CN111399650B - Audio-visual media evaluation method based on group brain network - Google Patents

Audio-visual media evaluation method based on group brain network Download PDF

Info

Publication number
CN111399650B
CN111399650B CN202010193942.1A CN202010193942A CN111399650B CN 111399650 B CN111399650 B CN 111399650B CN 202010193942 A CN202010193942 A CN 202010193942A CN 111399650 B CN111399650 B CN 111399650B
Authority
CN
China
Prior art keywords
blood oxygen
brain
oxygen concentration
brain network
audio
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202010193942.1A
Other languages
Chinese (zh)
Other versions
CN111399650A (en
Inventor
刘涛
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Zhejiang University ZJU
Original Assignee
Zhejiang University ZJU
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Zhejiang University ZJU filed Critical Zhejiang University ZJU
Priority to CN202010193942.1A priority Critical patent/CN111399650B/en
Publication of CN111399650A publication Critical patent/CN111399650A/en
Application granted granted Critical
Publication of CN111399650B publication Critical patent/CN111399650B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
    • G06F3/011—Arrangements for interaction with the human body, e.g. for user immersion in virtual reality
    • G06F3/015—Input arrangements based on nervous system activity detection, e.g. brain waves [EEG] detection, electromyograms [EMG] detection, electrodermal response detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0241—Advertisements
    • G06Q30/0242—Determining effectiveness of advertisements

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Strategic Management (AREA)
  • Development Economics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Economics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Game Theory and Decision Science (AREA)
  • Educational Administration (AREA)
  • Accounting & Taxation (AREA)
  • General Business, Economics & Management (AREA)
  • Marketing (AREA)
  • Finance (AREA)
  • Dermatology (AREA)
  • General Health & Medical Sciences (AREA)
  • Human Computer Interaction (AREA)
  • Neurology (AREA)
  • Neurosurgery (AREA)
  • Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)

Abstract

The invention discloses an audio-visual media evaluation method based on a swarm brain network, which comprises the steps of 1, respectively presenting audio-visual media to each tested person, and continuously measuring the blood oxygen concentration data of the specific position of the preset head of the tested person by utilizing a near infrared spectrum brain function imaging technology during the presentation period; the number of the tested persons is not less than 10 persons; step 2, after denoising the blood oxygen concentration data of the detected person, carrying out baseline calibration, and carrying out standardized conversion to the blood oxygen concentration z fraction of each detected person; taking each tested person as a node in a brain network of a group, and calculating a pearson correlation coefficient of brain activity among the tested persons according to the blood oxygen concentration z fraction of each tested person; calculating the population brain network density according to the pearson correlation coefficient of the brain activity; and 3, evaluating the transmission effect of the audio-visual media according to the obtained population brain network density, wherein the higher the population brain network density is, the better the transmission effect is. The method has objective data source and high ecological efficiency in the evaluation process.

Description

Audio-visual media evaluation method based on group brain network
Technical Field
The invention relates to the cross field of psychology, neuroscience and marketing, in particular to an evaluation method for evaluating the audio-visual media propagation effect based on brain network density.
Background
Audiovisual media is an important way of information dissemination and plays a crucial role in commercial activities. For example, effective commercials may promote increased product sales, establish good brand image, and help businesses realize increased profits. Therefore, how to accurately and objectively evaluate and predict the propagation effect of audiovisual media becomes a topic widely discussed in the academic world and the industry.
Traditional methods of assessing the effects of audiovisual media mainly include questionnaires, group discussions, behavioral experiments, and realistic data mining. However, these conventional measurement methods have certain limitations. For example, the research methods in the forms of questionnaires, interviews and the like are posterior type research methods, which are mainly based on the accurate recollection of a subject to a specific scene and feeling, and cannot measure the attitude behavior of a consumer in the process of watching or listening to audio-visual media in real time; moreover, due to The limitations of cognitive resources, most decisions and behaviors of people are made unconsciously, so The person under investigation often cannot answer The true reasons behind many decisions and behaviors accurately (neural and personal decisions of neural in business, 11, 284-.
Therefore, the self-reported data source of the traditional audio-visual media evaluation method cannot completely reflect the real idea of the consumer, and the key factors influencing the audio-visual media effect cannot be completely and accurately found. To address this problem, new methods for evaluating audiovisual media require more direct and objective measures to measure consumer decision processes and degrees of implicit status to make The evaluation of audiovisual media more effective (The voice of The customer, Marketing Science, 1993: 12(1), 1-27.).
Nowadays, rapid development of neuroscience and brain imaging technologies makes it possible to objectively and accurately measure brain activities of consumers, and also provides a new idea for improvement of an audiovisual media evaluation method. In short, in the conventional audiovisual medium evaluation method, the mind of the consumer is a black box, which cannot be accurately read, and by means of neuroscience, the researcher can scan the brain of the consumer to further know what the consumer wants and wants to do at the moment. This is of great significance for audiovisual media effect evaluation.
From the specific audiovisual media evaluation technology, the audiovisual media evaluation technology based on the group brain network has the following advantages: firstly, the traditional audiovisual media evaluation method relies on subjective data of self-report of a consumer, and objective brain activity data of a consumer responding to a certain video or audio can be obtained by means of measurement means of a brain imaging technology, so that attitude and opinion of the consumer to audiovisual media (marked actions and modulated neural responses of advanced pleasents ", PNAS, 2008: 105(3), 1050-. Second, in neuroscience experiments, activation of the cerebral cortex occurs simultaneously with the progress of watching video, without any delay in data (What is "neural activation". Therefore, neuroscience can help us to know the attitude of the consumers to the advertisements more scientifically and evaluate the effect of the advertisements more accurately. Finally, compared with the traditional evaluation mode, the audiovisual medium evaluation technology based on the group brain network has fewer participants to be recruited, and can obtain more subtle attitude response to audiovisual media based on fewer participants. Therefore, The audio-visual media evaluation method based on brain imaging has attracted increasing attention as a powerful complement to The conventional behavior method (neural imaging: The house and hope of neural in The business ", Nature Reviews Neuroscience, 2010: 11, 284-292.).
Some prior studies have shown that different characteristics of audiovisual media can cause different Neuro-responses in consumers (branches on the brain: Neuro-images of adapting ". Business Strategy Review, 2000: 11(3), 17-30.) and that different activation states of the consumer's brain can reflect the attitude and evaluation of the consumer with respect to audiovisual media. For example, a scholars measures brain activity of participants when watching advertisements by means of functional magnetic resonance imaging (fMRI), finds that the degree of activation of the ventral striatal region associated with the reward process can reflect the preference of people for advertisements, and that the activity of the ventral striatal region is a more powerful indicator for predicting the market reaction to advertisements than for evaluating the advertisement effect by using a conventional behavioral measurement method. In addition, some scholars record the brain activities of experimental participants when watching movie trailers by means of electroencephalogram (EEG) technology, and find that the activities of beta waves in high-frequency electroencephalogram components are related to personal preferences, and the activities of gamma waves are related to box-office conditions after showing movies.
The above studies have promoted the application and development of brain imaging techniques in audio-visual media evaluation, but the limitations of these techniques are not negligible at present. Firstly, in the aspect of equipment, fMRI is high in cost, an experimental participant is required to lie in an experimental instrument to watch audio-visual media, the physical activity of the participant is very limited, and the difference is larger than the daily real environment contacting the audio-visual media; the EEG technique also has problems of low spatial resolution, poor noise immunity, etc.
Secondly, in the aspect of a calculation model, a student (A ticket for your resources, Method for predicting content and using the new spatial information of moviegoers', Journal of Consumer Research, 2017: 44(1), 160 and 181.), predicts the box room situation after the movie is shown by the electroencephalogram activity synchronization degree when a group of participants watch the movie trailer, and finds that the higher the electroencephalogram activity synchronization degree when the participants watch the trailer, the higher the box room after the movie is shown. However, this calculation method completely ignores the problem of spatial resolution of the brain, and it is difficult to interpret the result.
Previous studies show that audio-visual media is a kind of information source, and the process affecting consumers is also a special communication form, and one of the signs of effective communication is that the recipients of information have the same understanding of the information source, and there is more consistent cognitive or emotional resonance between the recipients of information (Neural synchronization, Neural of Neural communication, 2012: 32(45), 16064-. For example, studies (Abstract art and clinical motor activation: an EEG study ", Frontiers in Human Neuroscience, 2012: 6, 311.) have shown that cognitive or emotional resonance affects consumer ratings of products displayed in advertisements, with higher levels of resonance giving higher ratings of products.
However, the existing measurement method based on the brain imaging technology has the problems of low ecological effectiveness of equipment, insufficient theoretical explanation, no establishment of a calculation model based on the basic principle of media propagation and the like.
Disclosure of Invention
The invention provides an audio-visual media evaluation method based on a colony brain network, which is characterized in that near infrared spectrum brain function imaging equipment is utilized to continuously measure the blood oxygen concentration data of the brain of a tested person in an environment with higher ecological efficiency, the colony brain network density is calculated by utilizing a network modeling method based on graph theory, and the propagation effect of the audio-visual media is evaluated by utilizing the height of the colony brain network density, so that the audio-visual media quality evaluation method which is more objective and accurate and has higher ecological efficiency and theoretical interpretation strength is provided.
The technical scheme provided by the invention for solving the technical problems is as follows:
a method for evaluating the transmission effect of an audio-visual medium comprises the following steps:
step 1: acquiring cerebral blood oxygen concentration data by using a near infrared spectrum cerebral function imaging technology:
respectively presenting the audio-visual media to each tested person, and continuously measuring the blood oxygen concentration data of the specific head position preset by each tested person by using a near infrared spectrum brain function imaging technology during presentation; the number of the tested persons is not less than 10 persons;
step 2: data processing:
denoising the blood oxygen concentration data of each measured person acquired in the step (1), calibrating a base line, and converting the data into a blood oxygen concentration z fraction of each measured person in a standardized manner; taking each tested person as a node in a group brain network, and calculating a pearson correlation coefficient of brain activity between every two tested persons according to the blood oxygen concentration z fraction of each tested person; further calculating the population brain network density according to the pearson correlation coefficient of the brain activity;
and step 3: evaluation of audiovisual medium propagation effect:
evaluating the transmission effect of the audio-visual media according to the population brain network density obtained in the step 2, wherein the higher the population brain network density is, the better the transmission effect is.
The invention combines a near infrared spectrum brain-attached spectroscopy (fNIRS) technology and a network modeling method based on graph theory to construct an evaluation method of the audiovisual media propagation effect.
fNIRS is a novel, non-invasive brain imaging technique. fNIRS has unique technical advantages over the most commonly used brain function imaging techniques at present (fMRI and EEG).
Compared with fMRI, fNIRS has a higher time sampling rate (several milliseconds to tens of milliseconds), has smaller limitation on the physical activity of a tested person, and can provide richer cortical blood oxygen metabolism information, including concentration information of oxyhemoglobin, deoxyhemoglobin, total hemoglobin and the like. At the same time, fNIRS is superior to fMRI in price, convenience of operation, portability of the apparatus, and compatibility.
Compared with EEG, fNIRS has higher spatial resolution (in the order of centimeters) and better noise immunity, and can be measured continuously at any time and any place for a long time.
In view of the above technical advantages, fNIRS is well suited for data collection and research in viewing scenes of audiovisual media for assessing the quality of audiovisual media.
Based on the theoretical basis of the effective transmission of the audio-visual media, the invention combines the method of graph theory to construct a model for judging the transmission effect of the audio-visual media from the perspective of the swarm brain network: the brain of each tested person is taken as a node in a group brain network, if the audio-visual media can effectively transmit information, the audio-visual media taken as an information source in the network can cause emotion and cognitive resonance in the tested persons, so that the tested person group and the brain region activation related to the emotion have a synchronization characteristic, and further, the group brain network density formed by each tested person and the brain region related to the shared emotion is higher.
The evaluation method provided by the invention can judge whether the audio-visual media can cause emotion and cognitive resonance among the tested persons or not and can effectively transmit the information effect, and has more scientific and practical value compared with the traditional evaluation model.
The population brain network density index comprises a brain network density overall index (overall brain network density) reflecting overall evaluation of the audio-visual media and a brain network density detail index (instantaneous brain network density) on each unit time point of the audio-visual media obtained based on analysis of a moving window technology.
The audiovisual media described in the present invention refers to dynamic images including audio or video, such as video advertisements, music, or movie trailers, that are transmitted through terminals such as televisions, PCs, and mobile phones. The audio-visual media are advertisements, trailers or music.
Preferably, the audiovisual medium is an advertisement or music. The evaluation method is more suitable for evaluating the propagation effects of complete and continuous audio-visual media and experience media.
The predetermined head specific position is a left or right inferior frontal gyrocerebral region.
Preferably, the predetermined head specific position is a right inferior frontal gyrocerebral region.
The right underforehead ventral gyrus (IFG) is one of the core brain areas of the mirror image neuron system, is closely related to the human sympathy capacity, and a group brain network connection density index calculated based on the brain activities of the area can be more accurately used for judging whether the audio-visual media can cause emotion and cognitive resonance among tested persons or not and can effectively transmit information or not.
Preferably, the predetermined head specific positions are Bradman zones 44 and 45.
The number of the tested persons is 18-32. Preferably, the number of the tested persons is 20 or 30.
According to the general experience of neuroscience experiments, the number of the tested persons in a single test is preferably 20 or 30.
During the process of presenting audio-visual media to the tested person, the blood oxygen concentration data of the specific position of the preset head of the tested person is continuously measured by utilizing the near infrared spectrum brain function imaging technology, and the method comprises the following steps: the tested person sits in front of the screen for playing the audio-visual media, wears a near-infrared optical brain function imaging device (fNIRS device), and after the fNIRS device starts to record the blood oxygen concentration data of the specific position of the preset head of the tested person, the tested object watches the audio-visual media presented on the screen, the fNIRS device records the blood oxygen concentration data of the tested person in real time during the period, and the fNIRS device is closed after the watching is finished.
The method for wearing the near-infrared optical brain function imaging equipment comprises the following steps: the 6 light sources and the 6 detectors are distributed in measuring channels of two lateral frontal gyrus areas, each semi-cerebral cortex comprises 3 light sources and 3 detectors, the interval between every two probes is 3 cm, the 6 measuring channels with double wavelengths (760nm and 850nm) are totally formed, and the sampling rate of the device is 10 Hz.
Preferably, the step 1 of collecting infrared spectrum data comprises the following specific steps: the measured person sits in front of a screen for playing audio-visual media, wears the fNIRS device, the fNIRS device starts to record the blood oxygen concentration data of the specific head position of the measured person, each measured person takes a rest for 30-60 seconds facing the empty screen to reach a resting state, then the measured person watches the audio-visual media presented on the screen, the fNIRS device records the resting blood oxygen concentration data of the measured person in real time and the blood oxygen concentration data during watching the audio-visual media, and the fNIRS device is closed after the watching is finished.
The rest blood oxygen concentration data takes the average value of the blood oxygen concentration data of the last 10-15 seconds of the rest period as the rest blood oxygen concentration data.
Before the audio-visual media are presented, the tested person goes through a rest stage, the brain activity when the audio-visual media are watched and the brain activity in a rest state are distinguished, and then the brain activation phenomena caused by watching the audio-visual media are determined.
The measurement channel (the midpoint position of one light source and one detector) is the channel of the left or right frontal subconcephalat region.
In step 2 of the invention, the denoising is to remove data noise, such as physiological noise of respiration, heartbeat and the like and machine systems of drift and the like, and can be realized by band-pass filtering (0.01-0.1Hz) by using Matlab software.
The baseline calibration is respectively carried out by taking the resting blood oxygen concentration of each tested person after denoising treatment as a baseline.
The method for measuring the resting blood oxygen concentration comprises the following steps: and respectively and continuously measuring the blood oxygen concentration data of each tested person, which is exposed to an empty screen and has a rest time of 30-60 seconds, by using a near-infrared spectrum brain function imaging technology, and taking the average value of the blood oxygen concentration data of each sampling point of the last 10-15 seconds of the rest time of each tested person as the average concentration of the rest blood oxygen of each tested person.
The blood oxygen concentration z fraction can be directly obtained through Excel software or obtained through calculation of a formula 1.
The calculation formula 1 is: z ═ x- μ)/σ;
wherein x is blood oxygen concentration data, mu is the average value of the average concentration of the resting blood oxygen of all the tested persons, and sigma is the standard deviation of the average concentration of the resting blood oxygen of the tested persons.
The pearson correlation coefficient of brain activity can be directly obtained by SPSS software or calculated by a calculation formula 2.
The calculation formula 2 is:
Figure BDA0002415904630000051
wherein, the blood oxygen concentration z scores of any two tested persons of X and Y, cov (X, Y) is the covariance of X and Y, Var [ X ] is the variance of X, and Var [ Y ] is the variance of Y.
The population brain network density can be directly obtained by Excel software. The method comprises the following steps of taking the brain of each tested person as a node in a group brain network, and calculating through a formula 2 to obtain the brain activity synchronism between every two tested persons; if a pair of tested persons has obvious brain activity synchronism, calculating that one edge of the brain network of the group exists; the same method can obtain the number of the edges coexisting in the brain network of the group; the sum of all its edges is taken as the connection density of the brain network of the population.
When the number of said audio-visual media is 1; the blood oxygen concentration data in the step 1 are instantaneous blood oxygen concentration data at different moments; the population brain network density in the step 2 is the instantaneous brain network density at different moments; obtaining instantaneous blood oxygen concentration data of the audio-visual media at different moments by utilizing a sliding window technology; and (3) calculating the instantaneous brain network density at different moments, wherein the population brain network density in the step (3) is the instantaneous brain network density obtained in the step (2), and the propagation effect is the propagation effect at different moments.
When the number of audiovisual media is greater than 1; the blood oxygen concentration data in the step 1 is the average blood oxygen concentration data of each audio-visual media; the population brain network density in the step 2 is the total brain network density of each audio-visual media; calculating the total brain network density of the audio-visual media according to the average blood oxygen concentration data of each tested person; the population brain network density in the step 3 is the total brain network density of each audio-visual media obtained in the step 2, and the propagation effect is the total propagation effect.
The average blood oxygen concentration data can be directly obtained by Excel software. The specific method is that each audio-visual media is taken as a complete experience unit, and then the average value of the blood oxygen concentration change in the experience unit is calculated.
The instantaneous brain network density, namely the detail index, aims at a certain audio-visual medium, and utilizes a sliding window technology to analyze how much each time point of the medium can cause group resonance, which time points can generate group resonance with higher degree, and which time points can not generate group resonance.
The overall brain network density, namely the group resonance proportion, is the percentage of the total consumers who can generate cognition and emotional resonance on a certain audio-visual medium in a specific consumer group.
The blood oxygen concentration data of the present invention includes oxyhemoglobin concentration, deoxyhemoglobin, or total hemoglobin concentration. Preferably, the blood oxygen concentration data is oxyhemoglobin concentration.
Oxyhemoglobin has a better signal-to-noise ratio than deoxyhemoglobin or total hemoglobin concentration.
The invention has the following beneficial effects:
(1) the invention has more objectivity by using the data acquired by the fNIRS and has higher ecological efficiency in the evaluation process. Compared with the method for evaluating the audio-visual media by relying on the traditional behavior measurement method, the method for acquiring the brain activity data of the consumer by using the fNIRS can record the most direct and objective reaction of the consumer to the attitude of the audio-visual media in real time, and the fNIRS equipment has small limitation on the body of the tester and is closer to the state of watching the audio-visual media in a daily life scene.
(2) The audio-visual media evaluation based on the group brain network modeling is more consistent with the basic principle of media propagation, and can accurately depict the degree of emotion and cognitive resonance generated by a certain audio-visual media in a group.
(3) Providing two technical indexes of audio-visual media evaluation: firstly, calculating the percentage of the consumers which can generate cognition and emotional resonance on a certain audio-visual medium in a specific consumer group according to a general index (group resonance degree); secondly, a detail index is used for analyzing the degree of group resonance caused by a certain medium at each time point by using a sliding window technology. The general index can judge the general propagation effect of the audio-visual media, and the detail index is the basis for further cutting, modifying and optimizing the audio-visual media at specific time points and segments so as to realize better propagation effect.
Drawings
FIG. 1 is a flow chart of the present invention.
FIG. 2 is a schematic diagram of a reference Brahman partition marking underrun channel.
FIG. 3 is a schematic diagram of the present invention for collecting near infrared optical data.
FIG. 4 is a graphical representation of the population brain network density for high-score ads and low-score ads in example 1.
Detailed Description
The present invention will be described in detail with reference to specific examples. The following examples will assist those skilled in the art in further understanding the invention, but are not intended to limit the invention in any way. It should be noted that variations and modifications can be made by persons skilled in the art without departing from the spirit of the invention. All falling within the scope of the present invention.
Example 1 prediction of the evaluation of video advertisements by a small range of consumers using audio-visual media evaluation techniques based on a swarm brain network
Before the first test began, we first recruited 12 testers to rate 30 segments of ads and picked out 10 lower scoring ads and 10 higher scoring ads as test material. The goods or services displayed in the advertisement comprise mobile phones, watches, chocolate, cosmetics, travel agencies and the like, the advertisement simultaneously comprises sound and dynamic images, and the average time of the advertisement segments is 32.95 seconds.
In the first test we recruited 20 female consumers (average age 20.15 years) and rated the 20 advertisements that were picked.
Step 1, acquiring cerebral blood oxygen concentration data by using a near infrared spectrum cerebral function imaging technology:
as shown in FIG. 3, 20 persons to be measured sit in front of the computer screen for playing advertisement clips and wear a fNIRS apparatus (NIRS port; NIRX Medical Technologies, LLC, USA) which records the concentration changes of oxyhemoglobin, deoxyhemoglobin and total hemoglobin in the infraforehead region of the person to be measured, the sampling rate of the apparatus is 10Hz, and as shown in FIG. 2, 6 light sources and 6 detectors are distributed in the infraforehead region of the left and right sides, the interval between every two probes is 3 cm, wherein the measurement channels (the midpoint positions of one light source and one detector) of 6 two wavelengths (760nm and 850nm) cover the Bradymann regions 44 and 45 of the infraforehead region of the left and right sides, respectively. The tested person has a rest time of 30 seconds before watching each advertisement; the fmirs device is removed at the end of viewing.
During the viewing process, the tester indicates his/her favorite degree (7-point scale rating, 1 for no favorite, 7 for very favorite) and the price (7-point scale) that he/she is willing to pay for the products in the advertisement by means of keyboard input.
The person being tested, after removing the fNIRS device, indicated how well he or she understood each advertisement viewed by way of a paper-pen questionnaire (7-point rating scale, 1 for no understanding and 7 for full understanding).
Step 2, processing data:
according to the general standard of brain activity data processing, oxyhemoglobin concentration changes are noted at the time of data processing.
Step 2a, denoising:
as partial data is excessively noisy or lost during data acquisition, the effective data is 14 people times, and Matlab software is utilized to firstly carry out band-pass filtering (0.01-0.1Hz) on the original data of each channel and each test in a preprocessing stage so as to remove physiological noises such as respiration, heartbeat and the like and noises of a machine system. The oxyhemoglobin value of the brain region is represented by the average of the data for the bradmann regions 44 and 45 corresponding to the left forehead front gross grams (IFG); the mean of the data for bradmann areas 44 and 45 corresponding to the right IFG represents the oxyhemoglobin value for this brain area.
Step 2b baseline calibration:
and (3) performing baseline correction on the oxyhemoglobin value of each sub-frontal gyrus area by using Excel software and using the oxyhemoglobin concentration average value of each tested person in the last 10 seconds of the 30-second rest period.
Step 2c, converting the normalization into the blood oxygen concentration z fraction of each tested person:
since the fNIRS data are recorded as relative values rather than absolute values, and cannot be directly averaged among the tested persons or among the channels, the fNIRS data are converted into the blood oxygen concentration z-fraction by calculating the average value and standard deviation of the last 10 seconds of oxygenated hemoglobin concentration of all the tested persons during the rest period by using Excel software.
Step 2d, calculating a brain activity Pearson correlation coefficient:
statistical analyses were performed using SPSS statistical software, and the significance level for all analyses was set at p < 0.05.
Each tester is taken as a node in the network, and the network strength is calculated, namely the average value of the weight of each edge of the network (the Pearson correlation coefficient of brain activity needs to exceed a network model threshold value r & gt 0.01, and p & lt 0.05), wherein the network strength represents the consistency degree of the response of the testers to the advertisement co-estrus.
Step 2e, calculating the population brain network density:
and calculating the density of the brain network corresponding to each advertisement by using Excel software, wherein the network density represents how many testers have similar understanding and emotional reactions to the advertisements, namely the group resonance proportion. The experimental results are shown in the following tables 1-2, and the brain network schematic diagrams of the high-score advertisements and the low-score advertisements are shown in fig. 4.
TABLE 1
Figure BDA0002415904630000071
Figure BDA0002415904630000081
TABLE 2
Figure BDA0002415904630000082
Step 3, evaluating the transmission effect of the audio-visual media:
data analysis results show that in the left underforehead gyrus area of the testers, the high-score advertisements are higher in brain network density than the population formed by the low-score advertisements.
In the left subconcephalic region [45.90, ± 15.11vs.33.50, ± 9.94, t (18) ═ 2.168, p ═ 044, Cohen's d ═ 0.97 ]; similar results were obtained in the right subconcentral region [47.40, ± 17.73vs.32.70, ± 11.93, t (18) ═ 2.176, p ═ 0.043, Cohen's d ═ 1.01 ]. A larger value of t indicates a larger difference between groups, Cohen's d represents an effect amount, and a larger value indicates a larger experimental effect.
In addition, the results of regression analysis showed that the strength of the right-sided subtotal gyrocerebral network and the degree of comprehension of the testers to the advertisement (R)20.206, beta 0.028, p 0.045) and attitude (mean of the price and the like willing to pay, R20.223, 0.031, 0.036). When the consumer watches the advertisement, the attitude and the understanding degree of the consumer to the advertisement can be reflected by the strength of the right underforehead echo network, whether the advertisement causes resonance in the consumer is judged, and the method can be used for measuring the advertisement effect.
Example 2 prediction of music evaluation by Wide Range of consumers Using Audio visual media evaluation techniques based on the swarm brain network
In the second test we recruited 20 female testers (average age 19.20 years) and evaluated the selected musical pieces. The test material included 30 popular music songs from the top 15 (i.e., the most popular songs) and the bottom 15 (i.e., the less popular songs) of the "2014 top 100 songs" list published by the billboard. The length of time for selecting materials of songs is 15 seconds, and the materials mainly comprise a refrain part. In addition, we also collected scores (0 to 10 points) for each song by netizens in China, and the score data comes from the bean-net of the most popular online review website.
Step 1, acquiring cerebral blood oxygen concentration data by using a near infrared spectrum cerebral function imaging technology:
the test apparatus and brain area used in example 2 were identical to those used in example 1 with reference to the collection method of example 1.
The test procedure was consistent with example 1, and the tester listened to the music piece and indicated his/her own likeability (7-point scale rating, 1 for little dislike and 7 for strong like), comprehension (7-point scale rating, 1 for little dislike and 7 for full understanding) and price (7-point scale) willing to be paid for purchasing the song by keyboard input.
Step 2, processing data:
the data processing method was the same as that in example 1. The results are shown in tables 3 to 4.
TABLE 3
Figure BDA0002415904630000091
TABLE 4
Figure BDA0002415904630000092
Figure BDA0002415904630000101
Step 3, evaluating the transmission effect of the audio-visual media:
the data is analyzed, and the result shows that the brain network density of the group formed by the first 15 (i.e. the most popular songs) of the list is higher than that of the second 15 (i.e. the most unpopular songs), the right-side forehead return (Brodmann area 44, 45) of the tested person is in significant positive correlation with the broad bean score of the songs (R is significantly and positively correlated with the broad bean score of the songs)2＝0.188，beta＝0.043，p＝0.017)。
In addition, the density of the right-side lower-forehead group brain network is obviously and positively correlated with the attitude of the tested person to the song (R2 ═ 180, beta ═ 047, and p ═ 019). Basically, the comprehension of a song can also be predicted (R2 ═ 127, beta ═ 032, p ═ 053). And (4) conclusion: the results of example 2 show that, in the process of listening to the audio, the crowd brain network density of the right forehead back can reflect the likeness of the mass consumer to the audio, and judge whether the audio causes resonance in the consumer.

Claims (10)

1. A method for evaluating the transmission effectiveness of an audio-visual medium is characterized by comprising the following steps:
step 1: acquiring cerebral blood oxygen concentration data by using a near infrared spectrum cerebral function imaging technology:
respectively presenting the audio-visual media to each tested person, and continuously measuring the blood oxygen concentration data of the specific head position preset by each tested person by using a near infrared spectrum brain function imaging technology during presentation; the number of the tested persons is not less than 10 persons;
step 2: data processing:
denoising the blood oxygen concentration data of each measured person acquired in the step (1), calibrating a base line, and converting the data into a blood oxygen concentration z fraction of each measured person in a standardized manner; taking each tested person as a node in a group brain network, and calculating a pearson correlation coefficient of brain activity between every two tested persons according to the blood oxygen concentration z fraction of each tested person; further calculating the population brain network density according to the pearson correlation coefficient of the brain activity;
the pearson correlation coefficient of brain activity is calculated by a calculation formula 2, wherein the calculation formula 2 is as follows:
Figure FDA0002919081990000011
wherein, the blood oxygen concentration z scores of any two tested persons of X and Y, cov (X, Y) is the covariance of X and Y, Var [ X]Variance of X, Var [ Y ]]Is the variance of Y;
the method for calculating the population brain network density specifically comprises the steps of taking the brain of each tested person as a node in the population brain network, and calculating through a formula 2 to obtain the brain activity synchronism between every two tested persons; if a pair of tested persons has obvious brain activity synchronism, calculating that one edge of the brain network of the group exists; the same method can obtain the number of the edges coexisting in the brain network of the group; taking the sum of all edges as the brain network density of the population;
and step 3: evaluation of audiovisual medium propagation effect:
evaluating the transmission effect of the audio-visual media according to the population brain network density obtained in the step 2, wherein the higher the population brain network density is, the better the transmission effect is.
2. The rating method of claim 1, wherein the audiovisual media is a video, music or movie trailer.
3. The method of claim 1, wherein the predetermined specific head position is a left or right subfrontal gyrocembria region.
4. The method of claim 1, wherein the predetermined head-specific locations are regions 44 and 45 of the bradmann area.
5. The evaluation method according to claim 1, wherein the number of the persons to be tested is 18 to 32.
6. The evaluation method according to claim 1, wherein the baseline calibration in step 2 is performed by taking the denoised resting blood oxygen concentration of each measured person as a baseline.
7. The evaluation method according to claim 6, wherein the method for measuring resting blood oxygen concentration comprises the steps of: and respectively and continuously measuring the blood oxygen concentration data of each tested person, which is exposed to an empty screen and has a rest time of 30-60 seconds, by using a near-infrared spectrum brain function imaging technology, and taking the average value of the blood oxygen concentration data of each sampling point of the last 10-15 seconds of the rest time of each tested person as the average concentration of the rest blood oxygen of each tested person.
8. The rating method according to claim 1, wherein the number of the audiovisual media is 1; the blood oxygen concentration data in the step 1 are instantaneous blood oxygen concentration data at different moments; the population brain network density in the step 2 is the instantaneous brain network density at different moments; the population brain network density in the step 3 is the instantaneous brain network density obtained in the step 2, and the propagation effect is the propagation effect at different moments.
9. The rating method of claim 1, wherein the number of audiovisual media is greater than 1; the blood oxygen concentration data in the step 1 is the average blood oxygen concentration data of each audio-visual media; the population brain network density in the step 2 is the total brain network density of each audio-visual media; the population brain network density in the step 3 is the total brain network density of each audio-visual media obtained in the step 2, and the propagation effect is the total propagation effect.
10. The evaluation method according to any one of claims 1 and 6 to 9, wherein the blood oxygen concentration data is an oxyhemoglobin concentration.
CN202010193942.1A 2020-03-18 2020-03-18 Audio-visual media evaluation method based on group brain network Active CN111399650B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010193942.1A CN111399650B (en) 2020-03-18 2020-03-18 Audio-visual media evaluation method based on group brain network

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010193942.1A CN111399650B (en) 2020-03-18 2020-03-18 Audio-visual media evaluation method based on group brain network

Publications (2)

Publication Number Publication Date
CN111399650A CN111399650A (en) 2020-07-10
CN111399650B true CN111399650B (en) 2021-04-20

Family

ID=71434399

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010193942.1A Active CN111399650B (en) 2020-03-18 2020-03-18 Audio-visual media evaluation method based on group brain network

Country Status (1)

Country Link
CN (1) CN111399650B (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112057088B (en) * 2020-08-18 2024-01-05 天津大学 Brain region positioning method related to audio-visual mode emotion voice processing based on searchlight
CN112232861A (en) * 2020-10-10 2021-01-15 上海外国语大学 Plane advertisement evaluation method and system based on neural similarity analysis
CN112545519B (en) * 2021-02-22 2021-06-04 之江实验室 Real-time assessment method and system for group emotion homogeneity
CN113476057B (en) * 2021-07-08 2023-04-07 先端智能科技(天津)有限公司 Content evaluation method and device, electronic device and storage medium

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050027173A1 (en) * 2003-07-31 2005-02-03 Briscoe Kathleen E. Brain injury protocols
CN109767088A (en) * 2018-12-27 2019-05-17 中国矿业大学 The quantitative assessment of fabric tactile texture and identifying system based on multi-mode brain-computer interface
CN110428884A (en) * 2019-08-09 2019-11-08 丹阳慧创医疗设备有限公司 A kind of training method of memory capability, device, equipment and system

Also Published As

Publication number Publication date
CN111399650A (en) 2020-07-10

Similar Documents

Publication Publication Date Title
CN111399650B (en) Audio-visual media evaluation method based on group brain network
Li et al. Using skin conductance and facial electromyography to measure emotional responses to tourism advertising
US11200964B2 (en) Short imagery task (SIT) research method
Tan et al. Repeatability of facial electromyography (EMG) activity over corrugator supercilii and zygomaticus major on differentiating various emotions
US8635105B2 (en) Consumer experience portrayal effectiveness assessment system
US8392254B2 (en) Consumer experience assessment system
US8548852B2 (en) Effective virtual reality environments for presentation of marketing materials
US20120259240A1 (en) Method and System for Assessing and Measuring Emotional Intensity to a Stimulus
US20100004977A1 (en) Method and System For Measuring User Experience For Interactive Activities
US20130185142A1 (en) Content based selection and meta tagging of advertisement breaks
WO2008154410A1 (en) Multi-market program and commercial response monitoring system using neuro-response measurements
JP2012511397A (en) Brain pattern analyzer using neural response data
US20030032890A1 (en) Continuous emotional response analysis with facial EMG
van den Broek et al. Considerations for emotion-aware consumer products
Ciceri et al. A neuroscientific method for assessing effectiveness of digital vs. Print ads: Using biometric techniques to measure cross-media ad experience and recall
Niu et al. User experience evaluation in virtual reality based on subjective feelings and physiological signals
Holman et al. Taking the Subjectivity out of UX Evaluation with Emotiv EPOC+
Lei et al. Neurophysiological measures in hospitality and tourism: Review, critique, and research agenda
Stuldreher et al. EEG measures of attention toward food-related stimuli vary with food neophobia
Borawska et al. The use of neurophysiological measures in studying social advertising effectiveness
Li et al. Unveiling the secrets of online consumer choice: A deep learning algorithmic approach to evaluate and predict purchase decisions through EEG responses
Ward An analysis of facial movement tracking in ordinary human–computer interaction
Nascimben et al. A minimal setup for spontaneous smile quantification applicable for valence detection
Sameeha et al. Analysis And Evaluation Of Consumers’ Cognitive Responses To Visual And Gustatory Stimuli For Neuromarketing Application
González-Morales et al. Neuromarketing applied to communication as neurocommunication in WOS: bibliometric reviews and a guide to the first twenty years of the 21st century, from 2001 to 2020

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant