Kevin Kantono - Open-Minded
Welcome to "AigoraCast", conversations with industry experts on how new technologies are transforming sensory and consumer science!
AigoraCast is available on Apple Podcasts, Stitcher, Google Podcasts, Spotify, PodCast Republic, Pandora, and Amazon Music. Remember to subscribe, and please leave a positive review if you like what you hear!
Kevin Kantono is a Sensory Scientist in Arla based in the Arla Innovation Center in Aarhus, Denmark. Kevin graduated with his doctorate degree from Auckland University of Technology in New Zealand and now is the principal sensory scientist for Arla's business transformation program Calcium. He’s also leading Arla’s sensory and consumer science research program in collaboration with local and international universities.
Transcript (Semi-automated, forgive typos!)
John: Kevin, welcome to the show.
Kevin: Thanks, John.
John: Thank you for being here. It's really a pleasure. I have to say. You're someone who I really considered to be a kindred spirit. I think that we have many overlapping interests. And I think the challenge will be to try to condense it down to 30 minute call so that we can cover a lot of topics I like to talk to you about. So, for starters, I think it would be interesting for our listeners to hear something about your background and how you got to be where you are. I think it's everyone's story is kind of unique and sensory. So if you can kind of take us through the history of your career as a sensory scientist.
Kevin: Yeah, definitely. Well, as everyone else, I started my degree in Auckland University of Technology (AUT). I did my doctorate and looking at the influence of sounds essentially oil in this particular music as a tool in delivering emotions to change taste and flavors of food. During that day, it was back in 2015 & 2016 and that's when I met the first few prospects out of understanding this as a modulating factor of food. So we decided to also hop on the boat and investigative as well.
Kevin: On top of that, we use the latest methods, obviously, in sensory science it was TDS. It was the first few things as well that came up. So we're interested to hop to see the temporal effects because we believe that music is temporal stimulus. So naturally, we wanted to see the temporal perception of food. We also obviously went to the emotion, but as well, because that was a trending. But then, we looked at catus, essence profile and on and on. Luckily enough, my professor is actually a professor in Psychoacoustics. So it's the field of psychology and acoustics and looking at the influence of emotion and also I've got a lot of guidance from him and from there, actually that's when I discovered that we can actually measure emotions objectively as well using electrophysiology measurements. So with that all combine that become my PhD and then I graduated. And from that, I become a developer in Sanford Limited in New Zealand and then now I'm sensory scientist in Arla.
John: Fascinating. Yeah, now I think you're really in the right area because sound I think has been underappreciated within sensory. There's a lot of ways that I think I mean, obviously you've got things like the sound of packaging in this kind of thing. But I think as we go into this new augmented existence that I see as sort of the end point of the fourth industrial revolution, that there'll be this augmented reality that we just live in. That we're going to have a chance to provide kind of auditory augmentation to our products. I mean, there's like a lot of ways that sound comes up in sensory, but what are some of the key applications and the key ways that you see sound interacting with some of the more common place activities you might expect in sensory science, especially in the food industry?
Kevin: Yeah, well, I mean, essentially like there are some simpler things like what you can do is that, for example, the popping bottle. Like the sound of a popping bottle is actually one of the most important qualities, right? It's the freshness and premium quality and I think one company actually licensed their sounds effect to make sure that they have that consistent sound, something popping a bottle experience. And I think like you mentioned before, John, now that we're going to AR, one of the really really cool one which is always up here example, is it was an ice cream company where you can actually download them and then scanned the product and then there will be some violinist playing for two minutes while you wait until the temperature hopefully, so that you can get this optimum sensory experience after the ice cream is tempered a bit as to compare to just having a different freezer. So this is definitely lot of application that is definitely under look here. And I think now with the AR and VR and our smartphones are getting better and better, I think it will be easier to deliver such stimulus very soon.
John: Yeah, this was part of figures, too. I mean, if you have, let's say in Alexa scale or something like that, I definitely think you'll have this. You'll have music that you play while you're making a recipe or that you're, I mean, it's interesting, actually, I like to hear your thoughts on sound versus music, because music is a very structured form of sound, right? What are your thoughts on, okay let's just to give you open platform to talk about music, sound and emotion, what are the things you think are most interesting that people should be kind of aware of in this you know, in this space?
Kevin: It's an ongoing challenge, I think, because, for example, there's so much studies right now, there's been published, including mine obviously. We delivered the sound stimulus through the headset. And it's actually a really high quality headset, like I think it's some specific brand. And has noise cancelling effect and also it's a very controlled environment, right? While in reality, oftentimes when we play music, it will just come from our loudspeaker, maybe in the living room and hopefully we'll hear it in our kitchen a little bit more. So I think it's still an ongoing discussion which one is more effective, at least in our research group in New Zealand, we also have a PhD student as well looking at this ecological validity part where she will actually plan sound with the music to actually see what happens, like, for example, and reality, say, if you have this kitchen chopping sound in the background, will it play some music. Do you still have that emotional effect delivered to you or not? So we are testing it and in some way, I guess it's still an ongoing discussion. We don't necessarily know yet which one would do better, really.
John: Now, that's fascinating. I mean, it's really exciting because I think it's clear that sound is relevant, right? That we've got, like it's interesting to hear you talk about cues. There are auditory cues for behavior. I mean, the example would be a coffee shop, right? You've got the sound of a coffee shop. That's not really maybe there's coffee shop music, but really there's the sound. There's the ambient noise of the coffee shop, is the experience cued, right? And the same thing with the sound of chopping on a cutting board. Now, what I think is really interesting is that when you think about brand harmony, right? There's certain amount like there's a lot tons of work going on and trying to make sure that the taste and smell experience of a product is congruent with the brand promise. Now, how do you see sound coming into that? Do you see it as a situation where there's that sound associated with the actual product experience, or do you see it more that we're going to layer in sound that we're going to provide through apps or through skills or whatever, that will now supplement the user experience? Like how do you see sound coming into real life?
Kevin: I mean, I think it's going to be a mixture of both, right? It's either going to be like in a product related cues which can like promote crispiness or creaminess or whatever, and also kind of like maybe a customized music that kind of reinforce that brand identity. One of the cool things, actually, the Arla Foods has done, obviously, we can sell that with suspense at some point and then we ask for advice like how would we build a multi-sensory brand and not only just through sensory narratives, but also through sounds. One of the projects that kind of became a thing right now, I'm actually a marketing campaign as well done for UK consumers. We created this really cool ASMR video promotions. It's actually on Facebook as well, maybe if you're in the UK, maybe you might see it. It's like we rolled this cream cheese on this black pepper and then you can hear the sound of the crackling black pepper like when sticks is on the cheese to kind of promote that perception of like, you know, this is going to be like in a spicy, fiery, yet creamy kind of cream cheese. So that's really, really cool. So we did embark on that ASMR journey as well but that's more on product as compared to using a sound or music to kind of modulate or change your perception or reinforce the brand image.
John: Now, that's fascinating. And I guess there's all sorts of kind of subconscious cues that are being activated through the use of which I think brings us to something else you mentioned that I think is really interesting, the objective measurement of emotion. So could you talk a little bit about your research in that area and what have you found to be kind of most informative, most effective? What are some lessons that you've learned on that phase of your journey?
Kevin: Yeah, I mean, I guess I was lucky enough that my secondary supervisor is a professor psychoacoustics. And I think the field of psychology is definitely much older than sensory. They have done a lot of different things. They're really well versed with their scales. For us a scale, a scale which is a line scale. For them a scale could be, of course, now look at emoji’s and stuff which is really, really cool but they even have this kind of like mannequins, like doll images to show that a person is happy or sad and so on and so on. But in terms of the objective measurements, especially the electrophysiology, which is now a topic of a topic of discussion now in the sensory science world. What I what I found is that emotion is divided into three different dimensions. At least that’s a theory. It's called the valence arousal dominance theory. So valence corresponds to your positive emotions. Arousal is like your engagement or your attentiveness t stimulus and dominance is how much you can control the emotion itself. The good thing about this, electrophysiology measurements, at least from my study, what we found, that it's really, really good in differentiating emotion on just honourary, like in a two dimensional level on the valence and the arousal dimension. It doesn't necessarily tell me if it's going to be a good thing like if this does one experience happiness or surprise or anger like it doesn't go that deep, but at least it gives me an insight which direction they are at right now. So and particularly in my case, we use skin conductance. We use heart rate and we also use blood and impulse. And at some point as well, we did something really, really funny with the participants where we give them kind of like a belt around their chest and in a few minutes is supposed to take up their respiration rate. So we try noninvasive biosensors essentially, in my case.
John: And these instruments that you're using, maybe my understanding is that every day wearables are still pretty far away from being able to provide the level of resolution needed for this kind of research. So is this sort of specialty equipment that you have to use? I mean, how far are you to get back to the ecological, how far do you see that technology from being ready for use in the real world?
Kevin: I mean, there's a lot of, I think instruments and offerings as well right now, these days and back when I was in my PhD, I was actually using the Nexus, my media software. There was a Dutch company that created it. It's mainly geared for research because you can see that the machine was quite bulky. You couldn't really fit in the pocket and then you can't really move as well. So essentially, when I asked my participants to do the experiment, they will need to have their left hand on the bench and I put the pillow as well, to make them comfortable and they can't really move because any movement or anything is actually going to be registered as a measure. I'm completely aware, however, that there are companies that they have this wearable devices where you can, for example, they wanted to run a measurement, for example, some 2 channels or 4 channels that could work. But unfortunately, still, when we're moving, our brain is active. So the question is, is the activity in the brain is it because the participant is moving or actually interested in that and that there would be something interesting on the shelf? But we're still a bit far away. But there are definitely companies that are working on it and trying to shortcut this and hopefully meaningful insights from the data.
John: That's fascinating. So that's not necessarily just a hardware problem, right? That if you've got that detangled of different sources, I mean, it's really a question that I guess identifiability of information.
Kevin: And also like what biosensors should you choose? What measurement should you go for is accurate enough, is connected enough or do we need to get a 32 channel ECG measurements in the brain because I'm also aware that, for example, in default, they do the ECG measurements using a proper machine and everything, and Camilla, she's a sensory employee there and looked at it, her PhD was in that and unfortunately enough of the participants, they need to rest their head of this kind of like a heat map kind of thing and they can't move. They can only stick their tongue out and then have a drop of sugar or sweetener solutions because she was looking at that and see the brain activity. So I think we're a bit far away until maybe one day that we can manage to know and understand that, for example, okay, this is just the activity where people are moving or reading or distracted or anything that we can't pinpoint what is exactly that we're looking for in this biosensors. It's a great tool. It gives you a lot of data because it measures three hundred and twenty data points in each seconds or something with the latency of, I don't know, one or two milliseconds or some stuff like that. But still, that's a lot of things that needs to be done.
John: Right, that’s fascinating. I mean, that is one of the big gaps right now, right? Is you've got scientific control and precise measurements of ecological validity. I think knowledge is bridging that gap. Actually, I'd be interested to hear your thoughts on another gap which is structured versus unstructured data. Or you might think this quantitative versus qualitative. I mean, how interested are you in that? I mean, are your measurement sound like they're mainly quantitative? But have you also gone in the qualitative direction with some of your research in this area?
Kevin: No, I'm actually mainly more of a quant guy. I haven't actually had a lot of experience in the qual, although, you know I’m more of a quant guy, I think. I mean, of course, I was involved in some qualitative interview and with the coding exercise and thematic analysis using mdv But that's only it, really. So not a lot on the qualitative side but I am aware that it is actually an emerging field in sensory science as well, I mean, in EuroSense, I think Pascal presented his results on how he can actually engage consumers and analyze the data set just purely from you know free comments style analysis and that's a really, really cool because we don't constrict ourselves in that specific sensory attributes, but also understand what the consumers are. The words that they're using and how to describe as well.
John: Fascinating. Well, it was the focus groups on some of these sounds and emotion. That's probably happening too, you know, just like I mean, I tend to be on the quantitative side, too. I mean, it's my personal information. Okay let's talk a little bit about innovation, because I know you have a lot of thoughts on that topic. And I believe that that's one of your responsibilities at Arla, right? Is to help to support the innovation. So how do you what are the kind of important ideas that you think that people should be thinking about as you're trying to keep innovation going inside their companies?
Kevin: Well, I mean, like, you know, of course, it's always fun to come back to fundamental sensory research. And I mean, I am an academic myself. I come from an academic background. I jumped to industry after my PhD. So I'm still an academic insight. I do love fundamental sensory research and I like to see more of it, actually. But unfortunately, we are running for business. And business is sometimes there are specific needs and specific deadlines for whatever reason. However, I think that I can, I'm listening from the EuroSense 2020 workshop on the sense of innovation. I think I can kind of resonate with the researchers as well, that they are actually strong academia’s and it kind of looks back on, especially here in Arla Foods, majority of our sensory science research are very applied research that, you know, for example, we collaborated with the FoodBioSystems Doctoral Training program in the UK and looking at optimizing the nutrition and sensory qualities of plant-based beverages. Another one was also looking at flavor volatiles from plant based materials but it's really applied research where we can literally use the results when it's completed.
John: That's fascinating because hearing you talk about that, it's made me realize there's another kind of dichotomy that's getting rich, which is business versus academia. And I do think that technology is helping to bring that together, because we have I mean, obviously there's more communication thanks to technology. I mean, for one thing, you and I are here. You're in Denmark and I'm standing here in Richmond yet we're talking as if you're here. But there's more and more communication going on between the various areas. So, actually it's interesting, what do you think then that the people in business can learn from academia? What do you think are the kind of key ways that academics approach problems that people are inside businesses should be keeping, trying to keep alive, keep in their minds, you know, to go step beyond what would be the minimum business going versus academic principles that you would espouse?
Kevin: I mean, I think, you know, like from as academia, we are well trained to be very like we always planned things. We plan things nicely. We do it systematically. We have, let's say, a control group, a placebo group and you have the treatment group. And of course, you want to do it, a pilot first and then run a thorough analysis, then determine the number of sample size and everything. That's just how traditionally how academics work. Unfortunately, sometimes in business you will not have that luxury and unfortunately, I think especially sometimes in business, what I found is that people push R&D quite fast. And I completely understand it's because of either because of product launches or a specific needs and there are things that needs to be done as well. We need really fast, but I think it's just a gentle reminder, I guess, for everyone, especially working in industry. I mean, now I understand that deadlines is actually something quite difficult and oftentimes as scientists, we have to kind of short cut them. But the question is then how do we short cut them in the most sensible way we thought without abusing or without damaging our fundamental credibility, especially in sensory and consumer science. For example, in a very interesting note, I don't know if you probably remember, there was a series of paper that was published in 2018 to 2020. And it was a discussion on classical profiling and especially on the number of repetitions. Of course, we don't want to be the bottleneck in the whole innovation system and there are papers that traditionally we do sensory profiling triplicates and that's it. However, the papers, a series of papers suggest that if your panel is good enough, maybe we should only do two times. And then that's not a radical one that says no instead of two times to once. So that's actually really difficult, at least for me. When I read the paper, I'm like, okay, well, this is super interesting. And you can actually see all the responses and I think the paper from a lot of statisticians saying that, you know, an MFA with an adequate might not be sufficient. You have to consider the power statistics and stuff. And I think that's what I mean a sensory practitioners in the industry are forgetting that we are forgetting our fundamental so I think that's where we should kind of focus on. Yes, we can take shortcuts. Definitely, I would say yes to that. But whatever we take the shortcuts for our stakeholders, we have to make sure that fundamentally it's a correctional help.
John: Right and that's fascinating. And what role do you see, I mean something I'm really interested in is the use of historical data to try to, I mean, most large companies know a tremendous amount about the product categories, right? I mean you have relationships between formulation, analytic measurements, sensory profiles, and that's sometimes hard to pin down. Right? But in some situations, you can make reasonable predictions. If you have four power analysis, for example, maybe you have a lot of studies that you may be able to look back and inform a power analysis without needing to collect new data. How enthusiastic or pessimistic are you as far as the use of historical data to help with some of these challenges that you've mentioned?
Kevin: Well, I mean, you know, especially in relation to that power analysis, but and also dependent repetitions and all, I think it's actually really, really crucial for us to actually investigate that. So, for example, if I were in that situation where, okay, we don't really have any time to do triplicate, we can duplicate or we can only do once, even like one repetition, is it going to be okay? Or is the data going valid? Well, I mean, like you said, John, we can look at historical data and that's definitely something that a place that I will definitely start, because then I can see how the panel performs over time, for example, will that be good enough? Or how are they differentiating this product? Are they sensitive enough? Are we confident enough that one repetition is sufficient to differentiate or discriminate the products and the confidence that you are most comfortable with. So yes and of course, yes, we are looking at historical data as well, looking at, you know, like, for example, as a diagnostic, say, in the dairy, if they say, oh, this is an issue in taste, for example, and then we can actually track that data set because those samples come to us time to time, right? Either for other QA checks or any recipe changes whatsoever. So we can actually have a look and then and of course, the ambition is obviously to build a predictive model. Well, I'm not there yet. So that's probably something that we can collaborate in the future. But, you know, it's something that one of our ambition to actually try utilize the historical data more and to see what's going to happen in the future.
John: Yeah, I definitely agree with that. And so we're actually almost out of time here. So let me just ask you, I got a question for you still, what are the technologies you're most excited about over the next couple of years? The things that you think have the most promise for us within in a sensory and consumer science?
Kevin: Yeah, I think AR and VR is one of the coolest thing. I mean, my partner, he had a headset for Christmas and that was actually one of my first time I actually played like a VR in a horror game. So it's actually a really immersive because you can hear the 360 sound and everything 360. It's super cool. So that's definitely something that sensory scientists should be aware in the future, that it's going to be a technology where instead of we set up a shop and consumer comes to us, we can literally set up a shop for them and then they enter using that VR thing in the convenience of their home, for example, or you know, reinforcing sensory experiences, saying, you know, whenever you eat this particular cheese, we can create this perfect environment and then you eat it somehow. I still don't know how yet on how that VR headset can help us, because technically you're kind of blind it because you can't really see the food in front of you. But that's definitely something to be excited about and something to solve.
John: That's fascinating. Okay, so let's get your advice for young sensory scientists. Now, you are yourself a somewhat young scientist, but so I think just some advice for someone who just finished, say, their master's or PhD, maybe five years, but 10 years behind you in path. What would you encourage them to do? What would be the kind of the key things when you look back things that have helped you the most in your career that you would recommend to them?
Kevin: Yeah, I mean, you know, I think for most of us, you know, whoever is completing a PhD or have completed a PhD, we're always in a crossroad, right? Do I continue with academia pathway or do I want to go to the industry pathway? You know academia pathway usually entails all stock and then hopefully you'll get a tenured assistant professorship and then you become a full professor and then that's it, right? Well, the pathway then you get to do a lot of fun things like what I do right now. So I think the advice, say if I would to give one to the younger me is that, you know, keep it open-minded that you know, like you don't necessarily have to do or go to academia like a normal. After doing your PhD, you can actually explore outside. Be open minded. Ask what other experiences that you think about that experiences that you might get and make sure you don't regret if you say like because I'm the kind of person that I don't want to regret, like going to academia and never going to industry, right? So I’m the kind of person that like, you have to do it. Don't say no before you do it kind of thing. So definitely the advice for everyone I think was kind of completing your PhD, you know, if you ever have that doubt, like, you know, should I go to industry? Is it too stressful? Is it to deadliney? You know you'll never know. You got to go in there. If you like, you stay. If you don't like it, maybe after 2 or 3 years, then you can always switch back to academia.
John: Right. I agree with that and I have a saying that no productive work is ever wasted. So if you go into it and you do something interesting, then no matter what you do 3 or 4 years from now, you'll have been improved by the experience.
John: As long as you're improving, I think it doesn't matter. Well, Kevin, this has been great. So how can people get in touch with you? Maybe they want to apply for a job at Arla or they just have a question, what would be the best way for them to connect with you?
Kevin: Yeah, I think definitely, well, first of all, if you wanted to apply a job in Arla, you can always visit our career website. So you just Google Arla careers, something will come up in Google. I believe it will be the first part, on the first page and personally, if you wanted to contact me, please feel free to contact me on LinkedIn. That would be the best way to go for.
John: Okay, great. We'll put the link in the show notes. Any last words of wisdom for our audience?
Kevin: No, I think, you know, to be honest, like well I think yes, for sensory practitioners out there, I think we are, like what John has said back in EuroSense, that we are in the wave of, you know, kind of like another kind of revolution, I guess. Like, you know, we are getting better and our computer power is getting better. We are now looking at, you know starting with the big stuff and that is actually one of the biggest trend. So that's definitely something that we have to look out for. Be very worry, if you take shortcuts, please make sure that you don't just take shortcuts for the sake of it, but make sure that you have the fundamental knowledge behind it and make sure everything's okay. And I think that's all really, you know my kind of summary of the whole thing.
John: Yeah. Fascinating. Well, I'm really excited by your research, and I think you're going to be a really important player in the field for years to come. So I'm really happy to have this conversation with you, Kevin. So thanks a lot.
Kevin: Yeah, pleasure is mine.
John: Okay great. Thank you. Okay, that's it. Hope you enjoyed this conversation. If you did, please help us grow our audience by telling your friend about AigoraCast and leaving us a positive review on iTunes. Thanks.
That's it for now. If you'd like to receive email updates from Aigora, including weekly video recaps of our blog activity, click on the button below to join our email list. Thanks for stopping by!