Historical Development
Alexander Thomas and Stella Chess's pioneering New York Longitudinal Study (NYLS), initiated in 1956, tracked 133 children from infancy into adulthood, establishing temperament as a measurable individual difference domain distinct from personality and intelligence. Their 1977 book "Temperament and Development" proposed that behavioral styles emerge early, show consistency across contexts and time, and interact with environmental factors to shape personality and outcomes.
This foundational work inaugurated modern temperament research, moving beyond earlier categorical typologies (Galen's humors, Sheldon's somatotypes) to dimensional assessment.
Nine Dimensional Model
Thomas and Chess defined nine independent dimensions: (1) Activity Level—the proportion of activity to passivity; (2) Rhythmicity—the regularity of biological functions (sleep, hunger, bowel movements); (3) Approach-Withdrawal—the initial response to novel stimuli (food, people, situations); (4) Adaptability—the speed and smoothness of adjustment to environmental change; (5) Intensity—the energy of response, whether positive or negative; (6) Mood—the predominant affective tone (positive vs. negative); (7) Persistence—the duration of attention and motivation in pursuing goals; (8) Distractibility—the ease with which external stimuli interrupt ongoing behavior; and (9) Sensory Threshold—the minimum stimulus intensity required to elicit a response.
Three temperament constellations emerged from NYLS data: Easy (40% of children, positive approach, adaptability, regular rhythms), Difficult (10%, negative approach, low adaptability, high intensity, negative mood), and Slow-to-Warm-Up (15%, withdrawal, low adaptability, mildly negative mood). The remaining 35% did not fit clear patterns.
Kagan's Behavioral Inhibition Paradigm
Jerome Kagan's research program, summarized in his 1994 book "Galen's Prophecy," identified behavioral inhibition—a temperamental trait characterized by wariness toward novel situations, delayed approach to unfamiliar people and objects, and physiological reactivity (elevated heart rate, cortisol). Using longitudinal designs from infancy through childhood, Kagan demonstrated that inhibited children exhibited greater amygdala reactivity and right prefrontal dominance in electroencephalography (EEG).
Longitudinal follow-up showed behavioral inhibition predicted anxiety disorders in adolescence (Kagan et al. 1988, Child Development) and social inhibition in adulthood. Approximately 15-20% of children exhibit high behavioral inhibition.
Kagan's work integrated physiology, behavior, and development, moving temperament research beyond parental reports to objective observation and neural measurement.
Rothbart's Effortful Control
Mary K. Rothbart's 2000 Advances in Child Development and Behavior paper synthesized decades of research into the Temperament in Childhood Questionnaire (TMCQ) and later the Child Behavior Questionnaire (CBQ), emphasizing three core dimensions: (1) Surgency/Extraversion (approach, activity, vocal intensity); (2) Negative Affectivity (fear, sadness, anger frustration); and (3) Effortful Control (inhibitory control, attentional focusing, low-intensity pleasure).
Effortful Control—the capacity to voluntarily suppress dominant responses and activate weak responses—emerges in the second year of life and shows substantial development through early childhood. Neuroimaging studies identify prefrontal-amygdala connectivity as the neural substrate of effortful control (Rothbart et al.
2011, Developmental Psychology). Effortful control predicts academic achievement, social competence, and reduced externalizing behavior.
Measurement Instruments
The Dimensions of Temperament Survey—Revised (DOTS-R), developed by Windle and Lerner (1986), provides brief assessment of temperament across the lifespan. The Revised Dimensions of Temperament Survey (DOTS-R) includes 54 items measuring 9 scales.
Parent report and teacher report versions allow cross-situational comparison. Behavioral observation protocols (Laboratory Temperament Assessment Battery, Lab-TAB) provide objective measurement beyond questionnaires.
Physiological measures (heart rate variability, cortisol response, EEG asymmetry) validate self-report and observer measures.
Stability and Change Debate
Central to temperament research is the question of developmental stability. Thomas and Chess reported moderate stability correlations (r = 42– 75) across their study, suggesting both consistency and plasticity.
Caspi's 1997 meta-analysis (Psychological Bulletin) of longitudinal research found behavioral inhibition showed 22-year stability (r = 30), child difficultness predicted adult externalizing problems, and childhood undercontrol correlated with adult impulsivity.
However, Kagan (1989, Social Development) emphasized that behavioral inhibition can decrease with age if environmental support is provided. Contemporary consensus (Roberts & DelVecchio 2000, Psychological Bulletin) is that temperament shows significant cross-time consistency (mean r =
54) but considerable malleability through experience and volition. Gene-environment interactions substantially moderate temperament trajectories (Caspi et al. 2002, Developmental Psychology on diathesis-stress models).