Dr. Dwain Allan

Jam Mayer

Resources

Dr. Dwain's Research

Stress Mindset Theory

Cynthia Breazel

Sherry Turkle

There will Never Be an Age of Artificial Intimacy

Increase in SMEs experiencing stress and anxiety

How Burnout & Inflation Continue to Impact Small Business Owners

Jam

Welcome to the Conversologist podcast, where we talk about the art and science of conversation in the digital space. We know that technology can be a powerful enabler, but communication and emotional connection still need to be at the core. I'm your host, Jam Mayer, and joining me today is Dr. Dwain Allan, who just designed a conversational AI that can change an entrepreneur's mindset about stress.

Dwain is a psychologist, roboticist, and research scientist at the University of Canterbury. He is also a design strategist and expert of choice for success-hearted entrepreneurial individuals and value creators. Welcome, Dr. Dwain. Should I be formal, or is Dwain enough? Thanks for taking the time today.

Dr. Dwain

Thanks. Great to be here.

Jam

I'm giddy and excited because this is very relevant, because obviously this is close to my heart because I am an entrepreneur myself. Can you just tell us who Dr. Dwain or Dwain is? What was your journey like, and what led you to designing this project?

Dr. Dwain

Well, that's a lot of questions actually, isn't it? Well, I have been called a lot of different things, everything from a polymath designer to a social roboticist to a brass-tacks business strategist. So, lots of different things. I would say that I'm a lifelong student of psychology, social science, and design as well. I've long been a champion of small business owners and entrepreneurs. I actually started out in the 2000s and teens, during the "golden age" of business incubators. Do you remember those? Business incubators?

Jam

Yep.

Dr. Dwain

Yeah. Well, I worked at one and I was involved in administering the commercialisation services. That was for technology startups and high growth companies. So that was teaching business model innovation and agile methodologies, and while I was doing that, I had the opportunity to learn from some really experienced entrepreneurs and professionals. And really, I was just making a lot of mistakes, learning lessons, and really starting to hone my craft. And while I was doing that, I was eventually invited by a mentor to go and contribute to academia. So that was just based on my experience already in industry, and that kind of led me to complete a master's degree and then on to my doctoral research, where I studied social robotics, human interface technology, and also psychology, all up to my current position, which is at the University of Canterbury.

I like to say that I'm spearheading a renaissance in human factors, psychology, and behavioural design.

 Jam

That's awesome. Okay, let's just now move on to this project. Talk to us like a five-year-old. How does this AI potentially change my mindset of stress?

Dr. Dwain

I'll give you a little bit of background as to why I even took on the project itself. My current research is looking at the underlying psychological factors that sort of shape people's perceptions, furnishes about emerging technology products. So that's your AIs; that's your conversational agents, the digital humans as well, and also interested in understanding how they can influence people's thoughts, feelings, and behaviours. So that's kind of my overall area of research. 

So that sort of combines with my background. I continue to work directly with entrepreneurs and owner-managers in a bunch of different industries. And obviously, I've observed during that time that it really doesn't matter how brilliant or successful they are; they all get to a stage where they are simply overwhelmed by stress. Often to the point that it threatens their peace of mind, sometimes their relationships and even their business. 

So, I became interested in the nature of stress, why it is uniquely challenging for most entrepreneurs, and then obviously, what could be done to help it. Right now, there seems to be a bit of a rise in stress-related illnesses impacting people, particularly in our country. But I think all around the world, governments have also come out to try and support it, putting a lot of money into trying to help support owner-managers with their stress.

I was particularly interested in owner-managers, which are people who have employees, because they have a particular set of stresses, right? Because you're dealing with employees and other people, you've got people that rely on you. From a research perspective, the entrepreneurs are called a "hard to reach population group." So that just basically means that we just don't know a lot about them, not at a research level. But it turns out that the sort of stress management interventions and all these types of things that various researchers and practitioners have tried to produce for entrepreneurs don't go down so well. Some entrepreneurs tend not to really get involved in them. Some have cited them as being insufficient, very time-consuming, and not really relevant. And when you think about it, it just makes sense. A lot of the stuff that is created in the stress reduction space. It's really designed for people who are employees or in larger organisations.

Jam

You mentioned we are hard to reach in terms of research. Why is that?

Dr. Dwain

It appears to be the case that many entrepreneurs don't see a lot of value in research-related activities. And it also depends, I would think. So we're talking smaller business owners here, right? Small to medium business owners. They are time poor; they've got a lot on, and some believe that taking time to invest in that type of thing takes them away from doing their business activities. At least that's what the literature says.

Jam

Why not include the one-man band entrepreneurs? Don't they have the same level of stress?

Dr. Dwain

Well, it depends on if you're talking about levels of stress. I'm sure they would think they have the same levels of stress because it's largely subjective. It's more the case because they have people who rely on them. In terms of when you're employing people, you're directly responsible for those people in some ways. I was very interested in this relational level of stress that is put on the owner-manager. I was curious to see how that might change the stress. And there's also something about how many of the individuals that exist for SMEs, because there are some that work for SMEs are focused on employees, not the owner-manager. And indeed some small medium-sized enterprises are investing in their staff like they are creating wellness programs and some of these things for their staff. But the question is, are they doing it for themselves?

Jam

Interesting, I probably am part of that statistic. I do have staff, but not full-time employees. I think it would be the same because I get stressed when say, for example, cash flow is a problem. I'm not actually thinking about the money. I'm actually thinking I am actually impacting five or seven people who have their own bills to pay and put food on the table.

Dr. Dwain

The problem seems to be that there is little known about the spreadsheet experiences of many of the different entrepreneurial types. We just don't know a lot, not in terms of research, for some of the reasons I've just mentioned. There's a lot of talk in the research community that there needs to be more development of interventions that are in one way or another cost-effective, so obviously, entrepreneurs don't have all the money in the world, time-efficient for the reasons we just talked about, you need to be able to engage with these technologies, accessible, and also relevant to entrepreneurs and senior owner-managers. 

And as I said, the issue is that we don't really know what is relevant, and so it opens up a whole new can of worms again, at least from a research perspective, because it's not really clear what constitutes an appropriate strategy for the entrepreneurial population group. But there is one possible strategy that I was interested in, and this comes down to this mindset thing. And it's this concept called "stress mindset theory." The idea of "stress mindset" is that it posits that the mindset, so the underlying beliefs that somebody has about something in regards to, say, stress will influence to some degree the manner in which they respond to and experience stress.

Dr. Dwain

Okay? So that basically means that the degree to which a person generally believes that stress holds really enhancing possibilities, like it can be useful. They define that as a stress-is-enhancing mindset or whether somebody gravitates towards the mindset that stress necessitates negative consequences. So they call that a stress-is-debilitating mindset. The literature on stress mindset theory is fairly new, but the data that we do have on it does seem to indicate that the dominant mindset held by most people is that stress-is-debilitating, that stress is harmful. So I mean, if I would ask you, Jam, would you agree with the statement that experiencing stress enhances my learning and growth?

Jam

Actually, I agree.

Dr. Dwain

Cool. What would you give it on, say, a scale of one to five?

Jam

Five being the highest?

Dr. Dwain

Being the highest.

Jam

Four

Dr. Dwain

Oh, great, so that's excellent. Most people don't agree with that statement. And when you think about it, it just makes sense, because if you look around, the idea of a mindset is influenced by authority figures and messages that you receive on a frequent basis. So if you consider the nature of stress and what we see in the media, what we read online, what people talk about, and how those people talk about stress in their day-to-day lives, if you're talking with colleagues, friends, partners, or whatever, it's usually framed in the negative. I'm really stressed, I'm stressing out. You can see how the culture itself that comes around this seems to perpetuate the stress-is-debilitating mindset. The stress-is-enhancing mindset, so far in the research at least, we can see, is related to higher life satisfaction, greater optimism, less anxiety as well. Whereas, the stress-is-debilitating mindset, it's been linked to greater sensitivity to destructive coping strategies. And that's stuff like excessive alcohol consumption, procrastination, and ruminating on worst-case scenarios. That's usually what a negative stress, a coping mechanism might be. So that's kind of how these two big constructs work.

And then you have these things called "stress mindset interventions," which are designed to inform people about the broad nature of stress. And really, what we're saying here is it's not wholly negative all the time, right? You need stress to achieve things. It's almost like they call it "eustress." There's eustress and distress. Those terms never really caught on. They're very academic. No one goes around saying, "Hey, I'm experiencing stress." It never really took off. But there's a positive view of stress. It's not always bad for us. And so the intervention itself will provide exercises to help people adopt a stress-is-enhancing mindset. It turns out that they do improve people's responses to stress and can enhance the performance and well-being of the individual who's undergone it. If you looked at stress as enhancing possibilities, then it tends to have a different view on how people actually deal with stressors in their life. So, stresses are things that actually happen to you on a day-to-day basis. We're talking about the nature of stress itself because it's dealing with that nature. It's thought to be applicable to a wide range of groups.

If we come back to the entrepreneurs, what I just said about how we don't really know what's the right strategy, I posited that this could be something useful, right? We could take this construct and maybe use it somehow.

Jam

Is it similar to being resilient? That's why I've got that more positive mindset of stress.

Dr. Dwain

It predicts resilience. So that's a bit of an academic term. People who generally hold an enhancing mindset that stress can be good for me and can have enhancing benefits. Remember, we're not saying all stress is good. We're saying that it also has positive attributes. This is about saying, "can you see it that way?" People who see stress that way tend to have greater optimism, resilience, and mindfulness as a result of seeing the enhancing qualities of stress. Here are about a thing called stress reappraisal. Very similar to what you're saying in some ways where it's the way someone sees the stress and they reappraise the stress of themselves. This is actually working for me, not against me.

 Jam

So how does this theory and stress—and I'm loving this conversation—how does this then tie to the conversational AI side of things?

Dr. Dwain

The good question would be probably before we even get into that is why would we want a mobile messaging-based conversational agent as a potential delivery mechanism for this type of intervention? The great thing is that we've got some research to support that. Now, there is stuff in the health field around huge amounts of success for using mobile phones as delivery channels, particularly for population groups which entrepreneurs fit into, who would otherwise not participate in traditional health interventions. So that's a big tick. And then also we already know from in the medical literature that conversational agents that includes digital humans as well, anything that's using text, speech, visual based conversations are really cost-effective and an accessible medium for delivering mental health interventions in particular via mobile phone. So those two things are really great. So taking those together and coming back to your question, combining the stress mindset theory with these developments in mobile messaging phone use to develop this mobile messaging-based conversational AI, the design of this thing itself was rather intensive. So the reason why I'll tell you that is because I come from the psychological sciences, right? so very similar to the behavioral sciences.

We are very literature and research driven. We will always approach a problem from the literature, right? So we'll have a look at all the different viewpoints that are already exposed. What's the data like? What can we learn from what's already happened? What are the theories of learning? And so I was in behavioural economics, I looked at the marketing literature interdisciplinary. Like, how could we best design the right intervention? Also, this involved designing it with a small segment of the target population. This involved classic design thinking techniques, so field visits, interviews, observations, and immersions. 

Dr. Dwain

We had an agent that we called State Shifter; didn't have a human name. It was called a "state shifter." And the program we called "de-stress your stress," like, de-stress your stress. 

It comprised of videos; we just had a classic text-to-speech robot voice with text. There was nothing fancy about this at all. It was very basic, and they were literally, if they watched a video, it was text and on screen, that is it, with a voice. They received the intervention through a structured sequence of these text messages, which were delivered by a WhatsApp messenger. So the agent just used rule-based responses via a decision tree structure.

So pretty standard, and the messages themselves just focused on delivering the video links, answering questions, and encouraging adherence because this is a very difficult thing to do to get people to continue doing it. We had one group of people, owner-managers, and they received this intervention. We didn't have multiple groups. The amount of people that actually went through this intervention was only 14 people. 

The objectives that we were trying to meet here were to verify whether or not this intervention was successful in producing changes in the stress mindset. Could we change them from a stress-is-debilitating mindset to a stress-is-enhancing mindset? Could the agent do that? The second thing was to establish whether or not this was going to be associated with improvements in productivity and work performance. So we wanted to see whether or not that was going to happen. 

And then the second big objective was to just determine whether or not it was actually acceptable amongst the sample of entrepreneurs or owner-managers, right? The intervention was delivered directly to the participants via their mobile phones, but in the context of their day-to-day lives. So imagine you getting a jam, it would be you'd just be going about your day.

So it was quite interesting to see whether or not this series of basic videos and text messages from an agent would be enough to hold their interest as they went about their daily lives. They got two videos each day, and it was for four days. They got one at 7:30 a.m. in the morning and one at 07:30 p.m. every day for four days. That's a big ask, really, for these owner-managers to get in there. The videos are only about eight minutes each, and they're only available online for a short period of time. 

So we were just bringing that in each time, and we had a bunch of measures that we had to measure. So we measured their stress mindset and their work performance. These are scales. They're just self-report scales where somebody will you just answer those? I asked you about one of the items just before the acceptability itself. We just captured that as the number of participants that requested ongoing support from the agent. So there was a bit in the last—I think it was the last video—where the agent said, "Hey, if you want me to help you out, let me know."

So we wanted to see whether or not there would be something that they'll be interested in doing. We also wanted to test how many messages that participants assess through WhatsApp. So how many did they actually go to? You can tell. We can measure that. 

And then the other thing we measured was a thing called the "friendship questionnaire" which has been used with digital humans before and really just helps us understand the degree to which participants feel that they've developed a relationship akin to a friendship with the agent.

Jam

Yeah, so let's just go back to delivery and having to design a few very level one chatbots for clients and for ourselves as well. So the decision tree, you just deliver those two videos twice a day, one in the morning, one at night. Was there anything in the design? And I'm kind of honing in on the conversation side of things. Was it just delivering it so it's more passive and for them to decide whether they have the time at 07:30 a.m. and what? An eight-minute video is quite long.

Dr. Dwain

And it was also intense too. It was full of academic knowledge because that was part of the intervention. So it was still based on the principles of a mindset intervention. So you can imagine these individuals having to engage with this 7:30 in the morning sharp and 7:30 at night. And we also built scarcity into it by saying the videos would be deleted in 2 hours. We wanted them to do it, and they're busy individuals.

Jam

No, I understand it. So was there any interaction at all? When I say interaction, not just playing the video play button and spending eight minutes watching the video. Were there any other questions or interactions?

Dr. Dwain

Yes, there were ongoing conversations. The chatbot would follow up with each individual afterwards and say, "Hey, how did you find that?" "Would you like to know something that wasn't in the video?" And then people would say, "Yes," "No," or "Maybe." And there was a decision tree there. And then people would often come back to it because, as you've probably experienced, right, people start to anthropomorphized the bot and have all sorts of relational questions and stuff. So yes, there were ongoing communications above and beyond the delivery of the videos.

Jam

It's a good thing I wasn't part of that experiment or the group because I'm a night owl. 07:30 A.m.. I would have probably always missed the first one because I'd be asleep by then. Anyway, is there any reason why you decided to use WhatsApp?

Dr. Dwain

Yeah, that was based on what we'd found in our initial field visits. It seemed like most of the owner-managers that we interacted with had WhatsApp. So I was meeting them where they're at, trying to make that barrier to engagement as low as possible for us.

Jam

When you design conversations, the question that you have to ask after they engage with a video or are not engaged in the video, "Is there any personality that they feel like they are talking to a person?"

Dr. Dwain

There was a strong personality designed into the agent itself. The agent actually makes it clear that it's a robot. I purposely had a slightly robotic voice built into it. There is some psychology behind it about why we would want to make it known and remind the individuals that it was a robot, or like a bot, and wasn't a human.

Jam

I'm really interested because this ties back to one of our past episodes. A lot of listeners are interested in whether a chatbot should have a personality or not. So I'm curious on what your thoughts are.

Dr. Dwain

Well, I certainly think that personality, or sometimes we call it, "ethos" or "character," that's inherent in a technology, should be there. I mean, humans will make their personality up even if it's not there. This is the core fundamental of the concept of anthropomorphism. We just can't help it. We'll project those onto it. But what you can do is actually control that a little bit more. So in human-robot interaction and human-agent interaction, we're not really that interested in bots and agents attempting to mimic humans so much. And if you think about the way in which they're commercially put together, they are often just trying to mimic a human, and that's okay. There's nothing wrong with that, particularly if it's in a call center and whatnot. But there are a couple of issues with that. So first off, there is the concept of vulnerabilities, right? 

There's a famous researcher person called Sherry Turkle. She talks a lot about the vulnerabilities that we humans have in regards to AIs and agents that are demonstrating that they're humans when they're not. And her concern is really around this idea of them being our best friends. She doesn't like that idea.

Is that actually a good idea, that they pretend to be our friends, to be something that it can't actually really be? And does that affect the way that humans make friendships? Not for or against that at all. How should we consider how we want to design our agents? 

And Cynthia Breazel, who's at MIT and leads the personalised robotics group, has said this for a really long time, which is that this concept of robots aren't humans; they're never going to be human. And the real power is in celebrating the difference between how they're not humans and what the relationship is like. So conversation, personality, voice—they're all great things, and we should continue to do that. But we should also consider: what does the agent provide that human beings can't? Like, where is the actual power, where is the service in it? And so, from the design space, it's a really interesting perspective.

Jam

This reminds me of the projects that we've done, however small or big they are, when we put some personality into trying to be not necessarily the best friend but human enough, and yes, we do tell them, "Look, we're a robot." You're not talking to a human being. As time passes, and when they're so used to how the robot or how the chatbot communicates with them, then people would say it didn't know that it was actually a chatbot, and they thought it was a real human being. So there was both a negative and positive effect. Some would say, "You lied to me." They feel really bad. And there were some that they really appreciated but also scared a little bit because a robot can actually be like a human being. Definitely something that we can talk about in another episode. I would love to know more about the results of this study.

Dr. Dwain

Okay. It was massive, like what we call an effect size, right? There was a mean change of, I think, something like 1.38 for stress mindset, which just basically means there was a really big difference between when they started it at what we call the "baseline" when they first showed up into the experiment and when they did their follow up measure. It was really big. Most of the people that showed up were on the negative side of "stress is debilitating." Every single person that was in the intervention had a shift, which is pretty interesting and quite rare. Very big change. There were corresponding differences again with work performance as well. Really huge effect sizes. And that shows us that the thing was quite effective. And the other probably most important piece about that was 14 out of the 14 participants —that's 100%—completed the intervention from start to finish. We didn't lose them.

Jam

Interesting.

Dr. Dwain

It's really interesting. And again, conversational design was a part of it. But coming back to the results—acceptability, there were twelve of the participants. So twelve out of the 14 requested ongoing engagement with the agent. So when the agent said, "Hey look, Jam, if you'd like me to help you out every now and then with this stuff or you can learn it yourself, that's cool." —did say stuff like this. Let me know; just go message me and WhatsApp and I'll sort you out. And so twelve of them did it. So that's pretty cool. That's really high as well. Really high. There was a relatively high degree of reporting the agent using their friendship scale. So there's a really weird scale. It's between twenty and one hundred, and the average or the mean was 81.9, which is really high. That means again, coming back to your point from before, to what extent did they perceive a relationship with the robots? So as far as we know, it's the first to provide evidence of the effectiveness of mobile messaging-based conversational AI-led stress mindset intervention. That's a heck of a mouthful, by the way, but that's exactly what it is.

Dr. Dwain

So that's the first, it's the first ever do it. And we can see that it's highly effective in changing individuals' mindset about stress, and that was also strongly associated with improved productivity and work performance as well, which means participants were engaged and they were adherent, which means they stuck with everything. And they also were overwhelmingly, I would say, positive in their evaluation of the robot by means of the friendship questionnaire. So it was pretty good. So it supports us conducting a larger scale experiment.

Jam

This just reminds me of: I think there was a study or there was a real product out in the market where in different audience obviously, it's for seniors, and they tend to be alone. And some of them actually wanted to end their life early if they could because there's no point. There is a mobile app or a chatbot who was their friend. This is a voice, not technically like your Alexa and Siri, and they were friends. And it's the same thing as your friendship scale. And I'm not surprised because once you start talking to them and get used to them, you feel like you've got a friend even though in your mind they're not real people. It's the same as having a helpline when you're stressed or depressed. It feels more comfortable to talk to a stranger than your partner or a loved one because there's no embarrassment and you don't feel the burden that you're needing help, and you're talking to a chatbot.

Dr. Dwain

This is where it comes back to what I was saying about the idea of how we express the power of the chatbot or the AI technology, as well as the conversational aspects that are human. When the agent was interacting with the humans, it spoke very quickly, saying something like, "I can't order your pizza and some other stuff if you need me to tell you what the weather is." I'm no good to you, basically. And then it would say, but you might be thinking, "Why would you even want to chat with a bot, let alone listen to one kind of thing?" And then it would say, "I have got the goods when it comes to utilising stress" or something like that. And then it would go on about how it had gone for countless months. It had undergone a gruelling schedule of eating content training. I've gobbled up lots of information on stress, and all this sort of carrying on and saying, "What's important for you to understand is that I've been able to synthesise a world of knowledge that most humans can't access, let alone experience." So it just basically means that I can deliver for you; that is what it would say.

So what I'm demonstrating here is that it was smarter than us, right? We're not necessarily smarter, but it's more appropriate. So even in any sort of situation, such as the one you were mentioning, where it's say a youth line, it's better than us because it's got access to lots of data, assuming it can produce that data and communicate it in a conversational friendly way.

Jam

When I was in Uni, I would tell my students, "Look, a chatbot quite simply starts out as a baby, and once you feed data into it depending on the interactions with real human beings, it starts to learn and learn." I mean, it's similar to us as human beings. Why would you talk to someone? Or let's say that's why you are an expert, because you've got all that experience all through these years versus someone who is just studying psychology. It's starting out as a ruthless scientist. It's kind of the same. But I know machines, obviously, are better in a way they can compute more data in a faster way.

Dr. Dwain

If you combine that, though, with what you said earlier, which was that there's a lack of judgment there because it's a machine, it's not like you're telling a friend, it could change the relationship. You never know if people don't want to do that. That's another aspect. So you can imagine a scoreboard there, and it's like, "What are these things really good at? What are they better than us at?" And how can we actually feature that a little bit more and tell the user that we do that in the actual conversational experience?

Jam

I do have another question. Let's say it becomes a product on the market that entrepreneurs like myself can use. Do you want it to become habit-forming so that it really helps the entrepreneur change their mindset, or not necessarily?

Dr. Dwain

Well, the mindset should be changed, essentially, right? That's the point. At the end of the intervention, you should have a different mindset, assuming you don't have the more effective mindset for dealing with stress. What's really powerful about the theory, and particularly the way that we are presenting it to the agent, is that it doesn't matter what the stressor is that you're experiencing, which is really cool, like it's not stressor dependent. It's just that you have a different outlook now. It's not saying that stresses should go and get more stressed and hey, it's all good and there is little sunshine out there. It's not, right? Sometimes things are really tough, but it's like, what's your outlook on this? How is it affecting you? How can you use that energy that you're feeling right now, because it's exactly the same thing. The stress you have when you have your first kiss with a partner and the stress you're feeling about giving a speech or something if you're not comfortable are very similar. They're the same thing once you've got the mindset. The question then is, "How do you keep the mindset?" So coming to your point about habits forming, and the answer is that we don't know.

We don't know how long. The research is so young. How long will you keep that mindset? Because the thing with mindsets is that they are shaped by media messages. Your day-to-day experiences, the people you interact with, the constant messaging that's constantly coming at you because—remember, it's out of your awareness most of the time. It's not like you walk around saying, "I've got this mindset." You don't.

Jam

Yeah.

Dr. Dwain

The technical term for it is "implicit belief." It's an implicit self-belief. The question is, "How can this be something I learned?" So in the intervention, the bot would teach people this three-step process for how to switch their mindset at will. Now, the problem is, of course, that the chances of people actually practicing that in real time are zero unless you're very dedicated to doing it, where you've got triggers and cues in the environment to tell you, "Hey, I need to be doing this thing at this time." This is the sort of essence of habit forming, but where the bot would say to people, hey, do you want to message me? Or how about I message you and let you know and remind you of the steps and can help guide you on that.

That's what people were responding to very heavily. It does have massive applications for habits.

Jam

For me, this is definitely something that an entrepreneur should have. I totally believe in the project. I hope that it will be on the market for an entrepreneur to actually use. That's all I can say. It will change a lot of lives. This is great. I love it. Thank you so much for your time.

Dr. Dwain

I really enjoyed the conversation. It was great. It was a good chat.

Jam

Thank you, Dwain, for all of that. Me being a nerd, I just absolutely loved the conversation. Is there anything like just a one or two-sentence takeaway for the listeners, for them to think about? Not necessarily around the research, could be around stress, could be around conversational AI.

Dr. Dwain

Yeah, like understanding that stress itself is not, by definition, inherently negative and can have enhancing outcomes as well. The way in which you conceptualize stress for yourself can have a lot to do with how you experience stress itself. And on the conversational AI front, I would encourage designers to start experimenting with expressing the machine and AI-type capabilities of the conversational AI, as well as the human aspects of things that we anthropomorphized, like the speech and the conversational aspects. Do both of those, and I think we're going to start to see a new era of relational artifacts.

Jam

Awesome. True art and science of conversation—that's what it's all about. Well, thank you again, Dwain, for your time. I would love to have you back as a guest.

Dr. Dwain

Thank you, Jam.

Jam

And thank you all for listening to this episode. There's so much to unpack, and I would love to know your thoughts and questions. You can leave a voice message on Anchor.FM, ask through the Wisdom app, where you'll find a link in the show notes, or comment on social. Hit that "follow" or "subscribe" button to be notified of the next episode in your favorite podcast app. Till the next episode. Thanks for listening to our nerdy thoughts, and remember to keep the conversation going.