A Human-Computer Interaction Simulation Evaluation Method and System

Abstract

The present invention relates to the technical field of human-computer interaction evaluation, and specifically discloses a human-computer interaction simulation evaluation method and system. The method includes: constructing a task structure diagram based on task data and action parameters, and generating a roll interaction instruction set; configuring environmental parameters, constructing a deception simulation environment, and generating a perception stimulus set; parsing the perception stimulus and the task structure diagram, reorganizing the interaction interface, and generating a configured interaction structure diagram; constructing an association model and simulating a trust induction process, and generating a state regulation factor; collecting human factor data based on the state regulation factor, and generating a structured behavior label sequence; combining the label sequence and the task structure diagram for reasoning analysis, generating an adjustment control strategy, updating the interaction configuration, and completing feedback adjustment. The present invention can dynamically model the change of the interaction-induced trust state, and improve the perception ability and adjustment accuracy of the evaluation process for individual psychological cognition.

Description

Technical Field

[0001] The present invention relates to the technical field of human-computer interaction intelligent evaluation and cognitive state modeling, and particularly relates to a human-computer interaction simulation evaluation method and system. Background Art

[0002] With the continuous development of multi-modal interaction systems and immersive virtual simulation technologies, the dynamic recognition of users' mental states and the modeling of trust levels in complex task environments have become key research directions in the field of human-computer interaction evaluation. Most of the existing technologies rely on linear analysis of static behavioral indicators or physiological data, lacking the ability to jointly model the changes in interaction structures and the induced effects of the perceived environment, and it is difficult to effectively reflect the trust change trajectories and control state evolution processes of individuals during task execution.

[0003] At the same time, traditional human factors evaluation systems often have the following limitations: First, there is a lack of a structured linkage modeling mechanism between interaction content and perceived stimuli, making it difficult to establish a causal relationship between interaction configurations and users' psychological responses; second, there is a lack of a unified task-interaction-state mapping model, and it is impossible to achieve an accurate mapping between behavior labels and cognitive states; third, it is impossible to dynamically generate adjustment factors according to real-time trust fluctuations during the evaluation process, and it is difficult to support the automatic adjustment and feedback update of personalized interaction strategies.

[0004] Therefore, there is an urgent need for an intelligent evaluation method and system that integrates a configuration interaction structure diagram and a set of perceived stimuli for trust modeling, and generates state adjustment factors through a simulation evolution method to achieve dynamic perception and interpretable prediction of users' cognitive trust states, thereby providing a basis for subsequent interaction configuration optimization and personalized strategy update. Summary of the Invention

[0005] The present invention provides a human-computer interaction simulation evaluation method and system to solve the problem of how to construct a trust induction factor association model and simulate the trust state evolution process based on a configuration interaction structure diagram and a set of perceived stimuli, so as to generate state adjustment factors to dynamically reflect the psychological change trends and control states of individuals in multi-modal interaction tasks.

[0006] To solve the above technical problems, the present invention provides a human-computer interaction simulation evaluation method, including:

[0007] Based on flight mission data and action control parameters, construct a task structure diagram, extract attitude requirements and operation instructions, and generate a roll interaction instruction set;

[0008] Configure simulation environment parameters based on the roll interaction instruction set, construct a deception simulation environment, and generate a set of perceived stimuli;

[0009] Analyze the set of perceptual stimuli and the task structure diagram, construct and reorganize the interactive interface module, and generate a configured interactive structure diagram;

[0010] Based on the configured interactive structure diagram and the set of perceptual stimuli, construct an association model, simulate the induction process, and generate a state regulation factor, where the state regulation factor is:

[0011]

[0012] where, is the set of state regulation factors; is the time label corresponding to the th stage; is the regulation index value obtained according to the complex path integral at the th stage; is the corresponding historical value of the trust induction factor; is the set representation method; is a natural number, representing the position of the jth state regulation factor in the sequence; n is the total number of elements in the state regulation factor combination structure;

[0013] Collect human factor data according to the state regulation factor and the configured interactive structure diagram, and process and generate a structured behavior label sequence;

[0014] Combine the structured behavior label sequence and the task structure diagram for reasoning and analysis, generate an adjustment control strategy, update the interactive configuration and complete the feedback adjustment.

[0015] Furthermore, constructing the task structure diagram includes:

[0016] Obtain flight mission data and action control parameters, establish a task structure diagram and perform module decomposition to form a task stage sequence;

[0017] Extract attitude change requirements and operation instruction information from the task stage sequence, and construct an interactive control parameter set;

[0018] Perform dynamic mapping and roll motion transformation on the interactive control parameter set to generate a roll interaction instruction set.

[0019] Furthermore, constructing the deception simulation environment includes:

[0020] Obtain the roll interaction instruction set, and match environmental simulation configurations such as lighting, sound effects, and vibration;

[0021] Based on the environmental simulation configuration, construct a multi-modal deception simulation environment, and generate lighting, noise, and vibration simulation data;

[0022] Combine the lighting, noise, and vibration simulation data to generate a set of perceptual stimuli.

[0023] Furthermore, constructing and restructuring the interactive interface module includes:

[0024] Obtain the perception stimulus set and the task structure diagram, and analyze the interactive load requirements and interface response characteristics;

[0025] Based on the interface response characteristics and interactive load requirements, construct a set of configured interactive interface modules;

[0026] Structurally restructure the set of configured interactive interface modules to generate a configured interactive structure diagram.

[0027] Furthermore, constructing a correlation model and generating a state regulation factor includes:

[0028] Obtain the configured interactive structure diagram and the perception stimulus set, and establish a trust induction factor correlation model, where the trust induction factor correlation model is:

[0029]

[0030] wherein, is the trust induction factor generated in the th task stage; is the normalization function; is the trust functional value calculated in the th task stage;

[0031] Simulate and evolve the trust induction factor correlation model to generate trust state induction process data.

[0032] Furthermore, extract key state indicators from the trust state induction process data to generate a state regulation factor.

[0033] Furthermore, collecting human factor data includes:

[0034] Obtain the state regulation factor and the configured interactive structure diagram, and perform synchronous collection of human factor data;

[0035] Clean and normalize the collected human factor data to generate a set of human factor indicators.

[0036] Furthermore, perform structural mapping and label encoding on the set of human factor indicators to generate a structured behavior label sequence.

[0037] Furthermore, perform reasoning analysis including:

[0038] Obtain the structured behavior label sequence and the task structure diagram, and construct a trust performance reasoning model;

[0039] Perform task performance scoring and trust association analysis based on the trust performance reasoning model to generate a regulation control strategy;

[0040] The regulation control strategy is applied to the configuration interaction structure diagram to update the interaction configuration and complete feedback regulation.

[0041] Furthermore, a human-computer interaction simulation evaluation system includes:

[0042] A mission structure diagram building module is used to build a mission structure diagram and generate a roll interaction instruction set based on flight mission data and action control parameters;

[0043] A deception simulation environment construction module, used to construct a deception simulation environment based on the rolling interaction instruction set and generate a perception stimulus set;

[0044] An interactive interface configuration module, used to parse the perceptual stimulus set and the task structure diagram, and generate a configuration interactive structure diagram;

[0045] A trust state modeling module, used for generating a state regulation factor based on the configuration interaction structure diagram and the perception stimulus set;

[0046] A human factor data collection and processing module, used to generate a structured behavior label sequence according to the state control factor and the configuration interaction structure diagram;

[0047] The interaction strategy updating module is used to generate a regulation control strategy and update the interaction configuration by combining the structured behavior label sequence with the task structure diagram.

[0048] The key innovative features of the present invention include:

[0049] (1) Configure the interaction structure diagram and the collaborative modeling mechanism of the perceptual stimulus set: Open up the structured modeling link between perceptual input, task structure and interface configuration to achieve a unified expression of the interaction space.

[0050] (2) Trust-inducing factor association model and state regulation function construction mechanism: Integrating topological mapping, state function simulation and psychological factor identification methods, innovatively establishing the "stimulus-response-trust" evolution channel.

[0051] (3) Feedback adjustment and strategy iteration mechanism based on behavior labels: The structured behavior label results are used in reverse for interactive configuration adjustment and task adaptation strategy generation, forming a complete "behavior identification-performance reasoning-configuration update" closed loop.

[0052] The following are its main beneficial effects:

[0053] (1) The present invention constructs a linkage mechanism between the task structure diagram and the roll interaction instruction set. By performing structured modeling on flight task data and action control parameters, it significantly improves the mapping accuracy between tasks and interaction instructions, avoids the control lag problem caused by inconsistent task decomposition granularity in traditional solutions, and is conducive to enhancing the timeliness of interaction instructions and the task restoration degree.

[0054] (2) The present invention realizes the synchronous linkage between environmental perception and user response by constructing a multi-modal deception simulation environment and generating a perception stimulus set, significantly enhancing the realism and interference control ability of the simulation interaction system, and is conducive to stimulating more representative human factor responses.

[0055] (3) The present invention first proposes a method for associative modeling of trust induction factors based on the configuration interaction structure diagram and the perception stimulus set. Combining composite mathematical models such as topological analysis and function mapping mechanisms, it constructs a trust evolution function and a state regulation generation mechanism in the S400 module, which can dynamically identify the phenomena of trust transfer and manipulation deviation in the cognitive process, and is conducive to accurately inferring the evolution process of individual mental states.

[0056] (4) The present invention establishes a synchronous acquisition standard for multi-channel physiological and behavioral indicators through a human factor data acquisition and labeling mechanism driven by state regulation factors. Combining standardization processing and label coding technology, it can extract efficient structured behavior labels from complex electroencephalogram, eye movement, electrocardiogram and posture data, which is conducive to realizing accurate identification and interpretable prediction of cognitive behavior states. Description of the Drawings

[0057] Figure 1 It is a schematic flowchart of a human-computer interaction simulation evaluation method provided by an embodiment of the present application;

[0058] Figure 2 It is a structural block diagram of a human-computer interaction simulation evaluation system provided by an embodiment of the present application;

[0059] Figure 3 It is a schematic diagram of an interaction module of a spherical rolling platform provided by an embodiment of the present application for simulating dynamic perception ability;

[0060] Figure 4 It is a schematic diagram of an interaction module of a test pilot simulating a typical task scenario provided by an embodiment of the present application;

[0061] Figure 5 It is an equivalent test schematic diagram of inducing human-computer interaction and trust state provided by an embodiment of the present application.

[0062] Figure 6 It is a schematic diagram of quantitative evaluation of human-computer interaction characteristics provided by an embodiment of the present application. Detailed Embodiments

[0063] Example 1: Refer to Figure 1 , which is a schematic flowchart of a human-computer interaction simulation and evaluation method provided by an embodiment of the present invention. This process may at least include steps S100 - S600:

[0064] S100. Based on flight mission data and action control parameters, construct a task structure diagram, extract attitude requirements and operation instructions, and generate a roll interaction instruction set;

[0065] S200. Configure simulation environment parameters based on the roll interaction instruction set, construct a deception simulation environment, and generate a perception stimulus set;

[0066] S300. Analyze the perception stimulus set and the task structure diagram, construct and reorganize an interaction interface module, and generate a configured interaction structure diagram;

[0067] S400. Based on the configured interaction structure diagram and the perception stimulus set, construct an association model, simulate the induction process, and generate a state regulation factor;

[0068] S500. Collect human factor data according to the state regulation factor and the configured interaction structure diagram, and process it to generate a structured behavior label sequence;

[0069] S600. Combine the structured behavior label sequence and the task structure diagram for reasoning and analysis, generate an adjustment control strategy, update the interaction configuration, and complete the feedback adjustment.

[0070] Step S100 at least includes steps S110 - S130:

[0071] S110. Obtain flight mission data and action control parameters, establish a task structure diagram and perform module decomposition to form a task phase sequence.

[0072] The system first obtains flight mission data defined for a complex control task process. The flight mission data includes structured information such as task type, phase identifier, spatial path requirements, attitude control expectations, etc., reflecting the operation objectives and time logic of the entire task process.

[0073] At the same time, combined with the action control parameters collected by the control device used by the pilot, the action control parameters include input value records corresponding to the joystick, push rod, multi-dimensional knob, touch panel, etc., reflecting the control requirements of the user for the control platform at different stages.

[0074] Specifically, the system logically combines the flight mission data and the action control parameters, and establishes a task structure diagram through time stamps, task target identifiers, and input signal encoding rules. The task structure diagram is a set of multi-node directed relationship diagrams, where each node represents a state transition stage in the task, and the edge represents the logical flow path of the control instruction.

[0075] Further, decompose the task structure diagram into modules, and classify them according to the time axis sequence, operation complexity, and control type to form several task stage units. The task stage units describe the operation requirements of the system in different control stages, including attitude changes, path responses, state feedback, etc. The finally obtained task stage sequence serves as the execution basis for the entire task process and is called by the subsequent control parameter extraction step.

[0076] S120. Extract the attitude change requirements and operation instruction information from the task stage sequence, and construct an interactive control parameter set.

[0077] After the task stage sequence is constructed, the system analyzes the control characteristics included in each task stage unit item by item.

[0078] Specifically, for each task stage unit, extract the attitude change requirements, including indicators such as spatial rotation direction, angular velocity, pitch angle change rate, holding time, attitude stability requirements, etc. These indicators are used to describe the attitude adjustment tasks that the interactive platform should execute in this stage.

[0079] At the same time, the system extracts the operation instruction information corresponding to the above attitude changes from the control action mapping rules. The operation instruction information includes the control lever displacement amplitude, input direction, operation mode (continuous / pulse), signal response priority, etc. This operation instruction information is constructed based on the mapping logic between the task and the platform, forming an instruction pair between the manipulation input and the platform response.

[0080] Based on the attitude change requirements and operation instruction information, construct an interactive control parameter set. The interactive control parameter set is a multi-field structure, including operation input fields, target state fields, execution constraint fields, mapping condition fields, etc., and is used to characterize the control response behavior of the platform in a specific stage.

[0081] The interactive control parameter set serves as the basic input for generating the roll interaction control instruction, and its structure directly determines the instruction format, control dimension, and platform interface coding logic in the next stage.

[0082] S130. Perform dynamic mapping and roll action conversion on the interactive control parameter set to generate a roll interaction instruction set.

[0083] After constructing the interactive control parameter set, the system calls the dynamic mapping engine to perform platform adaptation processing on the operation requirements therein.

[0084] Specifically, the system first analyzes the target attitude requirements and corresponding operation instructions in the set of interactive control parameters based on the platform control model, and maps them into control signals that can be executed by the roll simulation platform. The control signals include direction control instructions, angular velocity adjustment instructions, action start / end identification instructions, attitude holding duration, etc.

[0085] Furthermore, according to the stage time control information in the task structure diagram, the above control signals are dynamically serialized to form a roll action scheduling list, which includes operation start time, action sequence, execution window length, parameter correction coefficient, etc.

[0086] The system constructs a roll interaction instruction set based on the roll action scheduling list. The roll interaction instruction set is a multi-dimensional data structure with complete time clues and instruction content, meeting the real-time response requirements of the platform for dynamic instructions in the whole process task simulation.

[0087] The roll interaction instruction set will serve as an important input for the construction of the simulation environment and the generation of perceptual stimuli in the subsequent steps. Specifically, it will be used as the environmental configuration benchmark in step S200 and used to control the triggering parameters of simulation factors such as lighting, sound effects, and vibration in S210, enabling the simulation platform to have staged dynamic task response capabilities and synchronous external feedback capabilities.

[0088] Through the three-stage task processing flow of S100 of the present invention, the structured decomposition of the manipulation task, the accurate modeling of the control logic, and the high-fidelity instruction output of the execution actions can be realized, ensuring the structured, continuous, and real-time response capabilities of the simulation platform to complex interaction tasks. The generated roll interaction instruction set provides basic input support for the construction of perceptual stimuli, the reconstruction of the interface layout, and the formation of state feedback, realizing the closed-loop closure of the system function from task intention to behavior feedback.

[0089] Step S200 at least includes steps S210 - S230:

[0090] S210. Obtain the roll interaction instruction set and match the environmental simulation configurations such as lighting, sound effects, and vibration.

[0091] The system first obtains the roll interaction instruction set, which is an action-driven instruction sequence mapped from the task structure diagram and operation control parameters, and includes multiple action elements such as roll direction, duration, target angle, rate type, and holding state.

[0092] Specifically, the system constructs a simulation response table based on the time distribution and action characteristics in the roll interaction instruction set. The simulation response table records the types of environmental feedback required for each interaction action, and divides them into a lighting channel, a sound effect channel, and a vibration channel according to the feedback channels.

[0093] Furthermore, for different types of action units, such as slow pose transitions, rapid roll responses, long-term hold states, etc., predefined environmental feedback models are matched. The lighting model is used to simulate visual stimulus changes such as natural light intensity changes, glare interference, and lighting transitions; the sound effect model is used to simulate auditory stimuli such as background noise, operation prompts, environmental audio, and feedback sounds; the vibration model is used to simulate tactile simulations such as platform responses, inertial feedback, and mechanical feedback.

[0094] The system constructs initial environmental simulation configuration parameters based on the three types of feedback models of lighting, sound effects, and vibration matched according to the simulation response table. The environmental simulation configuration parameters are a structured set of environmental response settings, including stimulus type, change period, output intensity, trigger window, and synchronization flag, and are used to control the synchronization behavior of the multimodal feedback device.

[0095] The above environmental simulation configuration parameters are used as the basic input for S220 to construct a multimodal simulation environment.

[0096] S220 constructs a multimodal deception simulation environment based on the environmental simulation configuration and generates lighting, noise, and vibration simulation data.

[0097] After obtaining the environmental simulation configuration parameters, the system constructs a lighting simulation environment, a sound effect simulation environment, and a vibration simulation environment according to the feedback channel classification.

[0098] Specifically, the lighting simulation environment uses a programmable light source, a dynamic effect light-shielding mechanism, and a background layer control system to simulate different light intensities, lighting angles, occlusion relationships, and reflection change processes, and dynamically adjusts the degree of scene light interference and visual stimulus changes during task execution.

[0099] The sound effect simulation environment loads elements such as background sounds, voice announcements, and operation prompt sounds parsed from the environmental simulation configuration parameters through a three-dimensional spatial sound effect simulation system, dynamically generates audio output segments matching the task phase, and fine-tunes parameters such as sound pressure, frequency, and spatial directivity.

[0100] The vibration simulation environment uses a tactile feedback module or a multi-axis platform control module to generate a dynamic vibration feedback sequence according to the vibration type setting parameters. The vibration feedback can include multiple types of tactile stimulus modes such as uniform oscillation, pulse impact, and continuous jitter.

[0101] The system samples and caches the simulation data output by the lighting feedback module, the sound effect feedback module, and the vibration feedback module respectively, and generates corresponding lighting simulation data, noise simulation data, and vibration simulation data. The simulation data is physical feedback drive data that has been aligned with the time window of the roll interaction action, and has characteristics such as channel synchronization, intensity hierarchy, and time series consistency.

[0102] The above three types of simulated data serve as the original input for the S230 combined perception stimulus set.

[0103] S230 combines the simulated data of light, noise, and vibration to generate a perception stimulus set.

[0104] After generating the simulated data of the three types of light, noise, and vibration, the system enters the multi-modal data fusion stage.

[0105] First, the system aligns the simulated data of light, noise, and vibration in chronological order according to the phase timestamp and synchronization identifier of the roll interaction instruction set. Through the channel mapping table, the data items on different channels are uniformly converted into time synchronization units based on the task execution phase.

[0106] Furthermore, perform structured fusion on the synchronized multi-channel data to construct perception stimulus units. Each perception stimulus unit consists of a set of time identifiers, light values, sound pressure values, vibration amplitudes, etc., and has an identifiable perception state and feedback trigger intensity.

[0107] Finally, the system combines all perception stimulus units according to the phase task requirements and interaction scenario logic to generate a perception stimulus set. The perception stimulus set, as a structured multi-modal feedback data set, will serve as the input basis for the interaction interface configuration and interface response regulation in S300, and participate in the modeling as the stimulus input factor of the trust state association model in S400.

[0108] Through the three-stage processing flow of S200, the system generates a highly consistent multi-modal perception stimulus set based on the roll interaction instruction set, and constructs a linkage input mechanism for subsequent interaction interface reconstruction, state-induced modeling, and behavior decision-making reasoning. The generated perception stimulus set has the characteristics of a complete environmental response structure, highly synchronized feedback channels, and dynamic matching of action logic, laying a feedback channel foundation for the human-computer interaction simulation system to construct a real perception environment, and ensuring the system to achieve accurate response and process consistency in dynamic interaction.

[0109] Step S300 at least includes steps S310 - S330:

[0110] S310. Obtain the perception stimulus set and the task structure diagram, and analyze the interaction load requirements and interface response characteristics.

[0111] In this step, the system obtains the perception stimulus set output in S230 and the task structure diagram constructed in the S100 stage, and performs interaction content analysis.

[0112] Specifically, the perception stimulus set is the light, sound effect, and vibration response data generated under the drive of the roll interaction instruction set, which has been synchronized with the operation behavior timeline; the task structure diagram consists of multiple task stages, describing the task execution logic, operation requirements, and feedback expectations.

[0113] Based on the synchronization relationship between the task stage and the perception stimulus, the system establishes a task-feedback mapping index table for annotating the interface response type, feedback pressure level, and user interaction intensity level required for each task node.

[0114] On this basis, the interaction load requirements are extracted. The interaction load requirements are three types of indicators that need to be evaluated when the system adapts the structure of the interaction interface, including the number of interaction channels (such as touch, voice, buttons, etc.), the trigger frequency per channel, the number and duration of simultaneously activated modules, and the pressure concentration in the focus area of the interface under specific perception stimuli.

[0115] At the same time, combined with the feedback type and intensity distribution of the perception stimulus set, the interface response characteristics are extracted. The interface response characteristics are used to describe the adaptation ability of the interface under different perception inputs, including parameters such as module display and hiding mechanism, font dynamic reconstruction, color contrast switching, interaction path contraction and expansion, etc.

[0116] The above interaction load requirements and interface response characteristics are used as parameter inputs for S320 to construct the interaction interface module set.

[0117] S320 constructs a configured interaction interface module set based on the interface response characteristics and interaction load requirements.

[0118] After obtaining the interaction load requirements and interface response characteristics, the system constructs an interaction interface module set based on the multi-dimensional adjustment rules.

[0119] Specifically, the system divides the task execution process into multiple interface control cycles according to the stage sequence of the task structure diagram, and configures a corresponding interface module requirement list for each cycle. The requirement list is differentially modeled according to the information density to be displayed, the number of user control paths, and the perception feedback intensity in the task stage.

[0120] Combined with the interface response characteristics, the system calls the preset interface module component library and selects appropriate interface module elements for instantiation. The interface module components include a graphic display area, an interaction control area, a feedback prompt area, an auxiliary navigation area, a data chart area, and a status reminder area, etc.

[0121] During the construction process, according to the channel types and intensity parameters provided by the perceptual stimulus set, configure the perceptual adaptation strategies for each interface module. For example, increase the interface contrast during the stage of strong light changes, reduce the text density and increase the voice output during the stage of strong seismic sensations, and reduce the dynamics of image animations during the multi-sound channel stage, etc.

[0122] Furthermore, uniformly encode the interaction relationships, focus switching logics, and priority control strategies among the interface modules to form a structured set of configuration interaction interface modules. Each set unit represents the interface configuration state under a specific task cycle, including the required modules, module layouts, response mechanisms, trigger boundaries, and module linkage mappings.

[0123] The set of configuration interaction interface modules serves as the input for S330 to generate a configuration interaction structure diagram.

[0124] S330: Structurally reorganize the set of configuration interaction interface modules to generate a configuration interaction structure diagram.

[0125] In this step, the system obtains the set of configuration interaction interface modules, performs a structured reorganization operation, and constructs a configuration interaction structure diagram for driving interface interaction rendering and the logical linkage of interaction tasks.

[0126] Specifically, the system performs semantic clustering on the set of interface modules corresponding to each task stage, and based on the relevance of interaction functions and the coordination of visual layouts, integrates multiple modules into module groups and establishes the logical control relationships between the module groups.

[0127] Furthermore, based on the control sequence and stage trigger boundaries in the task structure diagram, the system nests the module groups into the structure diagram in the execution order. Each structure node corresponds to a module group, and its internal defines the display structure, input channel type, status response path, and callback event chain.

[0128] At the same time, the system combines the stimulus timing information in the perceptual stimulus set to mark the interface deformation rules of each structure node in the multi-modal feedback scenario, such as fade-in strategies, sliding rearrangement, module fade-in and fade-out, mutually exclusive area layout, etc.

[0129] Finally, the system performs hierarchical division and path parsing on the structure diagram of the entire module group to generate a configuration interaction structure diagram. This structure diagram expresses the organizational relationship between interface display and user interaction in the entire task process in a graph structure, including elements such as task stage indexes, module node distributions, transfer paths between nodes, and perceptual feedback response strategies.

[0130] The configuration interaction structure diagram will serve as the core input basis for subsequent S400 to construct a trust induction factor association model and state regulation factors.

[0131] Through the S300 three-stage process, the system can achieve the semantic decoupling and interface module adaptation and integration of the task structure diagram and the perception stimulus set, and generate a complete configuration interaction structure diagram. This structure diagram has the capabilities of staged layout, module linkage logic, and multimodal feedback adaptation, providing basic data support for real-time interface response, structure deformation, and behavior mapping in the interaction task, and providing a basis for the scenario input structure for the subsequent trust modeling module, strengthening the controllability and adaptability of the human-computer interaction system.

[0132] Step S400 includes at least steps S410 - S430:

[0133] S410. Obtain the configuration interaction structure diagram and the perception stimulus set, and establish a trust induction factor association model.

[0134] In this step, the system obtains the configuration interaction structure diagram from S330 and the perception stimulus set from S230 as the input for constructing the trust induction factor association model.

[0135] Specifically, the configuration interaction structure diagram includes multi-stage interface module nodes and their response paths; the perception stimulus set includes the channel information of light, sound, and vibration at each interaction moment. The system needs to judge the multi-dimensional non-linear association among the stimulus intensity, interaction load, and interface module complexity.

[0136] To this end, the following trust induction function model is constructed:

[0137] ① Trust induction functional model

[0138]

[0139] Where:

[0140] : The trust induction function functional, used to describe the total influence of the multi-channel stimulus and the interaction structure on the formation of user trust;

[0141] : The spatio-temporal integration domain defined by the interaction process (for example: task execution time interval × control channel set);

[0142] : The weight parameters of the environmental stimulus channels, corresponding to the importance of the light, sound effect, and vibration channels respectively;

[0143] : The brightness response function of the light stimulus channel, is the time series of the light intensity value within the task period;

[0144] : The sound pressure function of the sound effect stimulus channel, represents the sound pressure change value at the current stage;

[0145] : The amplitude function of the vibration stimulation channel, is the sequence of the vibration response in the time domain;

[0146] : The interface complexity function, and the input quantities are:

[0147] : The number of interface modules;

[0148] : The interface module switching frequency;

[0149] : The focus area concentration coefficient;

[0150] : In the spatio-temporal domain the infinitesimal integration element.

[0151] The trust induction functional model is used to describe the combined action intensity of multi-channel stimulation and interaction structure, and its result will be used as the input of the trust induction factor generation model.

[0152] ② Trust induction factor correlation model

[0153]

[0154] Wherein:

[0155] : The trust induction factor generated in the -th task stage;

[0156] : The normalization function, usually taking the sigmoid function, that is , mapping the induction value to interval;

[0157] : The trust functional value calculated in the -th task stage.

[0158] The generated value is used as the input parameter of the simulation evolution module in S420.

[0159] S420. Simulate and evolve the trust induction factor correlation model to generate trust state induction process data.

[0160] To further describe the dynamic change trend of the induction process, this step describes the trust propagation state.

[0161] ③ Trust evolution state field model (tensor form)

[0162]

[0163] Wherein:

[0164] : The trust state tensor component is used to describe the local curvature characteristics of a point in the trust state space in the direction;

[0165] : The affine connection, that is, the change directionality tensor of the trust induction factor, represents the th th th

[0166] : In the trust induction factor field, it is the coupling influence amount on the and th input dimension direction that triggers the th trust state output dimension

[0167] : The partial derivative symbol represents the local differential operation on the variable;

[0168] : Represents the partial derivative of the affine connection with respect to the stimulus input dimension , and is used to characterize the perturbation influence of the stimulus input change in this direction on the trust transfer direction.

[0169] , : The stimulus input dimension (e.g., feedback time, channel type, task phase number, etc.);

[0170] : It is the product term between affine connection tensors and is used to characterize the "composite deformation trend of the connection" in the trust state space, representing the propagation process of the channel driven jointly by on the channel;

[0171] : It is the product term between affine connection tensors and is used to characterize the "composite deformation trend of the connection" in the trust state space, representing the restrictive perturbation of the channel on the linkage to the channel;

[0172] All : Are all from those in S410 The tensor tension function constructed by the derivation of the trust factor field reflects the propagation and deformation trends induced by trust in the state space.

[0173] This formula is used to simulate the non-Euclidean state offset trajectory during the trust induction process under complex multi-channel interaction conditions.

[0174] ④ State induction evolution function

[0175]

[0176] Where:

[0177] : The cumulative trust state level at time point ;

[0178] : The trust induction factor generated in the th stage (derived from S410);

[0179] : The stimulus response intensity coefficient of the th task stage;

[0180] : The trust decay coefficient of the th task stage, used to characterize the time weakening of the trust influence;

[0181] : The start time of the th task stage;

[0182] : The exponential decay term, reflecting the decreasing trend of the induction influence over time;

[0183] : The time element.

[0184] The function is used to characterize the dynamic accumulation and decay trends of the trust level. The discrete sampling results of this function will enter S430 as the basis for extracting the state regulation factor.

[0185] S430. Extract key state indicators from the data of the trust state induction process to generate state regulation factors.

[0186] Based on the evolution results of S420, this step extracts state indicators that have practical regulatory significance for subsequent human factor data collection.

[0187] Specifically, identify the inflection points, mutation regions, and saturation regions in the trust induction process.

[0188] ⑤ State regulation index extraction function

[0189]

[0190] Wherein:

[0191] : The th state regulation index, indicating the influence degree of this state loop in psychological induction;

[0192] : The th task state's induction path loop on the complex plane (for example, the closed stimulation trajectory of the combination of repeated vibration + highlighted interface in a certain period);

[0193] : A complex function, the real part is the trust evolution value , and the imaginary part is the perceptual stimulus perturbation term;

[0194] : Indicates integration on the closed path ;

[0195] : The infinitesimal variable increment on the complex plane;

[0196] : Indicates taking the absolute value of the induction intensity under this path.

[0197] This integral form is used to measure whether a certain interface combination and stimulus sequence form a "psychological induction closed-loop" mechanism in continuous task states.

[0198] ⑥ State regulation factor combination structure

[0199]

[0200] Wherein:

[0201] : The set of state regulation factors;

[0202] : The th time label corresponding to the stage;

[0203] : The regulation index value obtained by complex path integration at the th stage;

[0204] : The corresponding historical value of the trust induction factor (output from S410);

[0205] : The set representation method, indicating that the regulation factors are structured with state number indexing;

[0206] is a natural number, representing the position of the j-th state regulation factor in the sequence;

[0207] n is the total number of elements in the state regulation factor combination structure.

[0208] This structure will be passed into S500 as the regulation signal source in the human factor data acquisition process to drive the dynamic configuration of parameters such as the acquisition window, feedback interval, and data cleaning threshold.

[0209] Explanation of the closed-loop connection of the front and back steps:

[0210] The value generated jointly driven by the perception stimulus set and the configuration interaction structure diagram in S410 is the kernel function term in the evolution model of S420;

[0211] The function result generated in S420 serves as the basis for S430 to extract regulation indicators;

[0212] The state regulation factor finally output by S430 will be directly used for the adjustment mechanism of the human factor data acquisition module (S510–S530) in S500;

[0213] The entire S400 module completes the logical closed-loop from interactive perception feedback → trust factor establishment → state simulation → parameter extraction.

[0214] Through the trust-induced evolution modeling method designed by the S400 module, which integrates functional analysis, differential geometry, and complex analysis theories, the system can accurately identify the comprehensive impact of the combination of perception stimuli and interface structures on the trust state under multi-channel conditions, extract key state control factors based on the evolution model, and achieve precise regulation of the subsequent human factor data acquisition path and label determination strategy. The established correlation model has the advantages of mathematical stability, input index consistency, and induction process continuity, significantly enhancing the system's structured modeling ability for user responses in complex interactive scenarios.

[0215] Step 500 includes at least steps S510 - S530:

[0216] S510. Obtain the state regulation factor and the configuration interaction structure diagram, and perform synchronous acquisition of human factor data.

[0217] In this step, the system obtains the state control factor from S430 and the configuration interaction structure diagram from S330, and jointly drives the synchronous collection process of human factors data. The state control factor is a dynamic psychological response control parameter calculated based on the multi-channel perception stimulation and interaction structure induced model, which is used to guide the sampling window, channel selection and data accuracy setting of the human factors collection process; the configuration interaction structure diagram is used to provide the human-computer interface layout structure, focus distribution path and interaction node information under the task stage, ensuring that the collection is highly aligned with the behavior context.

[0218] Specifically, the system dynamically adjusts the data acquisition window according to the task timing parameters and stimulus synchronization flags in the state control factor. In the high-load interaction state, the sampling window is shortened and the synchronization accuracy is improved to capture the rapidly changing human response; in the low-intensity interaction state, regular periodic sampling is used to ensure data continuity and integrity.

[0219] At the channel level, the system enables a multi-source parallel acquisition mechanism, covering but not limited to the following data channels:

[0220] ECG signal acquisition channel: records the user's heart rate variability and ECG waveform structure during the interaction process;

[0221] EEG signal acquisition channel: obtains electrical activity data in the head area to reflect the user's concentration and cognitive load level;

[0222] Eye tracking channel: Identify the user's gaze path, pupil contraction rate and gaze rest point during interface interaction;

[0223] Posture motion channel: Captures body posture adjustment, micro-motion response, and head and shoulder movement sequences based on inertial sensor units;

[0224] Facial electromyography channel (if adapted): records the user's micro-expression reactions and eyelid and mouth corner muscle activity trajectories;

[0225] Hand micro-motion channel (if adapted): captures hand pressure, click rhythm and stagnation characteristics during input operations.

[0226] All of the above channels are triggered for collection according to the dynamic frequency set by the state control factor, and combined with the mapping relationship between the interaction nodes and controls identified in the configuration interaction structure diagram, a data-event synchronization record table is established to achieve precise alignment of the task behavior chain and the physiological response chain.

[0227] S520. Clean and standardize the collected human factor data to generate a human factor indicator set.

[0228] After completing the multi-channel human factor data synchronous acquisition, the system enters the data processing stage, where operations such as cleaning, filtering, standardizing, and index extraction are performed on the original human factor data to form a human factor index set in a unified format.

[0229] First, the system performs anomaly detection and noise filtering processing:

[0230] Identify R-wave peaks and remove artifacts from the electrocardiogram signal, removing spikes caused by sensor detachment or noise interference;

[0231] Use high-frequency oscillation identification within a sliding window for the electroencephalogram signal, remove electromyogram interference, and perform artifact channel masking;

[0232] Extract fixation points and saccade segments from the eye movement data, and interpolate and complete saccade events and defocus drift points;

[0233] Perform three-axis filtering on the pose data, smooth the pose transition boundaries, and unify the motion direction and speed dimension;

[0234] Perform baseline drift removal on the electromyogram and hand data, and normalize the input pressure and micro-motion amplitude.

[0235] Subsequently, the system performs behavior interval slicing on the data of each channel according to the interface node sequence and interaction behavior number identified in the interaction structure diagram, and extracts the corresponding index dimensions. The indexes include but are not limited to:

[0236] Heart rate variability, average heart rate, peak response duration, heart rate recovery rate;

[0237] Electroencephalogram frequency domain energy distribution (α, β, θ bands), attention index, cognitive load index;

[0238] Fixation concentration, target area fixation ratio, line-of-sight jump frequency;

[0239] Posture stability coefficient, action amplitude change rate, asymmetric body offset index;

[0240] Facial electromyogram response amplitude, mouth corner elevation angle, blink frequency;

[0241] Hand stability coefficient, operation rhythm stability, average value of the input point pressure sequence, etc.

[0242] The above indexes are uniformly encapsulated into human factor index units, and the numerical expressions of each channel and each stage are recorded in the form of key-value pairs. The system constructs a human factor index set structure body with a unified structure, and each structure body unit aligns with the task stage number, interaction node code, data sampling time, index field list, and original interval index.

[0243] Finally, the system inputs the complete human factor index set into S530 for behavior label mapping.

[0244] S530. Structurally map and label-encode the human factor index set to generate a structured behavior label sequence.

[0245] After the construction of the human factor index set is completed, the system performs the construction and encoding operations of structured behavior labels. The goal of this step is to structurally map the abstract numerical responses in the physiological data into semantic labels with behavioral meanings for subsequent strategy reasoning and adjustment strategy generation phases.

[0246] The system first calls the behavior label rule library and loads the corresponding label mapping template according to the task type, interaction phase, and interface structure definition. This template defines the observable behavior label types (such as concentration, high load, stress response, lost, confused, unresponsive, etc.) in each task phase and sets the index thresholds and combination conditions.

[0247] Specifically, the system processes the human factor index set in each task phase as follows:

[0248] Match the combination of "sudden increase in heart rate + sudden drop in HRV" in the electrocardiogram index and label it with the "stress response" label;

[0249] Specifically, the definition of "sudden increase in heart rate"

[0250] : represents the instantaneous heart rate at time , with the unit of beats per minute;

[0251] : represents the heart rate change rate, usually defined as:

[0252]

[0253] where is the time interval between two consecutive measurements.

[0254] When (threshold), it is marked as a "sudden increase in heart rate" event.

[0255] Definition of sudden drop in HRV

[0256] : represents heart rate variability, which is the standard deviation or frequency domain feature of the R-R interval within a short time window;

[0257] : represents the change amount of heart rate variability, which can be defined as:

[0258]

[0259] If , it represents the "sudden drop in HRV" phenomenon.

[0260] Combined judgment: When both and are satisfied, generate the "stress response" label.

[0261] Match the combination of "high beta wave energy + alpha wave suppression" in the EEG index and mark it with the "high cognitive load" label;

[0262] Specifically, the definition of "high beta wave energy": : It represents the energy power of the EEG beta wave (13–30 Hz) frequency band, which is derived from frequency domain transformation (such as FFT); if (set the energy threshold), it is considered that the beta wave is active.

[0263] The definition of "alpha wave suppression": : It represents the energy of the alpha wave (8–12 Hz) frequency band; : The change amount of alpha wave power:

[0264]

[0265] If , that is, the alpha wave drops significantly, it is judged as alpha wave suppression.

[0266] Combined judgment: When both and are satisfied, generate the "high cognitive load" label.

[0267] Match the "non-target area fixation ratio > threshold" in the eye movement index and mark it with the "attention deviation" label;

[0268] Match the "rapid movement change + balance fluctuation" in the posture index and mark it with the "instability" label;

[0269] Match the "operation rhythm interruption + sudden pressure drop" in the hand index and mark it with the "operation hesitation" label;

[0270] Other labels such as "low motor response", "high concentration", "stable execution", etc. are generated based on rules.

[0271] The system establishes a two-way index between the above-mentioned behavior labels and their corresponding index structures, sorts them in chronological order, and constructs a structured behavior label sequence. Each label sequence unit contains the following fields:

[0272] Timestamp;

[0273] The task phase and interaction node number to which it belongs;

[0274] Label category;

[0275] Label trigger source metric set reference;

[0276] Confidence score (calculated based on historical matching frequency and metric variation degree).

[0277] This structured behavior label sequence serves as the inference input basis for S600, and penetrates through the subsequent task performance modeling and strategy adjustment modules.

[0278] Through the three-stage processing flow of this step, the system can accurately collect multi-dimensional human factor response data of users under the joint guidance of the structured interface diagram and the state regulation factor, form a structured behavior label sequence with high consistency, multi-channel, and strong semantics, and construct a clear and controllable individual behavior portrait for subsequent performance analysis and strategy adjustment. The system has the integrated ability of acquisition - processing - mapping - output, realizes the structural-level conversion from physiological signals to cognitive behavior labels, and effectively enhances the interpretability of the human-computer interaction evaluation system for individual responses under complex dynamic tasks and the ability to formulate response strategies.

[0279] Step S600 at least includes steps S610 - S630:

[0280] S610. Obtain the structured behavior label sequence and the task structure diagram, and construct a trust performance inference model.

[0281] In this step, the system obtains the structured behavior label sequence output by S530 and the task structure diagram information output by S130, and fuses them to construct a trust performance inference model. The structured behavior label sequence is a set of sequential behavior labels shown by the user during the task, including label type, associated metrics, stage identifiers, interaction nodes, label confidence, etc.; the task structure diagram contains the associated mapping of the logical order of task stages, action control parameters, and posture change requirements.

[0282] Specifically, the system establishes a task flow diagram according to the stage numbers and behavior paths in the task structure diagram, maps the structured behavior labels to the nodes of each stage in the task process, and forms a label-node comparison table. Based on this, the system calculates the following inference input vectors:

[0283] Node behavior label density vector: Statistically analyze the label frequency and label type distribution that appear on each task node;

[0284] Average node behavior confidence: Calculate the average confidence of each type of label on each task node;

[0285] Label cross-stage propagation path: Identify the same type of label paths that appear continuously on multiple task nodes;

[0286] Task phase transition context: Extract the order and triggering conditions between phases in the task structure diagram to support subsequent causal modeling.

[0287] On this basis, the system constructs a trust performance inference model based on the task behavior label graph. This model adopts a graph structure inference framework, takes task nodes as graph nodes, builds the phase transition relationship as an edge weight channel, and embeds information such as the corresponding label density, confidence level, and behavior type distribution in the graph nodes.

[0288] The system further establishes a node clustering weight function within the graph to identify significant regions of trust fluctuations in the task. The model automatically extracts the following core inference features:

[0289] High-frequency label coupling region;

[0290] Low-confidence fluctuation frequency band;

[0291] Non-target path drift label marking segment;

[0292] Multi-channel conflict trigger path, etc.

[0293] The above model will be used as the decision basis for scoring and policy generation in S620, and has the ability of self-learning, and can continuously optimize the inference accuracy in multiple rounds of tasks.

[0294] S620. Perform task performance scoring and trust correlation analysis based on the trust performance inference model, and generate a regulation and control strategy.

[0295] This step is based on the trust performance inference model constructed in S610, performs task performance scoring and trust correlation analysis, and finally generates a regulation and control strategy. This process takes task nodes as the smallest analysis unit, integrates structured behavior labels and node state features in the task execution path, and evaluates the performance and trust level of individuals in each phase of the task.

[0296] Specifically, the system first performs performance scoring processing on each task node. The scoring inputs are:

[0297] The number and distribution type of node behavior labels;

[0298] Label confidence;

[0299] Whether the task action parameters are fully executed;

[0300] Node residence duration and interactive interface response information.

[0301] According to the scoring template, the system outputs the following scoring results for each node:

[0302] Behavior completion degree score;

[0303] Operation coherence score;

[0304] Attention stability score;

[0305] Stress response sensitivity score.

[0306] Subsequently, the system constructs a trust association graph within the task according to the sequential logic between nodes and the temporal characteristics of behavior labels. This graph is formed based on the following rules:

[0307] If multiple consecutive nodes all have the "attention deviation" label, a weak trust connection segment is formed;

[0308] If a node shows "high cognitive load", but the node immediately following it is "high stability", a trust recovery segment is formed;

[0309] If "operation hesitation" and "high concentration" frequently alternate in multiple stages, it is marked as a trust fluctuation segment.

[0310] Based on the above results, the system identifies the overall trust state trend of the task and aggregates the scoring data to generate a regulatory control strategy structure. This strategy structure includes:

[0311] Identifiers of task phases that need to be intervened first;

[0312] Recommended interactive interface elements to be adjusted;

[0313] Recommended types of changes in stimulation channels to be increased;

[0314] Recommended task rhythm parameters to be slowed down or speeded up;

[0315] Recommended interactive complexity parameters to be reduced or enhanced.

[0316] The above regulatory control strategy will be used as the input of S630 to directly drive the update behavior of the interactive configuration.

[0317] S630. Apply the regulatory control strategy to the configured interactive structure diagram, update the interactive configuration and complete the feedback adjustment.

[0318] In this step, the system applies the regulatory control strategy output by S620 to the configured interactive structure diagram output by S330, completes the dynamic update of the interactive configuration, and establishes a feedback adjustment link to achieve closed-loop control.

[0319] Specifically, the system locates the corresponding interface node structure in the configured interactive structure diagram according to the target task phase identified in the regulatory control strategy, and processes the node parameters as follows:

[0320] Interactive complexity adjustment: According to the complexity coefficient indicated in the strategy, the system can increase or decrease the number of interface controls, display logic, and number of labels;

[0321] Stimulation channel update: If the strategy suggests enhancing the perceived stimulation intensity, the system can increase the light, vibration, or sound effect simulation signals during the corresponding task phase;

[0322] Interaction frequency adjustment: If the strategy suggests slowing down the rhythm, the system inserts waiting indicators, interaction confirmation prompts, or auxiliary guidance information at the corresponding nodes;

[0323] Interface focus reconstruction: The system reconstructs the layout of interface controls based on the focus areas identified from the user's previous gaze and action response data, reducing the user's operation load;

[0324] Feedback loop embedding: The system embeds real-time feedback sensing trigger points in all modified nodes to record the new response data generated by the user after the configuration update.

[0325] All update operations perform parameter replacement and module update without changing the task structure logic, ensuring the integrity of the task closed-loop.

[0326] The system further encapsulates the above update information into an interaction configuration update log and marks its corresponding task phase, adjustment trigger source, policy version, and user status identifier. This log can be used as an evolutionary training data source in subsequent evaluations to support the continuous optimization and learning of the inference model.

[0327] The system finally reloads the updated configuration interaction structure diagram into the task engine and restarts the task simulation to achieve the dynamic verification and effect feedback of the adjustment control strategy under the current user behavior state.

[0328] Through the execution of this module, the system can construct a trust performance inference model with context awareness based on structured behavior tags and task structure diagrams, mine the implicit trust changes and performance characteristics in user behavior, further generate dynamic interaction adjustment strategies and feedback them to the configuration structure diagram in real time, realizing the intelligent adaptation and personalized adjustment between task configuration and user status. This module has the capabilities of causal reasoning modeling, multi-label evaluation, and configuration closed-loop feedback, effectively improving the intelligence, adaptability, and dynamic evolution level of human-computer interaction simulation evaluations.

[0329] Embodiment 2: Figure 2 Show a structural block diagram of a human-computer interaction simulation evaluation system according to an embodiment of the present invention. As Figure 2 shown, the structure may include:

[0330] A task structure diagram construction module 10, which is used to construct a task structure diagram and generate a roll interaction instruction set based on flight task data and action control parameters.

[0331] Obtain flight task data and action control parameters;

[0332] Build a task structure diagram and decompose it into a sequence of task phases;

[0333] Extract the attitude change requirements and operation instruction information, and construct an interactive control parameter set;

[0334] Perform dynamic mapping and roll motion transformation on the interactive control parameter set to generate a roll interaction instruction set.

[0335] This module realizes the mapping from the task space structure to the instruction space, and provides a unified control baseline for environment simulation and interaction construction.

[0336] The deception simulation environment construction module 20 is used to configure the simulation environment parameters based on the roll interaction instruction set, construct a multi-modal deception simulation environment, and generate a perception stimulus set.

[0337] Obtain the roll interaction instruction set, call the environment configuration library, and match the lighting, sound effects, and vibration parameters;

[0338] Construct a simulation environment, including lighting change, noise simulation, and vibration interference generation;

[0339] Combine the simulation output results into a perception stimulus set as the input context for interface interaction.

[0340] This module enables the simulation system to present a highly complex and realistic interaction environment and supports the output of multi-source interference stimuli.

[0341] The interactive interface configuration module 30 is used to parse the perception stimulus set and the task structure diagram, construct and reorganize the interactive interface module, and generate a configured interactive structure diagram.

[0342] Extract the interactive payload and response characteristics from the perception stimulus set and the task structure diagram;

[0343] Construct a set of interactive interface modules, identify the distribution of interface controls and behavior nodes;

[0344] Perform a structured reorganization operation on the set of interface modules to generate a configured interactive structure diagram.

[0345] This module realizes the dynamic interface construction and personalized interaction design based on perception input and task goals.

[0346] The trust state modeling module 40 is used to generate a state regulation factor based on the configured interactive structure diagram and the perception stimulus set.

[0347] Obtain the configured interactive structure diagram and the perception stimulus set, and establish a trust induction factor association model;

[0348] Simulate the trust induction process based on user behavior characteristics and stimulus response;

[0349] Extract key status indicators (such as attention to stability, task responsiveness, etc.) to generate status regulation factors.

[0350] This module constructs the psychological state evolution mechanism in the human-computer interaction process and is the core support for subsequent human factor evaluation and behavior label reasoning.

[0351] The human factor data acquisition and processing module 50 is used to collect human factor data according to the status regulation factors and the configured interaction structure diagram, and process and generate a structured behavior label sequence.

[0352] Perform synchronous acquisition of human factor data such as electrocardiogram, eye movement, electroencephalogram, and posture;

[0353] Clean, filter, and normalize the original human factor data to extract a standardized human factor index set;

[0354] Map the human factor index set to a structured behavior label library to generate a behavior label sequence and mark the user's behavior status and deviation characteristics.

[0355] This module realizes the conversion from underlying physiological and behavioral signals to recognizable cognitive behavior labels.

[0356] The interaction strategy update module 60 is used to combine the structured behavior label sequence and the task structure diagram to generate a regulation and control strategy and update the interaction configuration to complete feedback regulation.

[0357] Obtain the structured behavior label sequence and the task structure diagram to construct a trust performance reasoning model;

[0358] Perform performance scoring and trust dynamic analysis on the task process to identify problem stages and behavior fluctuation segments;

[0359] Generate a regulation and control strategy based on the reasoning result and feedback it to the configured interaction structure diagram;

[0360] Adjust the interface complexity, stimulus intensity, and interaction frequency to achieve real-time task interaction adaptation and personalized optimization.

[0361] This module forms a system intelligent feedback closed-loop and realizes the adaptive evolution logic of configuration - acquisition - analysis - regulation.

[0362] This embodiment has the following beneficial effects:

[0363] (1) Achieved a highly integrated task - interaction - human factor full-link modeling. By integrating flight task parameters, simulation environment, and human factor data into a structured modeling system, the separation problem between traditional task simulation systems and psychological behavior evaluation systems is solved.

[0364] (2) An inferable and adjustable cognitive state recognition mechanism is established. By associating the trust induction factor model with the structured behavior label sequence, the system can identify the trust changes and operation instability states of individuals during the interaction process, and reversely adjust the interaction configuration to achieve dynamic adaptation.

[0365] (3) Support a closed-loop of traceable behavior data mapping and intelligent policy update. The structured behavior label sequence and the adjustment control strategy construct a full-process evaluation - feedback - re-evaluation closed-loop logic, supporting the rapid iteration and personalized customization of interaction strategies.

[0366] (4) The system modules have clear boundaries and unified data interfaces, facilitating integration and expansion. Data exchange is carried out between modules based on a standardized structure body, supporting the access of multiple sensing channels and simulation output forms, and having good system adaptability and expandability.

[0367] Obviously, the embodiments described above are only a part of the embodiments of this application, rather than all embodiments. The accompanying drawings show the preferred embodiments of this application, but do not limit the patent scope of this application. This application can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosed content of this application more thorough and comprehensive. Although this application has been described in detail with reference to the foregoing embodiments, for those skilled in the art, they can still modify the technical solutions recorded in the foregoing specific embodiments, or perform equivalent replacements for some of the technical features. Any equivalent structure directly or indirectly using the content of this application's specification and drawings in other related technical fields is similarly within the scope of the patent protection of this application.

Claims

1. A human-computer interaction simulation evaluation method, characterized in that: The following steps are involved: Based on the flight mission data and motion control parameters, the mission structure diagram is constructed, the attitude requirements and operation instructions are extracted, and the roll interaction instruction set is generated; Based on the rolling interaction instruction set, configure simulation environment parameters, build a deception simulation environment, and generate a perception stimulus set; Parsing the perceptual stimulus set and the task structure diagram, constructing and reorganizing the interactive interface module, and generating a configuration interactive structure diagram; Based on the configuration interaction structure diagram and the perceptual stimulus set, an association model is constructed to simulate the induction process and generate a state regulation factor, wherein the state regulation factor is: in, is the set of state control factors; For the The time label corresponding to each stage; For the The control index value obtained based on the complex path integral at each stage; is the corresponding historical value of the trust inducing factor; is a set representation; is a natural number, indicating the position of the jth state control factor in the sequence; n is the total number of elements in the state control factor combination structure; Collecting human factor data according to the state control factor and the configuration interaction structure diagram, and processing and generating a structured behavior label sequence; The structured behavior label sequence is combined with the task structure diagram to perform reasoning analysis, generate a regulation control strategy, update the interaction configuration and complete feedback regulation.

2. The method according to claim 1, characterized in that The task structure diagram includes: Obtain flight mission data and action control parameters, establish a mission structure diagram and perform module decomposition to form a mission phase sequence; Extracting posture change requirements and operation instruction information from the task phase sequence to construct an interactive control parameter set; The interactive control parameter set is dynamically mapped and converted into a rolling action to generate a rolling interactive instruction set.

3. The method according to claim 1, characterized in that Building a deception simulation environment includes: Obtain the rolling interaction instruction set and match the environment simulation configuration such as lighting, sound effects and vibration; Constructing a multimodal deception simulation environment based on the environment simulation configuration, and generating illumination, noise and vibration simulation data; The light, noise and vibration simulation data are combined to generate a perceptual stimulus set.

4. The method according to claim 1, characterized in that: Constructing and reorganizing the interactive interface modules include: Acquire the perceptual stimulus set and the task structure diagram, and analyze the interaction load requirements and interface response characteristics; Based on the interface response characteristics and interaction load requirements, a set of configuration interaction interface modules is constructed; The configuration interaction interface module set is structurally reorganized to generate a configuration interaction structure diagram.

5. The method according to claim 1, characterized in that Building an association model and generating state control factors include: The configuration interaction structure diagram and the perception stimulus set are obtained to establish a trust inducing factor association model, wherein the trust inducing factor association model is: in, For the Trust-inducing factors generated at each task stage; is the standardization function; For the The trust functional value calculated in each task stage; The trust inducing factor association model is simulated and evolved to generate trust state inducing process data.

6. The method according to claim 5, characterized in that Key state indicators are extracted from the trust state inducing process data to generate state control factors.

7. The method according to claim 1, characterized in that Human factors data collected include: Obtaining the state control factor and the configuration interaction structure diagram, and performing synchronous collection of human factor data; The collected human factor data is cleaned and standardized to generate a set of human factor indicators.

8. The method according to claim 7, characterized in that The human factor indicator set is subjected to structural mapping and label encoding to generate a structured behavior label sequence.

9. The method according to claim 1, characterized in that: Conducting inference analysis includes: Acquire the structured behavior label sequence and the task structure diagram to construct a trust performance reasoning model; Perform task performance scoring and trust association analysis based on the trust performance reasoning model to generate a regulation control strategy; The regulation control strategy is applied to the configuration interaction structure diagram to update the interaction configuration and complete feedback regulation.

10. A human-computer interaction simulation evaluation system, applied to the human-computer interaction simulation evaluation method according to any one of claims 1 to 9, characterized in that: include: A mission structure diagram building module is used to build a mission structure diagram and generate a roll interaction instruction set based on flight mission data and action control parameters; A deception simulation environment construction module, used to construct a deception simulation environment based on the rolling interaction instruction set and generate a perception stimulus set; An interactive interface configuration module, used to parse the perceptual stimulus set and the task structure diagram, and generate a configuration interactive structure diagram; A trust state modeling module, used for generating a state regulation factor based on the configuration interaction structure diagram and the perception stimulus set; A human factor data collection and processing module, used to generate a structured behavior label sequence according to the state control factor and the configuration interaction structure diagram; The interaction strategy updating module is used to generate a regulation control strategy and update the interaction configuration by combining the structured behavior label sequence with the task structure diagram.

Images