Published by Unseen Progress, an independent publisher of caregiver research. Last reviewed 2026-05-10. Part of the child anxiety research overview.
Short answer. Exposure hierarchies — ranked lists of feared situations from easiest to hardest — are the structural backbone of evidence-based anxiety and OCD treatment (March & Mulle, 1998; Kendall & Hedtke, 2006). They work best when the clinician leads the design, the child is engaged in ranking, the ladder starts at a tolerable rung, and the hierarchy is reviewed weekly. Parent-built hierarchies done in isolation often fail by starting too high, being too vague, or skipping rungs the child still needs.
The exposure hierarchy is the core operational tool in ERP for OCD (March & Mulle, 1998; Foa et al., 2005) and graduated CBT for anxiety disorders (Kendall & Hedtke, 2006). It functions as a roadmap: the child cannot face the top of the feared list cold, but can usually face a small, tolerable version of it, and that success builds the willingness for the next rung.
The empirical evidence is robust. The POTS trial (POTS Team, 2004) and CAMS (Walkup et al., 2008) both used manualised hierarchy-based exposure as their primary intervention. Inhibitory-learning research (Craske et al., 2008) has refined the design principles — variability and novelty across rungs improves long-term retention more than rigid SUDS-based ordering — but the hierarchy itself remains central.
The clinical literature is also clear about a critical limit: parents should not run a hierarchy alone. A hierarchy built without clinician input usually contains design errors — rungs that are too high, vague rungs that can't be measured, missing rungs in the middle, or rungs that inadvertently include safety behaviours that block habituation.
When parents ask "how do I build a hierarchy at home," they usually mean one of three things:
1. The clinician has assigned exposure homework and the parent is the one supervising it. This is the most common case — the hierarchy was clinician-built, and the parent's role is delivery. 2. The family is waiting for therapy to start or is between courses. The parent wants to keep momentum and is looking for a way to structure home practice. 3. The therapy has plateaued. The clinician's hierarchy seems stale, and the parent is trying to figure out what's missing.
Each of these has a different research-backed answer.
The first move is not to rank fears — it is to name them. A child afraid of dogs has different sub-fears (small dogs, large dogs, dogs off leash, dogs that bark, the idea of a dog approaching). Each sub-fear is a candidate rung. Kendall & Hedtke (2006) recommend a structured conversation that surfaces 10–20 candidate situations before ranking begins.
The child rates each candidate situation on a 0–10 SUDS scale: 0 is no anxiety, 10 is the worst they can imagine. Younger children can use a 1–5 "fear thermometer" with faces. The number is the child's, not the parent's — parental estimates of child fear are usually inaccurate (Comer & Kendall, 2004).
The most common design failure is starting too high. The bottom rung should be a situation the child rates SUDS 3–4 — enough anxiety to be a real exposure, not so much that the child cannot enter. Foa et al. (2005) and the inhibitory-learning literature emphasise that the first successful exposure is disproportionately important to the rest of the course.
Adjacent rungs should usually differ by 1–2 SUDS, not 4–5. A hierarchy that jumps from SUDS 4 to SUDS 8 with nothing in between usually stalls at the gap. The middle rungs are where most of the treatment time is spent, and they need to be granular enough to make progress weekly.
A subtle compulsion-like behaviour — carrying a phone, having a parent in the room, silently counting, holding a particular object — can reduce in-the-moment anxiety and block habituation. Hierarchies that look complete but include hidden safety behaviours produce flat SUDS curves and treatment plateaus. The clinician's role here is essential; parents alone often miss them.
A hierarchy is a working document, not a fixed plan. As rungs are mastered, new ones are added; rungs that turn out to be too easy or too hard are revised. The weekly review with the clinician is part of the treatment, not an add-on.
From recent parent threads:
The third quote captures the most common design mistake — the intuition that "facing the worst fear" is the goal. The research-backed answer is the opposite: a graded ladder produces durable change; cold-turkey exposure to the top of the hierarchy usually fails.
1. Don't build a hierarchy alone. The single most important step is to work with a trained clinician on design. Parents who try to build hierarchies from books usually produce ones with design flaws that limit progress. 2. Be a careful delivery partner, not the architect. Once the hierarchy exists, the parent's role at home is structured, warm, consistent delivery of the agreed exposures and response prevention. 3. Use the child's SUDS, not your own. Parental estimates of child fear are unreliable; the child's ratings are the operational data. 4. Track SUDS across sessions, not just within them. Starting SUDS dropping across weeks is the strongest signal of consolidation. 5. Flag plateaus early. A two-week stall on the same rung is not a sign to push harder — it is a sign to look for a hidden safety behaviour or compulsion.
---
Unseen Progress publishes long-form caregiver research and builds research-backed daily trackers for the families covered. See the full child anxiety research overview for the complete framework.