The science of speaking anxiety
Speaking anxiety is one of the most common performance fears among professionals. It’s also one of the most misunderstood — not because the science is unclear, but because most of what gets written about it ignores the science entirely.
What follows is what the research actually shows about why speaking anxiety occurs, why it persists, and what it takes to change it.
The brain detects threat before you know it
When you walk into a meeting room, stand up to present, or hear your name called in a group — something happens in your nervous system before conscious thought has had a chance to weigh in.
Research published in Nature Neuroscience has shown that the amygdala — a small structure deep in the brain responsible for detecting threat — responds to threat-related signals in as little as 45 to 88 milliseconds after exposure. For context, it takes roughly 200 milliseconds to consciously perceive what you’re seeing.¹
The neuroscientist Joseph LeDoux documented two distinct pathways for threat processing. The fast pathway sends sensory information directly from the thalamus to the amygdala — automatic, and bypassing the cortex entirely. The slow pathway takes the longer route through the thinking brain, allowing for rational appraisal. The fast pathway fires first.
By the time your prefrontal cortex has assessed that you’re safe, your body has already responded.²
This is why knowing you’re safe doesn’t stop the reaction. The reaction doesn’t wait for that assessment.
Anxiety is anticipatory, not reactive
A common assumption is that anxiety is a response to something happening. The research suggests otherwise.
Peer-reviewed studies have consistently shown that speaking anxiety is characterised by exaggerated responses to anticipated threats — not threats that are actually present. The nervous system learns to associate specific cues with danger. From that point forward, the cue alone is enough to trigger the defensive response. The original situation doesn’t need to be repeating. The prediction of it is sufficient.³
This is why the physiological response — elevated heart rate, shallow breathing, cognitive disruption — typically begins before the person has spoken a word. Often before they’ve entered the room. The nervous system isn’t responding to what’s happening. It’s responding to what it predicts is about to happen, based on what it has learned.
How the pattern gets encoded
Emotional memories are stored differently from factual memories.
Research in memory neuroscience has established that experiences with significant emotional content — particularly those involving social threat, humiliation, or being suddenly exposed — are encoded rapidly and durably by the amygdala and hippocampus working together. The amygdala signals the hippocampus to consolidate memories of threatening situations so the individual can avoid them in future. For most threats in human evolutionary history, this was lifesaving.
The complication arises when the encoded threat is social rather than physical — being judged, embarrassed, or exposed as inadequate in front of others. The nervous system encodes this with the same urgency. And once encoded, a neutral stimulus — a meeting room, a group of faces, the phrase “let’s go around the table” — can be enough to activate the full defensive response.
The original experience doesn’t need to remain consciously accessible for the pattern to keep firing. The encoding happens at an unconscious level. So does the retrieval.
What speaking anxiety does to performance
When the threat response fires during a speaking situation, the effects on cognitive function are measurable.
Research published across multiple peer-reviewed journals has documented that anxiety disrupts working memory — the system responsible for holding and manipulating information in real time.⁴ In speaking terms, this means words that were available moments before become harder to retrieve. Organised thinking becomes more effortful. The ability to track the room, manage structure, and speak fluently all compete for cognitive resources that are now being directed toward threat monitoring.
One study found that the effortful suppression of anxiety — trying not to show it — itself increased physiological arousal and reduced memory for the speech.⁵ Attempting to manage the anxiety consumes the very resources needed to perform.
This is not a weakness of character or a failure of preparation. It is the predictable consequence of a threat response firing in a context where it was never designed to help.
Why more practice often isn’t enough
The standard advice for speaking anxiety is to get more experience. Sometimes this helps. But research on conditioned fear responses explains precisely why it often doesn’t — and occasionally makes things worse.
Repeated exposure to the feared situation without the underlying threat association being updated produces what researchers call extinction — a new layer of learning that suppresses the original response, but doesn’t erase it. The original memory remains intact. Under sufficient pressure or stress, it can return — a phenomenon researchers call fear reinstatement.
This is why someone can present hundreds of times and still experience the same response. Practice builds skill and familiarity. It doesn’t automatically update an emotional memory.⁶
Memory reconsolidation: the mechanism of lasting change
The most significant development in anxiety neuroscience in the past two decades has been the discovery of memory reconsolidation.
When a long-term memory is retrieved — reactivated — it briefly returns to a malleable state in which it can be updated before restabilising. Research published in Neuroscience Bulletin and multiple peer-reviewed journals has confirmed that fear memories are susceptible to this process.⁷ During this window, the encoded threat association can be modified. The emotional charge attached to the memory can change.
This is fundamentally different from extinction. Where extinction creates a new competing memory, reconsolidation updates the original one. Memories updated through reconsolidation show significantly less tendency to return than those addressed through exposure alone.
What this means in practice
The research points in a consistent direction. Speaking anxiety is not a knowledge gap, a skill deficit, or a confidence problem. It is an encoded threat response operating below conscious awareness, triggered predictively, and resistant to approaches that only work at the conscious level.
Lasting change requires working at the level where the pattern is stored — accessing the encoded association between speaking and threat, creating the conditions for reconsolidation, and updating the original learning.
That is not a metaphor. It is what the neuroscience of fear memory describes as the mechanism of durable change.
References
¹ Rapid processing of invisible fearful faces in the human amygdala. Journal of Neuroscience, 2023. Documented amygdala threat responses beginning at 45–88ms, prior to conscious awareness.
² LeDoux, J.E. The Emotional Brain. Simon & Schuster, 1996. Established the fast subcortical “low road” pathway to the amygdala, confirmed by subsequent neuroimaging research published in Nature Neuroscience, 2016.
³ Anticipatory Threat Responding: Associations with Anxiety, Development, and Brain Structure. Biological Psychiatry, 2020. Demonstrated that anxiety involves heightened response to anticipated rather than immediately present threats.
⁴ The complex interaction between anxiety and cognition: insight from spatial and verbal working memory. Frontiers in Human Neuroscience, 2013. Documented anxiety-related disruption to verbal and spatial working memory.
⁵ Attentional Control Buffers the Effect of Public Speaking Anxiety on Performance. PMC, 2012. Found that suppression of anxiety increased physiological arousal and reduced speech memory.
⁶ Retrieval and Reconsolidation Accounts of Fear Extinction. PMC, 2016. Reviewed evidence that extinction suppresses but does not erase the original fear memory, accounting for fear reinstatement.
⁷ How Fear Memory is Updated: From Reconsolidation to Extinction? Neuroscience Bulletin, 2025. Reviewed the reconsolidation process and its implications for updating fear memories at their source.
