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!
Dr. Vanessa Rios de Souza holds a Ph.D. in Food Science and has over 10 years of experience in R&D, consumer and sensory research across multiple food product categories and functions, with a strong background in food science and food processing. She has extensive experience in industrial, academic, and research settings.
Vanessa works as a Computational Sensory Science Consultant at Aigora and before her role there she worked as a professor, researcher and as an independent CPG consultant for several years in Brazil, where she is from. She has presented her work in many conferences and has published 4 book chapters and 65+ peer-reviewed papers throughout her career.
Transcript (Semi-automated, forgive typos!)
John: Vanessa, welcome to the show.
Vanessa: Thanks, John. Yeah, thanks for having me.
John: It's really a pleasure so I would say you have been a great blessing to our team, and I mean, it's always better when you hire somebody and then they turn out to be much better than you already been expected them to be. So we're very happy to have you on the team and maybe take our listeners through the journey of how you ended up at Aigora.
Vanessa: Okay, yeah. This is kind of interesting, pleasure to be here, John and very grateful to be here in Aigora's team. Well, right before I finish my PhD in Food Science, I got a position as a professor in a University in Brazil. So of course, besides teaching, I continued working with research, and eventually I started working as a CPG consultant, as you mentioned, with my former sensory professor and it was really like enriching experience because it was kind of my first contact with R&D and sensory and consumer research in a more practical way. But of course, as I had my job as a professor, I couldn't dedicate as much as I wanted. Anyhow, so it's kind of hard to explain how I ended up here at Aigora. What I know is one thing led to another, and somehow, like the destiny also played a role. So I came to Canada two years ago to work as a research scientist at the University of Wealth and because I wanted to broaden my experience and improve my knowledge and so on. And at some point through a simple LinkedIn connection, I got in contact with you, John, and we made this partnership that I'm grateful for to work. So it's funny, because to be honest, I've never thought about myself working this kind of role. I mean, in this computational sensory science, but I'm so glad that we somehow crossed our ways because it is something that I really enjoy. It's amazing opportunity to help and promote digital transformation with so many companies and to learn about their particularity and liking and effectively improve their lives.
John: That's right. And I think also it's a bit of a lesson is the importance of networking, right? Because you contact me on LinkedIn. I think maybe I sent you a connection request because I tried to connect with people in the field and you accepted it. And you said, oh, it sounds really interesting what you're doing on Aigora, and it turned out to be around the time that we were hiring. So we just had some LinkedIn messages, and it became apparent that you would be interested in doing something like we do here and here you are.
Vanessa: Yes. Exactly.
John: Networking is very important. The older I get, the more I think networking is extremely important.
Vanessa: Yeah, you're totally true and kind of I was more like an academia, and it's not that we don't use a lot LinkedIn in academia, maybe using more but I was not very used to LinkedIn. I was more like researchgate and other things. But then when I started looking for opportunities in the sensory field and so on, it was amazing. There are so many people available to share their thoughts and their lives and to share experience and to give you advice so it's really amazing. Yeah, that's really important.
John: I totally agree. It's about time you had to physically be in an area where things were happening and now that area is just online. So as long as you can get online, you can connect. Let's talk about your research interest then, can you just kind of take us through the scope of your research? I mean, what was your focus when you were in graduate school and then what were you working on when you're a professor? And then now we can talk about how especially with your work with the kind of digital and computational side of sensory, what are your interests now? So you can kind of take us through the arc of your...
Vanessa: Yeah, sure. Well, the type of research I worked on before joining Aigora, well, I worked a lot with sensory as a tool in product and process development. My background is in food science, so I worked a lot in the R&D process. So, for instance, I used to work with non-termotechnologies for food processing and sensory was crucial to understand the sensory profile and consumer response of a product to the conventional thermal treatment compared to the new emerging technology. So I worked a lot in this field, and sensory was crucial. Well, I was also very interested in and worked a lot with temporal methods like TDS, NTI, particularly for sugar and sodium replacement to fully understand the changes a replacement would cause in the product sensory profile and consumer response. And again, the use of sensory more like a tool in the R&D process. But also I also did some research like on new methods and new approaches, analyzing methodologies and panel profiles so more like focus on sensory itself. So that is a little bit about what I've worked before I join Aigora, I would say and you asked me what I've learned here right after I joined?
Vanessa: It's been really awesome. So, yeah, it's been, like only three months but for sure, I've learned a lot here at Aigora. I guess first and most importantly, I have to mention that, of course, I've learned R language, and this is critical for my role. I would say that I don't know, I spend maybe like half of my time programming R.
John: It is statistical programming.
Vanessa: Yes, and I was not used to that. Unfortunately, I have in mind that, well, there are so many sensory software, why do I need to struggle with R? And I had this mistake in thinking that R was very difficult and it was made for people with very good skills in programming and coding and somehow this completely changed my mind. I guess after I start working here and also because of this role, I have a good understanding of things I haven't worked on directly before, and I can see great value on that, like automated reporting and dashboards and I understand now how that works and machine learning and database. Everything we do here in Aigora is very interesting to me. But maybe one thing that I became particularly interested in is the automated reporting dashboard. Maybe because I'm more involved in this type of project. So yeah, it's just awesome like to write a script and tailor the dashboard exactly according to the client's needs and wishes in this process of starting from scratch and then seeing a fully customized dashboard is amazing. Like seeing how the client is satisfied with that, how it can save their time, it's just like priceless.
John: Yeah, I do take a lot of pride in that, too, myself, because I know when I was at the Institute for Perception, so many hours were spent making decks. Right? And it was a lot of the same work again and again, and that can be automated in an interactive way. It's an incredible. It's just like a whole new way of working, right? Not only do you save a lot of time, but then you can investigate new things.
John: So I definitely value that as well. So let's talk a little bit about you becoming a data scientist because I think that's really an inspirational story. I think there are a lot of people out there like you who have a background in something like food science or maybe social science, actually, my wife, Ruth who was recently on the show as well. Ruth, she's a psychologist and she has trained herself in data science. I think a big message people need to hear is that they can do it, right?
Vanessa: Yeah, apparently.
John: When did you start studying R and how did you learn it? What was your kind of path? Because you're actually quite proficient at this point.
Vanessa: Yeah, this is funny, too, I would say. So, well, I'll never forget the day you John texted me asking if I knew coding. And I was honest and I replied saying, I have zero experience on that. And at that moment I thought, okay, if he had anything in mind or a possible collaboration or partnership, I just killed it. But I was honest.
John: No, you should have to be honest. Everyone should always be honest.
Vanessa: But then you texted me back asking, would you be interested in learning? And I guess it's been kind of everything changed. I replied, like saying, yes, I am and I always be open and interested in learning new things. So I guess because of you, John, you were my greatest motivator because of this possibility of consolidating like this, like collaboration and this amazing partnership. I decided that I would learn R. Well, my journey started two months before joining Aigora which was last August I guess. So, basically, my journey was like I started studying R through a very good book. You recommended me called R for data science. So almost every day, like early in the morning before work or later in the night, I spent like one to 2 hours studying, just going through the book and doing the practice. I also watched a few videos and I read a few things on the internet about R and also GitHub which I had no idea what it was at that time. Well, so I joined Aigora three months ago, of course, I had some background on that but it was with the daily practice and the support of this amazing and highly skilled team that I made the biggest improvement. So it is true when people say that you learn more or you fix it what you have learned with practice. Well, of course, I still have a lot to learn. I want something that you become like an expert in a very short period of time. But I'm confident with R and yeah, it is possible. It's an interesting journey.
John: Yeah, and I think it's something people really should embrace because I got the book R for Data Science in maybe, I don't remember what year I got it in. It might have been something like 2017, something like that. And I had been involved with statistics for regulatory engagement. And we had these enormous data sets where we had hundreds of measures that were all very similar and they had to be computed again and again. There'd be little changes. It was a nightmare. And that's when I guess that was probably 2015, that I started to really learn R to deal with the statistics for regulatory engagement. And in 2017, I just remember Amazon recommended it to me. Just recommended, hey, you might like this book. They were very right. And it changed my life. I did exactly what you did. I went through it and I typed every line of code in the book. In fact, you're helping us now with the data science for sensory and consumer science book that Thierry Walsh and Julien Delarue and I are putting together and our advice is type every line of code, right? You can't learn a language by just reading it. You have to type it.
John: It's just like a regular language. If you wanted to learn, say Portuguese, right? You have to move to Brazil. That's the way to learn Portuguese. The best way to learn Portuguese is go somewhere where people are speaking it, right? If you can get into a job where you are writing code every day, right? You do some studying on your own, get into a job. You'll learn the language. Now, let's talk a little bit about what can you do? What new capability would you say programming has given…
Vanessa: Right, yeah. That's interesting. Like I mentioned John, like many people, I was afraid of R. As I said, I always had this mistake thinking that R was difficult. And I was always thinking, why are you wasting time struggling with R if I can use sensory software to do my statistical analysis and my plot and other than better later than ever to realize I was so wrong. Well, my thinking is once you learn R, that's it of course, it requires time and long term commitment. But what I realized was it gives you a word of possibilities, including full control of your data. You don't have that in normal software’s. And it gives you an extraordinary flexibility and ability to play with your data so you are under control, like you are free, not attached to what the software can offer you. And I guess once you make your own analysis, when you use R, you understand much better. Like what you're doing. It's different. If you use like a statistical software to plot a PCA, you are going to push a button and you have a PCA. If you do R, you're going to understand what is behind that like what is going on. So it's really awesome.
John: Yeah. There's a bunch of things you said there that are really interesting to kind of explore. One is, I think just being brave and putting yourself out there and learning something new does lead to freedom. I think fortune favors the brave. When you're brave, you put yourself out there, you eventually earn freedom, and you get that through the programming. The other thing you said that I really like is the control. I think when you're really in the middle of coding, maybe you have this feeling, too. I feel like I have my hands on the data. That's the way it feels. And I never have that feeling when it's statistical software when it's some other interface. When I'm coding, I feel like it's in my hands, and I can do whatever I want with it. Right? Anything I want to do, I can do. I can write it at the whiteboard, and then I can do it. Yeah, and then there's the idea which you just mentioned that by writing the code for the PCA and by having specify what the arguments are, you looked at the documentation, what do these variables mean? What are the arguments mean? Options? What do they mean? It's a lot closer to doing it by hand, right? When I was in college, I had to do analysis of variants by hand. They made us compute with the pencil, all those squares and add them up and it was actually helpful exercise. And I think sometimes coding is more like that than just pushing buttons and something happens. I think it forces you to think at a deeper level.
Vanessa: Yeah, definitely.
John: Okay. So let's talk now about your interest going forward. So you're still obviously interested in sensory, interested in temporal methods. You're interested in consumer experience so how do you see the kind of new technologies? It could be the automated reporting. It could be machine learning. Could be smart speakers. It could be the use of historical data. What new opportunities do you see new ways that you would approach your research going into this new year?
Vanessa: Yeah. Interesting. Well, I guess because of my job today, I became really interested in like automated reporting and machine learning so now I see sensory research with a different perspective, I would say. Like machine learning, for instance, it's something that it's really interesting. It's a life changing for the R&D process. It allows the sensory and R&D team to quickly understand the changes to be made in a product like formulation, for instance, to improve performance without having to go to the lab and kind of like making different prototypes, like crazy. After the development of several samples, like in the attempt, based on the directions you had from the model, you go to the lab, you are going to develop some prototypes, and then the model can play a role again and help you to filter the most potential candidates for sensory and for consumer testing mainly. So what I see that it's changing the sensory world like focusing on machine learning because we have so many new technologies is the fact that the machine learning is really important tool to support the R & D team. To make things faster, more efficiently and cheaper, of course. So I see that it's really awesome. Like the role the machine learning, the AI can play in the R & D process.
John: Yeah. I totally agree with that. I think what it is we have all this historical information and we've always had this problem, how are we going to use it? Well, use your historical data to build a model or maybe build several models, usually one model at each stage in the product development process. Right?
John: And then as you said, those models can kind of run both ways. You can optimize according to the model to get ideas for prototypes. But once you have real prototypes, you can run them through the model to get predictions.
John: And so I think this idea that we're just going to replace sensory panels with machines, that's not really accurate. I think it's more that our work will be greatly supported by these predictive models, and we'll just be a lot more efficient. We use the resources more effectively, be more sustainable. Right? Because you won't be doing as many tests, won't be going back. Easily you've got these stages. You won't be going back to the previous stages as often. So I see it as we're going to have all these little places where you plug in basically an assistant, an AI assistant that helps the product development and I think that you're absolutely right. Now, what about something else you've been working on our smart speakers? What do you see as far as opportunities there with voice activated technology? Is that something you're particularly interested in or is that something that you would say is still a little bit in the future?
Vanessa: No, I guess this is something that is definitely interesting. I see this technology is really important for specific situations because we know that the response from the consumer or whatever at the time that the person is experiencing something is much more real than this person. I don't know to complete like a questionnaire after some time. So they're like things, let's say if you are cooking, you can be describing the perception right away or we discussed that with a client and they use patients and sick people so they are not really in a position that they can be like feeling like a questionnaire. So this voice activation technology can be really important to collect data from this special group. I guess there are many applications that this can be really useful.
John: Yeah, right. I see that coming up again as we go into the metaverse, right? Where we have augmented reality, we're going to have headsets on. We need to collect data from people. You can do some things with hand movements and whatnot. But being able to talk is one of the easiest ways to convey information. And we haven't even talked about natural language processing, making sense of what people are saying. But a lot of work has been done to allow machines to better understand human speech, and we should be leveraging that in our research. I think it gives us all these opportunities. Now, I think we're still at the early days, right? I mean, we have at this point conducted many surveys for our clients. However, there are still some limitations. It's not the same as a human, but it definitely provides value. So that's something I'm personally very excited about. Now, we're actually starting to run kind of towards the end of the call here, but I do want to talk a little bit about the metaverse because I know that's something you really didn't have any experience with. Crypto currencies, NFTs, you've kind of gotten involved in that a little bit through some of our trading. How interested are you in the virtual reality, augmented reality, chemical senses in the Metaverse? How much are these topics interesting to you versus these are just kind of some ideas that are out there? Is that something you would like to contribute to at some point?
Vanessa: Yeah, I don't understand like in deep like metaverse and virtual world but of course, I see value on that, particularly the technology including virtual reality and augmented reality for sensory. I mean, it's more than proven that when someone is tasting a product in a lab, it's not the same as once you create a whole environment to make it closer to the reality and the technologies are here to help that. So we can have much more reliable data using those things, like virtual reality, augmented reality and yeah, of course, I guess it's the present. It's not something that we are talking about in the future anymore.
John: Yeah, that's right and one thing that I think is interesting. I always thought of virtual reality in sensory, at least until fairly recently, as a way to collect data. Right? But we're actually going to be going into a world. I mean, we're in an augmented world right now, right? You're in Canada. I'm here. We're just talking. It doesn't feel that different to me from if you are across the table. In fact, our whole team, I haven't met, I think, five or six of our team members in person. So we are already in this kind of augmented reality and in order for that experience to be complete, we're going to have to do more than just have sight and sound. We're going to have to bring the chemical senses in.
John: I think that's a very important problem. So I was always thinking, okay, we're going to use virtual reality to collect data in the lab that maybe is closer to being ecologically valid. Or we're going to use smart speakers to collect data in home that's closer to the lab because it's more controlled, that sort of thing. But increasingly, I see, the life will actually be in augmented reality, it'll be all the time in that virtual reality. And we need to understand the sensory experience in that space. We need to be able to deliver a realistic sensory experience in that space and there's going to be a huge role for sensory scientists in that area. That's why I think it's the best time in the history of the world to be a sensory scientist. There's a huge door that's opening.
John: All these new problems for us to work on, it's very exciting. Well, Vanessa, we're almost out of time so let's talk about how can people get in touch with you? What is the best, I know you're active on LinkedIn, would that be the best way for people to reach out to you?
Vanessa: Yeah, I would say it is the best way to get in touch.
John: Okay, and I've noticed that you post basically every day some interesting article I actually find I learned a lot from just following you on LinkedIn. Okay, so what advice would you have for young researcher, young scientists, to be academic, but someone 25, 30 starting out, what would be your advice?
Vanessa: Yeah. Well, my first advice is make connections and be open and flexible. Connect and talk to as many people as you can in your field and LinkedIn is an excellent tool so it helped me a lot. I met so many amazing people on LinkedIn and be open to new possibilities and opportunities. I mean, think you are a good fit only for opportunities related to things you are good at, because maybe you are even better in things you don't even know. You haven't given it a try so I guess this would be my first piece of advice, and I guess my second advice, my second piece of advice for young researchers in the sensory and consumer science field is to learn R language. It will change your life and learn, or at least try to understand a little bit about artificial intelligence and machine learning and computational analytics, because this is not even more like the future. This is like the present, right? Yeah, this would be my advice for young researchers.
John: Okay, well, I think anyone be wise to take advice. You've really been incredible blessing to our team. You're incredibly talented person, Vanessa, and we're really lucky to have you. I can't believe how much you get done. You're the one of the most productive people I've ever met. Anyway, thank you very much for being on the show and thank you for being part of Aigora.
Vanessa: Thank you very much, John.
John: Okay, bye Vanessa.
Vanessa: Bye bye.
John: Okay, that's it. I 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!