The New York Longitudinal Study (NYLS) was conducted in the 1950s on infants aged 2-3 months. This study followed these children at regular intervals from infancy to young adulthood (Lamb, Bornstein, Teti, 2002). Based on this study, Thomas, Chess, Birch, Hertzig, and Korn (1963) categorized infants into three groups: easy, slow-to-warm-up, and difficult (Slentz & Krogh, 2001). They further described the characteristics of children in each category as follows: easy babies are regular in their daily schedules for eating, sleeping, and elimination; tend to interact in novel situations; are adaptable to change, mild in disposition, and positive in the mood. Slow-to-warm-up babies are not very active; tend to withdraw in novel situations; are not adaptable to change; have a mild disposition and negative moods.
Difficult babies are not very regular in their schedules; tend to withdraw in novel situations; are not adaptable to change; have intense dispositions and negative moods (Slentz & Krogh, 2001).
Data from the NYLS and a few other subsequent longitudinal studies have reported findings relating to the stability of some or all of the NYLS dimensions. Thomas & Chess, (1977) used ratings of parental interview protocols and found significant year-to-year correlation coefficients ranging to.52 in magnitude. For intervals of two years, correlations reached a max of.37. With 3-year intervals, the maximum correlation decreased to.29. In general, ratings of parental interview protocols in the NYLS reveals evidence of moderate stability over one and two-year intervals during early childhood, but low or non-significant stability for intervals ranging 3-years or longer (Guerin et al., 2003.) Hegvik et al. (1982) reported correlations between ratings over a 4.5-year interval spanning preschool and middle childhood and concluded that children’s temperament was more stable during this period when compared to infancy (Guerin et al., 2003).
In summary, data reveals an impressive level of stability in temperament both within and between periods of childhood as well as across the preschool and middle childhood periods (Guerin et al., 2003). According to another investigator, Rothbart (1981), although temperamental dimensions are more likely to be stable over time than are other aspects of behavioral individuality, stability is more likely within rather than between periods of rapid developmental change. According to him, the stability of temperament depends not only on environmental demands but also on intrinsic maturational processes (Lamb, Bornstein, Teti, 2002)
According to Kagan (1997) characteristics sometimes change as a result of efforts by parents to modify their children’s attributes (Lamb, Bornstein, Teti, 2002). Certain temperamental attributes are associated with differences in underlying biological or genetic functions but these attributes are not necessarily stable and unchanging (Lamb, Bornstein, Teti, 2002). Although variability in dispositional behaviors associated with emotional reaction, sociability, shyness, inhibition, and activity levels has strong psychobiological correlates, these systems are also plastic such that individual differences in these attributes are likely to remain more stable over time than are other behavioral characteristics, although they are not immutable and fixed (Lamb, Bornstein, Teti, 2002).
Buss and Plomin view temperament as having exclusively genetic than constitutional origins (Lamb, Bornstein, Teti, 2002). They stress that temperamental attributes remain highly stable throughout life, consistent with their genetic origins, and thus have enduring effects on personality development (Lamb, Bornstein, Teti, 2002.)
Guerin, DW, Gottfried, AW, Oliver, PH, Thomas, CW (2003). Temperament: Infancy through Adolescence. Springer.
Slentz, K, Krogh, S (2001). Early Childhood Development and Its Variations. Lawrence Erlbaum Associates.
Lamb, ME, Bornstein, MH, Teti, DM (2002). Development in Infancy. Lawrence Erlbaum Associates.