• A
  • A
  • A
  • ABC
  • ABC
  • ABC
  • А
  • А
  • А
  • А
  • А
Regular version of the site

Patrons' Brain Activity while Viewing Menu Items Can Predict Restaurant Sales

HSE neuroscientists aided in the discovery


In Brief

Assumption: Neuroscience has yet to reach a stage where it can serve practical marketing purposes.

In fact: Data on individuals’ brain activity can help predict next year's consumer behaviour.

More Details

A positive correlation has been discovered between the neural activity in the nucleus accumbens of study participants as they were presented with photographs of dishes from the menu of the Chaihona No. 1 restaurant chain and the sales performance of said dishes in the chain's establishments. This finding was made by a team of researchers of the HSE Institute for Cognitive Neuroscience, Lomonosov Moscow State University, and ETH Zurich (Switzerland).

Each of the 22 study participants was presented with a series of colour photographs featuring 78 dishes. The results suggest that peaks of neuronal activity in the brains of a limited group of subjects can be used to predict the consumer preferences of a broader population over an extended period. The study findings have been published in PLoS One.

The study was financed by a megagrant received by the International Laboratory of Social Neurobiology (075-15-2022-1037) as part of the Science and Universities National Project.

What Is It All About?

To date, numerous studies have highlighted the correlation between brain activity in various regions and consumer preferences and purchasing decisions. One study used functional MRI (fMRI) to predict the popularity of songs over a span of three years following the scanning process, based on the activity observed in reward-related regions of the subjects' brains, particularly the orbitofrontal cortex and ventral striatum, while they were listening to these songs. In other experiments, electroencephalography (EEG) was employed to predict preferences for television content, gauge the popularity of YouTube videos, and forecast the choice of products marketed on websites and the commercial success of movie trailers.

Nonetheless, fMRI remains the preferred method for neuroforecasting research due to its exceptional sensitivity in capturing the activity of subcortical structures within the brain. Previous fMRI neuroforecasting studies have successfully used brain activity to predict microlending performance, the effectiveness of advertising, internet video popularity, viral marketing success, and article forwards, among other applications.

According to the 'partial scaling' theory, neurocognitive processes in a small sample of subjects can effectively forecast the aggregate choice of a larger independent group of people, such as consumers of a certain product. Worldwide, pioneering neuroforecasting studies of aggregate choices have focused on the key region of affective processing, the nucleus accumbens (ventral striatum), according to Vasily Klyucharev, one of the study authors. Among other functions, this region is responsible for evaluating the anticipated pleasure to be derived from observed objects or events.

For instance, one study revealed that while the neural activity in both the nucleus accumbens and the frontal cortex could equally forecast individual choices, only bursts of activity in the nucleus accumbens, rich in the neurotransmitter dopamine, effectively predicted the aggregate consumer choices. Therefore, in their recent study, the HSE researchers also focused on the potential of neuroforecasting by observing the activity in the ventral striatum.

How Was It Studied?

First, the researchers recruited 372 potential participants via social media. The subjects had to meet the following inclusion criteria: no history of psychotropic drug usage, no substance abuse within the past month, and no history of neurological disorders. Furthermore, to align with the restaurant chain's target audience, the subjects needed to fall within the age range of 20 to 40, possess high-grade home appliances, own a car or real estate, and have spent a specific average amount during at least one visit to the restaurant chain within the past three months. As a result, 22 volunteers were ultimately selected to participate in the fMRI experiment.

During the fMRI session, which lasted approximately 35 minutes, each subject was presented with colour photographs of dishes from the restaurant's new menu. In total, 78 items from different food categories were shown, including main courses, appetisers, and desserts. The restaurant's offerings included European, Middle Eastern, and Russian cuisine. During each trial, the participants first viewed a photograph of the dish for five seconds, followed by its name and information on its contents for another five seconds. At the end of each trial, the participants were asked to indicate whether or not they would like to consume the dish after the experiment.

To simulate a real-life restaurant visit, the subjects were instructed to abstain from eating for three hours before the session. They were also informed that following the experiments, they would receive a complimentary voucher to enjoy one of the randomly selected dishes they had chosen during the study. Overall, the participants liked 79.5% of the dishes.

After the fMRI session, the participants were asked to rate the same food items on a five-point Likert scale, ranging from 1 ('not at all') to 5 ('very much'), by responding to the question 'Do you like it?'. They were also asked 'Are you satisfied with the price of this dish?' (from 1 to 5) and 'Have you ever tried this dish?' ('yes' or 'no').

What Was Found?

Analysis of the experimental data confirms that brain activity in the ventral striatum can indeed forecast the annual sales of a restaurant chain. Increased neural activity was observed in the brain region of interest when the subjects were viewing photographs of their preferred dishes. According to the authors, there was a pronounced release of dopamine in the right portion of the nucleus accumbens.

The choice of dishes made by the fMRI study participants showed a strong correlation with the annual sales of the restaurant chain. However, as the authors explain, global sales are affected by numerous factors that are hard to control. Nevertheless, the study results suggest that data on neural activity in specific brain regions of interest (such as the ventral striatum) from a small group of subjects can help predict the consumer behaviour of large populations.

Why Does It Matter?

The number of neuroforecasting studies focused on consumer behaviour remains limited. 'As far as we know, this study is the first of its kind,' the researchers note. Based on its results, restaurant business owners can potentially enhance the menu of their establishments, thereby stimulating sales. However, we should not overlook the significance of traditional methods of marketing research. Combining neuromarketing with a regular questionnaire survey can yield more comprehensive insights into consumer behaviour.

Study authors:

Andrew Kislov, Postgraduate Student, HSE School of Psychology; Head of Emonomy , a neuromarketing startup

Anna Shestakova, Director, HSE Institute for Cognitive Neuroscience

Vadim Ushakov, Institute for Advanced Brain Studies, Lomonosov Moscow State University

Mario Martinez-Saito, Research Fellow, HSE Institute for Cognitive Neuroscience

Valeria Beliaeva, ETH Zurich, Switzerland

Olga Savelo, HSE Institute for Cognitive Neuroscience

Alexey Vasilchuk, co-founder of Chayhona No. 1 restaurant chain

Vasily Klucharev, Leading Research Fellow, HSE Institute for Cognitive Neuroscience

Author: Marina Selina, July 04, 2023