Are some temperaments “better” than others?

by Alexandre Erler

Jerome Kagan’s latest book, The Temperamental Thread, is – as usual with Kagan – a fascinating read. It draws on the three decades of research done by Kagan on the topic of human temperament. In a famous series of studies, Kagan examined the way infants reacted to unfamiliar or unexpected events. He found that about 20 per cent of these infants were unusually responsive to such events, exhibiting vigorous motor activity and frequent crying. He calls these infants “high reactives”, and found after following their evolution during their subsequent years that they were biased to become timid, subdued toddlers and shy adolescents who become uneasy when they cannot predict or control the future. About 20 per cent of these high reactives proved unable to cope with their temperament and were subsequently diagnosed with social anxiety disorder, depression, or both. By contrast, another other group of infants showed a high threshold of excitability to the same events. Kagan calls them “low reactives”. They tended to become outgoing, relatively fearless children and relaxed adolescents who like risk and challenge [3]. In the wake of Kagan’s earlier work The Long Shadow of Temperament, The Temperamental Thread paints a rich and detailed picture of the differences between these two psychological types.

 

As he explains in a piece written a few months ago for the Independent [4], Kagan also found that specific differences in biological features underlied these temperamental differences. One key factor is the degree of excitability of the amygdala, a part of the brain located at the level of the ears, which plays a key role in the processing of emotional reactions, including fear. Low reactives possess a less excitable amygdala than high reactives, and these biological differences are usually determined by the presence of certain genes. Kagan is careful to stress that temperament does not determine one particular personality, yet it does constrain the range of possibilities open to individuals.

The extent of the biological contribution to temperament might be disputed, but let us assume that Kagan is right on this issue. I suspect many will be tempted to ask whether we shouldn’t see a high reactive temperament as a curse for its possessor, at least in most Western societies. Kagan’s reply to that worry is meant to be reassuring for high reactives: he argues that “many occupations require individuals with this temperament”, and suggests that they will be able to flourish if they choose the right social “niche” for them, e.g. occupations like computer programmer, scientist or writer, which will allow them to have control over their environment and avoid (unpredictable and anxiety-provoking) interactions with other people. (One might object to Kagan that many scientists today work in teams, but let us ignore this). He also points out that only a minority of high reactive infants go on to become extremely shy later in life. Finally, he adds that high reactive adolescents are more concerned with academic failure and thus more likely to study hard and gain admission to a good college. They are also less likely to engage in risky behavior such as experimenting with drugs or having unprotected sex.

While I would very much like to share Kagan’s optimism about the consequences of these temperamental differences, I cannot help asking myself a number of questions, including the following:

-   Isn’t Kagan underestimating the negative consequences of conditions like social anxiety or shyness in British or American society? People with social anxiety are more likely to drop out of school early on and to become the targets of bullying, and they find it more difficult to make friends. They are less likely to get promoted at work. Given the dominant cultural expectation that men should initiate courtship with women, socially anxious men also have reduced romantic and mating prospects. It is therefore a bit surprising to hear Kagan advise against testing all four-months old children for high reactivity so that parents could take steps to reduce the probability of later shyness or social anxiety, on the grounds that this would create unnecessary apprehension in parents, given that most high reactive children do not go on to develop social anxiety. [5] Surely the disvalue of the consequences of social anxiety, even among a minority of high reactives, is higher than the disvalue of parents worrying needlessly about the possibility that their children will develop such a condition? Especially as social anxiety has a negative impact not only on the sufferers themselves but also on their parents, who have to go through the pain of seeing how the condition seriously harms their child’s well-being. Furthermore, plain shyness tends to have consequences similar to social anxiety, only to a lesser degree: it makes it more difficult for people to obtain the social goods that we tend to value. [2] If early testing could reduce the occurrence of such conditions better than a “wait-and see” approach, then surely it would be a good thing to do (assuming it were technically feasible).

-   This leads us to a related question: isn’t a high reactive temperament more limiting in terms of life possibilities than a low reactive one, given that it increases the probability of developing conditions like shyness or even social anxiety and depression, with all their incapacitating consequences? If so, it would at least be morally desirable, perhaps even obligatory, for parents of high reactives to inform themselves about their children’s temperament and to make sure they help them overcome their inhibitions and develop adequate social skills by gradually exposing them to the unfamiliar (as Kagan himself suggests). Also, assuming our knowledge and control of genetics makes this possible for us in the coming decades, we might have a reason to bring future people within a certain range of comparatively low reactivity through the use of genetic technology. This could be justified by appealing to Joel Feinberg’s idea of the child’s right to an open future: according to Feinberg, good parents will ensure that their child enters the adult world “with as many open opportunities as possible, thus maximizing his chances for self-fulfillment” (Feinberg [1]).

To this it might be replied, first, that high reactives tend to be introverts, that is, people with a natural preference for solitary activities and a lesser degree of responsiveness to social rewards, and therefore that having a lesser natural ability to obtain such rewards does not constitute a handicap for them. But such a claim stands in need of much empirical support. The available evidence shows that high reactives are more vulnerable to depression than low reactives. It is tempting to suppose that this is (at least partly) related to a deficit in their social abilities: it is not clear from what Kagan says that all high reactives are natural introverts – and even introverts have ambitions about their career, and their happiness partly depends on the quality of their social and romantic life. Also, even in the case of high reactives who seem satisfied with living in their protective niche, we might legitimately wonder to what extent their lifestyle is really in keeping with their natural preferences, as opposed to adaptive preferences. One would like to know: isn’t it the case that high reactives simply cannot afford to desire and value certain things, given the constraints that their temperament imposes on them? Would they still look down on a promotion, on joining that particular student society, or on a relationship with the good-looking, popular girl in their class if they did not believe such things to be beyond their reach?

One might counter here that a low reactive temperament also closes off a number of life options, and therefore that it does not guarantee a more open future for its possessor than a high reactive temperament would – it just rules out different possibilities. Maybe so; but we should bear in mind here that the possibilities that count are possibilities that people in our society typically value. It might be more difficult for low reactives to develop social anxiety, for instance, but hardly any of them, I take it, will deplore this fact. And I am not entirely convinced that when it comes to desired goods and human flourishing, the two temperamental biases are on a par. Are low reactives at any disadvantage in terms of professional success compared to high reactives? This seems unlikely, given that confidence and social skills are just as important (if not more so) in this regard as educational attainment. And even if we focus specifically on academic achievement, I want to ask: why do low reactives tend to care less about their grades than high reactives? Are they simply naturally less inclined to value an academic career? If so, one cannot properly say that their temperament “limits” them in this. Or is there a correlation between low reactivity and low conscientiousness, or poor impulse control? One would like to see some data on this issue. Clearly, a propensity to fear and worry is not the only possible (and probably not the ideal) incentive for hard work. To what extent can low reactives who are bothered by their excessive carelessness learn to discipline themselves? And if they can, will this typically be an easier feat to accomplish than for a high reactive to become socially savvy and confident, or not?

-   Finally, those of us who favour an Aristotelian perspective on the virtues might wonder whether high reactivity does not also constitute a handicap when it comes to developing the virtue of courage. If, as Aristotle suggested, courage means experiencing fear to the right degree, at the right time and towards the right things, it might seem that high reactives are naturally more at risk of fearing the wrong things, or of fearing appropriate objects to an excessive degree, and thus that they will find it harder to become courageous. (Aristotle wouldn’t have been impressed by the idea that high reactives are capable of making as great an effort as low reactives at overcoming their fear, as for him the end result counts for courage, not just the effort one puts in.) True, even so, it might be the case that high reactives have an advantage over low reactives when it comes to developing other virtues, such as empathy. This is again an issue that would seem worth investigating.

To conclude, I do hope that my reading of Kagan has led me to worry unnecessarily about the life prospects of high reactives. But I believe that the various issues I have raised deserve to be discussed, no matter how politically incorrect they might seem. It is not in the interest of people with such a temperamental bias, or of our broader society, to keep such issues taboo.

 

REFERENCES:

[1] Feinberg, Joel. “The Child’s Right to an Open Future”. In Whose Child? Children’s Rights, Parental Authority, and State Power, edited by William Aiken and Hugh LaFollette, 124-153. Totowa, NJ: Rowman & Littlefield, 1980.

[2] Henderson, Lynne & Zimbardo, Philip. “Shyness”. In Encyclopedia of Mental Health. Academic Press, San Diego, CA, in press. Available online at http://www.shyness.com/encyclopedia.html (accessed October 29, 2010).

[3] Kagan, Jerome. The Temperamental Thread: How Genes, Culture, Time and Luck Make Us Who We Are. Dana Press, 2010.

[4] Kagan, Jerome. “Born worried: Is anxiety all in the genes?” The Independent, May 24, 2010. Accessed October 29, 2010. http://www.independent.co.uk/news/science/born-worried-is-anxiety-all-in-the-genes-1981022.html.

[5] Kagan, Jerome. “New Insights into Temperament”. The Dana Foundation website, January 1, 2004. Accessed October 29, 2010. http://www.dana.org/news/cerebrum/detail.aspx?id=1284.

 

  • Facebook
  • Twitter
  • Reddit

6 Responses to Are some temperaments “better” than others?

  • Before we get too involved in wondering how to lessen the handicap to high-reactives, we need a much better grip on the data. We need to know how the actual distributions compare, not just the averages. Averages, standard deviations and the like are after all properties of populations, not of individuals. It might be quite easily that there is a statistically significant difference between the high- and low-reactives in their likelihoods of having various personality types, but still have 40% of low-reactives be more outgoing than 50% of high-reactives. It would be a ludicrous overreaction in such a case to single out the high-reactives for special treatment: the main thing it would accomplish is stigmatization. On another high-class academic blog, Language Log, the linguist Mark Liberman has been on a crusade against overanalyzing such statistical results, and this post is a wonderful introduction to the issues with how misleading natural language can be about comparisons between populations: http://languagelog.ldc.upenn.edu/nll/?p=1297

  • There’s an unfortunate typo in there: it could be that 40% of high-reactives turn out to be more outgoing than 50% of low-reactives. So, the low-reactives would still on average turn out to be more outgoing than the high-reactives, but you could still have a little less than half of high-reactives be more outgoing than your average low-reactive. Without knowing the standard deviations of the two populations and the effect size of the difference between them, we have no idea how any one high-reactive compares to a low-reactive, just how the averages compare. And even with a statistically significant difference in the averages, you could have that a large number of high-reactives are more outgoing than the average low-reactive, to the degree that trying to address the disparity would be pointless busywork. Natural language is unfortunately very vague where the comparison between the properties of populations is involved, and neither the OP nor Kagan’s article for the Independent helps here (I haven’t read the book, which will almost certainly have the relevant information).

  • Marinus: I agree that we need a better grip on the data before we are able to make any confident claims (which I haven’t tried to do in this entry). When it comes to statistics, Kagan’s book states that 20 percent of the high-reactive infants he had observed had become extremely shy, fearful adolescents. On the other hand, only about 10 percent had developed the contrasting profile for their temperamental bias, i.e. an exuberant and highly sociable personality. Which leaves us with about 70 percent of high-reactives who end up somewhere in the middle (call them “moderately outgoing”). As for low-reactive infants, 40 percent of them ended up at the lower end of the spectrum (sociable, exuberant, with a minimally excitable amygdala), while less than 10 percent went on to become extremely shy, which puts slightly under 60 percent of them in the “moderately outgoing” category (see The Temperamental Thread, pp.48-49). I think these figures make it rather unlikely that a large number of high-reactives might be more outgoing than the average low-reactive, though of course we’d like to know more precisely where people in the “moderately outgoing” category fall on the spectrum. Kagan suggests that high-reactives in this category have learnt to cope with their shy tendencies, but their anxious feelings remain, which suggests that these people might be at a disadvantage in terms of subjective well-being compared to low-reactives in the same category, even if there is no significant difference between them in terms of behavioral outcomes (Kagan himself expresses such a worry on pp.50-51 of the book).

    Even if we assume that there’s no need to worry about people who fall into the “moderately outgoing” category, we might wonder what to think about the “extreme” ends of the spectrum. Being extremely shy seems typically undesirable (except maybe for a select few individuals like TS Eliot, Kagan’s favorite example), but the same does not seem to hold for minimal reactivity. True, if a minimally reactive child grows up in the wrong kind of environment, he might be at greater risk of engaging in criminal behavior. But clearly this is not the case of the 40 percent of low-reactives at the lower end of the spectrum (assuming of course that these figures are representative of the general population). On the contrary, being minimally reactive increases a child’s chances of growing up to be a resilient adult, presumably a desirable quality to have. Considerations like these help explain why a low-reactive temperament might seem “preferable” in a society like ours.

  • Anca Gheaus says:

    Hi Alex, I think it would be important to substantiate your claim that “the possibilities that count are possibilities that people in our society typically value.” This does not seem obvious. What if the people in our society typically value the wrong, or the less good, possibilities? Why not say instead that the possibilities that count are (the wide range of) possibilities to lead a good life? Whether there is an overlap between the latter and the possibilities that people in our society typically value, and how large the overlap is, is a different question.

    Moreover, suppose the 80% of ‘high reactives’ who do not develop social anxiety or depression are indeed introverts. To me it is not at all clear that, *even in a world which devalues introversion*, introverts really have lower prospects of leading a good life than extroverts. Perhaps they are more likely to have fewer, steadier and more substantial relationships? Perhaps they are less likely to attach too much value to transitory goods?

    This opens the question whether what needs to be changed is the individual temperament, or the ‘public values’. Even in the case of the 20% ‘high reactives’ who develop social anxiety or depression there is a possibility this happens only/mostly under undue social pressure, which as such stands in need of scrutiny.

    One more thing, on the Aristotle connection: it strikes me that a contemplative life is intrinsically more accessible to an introvert than to an extrovert, whereas an active life is something both an introvert and an extrovert may lead successfully depending on external circumstances – although perhaps there are no imaginable external circumstances that would make the ‘active life’ equally accessible to introvert and to extrovert individuals.

    But, as you say, all these possibilities invite further reflection.

    PS I wonder if it was my temperament that made me disinclined to pursue a relationship with the good-looking, popular girl in my class…

  • Anca:

    Thanks for your comment. You’re right that the phrasing of my sentence is not satisfactory, for the reasons you mention. I should have spoken, rather, of the things people typically value, and to which we have no good reason to object. Among those things, I take it, are fulfilling social interactions and relationships: human beings are fundamentally social beings. Now a propensity to anxiety can easily make it more difficult to secure that good, because anxiety tends to impair social performance, which makes it more difficult for those who have it to meet new people and also to connect with them. Hence my worry.

    The question whether we should question social preferences rather than try and help individuals to comply with them is a valid one, but I’m not sure it would be appropriate to just blame society for the problems that accompany social anxiety. There arguably are certain things we can legitimately expect from people, such as showing courtesy and indulgence towards those who have difficulties with their social abilities. But I don’t think we can legitimately require anyone to e.g. hang out with, or date people (s)he finds boring or unattractive. If so, a critique of social preferences, while it might be necessary, will nevertheless not be sufficient to tackle these problems.

    Finally, let me stress that I had no intention at all to suggest that introverts have lesser prospects of leading a good life than extroverts. Rather, I just wanted to ask to what extent is there a harmonious fit between the “natural” preferences of high reactives and their social abilities. If all high reactives were introverts, that would already suggest a better fit than if (as I suspect is the case) some were extroverts, as extroverts naturally crave social contact more than introverts do. My worry is not that certain people lack certain preferences (again, I have no wish to present extroversion as the “ideal” personality to have), but rather that a temperamental bias like high reactiveness is likely to make it more difficult for those who have it to satisfy the natural preferences that they will likely have. And again, I’d be glad to be proven wrong on this.

  • Just to remove a possible ambiguity in my previous comment: I am of course not claiming that socially anxious individuals ARE in fact boring or unattractive – only that they are, sadly, more likely to be perceived as such (especially men, when it comes to attractiveness) in informal interactions with unfamiliar people, as they find it harder to make their various qualities known to others.

Recent Comments

Authors

Affiliations