# Terminology Crosswalk — v0.2 Failure-Mode Layer

Status: **working methodological crosswalk**

Purpose: prevent the v0.2 failure-mode vocabulary from presenting established implementation/engagement concepts as novel terminology.

## 1. Exposure / engagement failure

### Established concepts

**Implementation fidelity**
Carroll et al. describe implementation fidelity in terms that include:
- adherence;
- exposure / dose;
- quality of delivery;
- participant responsiveness;
- programme differentiation.

Reference:  
Carroll C, Patterson M, Wood S, Booth A, Rick J, Balain S. (2007). *A conceptual framework for implementation fidelity.* Implementation Science, 2, 40.  
DOI: https://doi.org/10.1186/1748-5908-2-40

**Treatment fidelity**
The NIH Behavior Change Consortium framework distinguishes:
- study design;
- provider training;
- treatment delivery;
- treatment receipt;
- enactment of treatment skills.

Reference:  
Bellg AJ et al. (2004). *Enhancing treatment fidelity in health behavior change studies: best practices and recommendations from the NIH Behavior Change Consortium.* Health Psychology, 23(5), 443–451.  
DOI: https://doi.org/10.1037/0278-6133.23.5.443

**Digital engagement**
Perski et al. conceptualize engagement with digital behavior-change interventions as having both:
- behavioral aspects (extent of use);
- experiential aspects (attention, interest, affective experience).

They also place engagement in a system involving intervention content/delivery, context, mechanisms of action and target behavior.

Reference:  
Perski O, Blandford A, West R, Michie S. (2017). *Conceptualising engagement with digital behaviour change interventions: a systematic review using principles from critical interpretive synthesis.* Translational Behavioral Medicine, 7(2), 254–267.  
DOI: https://doi.org/10.1007/s13142-016-0453-1

### v0.2 implication

Our working phrase **exposure / engagement failure** should not be treated as a new theoretical construct.

Operationally it maps to evidence that the intended intervention was not fully:
- received;
- used;
- attended to;
- responded to;
- or retained over time.

For digital studies this can include:
- app use;
- device wear;
- message reading;
- diary completion;
- response to prompts;
- missing behavioral-sensor data when it reflects uncertain exposure.

Missing outcome data alone should not automatically be interpreted as low intervention engagement.

---

## 2. Mechanism failure despite exposure

### Established distinction

MRC process-evaluation guidance distinguishes **implementation** from **mechanisms of impact** and **context**.

Reference:  
Moore GF et al. (2015). *Process evaluation of complex interventions: Medical Research Council guidance.* BMJ, 350, h1258.  
DOI: https://doi.org/10.1136/bmj.h1258

### v0.2 implication

The working label **mechanism failure despite adequate exposure** means:

> the intervention component appears to have been delivered and meaningfully received, but the hypothesized behavioral pathway or incremental behavioral effect is not supported.

This is an **interpretive coding state**, not a causal proof that a psychological mechanism itself failed.

Preferred wording in future public material:

**no incremental behavioral effect despite documented exposure**

rather than:

**mechanism failed**

unless a study directly measures the mechanism of action.

---

## 3. Behavioral salience

### Closest established concepts

The working term overlaps with:
- experiential engagement;
- participant responsiveness;
- perceived usefulness/relevance;
- attention to intervention content;
- context-sensitive relevance.

Perski et al. explicitly distinguish behavioral and experiential dimensions of digital engagement.

### v0.2 implication

**Behavioral salience** should remain a working bridge construct, not a formal taxonomy category.

Working definition:

> the degree to which support remains noticed, experienced as relevant/useful, and actionable at the current behavioral decision point.

This term is useful for hypothesis generation but should not be described as an established validated construct without further literature work.

---

## 4. Maintenance is not simply prolonged acquisition

Kwasnicka et al. reviewed theoretical explanations of behavior-change maintenance and identified five broad themes:
- maintenance motives;
- self-regulation;
- resources;
- habit;
- environmental and social influences.

They argue that initiation and maintenance have distinguishable theoretical processes.

Reference:  
Kwasnicka D, Dombrowski SU, White M, Sniehotta F. (2016). *Theoretical explanations for maintenance of behaviour change: a systematic review of behaviour theories.* Health Psychology Review, 10(3), 277–296.  
DOI: https://doi.org/10.1080/17437199.2016.1151372

### v0.2 implication

The v0.2 distinction between:
- acquisition;
- continuous active treatment;
- explicit maintenance;
- post-intervention follow-up

is methodologically defensible.

A long intervention should not be called a maintenance intervention solely because it lasts many months.

---

## 5. Preservation rather than gain

### Relationship to maintenance theory

Maintenance concerns continuation of an adopted behavior over time, not necessarily continued escalation of that behavior.

The v0.2 operational category **preserved behavior** is therefore useful when:
- activity remains near the acquired level;
- the relevant comparator declines;
- no further increase occurs.

### Terminology caution

“Preservation rather than gain” is our operational label, not a standardized outcome taxonomy.

Future release wording should state the actual trajectory first, for example:

> activity was broadly maintained while the comparator declined

and only then apply the descriptive category:

> preserved behavior.

---

## 6. Comparator saturation

### Meaning

This is our operational label for trials in which the comparator already contains substantial active behavior-change support, for example:
- self-monitoring;
- feedback;
- goal setting;
- counseling;
- education.

The incremental intervention therefore has limited remaining behavioral contrast.

### Terminology caution

**Comparator saturation** is not presented as an established methodological term.

Safer public wording:

> active-comparator headroom was limited

or:

> the incremental contrast was small because both groups received a substantial behavioral backbone.

---

## 7. Maintenance add-on redundancy

This is also a working descriptive label.

It refers to the specific situation where:
- an acquisition component is continued into maintenance;
- the simpler maintenance condition preserves similar behavior;
- continued exposure to the added component produces no measurable incremental advantage.

PA010 is the current clearest example.

Public wording should remain descriptive:

> continued app diary/self-monitoring showed no incremental maintenance advantage over accelerometer-only maintenance.

---

# Revised expert-check target

The literature already supports separating implementation/exposure from mechanisms of impact.

Therefore the external question should **not** be:

> Is it legitimate to separate exposure failure from mechanism failure?

It should be:

> **Is our operational mapping of PA006, PA010, PA012 and PA013 onto implementation/exposure versus incremental behavioral-effect failure defensible, and are we using the most appropriate established terminology?**

Secondary question:

> **Is “preserved behavior” a useful descriptive maintenance outcome category when activity is maintained relative to a declining comparator, or should this be represented differently?**

These questions ask experts to correct our application of established concepts rather than validate an invented framework.