Reach a participant in the moment that matters.

A JITAI delivers the right support at the right time — adapting to a person's real-time state, location, and physiology instead of a fixed schedule. mEMA gives you five ways to build one, from a single survey answer to layered Garmin biometrics.

THE IDEA

Support that shows up only when it's useful.

Static, one-size-fits-all reminders either pester people when they're doing fine or arrive too late when they're not. A JITAI watches for the right conditions and intervenes at moments of vulnerability or opportunity — which is exactly what protects against alert fatigue and keeps participants engaged.

…and stays quiet the rest of the time. Frequency governors cap how often any trigger can fire, so you never over-treat.

The four parts of a JITAI, in mEMA terms.

If you know the JITAI framework, here's where each component lives in the platform.

Pick the channel that fits your protocol.

Every trigger reacts to one of three inputs — a survey answer, a location, or a biometric — and can route its action to the participant, to a care-team member, or to a researcher's inbox or phone.

A participant's answer to a specific question sends an email to study staff — the simplest way to route a flag to whoever is watching the inbox.

Same response logic, delivered as a text message when reaching someone quickly matters more than an email would allow.

The Support Team turns a flag into a real-time, in-app workflow. A trigger can drop a message to a caregiver or cohort member, prompt that person to complete a survey about the participant, or route straight back to the participant with their own message or follow-up survey — no third-party step required. A configurable delay (minimum 1 minute) governs delivery.

A geofence reacts to where a participant is and how long they stay there — not just the instant they cross a line. Fire when someone enters a place (On Entry) or leaves it (On Exit), but equally when they remain inside past a set dwell time (Req Time Inside) or stay outside one (Req Time Outside).

Trigger Mode sets the rhythm: Transition fires once on each in↔out state change, while Position fires continuously the whole time the participant sits inside (or outside) the area. Tune it with a Boundary Radius for entry, a Monitored Radius for exit, and an Outer Buffer that wakes monitoring as they approach. Group locations by region so participants in different areas get the right fences.

Pair a Garmin watch to the app and mEMA reacts to the body's own signals — stress, heart rate, near-real-time HRV, steps, respiration, and more. Build anything from a single-threshold rule to layered multi-condition logic. This is the channel with the most range, so it gets its own section next.

Start simple. Grow as complex as the science demands.

A trigger is an expression, and expressions compose. Begin with one comparison; add conditions, grouping, and logical operators only when your protocol calls for them.

The screenshots below are three real triggers from the Advanced Garmin Triggers builder, in rising order of complexity — each condition is a signal, an operator, a value, and a connector (and / or / none) that either continues the statement or ends it.

1 · One condition

The Breathing Exercise trigger fires on a single absolute threshold — 60-second stress at or above 10. Enough for many designs on its own.

2 · A few conditions, joined

The Grounding trigger — the rule shown on the cover — chains three signals: high stress and low step count, or a low heart rate. The closing parenthesis and End Statement on the last row mark where the expression finishes.

3 · Grouped, baseline-relative logic

The Complex Logic trigger nests two grouped clauses with parentheses. Note Uses Baseline Percent is ticked, so every threshold (120, 150…) is read as a percentage of the participant's own baseline rather than a raw value.

Operators > < >= <= === !== Connectors and or none (ends the statement)

Absolute vs. baseline

Leave Uses Baseline Percent unchecked and a trigger fires on an absolute value (stress ≥ 10, as in rules 1 and 2). Tick it — as in rule 3 — and the same rule adapts to each participant's own normal. Set it per trigger, or across all of them from the Configuration page.

Continuous vs. cumulative

Continuous signals are sampled over 60 s, 5 min, or 10 min windows. Cumulative signals — like steps — accrue across the day.

Near-real-time HRV

Trigger on heart-rate variability, calculated as pNN50 or Avg Standard Deviation. Tune baseline-queue length and window to match your measure.

From rule to action

A written trigger does nothing until you assign it. On the Time Triggers table you pair each rule with the survey to deliver, choose whether to Notify Team, and add an optional message. Below, Complex Logic is set to notify the team with “Please contact Participant!” — and the grey line beneath each row is the compiled expression mEMA actually evaluates.