Sciam. How uncertainty breaks the brain

 Deep dive: How Uncertainty Breaks the Brain

Maia Szalavitz · Scientific American · Neuroscience, psychology and the science of resilience

The most important idea is this: the brain doesn’t just react to what happens. It continually tries to anticipate what will happen, prepare for it, and make the world feel understandable. When life becomes persistently unpredictable, that system can become strained.

Szalavitz’s article explores how uncertainty affects stress, energy, learning and mental health, drawing on research into stress physiology, predictive processing and anxiety. APS +1

One qualification: the following is a deep dive into the article’s central themes and the underlying science, rather than a claim to reproduce every detail of the full article.

1. The brain is a prediction machine

Your brain is not a passive camera recording reality. It combines sensory information, memories, learned patterns and expectations to construct its best current interpretation of the world.

For example, when someone you care about usually replies to your messages promptly, you develop an expectation about their behaviour. If they suddenly stop replying, your brain starts generating possibilities:

• Have they become upset with me?
• Has something happened to them?
• Are they simply busy?
• Have I misunderstood the relationship?

Your brain is attempting to resolve a gap between what it expected and what it can currently explain.

This is related to the concept of predictive processing: the idea that perception and learning involve ongoing interaction between prior expectations and incoming evidence. It is an influential framework in neuroscience, although not every aspect of brain function is explained by one universally accepted predictive model. PubMed +1

The deeper point: we don’t merely want good outcomes. We also want to know what kind of world we are living in.

A disappointing but predictable situation can sometimes be easier to cope with than a situation that might turn out wonderfully or terribly and gives us no reliable clues.

2. Not all uncertainty is the same

Neuroscience distinguishes between different kinds of uncertainty. This matters because each calls for a different response. Nature Reviews Neuroscience +1

A. Randomness within a stable system

You know the rules, but the result varies. A fair coin can land either heads or tails. You can estimate the probabilities even though you cannot predict the next toss.

Helpful response: learn the odds, accept variation and avoid treating every outcome as meaningful.

B. A changing environment

The rules themselves change. Your job becomes insecure, a relationship shifts, or a previously reliable person becomes unpredictable.

Helpful response: update your expectations and remain flexible rather than relying on yesterday’s rules.

C. Not knowing what something means

Someone’s silence might mean anger, distraction, sadness or nothing in particular. You lack enough information to distinguish the explanations.

Helpful response: recognise competing explanations without prematurely committing to one.

Research on learning under uncertainty distinguishes stable variability from unexpected environmental change because the brain needs to decide both what to learn and how quickly to revise what it believes. Nature Reviews Neuroscience +1

This distinction has a practical consequence: you cannot solve every uncertainty by gathering more information. Sometimes the missing information exists. Sometimes the future genuinely has not been determined.

3. Why uncertainty can be more exhausting than bad news

Imagine two scenarios.

• You know that a difficult conversation with your manager will happen tomorrow.
• You suspect that your manager may be unhappy with you, but you don’t know why, when you will hear from them, or what the consequences might be.

The first scenario may be unpleasant, but you can prepare. In the second, your mind may repeatedly simulate possible outcomes.

Psychologist and neuroscientist Lisa Feldman Barrett, discussed in the article, explains that unresolved uncertainty can require the brain to maintain multiple possible plans and interpretations. This has a metabolic and cognitive cost. APS

That helps explain why uncertainty can feel like:

• Mental fatigue despite having done little physical work.
• Difficulty concentrating on the task in front of you.
• Repeatedly checking messages or looking for reassurance.
• Feeling unable to start because you don’t know which outcome to prepare for.
• Losing interest or motivation when the future feels impossible to navigate.

The brain is not necessarily working harder on one clear problem. It may be trying to keep several incompatible possibilities alive at once.

The key insight: unresolved uncertainty can consume resources even when nothing visibly happens.

4. Chronic uncertainty can become a biological stressor

Stress is not simply a feeling. It involves coordinated changes in the nervous system, hormones, attention and energy use.

When the brain perceives a threat, stress responses help prepare the body to act. Glucocorticoids, including cortisol in humans, play an important role in mobilising energy and regulating the stress response. In the short term, this can be protective. When stress is prolonged or repeatedly activated without adequate recovery, the cumulative effects can become harmful. APS +1

The problem with persistent uncertainty is that the threat may never be clearly resolved.

Consider someone waiting for news about a loved one’s health. There may be no immediate action to take, but the person cannot confidently conclude that everything is fine. They may repeatedly check their phone, imagine worst-case scenarios and struggle to concentrate.

The body can be at rest while the mind remains on alert.

The deeper mechanism

• Attention narrows: potential threats become more salient than neutral or reassuring information.
• Planning becomes harder: several possible futures compete for mental resources.
• Recovery becomes difficult: the mind keeps returning to the unresolved problem.
• Mood can deteriorate: exhaustion and negative expectations may reinforce each other.

These are possible responses, not inevitable outcomes. People vary considerably in how they respond to uncertainty, and not every period of stress produces lasting harm.

The critical distinction is between a stressful event and a state of prolonged uncertainty in which the person cannot find a reliable way to respond.

5. Why the brain sometimes prefers a bad certainty to a good uncertainty

This is one of the most revealing psychological implications.

Imagine that you are in a relationship and your partner’s behaviour has become inconsistent. Sometimes they are affectionate; at other times they are distant.

You may find yourself asking:

• Are they losing interest?
• Have I done something wrong?
• Should I confront them?
• Should I leave before I get hurt?

A definitive answer might be painful, but it would give you something to work with. Uncertainty keeps all these possibilities active.

This can lead to an uncomfortable paradox: we may cling to a negative explanation because it feels more manageable than having no explanation at all.

For instance, believing that someone dislikes you may seem easier than remaining unsure of how they feel. The belief creates a coherent story, even if the evidence is weak.

This does not mean that the brain always prefers bad news. It means that the desire for predictability can sometimes outweigh the desire for accuracy.

The practical lesson is to distinguish between two questions:

• What explanation would make this situation feel settled?
• What explanation is actually best supported by the evidence?

They are not necessarily the same.

6. Anxiety can become a self-reinforcing prediction loop

1. A threat is anticipated

“Something bad might happen.”

2. Anxiety rises

The possibility feels increasingly urgent and believable.

3. Avoidance brings relief

The person escapes the situation that feels threatening.

4. The prediction remains untested

“I was safe because I avoided it.”

The fear returns, often with its original credibility intact.

The Association for Psychological Science’s coverage of Szalavitz’s article discusses this mechanism using the example of a person who fears dogs after being bitten. Avoidance prevents the person from discovering that many dogs are safe. APS

This is important because avoidance works in the short term. It reduces distress, so the brain learns that avoiding the situation is useful.

But short-term relief and long-term learning are different things.

What helps interrupt the loop?

In appropriate circumstances, gradual exposure allows a person to encounter a feared but manageable situation without performing the usual avoidance behaviour. They can then gather new evidence about what happens.

For example, someone afraid of making mistakes might begin by expressing a minor disagreement, making a low-stakes decision without repeatedly checking it, or asking a question in a meeting.

The aim is not to prove that nothing bad can ever happen. It is to learn that the feared outcome is not inevitable and that uncertainty can be tolerated.

Exposure-based approaches have established clinical applications, particularly for anxiety disorders. The right approach depends on the fear and its severity; traumatic experiences and complex symptoms may require professional support.

7. Trauma: why some people recover and others do not

The article also addresses an important distinction between the immediate aftermath of trauma and longer-term psychological difficulties.

Following a frightening or life-threatening experience, people may have intrusive memories, nightmares, heightened alertness or difficulty making sense of what happened.

These reactions can be part of the brain’s attempt to process an overwhelming event. Their presence immediately after trauma does not, by itself, establish that someone will develop post-traumatic stress disorder (PTSD). APS

Why do outcomes differ?

There is no single explanation. Recovery can be influenced by the severity and duration of the experience, previous trauma, social support, ongoing threats, individual differences and access to appropriate care.

One useful way of understanding the difference is this:

• In recovery, the experience gradually becomes something the person can remember without continually experiencing it as a present threat.
• In persistent post-traumatic difficulties, reminders may continue to trigger intense threat responses, avoidance or a sense that the danger is still happening.

This is a simplification, not a complete account of PTSD. Trauma responses vary widely, and recovery is not always linear.

An important implication: resilience does not mean that an experience fails to affect you. It means that, over time, you can regain the ability to respond flexibly to life rather than remaining governed by the expectation of danger.

8. Why some people adapt better to uncertainty

People differ in how readily they revise their expectations.

Suppose you have worked in the same organisation for 20 years. You know the people, the systems and the informal rules. Suddenly, the organisation changes its leadership, restructures departments and introduces unfamiliar technology.

Your existing knowledge is not necessarily useless. But it may no longer be a reliable guide to every new situation.

One response is to double down on familiar assumptions. Another is to recognise that the environment has changed and that some previously reliable rules need revising.

Research into adaptive learning distinguishes between uncertainty that reflects normal variation and uncertainty that signals a change in the environment. Effective learning depends partly on recognising which kind of situation you are facing. Nature Reviews Neuroscience +1

This suggests three useful capacities:

• Flexibility: being willing to revise an interpretation when new evidence appears.
• Calibration: knowing how confident to be, given the available evidence.
• Tolerance: being able to function while an answer remains unavailable.

These capacities are related, but distinct. You might be good at changing your mind but still feel intensely anxious while doing so. Or you might tolerate ambiguity well but take too long to recognise that circumstances have genuinely changed.

The aim is not to become endlessly adaptable at any cost. Sometimes uncertainty signals a real problem that needs action. The skill is knowing when to adapt, when to investigate and when to accept that the answer is not yet available.

9. The energy problem: why exhaustion can make the world look worse

A particularly subtle point in the article concerns the relationship between mental energy and perception.

When people are under sustained strain, they may have less capacity for concentration, planning and reconsidering their assumptions. Barrett suggests that this can contribute to a bleak interpretation of circumstances, even when conditions are beginning to improve. APS

Imagine that you have spent months worrying about job security. Eventually, the uncertainty starts to resolve: your manager offers reassurance, your workload stabilises, and the immediate threat diminishes.

Yet you still feel pessimistic.

That does not necessarily mean you are irrational or failing to appreciate the improvement. Your expectations, habits of attention and stress responses may take time to adjust.

There is an important caution here: lack of pleasure, motivation or energy can have many causes, including depression, sleep problems and physical illness. They should not automatically be attributed to uncertainty or mental exhaustion.

A useful practical distinction

Ask yourself:

• Has the situation actually become worse?
• Has it improved, but I have not yet adjusted to the change?
• Am I so exhausted that I cannot assess it clearly right now?

Sometimes the most useful first step is not to solve the entire problem. It is to restore enough energy and perspective to assess the problem accurately.

10. Can we train ourselves to tolerate uncertainty?

The answer is broadly yes, although not by eliminating the brain’s desire to predict.

The goal is to become less dependent on certainty before taking action.

Here are six practical approaches that follow from the research discussed above. They are evidence-informed applications, not a claim that Szalavitz prescribes this exact six-step programme.

1. Separate what you know from what you fear

Write down the observable facts, your interpretation of them and what remains unknown.

Example: “They haven’t replied for six hours” is a fact. “They no longer care about me” is an interpretation.

This creates space between evidence and prediction.

2. Distinguish solvable uncertainty from unavoidable uncertainty

If you lack important information, identify one reasonable action that could obtain it. If the outcome depends on factors outside your control, recognise that further rumination may not produce more knowledge.

Ask: “Is there useful information I can obtain, or am I trying to think my way into knowing the unknowable?”

3. Practise small doses of uncertainty

Make a low-stakes decision without excessive checking. Allow a message to remain unanswered for a reasonable period. Try a new activity without needing to know in advance whether you will be good at it.

The point is to experience uncertainty and discover that you can continue functioning while it exists.

4. Generate alternative explanations

When your mind settles on one frightening interpretation, identify two or three plausible alternatives.

You do not need to convince yourself that the most positive explanation is true. You simply need to acknowledge that the evidence may not justify your certainty.

5. Protect your capacity to think

Sleep, physical activity, regular meals, restorative breaks and supportive relationships can help sustain your ability to cope with stress.

These do not remove external uncertainty. They can help preserve the resources needed to respond to it.

6. Decide what remains under your control

You may not control another person’s decision, an organisation’s restructuring or the timing of a medical result. You can often control your next question, your preparation, your boundaries and the support you seek.

This shifts attention from predicting the entire future to taking the next useful step.

11. The philosophical question: is the need for certainty itself part of the problem?

There is a deeper tension here.

Predictability is useful. Without it, learning, planning and coordinated social life would be much harder. We need expectations to navigate the world.

But expectations become problematic when we treat them as guarantees.

We may assume that if we behave well, people will treat us fairly; if we prepare thoroughly, we will succeed; if we love someone, they will stay; or if we make the right decision, we will avoid regret.

These expectations are understandable. They help us organise our lives. But they can also turn ordinary unpredictability into a perceived failure of the world to operate as it should.

A more flexible position is to hold expectations without making our wellbeing entirely dependent on their fulfilment.

You can plan for success while accepting that success is not guaranteed. You can trust someone while acknowledging that you cannot control their behaviour. You can make a careful decision without requiring it to be perfect.

Psychological flexibility is not the absence of expectations. It is the ability to revise them without losing your footing.

12. What the science does—and does not—tell us

It is worth separating robust findings from the broader explanatory framework.

Claim	How to interpret it
People learn by comparing outcomes with expectations.	Well-established in learning research.
Different kinds of uncertainty call for different learning strategies.	Supported by computational and experimental research.
Prolonged stress can affect brain and bodily functioning.	Well-established, although effects depend on context and duration.
Uncertainty can contribute to anxiety and avoidance.	Supported by research, but individual responses vary.
Predictive processing explains all mental activity.	Too strong. It is an influential framework, not a complete, universally accepted explanation.
Learning to tolerate uncertainty eliminates anxiety.	Not supported as a general promise. It may help, but anxiety has multiple causes.

Research reviews provide a useful foundation for these distinctions. Nature Reviews Neuroscience +2

The danger in popular neuroscience is making every human experience sound like the consequence of one brain mechanism. Uncertainty is important, but so are material conditions, relationships, past experiences, physical health and social support.

It would also be misleading to suggest that people can simply train themselves out of distress caused by war, poverty, illness, insecure housing or other serious external threats. Sometimes the environment needs to change; resilience alone is not an adequate solution.

13. The deepest takeaway

The article’s central question is not whether we can make the world predictable. We cannot.

It is whether we can develop a healthier relationship with the gap between what we expect and what actually happens.

That means learning three things:

• I can feel uncertain without assuming something is wrong.
• I can experience fear without treating the fear as proof that danger is imminent.
• I can act sensibly without knowing how everything will turn out.

The aim is not to stop predicting. It is to stop demanding that every prediction be confirmed before we feel safe enough to live.

A concise way to express the idea is this:

Resilience is not knowing what will happen. It is developing confidence in your ability to respond when it does.

That confidence is not a guarantee that everything will work out. It is the recognition that uncertainty, discomfort and even disappointment are experiences you can learn to navigate.

Further reading

• Adaptive learning under expected and unexpected uncertainty — Nature Reviews Neuroscience⁠. Explores how the brain learns from ordinary variability versus unexpected change.
• Understanding learning through uncertainty and bias — Communications Psychology⁠. Explains prediction errors, learning rates and why our updating of beliefs can be biased.
• How Uncertainty Breaks the Brain — Scientific American⁠. The original article may require a subscription.