Here we will consider whether temperamental differences are related to other aspects of children’s development. It must be emphasized that temperament is concerned with individual differences and therefore the impact on development centres on associations between temperament and variations in children’s cognitive and social development. There are several ways in which this can occur and these will be considered in turn.
Direct effect of temperament on development
A child with a short attention span and who is very impulsive is likely to experience difficulties in learning situations either at home or at pre-school groups (Tizard and Hughes, 1984). This example shows that temperamental differences may have a pervasive effect on children’s cognitive and social development through their impact on behavioural control and responsivity. In older children Keogh (1982) has identified a three factor model of temperament that is related to behaviour in school and which has implications for learning. The factors are Task Orientation, Personal-Social Flexibility and Reactivity.
Clearly factors such as task orientation will have a direct impact on the child’s ability to gain from learning experiences. Other temperamental influences will have more indirect effects on academic attainment. For example, reactivity is more likely to influence pupil–teacher and pupil–pupil interaction and thereby the social context within which learning takes place.
Direct effect of child temperament on parents
One of the central concepts in current thinking about child development is that of the child influencing its own development, i.e. not just being a passive receiver of externally determined experiences. Bell (1968) and Sameroff and Chandler (1975) are widely recognized as bringing this transactional model to the fore. Under this model the child plays a significant role in producing its own experiences both directly by its own selection of activities but, more importantly for the young child, by the influence its behaviour has upon caretakers (Sameroff and Fiese, 1990).
Indirect effect via ‘goodness of fit’
There has been a strand of thinking linked with the study of temperament that has emphasized that the significance of individual differences in temperament has to be considered in relation to specific environments. A child who is very low on adaptability and very high on rhythmicity using will have a more aversive experience if cared for by parents who are very erratic in their pattern of child care. The same child will be well suited to parents who are more regular in their routines of eating and sleeping. This suggests that the impact of temperament on development has to be analysed as an interaction between the child’s characteristics and features of the environment including parenting.
There have been several temperament theorists who have taken this position.
One of the most extensive research studies with this goodness of fit orientation is that of Lerner and colleagues:
The ‘goodness of fit’ concept emphasizes the need to consider both the characteristics of individuality of the person and the demands of the social environment, as indexed for instance by expectations or attitudes of key significant others with whom the person interacts (e.g. parents, peers or teachers). If a person’s characteristics of individuality match, or fit, the demands of a particular social context then positive interactions and adjustment are expected. In contrast, negative adjustment is expected to occur when there is a poor fit between the demands of a particular social context and the person’s characteristics of individuality. (Lerner, et al., 1989, p. 510)
As an illustration of this notion of the goodness of fit between the child’s temperament and parental behaviour Lerner et al. (1989) discussed some of the evidence concerning temperament and maternal employment outside the home. Of course a wide variety of social and economic pressures will be influencing the decision to work outside the home. However, in addition they suggest that there could be two plausible routes whereby difficult temperament could influence mothers’ decisions on whether to work outside the home. The first could be that mothers find the problems of rearing the child with difficult temperament too aversive and therefore opt to go out to work to avoid the hassles of daily child care.
The second route could be that the difficult child is so unpredictable in its eating and sleeping habits and protests intensely when left with unfamiliar people that the mother feels constrained not to go out to work because the child cannot fit in with the externally required constraints of the mother attending the work place at fixed times for fixed periods.
The goodness of fit approach suggests that which of these processes operates will depend on the fit between the child’s temperament and the mother’s tolerance. It will not be possible to predict the consequences of difficult temperament on the mother’s decision to return to work with knowledge of her attitudes towards child rearing and towards time keeping at work.
Lerner and Galambos (1985) found that mothers of children with difficult temperament tended to have more restricted work histories than other children.
One problem with this finding is that mothers’ reports on their infants’ ‘difficulty’ may be biased by factors that also affect work performance, such as depression. Hyde et al. (2004) examined this possibility in a study which found that the consensus infant temperament judgements of fathers and mothers were still a good predictor of mothers’ work outcomes. This study also found evidence that a mediating factor between infant temperament and maternal work outcome is maternal mood: difficult infants are likely to make mothers more depressed and diminish their sense of competence, thus affecting their work performance. The Lerner and Galambos (1985) study also found that it seemed to be harder for parents to make satisfactory day-care arrangements for difficult infants.
Indirect effect via susceptibility to psychosocial adversity
Temperament may also be related to differences in vulnerability to stress. Not all children are adversely affected by the experience of specific stresses, such as admission to hospital. Pre-school children repeatedly hospitalized are at risk for later educational and behavioural difficulties but only if they come from socially disadvantaged backgrounds (Quinton and Rutter, 1976).
It has proved more difficult to establish whether temperament does influence susceptibility to adverse experiences. Dunn and Kendrick (1982) have shown that an older child’s response to the arrival of a new sibling is systematically related to their temperament as measured whilst their mother was pregnant.
Most children respond to this event with some upsurge of behavioural disturbance, such as an increase in demands for parental attention or in crying. Which behavioural response is shown is related to prior temperament. Unfortunately their data do not suggest any clear pattern of any one aspect of temperament being more significant than any other. However, there were indications that increases in fears, worries and ‘ritual’ behaviours were associated with a high degree of temperamental Intensity and Negative Mood measured before the arrival of the second child.
Indirect effect on range of experiences
An important aspect of the transactional model of development is that as children become older they increasingly come to influence the range of environments they encounter and the experiences these create. During infancy, children with different temperament styles evoke different responses from the people they encounter, for example, active, smiling infants are more likely to be smiled at and played with than passive unresponsive infants. As children become more mobile and more independent they are able to select for themselves between alternative experiences, for example, a shy, behaviourally-inhibited child may avoid social encounters. This may accentuate temperamental characteristics: the avoidance of meeting other people prevents the child from becoming socially skilled and therefore more reluctant to engage in social behaviour in the future. This may have a wider impact on their development. For example, Rutter (1982) has demonstrated the way impulsive, active children are more likely to experience accidents, presumably as a result of their selecting more risky environments to play in.
These alternative mechanisms for the impact of temperament on the environments the child experiences can be classified into three types of gene-environment correlation. Scarr and McCartney (1983) have suggested that children’s genetic make-up comes to influence the environments they experience through three routes. These can be illustrated for temperament. One is passive gene-environment correlations which are produced when the child is being cared for by parents who share similar temperaments to the child. A child with a high intensity of reaction is more likely than other children to be cared for by a parent who has a similarly high intensity of reaction. Such parent–child pairs are likely to be creating experiences for the child which will be eliciting much aversive stimulation for the child. Evocative gene environment correlations are created when the child’s behaviour evokes specific types of responses from carers. This was illustrated in the earlier example of sociable children evoking more social stimulation from carers. The third type is active gene-environment correlation which arises from the child actively seeking environments that suit its behavioural predispositions. Children with a low threshold of responsiveness are likely to seek less extreme and more predictable environments.
An important feature of the Scarr and McCartney theory is that they propose that as the child becomes older the mix of these correlations will change. Initially the passive and evocative correlations will dominate. The evocative effects will remain fairly constant. The significance of passive effects decline in importance as the child encounters a wider range of people than just primarily the parents. Clearly active gene-environment effects are likely to become dominant as the child has greater and greater freedom to select its own activities.
Attachment and temperament
An important aspect of children’s early development is the quality of their attachment to their caregiver. A widely-used, standardised way of assessing this is a laboratory procedure called the ‘Strange Situation Test’ (SST; Ainsworth et al., 1978), consisting of a series of separations and reunions of child, caregiver and a stranger. Depending on how children behave during these episodes, their attachment is classified as either ‘secure’ or ‘insecure’. Insecure classifications are further subdivided into ‘avoidant’ or ‘ambivalent’ categories.
These different attachment styles are seen as important because they are associated with variations in children’s subsequent development; secure attachment is generally associated with more positive outcomes. Since the formation of attachment is bound up with how an infant behaves towards the caregiver during the first year of life it would seem likely that infant temperament is a significant element. It is surprising, then, that although some research has found that infant irritability and negative emotionality are linked with the avoidant type of insecure attachment, numerous studies have found no evidence that infant temperamental differences are associated directly with secure versus insecure attachment classifications in typical development (Goldsmith and Alansky, 1987).
One feature of caregiver behaviour during child’s first year that has been widely found to influence attachment quality is ‘sensitivity’ (De Wolff and Van IJzendoorn, 1997), namely, the extent to which the caregiver is attentive to the infant’s state, behaviour and communication, and responds appropriately. It might be expected that caregiver personality differences would thus be found to be associated with infant attachment security, but here again few direct effects have been found (Egeland and Farber, 1984).
What has been found, however, is that the combination of child and caregiver individual characteristics does predict attachment security (Belsky and Isabella, 1988; Notaro and Volling, 1999) lending support to a transactional model of the process.
Mangelsdorf et al. (2000) studied 102 mother-infant dyads in Michigan, U.S.A., to examine the contributions of maternal and infant characteristics to infant attachment. When the infants were 8 months old, their temperaments were assessed in a laboratory-based set of tasks, their mothers completed personality questionnaires (MPQ) for themselves and IBQ questionnaires on their infants, and then completed a brief teaching task with their infants. At twelve months of age, each infant’s attachment security was assessed with the SST.
Neither mothers’ nor infants’ characteristics, taken alone, were good predictors of infants’ attachment classification. However, when the joint effects of both mother and infant factors were examined, it was found that infants were classed as securely attached if they showed more positive emotions and fewer fearful reactions in the temperament assessments, but only if their mothers also showed more positive emotionality. These secure infants were also rated low in the IBQ on activity level and the amount of distress they showed to novelty but, again, only if their mothers also rated high on Constraint (self-control, conventionality) in the MPQ.
The researchers comment on these findings that:
The results of this investigation suggest that any individual characteristic of either child or mother may be less important than the relationship context within which that characteristic occurs. Mangelsdorf et al. (2000) p. 188.
- Temperament can directly influence other aspects of development, for example, attentional variation has an impact on cognitive development.
- Temperamental variation influences the parent’s response to the child.
- The goodness of fit between a child’s temperament and parental style can have an impact on the child’s attachment and long-term social adjustment.
- Temperament can influence a child’s vulnerability to the adverse effects of life events.
- Temperament can have a marked effect on the type and range of experiences to which the child is exposed.
Ainsworth, M. D. S., Blehar, M. C., Waters, E. and Wall, S. (1978) Patterns of Attachment, Hillsdale, N.J., Lawrence Erlbaum.
Bell, R. Q. (1968) ‘A reinterpretation of the direction of effects in studies of socialisation’, Psychology Review, vol. 75, pp. 81–95.
Belsky, J. and Isabella, R. (1988) ‘Maternal, infant and social-contextual determinants of attachment security’ in Belsky, J. and Nezworski, T. (eds) Clinical Implications of Attachment, pp. 253–99, New York, Lawrence Erlbaum.
De Wolff, M. S. and Van IJzendoorn, M. H. (1997) ‘Sensitivity and attachment: a meta-analysis on parental antecedents of infant attachment’, Child Development, vol. 68, pp. 571–591.
Dunn, J. and Kendrick, C. (1982) ‘Temperamental differences, family relationships and young children’s response to change within the family’ in Porter, R. and Collins, G. (eds) Temperamental Differences in Infants and Young Children, pp. 1–19, CIBA Foundation Symposium No. 89, London, Pitman.
Egeland, B., & Farber, E. A. (1984) ‘Infant-mother attachment: factors related to its development and changes over time’, Child Development, vol. 55, pp. 753–771.
Goldsmith, H. H. and Alansky, J. A. (1987) ‘Maternal and infant temperamental predictors of attachment: a meta-analytic review’, Journal of Consulting and Clinical Psychology, vol. 55, pp. 805–16.
Keogh, B. K. (1982) ‘Children’s temperament and teachers’ decisions’ in Porter, R. and Collins, G. (eds) Temperamental Differences in Infants and Young Children, pp. 269–85, CIBA Foundation Symposium No. 89, London, Pitman.
Lerner, J. V. and Galambos, N. L. (1985) ‘Maternal role satisfaction, mother–infant interaction and child temperament’, Developmental Psychology, vol. 21, pp. 1157–64.
Lerner, J. V., Nitz, K., Talwar, R. and Lerner, R. M. (1989) ‘On the functional significance of temperamental individuality: a developmental contextural view of the concept of goodness of fit’ in Kohnstamm, G. A., Bates, J. E. and Rothbart, M. K. (eds) Temperament in Childhood, pp. 509–22, Chichester, John Wiley.
Mangelsdorf, S.C., McHale, J.L., Diener, M., Goldstein, L.H. and Lehn, L. (2000) ‘Infant attachment; contributions of infant temperament and maternal characteristics’, Infant Behaviour and Development, vol. 23, pp. 175–196.
Notaro, P. C., & Volling, B. L. (1999) ‘Parental responsiveness and infant-parent attachment: A replication study with fathers and mothers’, Infant Behavior and Development, vol. 22, pp. 345-352.
Quinton, D. and Rutter, M. (1976) ‘Early hospital admissions and later disturbances of behaviour: an attempted replication of Douglas’s findings’, Developmental Medicine and Child Neurology, vol. 18, pp. 447–59.
Rutter, M. (1982) ‘Temperament: concepts, issues and problems’ in Porter, R. and Collins, G. (eds) Temperamental Differences in Infants and Young Children, pp. 1–19, CIBA Foundation Symposium No. 89, London, Pitman.
Sameroff, A. J. and Chandler, M. J. (1975) ‘Reproductive risk and the continuum of caretaking casualty’ in Harrowitz, F. D., Scarr-Salapatek, S. and Siegel, G. (eds) Review of Child Development Research, pp. 187–24, Vol. 4, Chicago, University of Chicago Press.
Sameroff, A. and Fiese, B.H. (1990) ‘Transactional regulation and early intervention’ in S. J. Meisels and J. P. Shonkoff (eds) Handbook of Early Childhood Intervention, pp. 119–149, New York, Cambridge University Press.
Scarr, S. and McCartney, K. (1983) ‘How people make their own environments: a theory of genotype-environment effects’, Child Development, vol. 54, pp. 424–35.
Tizard, B. and Hughes, M. (1984) Young Children Learning: talking and thinketing at home and at school, London, Fontana.
This extract from course ED209 is © Open University 2005