Psi man mascot

Attachment and Shyness

Attachment is the intense relationship to one person typical of the mother/infant relationship. Security of attachment to a primary caretaker–usually the mother–helps a baby gain confidence and the willingness to explore its environment. A baby who seems to be secure and confident, not overly shy or anxious, is called securely attached.

What is attachment? What attachment styles did Ainsworth observe?

For many years psychologist Mary Ainsworth studied attachment between babies and mothers. She distinguished between securely attached and anxiously attached children. The anxiously attached group was divided into two subtypes: anxiously attached avoidant babies and anxiously attached ambivalent babies.

Securely attached infants appear active and happy. They are willing to explore a new room if the mother is present. They warm up quickly to a stranger who talks with the mother first.

They are not greatly disturbed if the mother is briefly absent from the room (although this may depend on other factors such as mood and tiredness). If all the adults leave, so the baby is alone in the room, the securely attached infant becomes anxious and runs to the mother's side when she reappears.

Anxiously attached infants of the first (ambivalent) type do not explore a strange room full of toys. They cry and cling to the mother even before being separated from her. They act suspicious of strangers and get very upset if the mother leaves the room. When she returns, they may pout or continue crying.

This type of anxiously attached child is called ambivalent (caught between two opposites) because they seem depen­dent upon the mother and insecure when she is gone. But when she reappears, they do not act affectionate or loving toward her.

Anxiously attached infants of the second sub-type (avoidant) are not upset by separation from the mother. They do not cry when she leaves. When she returns, the baby may ignore her, react casually to her presence, or avoid her.

This type is called avoidant because the more distressed such a baby becomes, the less it seeks contact with the mother. That stands in sharp contrast to securely attached infants, who seek contact with the mother in stressful environments.

Continuity of attachment

The securely attached infant seems psychologically healthy, but what happens after a few years? Matas, Arend, and Sroufe (1978) decided to study continuity of attachment styles. Would the children who were securely attached at age 1 be strong and healthy personalities at age 2?

As a control procedure, different testers did the assessments each time. The later testers had no knowledge of the earlier results. The same children rated as securely attached at 18 months were described as "more enthusiastic, persistent, cooperative, and in general, more effective" at age 2.

What were early findings from research on continuity of attachment?

In follow-up studies, the same children were rated highly competent at the age of 4 in peer interactions at a nursery school. They got along well with other children. They showed a zest for pursuing new intellectual skills, ability to direct them­selves, and ability to tolerate novelty in the environment.

What were the results of a long-term study of 180 people?

Intrigued by these results, Sroufe and colleagues started the Minnesota longitudinal study of parents and children. The goal was to study of 180 individuals from birth to adulthood. The children were born into poverty. As of 2010 they were 34 years old.

Longitudinal studies like this are the gold standard of developmental psychology research. They are expensive and time-consuming, because individuals must be followed and tested again years later. However, only this type of research can provide data about how individuals change.

The other type of developmental research, used for trying to detect age trends, is cross-sectional research. In this type, people of different ages are sampled all at the same time.

That is quicker and less expensive than longitudinal research, but it cannot show how individuals develop through the years. Only longitudinal research could show long-term effects of early attachment styles.

What are advantages and disadvantages of longitudinal studies?

As the results of the Minnesota longitudinal study came in, they went contrary to expectations. There was no clear continuity between childhood attachment styles and adult relationship styles. "Evaluating the role of variations in early human experience turns out to be more complicated than first meets the eye," wrote the researchers.

They cited other factors that might have overwhelmed the effects of early attach­ment styles. Those other possible factors included maternal nutrition before birth, family stress, and challenges of growing up with a low income.

There were, however, some interesting correlations. Individuals who were anxious, avoidant types as babies showed more dependence on teachers and counselors in school, through middle school. Babies who were securely attached showed more comfort being vulnerable (showing trust) in teenage relationships.

Predictions of later behavior were more accurate when additional measures of family troubles were added. Vari­ables such as witnessing violence were powerful predictors of later trouble.

What factors correlated with better outcomes in adulthood?

There was also "abundant evidence" that some individuals could overcome early problems, especially if they had good relationships in adulthood. Children with behavior problems were most likely to be happy adults if they had (1) good romantic relationships, and (2) satisfactory work.

Some children showed "consistently poor adaptation" in three different preschool assessments. Among that group, those with secure attachment in early childhood were more likely to bounce back and have good school years.

Among those who still had problems during school years, those with early secure attachment were more likely to do better in adult years. In short, early secure attachment seemed to increase resili­ence or ability to overcome adversity, as a long-term effect.


Of all the temperamental factors studied, shyness is the most consistent personality trait from early childhood through adolescence and adulthood. Several lines of research suggest that shyness is determined more by biology than by upbringing or parenting style.

Kagan, Reznick, and Snidman (1988) reported a longitudinal study of shyness in which the same children were tested repeatedly from age 2 to 7. Shyness was defined as a tendency to become quiet, alert, and emotionally subdued in unfamiliar situations. The researchers found a high degree of stability in this trait among their subjects.

The same children shy at age 2 were shy at age 7. The researchers posited an "inherited variation in the threshold of arousal in selected limbic sites." In other words, in these children, the amygdala alarm system in the brain might be more active, so more situations were inter­preted as threatening.

What did Kagan's research group discover? How might the amygdala be involved?

Kagan and colleagues also noted that most of their shy subjects were younger children in multi-child families. Older children tended to be more outgoing.

The researchers speculated that older children might contribute to shyness in their younger brothers or sisters through such domineering tactics as grabbing away toys when the younger children were still vulnerable toddlers. There are other possibilities as well; males with older brothers tend to have less testosterone, on average, and this could contribute indirectly to shyness.

Biological and environmental effects are intertwined in normal development. A child with an overactive amygdala, or less testosterone, might develop learned avoidance behaviors if domin­ated by older siblings, resulting in shyness.

How might environmental and biological influences on shyness interact?

Karevold, Ystrom, Coplan, Sanson and Mathiesen (2012) studied 921 children from 1.5 to 12.5 years. They found "moderate stability" in shyness, with a tendency toward poor social skills and higher levels of anxiety and depression symptoms in those developing shyness later in childhood rather than in babyhood.

Those developing shyness later would presumably be those responding more to environmental effects, such as inter­actions with siblings or other children. Shyness due to inborn differences relating to the amygdala or testosterone should be detectable earlier in life. Those shy earlier in life had better outcomes in the Karevold, Ystrom, Coplan, Sanson and Mathiesen (2012) study.

Grose and Coplan (2014) reported a Canadian study in which shyness was measured twice: at age 8-9 years old (3,514 children) and at age 20-21 (1,447 who could be located for testing). Childhood shyness was, as usual, a significant predictor of adult shyness, but in this study it predicted social adjust­ment difficulties only for females.

A Swedish study in which babies were tested 16 years later showed, again, that extreme levels of shyness were predic­tive of later problems, but only in girls. This led the researchers to speculate that "culturally shared notions of gender-appropriate behavior influence the stability of inhibition."

In other words, shyness might be encouraged as part of a girl's personality in some cultures or families. That could make the trait more likely to last into adulthood and cause relationship difficulties.


Grose, J. & Coplan, R. J. (2014) Longitudinal Outcomes of Shyness From Childhood to Emerging Adulthood. Journal of Genetic Psychology, 176, 408-413.

Kagan, J., Reznick, J. S., & Snidman, N. (1988). Biological bases of childhood shyness. Science, 240, 167-171.

Karevold, E., Ystrom, E., Coplan, R. J., Sanson, A. V. & Mathiesen, K. S. (2012) A prospective longitudinal study of shyness from infancy to adolescence: stability, age-related changes, and prediction of socio-emotional functioning. Journal of Abnormal Child Psychology ,40, 1167-1177. doi:10.1007/s10802-012-9635-6.

Kerr, M., Lambert, W. W., Stattin, H., & Klackenberg-Larsson, I. (1994) Stability of Inhibition in a Swedish Longitudinal Sample. Child Development, 65, 138-146. doi:10.2307/1131371.

Matas, L., Arend, R. A., & Sroufe, L. A. (1978) Continuity of adaptation in the second year: The relationship between quality of attachment and later competence. Child Development, 49, 547-556.

Sroufe, L. A., Coffino, B., & Carlson, E. A. (2010) Conceptualizing the Role of Early Experience: Lessons from the Minnesota Longitudinal Study. Developmental Review, 30, 36-51. doi:10.1016/j.dr.2009.12.002.

Write to Dr. Dewey at

Don't see what you need? Psych Web has over 1,000 pages, so it may be elsewhere on the site. Do a site-specific Google search using the box below.