Most people think of avoidance as a targeted response to a specific threat. You avoid the thing that made you anxious. The thing stays small. The research shows the opposite: avoidance is inherently expansionary. The mechanism that makes a single avoided situation feel manageable is the same mechanism that, sustained over months and years, narrows the behavioural repertoire until the accumulated avoided territory constitutes the effective boundaries of a person’s life.

How avoidance of one thing automatically spreads to related things

Stimulus generalisation is the learning mechanism through which conditioned responses to one stimulus spread automatically to perceptually or conceptually similar stimuli — including stimuli the person has never directly encountered. Dymond and colleagues demonstrated that avoidance established for one stimulus in a relational network automatically produces avoidance of all indirectly related stimuli in that network. A traumatic ferry accident can produce avoidance of ferries, then boats, then harbours, then water, then maps showing coastlines — not through additional dangerous experiences at each stage but through the symbolic relational network that connects them as members of the same conceptual category.

Vandael and colleagues’ 2025 controlled experiment confirmed that avoidance behaviour specifically increases generalisation. Participants who avoided a conditioned stimulus were significantly more likely to identify related stimuli as threatening and show higher threat expectancy for them during generalisation testing than those who did not avoid. The proposed mechanism is perceptual memory blurring: relentless avoidance of the original stimulus prevents the formation of a precise perceptual memory of it, making it harder to distinguish from related stimuli, which are then treated as equivalently threatening. The walls are not built by additional experiences of danger. They are built by the expanding shadow of the original avoided stimulus across a widening category of related experience.

The accumulation of what avoidance takes from you

The clinical and behavioural literature frames the consequence of sustained avoidance as behavioural repertoire narrowing — the progressive reduction in the range of actions, contexts, and experiences the person is able to engage with. Excessive fear and avoidance in relatively safe situations leads to a narrowing of one’s behavioural repertoire and less engagement with valued aspects of living, reaching clinical levels across anxiety disorders, PTSD, and OCD. The narrowing is not linear. Each avoided domain expands through generalisation, generating new avoidance, which further narrows the repertoire.

The ACT framework operationalises this as a measurable discrepancy between what a person reports valuing and how consistently they are living in those valued domains. The Valued Living Questionnaire quantifies this gap across ten life domains: career, relationships, personal growth, leisure, and others. High avoidance consistently predicts large discrepancies between importance and consistency ratings — between how much something matters and how much someone is living in that domain. Research using this measure found that the discrepancy is directly predicted by experiential avoidance scores across clinical and non-clinical populations.

The psychological prison is not metaphorical in this framework. It is a measurable gap across life domains between the person the individual describes wanting to be and the person their avoidance is allowing them to be. The person who declines a networking event, avoids a difficult conversation, and postpones seeking a promotion experiences each avoidance as a locally rational decision producing immediate relief. Aggregated across months and years, these individually justified decisions produce a portrait of a life progressively separated from what was explicitly said to matter.

When avoidance becomes habit and stops requiring conscious maintenance

Research reviewed by Pittig, Treanor, LeBeau and Craske identified a critical transition in sustained avoidance that explains why the prison walls become progressively harder to dismantle. Initial avoidance is goal-directed and flexible — the person consciously decides to avoid the feared stimulus. With repetition, avoidance becomes habitual: automatic, stimulus-triggered, and insensitive to the changed value of the outcome. A person who has avoided social events for years will avoid automatically when invited, without engaging the deliberate evaluative process that would allow them to notice that the anxiety has reduced, that the feared outcome has not historically occurred, or that the cost of continued avoidance now exceeds the cost of attendance.

The transition from goal-directed to habitual avoidance is the mechanism that makes the prison self-perpetuating without requiring ongoing conscious maintenance. The person does not keep deciding to stay in the prison. Habitual avoidance produces the same behaviour automatically. And because habitual behaviour is neurologically less sensitive to outcome value, showing the person that avoidance is costly does not reliably produce behaviour change. The habit must be interrupted structurally — through new behaviour in avoided contexts — rather than through insight alone.

The five pathways through which avoidance proliferates

Pittig, Wong, Glück and Vervliet identified five distinct pathways through which avoidance expands beyond its original target: anxious appraisal of situations as more threatening than evidence warrants; avoidance despite accurate threat appraisal when immediate relief is sufficiently reinforcing; expansion through stimulus generalisation; social transmission through observation and instruction; and habituation. The social transmission pathway is particularly significant: Cameron, Schlund and Dymond demonstrated that avoidance generalises through observational learning and instruction without any direct conditioning experience. Some of the prison walls were not built through personal experience of threat. They were inherited — through social observation, parental communication, and cultural instruction — before the person had any direct experience to confirm or disconfirm them.

What dismantles the prison

Research by Bennett and colleagues demonstrated that the most effective intervention for generalised avoidance was not extinction of the original fear but reinforcement of competing behaviours in the generalised contexts. Participants who learned to approach rather than avoid in novel contexts showed reduced avoidance generalisation, and this reduction transferred to new contexts not included in the training. The mechanism that reverses the prison-construction process is not the removal of fear — it is the building of new behavioural repertoire in avoided domains, which gradually displaces avoidance as the default response and expands the accessible space of the functional world.

This is the inverse of the accumulation mechanism. Just as each avoidance episode narrows the repertoire incrementally, each new approach behaviour in a previously avoided domain expands it — and the expansion, like the narrowing, generalises outward through the symbolic relational network.

Book worth reading on this

Get Out of Your Mind and Into Your Life by Steven Hayes and Spencer Smith is the most accessible ACT-based treatment of the valued living gap — Hayes’s account of how the attempt to avoid psychological pain produces the progressive life constriction this article documents. Hayes writes with unusual directness about the mechanism through which avoidance eventually occupies the space that valued living was supposed to fill, and his framework for identifying which domains avoidance has colonised and what re-engaging with them requires is the most practically structured available treatment of the prison metaphor. For any entrepreneur who recognises the pattern of narrowing — domains of professional or personal life that have quietly contracted around what is comfortable rather than what matters — this book provides both the diagnosis and the most evidence-consistent account of what reverses it.

If the dynamics described here are significantly affecting your wellbeing, speaking with a psychologist is the right next step. UK: Samaritans (116 123, free, 24/7). Mind (0300 123 3393). BACP: bacp.co.uk/search/Therapists. Crisis Text Line — text HOME to 741741 (US, UK, Canada, Ireland). International: internationaltherapistdirectory.com.

This article is for educational and informational purposes only. Sources: Dymond, S., Schlund, M.W., Roche, B., De Houwer, J. & Freegard, G.P. (2012), Safe from Harm: Learned, Instructed, and Symbolic Generalization Pathways of Human Threat-Avoidance, PLOS One, 7(10), e47539. Vandael, C., Zenses, A.K., Baeyens, F. & Beckers, T. (2025), The Effect of Avoidance Behavior on Generalisation of Threat-Expectancy, Collabra: Psychology, 11(1). Smith, G.S., Dymond, S. et al. (2020), Relapse of Anxiety-Related Fear and Avoidance, Journal of the Experimental Analysis of Behavior, 113(1), 1–17. Wilson, K.G., Sandoz, E.K., Kitchens, J. & Roberts, M. (2010), The Valued Living Questionnaire, The Psychological Record, 60(2), 249–272. Pittig, A., Treanor, M., LeBeau, R.T. & Craske, M.G. (2018), The Role of Associative Fear and Avoidance Learning in Anxiety Disorders, Neuroscience & Biobehavioral Reviews, 88, 117–140. Pittig, A., Wong, A.H., Glück, V.M. & Vervliet, B. (2017), Pathways towards the Proliferation of Avoidance in Anxiety, Behaviour Research and Therapy, 96, 3–13. Cameron, G., Schlund, M.W. & Dymond, S. (2015), Generalisation of Socially Transmitted and Instructed Avoidance, Frontiers in Behavioural Neuroscience, 9, 159. Hayes, S.C., Strosahl, K.D. & Wilson, K.G. (2012), Acceptance and Commitment Therapy, Guilford Press. Hayes, S.C. & Smith, S. (2005), Get Out of Your Mind and Into Your Life, New Harbinger. Harris, R. (2009), The Confidence Gap, Penguin.