Subthreshold unconscious electroencephalogram signal acquisition system and method
By using a subthreshold unconscious EEG signal acquisition system based on the continuous flicker suppression paradigm, combined with improved masking material generation and binocular split vision technology, the problem of unconscious EEG signal acquisition has been solved, and high-precision monitoring and analysis of unconscious brain activity has been achieved.
Patent Information
- Application Number
- CN202411148599.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-21
- Publication Date
- 2026-03-03
AI Technical Summary
Existing EEG signal acquisition technologies are mostly designed for the conscious level, making it difficult to effectively acquire and analyze EEG signals in the unconscious state. They are also susceptible to interference, leading to significant errors.
A subthreshold unconscious EEG signal acquisition system based on the continuous flicker suppression paradigm was adopted, which combined an automatic masking material generation module, a stimulus induction module, an EEG signal acquisition and analysis module, and an optical stereoscopic hardware platform. The improved leaf-falling algorithm was used to generate masking images, and an unconscious state was induced by binocular split vision technology. The EEG signal acquisition cap and amplifier were used for signal acquisition and analysis.
It enables precise induction and real-time monitoring of subthreshold unconscious states, improves the signal-to-noise ratio and acquisition accuracy, ensures the repeatability and accuracy of experiments, and allows for in-depth analysis of unconscious brain activity.
Smart Images
Figure CN121587740A_ABST
Abstract
Description
(I) Technical Field:
[0001] This invention belongs to the field of electroencephalogram (EEG) signal processing technology, specifically relating to a subthreshold unconscious EEG signal acquisition system and method. (II) Background Technology:
[0002] Generally speaking, consciousness that an individual cannot directly perceive is collectively referred to as the unconscious, encompassing potential thoughts, feelings, motivations, and memories. Subliminal unconsciousness specifically refers to stimuli below the perception threshold—that is, situations where stimuli exist but the individual is unaware of them. These may influence an individual's behavior and decision-making, but do not enter the conscious level. Consciousness that an individual can perceive and control belongs to the conscious level. In recent years, with the rapid development of artificial intelligence, brain-computer interfaces, and high-performance hardware computing devices, research on EEG signals at the conscious level has achieved considerable progress. However, in contrast, research on methods for evoking and acquiring unconscious EEG signals remains relatively slow. The Continuous Flash Suppression paradigm (CFS) is a powerful interocular suppression technique with strong stimulus masking advantages, and is currently one of the important research paradigms in unconscious visual processing research.
[0003] In 1983, Marcel et al. discovered, using visual masking techniques, that unconscious semantic information stimuli significantly influenced individual behavior. By altering the semantic relationship between the target word and the masking word, they found that experimental individuals showed a stronger preference for semantically related target stimuli. Subsequently, Greenwald et al.'s research showed that under masking techniques, subjects responded faster when faced with the same emotional valence initiator and target words. Dehaene et al. were the first to demonstrate, using EEG signal acquisition technology, that unconscious stimuli could induce and activate the motor cortex of the brain. These studies have all explored unconscious brain activity. Existing EEG signal acquisition protocols primarily focus on acquiring cortical potentials for conscious tasks. While existing research indicates that information such as emotion and semantic understanding can be processed unconsciously, methods and tools for analyzing the neural mechanisms of the unconscious brain are still lacking. Furthermore, EEG signals acquired consciously are more susceptible to interference from irrelevant tasks, leading to some errors. (III) Summary of the Invention:
[0004] The purpose of this invention is to propose a subthreshold unconscious EEG signal acquisition system and method based on the continuous flicker suppression paradigm, so as to achieve the effect of acquiring subthreshold unconscious EEG signals from subjects.
[0005] To effectively solve the above problems, the present invention adopts the following technical solution:
[0006] A subthreshold unconscious EEG signal acquisition system includes an automatic masking material generation module, a stimulus induction module, an EEG signal acquisition and analysis module, and an optical stereoscopic hardware platform. The automatic masking material generation module presets different masking stimulus templates, which can generate colored or grayscale images and square or circular dynamic Mondrian images. The stimulus induction module is implemented using software algorithms based on Matlab and Psychophysics Toolbox, and a synchronous triggering mechanism is designed based on marker point technology to ensure precise synchronization between stimulus presentation and EEG signal acquisition. The EEG signal acquisition and analysis module uses a 64-channel EEG acquisition cap, two EEG signal amplifiers, and a headbox acquisition card structure. The EEG signal sampling frequency is 1000Hz, and a bandpass analog filter is used for filtering from 0.5Hz to 70Hz during acquisition. The optical stereoscopic hardware platform is constructed using standardized high-precision optical components, with basic components including a masking plate, guide rails, a lead reflector, and corresponding frames and support fixtures. The four modules or platforms mentioned above are independent in structure and function, but they may need to work together in some applications, such as collecting and analyzing EEG signals while using masking materials to study the effect of masking on cognitive processes.
[0007] A method for acquiring subthreshold unconscious EEG signals, characterized by comprising the following steps:
[0008] (1) Run the stimulus-evoked module algorithm on the optical stereoscopic hardware platform to obtain the initial resolution of the screen used in the current experiment. The layout of the screen image is defined as follows:
[0009] Ox×Oy≥(Ix×Iy)×D 2 +(Sx×Sy)×D 2 +l 2 ×D 2 ×2 (1)
[0010] Where Ox and Oy represent the width and height of the original screen resolution, respectively, while the angular resolution is defined as D (the product of the width or height and the angular resolution is the actual pixel length). When rendering the stimulus image, its width is defined as Ix and its length as Iy; the mask image is defined as having a width of Sx and a length of Sy. Furthermore, both the stimulus and mask frames are squares with a side length defined as l. Then, this invention fills the entire screen based on the experimental window size. A two-dimensional coordinate system is established, with its origin at the top-left corner of the experimental window, using pixels as the coordinate unit, extending to the right as the positive X-axis, and extending downwards as the positive Y-axis;
[0011] The optical stereoscopic hardware platform in step (1) is constructed using standardized precision optical components. The subject's two eyes are presented with stimulus material and masking material flashing at a certain frequency, depending on the dominant eye. This process is achieved by reflecting the material to the two eyes through two sets of mirrors, providing different visual fields.
[0012] (2) The position of the stimulus box and the mask box is determined by an adaptive screen parameter setting algorithm. Two square mask boxes and stimulus boxes of equal size are generated on both sides of the screen in a horizontally symmetrical manner, thereby completing the establishment of the visual experiment engine. The optical mirror is adjusted by a precision adjustment device so that the subject can achieve binocular fusion before the experiment.
[0013] In step (2), the visual experiment engine is implemented based on software algorithms of Matlab and Psychophysics Toolbox. After binocular fusion, it can achieve the effect that there is only one square frame in the eyes of the subject.
[0014] (3) The improved leaf algorithm is used to automatically generate masking material and display it in the masking box in step (2). The improved method is to randomly sample the neighborhood around each pixel in the stimulus image, thereby evaluating whether the pixel belongs to the foreground or the background based on the statistical characteristics of the neighborhood pixel values. Then, the Mondrian masking image with color blocks at different positions is generated by the position automatic generation module and the shape color block automatic generation module.
[0015] The improved leaf model in step (3) is an image noise generation model. Its characteristics include approximate scale invariance and some statistical model attributes of natural and synthetic images. The improved masking random color function adds three color output channels with assigned random numbers, which can generate color Mondrian images and further improve the masking types and effects.
[0016] (4) Using multi-angle and multi-color fusion stimulation materials displayed in the stimulation box in step (2) and inducing the subject's EEG signals, we can explore the effect of small changes in stimulation materials during unconscious visual perception on the breakthrough of consciousness and the functional connectivity of brain networks.
[0017] In an exemplary embodiment of this application, step (4) involves using a Chinese-style dynamic emotion face database to transform the material in three directions: front, side, and back, and to render it using different warm and cool tones.
[0018] (5) The EEG signal of the subject during the CFS experiment was collected in real time using the EEG extraction device. After signal amplification and noise reduction, the signal was transmitted to the computer for EEG signal preprocessing. The differences in the dominant brain regions were recorded and further analyzed when visual information gradually entered consciousness. At the same time, the analysis module can use machine learning algorithms such as random forest to visualize and analyze the functional connectivity of the brain.
[0019] The signals collected in step (5) are decomposed into Theta, Alpha, Beta and Gamma bands and low-frequency and high-frequency oscillation power. At the same time, the power spectral density, differential entropy and functional brain network characteristics of the subject's EEG signal are calculated, and frequency domain relative energy spectrum power analysis and time domain variance analysis are performed to fully understand the brain functional connectivity and brain network activity of the subject during the experiment.
[0020] The advantages of this invention are as follows: This invention combines precision optical stereoscopic imaging with EEG signal acquisition technology to achieve precise induction and real-time monitoring of subthreshold unconscious states, ensuring the accuracy of subthreshold stimulation and the repeatability of the experiment. The binocular vision technique in this invention allows for precise manipulation of the subject's visual information, effectively inducing unconscious states. This invention utilizes an improved leaf-falling algorithm to automatically generate Mondrian images with a high degree of masking effect. An adaptive image parameter setting algorithm ensures precise adjustment of the experimental interface. Furthermore, this invention utilizes high-precision amplifiers and acquisition cards to improve the accuracy and signal-to-noise ratio of EEG signal acquisition. This platform can perform in-depth analysis of EEG signals through machine learning algorithms such as random forests, providing strong technical support for understanding the brain's information flow and functional connectivity during the formation of consciousness. (iv) Description of the attached drawings:
[0021] Figure 1 This is a schematic diagram of the optical stereo mirror hardware platform of the present invention;
[0022] Figure 2 This is a diagram showing the positional relationship between the two masking color blocks in this invention;
[0023] Figure 3 This is a diagram illustrating the effect of the algorithm for automatically generating Mondrian masking images proposed in this invention.
[0024] Figure 4 This is a schematic diagram of the generation and real-time acquisition method of EEG signals in this invention;
[0025] Figure 5 This is a flowchart of the EEG signal acquisition and analysis module algorithm proposed in this invention;
[0026] Figure 6 This is a schematic diagram of the comprehensive principle of the unconscious EEG signal acquisition method based on the continuous flicker suppression paradigm proposed in this invention. (V) Specific Implementation Methods:
[0027] To explain in detail the technical content, structural features, objectives, and effects of the technical solution, the following description is provided in conjunction with specific embodiments and accompanying drawings.
[0028] This invention proposes a method for acquiring unconscious EEG signals based on a continuous flicker suppression paradigm, which includes:
[0029] (1) Run the stimulus-evoked module algorithm on the optical stereoscopic hardware platform to obtain the initial resolution of the screen used in the current experiment. The layout of the screen image is defined as follows:
[0030] Ox×Oy≥(Ix×Iy)×D 2 +(Sx×Sy)×D 2 +l 2 ×D 2 ×2 (1)
[0031] Where Ox and Oy represent the width and height of the original screen resolution, respectively, while the angular resolution is defined as D (the product of the width or height and the angular resolution is the actual pixel length). When rendering the stimulus image, its width is defined as Ix and its length as Iy; the mask image is defined as having a width of Sx and a length of Sy. Furthermore, both the stimulus and mask frames are squares with a side length defined as l. Then, this invention fills the entire screen based on the experimental window size. A two-dimensional coordinate system is established, with its origin at the top-left corner of the experimental window, using pixels as the coordinate unit, extending to the right as the positive X-axis, and extending downwards as the positive Y-axis;
[0032] The optical stereoscopic hardware platform in step (1) is constructed using standardized precision optical components. The subject's two eyes are presented with stimulus material and masking material flashing at a certain frequency, depending on the dominant eye. This process is achieved by reflecting the material to the two eyes through two sets of mirrors, providing different visual fields.
[0033] (2) The position of the stimulus box and the mask box is determined by an adaptive screen parameter setting algorithm. Two square mask boxes and stimulus boxes of equal size are generated on both sides of the screen in a horizontally symmetrical manner, thereby completing the establishment of the visual experiment engine. The optical mirror is adjusted by a precision adjustment device so that the subject can achieve binocular fusion before the experiment.
[0034] In step (2), the visual experiment engine is implemented based on software algorithms of Matlab and Psychophysics Toolbox. After binocular fusion, it can achieve the effect that there is only one square frame in the eyes of the subject.
[0035] (3) The improved leaf algorithm is used to automatically generate masking material and display it in the masking box in step (2). The improved method is to randomly sample the neighborhood around each pixel in the stimulus image, thereby evaluating whether the pixel belongs to the foreground or the background based on the statistical characteristics of the neighborhood pixel values. Then, the Mondrian masking image with color blocks at different positions is generated by the position automatic generation module and the shape color block automatic generation module.
[0036] The improved leaf model in step (3) is an image noise generation model. Its characteristics include approximate scale invariance and some statistical model attributes of natural and synthetic images. The improved masking random color function adds three color output channels with assigned random numbers, which can generate color Mondrian images and further improve the masking types and effects.
[0037] (4) Using multi-angle and multi-color fusion stimulation materials displayed in the stimulation box in step (2) and inducing the subject's EEG signals, we can explore the effect of small changes in stimulation materials during unconscious visual perception on the breakthrough of consciousness and the functional connectivity of brain networks.
[0038] In an exemplary embodiment of this application, step (4) involves using a Chinese-style dynamic emotion face database to transform the material in three directions: front, side, and back, and to render it using different warm and cool tones.
[0039] (5) The EEG signal of the subject during the CFS experiment was collected in real time using the EEG extraction device. After signal amplification and noise reduction, the signal was transmitted to the computer for EEG signal preprocessing. The differences in the dominant brain regions were recorded and further analyzed when visual information gradually entered consciousness. At the same time, the analysis module can use machine learning algorithms such as random forest to visualize and analyze the functional connectivity of the brain.
[0040] The signals collected in step (5) are decomposed into Theta, Alpha, Beta and Gamma bands and low-frequency and high-frequency oscillation power. At the same time, the power spectral density, differential entropy and functional brain network characteristics of the subject's EEG signal are calculated, and frequency domain relative energy spectrum power analysis and time domain variance analysis are performed to fully understand the brain functional connectivity and brain network activity of the subject during the experiment.
[0041] See Figure 1The diagram shows the principle structure of the optical stereoscopic hardware platform of this invention. The screen symmetrically displays a "masking frame" and a "stimulation frame" on its left and right sides, respectively. The central obstruction near the subject ensures that only the image on the screen is visible through the reflective mirror, thus achieving the effect of separate left and right vision. First, by adjusting four lead reflective intensifiers, the two frames seen by the subject's eyes before the experiment overlap, achieving "binocular fusion." Because the masking material possesses higher stimulus intensity, flicker characteristics, and is dynamically changing, the subject's visual attention is drawn to it. After the visual information received by both eyes is fused at the far end, the relatively static and low-intensity stimulus material remains invisible for a certain period.
[0042] This invention utilizes standardized, high-precision optical components, combined with supporting fixing and auxiliary devices, to construct an innovative stereoscopic vision hardware platform. It includes basic components such as a masking plate, guide rails, lead intensifying mirrors, and corresponding mirror frames and support poles. Specifically, the lead intensifying mirrors are fixed to the support poles, which are connected to the guide rails. The relative positions of the mirrors are adjusted by precision sliders on the guide rails. At the connection points between the support poles and the mirror frames, this invention employs high-precision adjustable tensioners, allowing free adjustment of the mirror angles and heights to achieve a complete overlap in size and position between the masking frame and the stimulus frame presented in the subject's eyes. Furthermore, to ensure subject comfort during the experiment and improve the accuracy of the results, the hardware platform is specifically equipped with a head support system to fix the subject's forehead and chin, providing convenient conditions for subthreshold unconscious EEG acquisition during observation and recording.
[0043] The visual experiment engine of this invention adopts an adaptive image parameter setting algorithm. First, the initial resolution of the screen is obtained. In terms of area, the layout of the screen image can be defined by the following relationship:
[0044] Ox×Oy≥(Ix×Iy)×D 2 +(Sx×Sy)×D 2 +l 2 ×D 2 ×2 (1)
[0045] Where Ox and Oy represent the width and height of the original screen resolution, respectively, while the angular resolution is defined as D (the product of the width or height and the angular resolution is the actual pixel length). When rendering stimulus images on the screen, this invention defines its width as Ix and its length as Iy; the masking image is defined with a width of Sx and a length of Sy. Furthermore, both the stimulus box and the masking box are squares with a side length defined as l.
[0046] Then, this invention fills the entire screen using the experimental window size as a reference. A two-dimensional coordinate system is established, with its origin at the top-left corner of the experimental window, using pixels as the coordinate unit, extending to the right as the positive X-axis, and downwards as the positive Y-axis. The relative positional relationships of the stimulus boxes in the left half of the experimental window can then be expressed as follows:
[0047]
[0048] The absolute distance relationship between the right half of the experimental screen window and the masking frame can be described as follows:
[0049]
[0050] Will and As the initial position coordinates of the masking box and the stimulus box, the screen layout and parameters can be adaptively adjusted.
[0051] This invention breaks away from the inefficient model where CFS researchers had to manually create Mondrian masking images. It develops and designs an automatic masking image generation algorithm based on a falling leaf model and integrates it into the CFS system platform. The core idea of this model is that each pixel in an image can be depicted as a drifting leaf, randomly moving in different directions. By randomly sampling the neighborhood around each pixel, this process is approximated by observing the shape, color, and other features of the surrounding leaves in a pile of drifting leaves, thus evaluating whether the pixel belongs to the foreground or background based on the statistical characteristics of the neighboring pixel values.
[0052] The automatic image generation algorithm based on the fallen leaf model in this invention is divided into an automatic position generation module and an automatic shape and color block generation module. Its design principle is as follows: Figure 2 As shown in the diagram, based on stability, given two disks, the probability that a given point e belongs to the same object at a distance x can be calculated as follows:
[0053]
[0054] In the above formula, P1(x) is the probability that the object contains one of the two points at this moment, and P2(x) is the probability that the object is not occluded at this moment. From this, we can derive the conditional probability model:
[0055] g(x,r)=Pr{x2∈A|x1∈A;||x1-x2||=x} (5)
[0056] Let the radius of disk A be r, which can be expressed as the dimensionless quantity ξ = x / r as follows:
[0057]
[0058] Based on the radius r of the disk min ≤r≤r max And the distribution of the radius is based on 1 / r 3 We can obtain:
[0059]
[0060] Therefore, the probability that a given point e belongs to an object with radius within the interval [r, r+dr] can be obtained as follows:
[0061]
[0062] Substituting equation (8) into equation (4) yields...
[0063]
[0064]
[0065] By performing a third-order polynomial best fit on equation (8), we can obtain:
[0066]
[0067] In equation (11), a0 = 1.0, a1 = -0.61, a2 = -0.051, and a3 = 0.052. For 2r... min ≤x<2r max Substituting equation (7) into equation (6), we can obtain the numerical estimate of B(x) as follows:
[0068]
[0069] This can generate, for example Figure 3 The Mondrian masking images shown are of different color blocks. Furthermore, this invention utilizes an improved masking random color function, adding three color output channels with assigned random numbers, thereby generating color Mondrian images and further improving the types and effects of masking.
[0070] This invention utilizes multi-angle, multi-directional, and multi-tone fusion stimulus materials in the EEG signal evoked paradigm to explore the consciousness breakthrough and brain network functional connectivity during unconscious visual perception. Table 1 shows the masking effect of this invention on visual emotion perception in an unconscious state, verified using a platform. Twenty-eight subjects were selected to conduct binocular split vision (CFS) and binocular synopsis (Non-CFS) experiments. The stimulus materials were derived from a Chinese dynamic emotional face database, divided into three different emotional valences: negative, neutral, and positive. Consciousness breakthrough judgment was used as the experimental task. When subjects visually perceived a face, they immediately pressed a button to report. The data in the table represent the subjects' consciousness breakthrough inhibition time (reaction time).
[0071] Table 1. Reaction times (seconds) under CFS and Non-CFS conditions (M±SD)
[0072]
[0073] Experimental results show that the EEG signal evoked platform based on the continuous flicker suppression paradigm proposed in this patent exhibits reliable and stable characteristics in both the hardware and software systems. The masking image automatic generation algorithm based on the leaf-falling algorithm demonstrates excellent masking performance and has achieved significant results in the verification experiment.
[0074] This invention employs electroencephalography (EEG) technology to explore the neural mechanisms underlying the generation of visual consciousness in subjects. Without causing trauma to the individual, bioelectrical currents are collected from the surface of the cerebral cortex via electrodes to simultaneously record the oscillatory activity of multiple neurons within the brain. This method possesses wide-range and reliable measurement characteristics of cerebral cortex activity, such as... Figure 4 The diagram illustrates the generation and real-time acquisition of EEG signals. While traditional methods for acquiring EEG signals at the conscious level have been extensively studied, methods for eliciting unconscious EEG signals remain relatively scarce. This invention utilizes a binocular competition paradigm, allowing for the artificial control of whether visual stimuli are processed consciously or unconsciously. This invention also proposes a data-driven approach to mining signal characteristics at different stages of consciousness, and presents a flowchart for EEG signal acquisition and analysis based on a continuous flicker inhibition paradigm, as shown below. Figure 5 As shown in the image.
[0075] The EEG signal acquisition and analysis system consists of a 64-channel EEG acquisition cap, two EEG signal amplifiers, and a headbox acquisition card. The signal sampling frequency is 1000Hz. During the acquisition process, a bandpass analog filter is used for filtering from 0.5Hz to 70Hz. The electrodes are arranged according to the international 10-20 electrode arrangement standard, with a ground electrode placed on the forehead and a reference electrode located on the top of the head. The acquired EEG information is transmitted to a computer through the I / O port for further analysis.
[0076] After acquiring EEG signals, data preprocessing was first performed. Then, the EEG signals were re-segmented according to different stages of consciousness formation during the CFS experiment and decomposed into different frequency bands (Theta, Alpha, Beta, and Gamma). Feature extraction, frequency domain relative power analysis, and time domain variance analysis were then performed. Subsequently, a decision tree was trained using a random forest model to perform pointer calculation and classification on the EEG signals at different stages, thereby determining the brain information flow and functional connectivity during the consciousness transition process, and further identifying the dominant brain regions in the process of consciousness transition from non-existence to existence.
[0077] The overall flowchart of this invention is as follows: Figure 6As shown, the proposed unconscious EEG signal acquisition method based on the continuous flicker suppression paradigm utilizes a standardized precision optical component-based optical stereoscopic hardware platform to achieve full-process brain-computer interface research from unconsciousness to consciousness. Furthermore, an improved leaf model is used to automatically generate masking materials, and an adaptive image parameter setting algorithm is employed to adjust the experimental window size. The masking effect on visual emotion perception in the unconscious state, verified in experiments, demonstrates the system's reliability and stability in both hardware and software. The analysis technique using EEG signals from different stages of consciousness formation during the experiment effectively determines the brain's information flow and functional connectivity during the consciousness transition process.
[0078] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or terminal device. Unless otherwise specified, an element defined by the phrase "comprising..." or "including..." does not exclude the presence of additional elements in the process, method, article, or terminal device that includes said element. Additionally, in this document, "greater than," "less than," "exceeding," etc., are understood to exclude the stated number; "above," "below," "within," etc., are understood to include the stated number.
[0079] Although the above embodiments have been described, those skilled in the art, once they understand the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the above descriptions are merely embodiments of the present invention and do not limit the scope of patent protection of the present invention. Any equivalent structural or procedural transformations made using the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.
Claims
1. A subthreshold unconscious EEG signal acquisition system, comprising an automatic masking material generation module, a stimulus induction module, an EEG signal acquisition and analysis module, and an optical stereoscopic hardware platform, wherein, The automatic masking material generation module presets different masking stimulus templates, which can generate colored or grayscale images and square or circular dynamic Mondrian images. The stimulus induction module is implemented using software algorithms based on Matlab and Psychophysics Toolbox, and a synchronous triggering mechanism is designed based on marker point technology to ensure precise synchronization between stimulus presentation and EEG signal acquisition. The EEG signal acquisition and analysis module adopts a 64-channel EEG acquisition cap, two EEG signal amplifiers, and a headbox acquisition card structure. The EEG signal sampling frequency is 1000Hz, and a bandpass analog filter is used for filtering from 0.5Hz to 70Hz during the acquisition process. The optical stereoscopic hardware platform is built using standardized high-precision optical components. The basic components include a masking plate, guide rails, lead reflector, and corresponding frame and stand fixing device. The above four modules or platforms are independent in structure and function, but may need to work together in some applications, such as acquiring and analyzing EEG signals while using masking materials to study the impact of masking effects on cognitive processes.
2. A method for acquiring subthreshold unconscious EEG signals, characterized in that... It includes the following steps: (1) Run the stimulus-evoked module algorithm on the optical stereoscopic hardware platform to obtain the initial resolution of the screen used in the current experiment. The layout of the screen image is defined as follows: Ox×Oy≥(Ix×Iy)×D 2 +(Sx×Sy)×D 2 +l 2 ×D 2 ×2(1) Where Ox and Oy represent the width and height of the original screen resolution, respectively, while the angular resolution is defined as D (the product of the width or height and the angular resolution is the actual pixel length); when rendering the stimulus image, its width is defined as Ix and its length as Iy; the mask image is defined as its width as Sx and its length as Sy; in addition, both the stimulus box and the mask box are squares and their side length is defined as l; then, the present invention fills the entire screen based on the experimental window size; Set up a two-dimensional coordinate system with its origin at the top left corner of the experimental window, using pixels as the coordinate unit, extending to the right as the positive X-axis, and extending downwards as the positive Y-axis; (2) The position of the stimulus box and the mask box is determined by an adaptive screen parameter setting algorithm. Two square mask boxes and stimulus boxes of equal size are generated on both sides of the screen in a horizontally symmetrical manner, thereby completing the establishment of the visual experiment engine. The optical mirror is adjusted by a precision adjustment device so that the subject can achieve binocular fusion before the experiment. (3) The improved leaf algorithm is used to automatically generate masking material and display it in the masking box in step (2). The improved method is to randomly sample the neighborhood around each pixel in the stimulus image, thereby evaluating whether the pixel belongs to the foreground or the background based on the statistical characteristics of the neighborhood pixel values. Then, the Mondrian masking image with color blocks at different positions is generated by the position automatic generation module and the shape color block automatic generation module. (4) Using multi-angle and multi-color fusion stimulation materials displayed in the stimulation box in step (2) and inducing the subject's EEG signals, we can explore the effect of small changes in stimulation materials during unconscious visual perception on the breakthrough of consciousness and the functional connectivity of brain networks. (5) The EEG signal of the subject during the CFS experiment was collected in real time using the EEG extraction device. After signal amplification and noise reduction, the signal was transmitted to the computer for EEG signal preprocessing. The differences in the dominant brain regions were recorded and further analyzed when visual information gradually entered consciousness. At the same time, the analysis module can use machine learning algorithms such as random forest to visualize and analyze the functional connectivity of the brain.
3. The method for acquiring subthreshold unconscious EEG signals according to claim 2, characterized in that... The optical stereoscopic hardware platform in step (1) is built using standardized precision optical components; the subject's two eyes are presented with stimulus material and masking material flashing at a certain frequency according to the dominant eye division. This process is achieved by reflecting different fields of vision to the two eyes through two sets of mirrors.
4. The method for acquiring subthreshold unconscious EEG signals according to claim 2, characterized in that... In step (2), the visual experiment engine is implemented based on software algorithms of Matlab and Psychophysics Toolbox. After binocular fusion, it can achieve the effect that there is only one square frame in the eyes of the subject.
5. The method for acquiring subthreshold unconscious EEG signals according to claim 2, characterized in that... The improved leaf model in step (3) is an image noise generation model. Its characteristics include approximate scale invariance and some statistical model attributes of natural and synthetic images. The improved masking random color function adds three color output channels with assigned random numbers, which can generate color Mondrian images and further improve the masking types and effects.
6. The method for acquiring subthreshold unconscious EEG signals according to claim 2, characterized in that... In an exemplary embodiment of this application, step (4) involves using a Chinese-style dynamic emotion face database to transform the material in three directions: front, side, and back, and to render it using different warm and cool tones.
7. The method for acquiring subthreshold unconscious EEG signals according to claim 2, characterized in that... The signals collected in step (5) are decomposed into Theta, Alpha, Beta and Gamma bands and low-frequency and high-frequency oscillation power. At the same time, the power spectral density, differential entropy and functional brain network characteristics of the subject's EEG signal are calculated, and frequency domain relative energy spectrum power analysis and time domain variance analysis are performed to fully understand the brain functional connectivity and brain network activity of the subject during the experiment.