THOMAS BORSATO, PHD

How we learn fear and safety

Conditioning, avoidance, extinction and freedom of choice
An important part of what we do every day was learned without us realising it.
The brain continuously observes what happens and builds predictions:
what is safe?
what is dangerous?
what is worth doing?
what helps me feel better?
This process begins very early and is one of the foundations through which we learn to adapt to our environment.

Learning means predicting
Conditioning is not only about famous laboratory experiments. It is a much broader principle.
If a sound repeatedly precedes something unpleasant, that sound alone may begin to trigger an alarm response.
But we also learn actions.
If I move away from a situation when I feel anxious and immediately feel better, the brain registers a very simple relationship:
anxiety → moving away → relief.
Next time, I will be more likely to use that strategy again.
This process is called negative reinforcement: a response becomes more likely because it reduces an unpleasant state.

Early experiences help build the map
During childhood, the brain must quickly learn to interpret its environment.
A facial expression, a voice, a distance or a bodily sensation can gradually acquire meaning.
Developmental research shows that the systems learning fear and safety change as we grow and involve circuits including the amygdala, hippocampus and prefrontal cortex. Early adverse experiences can influence how these systems distinguish threat from safety signals.
This does not mean that childhood experiences rigidly determine who a person will become.
It means they help build initial predictions about the world.
For example:
“When I express what I feel, conflict increases.”
“When I show vulnerability, I am at risk.”
“When I control everything, I feel safer.”
“When I move away, my anxiety decreases.”
In a particular environment, these responses may have been extremely useful.

Why does a response continue when it is no longer needed?
This brings us to one of the most interesting aspects of learning.
Imagine someone learned to avoid a situation because it was dangerous in the past.
Years later, the context changes. But whenever a similar cue appears, the person continues to avoid it.
As soon as they avoid it, they feel relief.
That relief tells the brain: “you did the right thing: keep doing it.”
The strategy therefore remains active precisely because it works in the short term.
Avoidance also prevents new information from being gathered.
If I always escape, I cannot discover that perhaps today I could stay, and that what I fear might not happen.
Human learning experiments show that safety behaviours can indeed protect a threat belief from extinction.
Many patterns of anxiety can also be understood through this logic.
They are not necessarily “senseless” responses. They are responses that continue to be reinforced.

Extinction does not mean erasure
When a person encounters a stimulus previously associated with danger and repeatedly experiences a different outcome, a process called extinction may develop.
But extinction does not simply erase the old learning.
The neuroscience of learning indicates that a new memory is formed, competing with the previous one.
We might represent it like this:
old learning:
“X means danger.”
new learning:
“X can be present without what I fear necessarily happening.”
Both memories can continue to exist.
This explains why a fear that seemed to have been overcome may occasionally return, for example after a long time, in a different context or following a new stressful event.
It does not necessarily mean that treatment has failed.
It means that the safety memory needs to be retrievable when it is needed.

What happens in PTSD
In post-traumatic stress disorder, this process is particularly important.
A classic neuroimaging study showed that people with PTSD could acquire extinction during learning, but had greater difficulty retrieving that extinction memory the following day.
This difficulty was associated with differences in the functioning of the amygdala, hippocampus and ventromedial prefrontal cortex.
One possible interpretation is therefore that PTSD is not simply a matter of having “too much fear”.
The system may have difficulty using the following information at the appropriate moment: “now it is different; now I am safe.”

Psychotherapy as new learning
This perspective has also changed how exposure therapies are understood.
In contemporary approaches, the aim is not simply to wait for anxiety to decrease during a session.
The goal is to create opportunities in which a prediction of threat can meet an experience different from the one expected.
This is the principle of inhibitory learning proposed in modern models of exposure therapy.
The person does not necessarily need to learn: “I will no longer feel anxious.”
They may learn something much more useful: “I can feel anxious and discover that I do not have to do what I have always done.”

EMDR, CBT and meditation: different paths
CBT, EMDR and meditation are not the same and should not be confused.
However, they may intersect around certain principles of learning.
In CBT with exposure, the person intentionally comes into contact with what they fear and has the opportunity to experience new outcomes.
In EMDR, traumatic memory is reactivated while the person also maintains contact with the present. A study in people with PTSD observed improved extinction learning and changes in brain fear circuits after EMDR.
Meditation follows a different path. It is not exposure therapy.
It may, however, exercise the ability to stay in contact with sensations, emotions and thoughts without immediately turning them into escape or control behaviours.
Two randomised studies have also linked mindfulness training to better retrieval of extinction memories. A randomised trial using 7-Tesla fMRI published in 2026 found improved extinction recall after four weeks of mindfulness, accompanied by lower physiological responses and reduced activity in subcortical threat circuits.
The latter study involved 55 healthy participants: it offers insights into learning mechanisms, not direct evidence of meditation's clinical effectiveness for PTSD.

From automatic behaviour to the possibility of choice
This is where learning and agency begin to meet.
Psychological change does not necessarily require old learning to be erased.
A person may continue to experience anxiety, fear, shame, tension or the urge to avoid.
The difference is that these experiences may gradually lose their power to determine behaviour automatically.
Before: “when I feel X, I must do Y.”
With new learning: “when I feel X, I can recognise what is happening and have more than one response available.”
From this perspective, greater psychological freedom does not mean the absence of difficult experiences.
It means having a wider range of possible actions in their presence.
This is where findings on learning, extinction and safety learning can enter into dialogue with the concept of agency and, as a theoretical hypothesis to be tested, with the Somatic Integration and Agency Model.

Key scientific references
Milad M.R. et al. (2009). Neurobiological basis of failure to recall extinction memory in posttraumatic stress disorder. Biological Psychiatry, 66, 1075–1082.
Lovibond P.F. et al. (2009). Safety behaviours preserve threat beliefs: Protection from extinction of human fear conditioning by an avoidance response. Behaviour Research and Therapy, 47, 716–720.
Craske M.G. et al. (2014). Maximizing exposure therapy: an inhibitory learning approach. Behaviour Research and Therapy, 58, 10–23.
Sevinc G. et al. (2019). Strengthened Hippocampal Circuits Underlie Enhanced Retrieval of Extinguished Fear Memories Following Mindfulness Training. Biological Psychiatry, 86, 693–702.
Rousseau P.F. et al. (2019). Fear extinction learning improvement in PTSD after EMDR therapy: an fMRI study. European Journal of Psychotraumatology, 10, 1568132.
Bouton M.E., Maren S., McNally G.P. (2021). Behavioral and neurobiological mechanisms of Pavlovian and instrumental extinction learning. Physiological Reviews, 101, 611–681.
Björkstrand J. et al. (2026). Effect of app-based mindfulness on extinction recall – a 7T-fMRI study. Scientific Reports, 16, 9957.
Whalen M.L. et al. (2026). A systematic review and meta-analysis of fear conditioning in children who have experienced childhood trauma or maltreatment. Behaviour Research and Therapy. DOI: 10.1016/j.brat.2026.105170.

An initial consultation
Request an appointment →
Thomas Borsato, PhD · Psychologist and psychotherapist
Order of Psychologists of Lombardy, no. 11270
Further reading
Meditation and neuroscience
EMDR and neuroscience
Conditioning, fear and safety
Home
Italiano · Français · English
EMDR Hypnosis Meditation Areas of practice SIA Model NVC Neuroscientific evidence Conditioning
This site was made on Tilda — a website builder that helps to create a website without any code
Create a website