Part 2 : Thinking Straight about Uncertainty - bootstrapping beliefs

In Part 1 of this series I suggested that our binary approach to stress was neither helpful or accurate and that our response to stress is highly nuanced - there is no typical response to stress because there is no typical stress.

In this Part 2 I investigate what happens when uncertainty increases and things become ambiguous. What happens when everything is unknown and uncertainty becomes irreducible? Under these 'Black Swan' conditions do we really lose control and have an 'amygdala hijack' and run for our lives or are things more complicated than that?

Acute Stress = Ambiguous Uncertainty

In Part 1 I described certainty as being a bucket full of the same coloured balls - when we put our hand in the bucket to choose a ball we can be certain of the outcome. In risky situations we have different coloured balls in the bucket but we can calculate the likelihood of the outcome so there is some degree of certainty. Now imagine we have the same bucket, but it only has one red ball and one green ball – everything else is unknown. Although the odds of getting a red and green ball are the same (50:50) as the risky scenario our amygdala interprets the ‘unknowns’ (or missing information) as a threat. Similarly even something as simple as flicking a coin can confuse us – people think the odds change with each flick because intuitively HHHHTTTT seems less likely than HTHTTHTH when in fact they are equally likely. This is called the ambiguity effect - a cognitive bias were we confuse missing information with negative information. In contrast to the risk condition described in the previous post, the ambiguity effect increases activity in the OFC (orbital frontal cortex) and the dorsomedial prefrontal cortex (DMFC) regions but the striatum seems to play less of a role. This quietening of the striatum is because it doesn’t know what success looks like and so cannot anticipate the reward. This creates anxiety and additional resources from the PFC to regulate behaviour.

The OFC is involved in higher order behaviour regulation. Faced with uncertainty people with large OFCs are less anxious and more optimistic. The OFC helps us to make choices relative to their perceived reward. We receive sensory information (the ‘WHAT’ from vision, taste, smell, touch etc) the OFC then assigns some value to them (e.g. if we are not hungry a zero value is assigned to food) which is projected to the ACC - a ‘conflict monitor’ that integrates beliefs and risks of attainable goals (people with low ACC volume are more susceptible to PTSD) which will then trigger the striatum to release a reward (dopamine). Therefore, the more we feel or sense the more likely we are to assign value to these experiences, the more pleasure we experience. This is trait optimism (people who “believe that they will experience positive outcomes in life but also maintain a vigorous pursuit of goals to ensure positive outcomes”) and neuroticism (behavioural regulation) . So, maybe those people who find it easier to connect to their current senses rather than future states, find it easier to derive some pleasure in ambiguous times. This may explain the positive effect that meditation (albeit a small to moderate one which might be subject to self-selection bias ) and exercise has on managing stress because we are more self-referential. In introspective or Default Mode  states we are looking for 'in-the-moment' stimulation and non-judgemental acceptance of negative emotions – the incorporation of both mindfulness and CBT?

The DMFC is used to distinguish self from others and make sense of others behaviour. This ability to perceive others intentions is called the Theory Of the Mind (TOM). TOM allows us to cooperate, show empathy and think strategically. When situations are ambiguous and we don’t have the answers we start to look to others (this may account for the psychological appeal of authoritarianism during uncertain times) – our own desires are suppressed as we try to become aware of others and start to feel their pain. The DMPFC is also associated with grit the “tendency to strive to achieve long-term goals with continual passion and perseverance” (albeit I am not convinced this is much different from self-efficacy as self-efficacy and grit scales use similar questions and doesn’t grit = growth mindset?) – having long-term goals helps us push through short term ambiguity? This boils down to our definition of stress – our ability to find alternative strategies to preserve our wellbeing. So it is not surprising that many studies have revealed that conscientiousness and neuroticism (from the Big 5 personality) are correlated with the DMPFC. So, the PFC gives us behavioural plasticity, an ability to self-regulate which helps us cope with stress and protect us from attention deficit, anxiety and major depressive disorders.

Acute stress – how much stress is acute stress?

But how much stress do we need to be subject to for us to lose control of our PFC and experience an ‘amygdala hijack’? This simple answer is that it depends on the nature of the stressor and an individual’s behavioural plasticity. Studies that show reduced activation in PFC regions occurs when people watch extremely violent material. Take this study for example. Young (18-25 year olds) female students were asked to watch selected scenes from a commercial movie called Irreversible. I have tried to watch it with my partner, so have a couple of my friends – none of us have made it to the end! It is extreme male to male and female violence. Hopefully something very few of us will experience and certainly not in the context of OCM.

Other tests use a powerful light source directed towards subjects, while an audience gives them negative feedback about their level of achievement which the subject can simultaneously watch on a monitor screen. Again, not representative of OCM. So, it seems to take extreme levels of stress for us to start losing cognitive ability but again it only impairs working memory not verbal explicit memory tasks, or implicit learning tasks that involve neutral or positive stimuli. 

Acute stress - It’s also a bit about timing

It’s not just the degree of stress that can affect our behaviour but its timing. Acute ‘on the spot’ stress (like an interview from hell) increases adrenaline (sharpens mental focus) and dopamine (focusses learning & attention) levels. This enhances decision-making performance for between 5-18 mins after induced stress. As cortisol (regulates metabolism and memory formation) concentrations peak, decision-making performance deteriorates. Again, learning is also affected by stress. Inducing chronic stress using a freezing shock shortly before or after learning can (maybe unsurprisingly!) improve memory. But if learning takes place 60 minutes before or after the stressful event, learning is impaired. This is because our brains ‘hook’ memories on to stressful events. I am more likely to remember a bad change programme than a good one!

Chronic Stress - Irreducible Uncertainty & Black Swans

But imagine if everything was unknown. We have no idea what the bucket contains – irreducible uncertainty. Or maybe events have extremely low probability of occurrence but when they do occur they have a huge impact – a black swan. Like the 2008 crash and the Covid-19 crisis we can no longer anticipate the future, we can only try to reduce the impact of threats we don’t understand.

Irreducible uncertainty is the randomness inherent in any complex system – particularly organisations. This is fortuitous luck (or not) – there is no rhyme or reason. A black swan is maybe something we thought would never happen and completely out of our control. These are the most stressful events both in how we feel (subjective stress) and our body’s reaction (objective stress e.g. pupil dilation). This stress is induced when the environment is utterly unpredictable, like Flash Gordon playing the Wood Beast Game. This is because the context carries little information on how the brain should respond – there is no opportunity to use our learning and no reward. In fact, what you don’t know becomes far more relevant than what you do know. Focusing on the future is of little help regardless of the clarity of our vision. And nothing from our past experiences help us. This type of stress is called chronic stress – stress suffered over a prolonged period – days, weeks or months.

Chronic Stress - A Brain changer

So, what is happening in the brain when it experiences chronic stress? Studies show that chronic stress and high levels of cortisol (stress hormone) will reduce neuroplasticity in the hippocampus and frontal cortex and even shrink them. Expose the hippocampus to chronic stress and it may take a couple of weeks to start shrinking. But the PFC is a sensitive soul. It may only take one stressful event and within a week it will start changing its shape. But, maybe more worryingly, stress on the amygdala increases its plasticity. So, creating stress at work will make people even more responsive (less resilient) to stress & uncertainty in the future. Indeed, people who report higher levels of life stress behave as if they believe that the environment is more uncertain. 

So, differences in responses to stress levels between individuals relates to variability in beliefs about uncertainty. This is what seems to happen with PTSD. PTSD seems to over activate our ACC – “ an all-purpose alarm what signals when ongoing behaviour hits a snag” (Behave p529) and subdue our vmPFC (our ability to “assign personal value to self-related information”). So people with PTSD find it hard to put their stress into context and extinguish their fears.

For those who have experienced chronic psychological stress (e.g. depression or come from an abusive background), seeing sad faces would be the norm. Seeing happy faces will trigger the amygdala because they are novel or unexpected. So, what triggers the amygdala is subjective – it depends on your life’s experiences. Ultimately our fears are only as deep as the mind allows. This is why Cognitive Behavioural Therapy (CBT) can be so effective in treating these cases. CBT not only allows victims to reappraise circumstances but gives them the belief that they can effectively reappraise.

So, chronic stress happens when our sensory signals consistently fail to match our mental models of what should be happening. If we cannot find a strategy that works and get stuck in our ways of thinking our ACC continues to signal incongruence. These constant waves of stress create allostatic (wear and tear on our psychological and physical state) overload.

Bootstrapping beliefs

So how do we discover new strategies when Black Swans and irreducible uncertainty debunk our mental models? When neuroscientist build models to simulate patterns in our stress responses to irreducible uncertainty they find the best predictors are those based on Bayesian learning algorithm. A Bayesian algorithm updates its prior knowledge (or beliefs in humans) based on emerging evidence from sensory data (likelihood that our prior knowledge is true) by forming new (posterior) knowledge (beliefs). In neuroscience this is called the Bayesian brain.  This challenges the traditional idea in neuroscience that information is progressively added to ‘incomplete’ perceived sensory information (a bottom up approach). The Bayesian brain concept flips that idea to suggest that we project our perceptions onto sensory data (a top-down approach) and our brain comes to a compromise on the reality. It maybe that we get a dopamine reward when we get the prediction right (or approximate and good enough) which in turn helps us build an accurate picture of OUR world. As mentioned in Part 1 the PFC only needs a small sample of ‘reality’ to conjure up meaningful entities and we end up seeing faces in funny places. In the Bayesian brain world the accumulation of evidence or information is in fact a strategy to reduce uncertainty.  

Our brains are too clever to be ‘prediction machines’ – they know certainty is an illusion. They are meaning-makers hungry for surprise and novelty which helps build a repertoire of strategies to deal with future events. 

Primed for uncertainty our brains must solve hard problems with little information and no time. So, it is not surprising that they seek stability by harvesting strategies to deal with uncertainty.

But is OCM ahead of the game?

But neuroscientist are just confirming what Ajzen hypothesised in 1974 and refined in 1985 – the Theory of Planned Behaviour (TPB) - probably one of best researched but least used change theories in change management. TPB links beliefs to behaviours. It suggests that we make logical reasoned decisions to engage in new behaviours. We need the belief that we have the ability (a beliefs of what we are capable of called self-efficacy) , motivation (the perceived benefits of the action based on a mental model of what good feels like) and opportunity to practice (the support from your colleagues to demonstrate the new behaviour). So, these initial or ‘prior’ beliefs control whether we attempt a new behavior. Based on this we create an intention (a reasoned decision or likelihood to achieve) and give the new behaviour a go. If we succeed, we form new beliefs (posterior) on our ability, I feel more motivated and I know I have the support to get even better. When I fail, if my prior beliefs in my ability, motivation & support are strong enough I might give it another go. Either way our

Bayesian updating helps our brains to bootstrap our beliefs based on new evidence. 

And as we know our stress response kicks in because our sensory signals consistently fail to match our mental models. ‘I believe therefore I achieve’ and ‘I achieve because I believe’. This iterative process between belief and behavior supports the idea of Bayesian updating and more critically stresses the importance of creating learning organisations for OCM success. Organisations should be building skills incrementally which allows them to quickly update beliefs about what they think they are capable of but also builds resilience because they learn how to affect their environment.

So, irreducible uncertainty and Black Swans disrupt our meaning frameworks or mental models. We have to regain meaning by challenging our assumptions (prior beliefs) and disregard outdated mental (business?) models. The more we can habituate questioning our assumptions the better we will be at adapting to stressful situations. We need to reject the familiar and seek the unfamiliar through surprise and novelty. This concept is called “predictive coding” whereby our brain becomes less sensitive to predictable redundant prior experience and more alert to unpredicted ones. People with autism are constantly bombarded by ‘surprises’ because they find it harder to encode prior experiences – hence they see the world more intensely and sometimes more accurately because their brain lacks cognitive bias or heuristics (rules of thumb) to guide them. 

Summary

So, there is little evidence to suggest that people go into a ‘flight-or-flight’ or ‘amyglada hyjacks’ due to psychological stress. In fact, I have not read a single definition of ‘fight-or-flight’ or ‘amyglada hyjacks’. We have known for a while there is no “single synchronised system” that links physiological, cognitive, emotional and behavioural responses to stress.

The evidence suggests that our response to stress is highly nuanced. When we are bored or subject to acute stress (interview from hell, graphic violence, noise stress, traumatic memories etc) there is reduced activation of the PFC, working memory and less goal directed behaviour. Resources seem to move towards the default mode network associated with ‘mind wandering’ and introspection. If we ‘cannot jump start’ our PFC by challenging our ways of thinking we get caught in habitual ruts. If this persists our stress levels turn toxic escalating into chronic stress and allostatic overload which manifests itself in a variety of maladaptive physical and psychological ways such as anxiety and depression. And even when we experience ‘tolerable’ stress levels (predictable uncertainty) we still need to be able to adjust our expectations and mental models to reduce levels of uncertainty.

So rather than using ‘flight-or-flight’ or ‘amyglada hyjacks’ language a more accurate and simple way to think about stress is our ability to find a strategy to preserve our future wellbeing. For organisations and change practitioners this means habituating change by constantly challenging prior beliefs (myths, assumptions, biases and heuristics) using an evidence-based approach and maybe #democratizingwork.

This gets to the root of organisational learning - understanding how it learns and how new behaviours are adopted. Maybe this is adopting triple loop learning where we take a more holsitic approach to thinking about what work really means?

Progress is impossible without change, and those who cannot change their minds cannot change anything. – George Berard Shaw

In Part 3 of this series I will be diving deeper into the concept of the 'bayesian brain' and its relationships with the theory of planned behaviour as a starting point for a new evidence-based and intergrated approach to OCM.