Ecological Momentary Interventions (EMIs): Considerations and Best Practices

Maybe I’m the stereotypical absent-minded professor. But if you are like me, you probably need reminders of upcoming meetings, past conversations, and the like.

This idea of reminders or nudges has gained popularity because of increased ease and automation that can help people become more effective and efficient. Think Gmail Nudges. Or Calendar Reminders.

Researchers are increasingly interested in using this idea of technology-enabled nudges to overhaul psychological interventions. We recognize that even after intensive training for behavioral change – whether diet, exercise, or a work habit – we need nudges to remind, encourage, and build up to desired behaviors. These nudges are training wheels of behavioral success.

But the best nudges don’t happen in an isolated context or a well-controlled laboratory. They occur in the messiness and chaos of daily life. And indeed, we see more researchers using ecological momentary interventions (EMIs) where nudges are combined with daily life rhythms to help people achieve their goals and live a better life.

How do we best implement EMIs, especially in a mobile app?

I summarize some pointers here from a recent paper, “Conducting mobile-enabled ecological momentary intervention research in positive psychology: key considerations and recommended practices,” published in the Journal of Positive Psychology that I co-wrote with Yerin Shim and Victoria Scotney.

We reviewed the key ideas and practices on EMIs in the field and created a framework to better design and evaluation EMIs. Synthesizing past work, the key dimensions we need to consider:

Mobile technologies

• Types of platform that can be used on different devices
• Types of passive sensors needed
• Degree of data wrangling

Ecological contexts

• Participant characteristics
• Time zones
• Incentives

Momentary characteristics of nudges

• Surveying appropriate timings for effective nudges
• Type of trigger scheduling

These dimensions need to be considered in light of the research design phases: design and selection (e.g., the population of interest, targeted behaviors, available resources) and evaluation (e.g., data collection procedures and methods, data analytic procedures and methods).

Clearly, there is more nuance to this, and I encourage you to dive into the paper. However, if you are putting together an EMI study for a paper or a grant, I hope you can use this framework to consider these dimensions carefully. You may just be able to create and design a more effective EMI that can hopefully nudge your participants to the desired behavior.

Wishing you all good success as we work toward better behavioral change!