at-risk children, either single-cause or cumulative risk models have been used. The latter extend the category of risk/no risk into a dimension of severity of adverse influences that is supposed to correlate with an increase in the probability or the severity of unfavourable outcomes. Rarely have initial predispositions or insults been differentiated from the kind of developmental context in which the child develops. Kopp (1983) points to this context and calls it favourable or unfavourable «external mediators of risk influences» (p. 1162). These external mediators consist of context characteristics at the macromeso-, and micro-level according to Bronfenbrenner & Crouter's (1983) ecological system. Such mediators include (a) the belief systems, cultural values and cultural repertoire in the specific subculture (Liden, Nichter & Murphy, 1982; Bronfenbrenner, 1988), (b) the quality of the social support system, (c) the availability and quality of educational programs (Kopp, RISK AND PROTECTIVE FACTORS 165 1983), (d) the economical situation of the family, (e) family coherence (Werner & Smith, 1982; Hetzer, 1938), (0 the number of persons available for social and emotional resources (Werner, 1985), (g) the quality of caregiving (Kopp, 1983), and (h) the emotional and social relationship to significant others (Liden, Nichter & Murphy, 1982; Parker & Asher, 1987), especially to the mother (Werner, 198.5). Favourable life conditions or context variables that reduce the predicted later occurrence of psychopathology or problem behaviour (Silbereisen & Kastner, 1987; Ulich, 1988), have been described by Rutter (1979) as «protection» variables. If two different kinds of contexts or even two kinds of interacting and counteracting environmental aspects are compared in studies of children with different degrees of perinatal risk, the research model is similar to what Bronfenbrenner (1988) would call a «person-context model». Research, however, in which protective factors are separately identified from adverse contexts and from initial risk factors, as was done post-hoc in the longitudinal study by Werner (1985), is still extremely rare. Protective or adverse developmental contexts are only effective in relation to subjectlinked variables (Bronfenbrenner, 1988; Brandtsatdter, 1982a). A continuum of adverse events may produce in the child a specific or generalized «vulnerability», a proneness or preparedness to disintegrate in the face of specific or general stress. Vulnerability can be described as reduced flexibility or adaprational potential (Prechtl, 1987). Also, the child in its specific characteristics may attract reactions from its environment that acerbate its adverse situation as, for example, in the case of a physically deformed, a sick, or an extremely irritable infant that may provoke rejection or even neglect (Sameroff & Chandler, 1975). These characteristics may be dependent on genetic endowment, on the intactness of the central nervous system, and on physical appearance and health. In addition, vulnerability may be defined by the individual's abilities and skills under development or by emotional and motivational reaction styles that become stabilised in interaction with significant others. The positive counterpart to vulnerability is «resilience» (Werner & Smith, 1982). Resilience is characterized as the ability to draw on personal or social resources, the ability to detect contingencies and predictability in complex situations, and the ability to react flexibly. Resilience is enhanced, for example, by self-efficacy beliefs, a positive self-concept and selfesteem. Resilience seems to be very similar to what Murphy and Moriarty (1976) termed Coping II, the competence of the child to maintain or to regain internal equilibrium or integration. Again, specific characteristics of the child, such as a particular age, gender, physical appearance (Elder, Caspi & van Nguyen, 1986), or temperament may provoke positive reactions that not only help the child build up its resilience but also promote its development by being «developmentally instigative» (Bronfenbrenner 1988). Vulnerability, and its positive counterpart, resilience, are usually conceived as relatively stable personal characteristics of the child (Parker & Asher, 1987; Compas, 1987; Werner, 1986; Liden, Nichter & Murphy, 1987; Garmezy, Masten & Tellegen, 1984). Models of development that account for differential characteristics of the child and for bidirectional influences between child and social environment are characterized by Bronfenbrenner (1988) as «process-person-context models». However, they are still more a desideratum than a fact in developmental research. The age variable in risk models Since cognitive and social cognitive abilities are major ingredients of resiliency, we might speculate whether the infant is more vulnerable to adverse circumstances than the older child. As it is usually defined, resilience is only a relatively late result of positive personality development. In this case, better than expected developmental outcomes in early childhood of previously at-risk infants could only be attributed to protective factors early in life, for example, to favourable characteristics of the social setting in which the infant developed. There again it is not improbable that the infancy period can be generally considered as less susceptible 166 H.RAUH to sub-optimal social influences. Such speculation leads to combining our already complex risk model with recent models for explaining stability and change of characteristics from infancy to later periods of life. Longitudinal studies with normal children have repeatedly shown that prediction of individual mental differences at school age or later is little better than chance when test scores of mental or motor development in infancy are used as predictors (Brim & Kagan, 1980). In populations of «at-risk» infants, prediction of developmental or behavioural problems from the perinatal situation reduces with age, particularly beyond the third year, even when extremely life-threatening circumstances are considered. In contrast, the quality of the social setting in which the child lives is an increasingly important predictor of psychological functioning with age (Rauh, 1984; Beek & van Geerdink, 1989this issue; Kalmar 1988). Many explanations can be advanced to account for these findings, none of which has yet been tested fully: (1) Biological insults or risks disturb neurological development only for some limited time. (2) Enduring effects are caused mainly by chronic or long-lasting adverse biological or social conditions, rather than by short, even though dramatic events. (3) Insults to the central nervous and to the sensory system have their main effects in infancy, when the sensorymotor system represents and expresses the level of development, and, at later life periods, only in highly restrictive achievement situations when attentional processes are specifically taxed. (4)With age and development, behaviour becomes more complex and less constrained by specific biological handicaps; the possibilities for compensatory behaviours increase. (5) Social influences affect mainly cognitive and social behaviour, and therefore become effective only after the infancy period. (6)With age, parental expectations and developmental tasks for the child emphasise less specific physical skills and more mental and social abilities. (7) Over time, parents of a «difficult» child might «burn out» and loose their ability to react flexibly to their child, especially under socially strained life conditions. Some models of development emphasize the importance of early influences, induding social influences, for channelling the life course, for instance psychoanalysis, bonding and attachment theory, models of sensitive or critical periods, and the wedge hypothesis by Clarke and Clarke (1976, 1986; Keller, 1989 in press). Other models assume that biologically based self-healing or self-righting processes redirect the infant's course of development even after relatively severe deviations (Bronfenbrenner, 1988; Kopp & McCall, 1986; Rauh & Rudinger, 1987; Rauh & Berry, 1989 in press). The «scoop model» by Kopp and McCall (1984) is of special interest in this respect. Analyzing longitudinal data of mental development in normal healthy and at-risk children, they found that the predictive power of infant tests was negligible for the normal children. In contrast, for the at-risk groups (preterm children, children with Down's syndrome), the predictive power of infant tests was significant even at an earlier age. This was true, for example, for the children with Down's syndrome at 12 months when they reached on average a developmental level of about six to seven mental months. Similar results can be found in the Zurich longitudinal study (Largo et aI., 1988). It seems that prediction from mental tests cannot be attributed solely to similarities in test item content. Kopp and McCall assume that normal children during their first two years of life are well buffered or «biologically protected» against adverse experiences, such as the relative inexperience and unskilfulness of their parents. This self-righting process appears to be reduced in sick or biologically-insulted children, leaving them more vulnerable to adverse influences. Beyond the infancy period, during which the parents have a chance to learn to become «good» parents while their children are still relatively insensitive to their mistakes, the social setting, social standards and expectations increase their channelling influences on the developmental course of the child. In systemic conceptions of development (Grossmann, in press; Gordon & Jens, 1987; Bell, 1986; Kagan, Kearsley & Zelazo, 1980; Keller, 1989 in press), development progresses by building and reorganizing systems or patterns of behaviour. At least two major shifts in RISK AND PROTECTIVE FACTORS 167 infancy are acknowledged by most researchers of infant development: one around three months of life, and the other around eight months (Prechtl, 1987; Rauh, 1987; Kagan, Kearsley & Zelazo, 1980). The neonatal behavioural repertoire gives way to more flexible and complex behaviours at about three months. Around eight months, active memory retrieval becomes possible for the child, a prerequisite for developing symbolic thinking (Kagan, Kearsley & Zelazo, 1980). Conceptions of developmental models differ in emphasizing either the importance of the period of reorganization and reconstruction (McCall, Eichorn & Hogarty, 1977) or the time at which the final accomplishment of the developmental step or task is achieved so that the behavioural system is at its optimal level (Blank, 1964; Keller, 1989 in press; Beek & van Geerdink, 1989, this issue). Most researchers with these theoretical inclinations tend to use periods of major changes or periods of accomplishments as well as measures for the effects of previous risk influences and as optimal predictors of later development. Risk effects, however, may also disappear as a result of the process of restructuring and reorganization itself. Thus a child can move in and out of «risk» during its course of development (Gordon & Jens, 1987). Stress and coping The concept of «stress» has been adopted from clinical and personality research to refer to a specific and potentially adversive event in an infant's life. Stress has been defined in quite different ways in the infancy literature: namely, as a more or less adversive environmental challenge (Grossmann, 1989 in press), as a positive or negative experience (Liden, Nichter & Murphy, 1987), as an unpleasant mental and emotional state resulting from strain, and as a force requiring individual adaptation and the reestablishment of homeostasis (Ramey & MacPhee, 1986). Based on Lazarus and Folkman (1984), Compas (1987) distinguished between task-induced stress and emotional stress. Task-induced stress comes from a sequence of experiences of failure leading to what appears similar to the concept of vulnerability, although perhaps in a transient form. Emotional stress is characterized by tasks that are either a challenge, or rather a threat to the individual. Such tasks are characterized by novelty, unpredictability and resistance to efforts of control by the subject. Stresses can result from external forces, for example, meeting a strange person, experiencing a strange situation, being attacked by a peer, entering into day-care. They can also result from internal forces, such as illnesses and new but not as yet mastered physical or mental competences. These situations test the adaptational system of the child to its limits. They may lead to reorganization of the system, or, in the worst case, to a transient or prolonged break-down of the system, as, for instance, in a temper tantrum. Mild stresses are everyday experiences in an infant's life. As cumulative stress experiences they can contribute to the formation of generalized traits of vulnerability and resilience. As the complement to stress, coping characterizes the kind of problem-solving evoked by a challenging or threatening situation. Compas (1987) distinguished problem- and emotionfocused styles of coping, even in children. .Active coping implies cognitive and behavioural effort to manage the internal or external demands that are appraised as taxing or exceeding the resources of the person. If there is no other way to master the situation, regulation of emotion and minimizing or avoiding the stressor may be most effective and self-protective. Although coping is situation-specific, with age the child may develop a cross-situational strategy to deal with difficult situations (coping style). Children can be characterized as being more or less apprehensive for experiencing a situation as stressing. A vulnerable child tends to accumulate the negative stress effects. A resilient child, on the other hand, appears to be either relatively immune to adverse stress effects or able to interpret even an adverse situation as a challenge. According to Ramey and MacPhee (1986), the incidence of later problem behaviour is a function of organic factors and stress (including adverse life situations) relative to available 168 H. RAUH coping skills and self esteem (or resilience) as well as social and material resources or protective factors. Successful coping, according to the same authors, implies (1) recognizing a situation as the source of the problem, (2) knowing which strategies or resources to call upon in order to modify the situation, and (3) having the confidence (or internal locus of control) to maintain a sense of self-worth while attempting to gain control. These characteristics, however, are highly problematic considering the limited mental capabilities of an infant. To which degree are they applicable in the first two years of life? Can already the infant be conceived of as the agent of its own development? Some of the positive characteristics of coping with a challenging situation do appear to be similar to typical behaviours of the positivelydevelopinginfant. Being attracted by novelty rather than the familiar, which marks the shift around three months, or extensive exploratory behaviour, as is typical toward the end of the first year, could be interpreted as an active search for challenging situations. Hand (1981) as well as Losche (1989) showed that infants and young children, in contrast to older children, tend to behave and to act at the upper limit of their competencies. Records from natural observations of their behaviour therefore correlate closely with tested levels of development. Older children tend to select the level of their behaviour more economically according to the specific demands of the situation. On the other hand, infants later diagnosed as autistic more often took recourse to lower-level behaviour than their «normal» counterparts. Similarly, Wishart (1986) described the typical achievement-behaviour of young children with Down's syndrome as «cognitive avoidance behaviour». Even when minimally challenged beyond their established competency, they tended to «give up» or to retreat to lower-level behaviour. Spontaneous exploratory behaviour or even adventurous approach to novel situations was very rare with these children. Our own observations on Down's syndrome infants corroborate this impression of an extreme lack of «self-confidence» from early age. It is extremely difficuft to explain theoretically what «self-confidence», «cognitive avoidance», or all the facets of coping are in the youngest infants, without speculative recourse to cognitive abilities that are at-best nascent in infants. Recently, Bronfenbrenner (1988) has also attributed to the infant some of those active characteristics that not only keep the child oriented to new and developmentally instigating challenges of his environment but also signal to the responsive adults its needs and interests, that in turn influence the infant's developmental context. He then qualified his attributions as follows: «To the extent that such an active and responsive personal orientation to the environment exists only in limited degree, the developmental progression may proceed at a slower rate, and along fewer developmental pathways»(p.62). Educational implications «Risk children» are those that deserve society's special and foremost attention and preventive help. At least this is the connotation of the term «risk». As the concept of risk was extended beyond life-threatening dangers to refer to risk of major social and behavioural problems, different kinds of deficit models resulted. The deficits are defined at different conceptual levels, the situational, the personal trait, the social and the societal system levels. The complex transactional and systemic models presented in the previous sections, however, challenge the conception of risk models as merely deficit models. The review of various meanings of risk and their implications showed that we are still a long way from understanding fully the processes of development and the impact of stressful life events, especially in infancy. The different conceptualisations of the infant as a recipient and victim of risks or adverse circumstances and, at least to some degree, an agent of its own development, have implications for prevention and intervention through improvement of care and education. Therefore, some suggestions as well as caveats seem to be in order. RISK AND PROTECTIVE FACTORS 169 It cannot be denied that the concept of «risk» has become fashionable and useful for attracting attention to particular groups of children and for attracting financial resources. However, as is true with most fashionable concepts, the risk concept is in danger of becoming overgeneralised and eventually impracticable. Children with definite handicaps, for instance, are not necessarily «risk children» with regard to their handicap or its immediate sequelae. Still they need care and education adapted to their needs. The kind of training, education or therapy offered to them and to their families as «risk prevention» could, however, in itself become a «risk factor»: Professional compensatory intervention with infants from poor social backgrounds may have iatrogenic influences on the family system and parental feelings of responsibility for their children's development. Gunn and Berry (1989, this issue) also point to such problems. Several screening procedures for detecting children at-risk have been suggested (Gordon & Jens, 1989; Keogh, Wilcoxen & Bernheimer, 1986; Liden, Nichter & Murphy, 1982; Ramey and MacPhee, 1986). It is difficult at this stage of our knowledge to pinpoint the optimal age and the optimal measures and methods for economically and effectively detecting children for preventive and compensatory intervention. Rarely have the whole family and the social setting been the focus of screening procedures. Nevertheless, an effective educational intervention could aim at the setting level and thereby help to provide living conditions and prerequisites for growth and development that are suited for children with a variety of characteristics. An important pair of concepts seems to be that of vulnerability and resilience, including coping. Improving coping skills and resilience in infants and their families may be the best prevention and compensation that can be offered. Again, we are just beginning to learn about these concepts with respect to infancy, and a lot needs to be done from the side of educational psychology to develop adequate provisions. If living conditions are fair and if the infant and his/her parents are healthy, «intuitive parenting» (Papousek & Papousek, 1989, this issue) offers the infant the necessary framework for psychologically healthy development. On the other hand, sensitive parents learn with and from their child. They adjust the task of parenting to their individual child and to their own temperament. Cultural and societal values enter already at this stage of dyadic interaction as is shown in the paper of Tourette, Robin & Josse (1989, this issue) on twinning. Educational intervention in this age period should be extremely careful not to streamline families and dyads into educational values which have not been thoroughly analyzed for their roots and culturallyvalued implications. Parents most in need of educational advice and help are those with deviant or sick infants. These infants need «optimal» parents from early on. These children offer little help to their parents to develop their parental skills. At the same time, these infants seem to be more vulnerable to even minor educational deficiencies in their parents (Kalmar 1987). All the accumulating knowledge on optimal parenting should be provided to their parents in an adequate form. Further, the living conditions of these parents should be stabilised or improved in such a way that they have the opportunity to learn and to carry out this difficult task. Mutual enjoyment of parents and children would be a major measure of success. The basis of such a suggestion is, that all infants, even the sick, the handicapped or the deviant ones, have a human right to develop optimally within the range of their possibilities. Education provides cultural and social values, contents, as well as skills for living and participating in a particular society. It may also help to provide a context for optimal development. Psychological measures of average levels of functioning, however, are not necessarily the optimal goals of development. Here there is still a rich field of discussion and cooperative research between the two disciplines of psychology and education. 170 H. RAUH References Appleton, T., Clifton, R., & Goldberg, S. (1975). The development of behavioral competence in infancy. In ED. Horowitz (Ed.), Review of child development research. Vol. 4 (pp. 101·]86). Chicago: University of Chicago Press. Beek, Y. van, & Geerdink, J. J. (1989 this issue). Intervention with preterms; is it educational enough? European Journal of Psychology of Education, 4, 251·265. Bell, R. Q. (1986). Age-specific manifestations in changing psychosocial risk. In D.C. Farran & J.D. MacKinney (Eds.), Risk in intellectual and psychosocial development (pp. 169·185). New York: Academic Press. Blank, M. (1964). A focal periods hypothesis in sensorimotor development. Child Development, 35, 817·829. Brandtstiidter, J. (l982a). Kern- und Leitbegriffe psychologischer Priivention. In J. Brandtstiidter & A. von Eye (Eds.), Pychologische Pravention. Grundlagen, Programme, Methoden (pp. 81· 115). Bern: Huber. Brandtstiidter, 1. (l982b). Methodologische Grundfragen psychologischer Pravention, In J. Brandtstadter & A. von Eye (Eds.), Psychologische Pravention. Grundlagen, Programme, Methoden (pp. 37·80). Bern: Huber. Brandtstadter, J. (1982c). Pravention als psychologische Aufgabe. In J. Brandtstadter & A. von Eye (Eds.), Psychologische Pravention. Grundlagen, Programme, Methoden (pp, 15·36). Bern: Huber. Brim, O. G. & Kagan, J. (1980). Constancy and change: A view of the issues. In O. G. Brim & 1. Kagan (Eds.), Constancy and change in human development (pp. 1·25). Cambridge, Mass.: Harvard University Press. Brockhaus (1980). Der grosse Brockhaus in 12 Bdnden (Vol. 9). Wiesbaden: EA. Brockhaus. Bronfenbrenner, U (1988). Ecological systems theory, (unpublished manuscript). Syracuse, N.Y.: Cornell University. Bronfenbrenner, U, & Crouter A. C. (1983). The evolution of environmental models in developmental research. In P. H. Mussen (Ed.), Handbook of child psychology, Vol.I: History, theory, and methods (w. Kessen, volume editor) (pp. 357-414). New York: Wiley. Bruner, 1. (1968). Processes of cognitive growth: Infancy. Worcester, Mass.: Clark University Press. Clarke, A. M., & Clarke, A. D. B. (1976). Early experience: Myth and evidence. London: Open Books. Clarke, A. M., & Clarke, A. D. B. (1986). Thirty years of child psychology: A selective review. Journal ofChild Psychology and Psychiatry, 27, 719-759. Compas, B. E. (1987). Coping with stress during childhood and adolescence. Psychological Bulletin. 101, 393-403. Duden (1980). Das grosse Worterbuch der deutschen Sprache in sechsBanden (Vol. 5). In Wissensdraftlicher. Rat der Dudenredaktion (Ed.), Mannheim: Bibliographisches Institut. Elder, G. H. (1974). Children of the great depression. Chicago: University of Chicago Press. Elder, G. H., Caspi, A., & van Nguyen, T. (1986). Resourceful and vulnerable children: Family influences in hard times. In R. K. Silbereisen, K. Eyferth, & G. Rudinger (Eds.), Development as action in context: Problem behavior and normal youth development (pp. 167-186). New York: Springer. Farran, D. C. (1986). Problems and perspectives for the concept of risk in psychosocial development: A summary. In D. C. Farran & 1. D. McKinney (Eds.), Risk in intellectual and psychosocial development (pp. 261·269). New York: Academic Press. Farran, D. C; & Cooper, D. H. (1986). Psychosocial risk: Which early experiences are important for whom? In D.C. Farran & J. D. McKinney (Eds.), Risk in intellectual and psychosocial development (pp, 187-226). New York: Academic Press. Garmezy, N. (1981). Children under stress, Perspectives on antecedents and correlates of vulnerability and resistance to psychopathology. In A. I. Rabin, J. Aronoff, A. M. Barclay, & R. A. Zucker, (Eds.), Further explorations in personality (pp. 196-269). New York: Wiley. Garmezy, N. (1983). Stressors of childhood. In N. Garmezy & M. Rutter, (Eds.), Stress, coping, and development in children (pp. 43-84). New York: Wiley. Garmezy, N. Masten, A. S., & Tellegen, A. (1984). The study of stress and competence in children: A building block for developmental psychopathology. Child Development, 55, 97-11!. Gordon, B. N. & Jens, K. G. (in press). A conceptual model for tracking high-risk infants and making early service decisions. Journal of Behavioral and Developmental Pediatrics. Grossmann, K. E. (in press). Follow-up of biological high and low risk groups: What can be learned from longitudinal and systemic research? In M. Rutter (Ed.), The power of longitudinal data:Studies of risk and protective factors for psychosocial disorder. Cambridge: Cambridge University Press. Gunn, P. & Berry, P. (1989 this issue). Education of infants with Down syndrome. European Journal of Psychology of Education, 4, 235-246. Hand, H. H. (1981). The relation between developmental level and spontaneous behavior: The importance of sampling context. New Directions of Child Development, 12, 55-67. Hetzer, H. (1937). Kindheit und Armut. Psychologische Methoden in Armutsforschung und Armutsbekdmpfung (2nd. ed.). Leipzig: Hirzel. Kagan, 1. (1980). Perspectives on continuity. In O.G. Brim & J. Kagan (Eds.), Constancy and change in human development (pp. 26-74). Cambridge, Mass.: Harvard University Press. RISK AND PROTECTIVE FAClDRS 171 Kagan, J., Kearsley, R. B., & Zelazo, P. R. (1980). Infancy. Its place in human development. Cambridge, Mass.: Harvard University Press. Kalmar, M. (1987). Biological and environmental determinants of mental development in «at-risk» children. (unpublished paper presented at the Italian-Hungarian Psychology Symposium, Cassino, Italy). Budapest: Institute of Psychology, Eotvas University, Keller, H. (in press). Kontinuitat und Diskontinuitat in der Entwicklung. In H. Keller (Ed.), Handbuch der fruhen Kindheit. Heidelberg: Springer. Keogh, B., Wilcoxen, A. G., & Bernheimer, L. (1986). Prevention services for risk children: Evidence for policy and practice. In D. C. Farran & J. D. McKinney (Eds.), Risk in intellectual and psychosocial development (pp. 287316). New York: Academic Press. Koluchova, J. (1976). The further development of twins after severe and prolonged deprivation: A second report. Journal of Child Psychology and Psychiatry. 17, 181-188. Kopp, C. B. (1983). Risk factors in development. In M.M.Haith & 1.1. Campos (Eds.), Infancy and developmental psychobiology. Handbook of child psychology. Vol./I (RH.Mussen, Ed.) (pp. 1081-1188). New York: Wiley. Kopp, C. B., & McCall, R. B. (1984). Predicting later mental performance for normal, at risk, and handicapped infants. In P. B. Baltes, & O. G. Brim jr. (Eds.), Life-span development and behavior. Vol. 4 (pp. 33-61). New York: Academic Press. Largo, R. H., Graf, S., Kundu, S., Hunziker, u., & Molinari, L. (in press). Predicting developmental outcome at school age from infant tests in normal, at risk and retarded children. Developmental Medecine and Child Neurology. Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal and coping. New York: Springer. Liden, C. B., Nichter, C. A., & Murphy, T. F. (1982). TRANSACT - A model for integrating clinical practice and longitudinal research. Seminars in Perinatology, 6(4), 340-352. Losche, G. (1989). Entwicklung autistischer Kinder in den ersten dreieinhalb Lebensjahren. (unpubl, dissertation manuscript). Berlin (W): Department of Education, Free University Berlin. McCall, R., Eichorn, D. H., & Hogarty, P. S. (1977). Transitions in early mental development. Monographs of the Society for Research in Child Development. Serial No. 42, 42, (3). Meyer-Probst B., & 'Ieichmann, H. (1984). Risiken in der Personlichkeltsentwicklung im Kindesalter. Rostocker Liingsschnittstudie. Leipzig: Thieme. Michaelis, R., & Haas, G. (1987). The concept of optimality in evaluating hazards of prenatal, perinatal and postnatal complications. In H. Rauh & H. C. Steinhausen (Eds.), Psychobiology and early development (pp. 145-154). Amsterdam: North- Holland. Murphy, L. B., & Moriarty, A. E. (1976). Vulnerability, coping, and growth. From infancy to adolescence. New Haven: Yale University Press. Mussen, P. H. (Ed.) (1970). Carmichael's manual of child psychology (Vol. I, //) (3rd. ed.). New York: Wiley. Mussen, P. H. (ed.) (1983). Handbook of child psychology. Vol. I- IV (4th. ed.), New York: Wiley. Oerter, R., & Montada, L. (Eds.) (1987). Entwicklungspsychologie. Ein Lehrbuch (2nd. ed.). Miinchen: Psychologie Verlags Union. Papousek, H., & Papousek, M. (1989, this issue). Intuitive parenting: Aspects related to Educational Psychology. European Journal of Psychology of Education, 4, 201-210. Parker, J. G., & Asher, S. R. (1987). Peer relations and later personal adjustment: Are low-accepted children at risk? Psychological Bulletin, 102, 357-389. Prechtl, H. F. R. (1980). The optimality concept. Early Human Development, 4, 201-205. Prechtl, H. F. R. (1987). Prenatal development and postnatal behavior. In H. Rauh & H. C. Steinhausen (Eds.), Psychobiology and early development (pp, 231-238). Amsterdam: North Holland. Ramey, C. T. & MacPhee, D. (1986). Developmental retardation: A systems theory perspective on risk and preventive intervention. In C. D. Farran & J. D. McKinney (Eds.), Risk in intellectual and psychosocial development (pp. 61-82). New York: Academic Press. Rauh, H. (1984). Friihgeborene Kinder. In H. C. Steinhausen (Ed.), Risikokinder. Ergebnisse der Kinderpsychologie und -psychiatrie (pp. 11-35). Stuttgart: Kohlhammer. Rauh, H. (1987). Verhaltensausstattung und erste Anpassungsleistungen des Sauglings: Anfange der geistigen Entwicklung. In C. Niemitz (Ed.), Erbe und Umwe/t (pp. 174-189). FrankfurtlM.: Suhrkamp. Rauh, H. (in press). Kontinuitat und Diskontinuitat in der Entwicklung - Ein Kommentar und eine Erganzung. In H. Keller (Ed.), Handbuch der friihen Kindehit. Heidelberg: Springer. Rauh, H., & Berry, P. (in press). Differentielle Entwickhmgsverlaufe bei Kleinkindern mit Down-Syndrom. In H. Teichmann, B. Meyer-Probst, & D. Roether (Eds.), Risikobewaltigung in der lebenslangen psychischen Entwicklung. Leipzig: Thieme. Rauh, H., & Rudinger, G. (\987).Early development of Down syndrome children as assessed by the Bayley Scales. In H. Rauh (Ed.), ( Arbeitsberichte des Instituts fiir Psychologie der FU Berlin). Berlin (W): Freie Universitat Berlin. Rutter, M. (1979). Protective factors in children's responses to stress and disadvantage. In M. W. Kent & J. E. Rolf (Eds.), Social competence in children. Primary prevention in psychopathology (Vol. Ill, pp. 49-74). Hanover, N.H.: University Press of New England. 172 H.RAUH Sameroff, A. J., & Chandler, M. J. (1975). Reproductive risk and the continuum of caretaking casuality. In F. D. Horowitz (Ed.), Review of child development research. Vol. 4. (pp. 187-244). Chicago: University of Chicago Press. Silbereisen, R. K., & Kastner, P. (1987). Jugend und Problemverhalten: Entwicklungspsvchologische Perspektiven. In R. Oerter & L. Montada (Eds.), Entwicklungspsychologie, Ein Lehrbuch (pp. 882-919). Miinchen: Psychologic Verlags Union. Slovic, P. (1987). Perception of risk. Science, 236, 280-285. Stone, J. L., Murphy, L. B., & Smith, H. T. (Eds.) (1973). The competent infant. New York: Basic Books. Tourette, C; Robin, M., & Josse, D. (1989 this issue). Differentiation of twins and maternal educational practices in infancy. European Journal of Psychology of Education, 4, 269-283. Ulich, M. (1988). Risiko- und Schutzfaktoren in der Entwicklung von Kindem und Jugendlichen. Zeitschrift fur Entwicklungspsychologie und Piidagogische Psychologie, 20, 146-166. Werner, E. E. (1985). Stress and protective factors in children's lives. In A. R. Nicol (Ed.), Longitudinal studies in child psychology and psychiatry (pp. 335-355). New York: Wiley. Werner, E. E. (1986). A longitudinal study of perinatal risk. In D.C. Farran & 1. D. McKinney (Eds.), Risk in intellectual and psychosocial development (pp. 3-28). New York: Academic Press. Werner, E. E. & Smith, R. S. (1982). Vulnerable but not invincible. A longitudinal study of resilient children and youth. New York: McGraw-Hili. Wilson, R. & Crouch, E. A. C. (1987). Risk assessment and comparisons: An introduction. Science, 236, 267-270. Wishart, J. G. (1986). Cognitive avoidance in infants and young children with Down's syndrome. (Paper presented at the International Down's Syndrome Congress, Brighton, England 1986). Edinburgh: Department of Psychology. Facteurs de risque et de securite chez Ie nourrisson Les recherches recentes en psychologie du nouveau-ne et du nourrisson ont mis en evidence des competences precoces non prevues jusqu'alors par /es scientifiques, et difficiles iJ concilier avec les modeles traditionnels du developpement de l'enfant. Simultanement, de nombreuses donnees longitudinales remettent en question /es anciennes conceptions sur l'effet persistant des experiences precoces sur Ie developpement ulterieur: Le concept de risque, emprunte iJ la pediatrie et iJ la psychiatrie, a ete primitivement introduit en psychologie du developpement pour identifier des groupes d'enfants pour lesquels les risques de mauvaise adaptation comportementale ou developpementale seraient accrus. Une revue de ce concept montre que des sens tres differents lui ont ete donne. Des concepts recents comme la vuinerabilite vs la resistance, comme caracteristiquesstables de la personnalite, et la maniere de faire face Ii une situation de stress en tant que variable ou processus situationnels sont discutes. Leur combinaison avec Ie concept de risque dans des modeles plus sophistiques du developpement precoce, tels que les modeles transactionnels ou la theorie des systemes fournissent des cadres plus detailles pour comprendre les enfants iJ risque. Les implicationspour la prevention, l'intervention et l'education precoce sont discutes pour certains groupes particuliers consideres comme iJ risques. Key words: Risk factors, Protective factors, Vulnerability, Resilience infancy, Stress, Coping. Revision received: May 1989 HeUgard Raub. Institute of Psychology, Department of Psychology, Free University Berlin, Habelschwerdter Allee 45, D-l000 Berlin 33. Current theme of research: Mental growth in infants and young children with Down syndrome. Coping with novelty in infancy: entrance into day-care. RISK AND PROTECTIVE FACTORS 173 Most relevant publications in the field of Educational Psychology: Rauh, H. (1973). Grundlagen und Probleme der Elementarerziehung. In H. -P, Bennwitz & F. Weinert (Ed.), CIEL - Ein Fordetungsprogram zur Elementarerziehung und seine wissenschaftlichen Voraussetzungen (pp, 11-73). Gottingen: Vandenhoeck & Ruprecht. Rauh, H. (1979). Vorschulerziehung. In H. H. Groothoff (Ed.), Erziehungswissenschaftliches Handbuch, Bd. 5, 1 (pp, 10-70). Konigstein: Athenaum. Rauh, H. (1984). Friihgeborene Kinder. In H. C. Steinhausen (Ed.), Risikokinder. Ergebnisse der Kinderpsychologie und Kinderpsychiatrie (pp. 11-35). Stuttgart: Kohlhammer. Rauh, H. (1985). Diagnose und Beratung in der friihkindlichen Entwicklung. In 1. Brandtstadter & H. Graser (Ed.), Entwicklungsberatung unter dem Aspekt der Lebensspanne (pp. 44-64). Gottingen: Hogrefe. Rauh, H., & Steinhausen, H. C. (Eds.) (1987). Psychobiology and early development. Amsterdam: North-Holland.