Born worried: Is anxiety all in the genes?

Sigmund Freud believed that anxious personalities are shaped in childhood. But after a lifetime of study, Harvard psychologist Jerome Kagan is convinced that it's all in the genes

Alice, a young scientist working in a biological laboratory, likes her work, and her supervisor values her conscientiousness and perfectionism.

But when her colleagues take a coffee or lunch break she usually sits quietly on the periphery of the group, fingering her hair and blinking rapidly if someone asks her a question. Alice dreads the occasions during the year when it is her turn to tell the staff about the experiments she is working on and what she has learned. She usually doesn't sleep well on the nights before these performances, and her mouth becomes dry and her palms sweaty when she stands up to give her report.

During the period between the two World Wars, when Sigmund Freud's ideas were dominant, most psychiatrists and psychologists would have attributed Alice's behaviour and moods to her childhood experiences. Perhaps Alice's mother had been hypercritical and she felt guilty over her anger towards the mother who had punished her occasional disobedience severely.

I had accepted such experiential accounts as an obvious truth when I was a graduate student at Yale University from 1950 to 1954. But I began to question my inflexible commitment to the sole influence of early experience in 1962, when Howard Moss and I were reflecting on the evidence we had gathered on a group of normal adults, born in the 1930s, who were members of a longitudinal study conducted at the Fels Research Institute in Ohio. About 15 per cent of these adults, who had been timid, fearful, and shy during their first three years, resembled Alice: they told me that they were shy, often felt unsure, avoided risky activities, and were reluctant to take on difficult challenges. When Howard and I wrote the book Birth to Maturity summarising the project, we suggested that these children inherited a constitutional disposition. This was our way of saying that their temperament was relevant – where temperament refers to a set of biological properties affecting brain chemistry that are usually, but not always, due to the presence of specific genes.

A second attack on my premise that only experience mattered occurred almost 20 years later, when Richard Kearsley, Philip Zelazo, and I were examining evidence we had gathered on the effects of day care on Chinese-American and Caucasian infants who either attended a day care centre we managed, or were raised only at home from age three to 29 months.

We were surprised by the fact that the Chinese-American infants, whether attending our centre or raised only at home, were more quiet, timid, and shy than the Caucasian-American infants. This observation also suggested that the two groups began life with distinctive temperaments. Finally, the writings of Alexander Thomas and Stella Chess, two psychiatrists who angered their psychoanalytic colleagues in the Sixties by suggesting children began life with distinctive temperaments, provoked brooding on this idea and the desire to study infant temperaments in my laboratory.

Humans are capable of possessing a very large number of temperamental biases, but my colleagues and I have been studying only two members of this extensive set. One temperament biases children to experience uncertainty, fear, or anxiety when faced with unfamiliar or unexpected events; the other bias protects children from these emotions. We have been following a large group of Caucasian children born healthy to intact, middle-class families in the Boston metropolitan area in 1989 or 1990, from four months of age until 2007-08, when they were 18 years, and have arrived at several conclusions.

First, about 20 per cent of these infants were unusually responsive to events that were unexpected or unfamiliar. When colourful toys they had never seen before were moved slowly in front of their faces, the four-month-olds begin to thrash their limbs, arch their backs, and cry, as if this experience was too arousing for their brains. These infants, whom we call high-reactive, tended to be shy when they met unfamiliar people and afraid when they entered unfamiliar rooms or encountered unfamiliar objects during their second year of life. Most were shy when interacting with unfamiliar adults or children at four-and-a-half years of age, and, at seven, many required a night light while sleeping, were afraid to spend the night at a friend's home, and were generally quiet in the school setting. As adolescents, many told us they worried excessively about the future and were anxious over events that most adolescents regarded as unrealistic. For example, they didn't like talking to strangers and felt anxious when they had to go to a party where they knew few people. Some were reluctant to take the subway from their suburban home to central Boston, or worried when they heard the sound of an ambulance. One high-reactive adolescent said that she didn't like spring because the weather was unpredictable. The central source of their tension was not knowing what might happen in the next hour, day, or week. As a result, about two-thirds of these adolescents plan most days and prepare for potentially threatening contingencies. Their friends see them as hypervigilant, careful, cautious, and reflective.

We have examined aspects of their biology and, to our pleasant surprise, found that their physiology is in theoretical accord with their behaviour. The critical structure, called the amygdala, is tucked inside the brain at the level of the ears. The amygdala is the personal antennae, for it detects changes in the outside world and inside the body and reacts to all new or unexpected events, or to anticipations of such experiences, by sending impulses to the heart, skin, circulatory vessels, and muscles, leading to a rise in heart rate, increased blood pressure, motor tension, altered breathing, and sweating of the palms. High reactives seem to possess a chemistry of the amygdala that renders it more excitable than it is with most children.

Carl Schwartz, a psychiatrist working at Massachusetts General Hospital, observed these youths at age 18 in a magnetic scanner and found the high reactives had a thicker cortex than others in a small section of the frontal lobe on the right side (called the medial prefrontal cortex) that communicates with the amygdala. Second, their amygdala remained active for a longer time to unfamiliar pictures, such as a person's head on an animal's body – suggesting that this structure resisted regulation. However, many high-reactive adolescents were no longer especially shy or timid. The consequences of their temperamental bias were now primarily on the inside, and reflected in private worries not always apparent to parents or friends. Jung's twin concepts of persona (the surface behaviour shown to others) and anima (one's private feeling tone) help us understand this phenomenon. It is easier for individuals to change their persona than to alter their anima – which contains patterns of private worries and feelings.

The majority of our high-reactives have not seen a psychiatrist and probably will not develop any psychiatric disorder. However, about 20 per cent have been unable to cope with their temperament and have been diagnosed with social anxiety disorder, depression, or both. Only one of our original group of 90 high-reactive infants committed suicide. The parents of these children may wonder whether this temperamental bias is a serious disadvantage in contemporary society, but fortunately many occupations require individuals with this temperament.

These adults like to work alone in settings in which they have control of most moments of the day. Therefore, vocations like computer programmer, writer, scientist, or historian appeal to them, and they make important contributions to society. The biography of the poet T S Eliot and the memoir of the biologist Rita Levi-Montalcini, who both won Nobel prizes, suggest they had Alice's temperament. I remember telling under- graduates that, if we had asked Eliot or Levi-Montalcini a few days after they had won the prize whether they would have wanted their parents to take them to a child psychiatrist to cure their excessive anxiety, both would probably have said "no", because they would recognise that their temperamental preference for solitary work made it possible for them to be creative.

The most important conclusion is that the power of an infant's temperament rests with its ability to limit what a child might become, rather than determining a particular personality.

Although only 20 per cent of our high-reactives were unusually shy, introverted 18-year-olds, over 90 per cent were not consistently exuberant, worry-free youths who loved risks, crowds, and new sensations. Knowledge of a person's temperament allows one to predict with confidence what they will not become, but is less able to predict the persona they will develop. Predicting the exact location where a rock will land after rolling down a mountain provides an analogy. Knowledge of the location, size, and shape of the rock, analogous to a temperament, permits an observer to eliminate many places where it will not rest, but it is impossible to know its final location because one cannot know the gullies and obstacles it will strike on its way down. These events are analogous to the diverse experiences each person encounters as they grow to maturity.

Dr Jerome Kagan is a professor of psychology at Harvard University. His latest book, 'The Temperamental Thread: How Genes, Culture, Time and Luck Make Us Who We Are' is published by Dana Press (£16.85). To order a copy for the special price of £13.54 (free P&P) call Independent Books