FEATURE: Wearable Tech - Fit for MRX?

Wearable tech is everywhere, and offers great potential for collecting data, but how useful is it right now for answering specific research questions? Kathryn Ambroze, Manager of Behavioral and Marketing Sciences at HCD Research continues our month-long focus on 'Measuring Brains & Bodies'.



Market researchers are constantly looking to incorporate innovative and new solutions in technology to monitor physiological data. Wearable technology is one of the most popularized approaches for collecting activity from the body: however, the verdict on its utility in industry applications is largely subjective. In this article, we will review how wearable devices can work in research, as well as expose some often overlooked but important limitations of these technologies. By taking a critical look into wearable technology, market researchers can decide for themselves if a device's capabilities meet their standards to help uncover their specific research questions.


The Opportunity

Wearable technology, largely because of the promotion of smartwatches and fitness monitors, has become one of the most recognizable approaches for collecting body signals. Many of these products are easily available on Amazon and can provide a multimodal approach to capturing activity during virtually any experience. Often used to monitor health (such as heart activity and sleep patterns) to wellness tracking (like steps and even food intake), many people have incorporated wearable technologies into their private lives. With the public adoption and advances in noninvasive, portable technology, market researchers recognize the opportunity to easily integrate these tools into research. While the accessibility and ease of using wearables may be attractive, there are some considerations to address before going all-in on the technology. While it seems like the perfect solution to continuous measurements or deeper insights, researchers must be aware and cautious of its limitations. Behind the friendly interface a lot of careful design is involved, which keeps its execution consistent for consumers but may have impractical implications for research. Leading with the research objectives can help distinguish the real value, if any, wearable technology has for a project.


Quality Matters

Wearable technology is rapidly expanding, and the progress made in such a short span of time is impressive: however, while the sleek design gives a cutting-edge feel, it is important to remember not all wearables are created equal.

The technology, consumer-grade or not, is still only as good as the quality of its parts. Researchers should be aware of the quality of the sensors being used, since that has a direct impact on the signal collected. For example, most research grade (non-wearable) physiological sensors used for tracking heart rate changes over time rely on ECG (electrocardiogram) which is the most accurate and reliable measure. Wearable heart rate devices typically use PPG (photoplethysmography), shining light into the skin and measuring light scatter by blood flow, as an alternative to ECG, but PPG can present certain experimental design implications, including optical noise, sensor location, skin tone, etc. Generally, you get what you pay for, but even the highest quality, most expensive wearables available can have similar shortcomings.

As with all technology, the quality is often associated with the price. Researchers should recognize that consumer-grade devices are designed for consumers, and the data are not 'research-grade': therefore, a lot of the information is presented in a way that is digestible for the layperson. While appropriate for personal use, the adjustments (such as data smoothing) made for consumer-grade wearables may present issues in research settings due to lack of specificity.

For example, gyroscopes are used in wearable technology to understand the object's location in space. If the gyroscope detects too much motion, it will pause the algorithm recording biofeedback measures, such as heart rate, in many consumer-grade wearables. This period, where the wearable stops collecting data, presents a huge gap in data recording and analysis, potentially missing valuable information during a consumer experience. If using consumer-grade wearable technology during market research, it is important to be aware of these caveats that can be detrimental to a research study if not understood.


Remember who is wearing the wearable

One of the biggest reasons for interest in wearables for consumer research is the appeal of being able to measure consumers easily in their natural environment. However, one of their biggest limitations is that the research will always be reliant on the participants' ability to appropriately use and wear the wearable. If the technology is tampered with, the signal and data quality will diminish. Having windows of time without data may involve missing key behaviors. Other data loss issues may present themselves if the battery life is inadequate. Having a longer power cycle can help mitigate issues with charging (and less responsibility about logistical efforts can keep the participant more in-the-moment). Further, the true experience of the participant is a big assumption unless a camera is recording the participant. Participants may say and do different things, which can also jeopardize the learnings.

The real world is a messy place, which can result in messy data. Setting parameters to minimize ambiguity or confusion is a crucial step in the design process, but ultimately, researchers must decide if the noise from wearable technology will overshadow any meaningful responses.


Check the detail available, and read the data raw

Market research needs clear, measurable concepts to extract insights. When looking into wearable technology, it's important to consider its capabilities and the level of detail necessary to perform strong analyses. Some important considerations are listed below:

• Space: Make sure the data capacity of the wearable can fit the scope of the study.
• Raw data: The exporting capabilities often consist only of automated graphs. Access to tabular data is something that researchers should consider when vetting out different wearable products since this will directly impact the statistical analysis performed.
• Aggregate adjustments: Some wearable technology uses momentary, rather than continuous, measures to account for normal variance. By using sliding time windows or an average of beats, the data is cleaned before the researcher even looks at it. Interval data is then impossible to collect because the output is already aggregated.

It is easy to get swept up in the excitement of wearable technology; however, the data collected from this tool does not provide a comprehensive evaluation. Furthermore, misconstrued information can lead to overpromising or inaccurate results. Keep in mind the research question when deciding the technology under consideration. Will this device actually provide additional insights? By leading with the research question and using the appropriate methodologies to fill in the necessary gaps, researchers will find more success.


Don't fall for the rose-colored glasses

Wearable technology has infiltrated the field of marketing, providing additional information about how consumers respond to real-world experiences. However, the consumer experience of wearables is intentionally intuitive: designers keep the system as simple as possible to avoid confusion. Behind the sleek surface, wearable technology is complex and intricate. To optimize the use of these items, it is important for market researchers to understand what is happening in the process of data collection so the output can make sense. It is not enough to trust the algorithms behind these tools - researchers must explore how they are developed and validated to ensure its full value. Behind every algorithm is a human, and no human is perfect. Learn the inadequacies of the tool and decide if the shortcomings jeopardize the findings, or if they can be minimized by being paired with another measure. By reflecting on these concerns prior to the research, there is an opportunity to compensate for what the tool is lacking - or find a new measure that has a stronger approach.

Are wearables 'fit' for MRX? Yes, no, and it depends. At the end of the day, researchers need to fully understand the problem they are trying to explore because the fanciest of wearables will not help if the data makes no sense. Be aware of the risks and limitations, and really reflect on what the recording is adding to the overall understanding of the research objectives. Being flexible and critical in the methodology and approach can help researchers use the right tool for the right question - with or without wearable technology.



Kathryn Ambroze is a behavioral neuroscientist with experience in consumer research and methodological innovation. She earned her bachelor's in Neuroscience and Business from Muhlenberg College and currently works at HCD Research as the Manager of Behavioral and Marketing Science focusing on methodological development and innovation applying neuroscience and psychological tools to consumer and market research. Kathryn is also currently studying at the University of Pennsylvania pursuing her Masters in Behavioral and Decision Science.


SEE ALSO other pieces from this supplement published as articles on DRNO:

Neuro Best Practice Guide - March 1 2022

Neuroscience Techniques and Best Practice #1 - February 24

Podcast / video interview: Neuro-Guru Dr Michelle Niedziela - February 3 2022

Using EDA to Track Response to SquidGame - February 2 2022

Myths and Realities in Neuromarketing - January 11 2022

View the whole supplement in pdf format - download for free