1. Background and significance

Severity of insomnia disorder and research needs: Insomnia is a chronic dysfunction with a high prevalence and serious prevalence, characterized by difficulty falling asleep, impaired sleep maintenance, and decreased sleep quality, accompanied by problems such as daytime cognitive impairment. In recent years, the incidence of insomnia has continued to rise, which has a serious impact on people's physical and mental health, and has become a public health problem that needs to be solved urgently. However, so far, the neuromodulation mechanism of sleep-wake and the pathological mechanism of insomnia have not been fully elucidated, and the existing clinical treatment mainly stays in symptomatic treatment, and there is a lack of precise intervention methods for the etiology. Based on this, the latest scientific research guidelines in China list "the neural mechanisms of sleep and wakefulness and the pathological mechanisms and diagnosis and treatment strategies of dysfunction" as a key direction, with special emphasis on in-depth research on the occurrence mechanism of sleep disorders such as insomnia, and explore new strategies for non-drug intervention. This highlights the urgency and strategic significance of conducting innovative research on insomnia mechanisms and therapies.

Cognitive factors of insomnia and higher-order brain dysfunction: Numerous studies have shown that the occurrence and maintenance of insomnia are closely related to abnormal cognitive-affective processes. Cognitive models point out that patients with chronic insomnia often have excessive mental arousal and cognitive control disorders, such as repeated worrying before bedtime, suppressing distracting thoughts, ruminating thoughts and other adverse cognitive strategies to maintain and aggravate the wakefulness state. It is difficult for patients to relax and "delegate" to the body when they fall asleep, and their autonomic purpose is too strongly involved in the sleep process, resulting in a failure of self-regulation. For example, the "purpose of falling asleep" itself may be counterproductive to trigger anxiety – as the "ambivalent intention" in psychotherapy suggests, the more people fear insomnia and force themselves to fall asleep, and the more they are overly nervous, the harder it is to fall asleep. At the same time, insomnia patients often have perceptual biases about sleep status: overestimating their nighttime awake and underestimating sleep duration, which further reinforces anxiety and loss of control. These findings suggest that insomnia is not only a simple neuroexcitatory problem, but also involves higher-order cognitive dysfunction, including abnormalities in self-awareness, purpose generation, emotion regulation, and perceptual feedback. Therefore, it is of great significance to deeply analyze the mechanism of insomnia from the perspective of cognitive neuroscience and clarify which cognitive processes (such as false beliefs, over-monitoring, emotional interventions, etc.) play a key role in it, so as to innovate treatment strategies.

The DIKWP model and its potential in sleep cognitive regulation: In order to integrate multi-level cognitive processes, Professor Yucong Duan originally proposed the DIKWP reticular cognitive model。 DIKWP stands for Data, Information, Knowledge, Wisdom and Purpose, and extends the traditional linear "Data-Information-Knowledge-Wisdom" pyramid model into a fully connected network structure with a "Purpose/Purpose" layer. In this model, the five elements are interconnected with each other, and there are 25 bidirectional interactions and feedback loops. Different from the traditional bottom-up, one-way information processing chain, the DIKWP model allows two-way flow and dynamic adjustment between upper and lower cognitive processes: on the one hand, low-level data can be extracted into information and knowledge and integrated step by step to form Wisdom; On the other hand, high-level wisdom and purpose can also downward affect knowledge acquisition and information selection, thus forming a closed cognitive feedback loop. This mesh topology provides a full-link perspective from the perception input to the Purpose output, enabling iterative updates and bidirectional interaction between the layers of semantic units. The introduction of the DIKWP model into the study of sleep-wake regulation is expected to carry out cognitive modeling of complex multi-pathway regulatory mechanisms: for example, physiological signals and environmental cues are regarded as "data/information" layer inputs, memory and experience are regarded as "knowledge" layers, individual's understanding and strategies of sleep are elevated to the "Wisdom" layer, and finally sleep/wake decisions and motivation are attributed to the "Purpose" layer drive. Under such a unified framework, the comprehensive influence and feedback regulation of multi-dimensional factors such as circadian rhythm, environmental factors, and cognitive activities on the sleep-wake cycle can be simulated, providing a new tool for studying the maintenance and imbalance of sleep homeostasis.

Artificial Consciousness Theory Combined with DIKWP's Innovative Significance: Artificial Consciousness research aims to allow machines to simulate or have human consciousness characteristics, including high-level functions such as self-awareness, purpose-driven, and subjective experience. Professor Yucong Duan's team applied the DIKWP model to the construction of artificial consciousness system, and proposed the "5×5 DIKWP" consciousness representation and evaluation methodto map the level of consciousness as a combination of states in the DIKWP semantic space. In particular, the team's patent proposes a "dual circulation" (DIKWP×DIKWP) artificial consciousness architecture: embedding a set of metacognitive cycles in addition to the basic cognitive processes to achieve self-monitoring, self-reflection and self-regulation of cognitive processes. To put it simply, one DIKWP cycle assumes the first-order cognitive function (perceiving information and generating purpose), while the second DIKWP cycle monitors and regulates the former as metacognition, which is equivalent to simulating a preliminary self-awareness. This architecture breaks through the limitations of traditional AI without self-regulation, and is regarded as an important way to move towards autonomous consciousness. Applying this artificial consciousness theory to the study of insomnia mechanism can deepen our understanding of patients' internal cognitive interactions: insomnia patients often fall into a vicious circle of over-involvement of subjective consciousness in physiological processes, which is essentially a state of imbalance of "dual cycle of consciousness", in which the awakened consciousness layer over-monitors and interferes with the sleep process that should operate autonomously。 Through the DIKWP × DIKWP model, the conflict between the patient's conscious and subconscious sleep regulation layers can be simulated: for example, the meta-purpose layer is overly focused on sleep conditions and causes anxiety, which interferes with the underlying sleep-driven information flow. This modelling helps to elucidate the mechanisms by which self-regulation is out of control in insomnia. Professor Yucong Duan's initial exploration has confirmed this idea: the idea of "you have to wake up early tomorrow morning" often leads to insomnia that night, which can be seen as the interaction between purpose and emotion under the framework of DIKWP - strong wake-up purpose and anticipation trigger anxiety and other emotions, which are continuously strengthened by comparison with memory knowledge, and finally maintain the body in a state of high alertness and difficulty falling asleep. This suggests that the disturbance of the purpose-emotional-cognitive feedback loop is one of the key causes of insomnia, and it is also where artificial consciousness theory can play a role.

Theory and application value: In summary, this project introduces the DIKWP cognitive model and artificial consciousness framework into the study of sleep-wake regulation and insomnia, and is expected to achieve the following innovative understanding and application breakthroughs: (1) comprehensively reveal the neurocognitive mechanism of sleep-wake from the perspective of multi-level semantic interaction, and provide a unified model support for how complex brain networks produce and regulate sleep; (2) to elucidate the abnormal patterns of insomnia patients in information processing, cognitive evaluation, and **purpose decision-making, and to reveal new cognitive markers and potential intervention targets in their pathological mechanisms; (**3) develop adaptive digital therapies on the basis of the above theories, and use artificial consciousness systems to simulate human self-regulation functions, so as to provide a personalized and intelligent non-drug intervention method for insomnia; (4) Promote the cross-integration of sleep medicine, cognitive science, and artificial intelligence technology, and build a prototype of a digital therapy platform that can be extended to other cognitive dysfunctions, which will bring an exemplary role in promoting the clinical diagnosis and treatment of sleep disorders and the development of the digital health industry in China.

2. Research objectives and overall technical roadmap

Overall research objectives: This project proposes a series of organically connected research objectives around the cognitive mechanism of sleep-wake regulation and insomnia disorder intervention. These include:

1. Construct a sleep-wake cognitive control mechanism model based on DIKWP model: By mapping sleep-related physiological signals and cognitive processes to the DIKWP semantic level, a cognitive framework that can simulate the interaction and regulation of multiple pathways in the sleep-wake cycle is established. The model will reveal the flow of information from sensory inputs (e.g., circadian rhythm signals, sleep drives) to high-level purposes (e.g., staying awake or falling asleep), and characterize the two-way feedback between levels.

2. Elucidate insomnia-related cognitive abnormalities and mechanisms: The constructed model was used to analyze the specific abnormal patterns of cognitive processes in insomnia patients, including perceptual imbalance (e.g., abnormal perception of fatigue and sleepiness signals), purpose disorder (e.g., excessive desire to fall asleep or worry), and information blocking (e.g., excessive nighttime thoughts due to insufficient processing of emotions/memories during the day). To identify how these abnormalities disrupt the normal balance of sleep regulation, and to reveal the mechanism of insomnia from a cognitive perspective.

3. Development of Insomnia Adaptive Digital Therapy Tools Based on Artificial Consciousness Architecture: Based on the DIKWP×DIKWP dual-cycle architecture, an artificial consciousness-driven cognitive purpose modulation module was designed to monitor and guide the sleep-related cognitive status of insomnia individuals in real time. The digital therapeutics system is able to autonomously adjust intervention strategies based on patients' physiological feedback and subjective reports, and achieve dynamic guidance and balance on patients' arousal levels, cognitive content, and emotional responses.

4. Design a closed-loop control path fused with EEG feedback: Integrate objective brain signal monitoring such as EEG into the digital therapy system to form a closed-loop control link of "monitoring-analysis-feedback". The patient's wakefulness level and sleep stage are evaluated in real time through EEG and physiological data, which are converted into data/information input in the DIKWP model, and the corresponding adjustment instructions (such as relaxation training, cognitive reconstruction prompts, etc.) are output after analysis by the artificial consciousness module, and the feedback is applied to the patient, so as to realize the two-way regulation of physiology and cognition.

Overall technical route: This project adopts the technical route of "model construction→ mechanism analysis→ system development →clinical validation", spanning two levels: basic research and application development. Firstly, the knowledge of cognitive science and sleep medicine was integrated at the theoretical level, and the correspondence between the DIKWP semantic network and the sleep-wake system was established, and the sleep cognitive control model was formed。 Then, through literature research, experimental data analysis and model simulation, the unique cognitive abnormal patterns of insomnia patients were identified, and the corresponding hypotheses were put forward and verified in the model. Then, according to the model mechanism, the artificial consciousness digital intervention system was developed: the semantic framework of the artificial consciousness operating system developed by the team was used to embed the DIKWP model into the software platform, and the metacognitive adjustment module and brain feedback interface were integrated to realize the intelligent guidance of user cognition-behavior. Finally, the system was tested and optimized in a clinical setting: a sample of insomnia patients was recruited, subjective sleep assessments (such as insomnia severity index) and objective physiological indicators (such as EEG band power) were collected through a period of intervention and application, the efficacy and safety of digital therapeutics were verified, and the model and algorithm were iteratively improved based on the results. In the whole process, we pay attention to multidisciplinary and closed-loop verification: not only cognitive models guide intervention design, but also feed back model modification through biological signals and clinical feedback, so as to ensure that the research output is both scientific and practical, and form a complete innovation chain from mechanism discovery to technical products.

3. Key technologies and research contents

This project intends to focus on the following key technical issues:

• (1) DIKWP semantic modeling and dynamic regulation mechanism of sleep-wake multidimensional interaction: The physiological, environmental and psychological factors related to sleep regulation were mapped to the five semantic levels of the DIKWP model, and a network interaction model was constructed to simulate the cognitive control process of sleep-wake transition. The research contents include: improving the connotation of semantic units at each layer (e.g., the data layer contains sensory input and physiological signals, the information layer contains sensory integration and primary perception, the knowledge layer contains memory and experience, the Wisdom layer corresponds to cognitive evaluation and strategy, and the Purpose layer represents action decision-making and motivation), as well as the two-way mechanism of interaction between layers (such as how the Purpose layer affects knowledge and information processing through top-down feedback, and how physiological signal changes trigger Purpose at the bottom-up). Fix). This model is used to depict the time sequence of information flow under normal sleep conditions: for example, when night comes, the darkness of the environment and the biological clock signal enter as "data", triggering the perception of drowsiness in the information layer, combined with the sleep habits of the knowledge layer and the degree of brain fatigue The Wisdom layer makes a global assessment of the need for sleep, and finally the Purpose layer generates and executes the Purpose of Sleep. The process of waking up in the morning is reversed. Furthermore, mathematical simulations or computational models (such as discrete time state machines or differential equations) are introduced to dynamically simulate the cognitive network, and the response of the system to different inputs (such as illumination and pressure) and the stability of multiple internal feedback loops are analyzed. The results of this part will reveal the network topological characteristics and key links of sleep-wake regulation, and provide a benchmark model for the subsequent analysis of insomnia mechanism.

• (2) Research and development of cognitive purpose modulation module based on artificial consciousness system ("consciousness arousal-information integration-feedback rebalancing" path control): design and implement an artificial consciousness kernel implanted in the digital therapy system, giving it the ability to mimic human self-regulation and is used to correct abnormal cognitive-purpose states in insomnia patients. Specifically, the study will focus on the definition of a "conscious arousal" indicator, which assesses the patient's current level of arousal and cognitive load through the analysis of real-time data (e.g., brain excitability based on EEG β/theta wave ratio, relaxation based on breathing heart rate). When the degree of arousal is abnormally high, the artificial consciousness module is "awakened" into active intervention mode. Secondly, the function of "information integration" was realized, which integrated patients' subjective reports (such as drowsiness and anxious thoughts) and objective signals, and used the DIKWP model to interpret and understand the internal state of patients. For example, if it detects that the patient's high-frequency EEG activity is still significant after closing his eyes, accompanied by subjective feedback that "the brain can't stop", the system can recognize that the patient is stuck in the cognitive loop of "worrying and not being able to sleep" at the knowledge/wisdom level. Then, the "feedback rebalancing" control is carried out - the artificial consciousness module generates a corresponding adjustment plan at the Purpose layer according to the above situational evaluation, and feeds back to the patient through the digital therapy interface to achieve the correction of his physical and mental state. This feedback may include instructing the patient to transfer consciousness (e.g., using the "ambivalent intention" technique to advise them not to force them to fall asleep to relieve tension), guiding relaxation training or mindfulness meditation to reduce arousal levels, providing cognitive restructuring suggestions to correct exaggerated perceptions of insomnia, and adjusting environmental parameters (lighting, music) to create a more sleep-friendly atmosphere. The core of this module is to use DIKWP× the metacognitive cycle in the DIKWP architecture to achieve self-monitoring and self-adjustment: on the one hand, the system monitors the changes in the user's physiological and psychological state, and on the other hand, continuously evaluates the intervention effect and updates its own strategy, and finally forms a closed-loop control to make the patient's wakefulness-sleep state tend to be balanced.

• (3) Identification and digital modeling of cognitive biomarkers of insomnia: combining cognitive models and experimental data, the objective and subjective indicators that can characterize insomnia can be mined, so as to provide a basis for personalized diagnosis and intervention. The study first focused on the characteristics of EEG: by collecting polysomnography EEG data from insomnia patients and healthy controls with cooperative sleep centers, conducting spectrum analysis and power spectrum comparison, verifying whether the typical abnormalities reported in the literature, such as increased power in the high-frequency β band and decreased slow-wave activity in the NREM phase, were significantly present, and the correlation between them and the severity of insomnia was quantified. In addition, dynamic brain network indicators, such as changes in functional connectivity in different brain regions during sleep, were explored as potential markers. Secondly, attention was paid to cognitive and behavioral indicators: for example, sleep diaries and questionnaires were used to evaluate the perceptual-objective sleep disperception (Sleep State Misperception), bedtime anxiety level, daytime fatigue cognitive function, etc., and the indicators that were significantly abnormal in the insomnia group were screened out. Then, these multi-dimensional markers were mapped to the DIKWP model for digital depiction, such as treating high-frequency EEG activity as too much "data" noise, discognizing perception and reality as distortion of information layer processing, and treating excessive attention to the consequences of insomnia as an unreasonable trade-off of the Wisdom layer, so as to enrich the model's ability to describe insomnia. Finally, an individualized evaluation algorithm was established based on the identified markers, and the degree of insomnia-related cognitive deviation could be quantified by inputting the individual's EEG and behavioral data, which could be used for hierarchical intervention decision-making of digital therapeutics. For example, when the algorithm detects that a user still has significant β waves higher than the baseline threshold after falling asleep, and the subjective report anxiety score is high, it will determine that the user is in the "high arousal-high anxiety" subtype, and the system will trigger more intense relaxation and emotional counseling interventions for this pattern. This research content will provide a quantitative basis for decision-making of digital therapeutics, and realize accurate classification and dynamic monitoring of different patients.

• (4) Development of digital system for insomnia intervention and design of dual regulatory mechanism of behavior and cognition: On the basis of the above theories and data, a software system of "digital therapy for insomnia" was developedto achieve simultaneous intervention on patient behavior and cognition. The design concept of the system is to integrate the effective elements of traditional cognitive behavioral therapy for insomnia (CBT-I) into an AI-driven platform to make interventions more efficient, personalized and accessible. The specific R&D content includes: (1) Behavioral intervention module: digital realization of classic behavioral strategies such as sleep hygiene education, stimulus control therapy and sleep restriction therapy. For example, the system can automatically suggest adjustments to the user's sleep diary (such as postponing bedtime to increase sleep drive), and help them strictly follow it through schedule management and reminder functions. Use the phone's location and time information to correct users when they detect a violation of sleep hygiene principles (e.g., being active in a bright environment late at night). (2) Cognitive intervention module: embed interactive cognitive reconstruction and relaxation training tools. The system guides users to record their worries before going to bed through the dialogue interface, and uses the built-in cognitive reevaluation algorithm to give feedback to help users look at sleep problems from a more rational and peaceful perspective. At the same time, it provides a variety of relaxation methods such as progressive muscle relaxation, breathing guidance, meditation audio, etc., and activates the aforementioned artificial awareness module when necessary to implement more intelligent guidance (such as automatically inserting a mindfulness practice when the user is overly worried). (3) Personalized self-adaptation: With the help of machine learning methods, the feedback and intervention effect data of users are continuously learned, and the intensity and combination of the above-mentioned behavioral and cognitive interventions are adjusted. For example, if a user responds well to relaxation training but has difficulty implementing sleep restrictions, the system tends to add relaxation sessions and implement sleep restrictions gradually. With this adaptive mechanism, user compliance and efficacy are maximized. It is worth emphasizing that this system is different from the mobile app of the general fixed program, and is more similar to a continuously evolving intelligent therapist: with the support of the artificial consciousness framework, it can perceive**, make decisions and intervene in real time** to provide treatment options that vary from person to person. Previous studies and clinical trials have shown that the efficacy of pure online digital CBT-I in improving insomnia symptoms is no less than face-to-face treatment. The system of this project will take this foundation and explore new ways to improve efficacy by introducing physiological feedback and consciousness control algorithms such as EEG. For example, it can monitor the user's night EEG in real time, and automatically push intervention content when it is found that it has been awake for a long time, shortening the latency period of falling asleep; Another example is the combination of wearable devices to monitor daytime activities and emotional states, providing users with round-the-clock sleep health management suggestions. The successful development of this digital system will provide a standardized and scalable digital therapeutics for insomnia.

• **(5) Digital therapeutics platform construction and clinical sample evaluation: After the completion of the research and development of core modules, an integrated digital therapeutics platform for insomnia will be built and clinical trial evaluation will be carried out. The platform will include user terminals (mobile apps or wearable device interfaces) and cloud services (human-aware decision engines, data storage and security management). First of all, it is technically necessary to ensure the stability of the system and the security of data privacy. Then, small-scale user testing is used to refine the interactive experience and personalized recommendation algorithms. On this basis, at least 100 patients with insomnia were recruited for intervention studies in cooperation with clinical medical institutions. Digital therapeutics were applied to participants using randomized controlled trials or controlled before-and-after designs, followed for weeks to months. In terms of objective indicators, polysomnography (**PSG) or wearable devices were used to record the changes in sleep structure before and after the intervention, including sleep latency, total sleep time, sleep efficiency, and the number of awakenings. For subjective measures, the Insomnia Severity Index (ISI) and the Pittsburgh Sleep Quality Index (PSQI) were used Isoscales assess symptom improvement. Patient compliance and satisfaction feedback were also recorded. Based on these data, the clinical effects of this digital therapeutics were evaluated, such as the average reduction in the insomnia severity index, the percentage points for the improvement in sleep efficiency, and whether a significant proportion of patients met the criteria for no clinical insomnia. Statistical analysis will verify the significance of this system in improving sleep compared to baseline or control. In addition, exploratory analyses were used to examine differences in efficacy between different subgroups (e.g., those with high anxiety vs. those with low anxiety) to identify room for model improvement. The final output of the project will include: a sleep-wake cognitive model tool, an artificial consciousness feedback module, a digital therapeutics software prototype, and a 100-case clinical application data and analysis report. These phased results not only verify the research hypothesis, but also lay the foundation for the generalization of the results.

4. Feasibility analysis

Research foundation and team advantages: This project is led by Professor Yucong Duan of Hainan University, whose team has profound research accumulation and innovation achievements in the field of artificial intelligence and brain cognition. As the initiator of the DIKWP artificial consciousness model, Professor Yucong Duan has been highly recognized at home and abroad, and has been authorized 114 invention patents (including 15 PCT international patents), covering many cutting-edge directions from large model training, artificial consciousness construction, cognitive operating system to AI governance. These patented technologies constitute a complete DIKWP cognitive technology system, which provides a solid reserve of independent intellectual property rights and methods for this project. The team's research on the basic theory of cognitive computing and artificial consciousness is a milestone in the introduction of the "Purpose" layer in the traditional DIKW model and the realization of full-link semantic feedback, proposing an interpretable and controllable cognitive architecture, which provides an innovative path for transparent decision-making and autonomy of AI systems. This original theory provides a direct reference and reusable framework for the construction of a sleep cognitive model and an artificial consciousness regulation system in this project.

Previous related work: In recent years, Prof. Yucong Duan's team has begun to integrate the DIKWP model with the biomedical field to explore new models of active medicine/digital therapeutics, and has achieved preliminary results in multiple application scenarios. For example, in an artificial consciousness system that simulates doctor-patient interaction, the team developed a DIKWP semantic-driven state machine model to realize the whole-process simulation and feedback control of cognitive decision-making in the diagnosis and treatment process. Based on the principle of the DIKWP model, the team successfully developed a prototype of the Artificial Consciousness Operating System (ACOS), which embeds the five-layer cognitive process in a semantic form into the computing framework, which can orchestrate and manage the cognitive process of complex tasks. This OS is equivalent to the "semantic operating system" of AI, similar to the traditional OS that manages hardware resources and process scheduling, while the DIKWP semantic OS manages knowledge resources and cognitive process scheduling, so that each module within AI can operate in an orderly manner at the semantic level. This capability is critical to the development of a digital therapeutics platform for this project – we will leverage the existing ACOS architecture to efficiently integrate the various components of the project (cognitive models, feedback modules, etc.) and reduce the barriers from theory to software implementation. In addition, the team also has accumulated experience in the direction of brain signal decoding and brain-computer interface: it has carried out research on pattern recognition and semantic interpretation of cranial nerve signals, and explored methods for mapping "energy field" signals such as EEG/EEG to the cognitive semantic layer. The information field-energy field coupling technology proposed by the team can correlate the internal cognitive state of the AI system with the external physiological signals, and realize cross-modal information fusion and regulation. This technology provides a feasible solution for the fusion of EEG feedback in this project, and our existing methods and algorithms can convert EEG features into input information of the DIKWP model, and adjust the feedback stimuli according to the results of the model. In short, the project team's pioneering work in artificial consciousness model, cognitive intervention algorithm and brain signal processing has laid a solid foundation and feasibility guarantee for the smooth implementation of this project.

Technical conditions and platform support: The relying unit has built relevant scientific research platforms, such as "Artificial Consciousness and Cognitive Computing Laboratory" and "Brain-Computer Interface and Neuromodulation Technology Research Center", etc., which have the software and hardware conditions required for this project. The team has high-performance computing servers for complex model simulation and machine learning training, and EEG acquisition and physiological sensing equipment for human-machine experiments. The previously developed artificial consciousness prototype system and cognitive operating system software will serve as an important development foundation, and it is expected that the semantic engine and scheduling module can be directly reused to accelerate the integrated research and development of digital therapeutics platforms. In terms of clinical research, we have established cooperative relationships with sleep medicine centers in many hospitals, which can obtain sufficient insomnia patient resources and expert guidance to ensure the smooth development of sample recruitment and intervention trials. In terms of management, the project team is composed of an experienced interdisciplinary team, including key personnel in the fields of artificial intelligence, cognitive neuroscience, sleep medicine and clinical psychology. Relying on the previous research results and existing conditions, the project has high feasibility in scientific exploration and technology realization.

5. Phased results and assessment indicators

The project plans to produce the following outcomes in phases and evaluate them with quantitative indicators:

• Phase 1: Sleep-Wake Cognitive Regulation Model – Complete the modeling of the sleep-wake cycle by the DIKWP semantic network, and realize a theoretical model with five layers of elements and two-way interactive feedback. Assessment indicators: submit the model framework design report and simulation analysis results, clarify the normal sleep information flow path and key nodes, and form 1 paper or report.

• Phase 2: Mechanisms of Insomnia and Cognitive Markers – Elucidate the pathways of cognitive abnormalities associated with insomnia and identify objective markers. Assessment indicators: At least 3 cognitive**/physiological markers of insomnia** (such as EEG power in specific frequency bands, quantitative indicators of subjective-objective sleep difference, etc.) were extracted, and their mechanism of action was explained in the model. Wrote 1 research paper on the cognitive mechanism of insomnia, which required sufficient data and credible conclusions.

• Phase 3: Artificial Consciousness Feedback Module – Develop a metacognitive modulation module based on the DIKWP × DIKWP architecture and validate the function in an experimental environment. Assessment indicators: The module can read simulated or real user data in real time and generate intervention decisions, so as to realize a complete cycle of consciousness awakening and feedback adjustment; Submit module software and test reports demonstrating that the level of arousal or anxiety scores of the participants can be reduced in the experimental situation (a statistically significant difference compared to the non-feedback situation).

• Phase 4: Digital Therapeutics for Insomnia Prototype System – A prototype of a complete digital therapeutics platform that integrates behavioral and cognitive interventions is completed. Assessment indicators: complete software development and internal testing, and launch the main functional modules (sleep monitoring, personalized suggestion push, relaxation training, cognitive reconstruction dialogue, etc.); At least 50 users participate in the trial and feedback on the experience; The software performance is stable and there are no major vulnerabilities. Submit at least 1 software copyright or patent application.

• Phase 5: Clinical Sample Application and Data Analysis – Conduct a digital therapeutics application study in no less than 100 patients with insomnia and collect before-and-after control data. Assessment indicators: complete clinical trial registration and ethical approval; At least 100 patients completed the intervention, of which ≥80% had good compliance; Statistical analysis showed that the main efficacy measures (e.g., ISI insomnia index) improved by 30% from baseline or better than ≥ control group, with clear statistical differences. Form one clinical research report and one academic paper to summarize the therapeutic effect and application boundary.

The results of each of these phases will be assessed as project milestones. In particular, the final clinical data and analysis results of 100 cases are a direct verification of the theoretical and technical effectiveness of the project, which is required to be detailed, reliable and convincing. If all the indicators are met, it is expected that the project will successfully complete the set goals and lay the foundation for subsequent in-depth research and popularization and application.

6. Promote your app

The results of this project have broad application prospects and social and economic value, and are intended to be promoted and transformed through the following ways:

• Clinical Rehabilitation: In collaboration with the hospital sleep center and the Department of Psychology, digital sleep therapy based on the DIKWP model was incorporated into the routine diagnosis and treatment of patients with insomnia. Through continuing education and training, clinicians and therapists are instructed to use this system to assist in the diagnosis and treatment of insomnia, improve efficacy and reduce dependence on medications. At the end of the project, it is planned to build a demonstration application base to provide digital therapy services to at least 1,000 insomnia patients, collect long-term follow-up data, and further verify the efficacy and improve the protocol.

• Mental health services: Promote this digital therapeutics to primary mental health services and community health areas, such as as a tool for counselors and sleep coaches to improve their ability to manage patients with insomnia and anxiety. Develop a public version of the sleep health app, which provides popular science, self-test and basic intervention functions to help subclinical insomnia improve sleep habits and prevent the occurrence of chronic insomnia.

• Smart Home & Wearable Market: Combining IoT and smart hardware vendors to embed research results into smart sleep environments. For example, we cooperated with smart mattress and sleep monitor manufacturers to integrate the project's sleep status recognition algorithm and personalized intervention strategy to realize the automatic optimization of the bedroom environment (lighting, temperature, aromatherapy, etc.) and timely guidance for users to create the Wisdom bedroom solution. In conjunction with wearable devices (smart bracelets, EEG sleep headbands, etc.), the model algorithm of this project is used to improve the accuracy of the device's monitoring of sleep stages and quality, and provide real-time intervention suggestions according to the detection results to enhance the added value of wearable devices.

• Industry incubation and standard formulation: Seek opportunities for industrial cooperation in the later stage of the project, promote the productization and registration approval of the digital therapy platform, and strive to become the first batch of insomnia digital therapy products in China. Combining the team's advantages in patents and standards, we will promote the inclusion of the DIKWP model in industry standards and guidelines in the field of digital health (such as digital therapeutics design specifications, AI medical product standards), and enhance China's voice in this emerging field. Widely publicize the results of the project through academic conferences, industrial exhibitions and other channels, and cooperate with relevant enterprises to accelerate technology transfer and application.

• Extended Application Development: Further expand the application of DIKWP× DIKWP model and artificial consciousness architecture to other chronic brain diseases, such as digital interventions for anxiety and depression, and rehabilitation training for cognitive impairment. Through the successful demonstration in the field of insomnia, we will build a set of general digital cognitive therapy development platform to support the research and development of personalized intelligent treatment plans for a variety of diseases, and form a series of products and services to benefit more patients.

To sum up, this project closely follows the key direction of national sleep disorder research, integrates the original DIKWP× DIKWP cognitive model and artificial consciousness technology, and innovatively proposes a new digital therapy scheme for insomnia disorders. The research content of the project is complete, the technical route is clear, and it has a solid preliminary foundation and feasibility. The expected results will deepen the understanding of sleep-wake and insomnia mechanisms, fill the theoretical gap from the perspective of cognitive regulation, and directly produce an applicable digital intervention system, which will contribute to improving the level of insomnia diagnosis and treatment and promoting the development of the digital health industry. We will be in line with a rigorous and realistic scientific attitude and pioneering and innovative spirit, make every effort to implement various research tasks, and strive to achieve high-quality scientific research and application results.

References:

1. Ancoli-Israel, S., & Roth, T. (1999). Characteristics of insomnia in the United States: results of the 1991 National Sleep Foundation Survey. Sleep, 22(Suppl 2), S347-S353.

2. Schmidt, R. E., et al. (2011). Cognitive and affective control in insomnia. Frontiers in Psychology, 2, 349.

3. Frankl, V. E. Paradoxical intention. (Wikipedia citation).

4. National Natural Science Foundation of China. (2023). Guidelines for the project "Research on the Mechanism and Intervention Strategies of Sleep and Sleep Disorders".

5. Yucong Duan . (2023). Exploring the DIKWP Model of Artificial Consciousness: The Biological Basis and Information Processing Mechanisms of Consciousness, Subconscious and Emotions (preprint)

6. Yucong Duan et al. (2025). "Integrating Consciousness Relativity Theory and Consciousness BUG Theory Based on Mesh DIKWP Model" (Technical Report)

7. Feng News. (2025). Prof. Yucong Duan: The DIKWP artificial consciousness model leads the future of AI, and 114 patents are expected to be implemented in the industry

8. Lin Wenyao et al. (2023). Efficacy of digital cognitive behavioral therapy for insomnia and depression: a systematic review and meta-analysis of randomized controlled trials. PeerJ, 11:e16137.

9. Huang Zhili et al. (2020). Guidelines for the diagnosis and treatment of insomnia in Chinese adults. Chin J Neurol, 53(5), 324-335

10. Lichstein, K. L., & Rosenthal, T. L. (1980). Insomniacs’ perceptions of cognitive versus somatic determinants of sleep disturbance. Journal of Abnormal Psychology, 89(1), 105-107.