阅读说明：本技术 基于脑电信号的创造力人格特质测量方法及装置 (Creativity personality trait measuring method and device based on electroencephalogram signals ) 是由 张丹 胡鑫 王非 陈菁菁 于 2019-09-16 设计创作，主要内容包括：本发明提供一种基于脑电信号的创造力人格特质测量方法及装置,方法包括：对自然语言素材的文本进行词汇切分,并获取切分的每个词汇在所述文本的上下文中的出现概率；获取受测者在听到所述自然语言素材的音频时的脑电信号,并对所述脑电信号进行切分,获取每个词汇对应的脑电响应片段；根据每个词汇的出现概率和每个词汇对应的脑电响应片段,获取所述出现概率的冲击响应函数；根据所述冲击响应函数,基于预先训练好的创造力人格特质预测模型获取所述受测者的创造力人格特质测试得分。本发明实现对创造力人格特质的自动测量,不易受外界因素影响,且测量更加准确。(The invention provides a creativity personality trait measuring method and device based on electroencephalogram signals, wherein the method comprises the following steps: carrying out vocabulary segmentation on a text of a natural language material, and acquiring the occurrence probability of each segmented vocabulary in the context of the text; acquiring an electroencephalogram signal of a testee when the testee hears the audio frequency of the natural language material, segmenting the electroencephalogram signal, and acquiring an electroencephalogram response segment corresponding to each vocabulary; acquiring an impulse response function of the occurrence probability according to the occurrence probability of each vocabulary and the electroencephalogram response segment corresponding to each vocabulary; and obtaining the creativity personality trait test score of the testee based on a pre-trained creativity personality trait prediction model according to the impact response function. The invention realizes the automatic measurement of creativity personality traits, is not easily influenced by external factors, and has more accurate measurement.)

1. A creativity personality trait measurement method based on electroencephalogram signals is characterized by comprising the following steps:

carrying out vocabulary segmentation on a text of a natural language material, and acquiring the occurrence probability of each segmented vocabulary in the context of the text;

acquiring an electroencephalogram signal of a testee when the testee hears the audio frequency of the natural language material, segmenting the electroencephalogram signal, and acquiring an electroencephalogram response segment corresponding to each vocabulary;

acquiring an impulse response function of the occurrence probability according to the occurrence probability of each vocabulary and the electroencephalogram response segment corresponding to each vocabulary;

and obtaining the creativity personality trait test score of the testee based on a pre-trained creativity personality trait prediction model according to the impact response function.

2. The method of measuring creative personality traits based on electroencephalography signals of claim 1, characterized in that the step of vocabulary-segmenting a text of a natural language material and obtaining an occurrence probability of each segmented vocabulary in a context of the text comprises:

segmenting words of a text of a natural language material based on a computational linguistics method to obtain each word in the natural language material;

and calculating the occurrence probability of each vocabulary in the context of the text based on a Chinese language probability model according to a pre-constructed text corpus.

3. The creativity personality trait measurement method based on electroencephalogram signals of claim 1, wherein the step of segmenting the electroencephalogram signals to obtain electroencephalogram response segments corresponding to each vocabulary specifically comprises:

acquiring the starting playing time of each vocabulary relative to the playing starting time of the audio;

and segmenting the electroencephalogram signal according to the playing starting time of each vocabulary to obtain an electroencephalogram response segment corresponding to each vocabulary.

4. The method for measuring creative personality traits based on electroencephalogram signals of claim 3, wherein the step of segmenting the electroencephalogram signals according to the play-starting time of each vocabulary comprises:

and for the electroencephalogram signal of each channel, subtracting the average value of the electroencephalogram signals of a second preset time before the starting playing time of any vocabulary from the electroencephalogram signal of the first preset time after the starting playing time of any vocabulary, and obtaining the electroencephalogram signal after the baseline correction of each channel.

5. The method for measuring the creativity personality trait based on the electroencephalogram signal of claim 1, wherein the step of obtaining the impulse response function of the occurrence probability according to the occurrence probability of each vocabulary and the electroencephalogram response segment corresponding to each vocabulary comprises the steps of:

constructing a system response model which takes the occurrence probability as input and takes the electroencephalogram response segment corresponding to the occurrence probability as output;

solving an impact response function of the system response model; and the impulse response function is used for expressing the electroencephalogram activity mode corresponding to the occurrence probability of the vocabulary.

6. The method for measuring creativity personality trait based on electroencephalograph signals of any one of claims 1-5, wherein the step of obtaining the creativity personality trait test score of the subject based on a pre-trained creativity personality trait prediction model according to the impact response function comprises:

extracting curve characteristics of the impact response function;

and taking the curve characteristics as the input of the creativity personality trait prediction model to obtain the creativity personality trait test score of the testee.

7. The method for measuring creativity personality trait based on electroencephalographic signals of any one of claims 1-5, wherein the creativity personality trait prediction model is a regression model;

correspondingly, the step of obtaining the creativity personality trait test score of the testee based on the pre-trained creativity personality trait prediction model according to the impact response function comprises the following steps:

constructing a regression model by taking the questionnaire score of the creativity personality traits of the individual as a target; wherein the regression model is a LASSO regression model;

adopting a traditional creativity personality trait measurement method to perform questionnaire scoring on creativity personality traits of training individuals;

and training the regression model according to the questionnaire scores of the training individuals.

8. The utility model provides a creativity personality trait measuring device based on brain electrical signal which characterized in that includes:

the system comprises a first segmentation module, a second segmentation module and a third segmentation module, wherein the first segmentation module is used for carrying out vocabulary segmentation on a text of a natural language material and acquiring the occurrence probability of each segmented vocabulary in the context of the text;

the second segmentation module is used for acquiring an electroencephalogram signal of a testee when the testee hears the audio frequency of the natural language material, segmenting the electroencephalogram signal and acquiring an electroencephalogram response segment corresponding to each vocabulary;

the calculation module is used for acquiring an impulse response function of the occurrence probability according to the occurrence probability of each vocabulary and the electroencephalogram response segment corresponding to each vocabulary;

and the measuring module is used for obtaining the creativity personality trait test score of the testee based on a pre-trained creativity personality trait prediction model according to the impact response function.

9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the steps of the method for creativity personality trait measurement based on electroencephalographic signals of any one of claims 1 to 8 are implemented when the program is executed by the processor.

10. A non-transitory computer-readable storage medium having stored thereon a computer program, wherein the computer program when executed by a processor implements the steps of the method for creativity personality trait measurement based on electroencephalographic signals of any one of claims 1 to 8.

Technical Field

The invention belongs to the technical field of electroencephalogram signal analysis, and particularly relates to a creativity personality trait measuring method and device based on electroencephalogram signals.

Background

The individual creativity personality traits refer to the sum of non-intellectual traits such as ideal, belief, will, emotion, moral and the like which are excellent in guiding and determining the functions of the individual to express and develop in the creativity and promoting the generation of the creativity. One of the significant features of individuals with high creative personality is that they have a more divergent thinking ability in addressing challenging events in learning, work or life, and can generate more diverse thoughts to more effectively address these challenges.

Although the study of creative personality traits has been a major concern in some areas of science, the measurement of creative personality traits has been challenging. Specifically, the conventional mainstream measurement means is developed mainly based on a questionnaire manner, such as a multipurpose Test (Alternative Use Test), a Remote association Test (Remote association Test), and the like. These test questionnaires provide final measurements by allowing subjects to answer questions related to creativity ability, the answers being evaluated by the panelist based on professional experience. Such creativity measurement mode is easily influenced by individual factors such as social approval, individual state and the like, and is more easily interfered particularly when a target individual is in an environment such as election competition and the like.

Meanwhile, most of the existing creativity measurement questionnaires lack standardized scores, the final score is determined by testers to a large extent and is easily influenced by personal levels or trends of the testers, for example, in a multipurpose test, how the testees answer can be considered as original and creative, and the standard of the creativity measurement questionnaires is not enough objective. In addition, due to the high participation degree of the testers, the creation of the force measurement questionnaire to obtain the results often requires a relatively long time for the subsequent questionnaire review, which consumes a lot of time and manpower components, and the result feedback timeliness is not good.

Disclosure of Invention

In order to overcome the problems that the existing creative personality trait measurement method is easily interfered by the environment, has no objective standard and wastes time and labor or at least partially solves the problems, the embodiment of the invention provides a creative personality trait measurement method and a device based on electroencephalogram signals.

According to a first aspect of the embodiments of the present invention, there is provided a creativity personality trait measurement method based on electroencephalogram signals, including:

carrying out vocabulary segmentation on a text of a natural language material, and acquiring the occurrence probability of each segmented vocabulary in the context of the text;

acquiring an electroencephalogram signal of a testee when the testee hears the audio frequency of the natural language material, segmenting the electroencephalogram signal, and acquiring an electroencephalogram response segment corresponding to each vocabulary;

acquiring an impulse response function of the occurrence probability according to the occurrence probability of each vocabulary and the electroencephalogram response segment corresponding to each vocabulary;

and obtaining the creativity personality trait test score of the testee based on a pre-trained creativity personality trait prediction model according to the impact response function.

According to a second aspect of the embodiments of the present invention, there is provided a creativity personality trait measuring device based on electroencephalogram signals, including:

the system comprises a first segmentation module, a second segmentation module and a third segmentation module, wherein the first segmentation module is used for carrying out vocabulary segmentation on a text of a natural language material and acquiring the occurrence probability of each segmented vocabulary in the context of the text;

the second segmentation module is used for acquiring an electroencephalogram signal of a testee when the testee hears the audio frequency of the natural language material, segmenting the electroencephalogram signal and acquiring an electroencephalogram response segment corresponding to each vocabulary;

the calculation module is used for acquiring an impulse response function of the occurrence probability according to the occurrence probability of each vocabulary and the electroencephalogram response segment corresponding to each vocabulary;

and the measuring module is used for obtaining the creativity personality trait test score of the testee based on a pre-trained creativity personality trait prediction model according to the impact response function.

According to a third aspect of the embodiments of the present invention, there is also provided an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor calls the program instructions to execute the method for measuring creativity personality trait based on electroencephalogram signals provided in any one of the various possible implementations of the first aspect.

According to a fourth aspect of embodiments of the present invention, there is also provided a non-transitory computer-readable storage medium storing computer instructions for causing a computer to execute the method for measuring creative personality traits based on electroencephalogram signals provided in any one of the various possible implementations of the first aspect.

The embodiment of the invention provides a creativity personality trait measuring method and a creativity personality trait measuring device based on an electroencephalogram signal, the method is characterized in that words in the natural language material are segmented and the occurrence probability of each word in the context is quantified, the brain-computer interface method is used for obtaining the brain electrical signals when the testee receives the natural language materials in a hearing way, according to the occurrence probability and the electroencephalogram response segment of each vocabulary, the electroencephalogram response mode corresponding to the occurrence probability is expressed by using an impulse response function, obtaining the creativity personality trait test score of the testee based on the creativity personality trait prediction model according to the electroencephalogram response mode of the testee, therefore, the automatic measurement of the creativity personality traits is realized, the influence of external factors is not easy to influence, objective electroencephalogram signals are collected, the creativity personality traits of a testee are reflected more truly, and the measurement result is more accurate.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and those skilled in the art can also obtain other drawings according to the drawings without creative efforts.

FIG. 1 is a schematic overall flow chart of a creative personality trait measurement method based on electroencephalogram signals provided by an embodiment of the invention;

fig. 2 is a schematic diagram of a measurement flow based on equipment in the creativity personality trait measurement method based on electroencephalogram signals provided by the embodiment of the invention;

FIG. 3 is a schematic flow chart of a creative personality trait measurement method based on electroencephalogram signals according to another embodiment of the present invention;

FIG. 4 is a schematic diagram of the overall structure of the creativity personality trait measuring device based on electroencephalogram signals provided by the embodiment of the invention;

fig. 5 is a schematic view of an overall structure of an electronic device according to an embodiment of the present invention.

Detailed Description

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and those skilled in the art can also obtain other drawings according to the drawings without creative efforts.

In an embodiment of the present invention, a method for measuring creativity personality trait based on electroencephalogram signals is provided, and fig. 1 is an overall flow diagram of the method for measuring creativity personality trait based on electroencephalogram signals, which is provided by the embodiment of the present invention, and includes: s101, segmenting words of a text of a natural language material, and acquiring the occurrence probability of each segmented word in the context of the text;

starting from the test requirement, the natural language material should include words with occurrence probability levels as rich as possible, and situations with large differences among individuals with possible occurrence probability levels, such as excessive professional terms, are avoided in the natural language material. At the same time, the natural language material needs to be rendered in as short a time as possible in order for the test to be reliably conducted with timeliness.

Therefore, the embodiment needs to select natural language material texts describing popular contents such as literary works of daily life or classical stories, news broadcasts and the like as the basis of the testing material library. The natural language material can be from Chinese textbooks, Mandarin level examinations, Chinese level examinations of foreigners, etc. The texts from the sources are easy to understand, have no uncommon words or professionally-defined concept terms, and are beneficial to ensuring that the testee accurately understands the contents. And then inviting personnel with professional broadcast training to read the natural language material texts, and taking the recorded read audio as a testing material library to be actually used.

Taking the test completed within 10 minutes as an example, the present embodiment suggests selecting 3-4 texts with a text space of 300-500 words as the actual material for one test. In order to avoid repeated familiarity effect of the testee in multiple tests, 100 or more sections of texts are prepared as a test material library.

The embodiment divides the text of the natural language material into a plurality of words and obtains the occurrence probability of each divided word in the context of the text. The probability of occurrence calculation needs to be performed according to a Chinese language probability model obtained based on a massive text corpus, such as a common Word2Vec model, an N-gram model and the like, and then a certain specific vocabulary i is calculated according to a specific context of a certain text, wherein a conceptual formula of the probability calculation can be expressed as follows:

the probability of occurrence of the word i is P (word i | the first i-1 words of the text).

This embodiment performs a measure of the personality traits of individual creativity by examining the subject's responses to words of different probabilities of occurrence in the natural language material. Compared with the highly abstract materials commonly used in the traditional psychological research, the natural language materials can be used for enabling the testee to be blended into the context as much as possible, so that the state more conforming to the real life situation is shown, and the test result of the embodiment has better practical application and popularization performance.

S102, acquiring an electroencephalogram signal of a testee when the testee hears the audio frequency of the natural language material, segmenting the electroencephalogram signal, and acquiring an electroencephalogram response segment corresponding to each vocabulary;

by means of scalp electroencephalography, the cranial nerve response of a subject is recorded while receiving audio of specific natural language material presented audibly. And segmenting the electroencephalogram signals corresponding to the acquired natural language materials to acquire an electroencephalogram response segment corresponding to each vocabulary.

Aiming at the application scene of the embodiment, a series of brain nerve signal specific feature extraction and machine learning methods are applied to quickly and accurately identify individual specific creativity personality trait information contained in electroencephalogram signals corresponding to different occurrence probability vocabularies, so that automatic measurement is realized. Compared with the traditional method, the personality trait measurement method based on the brain-computer interface technology relies on objective brain neuroelectrical activity information to carry out personality trait identification, a testee cannot disguise, the influence of the personal ability or tendency of the testee is avoided, a large amount of time and labor cost can be saved, and the method is a more objective and efficient method. However, most of the related research based on brain-computer interface technology focuses on the state factors such as emotion and thinking of people, and a method for measuring the character of the creative personality is not reported.

S103, acquiring an impulse response function of the occurrence probability according to the occurrence probability of each vocabulary and the electroencephalogram response segment corresponding to each vocabulary;

the complexity of continuous presentation of vocabularies in the natural language material and the occurrence probability level of each vocabulary being a continuous numerical value is an important difficulty and challenge for data analysis in this embodiment. In the embodiment, an Impulse Response Function (Impulse Response Function) based on a system representation method is adopted to embody the electroencephalogram Response characteristics of an individual to vocabularies with different occurrence probabilities. The impulse response function represents an individual electroencephalogram neural activity mode corresponding to the vocabulary occurrence probability input of a certain abstract unit. Individuals with different creativity personality traits correspond to different electroencephalogram activity modes expressed in an impact response function mode.

People with different creativity personality traits have different electroencephalogram response modes when facing specific external stimulation, for example, an inward person and an outward person can excite different modes of electroencephalogram signals when facing the same natural language materials, such as pictures, sounds, characters and the like. Representative electroencephalogram responses sensitive to creative personality include an early negative component EPN of about 200 milliseconds after the occurrence of audio stimulation of natural language material, a late electroencephalogram positive component LPP of about 400 milliseconds, a late negative component N400 of 400-. By effectively identifying the electroencephalogram response modes with different time and space characteristics when the testee faces a specific natural language material, the testees with different creativity personality traits can be effectively distinguished.

And S104, obtaining the creativity personality trait test score of the testee based on a pre-trained creativity personality trait prediction model according to the impact response function.

And analyzing the electroencephalogram response mode of the individual by using a pre-trained creativity personality trait prediction model according to an impact response function, namely the electroencephalogram response mode of the individual facing the audio stimulation of the natural language material, and determining the creativity personality trait test score of the testee.

In order to obtain electroencephalogram data which is effective enough to better measure the creativity personality traits of a testee, the test device in the embodiment comprises the following components:

1) the individual electroencephalogram data recording component can record not less than 16 channels of electroencephalograms simultaneously, the covered electrodes comprise FP1, FP2, Fz, F3, F4, T3, T4, Cz, Pz, Oz, O1, O2, C3, C4, P3 and P4, and the sampling rate is not lower than 200 Hz;

2) the air conduction earphone based on the air conduit can realize presenting voice auditory information in a non-electromagnetic interference mode;

3) the special hardware module can realize accurate time synchronization of the voice signal and the electroencephalogram equipment, can convert the played voice signal into an input signal voltage range (+/-within 200 millivolts) of the electroencephalogram equipment in real time and input the input signal into the electroencephalogram equipment, and synchronously records the input signal and the electroencephalogram signals of 16 channels, wherein the time synchronization error of the two types of signals is less than 2 milliseconds;

4) the computer can support test material playing, electroencephalogram data acquisition and electroencephalogram data analysis, a CPU is not lower than the 7 th generation i5 of Intel, the internal memory is not lower than 4GB, and the hard disk space is not lower than 256 GB.

The flow of the creative personality trait test using the above-described components is shown in fig. 2.

According to the embodiment, through segmenting the words in the natural language material and quantifying the occurrence probability of each word in the context, the electroencephalogram signal of a testee when the testee receives the natural language material in a hearing mode is obtained by using a brain-computer interface method, the electroencephalogram response mode corresponding to the occurrence probability is expressed by using an impact response function according to the occurrence probability and the electroencephalogram response segment of each word, and the creativity personality trait test score of the testee is obtained according to the electroencephalogram response mode of the testee based on a creativity personality trait prediction model, so that the automatic measurement of the creativity trait is realized, the influence of external factors is not easily caused, objective electroencephalogram signals are collected to reflect the creativity personality trait of the testee more truly, and the creativity personality trait of the testee is measured more accurately.

On the basis of the foregoing embodiment, in this embodiment, the step of performing vocabulary segmentation on the text of the natural language material and acquiring the occurrence probability of each segmented vocabulary in the context of the text includes: segmenting words of a text of a natural language material based on a computational linguistics method to obtain each word in the natural language material; and calculating the occurrence probability of each vocabulary in the context of the text based on a Chinese language probability model according to a pre-constructed text corpus.

Specifically, the vocabulary of the natural language material text which is read aloud is segmented by a computational linguistics method, and the segmentation is carried out by taking a minimum unit which retains Chinese semantics as a unit. According to a Chinese language probability model obtained from a massive corpus, calculating and recording the occurrence probability of the obtained single vocabulary in the context of the corresponding text, and taking the single vocabulary as core reference information for the subsequent electroencephalogram signal analysis.

The concept of occurrence probability is now further explained. For example, if "Xiaoming fruit shop buys an apple", where the probability of occurrence of the word "apple" in context is relatively high, and if the sentence is changed to "Xiaoming fruit shop buys a hammer", the probability of occurrence of the "hammer" is low because it is an unnatural contextual connection. The calculation of the occurrence probability depends on the language model learning from a massive text corpus, and the complex interconnection relation among different vocabularies in natural language materials can be learned through massive Chinese texts and expressed in the occurrence probability mode. In particular, it is learned based on mass language material that the population is a general, average expectation of the probability that a certain vocabulary will appear in a particular context.

The innovation of the application of the embodiment on the basis of the linguistic occurrence probability is that the common and average expectation of the occurrence probability of a certain vocabulary in a specific context is obtained based on the mass language materials. On the basis, the embodiment pays attention to individual differences, namely, people with different creativity personality trait levels have different reactions to the occurrence probability of a certain word under a specific context situation. In particular, highly creative people have a relatively higher expectation of words with a lower average probability of occurrence, and appear more forgiving, i.e., these words are more easily conceived by them; and low creativity people cannot easily think of words with low occurrence probability, and the psychological expectation of words with low average occurrence probability is insufficient, so that the words are more surprised or uncomfortable. Further, the surprise or the maladaptive reaction to the vocabulary can be better and quantitatively depicted through the electroencephalogram response components.

On the basis of the foregoing embodiment, the step of segmenting the electroencephalogram signal and acquiring an electroencephalogram response segment corresponding to each vocabulary in this embodiment specifically includes: acquiring the starting playing time of each vocabulary relative to the playing starting time of the audio; and segmenting the electroencephalogram signal according to the playing starting time of each vocabulary to obtain an electroencephalogram response segment corresponding to each vocabulary.

Specifically, the recorded reading voice frequency is subjected to time domain segmentation according to words in the natural language material text, and the starting playing time of each word relative to the audio playing starting time is obtained and is used as a basis for segmenting the electroencephalogram response segments corresponding to each word. Presenting natural language materials to a testee in an auditory way, and extracting multichannel electroencephalogram response segments induced by presentation time of all vocabularies according to the starting playing time of all the vocabularies in the natural language materials. The electroencephalogram response segments of the vocabulary and the occurrence probability of the vocabulary together provide key basic information for electroencephalogram analysis.

On the basis of the foregoing embodiment, the segmenting the electroencephalogram signal according to the play start time of each vocabulary in this embodiment includes: and for the electroencephalogram signal of each channel, subtracting the average value of the electroencephalogram signals of a second preset time before the starting playing time of any vocabulary from the electroencephalogram signal of the first preset time after the starting playing time of any vocabulary, and obtaining the electroencephalogram signal after the baseline correction of each channel.

Specifically, the present embodiment performs necessary baseline correction on the electroencephalogram signals of each channel. Wherein the first preset time period may be set to 800 msec, and the second preset time period may be set to 200 msec.

On the basis of the above embodiment, in this embodiment, the step of obtaining the impulse response function of each occurrence probability according to the occurrence probability of each vocabulary and the electroencephalogram response segment corresponding to each vocabulary includes: constructing a system response model which takes the occurrence probability as input and takes the electroencephalogram response segment corresponding to the occurrence probability as output; solving an impact response function of the system response model; and the impulse response function is used for expressing the electroencephalogram activity mode corresponding to the occurrence probability of the vocabulary.

On the basis of the foregoing embodiments, in this embodiment, the step of obtaining the creativity personality trait test score of the subject based on the pre-trained creativity personality trait prediction model according to the impact response function includes: extracting curve characteristics of the impact response function; and taking the curve characteristics as the input of the creativity personality trait prediction model to obtain the creativity personality trait test score of the testee.

The curve characteristics of the impulse response function include the amplitude and latency of each delay time, and the like.

On the basis of the above embodiments, the creative personality trait prediction model in this embodiment is a regression model; correspondingly, the step of obtaining the creativity personality trait test score of the testee based on the pre-trained creativity personality trait prediction model according to the impact response function comprises the following steps: constructing a regression model by taking the questionnaire score of the creativity personality traits of the individual as a target; wherein the regression model is a LASSO regression model; adopting a traditional creativity personality trait measurement method to perform questionnaire scoring on creativity personality traits of training individuals; and training the regression model according to the questionnaire scores of the training individuals.

Specifically, in order to construct a creativity personality trait prediction model capable of reliably predicting individual creativity personality traits, the electroencephalogram data analysis is carried out on individuals of not less than 100 persons, and electroencephalogram activity patterns related to the occurrence probability level of vocabularies expressed by the individuals through an impact response function are extracted. Of these, attention is focused, but not limited to, on the early negative component EPN of about 200 milliseconds after the occurrence of the stimulus, the late brain electrical positive component LPP of about 400 milliseconds, and the late negative component N400 of 400-700 milliseconds. Meanwhile, creativity questionnaire data represented by the multipurpose test and the remote association test of the individuals are collected, and the sum score of the questionnaire scores of the individuals is used as the basis for training the creativity personality trait prediction model, so that the learning and training of specific parameters of the creativity personality trait prediction model are completed.

The flow shown in fig. 3 is used for acquiring the electroencephalogram data of the individual, and the system impulse response function corresponding to the vocabulary occurrence probability of each volunteer is calculated. And constructing a regression model by taking the questionnaire score of the individual creativity personality traits as a target. And taking the curve characteristics of the impact response function as the input of the regression model to obtain the prediction score of the individual creativity personality traits. And comparing the prediction scores with the questionnaire scores, and adjusting the parameters of the regression model according to the comparison result. Considering that the number of impulse response function features from the brain electricity may be large, a sparse regression method represented by LASSO (Least Absolute shrinkage and Selection Operator) may be adopted.

After the training of the creativity personality trait prediction model is completed, the method can be applied to automatic measurement of creativity personality traits. And (3) for a new testee, carrying out data acquisition according to the flow shown in the figure 3 to obtain a system impact response function corresponding to the vocabulary occurrence probability, and calculating to obtain the testing score of the character of the creativity of the testee according to the trained character prediction model of the character of the creativity.

In another embodiment of the invention, an inventive personality trait measurement device based on brain electrical signals is provided for implementing the methods of the foregoing embodiments. Therefore, the description and definition in the embodiments of the aforementioned method for measuring the personality traits of the creativity based on electroencephalogram signals can be used for understanding the various execution modules in the embodiments of the present invention. Fig. 4 is a schematic diagram of an overall structure of a creative personality trait measuring device based on electroencephalogram signals, the device including a first segmentation module 401, a second segmentation module 402, a calculation module 403, and a measurement module 404, where:

the first segmentation module 401 is configured to perform vocabulary segmentation on a text of a natural language material, and obtain an occurrence probability of each segmented vocabulary in a context of the text;

the first segmentation module 401 segments a text of a natural language material into a plurality of words and obtains an occurrence probability of each segmented word in a context of the text. The calculation of the occurrence probability requires that a Chinese language probability model is constructed based on a massive text corpus, and then the occurrence probability of a certain specific vocabulary is calculated according to the context of a specific text.

The second segmentation module 402 is configured to obtain an electroencephalogram signal when the subject hears the audio of the natural language material, segment the electroencephalogram signal, and obtain an electroencephalogram response segment corresponding to each vocabulary;

the second segmentation module 402 records the cranial nerve response of the subject while receiving audio of specific natural language material presented audibly through scalp electroencephalography. And segmenting the electroencephalogram signals corresponding to the acquired natural language materials to acquire an electroencephalogram response segment corresponding to each vocabulary.

The calculation module 403 is configured to obtain an impulse response function of each occurrence probability according to the occurrence probability of each vocabulary and the electroencephalogram response segment corresponding to each vocabulary;

the calculation module 403 adopts an impulse response function based on the system representation method to embody the electroencephalogram response characteristics of the individual to words with different occurrence probabilities. The impulse response function represents an individual electroencephalogram neural activity mode corresponding to the vocabulary occurrence probability input of a certain abstract unit. Individuals with different creativity personality traits have different electroencephalographic response patterns in the face of specific external stimuli.

The measuring module 404 is configured to obtain a creativity personality trait test score of the subject based on a pre-trained creativity personality trait prediction model according to the impact response function.

The measurement module 404 analyzes the electroencephalogram response pattern of the individual using a pre-trained creativity personality trait prediction model according to the impulse response function, i.e., the electroencephalogram response pattern of the individual when facing the audio stimulation of the natural language material, to determine the creativity personality trait test score of the subject.

According to the embodiment, through segmenting the words in the natural language material and quantifying the occurrence probability of each word in the context, the electroencephalogram signal of a testee when the testee receives the natural language material in a hearing mode is obtained by using a brain-computer interface method, the electroencephalogram response mode corresponding to the occurrence probability is expressed by using an impact response function according to the occurrence probability and the electroencephalogram response segment of each word, and the creativity personality trait test score of the testee is obtained according to the electroencephalogram response mode of the testee based on a creativity personality trait prediction model, so that the automatic measurement of the creativity trait is realized, the influence of external factors is not easily caused, objective electroencephalogram signals are collected to reflect the creativity personality trait of the testee more truly, and the creativity personality trait of the testee is measured more accurately.

On the basis of the above embodiment, the first division module in this embodiment is specifically configured to: segmenting words of a text of a natural language material based on a computational linguistics method to obtain each word in the natural language material; and calculating the occurrence probability of each vocabulary in the context of the text based on a Chinese language probability model according to a pre-constructed text corpus.

On the basis of the above embodiment, in this embodiment, the second cutting module is specifically configured to: acquiring the starting playing time of each vocabulary relative to the playing starting time of the audio; and segmenting the electroencephalogram signal according to the playing starting time of each vocabulary to obtain an electroencephalogram response segment corresponding to each vocabulary.

On the basis of the above embodiment, the present embodiment further includes a baseline correction module, configured to, for the electroencephalogram signal of each channel, subtract, from the electroencephalogram signal of a first preset duration after the start playing time of any vocabulary, an average value of the electroencephalogram signals of a second preset duration before the start playing time of any vocabulary, and obtain the electroencephalogram signal after baseline correction of each channel.

On the basis of the foregoing embodiment, the calculating module in this embodiment is specifically configured to: constructing a system response model which takes the occurrence probability as input and takes the electroencephalogram response segment corresponding to the occurrence probability as output; solving an impact response function of the system response model; and the impulse response function is used for expressing the electroencephalogram activity mode corresponding to the occurrence probability of the vocabulary.

On the basis of the foregoing embodiments, the measurement module in this embodiment is specifically configured to: extracting curve characteristics of the impact response function; and taking the curve characteristics as the input of the creativity personality trait prediction model to obtain the creativity personality trait test score of the testee.

On the basis of the above embodiments, the creative personality trait prediction model in this embodiment is a regression model; correspondingly, the system also comprises a training module, a regression model and a training module, wherein the training module is used for constructing the regression model by taking the questionnaire score of the creativity personality traits of the individual as a target; wherein the regression model is a LASSO regression model; adopting a traditional creativity personality trait measurement method to perform questionnaire scoring on creativity personality traits of training individuals; and training the regression model according to the questionnaire scores of the training individuals.

The embodiment provides an electronic device, and fig. 5 is a schematic view of an overall structure of the electronic device according to the embodiment of the present invention, where the electronic device includes: at least one processor 501, at least one memory 502, and a bus 503; wherein the content of the first and second substances,

the processor 501 and the memory 502 communicate with each other via a bus 503;

the memory 502 stores program instructions executable by the processor 501, and the processor calls the program instructions to perform the methods provided by the above method embodiments, for example, the methods include: carrying out vocabulary segmentation on a text of a natural language material, and acquiring the occurrence probability of each segmented vocabulary in the context of the text; acquiring an electroencephalogram signal of a testee when the testee hears the audio frequency of the natural language material, segmenting the electroencephalogram signal, and acquiring an electroencephalogram response segment corresponding to each vocabulary; acquiring an impulse response function of the occurrence probability according to the occurrence probability of each vocabulary and the electroencephalogram response segment corresponding to each vocabulary; and obtaining the creativity personality trait test score of the testee based on a pre-trained creativity personality trait prediction model according to the impact response function.

The present embodiments provide a non-transitory computer-readable storage medium storing computer instructions that cause a computer to perform the methods provided by the above method embodiments, for example, including: carrying out vocabulary segmentation on a text of a natural language material, and acquiring the occurrence probability of each segmented vocabulary in the context of the text; acquiring an electroencephalogram signal of a testee when the testee hears the audio frequency of the natural language material, segmenting the electroencephalogram signal, and acquiring an electroencephalogram response segment corresponding to each vocabulary; acquiring an impulse response function of the occurrence probability according to the occurrence probability of each vocabulary and the electroencephalogram response segment corresponding to each vocabulary; and obtaining the creativity personality trait test score of the testee based on a pre-trained creativity personality trait prediction model according to the impact response function.

Those of ordinary skill in the art will understand that: all or part of the steps for implementing the method embodiments may be implemented by hardware related to program instructions, and the program may be stored in a computer readable storage medium, and when executed, the program performs the steps including the method embodiments; and the aforementioned storage medium includes: various media that can store program codes, such as ROM, RAM, magnetic or optical disks.

The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.

Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. With this understanding in mind, the above-described technical solutions may be embodied in the form of a software product, which can be stored in a computer-readable storage medium such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in the embodiments or some parts of the embodiments.

Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.