Ecological Momentary Assessment in Health Behavior Research
Ecological Momentary Assessment (EMA) or Experience Sampling has been used in psychology research since the 1940s emerging from an recognition of the accuracy limitations of recall. Our memories are systematically biased by emotional intensity, priming and psychological state drawing into question the validity of retrospective autobiographical reports. EMA offers a method for capturing time-varying subjective experiences close to when they happen reducing concerns of response biases and memory distortions.
More Accurate – Context-Aware
Ecological Momentary Assessment (EMA) aims to capture more accurate self-reports by asking people about their experiences closer to the time and the context they occur. The context could be the external environment, the participants’ physical or internal state. The mEMA app is context-aware. It taps into the sensors on the phone (e.g. GPS, light, sound, proximity, motion, humidity, barometric pressure) and to those from wearable devices (e.g. heart rate, HRV, GSR, altitude, UV exposure, etc.) so it can deliver a survey at just the right moment. Furthermore, mEMA allows users to capture photo, video and audio files to submit as a survey response allowing researchers to access contextual information of which the user may be unaware. All these data are then sent to the platform for further analysis. Additionally the mEMA platform is integrated with third party databases that allow us to provide data of the local weather at the time any assessment was taken.
This method of data capture is much more accurate than the traditional method as it provides us with information about the context and does not rely on notoriously biased retrospective self-reports. EMA has been found to out perform pencil-and-paper methods of data collection.
Captures Dynamic Processes
EMA allows for more frequent sampling (often multiple times a day) so time-series analysis can be performed providing a deeper understand of the processes at work rather than static snapshots at distant time point. Research has found that health behaviors, emotional experience and strategies for dealing with stress fluctuate significantly throughout the day across different daily contexts. Some suggest that the larger portion of this variability can be accounted for by change in the situation not the person (Hoppmann & Gerstof, 2013). EMA allows us to delve deeper into intra-individual variability as a valid developmental process. Furthermore, EMA methods allow long-term capture of data providing insights into events that may happen less frequently and would be hard to replicate in the lab.
captures Interpersonal Dynamics
EMA can also be used to understand interpersonal processes. Our daily experience are often linked with that of those closest to us, this is not easily captured with traditional methods. EMA studies where each spouse, for instance, provides timestamped self-reports throughout the day can illuminate these more complex dynamics.
History of EMA
Early EMA studies used paper-based daily diaries and asked participants to record their own behavior usually once a day. With the advent of pagers researchers were able to design signal-driven sampling studies, “beeping” participants at random times throughout the day to signal them to record data at that moment. This method, popularized by Czikszentmihalyi and colleagues became known as the Experience Sampling Method and aimed to capture participants’ subjective experience in the moment. As the technology developed so too did the methodology.
With the introduction of handheld (palm-top) computers many digital daily diary studies were carried out in a diverse range of health behavior fields. Additionally, ambulatory physiological monitoring has been included in some EMA protocols to capture such biometrics as EDA, heart rate and movement.
In 2015 64% of American adults owned a Smartphone, a powerful tool for collecting both self-report data and passive data from either in-phone, external or wearable sensors. Bluetooth and WiFi allow us to collect data from a variety of sources in the participant’s environment or from their physical body, combine it with their own perception of their experience and deliver the entire data package to researchers anywhere. This widely available technology has the ability to revolutionize the way psychologists, therapists, physicians and behavioral health researchers understand people.
The rapid development of wearable and in-home sensors is allowing integrated health monitoring solutions to be more easily created and just-in-time interventions made available. These technological developments allow capture of objective measures (e.g. activity, heart rate, etc.) alongside self-report data alleviating concerns about socially desirable responding.
Current EMA methods are being employed in a wide range of areas including: eating behaviors, drug and alcohol use, sexual behavior, emotion and wellbeing, medial and psychiatric disorders.
mEMA by ilumivu
The ilumivu’s EMA package represents the next generation in tools for EMA researchers by providing:
- capture data directly from participants’ own phone (iOS or Android), no need for them to remember to carry a second device
- data stored locally on phone until within cell or WiFi range then automatically pushed to secure central server for high data security
- integration of biometric (electrodermal activity, heart rate or actigraphy) with self-report data
- reducing software development costs by accessing the Survey Editor yourself, no programing experience required to add and edit your questionnaire to be displayed on the Smartphone apps. We understand that during the pilot phase there will be multiple iterations of your questionnaire. We don’t believe you should pay extra to refine your questions in response to user participation and feedback, that’s why we built the tools that enable you to edit your questions without input from ilumivu staff.
- reduce cost of data transcription and transcription errors, all data automatically sent to central database to be viewed in real-time