Five Questions With:
Pete Bacevice and Ar Ducao on Biometrics
July 16, 2024
Alex Dunham of Ark/HLW talks with Pete Bacevice of Pangeam and Ar Ducao of Multimer, two leading experts using biometric devices to study how humans respond to different physical environments. This episode uncovers how biometric devices can help us study space and explores some of the findings from a controlled study comparing how workers of the same organization react to two different office environments. Duration: 32min 10sec
Transcript:
Alex Dunham
Welcome to the Five Questions With Podcast. My name is Alex Dunham. I'm the managing director of Ark, a multidisciplinary studio focused on the future of work. Today we're bringing you a unique interview from our Ark Research Lab. We're going to be exploring five questions with Pete Bacevice and Ar Ducao.
Ar, please introduce yourself to our listeners.
Ar Ducao
Thanks, Alex. Hi. My name is Ar. I use they pronouns and I'm a co-founder and principal at Multimer, a bio spatial analytics spinoff from MIT Media Lab, a National Science foundation, Small Business Innovation Research Award winner with multiple patents. Multimer has worked with marginalized communities in Africa, Asia, Europe, and here in North America. I'm also the creator of a sci fi art house animation series called The Great Tit is a Bird, which follows black and brown, feminine and trans feminine people around the world, the losses they survive and the ways they shape our high-tech era, even as they're pushed to the margins.
Alex Dunham
Wonderful. Great to have you. Pete, tell us a little bit about yourself.
Pete Bacevice
Sure. Hi, Alex. My name is Pete Bacevice. I am Vice President of Research and Development at Pangeam, which is a workplace analytics platform that uses AI technology to measure the performance of the built environment. I've worked at Pangeam now for just over two years, and prior to that I was at HLW, where I spent eight years as Head of Research for the Practice and was part of the strategy team.
In addition to my professional work, I've also been a researcher at the University of Michigan's Ross School of Business, where I have largely focused on the intersection of human thriving and the built environment’s impact on that.
So much of my research over the years has influenced my professional practice and vice versa. So, it's really good to be back with you, Alex, and being able to talk about great work we did at HLW.
Alex Dunham
That's great. You as well. Thanks for joining us. Great to have you both today.
We wanted to speak with you about an exciting and emerging topic, really, biometric devices, specifically what they are, how they help us study human response to physical environments and how they might be used to better inform workplace design in the future.
So, I want to frame this discussion more specifically around a study you undertook in 2019 and 2020, one in which you actually attached these biometric devices to a set of office employees to gauge how they were feeling and responding throughout their day in the office. It was a fascinating experiment to hear about and certainly a very nuanced way to study space, at least in the industry I come from.
So, before we get into the details of the study itself, our first question and Ar, I’ll post this to you. Can you give us a little bit of background on what biometric data is and how your company, Multimer, has used this type of information in the past?
Ar Ducao
Yes. So, we can split the word biometrics into two: bio and metrics. Metrics for measuring something, and in this case biology, and in this case, human biological signals, and in our case, electrical signals. So, Multimer hones in on electrical activity from the brain, so electroencephalography (EEG); electrical activity from the heart, so electrocardiogram (EKG) data; and then other kinds of supporting data that can help us in sense making, so spatial data, GPS, Bluetooth beacon data, and notes that participants take while they're contributing data to the study.
Alex Dunham
And just to follow up on that, what kinds of reactions are you measuring in the body itself?
Ar Ducao
Particularly with a fusion of EEG and EKG data, we can look at stimulation, stress, attention, you know, looking at heart rate. The EEG sensor is a low spatial resolution. So if you think of the brain as a space and you think of EEG electrodes, you know, as providing the resolution, we're using a very easy to use, easy to put on, low resolution sensor that only touches your forehead, so the frontal cortex, but that can still give us a lot of insight, particularly since we're collecting a very high volume of data, 500 hertz of data over time.
Alex Dunham
Okay. Pete, as an extension to that, with your background in the built environment, what are your thoughts on how we can take that concept of biometric data and use it to assess our physical surroundings?
Pete Bacevice
There's a hunger for new insights into the performance of the built environment. Specifically, there's an interest in knowing how the built environment impacts us, both good and bad. Architects and researchers of the built environments often use things like surveys and observations and interviews to understand this impact. But of course, there's limitations to doing research this way, you know. You're limited to what you can recall and not every aspect of the built environment we remember. Sometimes we experience a space, and we don't know that's causing us stress. So, biometrics gives us a new way to understand the built environment at a much more granular level that we might not have been able to previously detect.
Alex Dunham
Okay. So how do you go about structuring a measurable study using these concepts? Can you give us an example?
Ar Ducao
Yes. So over time, understanding how to kind of fuse all these different pieces of data was key. And one of the big studies that we conducted that was an immediate predecessor to the HLW study was a National Science Foundation supported study of the activity of cyclists, pedestrians, some drivers in the Central Business District of Manhattan, so Manhattan South of Central Park. And so, all of these folks were wearing sensors when it made sense for them, you know, over the course of six weeks in the summertime. So, it was very temporally scoped, but they were collecting data often during work hours. We had a lot of bike couriers involved. With all of this gear, a lot of it on their helmets, a lot of it on their bikes, a lot of it strapped around their chests and when it was kind of safe and convenient to do so, they would record notes as well about what they were seeing as they move through the built environment of the city.
Alex Dunham
How did that ultimately evolve into the workplace study?
Pete Bacevice
So, I think what was really fascinating about that research was the way that the results were presented. We could literally see the Manhattan street grid and we could see the pain points on the street grid, according to cyclists. So, when a cyclist encountered a stressor, maybe traffic congestion or some sort of physical barrier, those would show up as data points on the grid. And I thought, well, this is a really insightful way to understand the performance of environments because you’re layering aggregate biometric data or the experience that a group of people have at the same point in the space, and you can begin to see where those stressors are, where those pain points are.
So, I thought, well, I want to bring this indoors. I want to try this in an office environment. Rather than thinking about street grid, what if we imagined a floor plan and we did a similar project where we had participants wear these sensors as they went about their daily lives in the office, experiencing different things that, you know, might be stressful or things that might not be and things that might cause relaxation or might make you happy.
We were looking to understand the variation of experiences in the built environment from that biometric perspective. To my knowledge, there really hadn't been research that had taken that kind of biometric technology and used it to understand the indoor environment. So, working with Ar, we developed a research plan, and in April of 2019, we recruited about ten volunteers at HLW’s office to wear the sensors as they went about their work life.
Basically, the sensors would be recording their brain waves during that week in the office. Some volunteers also contributed heart rate data as well. And over the course of that week, we had a really interesting dataset. But I think what made this an even more impactful project was the fact that HLW was leaving its space. We were looking for a new office. So, we actually had this quasi-experiment where we could introduce some controls to the study.
So, we had the same people that would be able to wear the sensors in two locations to the same organization. And we could compare old space versus new space. So once HLW moved into its new offices and people were settling in, boxes were being unpacked and all that sort of thing. That was in June of 2019. So, in January of 2020 we managed to sneak in right before the pandemic, apparently, we were able to do the second round of data collection.
Alex Dunham
Yeah, it's really intriguing. I certainly love the idea of bringing the scientific method into our work planning, probably doesn't happen often enough. It sounds like a very involved exercise, involving a lot of planning, a lot of coordination, and I'm sure it was quite tricky to pull off. So as we move into our second big question here, what are some of the challenges one would face in conducting a study like this?
Ar Ducao
So, this was our first formal indoor study. This is Multimer‘s first formal indoor study. We conducted experiments in our workplace, but this is a formal study aspect was a challenge, in that particularly the head sensors, you know, their ergonomic challenges for indoor folks, wearing something on their head all day.
We had become very experienced in outfitting outdoor gear, you know, namely bike helmets with sensors. But you're not going to have an office worker wearing a bike helmet for the entire workday. So, for the HLW study, we brought in little chef's hats and attached the sensors to the little hats. So, you know that kind of marked all the participants. And that was an issue as well, it's sort of a self-conscious thing.
Also, you know, with brainwaves, there is all this sort of, you know, cultural baggage around, measuring brain activity. And I think that can contribute, often rightly so, to a perception of creepiness around collecting this data. And then, of course, there are issues of data privacy, and some participants, every participant could opt in and opt out at any time. They could opt out by turning off their sensors, you know, and some participants didn't want to contribute EEG data at all. They just wanted to contribute data with the chest strap or their own watch. So, accommodating and understanding all these different challenges was also interesting when it came to analyzing the data.
Alex Dunham
Yeah, I'm sure there's a chasm between wearing a personal Fitbit or Apple Watch and wearing a foreign chef's hat and biometrics set, right?
Pete Bacevice
Yeah, for sure. You know, certainly the fact that we're capturing data during office hours when people are going about their daily work. So, they're wearing these sensors, they're wearing this headset. We don't want to be disruptive to them. We know from past research that wearing EEG sensors there is a discomfort. So, you know, what would that be like after wearing it for a week? And that also brings up another thing, just about the background research we did into this, because we knew that there wasn't really much of a precedent for doing this in the indoor environment, we had to think about, you know, what would the comfort level be for wearing a sensor for a week? What if they had to take the sensor off at some point, you know, we knew that we were generally getting continuous data, but we knew it wouldn't be, 8 hours a day times five days a week. So, you know, we were dealing with a little bit of messiness that comes with capturing data over a longer period of time.
There were, cultural things we had to take into account. And by cultural, I mean just organizational culture, willingness to take risks in order to learn something new or even conduct a study like this.
Alex Dunham
Well, I would say kudos to you both for obviously pushing the boundaries and experimenting with this. As an extension of the same question, are there ethics or larger governance implications when you do a study like this?
Ar Ducao
Yes, certainly. So, for our outdoor studies, particularly studies that received federal funding, approval, Institutional Review Board (IRB) approval is required. And so for our big Manhattan study, we had that IRB approval and some of those protocols we brought into the HLW study. Another aspect of governance that I think is getting a lot of well-deserved attention today that was less developed when we did the study and when Multimer was starting out is if you're collecting a very high volume of data from a single person, particularly electroencephalography data from a single person, especially if it's from a high resolution cap, say, of many electrodes, you know, that are implanted, that are very close to the brain, then that does start to potentially become personally identifying data as much as our, you know, birth date, our Social Security numbers are. And so, legislations are starting to take a look at larger issues of privacy. So, I feel proud that we you know, we're thinking about those questions back then when they weren't so acute and also thinking about how to minimize collecting other kinds of personally identifying data that could potentially be compromised.
Alex Dunham
Yeah, well, it sounds like it was well-considered and maybe even on the forefront of some of the things that have emerged in the years since.
So, Pete, I’m going to start our third question with you. What were the core outcomes of the study? What did you learn?
Pete Bacevice
Sure. Well, we were able to look at the brainwaves in each of the participants and how they changed throughout the week. Similar to the outdoor data we could see where there were stress points in the indoor environment.
So again, remember when I first saw the data superimposed over the street grid, we could see both areas of attention and areas of relaxation. So basically, translating the brainwaves into where the points where people are more stressed and where our points where people are less stressed. I'm oversimplifying, of course, but this is really sort of how you would take this data and use it to understand the built environment.
So, we're looking for a pattern in the indoor environment as well. Rather than it being a street grid, we're looking at a layout, we're looking at a floor plan. In the original HLW office, we saw an interesting pattern where the cafe area, the pantry, which in the old office was kind of buried in the back of the office, it was surrounded by the work areas. It was in a high traffic area, you know. We were seeing that the brainwave patterns of people that experience that space were not too different from some of the other work areas. And then when we compared the results to the new office where the cafe had a completely different design affordance, it was placed right at the front of the office when you first walked in. It's like walking into an open concept kitchen in your home, it's welcoming, there's a view to the outside. It was designed to be a more relaxing, a more social space that was a little more separate from the work areas, and we were able to see the change in the brainwave patterns among the participants that use the space in that second round. People were experiencing that cafe in a more relaxed way. So, what that one set of findings showed us was, you know, we had an assumption about how we could design the cafe better with a different kind of design affordance, meaning, you know, this is the way that you should be using the space.
You know, there are cues in the built environment that, you know, suggest how you can use something, that’s what an affordance is. So, when you introduce soft seating, a big food table, it's meant to evoke a certain kind of social relaxing behavior. And we just didn't have as much of that in the old office. And that brainwave data, let us see what the impact was and how people were experiencing that new space.
Alex Dunham
Excellent. We've often promoted in our work as strategists, you know, that the cafe is kind of one of these beating hearts of the modern office. So I'm glad to hear that you focused on that in particular. We know that's where a lot of community building happens, particularly in the post-pandemic environment, as people are being drawn back to the office for that socialization.
Ar, were there any other kind of anecdotal instances, other particular space types that jump off the page to you.
Ar Ducao
Definitely. So in the pre move space, most folks didn't have their own offices, but there was one area of, you know, private, fully enclosed offices and I believe one of those offices is this kind of the C-suite, as I understand it. And that definitely showed up as an attention hot spot in the first round of data.
Alex Dunham
Makes total sense to me. I'm sure anybody out there in corporate America can jive with that. Okay. Well, it's great to hear that there were particular things in mind. And I know that having done this study for a longer period of time or having done it for multiple rounds, I'm sure other things would jump out at you.
Let's turn our attention now to how we can move forward with this type of study and the data and results that have emerged. How are other organizations really supposed to leverage this type of information and begin to adjust either their physical space or how they're operating going forward?
Pete Bacevice
There are a lot of things that architects and designers intuitively know about how good space should be designed. And so, when I was describing the findings from the second round of the study, I had indicated the changes that were made to the new space, you know, the introduction of natural light or the generous use of that in the space. You know, we positioned the cafe very close to the big windows. You know, we designed it in a certain way.
Intuitively, designers know this. This comes from practice, this is practice-based knowledge. But increasingly, there's this desire to put hard numbers behind this intuition. And as sensing technology becomes more sophisticated and can capture more granular data and just even more specific kinds of data, both from people or data that that comes from the built environment itself or how people interact with it.
You're beginning to quantify the performance of space and begin to quantify these intuitions about what makes good space in ways that haven't really been possible before. You have a greater level of fidelity in the data that the built environment is producing and what the and what the occupants of the building environment are producing.
Alex Dunham
Yeah, I like that idea.
I mean, as strategists and you well know this, we're often using ethnographic methods and other kind of first-person observation interviews survey to source our information. But we know a lot of those means have internal biases built in. And I think one of the powerful things about biometrics is, as you've said in the past, they're a bit harder to fake to your point they're more granular and have a higher degree of fidelity.
Ar, more broadly, same question to you. Are you using this kind of data to influence institutions or organizations in other ways in terms of how they operate or design.
Ar Ducao
Yeah. So, I think in terms of collecting biometric data, there are biases in regard to who's collecting the data, you know, and some of the things that we've talked about, time of day, what type of biometric data, that kind of thing. And so, something that's been on my mind and I'm not sure if this answers your question directly, Alex, but going back to what Pete mentioned about designers. I think to a certain degree, we are designers, and to a certain degree it's really useful for designers to be able to, participate, get their hands on this technology, understand how this really works. So it's not just a matter of, you know, handing over your data, to larger powers or, analysts. It's really aspect of empowering, you know, the designers on the ground. And that was one aspect that I thought was so interesting about this study, that the participants were designers and they were the folks who are in the office, not burdened with lots of travel, working on their designs. So in terms of, influencing institutions, I think the most important influence to be sort of at a more grassroots level from the level of design and designers.
Alex Dunham
Okay. That makes sense to me. A related question to you both is the idea of what you would do differently next time. Are there things you would change if you were running that same study today?
Ar Ducao
Yeah. So one of the hardware dreams of our team was being able to integrate the hardware into something that more people already wear, you know, whether that be headphones, like what we're wearing now or glasses. We studied hearing aids quite a bit, you know, and I think we talked about that a little bit, Pete. So I think sort of the ergonomics, the personal style, was something that we're always thinking about, always trying to improve upon and, always can be improved upon. I think also maybe a little bit more of a co-design aspect. So, engaging participants from the very beginning of designing the study, as well as engaging the participants, in a longer tale after the study was complete, I think would have been very interesting.
Alex Dunham
Mm hmm. Often the difference between pre-designed strategy and after-the-fact change management.
Ar Ducao
Right, exactly.
Alex Dunham
I’m very familiar with that, as is Pete. Pete, any thoughts from your end on things to do differently?
Pete Bacevice
Well, I really agree with the ergonomic design opportunity with respect to the EEG capture technology itself. Actually, I think this brings up a point that I wanted to make earlier about just in terms of the kinds of research that were done using EEG to study the built environment.
One of the things that we learned when we were designing the study, like good researchers, we did our literature review where we looked at the precedents that were out there and what was what was surprising, but not surprising, was that there hadn't been much research done like this. And part of that's because of these limitations of ergonomics. It can be hard to wear a sensor like this for an extended period of time. So, there's been research done where participants would wear a sensor for a very small amount of time in a lab and in a controlled setting. So, you're not really mimicking the real-life experience.
In other studies that I've seen, there are ones where participants would walk a set route in an outdoor setting, again, wearing the sensor. But again, it's limited, it's a controlled period of time through a controlled environment. We know that the built environment or the way that we experience this is not like that. You know, we take different paths.
And so, if we could solve the ergonomic challenge and we can improve that data capture technology, it would really open up a lot of really exciting opportunities to learn more about the built environment, because we will have overcome that limitation.
Alex Dunham
That's an excellent point. And I'm sure there are plenty of market opportunities to explore that going forward, which is a segue to our fifth and final big topic question here. This study was completed, as we said earlier, way back in 2019 and 2020, when the world was, shall I say, a different place. And obviously technology has advanced a great deal since then. What are some of the emerging or even future technologies and tools available that might help better inform a design process?
Ar, I know you've been doing a lot of work and exploring a number of different things, motion capture and others. Can you talk a little bit about some of that work and where it's going?
Ar Ducao
Sure. So as I mentioned, I'm also an animator and my animation project is called The Great Tit is a Bird. And so for that purpose, you know, and we're here at NYU in Brooklyn today, I've really gotten into using motion capture systems and learning about motion capture systems. So I've encountered research on the use of inertial motion capture suits that don't involve like a big grid of, cameras above our heads. So those kinds of suits being used to study workplaces, primarily manufacturing workplaces, workplace ergonomics kind of more basic productivity. The kind of motion capture I use that’s here in the studio we're recording in today is an optical motion capture system. And that kind of system is very commonly used with biomechanics, you know, bio mechatronics, like for instance, sports science uses optical motion capture pretty intensely.
So it's been really fun to learn about, a different kind of biometric for a completely different kind of project. And then, you know, in terms of thinking about space, in this project, The Great Tit is a Bird, some of the questions that it's grappling with, are aspects of space that are difficult to quantify and even difficult to express.
So this is another project that spun out of MIT. And last month the creators of the project, the performers of the project, many of whom came out of MIT, and experience aspects of their time at MIT in an unsafe way. We kind of came together and talked about, these hard to quantify, hard to articulate aspects of safety, unsafety, particularly in a post-pandemic era.
Alex Dunham
Wow! Pretty neat stuff. I would love to do another discussion on that one just by itself.
But yeah, I have visions of people in the workplace related to motion capture coming in with the suits like they're on a marvel set years from now and all dialoging with each other. So we'll see if it comes to pass.
Pete, I know you're doing a lot of work in the sensor community as it pertains to the world of workplace. Do you want to talk a little bit about that technology and where that industry is going?
Pete Bacevice
Sure. So, you know, I really look back on this project that we did as like a watershed moment for me because being able to quantify the difference in how people experienced one space versus another, I think really has influenced how I approach my work today at Pangeam, which is using a different kind of sensor.
You know, I really enjoyed learning about the neuroscience and architecture interface, but what I'm working on now, or what our company is working on right now is a really interesting technology that can sense human presence in space and anonymously measure and encode different activities and behaviors and then compare that to the design itself and to quantify the fit between the human and the environment or the team in the environment.
So we're working on a technology that we currently have deployed in some organizations today, and we're working on some really exciting new technology where we're going to take the concept I just described and make it even more granular. Again, you know, we talk a lot about granularity in the collection of data about built environment performance. And so I'm really excited by, you know, where sensing technology is heading because once you realize what you can sense and I think Ar makes a really good point about even with biometrics there are concerns about privacy and PII (Personal Identifiable Information). You know, what I've seen in this community of research is everyone takes that really seriously and the advancements in sensing technology are able to kind of work around that so that you're not violating PII, but you can still detect the presence of space so that you can continuously improve it.
And again, we saw what, you know, just this one simple experiment could show, the impact of an improvement to a particular kind of space. Imagine what we could do across, the vast majority of the built environment where, and frankly, we spend, what, 80 to 85% of our lives in the built environment.
Alex Dunham
That's right. Yeah. It's certainly amazing how quickly the technology and the tools are advancing. I think in terms of as you've both spoken about today, the variety of data sources and the access to it, the fidelity, the granularity and also the interpretation. Right. What we're doing with it and how we're using it to make informed moves going forward. So we certainly have our work cut out for us in the design community.
I want to thank you both for joining me today. I think it's been a really stimulating topic to discuss. And I also want to thank you both candidly for convincing people to wear these headsets in the name of science and design. And I'm sure it was no small task.
So, thank you again and good luck to you both in the future.
Ar Ducao
Thanks, Alex.
Pete Bacevice
Thanks, Alex.