The ‘other-bodies’ problem in stress

Following on from the previous article about the ‘other-bodies’ problem in attention, here is a piece about the ‘other-bodies’ problem in stress.

So we know – because Hans Selye defined it that way – that the stress response is, fundamentally, an adaptation response. Our stress systems respond dynamically to allow us to maintain homeostasis and allostasis in the face of a changing external environment.

At the same time, I would say, most current approaches to defining stress don’t do a very good job of capturing this dynamically fluctuating nature. Just as in attention, where most researchers have characterised attention as a property of individual brains considered in isolation, so most stress research has focused on stress reactivity as a parameter of individual differences. An individual’s level of stress reactivity is (implicitly, although it’s never actually described as such) considered a static (time-invariant) feature. (I have definitely been as guilty of this as others.) And, similarly, stressors are conventionally assessed using questionnaires - which also provide a static, time-invariant, snapshot of the total stressors to which an individual has been exposed. For my current Fellowship I'm trying to develop better methods to capture how moment-to-moment changes in the outside world associate with moment-to-moment changes in our physiological stress systems.

But how can this help us to understand pathological stress – which is the concept of stress which most people are thinking of when they actually refer to stress? One really interesting idea here is that of ‘stress contagion’ – how we are directly, and immediately, influenced by people around us. Sara Waters and Wendy Berry Mendes have done some really interesting studies looking at this recently. In one, for example, mothers were initially separated from their infants and assigned to either a low-arousal positive/relaxation condition, or a high-arousal negative/stress task. Upon reunion, infants were placed either on their mothers’ laps or in a high chair next to the mother. They found that infants’ physiological changes on reunion reflected the experiences that their mother had been through during the separation – so infants whose mothers had had the relaxation condition showed decreases in arousal, and infants whose mothers had had the stress task showed increases in arousal. Infants who sat on their mothers’ lap showed stronger coviariation, suggesting that physical touch can play a role in stress contagion.

Intuitively (always a dangerous word!) this seems possible: when I’m in the car and everyone in the family is shouting at each other, it certainly feels like it increases my stress levels. But we still don’t know, in terms of physiological stress (for cortisol, sure) whether moment-by-moment fluctuations in physiological stress do actually associate with moment-by-moment fluctuations in the stress levels of people around us. (As well as the direction of the effect – whether children influence parents’ stress more than vice versa.) If so, this would be an immediate, and environmental, mechanism by which the children of parents who are more stressed become stressed themselves – and it would offer potentially important intervention targets.

Another interesting - but, I think, under-explored - aspect of interpersonal stress dynamics is parental responsivity. We know that parental sensitivity is thought to be an important influence on childrens’ long-term levels of physiological stress and cognitive performance – albeit, again, largely from studies that have treated parental responsivity as a single, time-invariant feature of individual differences. But we don’t know much about how the dynamics of responsivity work.

How, exactly, do more sensitive parents show different patterns of co-fluctating stress with their children during the day? For example, do the children of parents who are less sensitive show larger moment-to-moment variability in physiological stress - because they don’t have a sensitive parent to 'catch them' and help them calm down when they approaching the ‘red zone’? And does this larger moment-to-moment variability somehow ‘burn out’ the stress system, learning to maladaptive long-term stress patterns?

Again, this would have important intervention implications if it were true. But one of the reasons we don’t know the answer to these questions is that most studies have, for practical reasons, only looked at children and parents over short time-frames in lab-based studies. Even parental sensitivity is, almost invariably, defined from video short sections of shared parent-child play – and when the cameras are watching, so parents are on best behaviour anyway. What we don’t have data on is what happens when a parent ignores, or shouts at, a crying or distressed child – because they don’t do this when the cameras are watching.