The State of Sensory Science

Sensory scientists seek to measure, interpret, and understand how humans perceive the world around them—as well as the consumer products within it—through the lens of sight, smell, taste, touch, and hearing. This unique field brings together multiple disciplines, from neurobiology to anthropology, to help us better understand not only what human beings like and dislike, but also how differently we perceive various stimuli. As companies across the globe work to glean reliable, repeatable insights regarding new foods or consumer packaged goods, they rely on sensory and consumer scientists to guide everything from new product development to marketing. It is a field that is in a unique position to help industries, across verticals, become more human-centric, bringing key data about human perception to better inform every strategic business decision.

With the advent of new technologies and methodologies, sensory science continues to evolve as a discipline. In this virtual roundtable Q&A, Food Technology spoke with Danielle van Hout, president of Aigora, Artificial Intelligence for Sensory and Consumer Science; Dolores C. Oreskovich, director of innovation, consumer/sensory insights, for Nestlé Health Science; and Julien Delarue, associate professor of sensory and consumer science at the University of California, Davis, about the state of sensory science today. They discuss what originally sparked their interest in the sensory sciences, how the field has evolved over the past decade, and how emerging technologies like artificial intelligence/machine learning may change their work going forward. Their comments have been condensed for clarity and brevity.

Tell me a little about your current role.

Oreskovich: As the director of innovation, consumer/sensory insights, for Nestlé Health Science, I’m responsible for the strategic and tactical direction of the global team, providing insights for the creation and growth of new global products and brands. Our key function is spurring innovation—from early ideation to post-launch assessments that delight our consumers and patients.

Van Hout: Before Aigora, I worked almost 25 years at Unilever in research and development. My role there was about scoping out the whole area of digital and e-science methods to see what they could bring to Unilever’s sensory consumer science. We focused the knowledge and skills required to set up studies in ways so that we just don’t have value from them now but can also harvest knowledge from them for the future. That’s also what I do now in my current job.

Delarue: Most of my career has been in academia, with almost 20 years in France at AgroParisTech and now three years at UC Davis. I’ve also worked with industry because my research is very much applied. I’m interested in measuring consumer perceptions and preferences, using that information to make better, more sustainable food, and also find ways to make that information more actionable for both research and in new product development.

What inspired your interest in sensory science?

Delarue: I’m fascinated by the sense of smell and how we perceive flavors. Originally, I wanted to become a flavor scientist and did my PhD on flavor chemistry. After working for a bit as a flavor specialist, I discovered the realm of human perceptions. That’s what moved me over to the consumer side of research.

When you are in a company working in sensory, it builds a bridge between all other sources of data. It’s important for marketing, for product development, and also for quality.

Van Hout: For me, I was intrigued by the whole element of having humans act as instruments. I came by sensory science by accident. It was a course offered at school, but I wasn’t sure what it entailed. When I learned what it was about, I thought it sounded very interesting. At that point, I became very interested in what happens when you use people as instruments, either as representative consumers or as panelists. It’s not an easy thing to do. You need to understand how the brain works, how people make decisions, and how the senses come together to create a perception. Then, you need to figure out how to design studies in ways where you can get at the data that really measures what you are trying to learn. That remains a challenge.

I’d also add that when you are in a company working in sensory, it builds a bridge between all other sources of data. It’s important for marketing, for product development, and also for quality. The fact that sensory science is in this central position also makes it very interesting.

Oreskovich: Everyone comes into sensory science through different pathways, and my journey has been shaped by my unique experiences and interests. I was first interested in veterinary medicine, and my early education was in animal sciences, which sparked my interest in meat science. During my master’s program, I frequently conducted taste tests to explore the sensory properties of different cuts of meat, which inspired my interest in the intricate relationship between sensory perception and consumer behavior. This led me to pursue a PhD in food science and human nutrition, specializing in sensory science. This initial inspiration continues to guide my career as I strive to explore new and innovative ways to deliver strategic insights and make meaningful contributions to sensory and consumer sciences.

What are the biggest challenges involved in sensory science?

Delarue: How to measure things. One approach is to try and control everything we do—make our measurements work in a very standard way. We’re quite successful in doing this. It’s possible to have reliable, repeatable measurements when we do analytical sensory evaluation like descriptive analysis. But one must also consider that people’s perceptions are influenced by many different factors—factors that aren’t necessarily noise or interference because they are part of the everyday experience for the consumer. Sensory scientists need to take that into account in their work because a product may feel differently to consumers depending on the context. And if we miss that information, we can draw false conclusions about what we are measuring.

I’d also add that people are very different. We don’t always realize how different we are in terms of our perception. This can be quite challenging in a panel because we may have two panelists give different answers and it doesn’t mean one is correct and one is wrong. As a consequence of these individual differences, we need to look at many participants to get a correct picture and we need good data science, or even artificial intelligence to analyze the data. We are looking to measure subjective experience—and, of course, it is a challenge to measure subjectivity. And there is incredible diversity in the chemical senses like smell and taste.

A product may feel differently to consumers depending on the context. And if we miss that information, we can draw false conclusions about what we are measuring.

Van Hout: I completely agree. Sensory science relies on different statistical techniques and models that have been developed quite rapidly. It’s important to work on designing the best methods to answer the question you are trying to answer.

The other challenge is how to communicate those findings between different groups because, as sensory science is in this central position, you need to be able to speak the language of marketing or product development so your findings can be applied. Sensory people report to people who do not have that kind of background and that can make it difficult to make changes.

Oreskovich: Understanding human sensory perception poses challenges including subjectivity making it difficult to measure and quantify experiences accurately due to personal biases and individual differences. Complexities of perception involve intricate mechanisms and interactions between sensory information, cognition, and prior experiences. Variability across different contexts, situations, and individuals as well as factors that can impact perception (attention, expectations, and cultural influences) further complicates making generalizations across different populations. Addressing these challenges can enhance our understanding of sensory perception and consumer behavior.

How has the field changed over the years?

Oreskovich: It’s changed tremendously. Initially, we used either QDA, or Spectrum descriptive analysis methods while new methods like texture analysis and Free-Choice Profiling were being perfected. Early on we began to understand the impact of the testing environment on test results leading to more control of context and respondent bias. Temporal assessment, Drivers of Liking, Design of Experiments (DOE), category mapping, and formula optimization emerged as key areas of research, with many introducing modeling to predict liking. Researchers also began exploring consumer perception, leading to a broader understanding of the influence of attitudes, emotions, and behaviors on preferences. The field recognized Type 1 and Type 2 thinking, resulting in the development of implicit methods to uncover the importance of sensory cues to enhance the development of more relevant and preferred products.

Today, the field has expanded to include understanding consumer expectations and how to meet them. Holistic consumer centricity has become a paramount consideration, maximizing user product sensory experiences at every touchpoint from point of purchase to disposal. These changes reflect the dynamic nature of the field as researchers adapt to evolving consumer needs.

Delarue: Today, we are doing a lot of work regarding sustainability. That’s a change. I’d also say, on the scientific side, we’ve had to change the way we measure perceptions and preferences. In the early days, we’d try to get people in our studies to all give the same answer so we could say we had a reliable measurement. Now that we know so much more about how our senses work, we can look at each person as an individual. We no longer look for one preference, one perception; we look at many, and it’s changed everything we do.

We also see a change in where sensory science is used. It’s no longer just in the food or food science industry. Companies now see the value in sensory science in cosmetics, personal care, the car industry, textiles—all consumer goods, really. Today, in both France and the United States, about 40% of sensory science professionals work in industries outside food.

Van Hout: It’s a huge change. And when sensory science started being used more broadly across industries, we became more aware of its influence on the customer experience. With a car, of course, how it smells is not the most important thing about it. But it has an effect. So, you need to think about how you use that information. Because of that, in foods, we are also having to think a bit more [about] our findings, the dynamics of perceptions, and how to best use our data in product development.

What are the benefits of collaborations between basic science and commercial labs?

Delarue: In Europe, there is a lot more collaboration between industry and academia as compared to the United States. Universities in the states are very protective of intellectual property, which can make collaboration challenging. Whereas in Europe, governments subsidize these important collaborations. You need to keep a good balance between sensory science and its applications. It’s a win-win situation when you can build collaborations between industry and academia because it helps build important knowledge, which not only furthers the science but [also] improves the skills of industry employees. It’s also a good way to help program graduates find jobs. The more we manage to train graduate students with highly skilled sensory scientists, the more industrial programs benefit.

Van Hout: I had the opportunity to work in a company that was very willing to work with academics. During my time at Unilever, I worked closely with university colleagues to develop methods so that we could implement them in the company. From my perspective, both industry and academia have brains and skills, but, in industry, when you see there is a gap, some kind of problem you can’t resolve, you go to academia. You know they are not having to be quick and dirty. They are taking all the important variables into account so they can publish and validate a measure. When that’s successful, industry can then take it back into their labs. These collaborations are very useful, especially because sensory science isn’t just about food or commercial applications. It’s also being used in health industries and other vital businesses. That said, it is also important that academics have a relationship with industry so they can get insights into what to research and how to improve things. They may not know what to pursue if they don’t understand the business situation.

Oreskovich: We have a lab, and we move through many different iterations of consumer and sensory testing—learning and adjusting at an iterative speed. Academic labs are more likely spending time on fewer/bigger projects and generating new knowledge and understanding of fundamental principles, which can, in turn, benefit labs like ours. Labs like ours, on the other hand, provide practical insights and experiences that may help guide the research priorities of academic labs. Also, commercial labs can share their practical expertise in implementing and adapting these methodologies in real-world settings, optimizing them for commercial use.

How are advances in artificial intelligence/machine learning (AI/ML) changing the field?

Oreskovich: I see AI as a tool that is an enabler—similar to statistics. One use for AI is to predict consumer liking and preference outside the testing space of Design of Experiments (DOE) studies by allowing for more efficient and cost-effective product development without the need for additional sensory and consumer testing. AI algorithms can analyze large datasets and identify patterns and correlations, providing insights into consumer preferences. AI can also uncover complex relationships between sensory attributes and consumer liking that may not be immediately apparent. By leveraging machine learning techniques, AI models can identify subtle nuances and interactions that can influence consumer preferences, leading to more accurate predictions and targeted product improvements.

Van Hout: Over the past decade, we’ve explored how to best use these new technologies. It’s not as easy as just applying AI and digital to our projects because our data is quite different than other types of data. All the different people, all the different influences, it isn’t possible always to account for those with an algorithm. At Aigora, we’ve been working with companies to identify gaps and see where these tools can be best applied. We’ve also looked at how we can make sensory science data more compatible to AI/ML approaches.

In the end, you want to create a big knowledge base as a company. You want to learn from every study that you do. But how do you compare results from one country to another? How do you compare methods? At first, AI was all about the models. It still is. But we need to make sure the data we put into these models are good enough to get meaningful results. It all comes back to improving the data, making sure we are measuring the right things. Before AI/ML can really fulfill its promise, we need to ensure the information used in these models is better.

Delarue: AI is a real game changer and allows us to have access to more diverse sources of information. We can use text from open comments, speech, web scraping; we can deal with more varied data. But the main question now is what is the goal when you are using AI? What are we using it for? There have been some breakthroughs in using AI to predict, for example, the sense of smell out of the structure of different chemical compounds. That’s fascinating, but these models are limited to very specific situations.

But before we can really make use of these models, we need to really think about what we are trying to predict. Is it the average consumer or person—which we know does not exist? How can we account for the subjectivity in these people? The context? So, it would be nice to use AI to predict more complex things, but we need to overcome the challenge of managing all the diversity and complexity in an individual’s perception.

Sensory science is a multidisciplinary field that involves psychology, anthropology, sociology, statistics, and psychophysics.

What advice do you have for students and young professionals entering the sensory science field?

Delarue: I advise my students to broaden their perspective. Study humanities and social sciences. Understanding data science is absolutely critical. And then, for more general advice, I think it’s good to have diverse early professional experience in both industry and in academia. Even if you want to ultimately work in academia, it’s still important to understand how your results may be used in the industry and the overall challenges and stakes in the industry.

Van Hout: Building on that, I’d say it’s important to be open to new things. I’ve had many different roles and even the jobs I didn’t like so much, I’ve still learned so much. Try to keep developing yourself, not only in terms of your skills, but in terms of your interactions and how you build relationships with colleagues. I also agree with Julien about working in both industry and academia. I’ve always worked in industry. I even did my PhD while working, but I learned so much by being in an academic situation.

Oreskovich: Sensory science is a multidisciplinary field that involves psychology, anthropology, sociology, statistics, and psychophysics. It makes the field interesting, but it can also make it difficult to follow a single career path. Embrace your individuality by taking ownership of your career, defining your own success. Continuously learn and grow by staying updated with the latest research, attending conferences, and seeking out mentors. Be prepared for opportunities by staying updated with industry trends, developing a strong network, and improving your skills. Believe in yourself and have confidence in your abilities by taking calculated risks and embracing new challenges. A successful career in sensory and consumer sciences requires continuous learning, personal growth, and adaptability. Stay passionate, curious, and open to new opportunities, and you will maximize your career potential.ft