Nazimah Hamid - Something More Complex
Welcome to "AigoraCast", conversations with industry experts on how new technologies are transforming sensory and consumer science!
Nazimah Hamid is a Professor of Food Science at Auckland University of Technology. Her research encompasses how processing techniques can influence physical, chemical, and flavour qualities of food. Her expertise in sensory and flavour science uses a combination of sensory and instrumental flavour analysis to examine and predict the relationships between food composition, sensory perception, and flavour of a variety of processed and minimally processed foods. She has worked with a variety of foods - from sea urchin roe, black foot abalone, clams, apricots, and cherries in New Zealand to ‘Durian’ (commonly referred to as the King of Fruits) and jackfruit in Malaysia, and earlier raspberries in Scotland. She also researches the role of auditory cues on flavour and the consumer perception of food.
Transcript (Semi-automated, forgive typos!)
John: Nazimah, welcome to the show.
Nazimah: Thank you for inviting me, John.
John: It's a pleasure. I'm really excited to talk to you. I mean, we definitely share an interest in predictive modeling, so I want to get into that for sure. And a few other topics that are interesting is sound obviously is also shared interest. So let's actually start with, for our listeners who maybe haven't had a chance to meet you yet or are not as familiar with you and your background, I think it would be great to start with your story of how you got, you know when you first got interested in sensory and the path you've taken on to your professorship.
Nazimah: Okay, so the story started in Malaysia, which is where I was born and raised. I eventually got a scholarship to study in the U.K. for my first degree in food science at the University of Nottingham. And then I went on to do my master's and PhD at Strathclyde University in Glasgow, Scotland. And when I was there I was first introduced to sensory science and I did my PhD there with Alistair Patterson. They always used instrumental analysis, coupled with sensory evaluation to look at a variety of whiskeys then. But I wasn't too keen on work on whiskeys. And so I opted to work on raspberries and my life has changed since because it was a bit difficult changing from chemistry, having done more food chemistry, and doing sensory science. It was really difficult for me at first, but I thoroughly enjoyed it. I went back to Malaysia and started my academic career there, but I didn't get an opportunity to teach much sensory science because, in the 1990s, it wasn't a big thing in Malaysia. You know, if you taught food science at the university, it would be more food chemistry, food processing, food microbiology. Sensory was just a small part of it. I eventually moved to New Zealand and the position I took up at the University of Otago, which is the oldest university in New Zealand, I managed to get a position as a sensory science lecturer. And so I came back to it when I arrived in New Zealand, and I've not looked back since. I was in Otago for a while and after that, I moved up to Auckland University of Technology. And this is the youngest university in New Zealand. It's only 20 years old. And when I first arrived, there wasn't even a sensory lab. So we had a sensory evaluation paper, but no sensory lab. It was a big thing. I only got a lab established two years after arriving and it's been good after that. I have a small space, but my interest in flavor chemistry and sensory science is something that I've been able to do with a lot of research here. Now, having facilities that were not here before.
John: So you've been setting up the program there, then?
John: And you have colleagues that you work with there as well? Or are you the main sensory researcher?
Nazimah: Yeah, I'm the main sensory person here. When I was establishing the lab, it was quite good because Hilda and actually came to New Zealand for a talk and she gave me some tips on what I should do with the space. So she's been very good in all that good advice.
John: Yeah. Now, that's, wonderful. Okay, so let's go ahead and dive into a little bit of the predictive modeling because I think that's really interesting. And I think it would be, I mean, first off, when did you first get into the predictive model? Is that really from early in your career? Is that a more recent addition to your research?
Nazimah: That was really during my PhD and that was in 1993.
John: Oh, I see.
Nazimah: Yes. John Piggott, who was my supervisor, was a University of Cambridge graduate in chemical engineering. And you can see where the data mining techniques came from, a lot of it in, in his work on whiskeys. And so we try doing that in raspberries and I was actually formulating a juice using various kinds of enzymes and looking at the flavor profile and the sensory characteristics. At that time, I was introduced to you know, the PCA, which was something very new but very common now. So it started from there. So my basics when I did my PhD and it's still in tools, but it's much more powerful now. What you can do now is far more powerful than what you could do in the 1990s.
John: And were you mainly interested in or are you mainly interested in predicting sensory properties from sort of chemical and analytic measurements? What are the particular models that you're most interested in?
Nazimah: Yeah, at a moment, it's quite interesting to know how processing can influence a flavor. Yeah, but just knowing what's happening with flavor does not tell you what's happening to the chemical changes in the food. And so I find that it's really important to look at those chemical changes and then relate that to sensory because if we know that the processing is changing certain chemical constituents, you have an idea on how you could optimize the processing to improve the quality.
John: Right. Now, that's good. So it isn't just a computational approach, you also have a kind of first principles approach that you have knowledge of. That the processing is impacting the chemical makeup and you're using that in your modeling, that you've got this pathway that you are investigating. Yeah, well, that's fascinating. I think that sort of thing, that's what we have to do because we don't typically have so much data that we can just overwhelm problems with a combination of data and computational power. So, yeah, I think that definitely is the right path that you're on. So can we talk a little bit about a term that I actually didn't know before you and I started talking to each other, and so you have to help me with pronunciation, make sure I get it, right? It sounds very interesting, which is this, Flavoromics? Is that correct?
Nazimah: Yes, that's correct. That probably was coined in the US so there's been some work at the University of Minnesota. And I know a professor of flavor chemistry there, Renesas. And yes, he looks at flavoromics and there are universities in the US that do that as well. And it comes from the feel of metabolomics. But you are looking at identifying chemical compounds that could potentially help you identify flavor drivers. So it's an untargeted approach. So rather than looking at just sugars, organic acids, or volatile components you can use a lot of highly sensitive instruments like LC-MS, GC-MS, and even NMR to look at metabolites and look at the entire chemical makeup of the food system rather than you know, chemical compounds that we know can contribute to flavor, because if you think about flavor, it's a process that, you know, when you look at the chemistry, you can get flavors from flavor precursors. Breakdowns of effects. The breakdowns of protein, and not just in meat products. In fact, a lot of your volatile constituents in fruits, you know, broken down. I mean the flavor precursors have got fat and protein origins as well. So I think it's really interesting. Because we have a lot better techniques, multivariate techniques, then maybe in the 1990s, you can have software that can deal with large datasets. Because you need to process this large amount of data. And so that will involve techniques like, you know, definitely chemometrics, multivariate statistics and wants data mining techniques which you're familiar with. And that's why when I work with people on this, it's not just me myself. I would want to work with a statistician. I work with a psychologist. It's all important that you work with others. So I've been quite fortunate in that sense because, you know, that's helped me a lot in being able to explain why those sensory changes occur in food and how you can go about in terms of processing to help optimize the flavor.
John: That's fascinating. That's right. Well, first off, there's a bunch of points you just made that is important, which is, for one thing, I think most of the really exciting work that's happening in the world right now is kind of multidisciplinary where you have teams of people, different backgrounds, working together to do exciting things. And I think that with sound you just mentioned, that's another area that's coming into our world that I guess is really exciting. But I think it's definitely worth it because so often, you know, I mean, you see this again and again in sensory where we have a problem And other people actually if we can just realize that the problem we have can be reformulated a little bit differently and other people have studied it extensively, like design, right? There's the whole field of design and user experience. They've actually done a lot of really interesting work, which is directly applicable to us and vice versa. We can help people who are working on user experience and design because we've learned a lot in our own, you know, often food sensory approach. I mean, there's food and non-food. It's funny to think about the fact that in sensory, the big categories are typically food and non-food, as if like non-food, which includes everything other than food. The best in a category, non-food. It's just good enough. Anyway, it's kind of a funny thing. It'd be like if the countries were New Zealand and not New Zealand or something like that. But we have learned a lot and I think that we have some of our knowledge is specific to taste and smell, but some of it is very generalizable. So it's quite interesting. So let's talk a little bit about no kind of go all over the place, but I'm excited to talk to you about a number of things. So your research on sound and the impact of the interaction, because I think, you know, just like the multidisciplinary research is interesting, I think the multimodal research that a lot of the action where you think about technology is allowing us to control maybe not just the taste of things, but also the sound that our consumers are experiencing, maybe the sound of the product, but also maybe the sounds that are happening in the background as increasingly, you know, people will have devices on them that allow us to control the sonic experience. So can you talk a little bit about your research on sound and some of the key insights that you've come across in that work?
Nazimah: Okay, so the work on sound really started at my present university, and the reason why I went into sound was because my PhD student then Kevin, whom you have interviewed and he was my student 10 years ago at this university. He loved music and it was interesting in my first sensory class, I had a student who owned a gelato shop on Queen Street, which is our high street equivalent in Auckland, and that's how we came together. I had very little research funding and I looked at what students were interested in and with this student who has a gelato shop, it's one of the most popular shops in Auckland now. He provides that the gelato free. And that's why the works started on chocolate gelato.
Nazimah: Yes. And so I also roped in a psychologist and he specializes in psychoacoustics, Daniel and that's how it started coming up. So it was something affordable that we could carry out without a big research grant and a student who was passionate about music and that's how it all started. And it was quite interesting to see what sort of research was out there. They were looking at musical instruments, the pitch, the tempo, the loudness of sounds. And really, no one was really looking at a musical piece, you know, a piece of music which is what we hear. We don't listen to just a particular instrument in music which is what people listen to now. It's something more complex. And we said, okay, let's do something on music because the only researchers who worked on music was, I think Professor Seo from University of Arkansas and so we looked at their work as well, and they were the only players who looked at the effect of different musical genres. So we decided we'll take a different approach rather than saying, if you listen to classical music, your food will be sweet, you know? I mean, that's what people always thought. You know, always good music. But it was quite good because my psychologist friend, he isn't a classical music fan. He said, no, that's not music to my ears. I love heavy metal. And yes, he loves heavy metal. It doesn't sound quite right because if you play classical music, I don't like it. I might not even like the food. And so that's how we looked at the valance, how wearing music in terms of valence effect that the food perception. So we had a really positive effect because we found that if people liked the music and, you know, they actually had an increase in blessedness as well. And the story went on and with Kevin's PhD finalized with music and we found that a lot of it was due to emotions.
John: Right, which can vary from person to person. Yeah, it's so easy to fall into the trap of thinking that something is hold universally for a population, you know. Even in early sensory research, I mean, I think the reason why a lot of the mainstream products in the United States are very bland is that the market research in the food science research in the 70s, for example, or in the 60s, was looking at average liking as a metric. And if you want to maximize average liking, the main thing is to not offend anybody. To create a product that nobody hates. That will do okay on average liking. Now, if you want to delight, if you've got two groups right and you want to have two products, one of which delights each group, then each of those products won't have an average liking score. That's as good as the one that's just in the middle that doesn't offend anybody, but those two products are going to be a lot more interesting. So I think that what you've done here with the research on sound has been bringing awareness of the individual differences and how they're important that people don't like the same music. I mean, I think it's the one that is an appeal to the idea that music is universal in some way and everybody responds the same way. But, you know, chemical senses are also very old and we don't all respond the same way to the same chemicals, right? So why would we respond in the same way to the same sound? So that's quite interesting. Yeah, that's really fascinating. Okay, well, that's great. So now let's talk a little bit if we can go back to the predictive modeling, something I didn't get a chance to fully ask you about what's the sorts of problems where you feel like you've had very good success and then other situations where you feel like it's more challenging? Where you feel like, you know, if you were going to recommend somebody get into predictive modeling, what sort of problems would you recommend that they pursue versus what sort of problems would you recommend that they be careful about that they might need a lot of data or they might need very experimentally controlled conditions or something like that to mine very good quality data. When you look back in your career, what have been the most successful projects and what have been the products where you feel like, okay, that was probably too complex of a situation to really make headway with the model.
Nazimah: That's a really interesting question because you've seen that I work with a wide variety of foods. I started off with meat foods when I looked at it and I started with looking at the effect of non-thermal processing. And that was really, really challenging. I'll start with the challenging one first. And if you were to do an experiment like that with processing, you'd need at least 12 cows. You need a lot of animals because the results are so variable and so if you have a few, it's not going to work.
Nazimah: And then you use predictive modeling. So you need a good number of samples. And with animal meat, it's really highly variable because with cows it's fine. You know what they're eating. So if they've got a diet that control, you will be able to do that predictive modeling in terms of sensory and flavor. However, it's interesting because I did some work on black food abalone, which is a New Zealand power. That's in Mauri and you have this animal, which is essentially a snail, but it's very you know, it's anything. So, when you don't have that control on what the animal eats, it's really challenging. And so that's why I moved to fruits, back to fruits, because from the tree and it's no problem. You don't need to slaughter 12 animals and make sure they eat the same diet. With plants, I think it's easier because you have that farm, you can source the fruit from the same farm. It's growing on the same type of soil. And so then, you know, it will work better. That's what I found when it comes to predictive modeling. So it's really, really tricky the type of food. You know, that you work on.
John: Right. That's fascinating. Yeah, that the amount of control, if you don't want variability, that's not in some sense. So, yeah, that's a very good way to think about it. Yeah, okay, that's very helpful. Okay, anything else that comes to mind as far as we look across the range of categories?
Nazimah: I guess those are some of them. So I've had success on with fruits than I did with animals, though even I've seen some work where they've looked at wine. Yeah, because it's controlled. Something that you can control better and you'll be able to carry out this sort of you know research. This flavoromics research.
John: Right. And I would add to that I think that whatever insights you gain when they're put into practice, it needs to be that the product that's being produced can be produced in a controlled way. Because maybe with this abalone, you could do some research where you control what they eat, etcetera. But it wouldn't be that valuable because, in real life, people aren't going to do that. And so the insights you might learn in a very controlled experiment with the snails, it may not actually be that useful to people who are in the industry because they're not going to have that level of control. It's not going to be, you know. So yeah, that's a really good point.
Nazimah: It's also because when you eat things like sea urchin, roe, and abalone, people eat it raw. It's not you know, it's processed. So you can't really change much. It's really dependent on what the animals are eating.
John: Right, yeah, that's very interesting. Okay, well, we are actually almost out of time, so this has been extremely helpful and really interesting. So I would like to get your advice for the young sensory researcher. I think that's always a good question, you know, to wrap up. I think the kind of typical listener to this podcast to someone in a sensory science department or maybe there could be a student, could be a graduate student, could be starting their career, what advice would you give that person right now? What can they do to you know, enjoy different metrics, but yes, what would your advice be?
Nazimah: Yeah, my advice is to be multidisciplinary and work with people from different disciplines, not just specializing in sensory, but understanding the chemistry of flavor and you will be able to do more things. You'll be able to think differently. And I think that's what we should be training, more sensory and flavor scientists rather than having a sensory scientist and a flavor scientist just get someone with some knowledge in between who can talk to those two people. I think that will, you know, be good for the future sensory scientist. And of course, do more statistics as well. Yeah. Get into, you know, use more all this great software that can do all these metrics, multivariate statistics, data mining techniques, rather than just sticking to the normal, you know, statistics that people use to end the lifestyle because you'll have a bigger set of data to analyze in future.
John: So really embrace data science. I mean, it sounds like that you're talking about using different software. I mean, yeah, that's a topic for another day. I could ask you a whole series of other questions. So, this has been great, Nazimah. How can somebody get in touch with you? Someone who listens to the show, maybe they want to apply to work in your lab or they want to apply as a student or they just have questions, what would be the best way for someone to follow up?
Nazimah: You can either contact me on LinkedIn or my email address at the university.
John: Okay, and we'll put the university website and your LinkedIn. So it's been great, Nazimah. Any other comments or questions before we wrap up?
Nazimah: No, but thank you for giving me this opportunity to share my experiences with you.
John: It's been totally my pleasure. Meeting people like you is why I'm doing this podcast is of so much fun for me so it's always a pleasure.
Nazimah: Yeah, you've given me good insights, too. And if not for covid, we wouldn't be talking today because I'm not really technology savvy. And I didn't realize you could reach out to more people when in lockdown.
John: Now, it is true. It is kind of a funny side effect. Yeah, but that's a good thing. I mean, there are some silver linings to the whole covid thing. There are some things we've learned that I think have been good. And this is one, connecting with you, you're on the other side of the planet. It's crazy.
John: Different day.
Nazimah: There's a population of five million here. Anyway, yeah, thanks, John.
John: Okay, thank you very much, Nazimah. 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!