Neurophysiological Basis and Mechanisms of the Reticular DIKWP

Neurophysiological Basis and Mechanisms of the Reticular DIKWPModel

Yucong Duan

Benefactor: Zhendong Guo

International Standardization Committee of Networked DIKWP for Artificial Intelligence Evaluation(DIKWP-SC)

World Artificial Consciousness CIC(WAC)

World Conference on Artificial Consciousness(WCAC)

(Email: duanyucong@hotmail.com)

Abstract

This paper provides a comprehensive analysis of the neurophysiological basis of the brain from the perspective of the Data-Information-Knowledge-Wisdom-Purpose (DIKWP) model. Without directly adopting traditional neuroscientific terminology for modeling, we use the DIKWP model and its semantic mathematics as the sole semantic space framework to inversely map and reinterpret current research findings in neuroscience, brain science, and even gene expression. First, we elaborate on the extensions and differences of the DIKWP model compared to the traditional Data-Information-Knowledge-Wisdom (DIKW) hierarchical model: DIKWP introduces the "Purpose" dimension and replaces the simple hierarchical structure with a reticular interconnected structure, enabling dynamic interactions among the five elements in the semantic space (DIKWP Semantic Mathematics: Embracing a Networked Model – Research Notes). Next, we map the five elements of the DIKWP model to key brain structures one by one, constructing a semantic mapping network from sensory input to cognitive decision-making: Data (D) corresponds to sensory cortices and perceptual pathways, Information (I) to the limbic system and primary prefrontal regions for information processing, Knowledge (K) to the hippocampus and default mode network (DMN) for long-term memory integration, Wisdom (W) to the parietal lobe and advanced prefrontal regions for multimodal fusion, and Purpose (P) to the medial prefrontal cortex, orbitofrontal cortex, and anterior cingulate cortex for motivation and intentional control.

We further explore the potential neurophysiological correspondences of the 25 interaction mappings formed by pairwise combinations of the five DIKWP elements, including bottom-up perception-cognition transmission (e.g., D→I, I→K), top-down intentional feedback regulation (e.g., P→W, W→K), and bidirectional parallel interactions (e.g., I↔K, W↔P). Based on this, we provide dialectical interpretations and model-based explanations of these mappings according to the intrinsic evolutionary rules of the DIKWP semantic space, integrating empirical observations (e.g., synaptic plasticity, brain network dynamics) and hypothetical mechanisms (e.g., predictive coding) to illustrate how information flows along the DIKWP network in the brain.

Finally, we creatively extend the model by proposing hypotheses such as the "semantic field" and "intention-driven pathways" to reinterpret cognitive processes like language comprehension (e.g., I→K→W) and behavior initiation (e.g., W→P→D) using the DIKWP network. We also explore the implications of the DIKWP framework for artificial consciousness research, proposing hypotheses such as the "semantic emergence threshold" and "subjective BUG chains," and discuss potential neurophysiological analogies for computable purpose-generation systems. Through this research, we aim to demonstrate that the DIKWP network model, as a unified semantic space framework, can effectively map and explain complex brain cognitive processes, providing new insights for understanding biological brains and constructing brain-like intelligent systems.

Introduction

From cognitive science to artificial intelligence, the classic DIKW model (Data-Information-Knowledge-Wisdom hierarchy) is often used to describe the stepwise transformation of information from raw data to wise decision-making. However, this hierarchical DIKW pyramid has significant limitations in explaining the complex dynamic interactions of human cognition (DIKWP Semantic Mathematics: Embracing a Networked Model – Research Notes). Real-world brain processing is not a simple unidirectional flow but is filled with feedback loops and parallel interactions: high-level intentions influence perception, and emotional memories alter the interpretation of information. This complexity drives the search for new models to characterize these processes.

The DIKWP model addresses this need. Building on the DIKW framework, it adds a "Purpose" layer, forming a five-element system of Data (D)-Information (I)-Knowledge (K)-Wisdom (W)-Purpose (P), and transforms the relationships among elements from a linear hierarchy to a reticular network (DIKWP Semantic Mathematics: Embracing a Networked Model – Research Notes). This extension reflects the fact that cognitive processes are not linear but involve multidirectional coupling among levels. The "semantic mathematics" underlying the DIKWP model provides a formalized tool for representation, allowing each element and its transformations to be expressed mathematically, thereby strictly defining the rules of information flow within the semantic space.

In this paper, we use the DIKWP model as the sole semantic space framework to reinterpret and examine brain neural activity. Our approach is "inverse semantic mapping": rather than constructing a model directly using traditional brain region functional terminology, we first ground it in the DIKWP model's implications and then refract existing neuroscientific discoveries into this semantic space for reinterpretation. This method helps avoid the limitations of previous models that over-relied on empirical terminology, ensuring explanations are based on the self-consistent semantic evolutionary rules of the DIKWP system.

Specifically, we first briefly describe the semantic framework of the DIKWP model and its networked characteristics, then delve into how brain anatomy and function can be mapped to the five elements of DIKWP. We select typical structures such as sensory cortices, the limbic system, the hippocampus-default mode network, the parietal-prefrontal network, and the anterior cingulate-orbitofrontal cortex to correspond to D, I, K, W, and P of DIKWP, respectively, and explain the rationale behind these mappings. For example, sensory cortices process raw environmental signals, which can be viewed as the implementation of the Data layer in DIKWP, while the anterior cingulate cortex, involved in motivational decision-making, can be seen as the neural counterpart of the Purpose layer.

Next, we explore the neural mechanisms underlying the interactions among the five DIKWP elements. The DIKWP model allows for 25 (5×5) possible interaction combinations, including bidirectional effects in any direction. Do these interaction mappings have counterparts in the biological brain? We attempt to explain this using evidence from neural circuits and network dynamics. For instance, how sensory input is transformed into knowledge and modulated by higher-level wisdom and purpose; how information at different levels is processed in parallel and integrated for decision-making. In this section, we reference mainstream neuroscientific terminology (e.g., synaptic plasticity, predictive coding, default mode network DMN, integrated information theory IIT) to aid explanation. However, it is important to emphasize that these terms serve only as references, and their meanings are refracted within the framework of the DIKWP model, rather than serving as the starting point of our model. The validity of all mappings is judged by the evolutionary rules of the DIKWP semantic space itself.

Finally, in the discussion, we engage in creative model extrapolation and prediction. Leveraging the unique perspective of the DIKWP model, we propose the concept of a "semantic field" to describe the holographic information space composed of D, I, K, W, and P, and "intention-driven pathways" to characterize the guiding role of purpose in cognitive flow. Based on this, we remodel typical cognitive functional pathways, such as interpreting language comprehension as a "I→K→W" DIKWP semantic cascade and action initiation as a "W→P→D" reverse chain. Furthermore, we discuss the implications of these ideas for artificial consciousness, proposing potential physiological analogies and new hypotheses, such as the "semantic emergence threshold hypothesis" (subjective experience may arise when the complexity of DIKWP interactions within a system exceeds a threshold) and the "subjective BUG chain" (minor semantic deviations in a cognitive system amplify across levels, leading to a chain reaction of subjective errors). These predictive perspectives may provide reference models for future research on computable purpose generation and brain-like subjective experiences in artificial intelligence.

In summary, this paper aims to demonstrate that the reticular semantic space of the DIKWP model can provide coherent and innovative explanations for the anatomical structures and cognitive functions of the brain. Such cross-domain mapping not only enriches our understanding of neurophysiological mechanisms but also offers a unique theoretical foundation for the design of next-generation brain-like intelligent systems.

Overview of the DIKWP Model Semantic Framework

The DIKWP model is an extension and reconstruction of the traditional DIKW model, with its core innovation being the introduction of the highest-level "Purpose (P)" element and the transformation of inter-element relationships from a linear hierarchy to a networked structure (DIKWP Semantic Mathematics: Embracing a Networked Model – Research Notes). Under the DIKWP framework, the five elements—Data (D), Information (I), Knowledge (K), Wisdom (W), and Purpose (P)—are not merely hierarchical but exist as interconnected semantic units within a dynamic network. This means any element can directly interact with and influence any other, unconstrained by a fixed sequence. For example, knowledge can be accumulated from information but can also, in turn, influence new information processing; purpose is not just the final output but can proactively intervene in data collection and interpretation. This flexibility addresses the limitations of traditional hierarchical models, enabling the DIKWP model to capture the multidirectional, parallel phenomena observed in real-world cognition (DIKWP Semantic Mathematics: Embracing a Networked Model – Research Notes).

Formally, the DIKWP model can be conceptualized as a directed graph network with transformation functions. Each node represents one of the five semantic components (D, I, K, W, P), and bidirectional transformation functions  T XY  may exist between any pair of nodes, denoting the semantic conversion from element X to element Y. The entire system forms a global semantic field through the interactions of these five layers. According to related research, the DIKWP information field (semantic field) is a holographic information space jointly constituted by the data, information, knowledge, wisdom, and purpose layers (A Neuroscience Research Framework for Consciousness Based on DIKWP Information and Energy Fields - Zhihu Column). In other words, the semantic field encompasses multilevel information collections, from molecular-level gene expression and neural activity to macroscopic biochemical indicators and psychological-cognitive states, weaving together a meaningfully interconnected whole (From the Perspective of the DIKWP Model: Interpreting Duan Yucong’s Active Medicine Theory of Information and Energy Fields). The DIKWP model posits that it is within this semantic field that information from different levels converges and integrates, giving rise to complex cognitive functions.

It is crucial to emphasize that although the DIKWP model borrows concepts from neuroscience and information science, within this framework, the meanings of the elements are strictly defined by the axioms of semantic mathematics. The model operates under a set of evolutionary rules and constraints that ensure the determinacy and consistency of semantic transformations. For example, the conversion of data to information must satisfy specific contextual association rules; knowledge formation requires structural integrity; wisdom-based decisions must account for experiential consistency; and purpose generation must align with motivational teleology. If these rules are likened to "semantic physical laws," then the flow of information in the DIKWP semantic space resembles particles moving within a force field, governed by intrinsic principles. It is precisely this foundation that allows the DIKWP model to provide a unified interpretation of diverse neuroscientific findings without being encumbered by superficial terminological complexities.

In summary, the semantic framework of the DIKWP model offers a macro yet nuanced perspective: macro in that it covers the entire cognitive chain from sensory input to intentional output through five abstract categories, and nuanced in that it permits bidirectional, fine-grained interactions between each category. Next, within this framework, we will attempt to project the key structures and functional phenomena of the human brain onto the components of the DIKWP model, establishing mappings between neural structures and semantic elements.

Semantic Mapping of DIKWP Elements to Brain Structures

In this section, we sequentially map the five components of the DIKWP model (D, I, K, W, P) to typical neural structures or functional networks in the human brain. This mapping is not a simplistic one-to-one correspondence but is grounded in analogies of semantic function: that is, when the function of a brain region or network in cognitive processes aligns closely with the definition of a DIKWP semantic element, we consider that region to implement the corresponding semantic component. It is important to note that this mapping is dialectical and reinterpretive—we cite neuroscientific findings to support the plausibility of the mappings, but the logic of these mappings ultimately rests on the semantic rules of the DIKWP model itself, not on the direct application of neuroscientific concepts.

D (Data): Sensory Signals and the Sensory Cortex

In the DIKWP model, Data (D) represents unprocessed raw facts or observations, the input starting point for the cognitive system. Mapping this to the brain's biological structures, the data layer corresponds to various sensory input pathways and their primary processing areas. For example, visual light signals are captured by retinal photoreceptors, transmitted via the optic nerve to the primary visual cortex (V1); auditory sound waves are converted into neural impulses by cochlear hair cells and relayed to the primary auditory cortex; somatosensory inputs from peripheral receptors are relayed through the spinal cord and thalamus to the primary somatosensory cortex (S1). These primary sensory cortices are where environmental stimuli are first translated into the brain's own "language"—patterns of neural electrical activity (Sensory Cortex | Definition, Location & Function - Lesson - Study.com).

Processing in the sensory cortex is largely low-level and local. For instance, neurons in the primary visual cortex detect basic lines and edges in the visual field, while those in the primary auditory cortex discriminate pure tone frequency components. This stage can be said to correspond to the semantic domain of the data layer in the DIKWP model—merely an objective recording of the environment, devoid of higher-level meaning. As research indicates, the primary sensory cortex is where the brain begins processing sensory information, corresponding to the basic feature encoding of sensory inputs (Sensory Cortex | Definition, Location & Function - Lesson - Study.com). These encodings resemble raw data in computational systems: they form the basis for further information processing but do not themselves contain deep knowledge about the environment.

Beyond the central sensory cortices, the data layer more broadly includes the peripheral sensory nervous system. For example, the retina, cochlea, skin and muscle receptors, and olfactory and taste receptors convert physical signals into neural signals, serving as the biological interfaces for acquiring environmental "data." These peripheral data converge via neural pathways to the central nervous system for cortical processing. The DIKWP model emphasizes that the data layer does not exist in isolation; it can also be modulated by higher layers (e.g., attention can selectively influence sensory input). However, in essence, data corresponds to the information at the perception stage. It answers the question: "What raw stimuli occurred?"

In summary, the neural correlates of Data (D) can be generalized as: peripheral receptors + primary sensory pathways + primary sensory cortices. Together, they perform the function of projecting the external world into neural signals, providing the cognitive system with its most primitive material. Within the DIKWP framework, these brain structures realize the acquisition and representation of data, serving as the foundational "materials" for subsequent I, K, W, and P layers.

I (Information): Preliminary Processing and the Limbic System

Information (I) in the DIKWP model refers to meaningful patterns derived from interpreting and contextualizing data. Compared to data, the information layer adds structure and relational significance. In the brain, information-level processing involves further refinement and interpretation of sensory data, including perceptual integration, pattern recognition, and preliminary association with emotions and context. Neuroanatomically, this level of function is primarily realized by sensory association cortices, parts of the limbic system, and primary regions of the prefrontal cortex.

A classic example is the sensory association cortex, such as the occipitotemporal association cortex, which integrates visual shape, color, and motion to form object recognition, or the parietal association cortex, which integrates information from different sensory modalities to create spatial orientation and body mapping. These association cortices transform fragmented "data" from primary sensory cortices into meaningful "information," such as combining lines into a specific object's shape or a sequence of sounds into recognizable speech. This clearly falls within the semantic domain of Information (I): recognized patterns, objects, and events that carry more meaning than raw sensory data (DIKWP Semantic Mathematics: Embracing a Networked Model – Research Notes).

Another key structure is the limbic system, particularly the amygdala and its associated circuits. The amygdala plays a critical role in rapidly evaluating the emotional significance of sensory information: it can "tag" perceived stimuli (e.g., as dangerous/safe, pleasant/aversive) within milliseconds, thereby transforming raw data into information directly relevant to the organism. This is equivalent to adding a dimension of "value" or "emotion" to the data, turning it into information. For example, a blurry image processed by the amygdala and labeled as a "threat" is no longer mere sensory input but becomes information that triggers bodily responses. The limbic system governs the preliminary processing of emotions, motivations, and memory (Limbic System: What It Is, Function, Parts & Location - Cleveland Clinic); it imbues data with meaning tied to survival and experience. This aligns perfectly with the DIKWP definition of the information layer as "converting data into meaningful patterns through the recognition of differences and context" (DIKWP Semantic Mathematics: Embracing a Networked Model – Research Notes).

Additionally, certain parts of the prefrontal cortex contribute to information-level processing. For instance, the dorsolateral prefrontal cortex (DLPFC) supports working memory, enabling the short-term retention and manipulation of sensory-derived information, allowing the brain to integrate current data with immediate goals. After receiving information from sensory association areas, the DLPFC filters, enhances, or suppresses it based on task demands—another form of contextual interpretation: assigning varying importance to current information based on goals (implicitly from the P layer). This process prepares information for further conversion into knowledge or decision-making triggers. At this stage, the brain has developed a preliminary understanding and judgment of perceived events, such as identifying what an object is, what is currently happening, and what it means for the self.

The neural correlates of Information (I) can be summarized as: sensory association cortices + emotional evaluation systems (e.g., the amygdala) + short-term memory and attentional control systems (e.g., the DLPFC). These structures transform sensory data into information with interpretative significance. From the DIKWP perspective, the brain no longer deals with vague stimuli but with representations of "a specific event occurring" and "what this event means" at a preliminary level.

K (Knowledge): Memory Integration and the Default Mode Network

Knowledge (K) in the DIKWP model refers to structured information capable of long-term storage, formed by integrating information into coherent frameworks or schemas. The content of the knowledge layer includes semantic memory, heuristic rules, mental models of the world, and more. In the brain, the structures most closely associated with the knowledge layer are the hippocampus and the connected default mode network (DMN).

First, the hippocampus and medial temporal lobe system are widely recognized as critical for long-term memory formation. Transient information must undergo consolidation to become enduring knowledge, and the hippocampus bridges perceptual experiences and long-term memory (Transcription Factors in Long-Term Memory and Synaptic Plasticity). When we learn new knowledge or experience new events, the hippocampus recodes sensory and episodic information, establishing connections with existing memories, and ultimately stores them in the neocortex. In this process, "information" ascends to "knowledge": information is endowed with broader context and systems, enabling its integration into long-term experiential frameworks. This is manifested in biological processes like synaptic plasticity: neuroscientific research shows that long-term memory formation relies on gene expression-induced changes in synaptic connections (Gene expression parallels synaptic excitability and plasticity...)—physiologically supporting the idea that transforming information into knowledge requires structural reorganization. In other words, hippocampal-driven memory consolidation turns fragmented information into stable, structured representations (knowledge), closely aligning with the definition of knowledge in the DIKWP model.

More broadly, the default mode network (DMN), to which the hippocampus connects, is deeply associated with the knowledge layer. The DMN is a set of regions highly active during rest and introspection, including the medial prefrontal cortex, posterior cingulate/medial parietal cortex, and medial temporal lobe (including the parahippocampal gyrus). The DMN is implicated in autobiographical memory, mental scene construction, and conceptual integration. When we recall the past or envision the future, engaging in "mental time travel," the DMN is strongly activated. Researchers propose that the DMN integrates memories, language, and semantic representations into internal narratives (Issue: Neuron - Cell Press). For example, when comprehending a story or mind-wandering during rest, the DMN retrieves and weaves together memory fragments and knowledge to create self-relevant scenarios. This can be regarded as the manifestation of the brain's knowledge network in operation: it associates and refines stored knowledge to form higher-level understandings of reality. As one scholar notes: "The default mode network integrates and broadcasts memories, language, and semantic representations, creating coherent 'internal narratives'" (Issue: Neuron - Cell Press). These narratives are essentially expressions of knowledge and its applications, as they transcend current perceptual information, drawing on long-term knowledge reservoirs to enrich present mental experiences.

It is worth noting that the neural realization of knowledge is not limited to DMN activity during rest. In many cognitive tasks, our brains also activate knowledge-related networks to interpret new information. For instance, when reading a passage, we immediately recruit semantic memory (language-semantic regions in the left temporal lobe and related DMN areas) to understand word meanings and connect sentences; when solving problems, we draw on past experiences (interactions between the prefrontal-parietal network and the hippocampus) to seek solutions. These processes all exemplify interactions between new information (I) and existing knowledge (K). The DIKWP model emphasizes bidirectional effects between I→K and K→I: new information can expand or revise knowledge (learning), while knowledge can frame and interpret information (understanding). From a neural mechanism perspective, this corresponds to reciprocal information flow between sensory/association cortices and the hippocampal-cortical memory system. For example, studies show that the hippocampus and DMN are highly functionally coupled during recollection (The Hippocampus Is Coupled with the Default Network during...), indicating that when retrieving knowledge, these knowledge networks interact with sensory and associative cortices, allowing us to "re-experience" relevant information. Conversely, during new memory encoding, the hippocampus's reliance on sensory input is also evident. These back-and-forth communications among brain networks support the formation and retrieval of knowledge.

In summary, the neural correlates of Knowledge (K) include: the hippocampus and medial temporal memory system—responsible for acquiring and consolidating new knowledge, and the default mode network—storing and refining existing knowledge, constructing internal scenarios. Additionally, widespread neocortical regions store specific knowledge content (e.g., language, general knowledge, skills) in the form of semantic networks. Together, they constitute the brain's "knowledge base" and "knowledge processor." Through these structures, the semantic concept of "knowledge" in the DIKWP model gains a tangible neural foundation: as neural connections and activity patterns that carry our long-term memories and cognitive schemas.

W (Wisdom): Multimodal Integration and Advanced Decision-Making

Wisdom (W) in the DIKWP model represents the profound application of knowledge and experience—the ability to make judgments, decisions, and creative integrations in complex environments. The wisdom layer synthesizes knowledge from different domains with the current context, incorporating considerations of value, ethics, and uncertainty handling. In the brain, no single region can be simplistically labeled the "wisdom center," but the essence of wisdom lies in the integration of multifaceted information and balanced decision-making. Neuroscientific research suggests that wisdom may involve the collaboration of evolutionarily older and newer brain structures: it relies on both rational analysis and emotional/social judgment (Neurobiology of Wisdom?: A Literature Overview - PMC). Thus, we map the wisdom layer to a distributed anterior-posterior network system, particularly the highest-level regions of the prefrontal cortex, multimodal association areas like the parietal lobe, and their interactions with the limbic system.

One key aspect is that wise decision-making requires a balance between reason and emotion. From a neural perspective, this means the prefrontal cortex (especially regions responsible for rational reasoning) and the limbic system (responsible for emotions and motivations) achieve optimal collaboration. A literature review posits that "wisdom involves the optimal balance between the functions of evolutionarily older brain regions (limbic system) and the neocortex (prefrontal cortex)" (Neurobiology of Wisdom?: A Literature Overview - PMC). Specifically, the lateral prefrontal cortex participates in logical reasoning, planning, and impulse inhibition, providing the cool-headed analytical side; the medial prefrontal cortex and cingulate cortex are associated with emotions and social cognition, offering the empathetic, value-judging side (Neurobiology of Wisdom?: A Literature Overview - PMC). Wise decisions often strike an appropriate compromise between emotional intuition and rational judgment, such as considering both moral feelings and objective assessments in ethical dilemmas. Brain imaging studies support this: individuals with higher wisdom scores exhibit more interaction and modulation between prefrontal (cognitive control) and limbic (emotional resonance) activity when facing moral dilemmas (Neurobiology of Wisdom?: A Literature Overview - PMC). This suggests wisdom is not purely rational supremacy but a product of emotion-cognition synergy.

Another aspect is that wisdom involves the synthesis and abstraction of multimodal information. We need to integrate knowledge from diverse sources (visual, linguistic, social cues, etc.) to generate insightful judgments. In the brain, the parietal association cortex, particularly the inferior parietal lobule/angular gyrus, plays a role in multimodal integration and semantic abstraction. Along with the prefrontal cortex, it forms a network for high-level cognitive tasks. Research identifies the fronto-parietal network as the brain's executive control network, responsible for flexibly coordinating cognitive resources based on goals (Fronto-Parietal Network - an overview | ScienceDirect Topics). This network includes the dorsolateral prefrontal cortex, inferior parietal lobule, and intraparietal sulcus, deemed "critical for rule-based problem-solving, working memory maintenance, and decision-making in goal contexts" (Frontoparietal network - Wikipedia). In wisdom-related activities, we frequently engage this network to handle complex problems, such as strategically planning life decisions or creatively solving interdisciplinary challenges. Here, the parietal cortex provides cross-modal integrative perspectives, while the prefrontal cortex enables judgment based on long-term goals. Their collaboration facilitates the weighing of multiple factors—a hallmark of wise decision-making.

Additionally, some studies link wisdom to higher virtues and meaning, such as compassion, open-mindedness, and tolerance for uncertainty. These may involve regions like the frontal pole (the most anterior part of the prefrontal cortex) for long-term planning and self-reflection, and the posterior cingulate cortex for self-relevance evaluation. These structures also belong to the broader DMN, active during deep introspection and life-meaning integration. Thus, the realization of the wisdom layer likely requires effective coordination between task-positive networks (fronto-parietal executive network) and task-negative networks (DMN introspective network). Wise individuals can detach from immediate situations for holistic thinking (activating DMN-related regions) while taking action based on current realities (activating executive networks); the two networks collaborate without overemphasizing either. This whole-brain cooperative characteristic may be the neural signature of wisdom.

In summary, the neural correlates of Wisdom (W) comprise a distributed yet tightly interactive network, primarily including: advanced prefrontal regions (e.g., lateral PFC for rational analysis, frontal pole for abstract reflection, medial PFC for value/social assessment), parietal association areas (multimodal semantic integration and attentional allocation), and their interaction loops with the limbic system. Through these neural mechanisms, the brain applies knowledge to complex situations, making decisions that transcend mechanical reactions. From the DIKWP perspective, the brain's wisdom layer realizes the "refinement" and "tailoring" of knowledge, ultimately aligning with purpose-driven behavior.

P (Purpose): Motivation Generation and Intentional Control

Purpose (P) is the top-level element of the DIKWP model, representing the system's motivations, goals, and intentions—it determines the direction of cognitive activity. For the human brain, the purpose layer corresponds to the generation of volition and plans, as well as high-level control over actions. Neuroscience attributes these functions to the prefrontal-limbic connectivity circuits, particularly regions involved in reward evaluation, decision-making, and behavioral monitoring. Among these, the medial prefrontal cortex (mPFC), orbitofrontal cortex (OFC), and anterior cingulate cortex (ACC) are widely recognized as closely tied to motivation and decision control.

The medial prefrontal cortex (mPFC) plays a key role in goal-setting, self-referential processing, and long-term evaluation. The mPFC is also part of the default mode network; it becomes highly active when people contemplate their futures, set long-term goals, or make value-based judgments. It applies an individual's experiences and values to current choices, serving as the interface between knowledge/wisdom and goals. For example, during career planning, the mPFC integrates self-concept and social values to form intentions like "what I want to become." Such processes exemplify the transformation from wisdom (W) to purpose (P): high-level synthetic judgments yield concrete intentions and objectives.

The orbitofrontal cortex (OFC) focuses on evaluating the rewards and punishments of behavioral options, closely tied to motivation and decision-making. OFC neurons encode the value of different choices, including immediate rewards and delayed consequences, helping us compare which option better aligns with our purpose during decision-making (Orbitofrontal Cortex - an overview | ScienceDirect Topics). Once a goal is established (signaled from the P layer), the OFC participates in planning strategies to achieve it, predicting the pros and cons of each step to guide action plans. Research shows that the OFC is active during context switching and experience updating, continuously adjusting goal evaluations based on recent outcomes (Emotion, motivation, decision-making, the orbitofrontal cortex...) (Orbitofrontal Cortex - an overview | ScienceDirect Topics). This can be interpreted as: at the purpose layer, the system does not rigidly execute fixed plans but flexibly monitors effects and revises intentions to better achieve the ultimate purpose.

The anterior cingulate cortex (ACC) can be regarded as the "supervisor" and "dispatcher" of purpose execution. It plays critical roles in attention allocation, conflict monitoring, error detection, and motivation-driven behavior maintenance (Anterior cingulate cortex - Wikipedia). When striving toward a goal, the ACC monitors the gap between current behavior and the goal, triggering corrective mechanisms upon detecting errors or conflicts (Anterior cingulate cortex - Wikipedia). Simultaneously, the ACC has extensive connections with other brain regions: its dorsal portion links to executive areas like the prefrontal and parietal cortices, while its ventral portion connects to emotional/motivational areas like the amygdala and nucleus accumbens (Anterior cingulate cortex - Wikipedia). This anatomical positioning allows the ACC to integrate top-down information (goals, rules) and bottom-up information (error feedback, emotional signals), mediating between the two (Anterior cingulate cortex - Wikipedia). For example, during endurance running, the ACC receives discomfort signals from the body (urging us to quit) while referencing volitional goals (to finish the run), thereby generating motivational decisions to continue or stop. This exemplifies a typical function of the purpose (P) layer in the brain: allocating cognitive and behavioral resources based on high-level intentions, overcoming interruptions, or adjusting strategies. ACC activity correlates with an organism's motivational level, significantly engaging when tasks require extra effort (Anterior cingulate cortex - Wikipedia)—highlighting its role in sustaining motivation. It could be said that the ACC ensures our purposes transcend mere ideas, driving attention and effort to translate into action.

Beyond these core regions, the dopaminergic system (e.g., the nucleus accumbens and ventral tegmental area), as the brain's reward circuitry, profoundly influences purpose formation and execution. By modulating pleasure and drive, it makes certain purposes more appealing. For instance, enhanced dopamine signaling increases motivation to pursue rewards (stronger purpose-directed behavior). Of course, these neuromodulatory mechanisms work in close concert with cortical regions, ensuring the purpose layer encompasses both concrete goal representations (in prefrontal networks) and emotionally driven trade-offs (in limbic-reward systems).

In summary, the neural correlates of Purpose (P) primarily involve: the medial/orbitofrontal prefrontal cortex (goal-setting and value assessment), the anterior cingulate cortex (conflict monitoring and motivation maintenance), and reward-related structures (e.g., the nucleus accumbens, encoding motivational drive). Through the collaborative work of these structures, the brain generates subjective volition, sets goals ranging from short-term to long-term, and continuously monitors and calibrates behavior to achieve these goals (Anterior cingulate cortex - Wikipedia). Within the DIKWP model, these neural processes correspond to refining intentions from the wisdom layer and then translating them into regulatory control over data and information processing. It is the guidance of the purpose layer that enables our cognitive system to operate directionally, rather than merely reacting passively to stimuli.

(Through the above analysis, we have mapped the brain's major cognitive structures to the five semantic elements of the DIKWP model. This mapping provides a novel methodological foundation for understanding the brain: dividing brain functions based on semantic roles rather than traditional anatomical classifications. From this perspective, different brain regions may be viewed as functional clusters realizing the same semantic function, while a single region may participate in multiple semantic levels of processing. More importantly, we see that these elements do not operate independently but interact closely. In the next section, we will delve into the interactions among DIKWP network elements and their neurophysiological correlates.)

Neural Mechanisms of DIKWP Element Interactions

A distinctive feature of the DIKWP model is its networked interactions: information flow and influence can occur between any two elements, not just unidirectional transmission between adjacent levels. The pairwise combinations of the five elements form 25 potential interaction mappings (including bidirectional interactions counted as separate mappings and, in a sense, self-feedback). Do these interactions have corresponding physiological mechanisms in the brain? If so, how can we describe them using DIKWP semantics? This section will address these questions across several dimensions.

First, we examine bottom-up mappings, i.e., the progressive transmission from data to purpose. Next, we explore top-down mappings, i.e., feedback regulation from purpose to data. Then, we discuss cross-level bidirectional interactions and parallel processing, such as the cyclic influence between knowledge and information or the interplay between wisdom and knowledge. Through this discussion, we aim to outline how information flows within the brain in the DIKWP semantic space, interpreted through specific neural theories (e.g., predictive coding, global workspace theory).

Bottom-Up Semantic Transmission: From Sensation to Purpose

Bottom-up information flow corresponds to the traditional sensory-cognition-decision pathway in cognitive science, where external stimuli enter through the senses, triggering a series of progressively complex processing steps that ultimately lead to decisions and actions. In DIKWP semantics, this pathway can be summarized as D → I → K → W → P. We have previously described the brain structures and functions corresponding to each step: sensory data (D) forms in the sensory cortex, is processed into information (I) in association areas and the limbic system, consolidated into knowledge (K) via the hippocampus and neocortex, integrated and elevated into wisdom (W) by the prefrontal-parietal network, and finally crystallized into purpose (P) as intentions emerge in the prefrontal-cingulate network. Now, we examine the continuity of this transmission in the brain and the mechanisms linking each step.

When a new environmental stimulus appears—for example, a sudden alarm sound—the bottom-up pathway activates: sensory input (auditory data, D) quickly reaches the auditory cortex and amygdala. The auditory cortex performs an initial analysis of sound frequency and loudness, while the amygdala rapidly assesses its emotional significance (e.g., labeling it as a "danger signal"). Thus, the sound is no longer purely acoustic data but becomes information (I) with the meaning "requires attention." In higher auditory cortical regions and frontal language areas, the brain may recognize the alarm as a fire alarm, such as by matching the sound pattern to stored memories (invoking knowledge, K: past learning about fire alarm sounds). At this point, the brain not only knows "there is a sound" but also identifies its source and meaning (fire alarm, danger, evacuation needed). This knowledge activation situates the individual in context: memories and knowledge related to fire scenarios are retrieved (e.g., escape routes, past drill experiences).

Next, the wisdom (W) layer comes into play: integrating the current building layout, the status of family members nearby, and the fire situation, it applies retrieved knowledge to the unique present context to make the best behavioral decision (e.g., deciding whether to notify others first or evacuate immediately, choosing which path to take). This involves situational judgment, prioritization, and possibly overcoming panic to think calmly. Ultimately, after comprehensive consideration, a clear intention (purpose, P) forms—for example, "lead the family to evacuate via the nearest safe exit." This intention is translated into a concrete action plan (initiating evacuation) under the drive of the ACC and premotor areas.

This example demonstrates that each step of the bottom-up process is supported by corresponding neural mechanisms and achieves the progressive elevation of DIKWP semantics: sound waves (data, D) → perceived alarm sound and triggered fear (information, I: auditory association cortex + amygdala) → recognized fire alarm (knowledge, K: hippocampal memory retrieval + related cortices) → calm evaluation of evacuation plan (wisdom, W: parietal integration of environment + prefrontal assessment of options) → resolved decision to act (purpose, P: ACC activation of volition + motor system initiation). Each step's output provides the necessary semantic elements for the next, ultimately achieving the transformation from sensation to intention.

This bottom-up transmission can also be mapped to the brain's hierarchical processing system in neuroscience. For instance, the hierarchical flow model of sensory information proposed by Felleman and Van Essen suggests that sensory information is processed through a series of progressively abstract cortical regions before reaching higher prefrontal levels (DIKWP Semantic Mathematics: Embracing a Networked Model – Research Notes). Although the DIKWP model is not strictly an anatomical hierarchy, the above example shows that, under normal conditions, information indeed follows a D→…→P direction. This ensures that our actions are based on perception and understanding of the external world, not arbitrary assumptions.

It is worth noting that the bottom-up process does not always fully traverse all five steps (D, I, K, W, P). In reality, partial pathways may occur—for example, rapid responses to simple stimuli may bypass complex knowledge and wisdom processing, going directly from stimulus (D) to reflexive motivation (P). Even so, the "absent" layers typically operate in the background as existing content. For instance, withdrawing a hand upon touching a hot object is a rapid "D→P" pathway (via spinal reflexes and brainstem regulation), but our wisdom and knowledge intervene afterward to learn from the experience and prevent recurrence. These examples illustrate the connectivity between bottom-up transmission and higher layers.

In summary, bottom-up interaction mappings link the five DIKWP elements into a perception-driven cognitive chain. The brain achieves the transformation from environmental data to internal purpose through progressive processing. The neural basis of this pathway is supported by extensive research, such as the anatomical projections of sensory pathways and the changing receptive fields of neurons across cortical levels, all aligning with the increasing integration of information from low to high levels (Sensory Cortex | Definition, Location & Function - Lesson - Study.com) (Fronto-Parietal Network - an overview | ScienceDirect Topics). The DIKWP model assigns explicit semantic labels to this hierarchical processing, enabling a higher-level abstraction to describe this natural functional chain in the brain: the ladder of data leads to the peak of wisdom, which then grounds itself in purpose-driven action.

Top-Down Feedback Regulation: From Intention to Perception

Equally important as bottom-up flow is top-down signal transmission. The brain does not passively process input but actively regulates its own operations to serve current goals and expectations. The DIKWP model's networked structure emphasizes the influence of higher-level elements (e.g., W, P) on lower-level ones (e.g., D, I). Classic top-down mappings include: P→W (purpose influences wisdom-based judgments), P→K (purpose drives specific knowledge retrieval or learning), W→I (wisdom alters the interpretation of current information), K→D (existing knowledge influences perceptual sampling), and more. Among these, P→D (purpose directly acting on the data layer) and K→I (knowledge acting on information processing) are particularly intuitive examples. We will illustrate these with neural mechanisms.

Attention is the most direct manifestation of top-down modulation of perception. The brain selectively focuses on certain sensory data while ignoring others based on current purpose (P). This is essentially a P→D mapping: high-level intentions instruct the sensory system to "look here, not there." Neurologically, attention is implemented via feedback from higher cortical areas (e.g., parietal and frontal eye fields) to early sensory cortices. When we decide to search for a red object, higher visual areas send feedback signals to the primary visual cortex to enhance red features, making red objects more salient in the visual field. Experimental evidence shows that attention can amplify the responses of corresponding sensory neurons while suppressing irrelevant stimuli (Applying the efficient coding principle to understand encoding of ...). This corresponds to P (the intention to find a red target) influencing D (the selection of visual input data). Similarly, listening to a familiar voice in a noisy environment involves tuning auditory cortex sensitivity to that voice—a case of the brain filtering auditory data (D) based on social purpose (P).

Expectations and predictions are also manifestations of P→D. If we anticipate a signal, baseline activity in sensory cortices adjusts to capture the expected data more sensitively. The predictive coding framework deeply explains the role of top-down feedback in perception. According to this theory, the brain generates predictions about sensory input at higher levels (based on knowledge, K, and current inferences, W), then transmits these predictions top-down to lower sensory areas to compare with actual input (Distinct Top-down and Bottom-up Brain Connectivity During Visual ...). Lower areas only relay "prediction errors" upward. This means that higher-level knowledge and intentions actively shape how we interpret sensory data—we only "notice" and update our models when input deviates from predictions. Under this framework, knowledge (K), wisdom (W), and purpose (P) collaboratively guide the processing of information (I) and data (D), endowing the cognitive system with an active construction property rather than complete passivity (Distinct Top-down and Bottom-up Brain Connectivity During Visual ...). For example, viewing an ambiguous image with a preconception (K) biases perception toward confirming that preconception; only when the mismatch is too great do we adjust our prior beliefs. This can be seen as the combined effect of K→D and K→I: knowledge pre-frames perception, making us "see what we expect to see."

Another example is memory-guided perception. When walking down an unfamiliar street with the purpose (P) of finding a restaurant, we retrieve memories of what restaurants look like (K) to compare with the shops we see (D). Here, the brain may directly generate representations in visual cortex based on memory to aid matching with real input. This mental imagery/recollection is a form of top-down activation: the hippocampus and higher visual areas produce expected images in the absence of actual visual stimuli, then use discrepancies with real input to guide search direction. Brain imaging studies show that when people search for a target in a scene, top-down signals from parietal regions alter visual cortex activity patterns to resemble the target's features (Distinct Top-down and Bottom-up Brain Connectivity During Visual ...). These are instances of knowledge and goals shaping perception—K, P→D.

Beyond perceptual feedback, higher-level influences on mid-level information processing are also widespread. For example, an expert, due to extensive domain knowledge (K), interprets information (I) differently from a novice: the expert quickly identifies patterns and filters noise, while the novice may miss key points. Neural evidence shows that in experts, relevant knowledge networks (e.g., cortical regions encoding professional expertise) activate immediately upon encountering domain-related information, influencing how visual or auditory association areas encode new input. This explains phenomena like "experience-driven perception" or "prior knowledge shaping sensation." Specifically, K→I makes the brain's processing of familiar information more efficient and precise. For instance, a chess master viewing a board quickly recognizes typical configurations (knowledge) and focuses on critical positions (information processing), rather than scanning square by square. Similarly, the influence of wisdom (W) on information manifests in cognitive framing: experienced individuals interpret new events through principles and lessons (wisdom), potentially deriving different meanings (information). This is essentially W→I—using high-level experiential judgment to selectively interpret current information. For example, an optimist and a pessimist (with different life wisdom, W) may focus on positive or negative aspects of the same event, respectively.

Another classic top-down modulation is the effect of motor imagination and preparation on sensation. When we perform an action (driven by purpose, P), the motor cortex sends an "efference copy" to sensory areas, allowing us to anticipate the sensory consequences of our actions (e.g., not being startled by our own voice or unable to tickle ourselves). This can be seen as P→I/D: intention pre-informs the sensory system about what to expect, ensuring we correctly distinguish self-generated from external sensations. This mechanism involves the insula and anterior cingulate cortex in interoceptive monitoring networks, which dampen sensory responses to self-produced actions (e.g., reduced sensation when tickling oneself) because a high-level intention signal tells sensory cortex, "This is self-caused; no need to overreact."

In summary, top-down mappings endow the brain with "expectation-driven" properties. Through these feedback loops, cognition can actively filter input, fill in missing information, accelerate processing, and correct errors. In the DIKWP model, this is reflected as higher-level elements influencing the semantics of lower-level ones: purpose dictates what data to attend to, wisdom alters information interpretation, and knowledge provides perceptual frameworks. Such bidirectional information flow makes the DIKWP model a truly cyclic network rather than a linear pipeline. Returning to the fire alarm scenario, high-level purpose (evacuation) feeds back to perception: heightened alertness to sounds and exit signs (P→D); knowledge and wisdom also shape how we process current information: someone with drill experience may calmly search for exits (K→I), while a panicked person may ignore instructions (W→I, negative case). These examples prove that cognition without top-down regulation is unimaginable.

Cross-Level Interactions and Parallel Processing: Loops and Holistic Emergence

Beyond the typical bottom-up and top-down processes, the brain's information processing is also characterized by parallelism and direct cross-level interactions. The network structure of the DIKWP model allows direct connections between non-adjacent elements, such as D↔K, I↔P, and W↔K, among others. These interactions are supported by empirical evidence in the brain. Below, we discuss several representative cross-level interactions and how they enable rapid, holistic cognitive responses.

Direct Sensory-Knowledge Links (D→K, K→D)

At times, sensory input can instantly trigger vivid memories or knowledge recall. For example, a specific smell might immediately evoke childhood scenes, or a single phrase might instantly bring to mind a piece of learned knowledge. This suggests that the data layer (D) directly activates the knowledge layer (K), with intermediate information processing seemingly bypassed. The neural mechanism may involve:

Olfactory pathways, which are uniquely connected to hippocampal circuits, allowing smell data to directly access memory systems with minimal intermediate processing.

Trained neural circuits, where specific perceptual patterns in experts have established direct synaptic links to stored knowledge, enabling rapid recognition.

Conversely, knowledge can also directly influence perception. For instance, experts exhibit heightened sensitivity to subtle signals because their knowledge reinforces neural circuits for specific data patterns. Neuroscientifically, this resembles conditioned reflex pathways or learned pattern recognition—where certain stimuli bypass conscious analysis to directly invoke stored knowledge. In DIKWP semantics, this is a D-K direct link. Such connections enhance cognitive efficiency, allowing reflexive knowledge retrieval in familiar contexts without exhaustive analysis.

Emotion-Purpose Interactions (I↔P)

Emotions, as a special form of information, can directly influence motivational purposes. Intense emotional information (I) may override rationality, causing abrupt shifts in purpose (P)—for example, fear driving flight despite prior plans. Conversely, strong purposes (P) can suppress emotional reactions (I), such as soldiers overcoming fear through使命感. This indicates rapid bidirectional I↔P interactions that may bypass knowledge or wisdom layers.

Neurally, the limbic system (emotional center) and prefrontal/cingulate regions (purpose center) are directly connected. For instance:

The ventral anterior cingulate cortex (ACC) links to the amygdala and can downregulate its activity (Anterior cingulate cortex - Wikipedia).

Extreme emotional stress can impair prefrontal function via stress pathways.

These I↔P mappings are particularly salient in emergencies or high-stakes motivation scenarios.

Wisdom-Knowledge Integration (W↔K)

Though wisdom (W) and knowledge (K) are semantically distinct, they are deeply intertwined in practice. Refining wisdom requires constant reference to knowledge, while wisdom reorganizes knowledge structures. Thus, W↔K represents a tightly coupled loop.

In the brain, this corresponds to recurrent circuits between the prefrontal cortex (wisdom/decision-making) and posterior cortices/hippocampus (knowledge storage). For example:

Working memory serves as a temporary "stage" for wisdom, drawing content from long-term memory (knowledge).

New insights (wisdom) often consolidate into new knowledge, while learning (knowledge) relies on wisdom to derive meaning.

This W↔K cycle is supported by prefrontal-hippocampal loops and interactions between the default mode network (DMN, for knowledge/introspection) and executive networks. Research shows that during complex problem-solving, the DMN and prefrontal networks alternate in activity, possibly reflecting wisdom assimilating knowledge and vice versa (Neurobiology of Wisdom: A Literature Overview - JAMA Network). Such iterative interactions can lead to emergence—sudden insights or revised knowledge frameworks after multiple W↔K cycles.

Parallel Processing and Multitasking

A hallmark of the human brain is its ability to handle multiple information streams concurrently. The DIKWP model accommodates parallel pathways, such as walking (driven by W→P→D for motor control) while contemplating a problem (another I→K→W stream). These processes coordinate via overarching goals (e.g., avoiding obstacles).

Neurally, parallel processing involves concurrent activation of distinct functional networks. For example:

A motor subnetwork (primary motor cortex, cerebellum) handles locomotion autonomously, from proprioceptive data (D) to step execution (P).

A cognitive subnetwork (prefrontal cortex, hippocampus) processes abstract problems (I→K→W).

The ACC and parietal cortex monitor conflicts (e.g., interrupting thought to focus on walking when needed).

This aligns with the global workspace theory, where specialized modules process in parallel, and results requiring global attention are broadcast across the brain (What is Bottom-Up and What is Top-Down in Predictive ... - Frontiers). The DIKWP model captures this distributed parallelism: local interactions can rapidly influence the entire network when necessary (e.g., sudden pain (D) hijacking attention at the purpose layer (P)).

Feedback Loops and Adaptive Control

Feedback loops permeate all levels of interaction, stabilizing cognition (DIKWP Semantic Mathematics: Embracing a Networked Model – Research Notes). Feedback can:

Calibrate high-level outcomes (e.g., behavioral results updating knowledge (K)).

Adjust low-level processes continuously (e.g., goal-maintained attention).

For instance, if a planned route (W→P→D) is blocked (unexpected D), feedback loops enable wisdom (W) to revise the purpose (P) and redirect data collection (new path D). This closed-loop control ensures cognition adapts dynamically to reality.

Conclusion

The 25 DIKWP interactions are not theoretical abstractions but reflect the brain's multilayered, parallel, and recurrent information processing. Modern neuroscience theories—predictive coding (cyclic high-low interactions), global workspace (cross-module communication), and emotion-cognition interplay (limbic-cortical links)—are unified under the DIKWP semantic framework. The model thus offers a generalized cognitive architecture: the brain is a complex network of DIKWP elements whose couplings enable the holistic emergence of intelligence. Creativity arises when memory (K) and wisdom (W) collide; extraordinary resilience emerges when emotion (I) aligns with purpose (P). These phenomena are products of the brain's cross-level interactions, captured elegantly by the DIKWP model.

Prospects for the Mechanisms of Artificial Consciousness: Semantic Threshold and BUG Chain Hypotheses

The DIKWP model not only aids in explaining human brain mechanisms but also inspires the exploration of artificial consciousness. In this section, we propose two hypothetical model predictions regarding consciousness and autonomous intention, based on the preceding reasoning, and discuss their physiological implications. These ideas aim to stimulate further theoretical development and experimental validation.

Hypothesis 1: Semantic Emergence Threshold

We conjecture that in the DIKWP network, there may exist a critical complexity or integration level beyond which subjective experiences akin to consciousness emerge. This resembles the "Phi threshold" for consciousness in Integrated Information Theory (IIT) (Consciousness Theory - Zhihu Column), but within the DIKWP framework, we focus on the richness of semantic interactions. Specifically, when the interactions among the five DIKWP elements reach a certain intensity and global coordination, the system transitions from mechanical data processing to holistic self-experience. In other words, a sufficiently complex DIKWP network equals the condition for consciousness.

From a physiological perspective, this may relate to the degree of brain network integration. In unconscious states (e.g., deep anesthesia or sleepwalking), human brain regions remain active but lack global coordination, and the closed-loop interactions of the DIKWP network may be disrupted or weakened. Upon regaining wakefulness, the modules re-establish tight communication, the information field becomes highly connected, and the "semantic threshold" is crossed, restoring consciousness. This aligns with clinical observations: loss of consciousness often coincides with the disintegration of functional brain connectivity, while recovery involves its reorganization (124 Scientists Co-Sign: Integrated Information Theory is "Pseudoscience" - News - Paper). Thus, we might hypothesize a measurable indicator, such as the number of simultaneously active DIKWP transformation functions or the connectivity entropy of the semantic field, which, when exceeding a certain value, signals the system's entry into a conscious state.

The significance of this hypothesis lies in treating consciousness as an emergent property of the DIKWP network, emphasizing the necessity of rich semantic interactions. To replicate consciousness in artificial systems, we must simulate a similar level of DIKWP interaction complexity. Simple perception-action loops may never achieve "self-awareness," but if a system incorporates modules for data, information, knowledge, wisdom, and purpose, and enables their bidirectional communication (e.g., artificial neural networks with multi-layer recurrence and broad attention mechanisms), it might cross the semantic emergence threshold, exhibiting rudimentary signs of subjective experience. While speculative, this offers a practical engineering approach: enhancing semantic richness and interaction depth within AI may bring us closer to machine consciousness than merely increasing computational power.

Hypothesis 2: Subjective BUG Chain

The term "BUG chain" originates from computer science, referring to minor errors that accumulate into systemic failures. In cognitive systems, we hypothesize a similar chain reaction of subjective biases: a minor misinterpretation at one DIKWP layer (a semantic "bug"), if uncorrected, propagates and amplifies across layers, ultimately distorting subjective experience or behavior—a "fallacy in subjective reality."

For example, an erroneous belief (e.g., a prejudice) formed in the knowledge layer (K) might initially seem like a localized bug. However, when this belief is used as a premise in wisdom (W)-based decisions, it skews judgment systematically, leading to irrational choices in purpose (P) and manifesting as behavioral errors (D). This chain—D→I→K (erroneous knowledge) → W (flawed judgment) → P (faulty motivation) → D (aberrant behavior)—demonstrates how small errors snowball into major mistakes. Psychopathological delusions fit this pattern: an initial hallucination (noise in the data layer, D) misinterpreted as real information (I) is accepted into knowledge (K), constructing a paranoid belief system (imbalanced wisdom, W), and culminating in abnormal actions.

In artificial intelligence or artificial consciousness systems, we must guard against similar BUG chains. For instance, if an adaptive AI misinterprets low-level sensory input and its high-level decisions rely on this error, the outcome could be catastrophic (e.g., an autonomous vehicle misidentifying an object and planning a hazardous route). The DIKWP model warns us: due to strong inter-layer coupling, errors propagate and amplify across the network. Thus, reliable cognitive systems require multi-level error monitoring and correction mechanisms. The brain partially achieves this through feedback, but it is not foolproof (humans succumb to cognitive biases). For artificial systems, we might emulate the brain by incorporating:

Cross-modal validation (different data sources cross-checking the D layer),

Knowledge consistency checks (self-coherence in the K layer),

High-level reflection (W-layer simulation before decision-making).

The "subjective BUG chain" hypothesis highlights the systemic origins of subjective errors. It underscores that module accuracy cannot be assessed in isolation; the entire closed loop must be considered. Even if individual modules are highly accurate, rare systemic bias pathways can amplify into catastrophic outcomes. The DIKWP model offers a way to identify these pathways: tracing self-reinforcing loops or single-point failure propagation routes (akin to fault-tree analysis in circuits). This informs the design of redundancy and damping mechanisms for artificial consciousness or complex AI. For instance, critical decisions could involve independent secondary judgment loops (a parallel W→P pathway) to prevent single-path bugs from spiraling out of control. Such principles are already applied in aviation safety systems and may become key to developing trustworthy artificial consciousness.

Summary

These two hypotheses extend the DIKWP model into frontier exploration. The semantic emergence threshold describes a possible quantitative-to-qualitative transition (unconscious to conscious), while the BUG chain illustrates how systemic bias accumulation leads to qualitative shifts (normal to aberrant). Together, they remind us that pursuing artificial consciousness requires both fostering genuine subjective intelligence and preventing失控 and fallacies. Though still theoretical, as neuroscience and AI advance, we hope to translate these hypotheses into testable models and experiments under the DIKWP framework.

Conclusion

This report systematically analyzes and creatively interprets the neurophysiological basis and cognitive mechanisms of the brain through the lens of the networked DIKWP model. We first theoretically clarify the structure and内涵 of the DIKWP model, emphasizing its advantages over traditional hierarchical models: by coupling data, information, knowledge, wisdom, and purpose bidirectionally, the DIKWP model more fully captures the complex dynamic interactions in real-world cognition (DIKWP Semantic Mathematics: Embracing a Networked Model – Research Notes).

Next, we map the model to brain anatomy and function, proposing correspondences such as:

D (Data): Sensory cortices (e.g., primary visual cortex for raw visual input (Sensory Cortex | Definition, Location & Function - Lesson - Study.com)),

I (Information): Limbic system (adding emotional context (Limbic System: What It Is, Function, Parts & Location - Cleveland Clinic)),

K (Knowledge): Hippocampus and default mode network (memory integration (Issue: Neuron - Cell Press)),

W (Wisdom): Prefrontal-parietal network (decision-making (Fronto-Parietal Network - an overview | ScienceDirect Topics)),

P (Purpose): Medial/orbitofrontal cortex and ACC (motivational control (Anterior cingulate cortex - Wikipedia) (Orbitofrontal Cortex - an overview | ScienceDirect Topics)).

We then delve into the neurophysiological representations of the 25 interaction mappings among DIKWP elements. Bottom-up sensory conduction (D→I→K→W→P) aligns with the brain's hierarchical processing, while top-down intentional regulation (P→W→K→I→D) manifests in feedback processes like attention and prediction (Distinct Top-down and Bottom-up Brain Connectivity During Visual ...). Cross-layer direct connections (e.g., knowledge influencing perception) and parallel processing are also prevalent (DIKWP Semantic Mathematics: Embracing a Networked Model – Research Notes).

The DIKWP model unifies diverse brain theories:

Predictive coding reflects high-level (W/K) corrections to low-level (I/D) processing (Distinct Top-down and Bottom-up Brain Connectivity During Visual ...),

Global workspace theory corresponds to the broadcast integration of knowledge and wisdom,

Emotion-cognition interactions embody bidirectional regulation between information and purpose.

Innovatively, we propose:

The semantic field concept captures the holistic meaning-association space in cognitive systems,

Intent-driven pathways highlight how goals shape cognitive flow,

DIKWP-based reconstructions of complex functions (e.g., language comprehension as I→K→W, action initiation as W→P→D).

For artificial consciousness, we hypothesize:

The semantic emergence threshold suggests that sufficiently complex DIKWP networks may spontaneously generate consciousness (Consciousness Theory - Zhihu Column).

The subjective BUG chain warns that minor semantic biases can amplify into systemic errors, necessitating multi-layer monitoring.

In summary, the DIKWP model offers a novel, unified framework for understanding the brain and consciousness. By viewing the brain as a semantic network woven from data, information, knowledge, wisdom, and purpose, we bridge disciplinary divides, integrating insights from neuroscience, psychology, and AI. While many claims require further experimental validation (e.g., defining the "semantic field," identifying neural correlates of DIKWP interaction strength), the model provides fresh perspectives on the fundamental question: How does the brain give rise to the mind?

As brain science and AI progress, the DIKWP model may evolve into a vital theoretical tool for unraveling intelligence and guiding brain-inspired technologies.

References

Duan, Y. (2024). DIKWP Semantic Mathematics: Embracing a Networked Model. Research Notes.

Anterior cingulate cortex. Wikipedia.

Rolls, E. T. (2019). The orbitofrontal cortex: reward, emotion and depression. Brain Communications.

Limbic System: What It Is, Function & Parts. Cleveland Clinic.

Jeste, D. V., et al. (2011). Neurobiology of Wisdom: A Literature Overview. Archives of General Psychiatry.

Yeshurun, Y., et al. (2021). A review and synthesis: 20 years of the default mode network. Neuron.

Lubin, F. D. (2011). Epigenetic gene regulation in the adult mammalian brain. Neurobiology of Learning and Memory.

Keller, G. B., & Mrsic-Flogel, T. D. (2018). Predictive coding: a canonical cortical computation. Neuron.

Scolari, M., et al. (2015). The Frontoparietal Network: critical for executive control and decision-making. ScienceDirect Topics.

Duan, Y. (2025). Interpreting Active Medicine’s Information and Energy Field Theory from the DIKWP Model Perspective. ResearchGate.

Li, Y., et al. (2022). Controversies in Integrated Information Theory (IIT). ScienceNet.

Sensory Cortex. Study.com.

Duan, Y. (2024). DIKWP Semantic Mathematics: Embracing a Networked Model. (Case studies on medical imaging diagnosis.)

人工意识与人类意识

人工意识日记

玩透DeepSeek：认知解构+技术解析+实践落地

人工意识概论：以DIKWP模型剖析智能差异，借“BUG”理论揭示意识局限

人工智能通识 2025新版 段玉聪 朱绵茂 编著 党建读物出版社

主动医学概论 初级版

世界人工意识大会主席 | 段玉聪

邮箱|duanyucong@hotmail.com

世界人工意识科学院

邮箱 | contact@waac.ac