Method for characterizing emotional state caused by presentation of product

Abstract

The present disclosure discloses a method for characterizing an emotional state caused by presentation of a product. The invention relates to a method for characterizing sensory stimuli induced by the presentation of a product, comprising the steps of: subjecting a human subject to visual stimuli, then subjecting it to sensory stimuli to be characterized, and subjecting it to an image showing several emoticons; acquiring and processing a gaze tracking signal (ETC) during presentation of the image to determine a first emoticon (EP1) gazed for the longest duration, the first emoticon (EP1) determining a set of topics that may be experienced during the sensory stimulation; and acquiring and processing physiological measurement signals (FP1S, FP2S, AF7S, AF8S, F3S, F4S, SS1, SS2) during the visual and sensory stimuli to identify a subject experienced within the set of subjects.

Description

Technical Field

[0001] This invention relates to the characterization of a subject, and more particularly to the emotional state of a human subject induced by sensory stimuli perceived during the presentation of a product. Sensory stimuli can be visual, olfactory, tactile, auditory, or gustatory. Background Technology

[0002] Various methods have been proposed for assessing the perception of stimuli. Thus, self-assessment questionnaires have been proposed to identify emotional states (reference [1]). The use of physiological measurements to characterize emotional states is also known, but these do not allow for the distinction between positive and negative valence emotions (reference [2], [Ma et al., 2020]). For example, behavioral measurements of facial expression analysis have been proposed to distinguish facial responses based on six basic emotions (reference [3]). However, the number of such emotions remains very limited, with negative emotions being the dominant factor.

[0003] Another approach uses mood mapping to categorize the olfactory products being tested into emotional states (reference [4]). This categorization is based on participants' selection of the emotion term that best reflects the emotion evoked by the product from eight categories. According to another approach, participants must associate intensity (from 0 to 10) with each emotion term from a list provided for each product to be evaluated (reference [5]). Conscious nonverbal approaches based on the selection of emotion images have also been proposed, and methods using electroencephalography (EEG) to measure a person's state of relaxation have been described (reference [6]).

[0004] It is also noted (reference [7]) that the use of questionnaires, such as those described in reference [8], to analyze emotions related to taste perception is unconventional and particular, because consumers tend to cognitively associate terms with products without directly experiencing them (reference [9]). Emotions are described as short-term affective responses to stimuli with reinforcing potential. Therefore, research is needed to understand how consumers naturally express emotions related to smell or taste.

[0005] Furthermore, the emergence of the Internet and social media has provided new mediums for the nonverbal expression of emotions through emojis (i.e., pictographs symbolizing emotions). On the X platform (formerly Twitter), users mentioned mood and emotion in 25% of their tweets, using emojis more frequently than text. Emojis have proven to be not only a partial substitute for standard language (reference

[10] ), but also an automatic and easy way to convey emotions, regardless of gender or age group (references

[11] ,

[12] ,

[13] , and

[14] ). In South Korea, emojis are the third most widely used language after Korean and English (reference [7]). Thus, emojis have revolutionized our communication patterns to some extent. When people use emojis, they are expressing actions resulting from the emotions they are experiencing.

[0006] Electroencephalography (EEG) is a method commonly used by researchers to assess the perceived pleasure of human subjects. More specifically, frontal alpha asymmetry (FAA) has been associated with emotion processing and affective psychopathology. High levels of right cortical activity have been associated with negative stimuli and behaviors as well as withdrawal stimuli and behaviors, while increased left cortical activity has been associated with positive stimuli and behaviors as well as proximity stimuli and behaviors (reference

[15] ).

[0007] Eye-tracking technology is frequently used to assess cognitive and decision-making processes (references

[16] ,

[17] ,

[18] ). Eye tracking can be correlated with the selection of emotional emoticons, thereby revealing decision-making processes. Eye tracking is also commonly used in various emotion recognition methods (reference

[19] ).

[0008] To analyze emotions, it has been proposed to integrate three components: the physiological component that defines the intensity of the emotion, the subjective component that determines the nature of the emotion, and the behavioral component that identifies the actions that respond to the emotion. However, this approach is only effective if these three components can be integrated without complicating the experimental protocol (reference

[20] ).

[0009] Due to the dominance of conscious / subjective components, all the methods mentioned above for characterizing emotional states do not allow for a comprehensive analysis of emotional responses in human subjects. Furthermore, the objective physiological measurements used do not allow for the differentiation of feelings beyond the valence scale. For example, behavioral measurements using facial and vocal analysis only allow for the study of a limited number of emotions, which may be insufficient to achieve certain objectives.

[0010] To overcome these limitations, the applicant proposed a method for characterizing perceptual themes triggered by olfactory stimuli, as described in patent FR3129072 (or patent application US2023 / 148941). In this method, the subject's experience is revealed from interaction with a product in a given environment. This method examines the perception of a group of participants by integrating subjective and physiological components. Questionnaires are used to assess the subjective components, while the physiological components involve recording blood volume pulse (BVP) and measuring skin conductance level (SCL). Videos are then used to calibrate the participants' emotions in a virtual reality environment. These videos are presented to the participants, who are asked to rate them emotionally while obtaining physiological measurements. The emotional terms chosen by each participant are then associated with the physiological responses thus measured. This data is collected into a database, which is then used to identify the emotions each participant experiences when smelling the product. For this purpose, physiological responses measured during the olfactory perception of the product are compared with physiological responses calibrated using virtual reality videos. In this way, the emotion identified in response to the product is the emotion for which the physiological response obtained during calibration is closest to the physiological response measured under the presentation of the product.

[0011] The resulting database linked subjective and physiological responses to emotions, serving as the basis for grouping emotions into different domains. However, this approach has certain limitations. In particular, narrow groupings of opposing emotions related to pleasure can be observed. Furthermore, this approach does not intend to use direct physiological measurements, such as electroencephalography (EEG), to directly measure pleasure induced by the product. This limitation appears to impair the accuracy of classifying product perceptions into different domains, especially for emotions such as energy and fear. Physiological responses such as BVP amplitude, heart rate (HR), and SCL appear similar to certain opposing emotions due to the lack of pleasure measurements, making it difficult to distinguish these emotions.

[0012] The experimental protocol implemented in the applicant's proposed method also appears to constitute another limitation. In fact, in this protocol, participants were asked to close their eyes while the experimenter placed the bottle under their noses. This design restricts the application of the method to olfactory perception only, thus making it unsuitable for products that require tasting.

[0013] Therefore, the methods developed by the applicant as described above need to be adapted to overcome these limitations, and in particular to make them suitable for testing products such as food and beverages. Summary of the Invention

[0014] The implementation relates to a method for determining topics evoked by a product to be represented, comprising the steps of: (a) presenting the product to be represented to a human subject in a manner that subjectes the subject to sensory stimulation; (b) acquiring physiological measurement signals from the subject before and during the presentation of the product to be represented; (c) presenting the subject with an image displaying a set of emojis during or after the presentation of the product to be represented, the subject being asked to visually select an emoji from the image; (d) acquiring gaze-tracking signals from the subject during the presentation of the image; (e) processing the gaze-tracking signals to determine a first number of emojis for which the subject gazes for the longest duration, the emojis for which the subject gazes for the longest duration determining a set of topics that may be evoked in the subject due to the presentation of the product to be represented; and (f) processing the physiological measurement signals acquired before and during the presentation of the product to be represented to identify topics evoked in the subject among the determined set of topics.

[0015] By performing various physiological measurements and analyzing eye movements while viewing images displaying emojis, products can be characterized thematically by identifying themes that correspond to emotional states evoked by sensory stimuli from the product's presentation. These emotional states are not based on subjective responses provided by the subject, but rather on physiological measurements directly measuring feelings of pleasure and emotional intensity. These physiological measurements also allow for the assessment of such themes evoked by sensory stimuli, such as those occurring during food tasting, and can simultaneously affect multiple senses, including taste, smell, sight, and hearing. The use of emojis allows for overcoming cultural differences and language barriers that can vary from one subject to another. Furthermore, the identified themes can be of various types, such as mood types or even film genres.

[0016] According to one implementation, processing of physiological measurement results includes using a table listing topics, in which each topic is associated with an emoji number and a point defined by a pleasure score and an emotional intensity score.

[0017] The use of this topic table makes it easy to identify topics from physiological measurement results.

[0018] According to one embodiment, processing of physiological measurement results includes the following steps: determining a pleasure score and an emotional intensity score based on the physiological measurement results; identifying all topics in a table corresponding to a first determined emoticon number; and calculating the distance between each of the points defined by the determined pleasure score and emotional intensity score and the points associated with the topics identified in the table, with the minimum calculated distance corresponding to the topic identified by the subject.

[0019] Determining pleasure and emotional intensity scores based on physiological measurements allows for the establishment of a connection with the topic table and facilitates the identification of topics that can represent sensory stimuli that can act on different senses.

[0020] According to one implementation, when another emoji has been gazed at by the subject for at least 50% of the gaze time corresponding to the emoji number corresponding to the first emoji number with the longest gaze duration, the processing of the gaze tracking signal provides a second emoji number with the longest gaze duration.

[0021] The possibility of providing a second emoji allows for a more detailed representation of the themes experienced by human subjects triggered by sensory stimuli.

[0022] According to one embodiment, the acquisition and processing of physiological measurement signals includes the steps of: acquiring electroencephalogram (EEG) signals from electrodes placed on the head of a subject during the presentation of a product to be characterized, the electrodes including a right frontal lobe electrode pair and a left frontal lobe electrode pair; processing the EEG signals to determine a percentage deviation between the left and right electrodes of each of the frontal lobe electrode pairs; for each percentage deviation, determining a pleasure index based on whether the percentage deviation falls within a range of percentage values ​​specified for a pleasure index; and determining a pleasure score based on the pleasure index.

[0023] The use of electroencephalogram (EEG) signals advantageously enables the assessment of feelings of pleasure. Furthermore, the use of percentage ranges associated with pleasure index values ​​allows for easy assessment of these feelings of pleasure in a manner compatible with other physiological measurement processing methods.

[0024] According to one implementation, a pleasure score is determined by adding a pleasure index to a pleasure level, which is based on whether the pleasure value indicated by the subject falls within a range of pleasure values ​​defined for a set of pleasure values.

[0025] Combining several measurements with values ​​provided by the subjects allows for the consideration of subjective aspects, which enables a more precise determination of pleasure scores.

[0026] According to one embodiment, the acquisition and processing of physiological measurement signals includes the following steps: acquiring skin conductance activity (SCEA) signals from a skin conductance activity sensor placed on a subject during a rest period and during a presentation period of the product to be characterized; processing the SCEA signals to determine the average and maximum deviations of the subject's SCEA amplitude and heart rate values ​​between the rest period and the presentation period of the product to be characterized; determining the percentage deviation of the average SCEA value, the percentage deviation of the maximum deviation of the SCEA amplitude, and the percentage deviation of the heart rate value between the rest period and the presentation period of the product to be characterized; and for each percentage deviation of SCEA and heart rate, determining an emotion intensity index based on whether the percentage deviation falls within a set of percentage ranges determined for the percentage deviation; and determining an emotion intensity score based on the emotion intensity index.

[0027] Combined measurements of skin conductance and cardiac activity, along with comparisons of measurements taken before and during the sensory stimulus to be characterized, allow for the objective determination of the intensity of emotion felt by the subject. Similarly, the use of a range of values ​​facilitates the easy determination of index values ​​and emotion intensity scores.

[0028] According to one implementation, an emotion intensity score is determined by adding an emotion intensity index to an emotion intensity level, which is based on whether the emotion intensity value indicated by the subject falls within a range of emotion intensity values ​​in a set of emotion intensity ranges defined for emotion intensity values.

[0029] Combining multiple measurements with values ​​provided by the subjects allows for consideration of subjective aspects, thus enabling a more accurate determination of the emotional intensity score.

[0030] According to one implementation, the subject belongs to a group of participants, and the method includes the following steps: performing steps (a) to (f) for each participant in the group of participants to determine a topic for that participant; and determining a perceived topic representation of the product to be represented based on the topics determined for all participants in the group of participants.

[0031] Therefore, by applying this method to groups of participants, sensory stimuli can be characterized at the topic level, such as sensory stimuli caused by the perception of a product.

[0032] According to one implementation, the sensory representation results include a list of topics, in which each topic is associated with a topic score determined based on the number of times a topic is identified for a participant in a group of participants.

[0033] Defining a list of topics allows for the effective characterization of sensory stimuli, such as those arising from the perception of a product.

[0034] The implementation also relates to a system comprising: a computer, an eye-tracking device connected to the computer, a physiological parameter measuring device connected to the computer, and a display screen connected to the computer, wherein a processor is configured to implement the previously defined method.

[0035] According to one embodiment, the physiological parameter measuring device includes: a set of electroencephalogram (EEG) electrodes, and a device for measuring skin conductance and cardiac activity. Attached Figure Description

[0036] The invention will be better understood with the aid of the following description of exemplary embodiments with reference to the accompanying drawings, in which the same reference numerals correspond to structurally and / or functionally identical or similar elements.

[0037] Figure 1 This schematically illustrates a system for acquiring and processing physiological measurements and responses of human subjects according to one embodiment.

[0038] Figure 2 This refers to an image presented to a human subject according to one embodiment to determine the emotions felt by the human subject.

[0039] Figure 3 This is a schematic top view of a set of EEG sensors used in a method according to one embodiment.

[0040] Figure 4 This is a schematic front view of a device used to measure HR, SCL, and BVP in a method according to one embodiment.

[0041] Figure 5 The diagram schematically illustrates the functions of an acquisition and processing system according to one implementation.

[0042] Figure 6A , Figure 6B It is a representation illustrating the characterization of an emotional state caused by sensory stimulation, determined by an acquisition and processing system according to one embodiment. Detailed Implementation

[0043] Figure 1 An Acquisition and Processing System (APS) is illustrated for acquiring and processing physiological measurements and responses from a cohort of human subjects participating in product testing to characterize their perception of the product. The APS includes a computer PRC connected to a display DSP, a set of EEG sensors (EGS), a physiological parameter (PHS) allowing the measurement of physiological parameters such as skin conductance and cardiac activity, and an eye-tracking system (ETR).

[0044] The Eye Tracking (ETR) system is configured to monitor and analyze human eye movements. It includes an optical camera that projects near-infrared light onto the cornea, thus aiding in eye position detection. It provides measurements indicating where a person's gaze is focused at any given moment.

[0045] According to one implementation, the first type of measurement is performed by placing the test participant in front of the display DSP and displaying an image (e.g., on the screen). Figure 2 The image DEM shown is performed for a duration of, for example, 10 seconds. During the time the participant can observe the image DEM displayed on the screen DSP, the eye-tracking system ETR is active and acquires the path of the center point observed by the participant. The image DEM presentation represents emojis E1 to E15 representing the primary emotions selected based on their relevance to the type of product being tested. Figure 2 In the example, the image DEM displays approximately fifteen emojis, including:

[0046] Emoji E1 represents a neutral emotional state that can correspond to a range from boredom to some degree of pleasure.

[0047] The E2 emoticon represents emotional states of surprise ranging from very pleasant surprise to very unpleasant surprise, including a neutral state of distraction.

[0048] The E3 emoticon represents an emotional state ranging from ecstasy to astonishment.

[0049] Emoji E4 represents emotional states ranging from seeking adventure to escapism, a state of flight.

[0050] The E5 emoji represents emotional states ranging from excitement to frenzied passion.

[0051] The E6 emoji represents an emotional state ranging from a feeling of energy restoration to an energy boost.

[0052] The E7 emoji represents an emotional state ranging from enthusiasm to skepticism, encompassing interest.

[0053] The E8 emoji represents an emotional state ranging from disgust to boredom to fervent aversion.

[0054] The E9 emoji represents an unpleasant emotional state ranging from dissatisfaction to rejection.

[0055] Emoji E10 represents a range of emotional states, from melancholy to nostalgia.

[0056] The emoticon E11 represents an emotional state ranging from the feeling of committing a pleasurable crime to a mild, indulgent sense of criminal pleasure.

[0057] The E12 emoji represents a range of emotional states, from erotic to sophisticated sensuality.

[0058] The emoji E13 represents a range of emotional states from happiness to calm and relaxed.

[0059] The E14 emoticon represents a range of gentle emotional states, from intense pleasure to simple comfort.

[0060] The E5 emoji represents a range of emotional states of joy, from ecstasy to serene bliss.

[0061] Figure 3 A set of EEG sensors was represented in the form of a headset EGS, with twelve electrodes distributed on the headset EGS, each allowing the capture of electrical activity from one hundred neurons. The electrodes were placed on the headset to obtain information about the emotional state of the user wearing the headset on the head UHE. Thus, these electrodes included six left frontal electrodes F3, FP1, AF7 and right frontal electrodes F4, FP2, AF8, two left parietal electrodes P3 and right parietal electrodes P4, two left occipital-parietal electrodes PO7 and right occipital-parietal electrodes PO8, and two left occipital electrodes O1 and right occipital electrodes O2 (and a ground electrode GND at the central frontal lobe). The study highlighted the effectiveness of EEG signals in assessing mood indices. Specifically, these studies focused on asymmetric activity in the frontal lobe (references

[21] ,

[22] ,

[23] ). This asymmetry consisted of observed differences in activity between the right and left frontal lobes in the alpha band (8 Hz to 12 Hz). This difference may indicate, for example, a response to an approach or withdrawal motive. According to the frontal asymmetry theory of EEG, the FAA, an increase in activity in the right hemisphere indicates a withdrawal motive, while an increase in activity in the left hemisphere indicates an approach motive (references

[24] ,

[25] ). In addition, attention can be measured using the θ / α exponential algorithm. This measurement focuses on the parietal electrodes P3 and P4 for θ and the frontal electrodes F3 and F4 for α (reference

[25] ).

[0062] Figure 4 A device called PHS (Photoplethysmography) for measuring electrical skin activity (EDA) and cardiac activity is shown. The PHS device can be a photoplethysmometer that uses optical sensors to measure infrared light to detect volumetric changes in blood circulation at the fingertip. Figure 4In one example, the device PHS is in the form of a housing for use on the index and middle fingers of the user's hand, attached to the UHA. For example, the device PHS includes an infrared light sensor S1 placed on the index finger to measure skin electrical activity and an infrared light sensor S2 placed on the middle finger to measure heart activity. For example, measurements performed by the device PHS can be achieved at a sampling rate of approximately 64 samples per second.

[0063] Sensory information from smelling or tasting food is processed by the olfactory system. The signals are then sent to the brain to regulate the activity of the autonomic nervous system (ANS) (reference

[26] ). Heart rate is a recognized measure of ANS activity (references

[27] ,

[28] ,

[29] ,

[30] ). The heart is regulated by the ANS, particularly by its sympathetic (excitatory) and parasympathetic (inhibitory) branches. These branches exert regulatory control over heart rate by influencing the activity of the sinoatrial node (the heart's natural pacemaker). Measurements of cardiac activity are one component used to assess a user's emotional arousal through sympathetic nervous system activity (reference

[31] ).

[0064] Electrical skin activity (EDA) refers to the electrical conductance or potential of the skin (especially the hands), which can be measured as a reflex response to emotional stimuli. EDA is used as an electrophysiological indicator reflecting the activity of the sympathetic nervous system (SNS). This activity originates from sweat glands, which are responsible for secreting watery sweat and distributing it throughout the body, including the palms. While the primary function of these glands is thermoregulation, resulting in vasodilation of the skin in response to high temperatures, they are also innervated by two types of neurons. Neurons that use acetylcholine as a neurotransmitter are responsible for the electrical skin activity of sweat glands in response to SNS activation. In contrast, neurons that use adrenaline as a neurotransmitter cause vasodilation in response to elevated temperatures. Therefore, sweating without vasodilation, known as “emotional sweating,” occurs due to EDA in response to SNS activation (reference

[32] ).

[0065] According to Figure 5In the illustrated implementation, the APS system implements a method for characterizing emotional states induced by sensory stimuli based on physiological measurements obtained during a test session involving several participants, conducted according to a specific protocol. According to an example of the protocol, each test participant is placed in front of a system, such as the APS system, with an EGS headset on their head and a PHS device in one of their hands. A bottle containing a sample of the product to be tasted is placed in front of each participant. The eye-tracking system ETR is calibrated on a display DSP to track eye movements while maintaining a static head position. If necessary, the signals provided by the electrodes of the EGS headset and the PHS device are also checked and calibrated before the test begins. Thus, the test begins with the participant being asked to gaze at a crosshair displayed on the DSP screen for 30 seconds to calibrate the system ETR. This first 30-second step is followed by a 5-second instruction to grasp the first bottle of the product sample and taste its contents. The participant is then asked to hold their gaze for 10 seconds. An image DEM is then displayed for 10 seconds, allowing the participant to see an emoji that best corresponds to the sensation experienced while tasting the sample. This step is followed by a new fixation step, during which participants are then asked to fixate on their gaze for 10 seconds. During the final step, participants may be asked to complete a questionnaire.

[0066] Questionnaires may include the following questions:

[0067] 1. Which emoji would you like to choose?

[0068] 2. Compared to your expectations, how much pleasure value do you assign to the product (from 0, which indicates very unpleasantness, to 10, which indicates very pleasantness)?

[0069] 3. When you taste the product, how many emotional intensity values ​​do you assign to your emotional feelings (from 0, which indicates no emotional intensity, to 10, which indicates a very strong emotional intensity)?

[0070] Based on the pleasure score and emotional intensity score provided by each participant, the computer PRC function RP determines the pleasure NP and emotional intensity (EI) NEI levels as follows:

[0071] If the value is less than 3, the level is set to -2.

[0072] If the value is between 3 (inclusive) and 5 (exclusive), the rating is set to -1.

[0073] If the value is 5, the level is set to 0.

[0074] If the value is between 5 (excluding 5) and 7 (including 7), the level is set to 1.

[0075] If the value is greater than 7, the level is set to 2.

[0076] Computerized PRC determines several parameters based on data derived from measurement results. These parameters include two main ones: the Pleasure Index (PS) and the EI Index (EIS). The Pleasure Index (PS) considers the evaluation of pleasure and frontal alpha wave asymmetry (FAA) within a 10-second window after tasting the product. The EIEIS Index is determined based on the possible amplitude and average change in heart rate, EDA levels before and after tasting, between a 30-second window after tasting and a 10-second window before tasting.

[0077] For this purpose, throughout the testing process, signals FP1S, FP2S, AF7S, AF8S, F3S, and F4S from electrodes FP1, FP2, AF7, AF8, F3, and F4 of the headset EGS, signals SS1 and SS2 from sensors S1 and S2 of the device PHS, and signal ETC from the eye-tracking system ETR were recorded and processed. Signals FP1S, FP2S, AF7S, AF8S, F3S, and F4S from the headset EGS were preprocessed by the function EPPS in the processor PRC. The function EPPS samples signals FP1S, FP2S, AF7S, AF8S, F3S, and F4S at a frequency of 256Hz, and then filters the obtained samples through a bandpass filter from 1Hz to 25Hz. Following the first filtering is a three-step filtering process. In the first step, artifact subspace reconstruction (ASR) is used to remove large-amplitude artifacts. In the second step, Independent Component Analysis (ICA) is applied to the filtered signal to extract the sources of the components affecting the signal. In the third step, the Multiple Artifact Suppression Algorithm (MARA) is performed to automatically classify the ICA components as artifacts. This independent component classification is performed using a pre-trained model. Thus, components unrelated to brain activity, such as blinking or pulse, can be removed. Once the signal is filtered, different frequency calculations, such as Individualized Alpha Frequency (IAF) analysis, are applied to obtain signals associated with alpha, beta, gamma, delta, and theta waves (reference

[33] ). The signals associated with different waves are then processed to calculate, for example, the EEG index of frontal alpha asymmetry (FAA) according to methods described in existing literature, and the power spectral density is obtained using frequency bands defined by the Welch method. The DP%, DA%, and DF% functions convert the obtained measurements into corresponding percentage values ​​of deviation between the left and right electrodes (FPD, AFD, FD).

[0078] Therefore, the function DP% calculates the percentage deviation FPD between the left and right signals from electrode pairs FP1-FP2, the function DA% calculates the percentage deviation AFD between the left and right signals from electrode pairs AF7-AF8, and the function DF% calculates the percentage deviation FPD between the left and right signals from electrode pairs F3-F4. The percentage deviations FPD and AFD are processed to obtain the following indices FPI and AFI:

[0079] If the percentage deviation is less than -15%, the corresponding index is set to -2.

[0080] If the percentage deviation is included between -15% (inclusive) and -5% (exclusive), the corresponding index is set to -1.

[0081] If the percentage deviation is included between -5% (inclusive) and 4% (exclusive), the corresponding index is set to 0.

[0082] If the percentage deviation is included between 4% and 15% (inclusive), the corresponding index is set to 1.

[0083] If the percentage deviation is greater than 15%, the corresponding index is set to 2.

[0084] Process the percentage deviation (FD) to obtain the following index (FI):

[0085] If the percentage deviation is less than -15%, the corresponding index is set to -3.

[0086] If the percentage deviation is included between -15% (inclusive) and -5% (exclusive), the corresponding index is set to -2.

[0087] If the percentage deviation is included between -5% (inclusive) and 4% (exclusive), the corresponding index is set to 0.

[0088] If the percentage deviation is included between 4% and 15% (inclusive), the corresponding index is set to 2.

[0089] If the percentage deviation is greater than 15%, the corresponding index is set to 3.

[0090] Then, the indices FPI, AFI, and FI are added to the pleasure level NP to obtain the pleasure score PS.

[0091] Analysis of skin conductance activity (EDA) and cardiac activity was performed by a computer-based PRC using signal SS1 from sensor S1. A preprocessing function SCP1 sampled signal SS1 at 32Hz. Function SCP1 smoothed the skin conductance level (SCL) signal associated with EDA to remove small-amplitude artifacts. Function SCPR processed the output signal of function SCP1 to obtain the mean SMV and maximum amplitude deviation (SAV) over the time period of interest, i.e., during the period when the cross is displayed on the screen DSP and during or after tasting the product sample. Based on the values ​​SMV, SAV, functions DM%, DA%, and [missing information], the percentage deviations SCM, SCA, mean SMV, and maximum amplitude deviation SAV were determined between rest periods and during or after tasting the product sample, respectively.

[0092] For heart rate analysis, the computer PRC function SCP2 can also apply sampling processing and sample smoothing to the signal SS2 at 32Hz to remove low-amplitude artifacts. The function HRPR processes the smoothed sample to determine the time between each heartbeat, which allows for the calculation of the instantaneous heart rate HRV during rest and after tasting the product. The function DH% processes the instantaneous heart rate to obtain the percentage change in heart rate HRP between the rest period and the period during and after tasting the product sample.

[0093] The functions ICM and ICS process the percentage change SCM and SCA to determine the mean SMI and the maximum magnitude deviation SAI index, as shown below:

[0094] - If the percentage change (ICM) is less than 0%, the average index (SMI) is set to -3.

[0095] - If the percentage change (ICM) is between 0% and 4% (inclusive), then the average index (SMI) is set to 0.

[0096] - If the percentage change (ICM) is greater than 4%, the average index (SMI) is set to 2.

[0097] - If the percentage change (ICS) is less than 8%, the maximum magnitude deviation index (SAI) is set to 0.

[0098] - If the percentage change (ICS) is between 8% and 15% (inclusive), the index (SAI) is set to 1.

[0099] - If the percentage change (ICS) is greater than 15%, the index (SAI) is set to 2.

[0100] The function IHR processes the percentage of heart rate change (HRP) to determine the exponential heart rate change (HRI), as shown below:

[0101] - If the percentage change in HRP is less than -6%, the index HRI is set to -2.

[0102] - If the percentage change HRP is between -6% and 6% (inclusive), the index HRI is set to 0.

[0103] - If the percentage change in HRP is greater than 6%, the index HRI is set to 2.

[0104] The mean amplitude deviation index (SMI), the maximum amplitude deviation index (SAI), and the index HRI are added to the EI grade (NEI) to obtain the EI score (EIS).

[0105] Therefore, when calculating pleasure PS and EIEIS scores, the importance of assigning physiological measurements is greater than the importance of assigning pleasure values ​​VP, VEI, and EI provided by the participants.

[0106] The system ETR records the coordinates ETC of the right and left eye gazes on the screen at a sampling rate of 60Hz. The computer PRC function ETPS receives the coordinates ETC and determines the moving averages of the X-coordinates and Y-coordinates of the right and left eyes at each time moment to obtain the average coordinates (Xmoy and Ymoy) of the center point of the gaze. The function ETPR determines the time spent IEe in each region of interest (IZ) of the emoji e based on the trajectory of the center point of the participant's gaze, with the emojis arbitrarily numbered from 1 to 15. For this purpose, it is assumed that the region of interest (IZ) of each emoji E1 to E15 extends over the entire display area of ​​the emoji in the image DEM and along the edges a few millimeters around the emoji. The gaze duration for each emoji is accumulated. According to one implementation, the region of interest of the emoji is defined by a circle centered on the emoji and having a diameter equal to the distance from either of the two adjacent emojis.

[0107] The ETPR function can also determine the number of emoji gazes for each emoji, which is understood as counting emoji gazes when the gaze remains in the region of interest (IZ) of the emoji for more than a certain minimum threshold. This minimum threshold is, for example, set to 200 ms. Therefore, the total number of emoji gazes can also be determined. The ETPR function also determines the hesitation score (HI) based on the total time spent in the emoji's IZ divided by the total number of emoji gazes.

[0108] The function EMP processes the gaze duration IEe for each emoji e to determine whether one or more emojis are relevant to the participant's emotional experience. Since each emoji is numbered, the function EMP assigns a first number EP1 to the most relevant emoji, meaning the emoji that was gazed at for the longest time. A second most relevant emoji number EP2 can also be determined. The second most relevant emoji is the one that the participant gazed at for at least 50% of the time spent in the region of interest of the most relevant emoji. The second most relevant emoji is considered if the hesitation score HI is less than or equal to 1.

[0109] Pleasure score (PS), Emotion intensity score (EIS), and the first most relevant emoji number (EP1) and possibly the second most relevant emoji number (EP2) are provided to the PRC computer using the function SCPR in Table ET1. Table ET1 contains a list of emotions, each associated with an emoji number and a Pleasure and EI score. Emotions in Table ET1 are also grouped by emotion domain. Table ET1 can be defined according to the following example:

[0110] Table 1

[0111]

[0112]

[0113]

[0114] In this example, Table ET1 associates each emotion with an emoji number ranging from 1 to 15 and a Pleasure and EI index ranging from -8, corresponding to very negative pleasure or emotion intensity, to +8, corresponding to very positive pleasure or emotion intensity, where 0 corresponds to a neutral feeling. Therefore, Table ET1 defines a set of points in a three-dimensional space (x = emoji number, y = pleasure score, z = EI score).

[0115] Based on the emoji number EP1 and the possible EP2, as well as each participant's Pleasure PS and Emotion Intensity EIS scores and Table ET1, the SCPR function performs a classification for each participant. The first classification allows the participant's response to be categorized into one of the emotion domains specified in Table ET1, while the second classification allows the participant's emotion to be categorized within that domain.

[0116] To this end, the function SCPR considers the plane (x = EP1, y, z) in the three-dimensional space corresponding to the participant's emoticon number EP1, and determines the distance between the participant's point (x = EP1, y = PS, z = EIS) and each of the points defined in Table ET1 located in this plane. The emotion corresponding to the point p closest to the participant in the plane corresponding to the participant's emoticon number EP1, as defined by Table ET1, is used to determine the emotion EM1p felt by the participant. Then, based on the emotion EM1p, the function SCPR identifies the emotion domain U1p for participant p in Table ET1.

[0117] If the second emoji digit EP2 is selected by analyzing the signal ETC, the second emotion EM2p for participant p is determined in the same way by considering the point in table ET1 that is closest to the participant's point in the plane (x = EP2, y, z) defined by the second emoji digit EP2. Therefore, this emotion classification method makes it possible to identify the specific emotion that best corresponds to the content felt by the participant. The function SCPR then identifies the emotion domain U2p for participant p in table ET1 based on emotion EM2p.

[0118] According to one implementation, if a participant's point is too far from other points in the same emotion group corresponding to emoji numbers EP1 or EP2, it means that the emotion group being discussed does not correspond to the participant's response. For this purpose, a maximum distance threshold can be applied. This maximum distance threshold can be defined for each emotion group. For example, it can correspond to the maximum distance between points within the emotion group. Therefore, the distance threshold can be defined as specified in the following table:

[0119] Table 2

[0120] emoji numbers Maximum distance threshold 0 0 1 4.00 2 11.70 3 15.26 4 8.06 5 13.60 6 13.30 7 10.00 8 16.12 9 16.12 10 13.60 11 10.00 12 7.21 13 12.53 14 14.42 15 14.00

[0121] The analysis is then redirected to another emotion group (or emoji number). This other group may correspond to the second emoji EP2 selected via signal analysis ETC. If there is no second emoji, then no emotion is assigned.

[0122] The processor PRC then determines a score for each emotion listed in Table ET1 based on the number of times emotions (EM1p, EM2p) are identified among participants in a group of participants. For this purpose, the processor PRC can apply a weight of 2 when only one emotion, EM1p, is identified among participant p, and a weight whose sum equals 2 when both emotions, EMP1 and EMP2, are identified. These weights can be set to 1 and 1, or 1.25 and 0.75, or even determined based on the corresponding gaze duration of the two corresponding emoticons EP1 and EP2.

[0123] The processor PRC also determines the score for each emotion domain listed in Table ET1 based on the number of times each emotion domain (U1p, U2p) is identified among participants in a group of participants. For this purpose, emotion weighting rules can be applied to the emotion domains. The score for each emotion domain can also be determined by summing the scores obtained for emotions belonging to that emotion domain, as specified in Table ET1.

[0124] The emotion score and emotion domain score obtained for a product can be presented as a mapping representation of emotions and emotion domains. In this representation, an emotion domain is represented as a rectangle, the area of ​​which corresponds to the score associated with that emotion domain. Then, each emotion within each emotion domain is also represented as a rectangle, the area of ​​which corresponds to the score associated with that emotion, and this rectangle is inscribed within the rectangle corresponding to the emotion domain.

[0125] Figure 6A , 6B This is an example of a list of scores mapped to a product. Figure 6A The representation includes a large rectangle EEX corresponding to the emotional arousal domain, a smaller rectangle ETQ corresponding to the emotional calm domain, an even smaller rectangle IBT corresponding to the stimulating energy domain, four rectangles of the same size or even smaller corresponding to the negative aversion domain, the indulgent temptation domain, the emotional satisfaction domain, and the nostalgic reflection domain: NAV, IDT, ECT, and NRF, and finally, an even smaller rectangle FAT corresponding to the transient attention domain. Rectangle EEX includes rectangles of different areas, including rectangle WD corresponding to the emotion of surprise, rectangle ES corresponding to the emotion of avoidance, rectangle EY corresponding to the emotion of ecstasy, and two rectangles of the same size, AS and PS, corresponding to the emotions of astonishment and passion. Rectangle ETQ includes rectangles of different surfaces, including rectangle TD corresponding to the emotion of gentleness, and two rectangles of the same size, CF and IP, corresponding to the emotions of comfort and intense pleasure. Rectangle IBT includes rectangle EG corresponding to the emotion of motivation. Rectangle NAV includes rectangle DS corresponding to the emotion of dissatisfaction. Rectangle IDT includes rectangle GP corresponding to the emotion of sinful pleasure. Rectangle ECT includes rectangle JY corresponding to the emotion of joy. Rectangle FAT includes rectangle DT corresponding to the emotion of distraction.

[0126] Figure 6BThe representation includes a large rectangle ETQ corresponding to the domain of emotional calm; two smaller rectangles EEX and FCT corresponding to the domains of emotional arousal and satisfaction; an even smaller rectangle FAT corresponding to the domain of fleeting attention; two rectangles of the same size or even smaller, NAV and IDT corresponding to the domains of negative aversion and indulgent temptation; and finally, an even smaller rectangle RTM corresponding to the domain of sensual temptation. Rectangle ETQ includes rectangles with different surfaces, including rectangle TD corresponding to gentle emotions, rectangle CM corresponding to calm emotions, and two rectangles of the same size, RX and WB, corresponding to relaxed and happy emotions. Rectangle EEX includes rectangles with different surfaces, including rectangle GW corresponding to distance emotions, rectangle ES corresponding to avoidance emotions, and rectangle EY corresponding to ecstasy emotions. Rectangle ECT includes rectangle JY corresponding to joyful emotions. Rectangle FAT includes rectangle DT corresponding to distracted emotions. Rectangle IBT includes rectangle EG corresponding to motivating emotions. Rectangle NAV includes rectangle DS corresponding to dissatisfaction emotions. Rectangle IDT includes rectangle GP corresponding to sinful pleasure emotions. The rectangular RTM includes rectangular DLs corresponding to refined emotions.

[0127] Those skilled in the art will recognize that this invention is susceptible to various modifications and applications. In particular, this invention is not limited to determining a list of emotions felt by participants in product testing. Rather, this invention can be applied more generally to determining what could be any kind of subject, such as, for example, film genres (visual and auditory), music genres (auditory), product sensations (tactile), video game concepts (visual and auditory), or even vehicle interior concepts.

[0128] Furthermore, it is unnecessary to use a topic table such as Table 1. In fact, other data structures can be easily developed to determine topics from physiological measurements. For example, a list of movie genres (action, adventure, comedy, thriller, horror, romance, etc.) can be correlated with physiological measurement values. Additionally, other types of physiological measurements, such as respiratory rate, shaking movements, facial expressions, skin temperature, etc., can be used. Similarly, other types of scores can be determined from physiological measurements, revealing that these scores are highly dependent on the type of physiological measurement.

[0129] It is also possible to implement a method to identify topics other than those mentioned above by calculating the distance between points.

[0130] In addition to the classification methods described above that use the range of values ​​associated with each score, methods for determining scores can also be implemented. Furthermore, classification can be implemented not based on percentage changes, but directly on values ​​of physiological measurements, and by setting a range of possible values ​​for the index generated by that classification.

[0131] Furthermore, it is unnecessary to consider the specified emojis or values ​​of pleasure and emotional intensity provided by the participants, as these indications merely allow for refinement of the results.

[0132] The methods described above can be applied to neurofeedback systems, for example, to guide people in decision-making or to determine a person's emotional state, such as anxiety or stress. Therefore, it is not necessary to apply the process to a group of participants and provide general results based on individual outcomes obtained by the participants.

[0133] References

[0134] 1.Bradley,Margaret M.,et Peter J.Lang.1994.<<Measuring Emotion:TheSelf-Assessment Manikin and the Semantic Differential> >.Journalof BehaviorTherapy and Experimental Psychiatry 25(1):49-59.doi:10.1016 / 0005-7916(94)90063-9.

[0135] 2.Bensafi,M.,C.Rouby,V.Farget,B.Bertrand,M.Vigouroux,etA.Holley.2002.<<Psychophysiological Correlates of Affectsin Human Olfaction> >.Neurophysiologie Clinique / Clinical Neurophysiology 32(5):326-32.doi:10.1016 / S0987-7053(02)00339-8.

[0136] 3.Jack,Rachael E.,Oliver GBGarrod,Hui Yu,Roberto Caldara,etPhilippe G.Schyns.2012.<<Facial Expressions of Emotion Are Not CulturallyUniversal> >.Proceedings of the National Academy of Sciences 109(19):7241-44.doi:10.1073 / pnas.1200155109.

[0137] 4.Warrenburg, S. 2005.<Effects of Fragrance on Emotions:Moods andPhysiology> >.Chemical Senses 30(Supplement 1):i248-49.doi:10.1093 / chemse / bjh208.

[0138] 5. Porcherot, Christelle, Sylvain Delplanque, Sophie Raviot-Derrien, Bénédicte Le Calvé, Christelle Chrea, Nadine Gaudreau, and Isabelle Cayeux. 2010. <HowDo You Feel When You Smell This?Optimization of a Verbal Measurement of Odor-Elicited Emotions> >.Food Quality and Preference 21(8):938-47.doi:10.1016 / j.foodqual.2010.03.012.

[0139] 6.Churchill,Anne,and John Behan.2010.<<Comparison of Methods Used toStudy Consumer Emotions Associated with Fragrance> >.Food Quality and Preference 21(8):1108-13.doi:10.1016 / j.foodqual.2010.07.006.

[0140] 7.Vidal,Leticia,Gastón Ares,and Sara R.Jaeger.2016.<<Use of Emoticonand Emojiin Tweets for Food-Related Emotional Expression> >.Food Quality and Preference 49:119-28.doi:10.1016 / j.foodqual.2015.12.002.

[0141] 8.Jaeger,Sara R.,Armand V.Cardello,et Howard G.Schutz.2013.<<EmotionQuestionnaires:AConsumer-Centric Perspective>>.Food Quality and Preference 30(2):229-41.doi:10.1016 / j.foodqual.2013.05.015.

[0142] 9.Thomson,David M.H.,et Christopher Crocker.2015.<<Application ofConceptual Profiling in Brand,Packaging and Product Development>>.FoodQuality and Preference 40:343-53.doi:10.1016 / j.foodqual.2014.04.013.

[0143] 10.Truss.2004.<<Truss 2004 An educational companion to eats shoots&leaves.pdf>>.

[0144] 11.Cowie,R.,E.Douglas-Cowie,N.Tsapatsoulis,G.Votsis,S.Kollias,W.Fellenz,et J.G.Taylor.2001.<<Emotion Recognition in Human-ComputerInteraction>>.IEEE Signal Processing Magazine 18(1):32-80.doi:10.1109 / 79.911197.

[0145] 12.Huang,Albert H.,David C.Yen,et Xiaoni Zhang.2008.<<Exploring thePotential Effects of Emoticons>>.Information&Management 45(7):466-73.doi:10.1016 / j.im.2008.07.001.

[0146] 13.Huffaker,David A.,et Sandra L.Calvert.2006.<<Gender,Identity,andLanguage Use in Teenage Blogs> >.Journalof Computer-Mediated Communication 10(2):00-00.doi:10.1111 / j.1083-6101.2005.tb00238.x.

[0147] 14.Wolf, Alecia.2000.<<Emotional Expression Online:GenderDifferencesin Emoticon Use> >.CyberPsychology&Behavior 3(5):827-33.doi:10.1089 / 10949310050191809.

[0148] 15.Barros,Catarina,Ana Rita Pereira,Adriana Sampaio,Ana Buján,etDiego Pinal.2022.<<Frontal Alpha Asymmetry and Negative Mood:A Cross-Sectional Study in Older and Younger Adults> >.Symmetry 14(8):1579.doi:10.3390 / sym14081579.

[0149] 16.Palacios- Almudena, Javier Marín-Morales, Manuel Contero, and Mariano 2023.<<Predicting Decision-Making in Virtual Environments:AnEye Movement Analysis with Household Products>>.Applied Sciences 13(12):7124.doi:10.3390 / app13127124.

[0150] 17.Rojas,Juan-Carlos,Javier Marín-Morales,Jose Manuel Ausín Azofra,etManuel Contero.2020.<<Recognizing Decision-Making Using Eye Movement:ACaseStudy With Children>>.Frontiersin Psychology 11:570470.doi:10.3389 / fpsyg.2020.570470.

[0151] 18.Wedel,Michel,Rik Pieters,et Ralf Van Der Lans.2023.<<Modeling EyeMovements During Decision Making:AReview>>.Psychometrika 88(2):697-729.doi:10.1007 / s11336-022-09876-4.

[0152] 19.Lim,Jia Zheng,James Mountstephens,et Jason Teo.2020.<<EmotionRecognition Using Eye-Tracking:Taxonomy,Review and Current Challenges>>.Sensors 20(8):2384.doi:10.3390 / s20082384.

[0153] 20.Zhu,Jianping,Lizhen Ji,et Chengyu Liu.2019.<<Heart RateVariability Monitoring for Emotion and Disorders of Emotion>>.PhysiologicalMeasurement 40(6):064004.doi:10.1088 / 1361-6579 / ab1887.

[0154] 21.Golnar-Nik,Parnaz,Sajjad Farashi,et Mir-Shahram Safari.2019.< <TheApplication of EEG Power for the Prediction and Interpretation of ConsumerDecision-Making:A Neuromarketing Study> >.Physiology&Behavior 207:90-98.doi:10.1016 / j.physbeh.2019.04.025.

[0155] 22.Hyun,Yong J.,Byungho Park,Sanghak Lee,Sungjun Park,and Moon-YongKim.2018.< <ANeuro-Scientific Analysis of Affect and Cognition in ConsumerResponse to the Marketing Stimulus-EEG Analysis of the Relationship betweenPrefrontal Hemispheric Lateralization and Occipital Alpha Blocking in AdExposure> >.SSRN Electronic Journal.doi:10.2139 / ssrn.3211155.

[0156] 23.Lin,Meng-Hsien(Jenny),Samantha NNCross,William J.Jones,and TerryL.Childers.2018.<<Applying EEG in Consumer Neuroscience> >.EuropeanJournalofMarketing 52(1 / 2):66-91.doi:10.1108 / EJM-12-2016-0805.

[0157] 24.Huffmeijer,Renske,Lenneke RAAlink,Mattie Tops,MarianJ.Bakermans-Kranenburg,and Marinus H.Van IJzendoorn.2012.< <Asymmetric FrontalBrain Activity and Parental Rejection Predict Altruistic Behavior:Moderationof Oxytocin Effects> >.Cognitive,Affective,&Behavioral Neuroscience 12(2):382-92.doi:10.3758 / s13415-011-0082-6.

[0158] 25.Sandoval,Pamela Simon,et Jesus García-Madariaga.2024.<<Impact ofEmotional Appealon Non-profit Advertising:A Neurophysiological Analysis> >.Journalof Consumer Behavior 23(1):203-17.doi:10.1002 / cb.2168.

[0159] 26.Masuo,Yoshinori,Tadaaki Satou,Hiroaki Takemoto,and KazuoKoike.2021.< <Smell and Stress Response in the Brain:Review of the Connectionbetween Chemistry and Neuropharmacology> >.Molecules 26(9):2571.doi:10.3390 / molecules26092571.

[0160] 27.Bansafi,M.2002.<<Autonomic Nervous System Responsesto Odours:TheRole of Pleasantness and Arousal> >.Chemical Senses 27(8):703-9.doi:10.1093 / chemse / 27.8.703.

[0161] 28.Kadohisa,Mikiko.2013.<<Effects of Odor on Emotion,withImplications>>.Frontiersin Systems Neuroscience 7.doi:10.3389 / fnsys.2013.00066.

[0162] 29.Marci,Carl D.,Debra M.Glick,Rebecca Loh,et Darin D.Dougherty.2007.<<Autonomic and Prefrontal Cortex Responsesto Autobiographical Recall ofEmotions>>.Cognitive,Affective,&Behavioral Neuroscience 7(3):243-50.doi:10.3758 / CABN.7.3.243.

[0163] 30.Rodriguez,S.,M.Fernandez,A.Cepedabenito,et J.Vila.2005.<<Subjective and Physiological Reactivity to Chocolate Imagesin High and LowChocolate Cravers>>.Biological Psychology 70(1):9-18.doi:10.1016 / j.biopsycho.2004.10.001.

[0164] 31.Ghamari,Mohammad.2018.< >.InternationalJournalof Biosensors&BioeleXctronics 4(4).doi:10.15406 / ijbsbe.2018.04.00125.

[0165] 32.Grapperon,J.2012.<<La mesure de la réactionélectrodermale> >.

[0166] 33.Moya,Ingrit,Jesus García-Madariaga,and Maria-Francisca Blasco.2020.<<What Can Neuromarketing Tell Us about Food Packaging?> >Foods 9(12):1856.doi:10.3390 / foods9121856.

[0167] 34.García-Madariaga,Jesus,Maria-Francisca Blasco Lopez,Ingrit MoyaBurgos,and Nuria Recuero Virto.2019.<<Do Isolated Packaging VariablesInfluence Consumers’Attention and Preferences?> >Physiology&Behavior 200:96-103.doi:10.1016 / j.physbeh.2018.04.030.

[0168] 35.Plutchik,Robert.1982.<<APsychoevolutionary Theory of Emotions> >.Social Science Information 21(4-5):529-53.doi:10.1177 / 053901882021004003.

[0169] 36. Zheng, Wei-Long, Jia-Yi Zhu, et Bao-Liang Lu. 2019.<IdentifyingStable Patterns over Time for Emotion Recognition from EEG> >.IEEETransactions on Affective Computing 10(3):417-29.doi:10.1109 / TAFFC.2017.2712143.

Claims

1. A method for determining topics arising from a product to be characterized, comprising the following steps: (a) Presenting the product to be characterized to the subject in a manner that subjectes a human subject to sensory stimulation; (b) Obtain physiological measurement signals (FP1S, FP2S, AF7S, AF8S, F3S, F4S, SS1, SS2) from the subject before and during the presentation of the product to be characterized. (c) During or after the presentation of the product to be characterized, the subject is presented with an image (DEM) showing a set of emojis (E1 to E15), and the subject is asked to visually select the emojis from the image; (d) Acquire eye-tracking signals (ETC) from the subject during the presentation of the image; (e) Process gaze tracking signals to determine a first emoji number (EP1) of the longest duration of gaze by the subject, wherein the emoji of the longest duration of gaze determines a set of themes that may be evoked in the subject by the presentation of the product to be represented; as well as (f) Process the physiological measurement signals acquired before and during the presentation of the product to be characterized to identify themes evoked in the subject within a defined set of themes.

2. The method according to claim 1, wherein, The processing of the physiological measurement results (FP1S, FP2S, AF7S, AF8S, F3S, F4S, SS1, SS2) involves using a table (ET1) listing topics, wherein each topic is associated with an emoji number and a point defined by a pleasure score and an emotional intensity score.

3. The method according to claim 2, wherein, The processing of the physiological measurement results (FP1S, FP2S, AF7S, AF8S, F3S, F4S, SS1, SS2) includes the following steps: The pleasure score (PS) and emotional intensity score (EIS) were determined based on the physiological measurement results. Identify all topics corresponding to the first determined emoji number (EP1) in the table (ET1); and Calculate the distance between each of the points defined by the determined Pleasure (PS) score and Emotional Intensity (EIS) score and the points associated with the topics identified in the table, wherein the minimum calculated distance corresponds to the topic identified for the subject.

4. The method according to any one of claims 1 to 3, wherein, When another emoji (E1 to E15) has been gazed at by the subject for at least 50% of the gaze time corresponding to the first emoji number (EP1) that has been gazed at for the longest duration, the processing of the gaze tracking signal (ETC) provides a second emoji number (EP2) that has been gazed at by the subject for the longest duration.

5. The method according to any one of claims 1 to 4, wherein, The acquisition and processing of the physiological measurement signals (FP1S, FP2S, AF7S, AF8S, F3S, F4S, SS1, SS2) includes the following steps: During the presentation of the product to be characterized, electroencephalogram (EEG) signals (FP1S, FP2S, AF7S, AF8S, F3S, F4S) were acquired from electrodes (FP1, FP2, AF7, AF8, F3, F4) placed on the head of the subject, the electrodes including a right frontal lobe electrode pair and a left frontal lobe electrode pair (FP1-FP2, AF7-AF8, F3-F4); The EEG signals are processed to determine the percentage deviation (FPD, AFD, FD) between the right and left electrodes in each of the frontal lobe electrode pairs; For each percentage deviation, the pleasure index (FPI, AFI, FI) is determined based on whether the percentage deviation falls within a range of percentage values ​​specified for the pleasure index. as well as The pleasure score (PS) is determined based on the pleasure index.

6. The method according to claim 5, wherein, The pleasure score (PS) is determined by adding the pleasure index (FPI, AFI, FI) to a pleasure level (NP) determined based on whether the pleasure value (VP) indicated by the subject falls within a range of pleasure values ​​in a set of pleasure ranges defined for the pleasure value.

7. The method according to any one of claims 1 to 6, wherein, The acquisition and processing of the physiological measurement signals includes the following steps: During the rest period and the presentation period of the product to be characterized, skin conductance signals (SS1, SS2) are acquired from the skin conductance sensors (S1, S2) placed on the subject. The skin conductance signal is processed to determine the mean (SMV) and maximum amplitude deviation (SAV) of the subject's skin conductance and heart rate (HRV) values ​​between the resting period and the presentation period of the product to be characterized; Determine the percentage deviation of the mean skin electrical activity value (SCM), the percentage deviation of the maximum amplitude deviation of skin electrical activity (SCA), and the percentage deviation of the heart rate value (HRP) between the rest period and the presentation period of the product to be characterized; as well as For each percentage deviation in electroskin activity and heart rate, an emotional intensity index (SMI, SAI, HRI) is determined based on whether the percentage deviation falls within a range of percentage values ​​in a set of percentage ranges determined for the percentage deviation. as well as The Emotion Intensity Score (EIS) is determined based on the Emotion Intensity Index.

8. The method according to claim 7, wherein, The Emotion Intensity Score (EIS) is determined by adding the Emotion Intensity Index (SMI, SAI, HRI) to an Emotion Intensity Level (NEI) determined based on whether the Emotion Intensity Value (VEI) indicated by the subject falls within a range of Emotion Intensity Values ​​defined for the Emotion Intensity Value.

9. The method according to any one of claims 1 to 8, wherein, The subjects belong to a group of participants, and the method includes the following steps: For each participant in the group of participants, steps (a) through (f) are performed to determine a topic for that participant; and The perceived topic representation of the product to be represented is determined based on the topics identified for all participants in the group of participants.

10. The method according to claim 9, wherein, The sensory representation results include a list of topics, each of which is associated with a topic score determined based on the number of times the topic is identified for a participant in the group of participants.

11. A system comprising: Computer (PRC), An eye-tracking device (ETR) connected to the computer. Physiological parameter measurement devices (EGS, PHS) connected to the computer, and A display screen (DSP) connected to the computer, the processor being configured to implement the method according to any one of claims 1 to 10.

12. The system according to claim 11, wherein, The physiological parameter measuring device includes: A set of electroencephalogram (EGS) electrodes, and Devices used to measure skin electrical activity and cardiac activity (PHS).

Images