KR20020015547A - recording medium for sensitivity treatment - Google Patents

Abstract

PURPOSE: A method for creating the data on the psychical cure through the EEG(electroencephalogram) analysis and the apparatus thereof are provided to make the optimal psychology by creating the audio-visual data of a signal equal to the EEG and actively changing the EEG. CONSTITUTION: The EEG of every emotion and behavior type is measured through an EEG measurement sensor(S1). The fine signal of the measured EEG is amplified through an amplifier(S2) and the amplified analog signal is converted into the digital signal through an A/D converter(S3). The EEG is analyzed by the basis of nonlinear dynamic theory(S4). The lyapunov dimension converging into lyapunov index is calculated by the EEG analysis value. The average of the lyapunov dimension between another subjects who are selected in the emotion and behavior type are calculated and the quantitative EEG signal equal to the time series EEG is output(S6). The emotional psychical cure data that is the same dimension to the analyzed EEG is created through a digital signal processing(S7). The audio-visual data for stimulating the EEG is created by mapping the frequency signal generated through the digital signal processing to the visual data(S8).

Description

Method for generating information for emotional treatment through electroencephalogram analysis, EEG analysis device and recording medium recording information for emotional treatment

The present invention relates to emotional treatment information generated based on EEG analysis using a nonlinear dynamic theory, and a recording medium having the information. Specifically, the human brain wave component is analyzed using a nonlinear dynamics system, Developing various stimulation media that can stimulate EEG and actively changing EEG to create an optimal psychological state. To a recorded recording medium.

More specifically, complex brain waves (electroencephalographs) are analyzed in detail and precisely, and the information is changed in the form of the same signal, thereby breaking away from the conventional form of brain wave stimulation by a simple signal sound, and hearing information and color with melody or rhythm. Method of generating information for emotional treatment through EEG analysis that provides optimal visual information by controlling human emotions and providing a supplementary role to intentional behaviors such as concentration and sleep promotion The present invention relates to an EEG analysis apparatus and a recording medium on which information for emotional treatment is recorded.

Recently, biofeedback and brain waves, which improve the user's mental state through the modulation of brain waves by external stimulation, have been widely used in the field of emotional engineering to evaluate emotions and apply them to product development. have.

Therefore, there is an urgent need for a technique capable of evaluating changes in human sensitivity to the surrounding environment. Accordingly, as an EEG analysis apparatus, a linear analysis method such as a Fourier transform and a nonlinear analysis method such as correlation dimension estimation are used.

In EEG analysis, the correlation dimension is known as the most commonly used method of chaotic mechanical analysis. In the Chaos system, the correlation index converges to a constant value, and the convergence value becomes a correlation dimension. Whether the time series is stochastic or chaotic can be determined.

In addition, Fourier transform can be very useful for obtaining information about specific frequency components such as Delta (δ), Theta (θ), Alpha (α), and Beta (β) waves. do.

The human brain always emits a minute current and measures this current in the form of a waveform in an EEG measuring device, which can be measured to emit in different waveforms according to the activity of the brain. By analyzing the measured EEG and classifying only the main component of the EEG component can be classified into four kinds according to the activity of the brain.

If we look at the kinds of the main components of EEG sequentially from the low frequency cycle, the first EEG is a delta wave, which is emitted when a person is in deep sleep, and has a frequency below 4 Hz. In the state of relaxation, the frequency of 4 ~ 8Hz is emitted, and the third brain wave alpha (α) wave is the highest level of consciousness, the body and the mind is in harmony, the brain wave of 8 ~ 13Hz is generated. The last brain wave, beta (β), is a state in which the conscious activity is carried out while the consciousness is completely awake and has a frequency of 13 to 30 Hz.

As an external stimulation device that stimulates the human brain by using the main components of EEG mentioned above, a product called MC Scare is commercially available, and these products extract only the main components from EEG components generated during simple human actions such as sleep and meditation. It is a simple delivery device that stimulates EEG with only its main component.

In addition, in this type of device, since the main component is a low frequency component that does not belong to the human hearing level, it is divided into multiple pieces of information to artificially provide different information to the right ear and the left ear to recognize the main component by the difference. Since the form of the information does not escape the form of a simple signal, it causes the antipathy to the information when the actual user listens.

In addition, the delta (δ) wave (frequency below 4 Hz), theta (θ) wave (4 to 8 Hz), alpha (α) wave (8 to 13 Hz), beta (β) wave ( The existing method of actively changing human brain waves by extracting only the main components of 13-30Hz) does not take into account individual differences between people. there was. In addition, since the characteristics of EEG signals vary according to the state of people, the characteristics of EEG signals are not considered. Therefore, it is difficult to obtain reliable results by EEG stimulation.

Therefore, there is a need for a new type of EEG stimulating medium that eliminates these conventional drawbacks.

Accordingly, the present invention has been made to solve the above-described problems, and by analyzing and estimating human brain EEG based on nonlinear dynamic theory (Chaos / Fractal theory), the fractal dimension, Lyapunov index (Lyapunov) index And a method for generating information for emotional therapy through electroencephalogram analysis to generate digital information for emotional therapy based on a quantified dimension in the same form as an actual time series EEG, and an EEG analysis device used therein. And it is an object to provide a recording medium on which information for emotional treatment is recorded.

Another object of the present invention is to analyze the complex EEG (Electroelectric Potential) in detail and change the information in the form of the same signal-mapping the EEG signal to the musical component, specific color and shape-the existing simple signal sound By escaping the form of EEG stimulation, it provides the auditory information with the melody or rhythm and the visual information with the color and the form to control human emotions and play a subsidiary role in intentional actions such as improving concentration and promoting sleep. It is an object of the present invention to provide a method for generating information for emotional treatment through EEG analysis, and a recording medium on which the EEG analysis device and the information for emotional treatment used therein are recorded.

Another object of the present invention is to analyze the brain waves using nonlinear dynamic theory and inversely use the estimated dimensions to create a signal closest to the brain waves, and then map these signals to frequencies corresponding to each note of music. By making music of the same signal as EEG and actively changing the EEG, the method of generating information for emotional therapy through EEG analysis that enables the optimal psychological state formation, and the EEG analysis device and the recording medium in which the information for emotional therapy is recorded For the purpose of providing it.

In addition, the object of the present invention is to calculate the average value of the Lyapunov dimensions between different subjects calculated by emotion / action type in consideration of individual differences between subjects, and the same source as the EEG signal calculated by emotion / action type. A method of generating information for emotional therapy through EEG analysis to obtain a desired brain state by EEG stimulation by generating a signal having a digital signal processing, and an EEG analysis device and a recording medium on which information for emotional therapy is recorded For the purpose of providing it.

1 is a schematic block diagram for explaining an EEG analysis apparatus using a nonlinear dynamic theory according to the present invention.

2 is a flowchart sequentially illustrating a method for generating information for brain wave stimulation information for emotional treatment by EEG analysis according to the present invention.

3 is a form of a chisel of human brain waves during sleep.

4 is a form of a chisel of human brain waves during meditation.

5 is a form of a chisel of human brain waves when excited.

6A-6B illustrate an example of EEG analysis using an extractor of nonlinear dynamic theory.

7 shows a complex brain wave form seen in time series.

FIG. 8 is a power spectrum of the EEG component appearing in time series as in FIG. 7 in frequency. FIG.

<Description of Symbols of Major Parts of Drawings>

11: electroencephalogram measurement sensor 12: amplifier

13: A / D converter 20: EEG analysis unit

21: Liafnov index calculation unit 22: correlation index calculation unit

23: frequency analysis unit 30: average value calculation unit

EEG analysis apparatus using a non-linear dynamic theory of the present invention for achieving this object, EEG measuring means for measuring the EEG for each type of emotion / behavior of the subjects; EEG analysis means for quantifying the complexity of the EEG in the time band by analyzing the measured EEG signal based on the nonlinear dynamic theory; EEG signal quantified in the same form as EEG signal in real time series by calculating the average value of quantified dimensions (Lyafnov dimension or correlation dimension) between different subjects calculated by emotion / action type considering individual differences among subjects Characterized in that the average value calculating means for outputting.

In addition, the method for generating information for emotional treatment through the EEG analysis in the present invention, the first step of detecting the brain waves of the test subjects by emotion / emotion type by a sensor to measure the EEG signal; A second step of analyzing the measured EEG signals to produce a Lyapunov dimension (or correlation dimension) that converges to the Lyapunov index (or correlation index); A third step of calculating average values of Lyapunov dimensions (or correlation dimensions) between different test subjects calculated for each emotion / action type in consideration of individual differences between test subjects; A fourth step of generating, by digital signal processing, a signal having the same origin as the EEG signal calculated for each emotion / action type by the average value of the Lyapunov dimension (or correlation dimension); Audio-visual information (data) for EEG stimulation for emotional treatment of the same signal as the EEG signal by emotion / action type by mapping the signal by emotion / action type generated by digital signal processing with the audio-visual information corresponding to the signal Characterized in that the fifth step of generating a.

In addition, the emotional treatment information recording medium carrying the EEG stimulation audiovisual information for the emotional treatment generated by the emotional treatment information generation method through the EEG analysis in the present invention is characterized in that it is implemented.

Hereinafter, the present invention will be described in detail with reference to the accompanying drawings.

1 is a schematic block diagram illustrating an EEG analysis apparatus using a nonlinear dynamic theory according to the present invention.

Looking at the configuration of the EEG analysis apparatus with reference to Figure 1, EEG measuring unit 10 for measuring the EEG by emotion / behavior type of the test subjects, and the measured EEG signals based on non-linear dynamic theory to analyze the EEG in the time band Calculate the average value of the quantified Lyapunov dimension or correlation dimension between different subjects calculated for each emotion / action type considering the individual difference between the subjects and the brain wave analysis unit 20 for quantifying the complexity of The average value calculator 30 outputs an EEG signal quantified in the same form as the EEG on the actual time series.

First of all, the characteristics of the nonlinear dynamics system have the complexity of orbital instability and long-term unpredictability due to the sensitivity of initial conditions, but quantification is possible due to the nature of magnetic similarity. The characteristics of these nonlinear dynamic theories can be used to quantify the complexity of EEG in the time band.

In the case of human brain waves, the brain waves are measured in the time zone when working or under stress than in sleep and meditation. The latter case is more complicated than the former case. Due to the difference is a little difference. In other words, when the brain wave is measured in the time band according to the human emotion state or behavior type, each case has a different type of brain wave. These brain waves can be analyzed and quantified using nonlinear dynamic theory to define human emotion and behavior.

More specifically, the vibration of EEG with various components is due to a method of transmitting information through electrical interactions between closely connected and complex neurons.

Therefore, by observing the frequency characteristics of EEG using the power spectrum, a rule for determining temporal change can be found. That is, the brain potential, which is the information of EEG, has a certain shape of geometric trajectories for various measurement behaviors such as human emotion and sleep, and this trajectory is called an attractor.

The shape of the chisel mentioned above has the characteristics of each of the neurons, and when the chisel is made in the same form, the signals of the EEG signal and the origin are generated. In other words, it is possible to generate the characteristics of a set of neurons in a specific way.

In order to estimate signals of the same origin, various types of brain potentials must first be measured. The brain waves that occur when feelings such as when you are happy, when you are sad, when you are angry, when you sleep, or when you meditate are classified in detail, and each brain wave corresponding to them is measured and recorded.

Then, the EEG measured in time is analyzed based on the nonlinear dynamic theory, and the approximate values for estimating the characteristics of each EEG in the form of correlation dimension / Lyapunov dimension / Fractal dimension are estimated. Generate the signal with the same method as the digital signal processing. The generated signals are mapped to specific colors and shapes to produce visual information, and in the same way, they are mapped to musical components to create auditory information in the form of music instead of simple beeps.

Accordingly, the present invention analyzes and estimates a complex brain wave of a human based on a nonlinear dynamic theory (chaos / fractal theory), and reconstructs the brain wave in actual time series by reconstructing the fractal dimension, the Lyapunov index, and the dimension. It is possible to derive the same type of information as.

First embodiment

As shown in FIG. 1, each clinical experiment is performed separately for each emotion / action type, and in order to minimize the difference between individual individuals in consideration of characteristics in which EEG signals are different according to a human state, EEG information of the subject "is measured.

The EEG time series signals of various test subjects measured by the EEG measurement sensor 11 are amplified by the amplifier 12. At this time, the measured voltage of the EEG measured on the time band is very minute, and then amplifies the EEG signal, and then converts the amplified signal into a digital form through the A / D converter 13 and analyzes it on a computer. In this case, for the EEG analysis, the Lyapunov exponent and the dimension calculator 21 calculate the Lyapunov exponents of the time series of the measured EEG signals, calculate the Lyapunov dimension converged to the Lyapunov exponent, and calculate the average value. The unit 30 calculates an average value of Lyapunov dimensions between different test subjects calculated for each emotion / action type in consideration of individual differences among test subjects, and outputs an EEG signal quantified in the same form as an actual time series EEG. .

More specifically, the calculation of the Lyapunov exponent and dimension by EEG analysis shows the time series data obtained by measuring the value of X at each time interval t (X (t-1), X (t), X (t +). 1), ..., etc. If there is any deterministic law in the whole EEG, the X value at any time will depend on the past X values, and how many X values will depend on the system's variables. It is closely related to the number of s, the dimensional value of the space containing the odd attractor.

If the micro displacement in X (t) is δX (t), it is expressed as the following equation.

By expanding the Taylor series, the linear approximation gives an idea of X (t) as shown in Equation (2).

Here, DF [X (t)] is a Jacobian matrix at point X (t).

As δX (0), U1 (t), U2 (t), ..., Un (t) are given as a combination of unit vectors orthogonal to each other in an n-dimensional junction space.

U1 (0) = (1, 0, ..., 0)

U2 (0) = (0, 1, ..., 0)

.

.

.

Un (0) = (0, 0, ..., 1)

If you look at the change of each vector,

ei (t + 1) = DF [X (t)] ui (t) gives ei (t + 1) and ui (t + 1) in the same way. Using the values thus obtained, the Liafnov spectra lambda i (i = 1, 2, ..., n) are obtained as follows.

Here, lambda i is called the Lyapunov exponent, and if the maximum Lyapunov exponent is positive, it has an orbital instability characteristic of chaos.

In the Liafnov Spectrum, we can also find the Liafnov dimension, an indicator of the fractal dimension of the chisel. When adding Lyapunov exponents in big order, add the largest nonnegative integer to j.

At this time, the Lyapunov dimension DLyapunov is given as follows.

In addition, the upper limit of the KS entropy indicating the unpredictability of the nonlinear dynamic theory can be evaluated as the sum of positive Lyapunov exponents λ1, λ2, ..., λk. That is, it is expressed as Equation 5.

Analysis of the EEG as described above, it can be seen that it has the following characteristics.

Referring to FIGS. 3 to 5, FIG. 3 is a form of a chisel in sleep, FIG. 4 is a form of a chisel in meditation, and FIG. 5 is a form of a chile in human brain waves when excited.

As shown in FIGS. 3 to 5, the patterns of the draggage are shown differently according to human emotions / action types, and instability increases from FIG. 3 to FIG. 5, and the Lyapunov dimension also has a large amount. That is, in FIG. 3, the predictability is easy, but in FIG. 5, the predictability is difficult.

Therefore, when music is created based on the theory proved above, lullabies are composed of simple patterns, such as brain waves during sleep, to create music that changes continuously and continuously. In this case, when the music is composed, the change of the sound becomes large and the height of the sound is also severely changed as shown in FIG.

Second embodiment

Here, the description of the same components as in the first embodiment will be omitted.

Referring to FIG. 6A to FIG. 6B, in the case of analyzing by using an attractor as another method of EEG analysis in nonlinear dynamic theory, the space composed of the time delay axis from the variable reflecting the information of the signal is temporally. Corresponds to the trajectory of a particular geometry formed in. For example, having a trajectory of a specific geometry as shown in FIG. 6b means that a signal having the same trajectory has the same source.

Therefore, the correlation index and the dimension calculation unit 22 for calculating and outputting a correlation index of the time series with respect to the measured EEG signal are configured to calculate the correlation index of the EEG signal. At this time, the value of the correlation index converges to a constant value, which is the correlation dimension.

Thereafter, the average value of correlation dimensions between different test subjects calculated for each emotion / action type is calculated in consideration of individual differences among test subjects, and the EEG signals quantified in the same form as the actual time series EEG signals are output.

Third embodiment

The complex brain waves are shown in FIG. 7 in time series. In the case of analyzing the EEG using 1 / f noise among the methods of analysis using the nonlinear dynamic theory, the power spectrum is shown in FIG. 8 when the EEG components appearing in time series are shown in frequency as shown in FIG. 7.

In this case, the dotted line of FIG. 8 may be represented in the form of 1 / f ^a , where a corresponds to an absolute value of the slope, and as the value of a increases, it means that the EEG has various types of information. According to the emotional and behavioral differences.

Accordingly, when applied to the present invention, the frequency is analyzed based on a predetermined time with respect to the magnitude of the EEG signal measured by the test subject through the frequency analyzer 23 for the EEG analysis, and the analyzed frequency value is compared with a predetermined reference. By estimating the EEG signals for each emotion / action type by comparing them with the values, the EEG signals are quantified in the same form as the EEG signals in real time series.

As described above, the nonlinear dynamic theory is used to analyze the EEG and inversely use the estimated dimension to make the signal closest to the EEG.

By mapping these signals to frequencies corresponding to each note of music, music of the same signal as the brain waves can be produced. In other words, the music has four octaves of 'Ra' sound at 440 hz. However, because the piano and the violin are different between the 'la' sound, the sound of each instrument based on musical theory and applied to the dimension obtained above makes it possible to make music with the same dimensions as the brain waves.

Human brain waves have different types of brain waves for each emotion (hee, noh, babies, rock, etc.) and vary according to the type of behavior (sleep, meditation, work performance, etc.). Therefore, when measuring and analyzing the EEG of each type to obtain the dimensions, all become different dimensions. When information is created with these dimensions, it is possible to generate auditory information with various melodies or rhythms and visual information with colors and shapes.

FIG. 2 is a flowchart sequentially illustrating a method of generating information for brain wave stimulation for emotional treatment by brain wave analysis using nonlinear dynamic theory according to the first and second embodiments.

First, the human brain wave is measured and measured by emotion (hee, noh, child, rock, etc.) / Action type (sleep, meditation, work performance, etc.) by the electroencephalography sensor 11 attached to the human brain. The amplified microsignal of a human brain wave is amplified by the amplifier 12, and the amplified analog signal is converted into a digital signal by the A / D converter 13 (S1 to S3).

Then, considering that the time series variables have a certain geometric trajectory, correlations / liafnov dimensions / It is possible to estimate the path that the trajectory in the state space such as the fractal dimension moves over time, so that the overall trajectory of emotions such as joy, sadness, anger, sleep, meditation, etc. It calculates (S4-S5). By this, the approximate value of each EEG can be estimated by emotion / action type-emotion, joy, sadness, anger, sleep, meditation, etc. By considering the average value of the approximate values between different test subjects calculated for each emotion / action type, the EEG signal quantified in the same form as the actual EEG on the time series is output (S6). In other words, the measured EEG signals of various subjects are analyzed to calculate the Lyapunov dimension (or correlation dimension) that converges to the Lyapunov index (or correlation index), and by emotion / action type by considering individual differences among the subjects. By calculating the average value of the Lyapunov dimension (or correlation dimension) between the different test subjects to calculate the EEG signal quantified in the same form as the actual EEG on the time series.

Based on this, a signal having the same origin as the signal of the EEG estimated by emotion / action type is generated by a method such as digital signal processing (S7).

The digital signal processing method of the present invention means numerical processing using a digital computer or a digital integrated circuit.

Specifically, it refers to a process of filtering, amplifying, removing noise, generating a signal, detecting a signal, and detecting a characteristic of a signal.

The generated signal is mapped to specific colors and shapes to produce visual information, and in the same manner, the signal is mapped to musical components to produce auditory information in the form of music instead of simple beeps (S8).

The process of generating a signal having the same origin as that of an EEG signal estimated by emotion / action type by digital signal processing and mapping the digital signal to audiovisual information will be described in more detail with reference to Table 1 below. .

The EEG signal estimated by the Lyapunov dimension can be expressed in a digital module by digital signal processing. Mapping these signals in binary form to each color or scale, referring to Table 1, maps the EEG stimulation signals with the same origin as the EEG signals calculated for each emotion / action type, along with auditory information with melody or rhythm. It can be provided in the form of visual information with color and shape.

For example, referring to Table 1, a binary signal having the same origin as an EEG signal generated by digital signal processing, '0 0 0' means 'degrees', '0 0 1' means 'le' sounds, '0 1 0' generates auditory information for brain wave stimulation for emotional treatment by mapping with musical scales such as 'mi' sound. Also, '0 0 0' is mapped to colors such as 'white', '0 0 1' is 'red', and '0 1 0' is 'yellow' to generate visual information for brain wave stimulation for emotional treatment. .

Therefore, a signal having the same source as the EEG signal estimated by emotion / action type according to the human condition is generated by a method such as digital signal processing, and audiovisual information for brain wave stimulation for the emotional treatment is provided to the actual user. First, the emotional treatment auditory information may be produced and supplied in the form of tape / CD, and the visual and auditory information may be disseminated in the form of video.

That is, the emotional treatment information recording medium carrying the EEG stimulation audiovisual information for the emotional treatment generated using the emotional treatment information generation method through the EEG analysis may be implemented.

For example, it is available to netizens at any time by registering each information on a web site through the Internet, and is also available in the form of a portable storage device that is programmed and stored in a memory chip for a specific treatment. It has a possible convenience (S9).

Those skilled in the art will appreciate that various changes and modifications can be made without departing from the technical spirit of the present invention. Therefore, the technical scope of the present invention is not limited to the contents described in the accompanying drawings and the detailed description of the specification.

As described above, according to the present invention, the complex EEG (Electron Dislocation) is analyzed in detail and precisely, and the information is changed in the form of the same signal, thereby breaking away from the conventional EEG stimulation form by the simple signal sound and having a melody or rhythm. By providing auditory information and visual information with color and form, it is possible to create an optimal psychological state by controlling human emotions and playing an auxiliary role in intentional actions such as improving concentration and promoting sleep.

In addition, the average value of the Lyapunov dimensions between different test subjects calculated by emotion / action type in consideration of individual differences among test subjects is calculated, and the signal having the same source as the EEG signal calculated by emotion / action type is digitally obtained. By generating by signal processing, a desired brain state can be obtained by brain wave stimulation.

Claims (7)

A first step of detecting an EEG of the test subjects by emotion / action type by a sensor and measuring an EEG signal; A second step of analyzing the measured EEG signals to calculate a Lyapunov dimension that converges to Lyapunov index; Calculating a mean value of Lyapunov dimensions between different test subjects calculated for each emotion / action type in consideration of individual differences between test subjects; A fourth step of generating, by digital signal processing, a signal having the same origin as an EEG signal calculated for each emotion / action type by the average value of the Lyapunov dimension; Audiovisual information for electroencephalogram stimulation for emotional treatment of the same signal as the EEG signal for each emotion / action type by mapping a signal for each emotion / action type generated by the digital signal processing with audiovisual information corresponding to the signal The information generating method for the emotional treatment through the EEG, characterized in that consisting of a fifth step of generating. A first step of detecting an EEG of the test subjects by emotion / action type by a sensor and measuring an EEG signal; A second step of analyzing the measured EEG signals to calculate a correlation dimension that converges to a correlation index; Calculating a mean value of correlation dimensions between different test subjects calculated for each emotion / action type in consideration of individual differences between test subjects; A fourth step of generating, by digital signal processing, a signal having the same origin as the EEG signal calculated for each emotion / action type based on the average value of the correlation dimension; Audiovisual information for electroencephalogram stimulation for emotional treatment of the same signal as the EEG signal for each emotion / action type by mapping a signal for each emotion / action type generated by the digital signal processing with the audiovisual information corresponding to the signal The information generating method for the emotional treatment through the EEG, characterized in that consisting of a fifth step of generating. The method according to claim 1 or 2, wherein the first step comprises: an EEG signal sensing step of detecting an EEG of test subjects by emotion / action type by a sensor; A signal amplifying step of amplifying the sensed signal through an amplifying means; And converting the amplified analog signal into a digital signal through an A / D converting means. EEG measuring means for measuring the EEG by emotion / type of behavior of the test subjects; EEG analysis means for quantifying the complexity of the EEG in the time band by analyzing the measured EEG signal based on the nonlinear dynamic theory; It is composed of an average value calculating means for calculating the average value of the quantified dimensions between different test subjects calculated by emotion / action type in consideration of individual differences between subjects and calculating the quantified EEG signal in the same form as the actual time series. EEG analysis device using a nonlinear dynamic theory. The method according to claim 4, wherein the EEG analysis means calculates the Lyapunov exponent of the time series for the measured EEG signal to calculate a Lyapunov dimension converging to the Lyapunov exponent. EEG analysis device using a nonlinear dynamic theory, characterized in that. 5. The nonlinear dynamic theory of claim 4, wherein the EEG analyzing means is a correlation index and a dimension calculating means for calculating a correlation dimension converged to the correlation index by calculating a correlation index of a time series of the measured EEG signal. EEG analysis device using the. An information recording medium on which audiovisual information for brain wave stimulation for emotional treatment generated using the method according to claim 1 or 2 is recorded.