Short CommunicationEEG correlates of perceived food product similarity in a cross-modal taste-visual task
Introduction
Nowadays, healthy eating trends are substantially influencing the choice of consumers. According to Nielsen research (2016) up to 70–80% of the world population make their food choices taking into account various healthy food attributes such as organic origin (absence of GMOs, artificial flavorings and colorants), low levels of cholesterol, salt, sugar and fat. Following this trend, food manufacturers, including sweet producers, are trying to include in their product lines items that have these healthy attributes (or at least are believed to have them). This often leads to the emergence of new innovative products made from unconventional ingredients or/and using innovative technologies.
When manufacturers launch such new products, they face questions regarding product positioning and potential consumers’ perception of the product. The most common tools used in traditional marketing to understand consumer perception of food products are hall-tests, surveys, observations, interviews and focus groups. However these approaches require a large sample to be used in order to ensure the reliability of the results, making such investigations costly (Wright, 2005). At the same time, data collection and processing using traditional marketing methods are also highly prone to researcher’s subjectivity (Wright, 2005), leading to low results reliability. Furthermore, previous studies have shown that quite often consumers are not able to express their preferences consciously and explicitly (Stasi, 2018), suggesting why results from such studies may fail to explain real consumer behavior.
Neuromarketing approaches, which aim at recording and interpreting product-relevant brain activity using neurophysiological techniques like EEG (electroencephalography) or fMRI (functional magnetic resonance imaging), may help to overcome these limitations and provide traditional marketing techniques with new complementary information on true consumers’ preferences (Ariely and Berns, 2010, Vlăsceanu, 2014). However, to our knowledge, there is no proposed neuromarker that could be used to assess perceived similarity between innovative and familiar food products. Thus, the purpose of the present study was to suggest a neuromarketing approach that may be reliably applied for investigation of perceived similarity between different food products on relatively small sample sizes. We focused here on EEG since it is widely-used in neuromarketing investigations thanks to its non-invasiveness, high-temporal resolution, portability and relatively low cost.
One of the possible approaches may be based on the Stroop-effect that could be observed in a cross-modal taste-visual task when there is a mismatch between product visual appearance and its taste. Multiple previous EEG studies have shown that processing of incongruent vs congruent stimuli consistently evokes a negative EEG wave with an amplitude peak at ~300–500 ms after the stimulus onset. This particular neuronal activity is often referred to as N400 wave and it was first described by Kutas and Hillyard (1980) during processing of semantically incongruent sentences. Subsequent studies have shown that similar N400-like potentials could be observed for other meaningful nonlinguistic stimuli (Duncan, 2009). For example, with a similar paradigm Xiao (2014) investigated the correspondence of taste and visual appearance of acidic and sweet foods. In congruent trials, the product images displayed on the screen corresponded to the taste experienced by the participants (i.e. an image of ice cream with the taste of sugar). In non-congruent trials, the visual stimulus did not match the taste stimulus (i.e. an image of a lemon with a taste of sugar). In agreement with their expectations, the researchers observed that a mismatch between taste and image was accompanied by a negative deflection amplitude between 430 and 620 ms after the stimulus onset, which they called ND430-620. Another important study of Pergola, Foroni, Mengotti, Argiris, and Rumiati (2017) has also shown that N400-like components can be efficiently used for testing food perception. In their study participants were primed with either sensory (i.e. “It tastes sweet.”) or functional (i.e. “It is suitable for a wedding meal.”) food description followed by presentation of either a natural or transformed food product. The authors have found that N400-like amplitude was significantly larger for transformed food than for natural food when stimuli were presented after sensory primes, and the opposite results were observed for functional primes. Together these findings suggest that N400-like potential may be a good candidate for testing incongruence of different food attributes and may help to reveal similar/dissimilar products.
Another potential approach for evaluating food products similarity may be based on measuring the power of evoked gamma oscillation (30–80 Hz). Synchronization of neural discharges in the gamma range is thought to ensure a distributed processing of information in various brain areas to form a common consistent perception of an object (Basar-Eroglu et al., 1996, Tallon-Baudry and Bertrand, 2000). Presentation of semantically congruent visuo-auditory stimulation has been reported to enhance the gamma-band activity (Schneider et al., 2008, Senkowski et al., 2007, Yuval-Greenberg and Deouell, 2007). For example, in the work of Yuval-Greenberg and Deouell (2007) higher synchronization of the evoked gamma waves were associated with the presentation of congruent images and vocalizations of animals, and incongruent trials were characterized by lower synchronization. Thus, we decided to test whether the power of the evoked gamma waves may also help to distinguish between congruent and incongruent stimuli in the taste-visual cross-modal task. To the best of our knowledge, it is the first study investigating evoked gamma oscillations in the cross-modal information processing involving taste.
Section snippets
Test product
Multi-cereal candies (MCC) with creamy and chocolate flavours were considered as test products because our preliminary analysis of the customers feedback has revealed that this product is perceived rather ambiguously. Some consumers indeed perceive it as a candy, others as cookies, granola, cereal min-bars or waffles. Although consumers believe that MCC healthier than regular sweets or candies, they often noted that they are nevertheless very rich in calories (200–400 kcal/100 g) and may not be
Hall test
The results of the Hall test have shown that participants equally liked both creamy and chocolate MCC. During the “Open Eyes” testing the mean liking scores were 3.96 ± 1.1 and 3.92 ± 0.9 correspondingly (uncorrected p > 0.82). However, under “Blind” condition testing the liking scores tended to be higher (uncorrected p = 0.082) for both types of MCCs with creamy flavour (mean liking score = 4.42 ± 0.67) prefered (p < 0.03) to chocolate one (mean liking score = 4.06 ± 0.80).
Under “Open Eyes”
Discussion
In this work we considered two potential neuromarkers to measure perceived similarity of food products: the amplitude of N400-like ERP and the power of induced gamma oscillations during 400–600 ms period after stimulus presentation. Previous physiological studies on cross-modal information processing (Qiu et al., 2006, Rebai et al., 1997), including food perception (Pergola et al., 2017, Xiao et al., 2014) suggested that presentation of incongruent stimuli is accompanied by an increase in the
Conclusions
In this work, we have suggested a novel approach to estimate perceived product similarity based on the power of induced gamma oscillations following presentation of congruent or incongruent visual stimuli. To the best of our knowledge, it is the first attempt to investigate evoked gamma oscillations in the taste-visual cross-modal paradigm. The results obtained using such an approach may be compared to traditional hall-tests but, unlike hall-tests, do not require large subject samples, do not
Authors contribution
SK designed experiments, contributed to data collection, analyzed the results and wrote the manuscript. MD contributed to the experimental design, collected experimental data, analyzed the results and wrote the manuscript.
Data availability
Experimental data will be available on request to the corresponding author.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgments
The authors greatly appreciate the precious comments on this work from Irina Shafranskaya, A/Prof. at the Higher School of Economics (Russia) and would like to thank all volunteers who participated in this study.
References (21)
Gamma-band responses in the brain: A short review of psychophysiological correlates and functional significance
International Journal of Psychophysiology
(1996)Dissociable contributions of prefrontal and parietal cortices to response selection
Neuroimage
(2002)Event-related potentials in clinical research: Guidelines for eliciting, recording, and quantifying mismatch negativity, P300, and N400
Clinical Neurophysiology
(2009)- et al.
Comparison of the N300 and N400 ERPs to picture stimuli in congruent and incongruent contexts
Clinical Neurophysiology
(2002) - et al.
A neural signature of food semantics is associated with body-mass index
Biological Psychology
(2017) - et al.
Brain mechanism of Stroop interference effect in Chinese characters
Brain Research
(2006) Differences in health and taste attitudes and reported behaviour among Finnish, Dutch and British consumers: A cross-national validation of the Health and Taste Attitude Scales (HTAS)
Appetite
(2001)- et al.
Enhanced EEG gamma-band activity reflects multisensory semantic matching in visual-to-auditory object priming
NeuroImage
(2008) - et al.
Good times for multisensory integration: Effects of the precision of temporal synchrony as revealed by gamma-band oscillations
Neuropsychologia
(2007) Neuromarketing empirical approaches and food choice: A systematic review
Food Research International
(2018)
Cited by (12)
Brain responses and connectivity to visual meal compositions: An EEG investigation into food liking
2023, Food Quality and PreferenceEmerging biometric methodologies for human behaviour measurement in applied sensory and consumer science
2023, Digital Sensory Science: Applications in New Product DevelopmentInfluence of the crossmodal congruency between color and flavor on product evaluations: Evidence from behavioral and oscillatory brain responses
2022, Food Quality and PreferenceCitation Excerpt :In recent years, the EEG technology has been increasingly used in food-related research (Carbine et al., 2018). Even though it is challenging to identify specific emotional constructs purely through the EEG signals, both general emotional activation and many cognitive processes can be examined by using this measure (Niedziela & Ambroze, 2021), such as perception (Domracheva & Kulikova, 2020), attention (Carbine, Larson et al., 2017), working memory (Kumar, Higgs, Rutters, & Humphreys, 2016; Marmolejo-Ramos et al., 2015), and inhibitory control (Carbinem, Christensen, et al., 2017; Kong, Zhang, Chen, & Liu, 2015). One common way to analyze the EEG data is to conduct time-domain analyses in which post-stimuli potentials across different trials were averaged for each electrode.
Implicit and explicit measures of the sensory and hedonic analysis of beer: The role of tasting expertise
2022, Food Research InternationalCitation Excerpt :Instruments for measuring such processes can also be divided into explicit, including self-reported measures at a certain point after the sensory or hedonic experience, or implicit, which are based on continuous registration of physiological measures, such as brain activity. Even if explicit measures are quick to gather and easy to use and analyze, they can also be cognitively biased, and it is for this reason that implicit measures have recently attracted attention in food science (Lagast, Gellynck, Schouteten, De Herdt, & De Steur, 2017) and neuromarketing (e.g., Domracheva & Kulikova, 2020; Walsh, Duncan, Bell, O’Keefe, & Gallagher, 2017). The main goal of the present study was to use both types of measures to evaluate differences in the explicit (sensory and quality judgments) and implicit (brain activity) sensory and hedonic assessment of beers, as well as to explore the strength of the relationship between such measures depending on beer tasting expertise (for recent reviews regarding beverage testing expertise, as well as implicit and explicit methods of assessing emotions elicited by food, see e.g., Honoré-Chedozeau, Desmas, Ballester, Parr, & Chollet, 2019, and De Wijk & Noldus, 2021; Lagast et al., 2017, respectively).
Public stereotypes of recycled water end uses with different human contact: Evidence from event-related potential (ERP)
2021, Resources, Conservation and RecyclingCitation Excerpt :The N400 component of the ERP was chosen to reflect the stereotypes of recycled water end uses. The N400 component was labeled as such because it is a relative negativity peaking around 400 ms (300–500 ms) (Domracheva and Kulikova, 2020) over centro-parietal sites, with a slightly right hemisphere bias (Kutas and Federmeier, 2011). The N400 component is always observed in a lexical decision task.