身体活动依从性 — 数字干预微地图 v0.2
A working 15-study structural map of selected adult digital physical-activity interventions, extending v0.1 with phase, active-comparator structure, adaptation, engagement/exposure and maintenance interpretation.
This sequence is a hypothesis-generating bridge, not a validated causal model. v0.2 asks what the digital layer adds beyond an already active self-monitoring or feedback backbone, whether participants meaningfully receive that added support, and how the result changes across acquisition, continuous treatment and maintenance phases.
These examples show why the absence of added benefit cannot automatically be described as failure of digital behaviour change. The interpretation depends on each study's design, comparator and exposure data.
Both groups received tracker feedback. Extra text prompts showed a transient early advantage but no sustained difference in the primary activity outcomes at six weeks. Message burden is a possible constraint, not a demonstrated causal mechanism.
After a shared acquisition package, continuing the app diary during the maintenance comparison did not outperform a simpler accelerometer-only condition. This does not negate the earlier acquisition response.
The JITAI addition had no clear average benefit over its step-counter comparison; perceived usefulness moderated response. Exposure limitations constrain the interpretation.
How can studies distinguish support delivery, actual exposure, incremental effects above active comparators, and protection against decline during maintenance?
Examples from a selected, hypothesis-generating set; not a systematic estimate or validated mechanism model. Study-linked records: comparison matrix and cross-era analysis.
Separates acquisition, continuous active treatment, explicit maintenance and post-intervention follow-up.
Shows what both groups already receive before attributing value to the incremental digital component.
Distinguishes fixed packages, static tailoring, dynamic goal adaptation, JITAI and response-adaptive sequencing.
Records evidence that support was actually used, read, worn or otherwise received; missingness alone is not treated as proof of disengagement.
Separates sustained gain, preserved behaviour, attenuation, no maintenance advantage and designs that do not actually test maintenance.
Null findings are described at the level supported by the study. “No incremental behavioral effect despite documented exposure” is preferred over unmeasured claims that a mechanism failed.
The table is rendered from the working study dataset micro_map_manifest_v0.2.csv. Filters affect display only; they do not modify the data.
| ID | Study | Phase | Shared backbone / active comparator | Adaptation | Engagement / exposure | Maintenance interpretation | Source |
|---|---|---|---|---|---|---|---|
| Loading study data… | |||||||
The more informative contrast is increasingly what is added around monitoring and feedback, not whether monitoring exists at all.
Recent rows include contextual JITAI and response-adaptive SMART designs, while earlier rows more often vary fixed components or static tailoring.
When both groups already receive substantial monitoring, feedback or counseling, a null incremental result has a narrower meaning.
Delivered support is not equivalent to received support. Weak or uncertain exposure limits what can be inferred from a null behavioral result.
Preserved activity relative to a declining comparator is kept distinct from continued gain.
Changes in walking or activity do not automatically imply corresponding changes in fitness, weight or other physiological outcomes.
v0.2 is a working public version and has not been frozen. External methodological validation is still pending.
External feedback is invited on whether the operational mapping of PA006, PA010, PA012 and PA013 onto implementation/exposure versus incremental behavioral-effect failure is defensible, and whether preserved behavior is a useful descriptive maintenance outcome category.
This is a selected-study structural map for comparison, correction and hypothesis generation. It is not a systematic review, meta-analysis, causal model, risk-of-bias assessment or intervention-effectiveness ranking.
The working construct behavioral salience is used only as a bridge for hypothesis generation. It is not presented as an established validated theory.
Co-maintained by Catherine Ge and Starr Choi / SISFIT Open Research. Frozen v0.1 remains unchanged.