Resumen de Identifying the neurocognitive bases of creativity to increase human and computational creativity
Abstract In this Doctoral Thesis, the neurocognitive structure that supports human creativity has been identified based on the joint analysis of more than 800 bibliographical references that show the most important investigations carried out to date. Based on this structure, a neurocognitive paradigm of human creativity has been described, and a neurocognitive model of creative process has been proposed. Finally, a computational neurocognitive creative paradigm has been also proposed, and the structure of a creative computational multi-agent system has been designed.
The research that has been carried out on the subject is very specialized and focuses on very specific aspects of creativity, and in many cases they have little relationship with each other. For this reason, and in order to have a joint and holistic idea of the neurocognitive processes of human creativity, it is necessary to study all these investigations in an interconnected way. This joint idea would allow directing more specific investigations in order to be more effective. For this reason, the first thing that has been done has been to classify, group, analyze, intertwine and structure, in an orderly manner, the most important investigations that have been carried out to date.
However, the work carried out goes much further, since by structuring and intertwining the existing research it has been possible to identify certain patterns, correlations and parallelisms, and make certain deductions, which as a whole, have made it possible to identify the fundamental neurocognitive bases of human creativity. And based on this, it has been possible to model a neurocognitive paradigm of human creativity and a computational paradigm of creativity. Finally, a neurocognitive model of the design process and a multi-agent computational A. I. system has been proposed.
To achieve the stated objectives, this doctoral thesis has been structured in 8 chapters, the results of which are summarized below.
Chapter 1.
Introduction In the first chapter, the main and secondary objectives of the thesis are established, the state of the art of previous research is shown, the methodology followed is established, and the most important contributions are shown.
Chapter 2.
Creativity.
Analysis of the main methods to stimulate creativity In this chapter, we have first reflected on the concept of creativity, showing some of the most representative definitions that have been made throughout history. An analysis of these definitions reveals that it is a very ambiguous and nuanced concept, although in essence it is a complex process of generating and evaluating ideas. All the definitions compiled are different from each other, and collect only some aspects of the creative process. However, among all of them, two stand out that are especially illustrative. One states that the most important component of creativity is the discovery of new relationships between different concepts, in order to transform established rules, whenever a certain problem is solved. Another is much more complete, since it proposes that creativity is a process by which the human being can voluntarily alter the functioning of his cognitive system, and therefore change the habitual way of perceiving and interacting with the environment, and the habitual way to establish concepts and processes.
Therefore, the most correct and complete definition of the creative process would be the one that could include all the definitions collected, as well as those that could be collected in the future.
Given the difficulty of defining the concept of creativity, it is necessary to approach the problem in a different way. The most appropriate thing would be to identify an adequate taxonomic framework that contains the greatest number of nuances in the definition of the concept of creativity, and that at the same time allows it to be analyzed rigorously and in detail. For this reason, and as a result of analyzing all the taxonomies proposed to date, the most complete and appropriate has been chosen: the 4-P taxonomy (Person, Process, Product, Press), devised in the 1960s, and recovered today by artificial intelligence researchers.
Person. What makes the agent creative.
Process. The actions performed by the agent to generate creative results.
Product. The creative result (object, entity or process).
Press. The cultural environment in which the result is evaluated.
This taxonomy provides the most adequate and complete framework to define and analyze human creativity, and provides the most adequate reference to re-direct future research.
Within this taxonomy, the main criteria for evaluating the level of creativity of the results obtained are novelty, value and surprise, although there are also other criteria such as effectiveness, elegance, generalization, imagination, appreciation, etc.
However, no matter how well the criteria are chosen, the evaluation process will always be a problem, since if their number is reduced, some non-creative ideas could be classified as such. On the other hand, if the number of evaluation criteria is expanded, some creative ideas could be discarded. Furthermore, depending on how the evaluation criteria are chosen, certain ideas may be rewarded to the detriment of others. Finally, ideas can also be more or less creative depending on their conceptual or temporal environment. Let's take an illustrative example: the idea of roasting meat (to protect its interior) is not a creative idea since it is done regularly, whereas the idea of roasting wood (to protect its interior) could be considered very creative. Another complementary example would be Van Gogh's paintings, which were not valued at the time, but years later are highly valued.
Taking into account the chosen taxonomic framework, a compilation and analysis of the most effective methods that stimulate human creativity has been carried out. In the same way, the advantages and disadvantages of each one have been shown, as well as the most suitable environments to use them.
The choice of methods has been made for their differentiating character, and for their effectiveness (number of ideas generated, and their novelty), under the 4-P taxonomy. It should be noted that all these methods have been designed following methodological guidelines identified by analyzing the creative results of the human brain, as if it were a black box, and without knowing its neurocognitive structure.
The fact that, throughout history, these methods have been created to stimulate human creativity shows that the human brain has certain neurocognitive limitations, and that its functionality must be redirected so that it can obtain creative results.
In essence, what the methods implemented to enhance human creativity do is force the search tree to be traversed by unusual paths We tend to solve problems in more or less the same way, since we tend to use the same procedures. For this reason, the aforementioned methods force us to carry out cognitive processes different from the usual ones, when it is evident that these are sterile. This shows that the creative process is much more flexible and complex than could be derived from the analysis of the conceptual models on the structure of the creative process proposed to date.
Analysis of these methods provides an overview of the strengths and weaknesses of the human brain in creative problem solving. It also raises questions that can only be answered by investigating the neurocognitive activity of the human brain. The methods to stimulate creativity have been designed and modified, throughout history, through trial-and-error. New methods have been proposed, and existing ones have been improved, flying blind, making continuous proposals and subsequently observing their effectiveness. And this has been a very slow process, and in many cases ineffective.
When an effective method is achieved, attempts are made to speculatively identify the reasons for its success, although it is never certain. For this reason, many times, when based on these speculations an attempt is made to improve a method, what is done is to make it worse (as has repeatedly happened with the different Brainstorming proposals). Therefore, other reasons are needed to explain and justify the success of some methodologies to amplify human creativity, with the aim of improving them, or proposing more effective ones. These reasons could come from the knowledge of the neurocognitive factors of creativity.
Based on the analysis of the behavior of creative people, and based on the analysis of the methods to stimulate creativity, different conceptual models have been identified on the structure of the creative process. Taking into account all this accumulated information, in this Doctoral Thesis a conceptual model of the creative process has been proposed, completing and detailing the previously proposed models. According to this model, the creative process is a complex process of generating and evaluating ideas. This process has a sequential component, and also a transversal (parallel) component, and at all times top-down and bottom-up decisions are made. The creative process takes place at various levels of abstraction in each of its stages, making decisions of all kinds, from those that are more abstract and general, to those that are more concrete and detailed. From the beginning of the creative process, ideas are proposed, so the imagination process is fundamental, and above all it is based on the surprising association of previously known concepts and ideas. The creative process is also very flexible, and in many cases indeterminate, which means that small changes in the input information can generate very different ideas.
These conceptual models of the creative process are what have been framing the different investigations on the neurocognitive structure of creativity. In turn, the different discoveries made are substantially modeling the previously established conceptual models.
Chapter 3.
Functional structure of the human brain and its relationship with the creative process This chapter has identified the general neurocognitive structure of the human brain that allows the generation of the fundamental and basic processes of its creative activity.
The fundamental neurocognitive factors that have been identified are the following: - Capacity of the human brain to generate neurocognitive processes in series and in parallel. Creative activity requires going through the search tree for solutions in series and in parallel, at different levels of abstraction, in order to arrive at surprising and creative solutions, and the human brain has an enormous facility for doing this. We have identified that parallel cognitive processes are fundamentally developed through large-scale networks, which integrate different cortical regions with varied functionality. Serial cognitive processes are based on gradients of neuronal activity, from sensorimotor nodes to high-level cognitive networks.
- Functional structure of the human brain based on the activation of large-scale neural networks, which integrate different specialized areas strategically distributed topographically. The operation based on large-scale networks allows the generation of high-level cognitive processes and various levels of abstraction. In this way the same networks are activated to solve apparently different problems, and each time they are activated they do not involve exactly the same specialized areas of the brain. This allows the creation of analogies at a high semantic level, and the generation of general processes at a high level of abstraction, which have a common structure, but with different details for each specific activity. To understand the general cognitive activity of the human brain, in this chapter the most important networks and sub-networks that have been identified to date have been specified and analyzed, and the most relevant ones in the generation of creative processes have been identified.
- Existence of a multimodal integration network (MIN), which integrates the hierarchical flow of information between the different large-scale networks, at different levels of abstraction, from sensorimotor regions to multimodal associative networks with a high cognitive level. The generation of creative ideas requires parallelism and cognitive processes to be carried out at different levels of abstraction, and the multimodal structure of the brain allows this in a natural way.
- Existence of a certain topographical structure of the cortex. According to the Tethering hypothesis, the human brain has expanded as a result of the need for greater neurocognitive capacity to deal with increasingly complex information, in increasingly complex societies. The expansion was done in such a way that the different sensorimotor regions separated from each other, and the cortical mass grew around each of these regions, creating new concentric cortical areas. As they moved away from these nodes, the different areas could gradually engage in activities with a higher cognitive level since they freed from the restrictions of the sensorimotor systems. In this way, the different large-scale functional networks were integrating different peripheral areas of the different sensorimotor nodes, according to its functionality. Therefore, the information processed by the sensorimotor nodes is gradually processed and transformed as it moves away from them. The information is becoming more and more abstract, and is structured by levels. Therefore, the topographical evolutionary structure of the human brain allows the generation and manipulation of abstract thoughts, which is essential for the generation of creative ideas. It also enables the creation of self-generated thoughts, which are fundamental to the creative process.
- Existence of basic neurocognitive structures to support the fundamental components of creativity: generation and evaluation of ideas, imagination, and semantic manipulation. The most important ones are specified below: - Existence of a basic neurocognitive structure (Triple network model) to generate and evaluate ideas fundamentally based on the interrelation of three large-scale networks: the Executive control network (ECN), the Default mode network (DMN), and the Salience network (SN). Within this model, the ECN is activated when the SN detects a relevant external event that requires cognitive attention to perform a specific goal-oriented task. Instead, when there are no external tasks (or when the tasks are trivial or require previous experience) DMN is activated. This basic creative structure justifies many patterns traditionally observed in the process of generating creative ideas, and perhaps the best known of all is what is known as “eureka”. In this pattern, great ideas seem to come out of nowhere, without much effort. However, it is based on the neurocognitive model of the triple network model.
When the problem is very well defined, or simple, the ECN usually kicks in and usually solves the problem. On the other hand, if the problem is poorly defined, or especially difficult, and at the same time very demanding, the ECN does not usually solve the problem. In this case, through the analytical cognitive processes of the ECN, an attempt is made to solve a certain problem over and over again, but the solutions obtained are not satisfactory. The unsuccessful process seems to have no end, and finally certain breaks are usually taken, trying to "disconnect" with the problem. And just at that moment, in the middle of the break, is when an idea appears in the mind, as if it came from nowhere.
What has actually happened is that the SN identifies goal-oriented external events over a long period of time. In this case, the ECN is activated (and the DMN is deactivated) and tries to find a solution. If the solution is unsatisfactory, the ECN tries to find another, through an iterative and continuous process over time. The process is usually unsuccessfully prolonged over time, and finally a break must be taken. In this case, the SN detects that there are no external events, so the DMN is activated and the ECN is deactivated. The DMN tries to solve the problem, unconsciously, by manipulating both the input information and the partial information generated with the ECN. The DMN also causes the mixing of the available information with personal and sensory information, so it is able to go through the search tree by another path, and in this way an unusual creative solution can be reached, difficult to reach by conventional paths. . The SN now detects a new outgoing event (a possible internally generated solution) and the ECN is activated so that it can be conveniently evaluated consciously, and with the eventual help of the DMN. The ECN may never be able to come up with this solution as it uses rigid and more conventional strategies. In contrast, the DMN uses more flexible strategies, subtly modifying the information in progress, and mixing it with personal information stored in memory, and with sensory and cognitive information in progress.
The left dorsolateral prefrontal cortex (DLPFC) plays a fundamental role in this process, as it is responsible for prioritizing self-generated thoughts in low-demand situations (context-dependent prioritization of off-task thoughts). The MTL (component of the DMN) also contributes to associative processes allowing the generation of creative ideas.
The DMN has a double functionality, since it contributes to the generation of creative ideas (due to its low level of information control), and it also contributes to the evaluation of ideas (since it is involved in the process of affective and visceral evaluation). Alternatively, the ECN contributes to the process of analytical evaluation of creative activity. The DMN is also involved in what is called “inner self evaluation”, that is, in the evaluation of a problem based on the most internal and personal information. It is also involved in the deep pleasure generated in the contemplation and evaluation of works of art that connect directly with the deepest part of our being (aesthetic catharsis), which is a fundamental ingredient for human creative capacity.
- Existence of a basic neurocognitive structure to imagine things and plan the future. Creativity implies the future visualization of something that does not yet exist. Therefore, in the creative process, although in a blurred and ambiguous way, plans must be made, in the same way that people project their personal goals. When imagining and planning for the future, two different processes must be carried out: - Process simulation. It is the process of visualizing the steps that must be taken to obtain a certain goal.
- Outcome simulation. It is the process of visualizing the effects that are obtained once a certain goal has been reached.
These two processes are constantly performed throughout the creative process. On the one hand, the necessary steps that must be carried out to achieve a certain objective must be visualized, and on the other hand, the desired result must be also visualized (in order to reach a solution), to experience the pleasure of the possible associated rewards (recognition, personal achievement, academic achievements, money, etc.).
- Existence of a basic neurocognitive structure to carry out semantic processes at different levels of abstraction. As has been seen, throughout the creative process information must be manipulated with different levels of abstraction, from more general aspects to more specific aspects. The most general aspects can be shared by a greater number of entities, objects or ideas, which allows the search tree for solutions to be traversed by more varied paths. Therefore, a correct semantic structure of the information and a correct system for processing it are needed.
According to recent studies, semantic memory is generated in the brain through a transmodal center. Classical neurocognitive paradigms assumed that modality-specific regions interact directly to generate multimodal concepts. However, it has been discovered that the brain has an intermodal semantic center in which semantic functions are carried out. Based on this, concepts are created based on a combination of cross-modal and modality-specific representations. Of course, modality-specific information is a basic source for concept construction, but it is insufficient, since the different “attributes” (characteristics of objects) are combined in a complex and non-linear way to form a concept. In this context, the cross-modal semantic center can provide two important "semantic functions": - It allows for correct mappings between the modality-specific "attributes" and a certain relevant concept.
- It allows for cross-modal representations that provide the basis for making semantic generalizations taking into account conceptual similarities, rather than superficial similarities (which constitute the central function of semantics).
Chapter 4.
The fundamental role of the DMN in the creative process In the previous chapter, the fundamental neurocognitive factors of human creativity have been analyzed, and it has been observed that the Default mode network (DMN) has a main role. For this reason, a chapter has been dedicated to its study, and several factors have been identified that directly involve it in the creative activity of the human brain. Among these factors, the following stand out: - The DMN is directly involved both in the generation of ideas (through its management of self-generated thoughts), and in the affective and visceroceptive evaluation of ideas (through its ability to manage memory, in memory-related tasks), autobiographical, episodic, and semantic, mind wandering, perspective taking, or future thinking).
- The DMN is at the end of the main gradient, so it handles abstract information, shared by the different tasks in which it may be involved, which only differs from each other in specific details. This allows it to also engage in high-level cognitive processes, which are apparently unrelated to each other, but share a common abstract structure. This also allows it to get involved in autobiographical cognitive processes and those related to the imagination of fictitious situations and scenarios. As a result, this fact gives the DMN a leading role in creative processes.
- DMN nodes are topographically located at the maximum possible distance from unimodal sensorimotor regions. This allows the DMN to manipulate information at the highest level of abstraction possible, with little or no relation to sensorimotor information. Likewise, it allows you to perform functions of a high cognitive level, away from the “here and now”. Among these functions, its ability to create self-generated thoughts stands out, which is the natural activity the brain prioritizes (in the absence of external tasks), and which are the basis for the generation of ideas in creative processes. On the other hand, the strategic topographical location of the DMN allows it to be involved in a wide variety of tasks that apparently have little in common.
- The DMN has a complex functionality and a high cognitive level. It was commonly known that the DMN deactivated during tasks, and also that its activity increased when it was not performing tasks. This generated an initial idea that emphasized task-negative functions of the DMN. However, recent studies show that DMN increases its activity by performing tasks in which cognition and behavior must utilize the memory of related previous experiences. In fact, it has been proven that the DMN increases its activity (under conditions in which it should remain deactivated) on at least the following occasions: - When the decisions to be made depend on the information that has been generated in a previous trial.
- When task-relevant stimuli should make use of long-term memory.
- When the context of a certain task is retrieved from memory.
- When a certain action rule has already been coded that can be used in subsequent actions.
- When performing very demanding semantic memory or working memory tasks.
All these circumstances occur continuously, and in an intertwined way, throughout the creative process.
- The DMN collaborates closely with the ECN, and together they have differential contributions during the different stages of the design process. The DMN is more intensely activated during the idea generation stage and, conversely, the ECN is more intensely activated during the idea evaluation stage. The DMN participates in a wide variety of imaginative processes, such as those commonly referred to as: mind wandering, mental simulation, and episodic future thinking, which are highly relevant during the creative process. Consequently, functional connectivity between the inferior prefrontal cortex (IPFC) (ECN region) and the DMN may reflect top-down control of bottom-up processes. In other words, the cognitive control mechanisms of the IPFC are in charge of directing and monitoring the spontaneous activity of the DMN.
- The DMN has patterns of meta-functionality. DMN participates in generic, high-cognitive activity patterns that can be shared by a wide variety of different and heterogeneous activities. Among these meta-functionalities observed in the investigations, the following stand out: - Delayed matching, according to which the functionality of the DMN can vary according to the way in which cognitive processes emerge within the temporal structure of the current task.
- Contextual modulation, according to which the DMN can vary its pattern of neuronal activity based on previous experience.
- Rule-based decision making, according to which the DMN increases its activity when internal rules are used to perform tasks.
These meta-functionalities are essential in the creative process.
- The DMN collaborates in the elaboration of detailed cognitive representations when performing tasks. The DMN integrates information from regions lower in the cortical hierarchy to create a representation of the ongoing neurocognitive context. As stated, the cerebral cortex is structured in a specific spectrum of regions ranging from unimodal to transmodal. In this spectrum, as we move away from unimodal cortex, neural activity encodes detailed thought patterns during periods of working memory maintenance. This suggests that the DMN is involved in the elaboration of detailed cognitive representations through interactions with lower regions of the cortical hierarchy. In addition, the DMN plays a critical role in detailed experiences in working memory processes during memory retrieval, and is involved in the level of information specificity, or the level of information detail. All of these characteristics are fundamental to the creative process.
- The DMN integrates subcortical components that allow it to develop a complex activity. It has recently been discovered that DMN integrates the basal forebrain and the anterior and middorsal thalamus, and this implies certain additional functional characteristics: - Allows it to be involved in memory processes, since both the anterior and mediodorsal thalamus, as well as the basal forebrain, constitute connection points of the limbic system.
- The basal forebrain produces acetylcholine, which has an important neuropharmacological effect on memory processes.
- DMN is associated with the mesolimbic dopaminergic pathway through its connection with the ventromedial prefrontal cortex (VMPFC) and the midbrain (MidB), which allows you to engage in emotional modulation.
- The basal forebrain and thalamus can act as a relay at the subcortical level, integrating primary functional networks and brainstem inputs with high-level associative networks. This explains why the activity of the DMN "seems to arise from within us", since it is related to the most intimate self-referential activity of our self (specifically the MPFC).
- Finally, it has been discovered that DMN collaborates with other regions of the brain, enabling several cognitive functions that are essential for creativity. These cognitive structures include: - The theory of blind variation and selective retention (BVSR), which indicates that the DMN could generate an uncontrolled process of information generation (blind variation) based on random conceptual combinations, and the ECN could carry out a controlled process of evaluation of the information generated (selective retention).
- The deactivation failure hypothesis, which suggests that high creative ability may be based on a “deactivation failure” of certain regions of the DMN during tasks that require external focused attention. Highly creative people have been shown to be unable to suppress activity in the precuneus (DMN region) when performing working memory tasks (in which the ECN is involved). This suggests that the creative capacity, as previously stated, benefits from the coactivation of regions of the ECN and the DMN.
- The flexible control hypothesis, which suggests the existence of a greater coupling between the ECN and the DMN throughout the creative process, reinforcing the idea that the creative process requires flexible cognitive control.
In summary, the DMN plays a fundamental role in high-level cognitive processes and especially in human creativity. For this reason, and based on the new functional paradigm of the DMN, supported by new experimental discoveries, it has been proposed to change its name to DCN (Deep cognitive network).
Chapter 5.
Identification and analysis of the neurocognitive bases of human creativity In this chapter, and complementing the fundamental neurocognitive factors previously identified, the general set of neurocognitive factors that underpin creative processes in the human brain has been identified.
These factors have been identified using different and complementary strategies: - Analysis of recent discoveries on the structure and functionality of the human brain.
- Results of specific experiments related to the identification of brain activity when creative activities are carried out (usually divergent thinking tests). In these studies, certain neurocognitive patterns have been detected at the time of performing creative activities (for example, the activation of certain brain areas, such as the insula or the angular gyrus, or of certain large-scale networks, such as the DMN or ECN).
- Results of specific experiments not directly related to creativity, but that have a direct relationship with it (for example, dopamine D2 receptors in the thalamus have been studied in a very specific way, without suspecting that they could be related to human creativity).
- Parallels detected between neurocognitive research of a general nature, with the behavior observed in highly creative people, or with the conceptual and methodological structure of creative processes (for example, the hierarchical structure of the brain, based on linked sequences of top-down and bottom up processes, facilitates top-down control of bottom-up emerging thoughts, when performing creative tasks).
As a result of these strategies, 17 neurocognitive factors have been identified that support human creativity: - The structure of the human cognitive system is based on a set of large-scale neural networks, which integrate different regions (specialized in certain high-level cognitive processes) strategically located throughout the cortex.
- Multimodal organization of the brain that allows information flows, transforming it into various levels of abstraction.
- Strategic topographic distribution of large-scale networks and their constituent nodes (located at the maximum distance from the sensorimotor nodes), allowing the existence of abstract thoughts and the creation of self-generated thoughts.
- Strategic collaboration between the Executive control network (ECN), the Salience network (SN) and the Default mode network (DMN), creating the central mechanism of human creativity (Triple network model).
- The DMN is located at one end of the main gradient, it handles information at a higher level of abstraction, and it has a very complex functional pattern. The DMN is involved in a large number of cognitive processes related to creativity, and plays a fundamental role in the processes of generating and evaluating ideas.
- The medial temporal lobe (MTL) plays a fundamental role in the creation of self-generated thoughts.
- The mechanism of Task Induced Deactivation (TID) can have operational failures, and this allows the mixture of uncontrolled information that is manipulated separately by the ECN and the DMN. The variability of this mechanism is directly related to the working memory capacity (WMC).
- The insula (and the SN in general) partly control the functionality of the DMN and the ECN, when performing creative tasks.
- The angular gyrus is directly involved in the integration of some neurocognitive processes related to creativity. The left AG plays an important role in human creativity because it is part of two networks related to human creativity (semantic network and DMN). Creative people also have high connectivity between the DMN and the inferior frontal gyrus, which is interpreted as a high capacity to exert top-down control over DMN-generated imaginative processes.
- Memory mechanisms are fuzzy, and do not allow information to be retrieved accurately. The information generated in a certain experience is stored in a filtered way in the memory, and also, when you want to retrieve it, it has been transformed, and this helps to generate creative processes.
- Episodic memory facilitates creative processes, since it supports semantic memory, and it has been proven that under a specific episodic induction process, creative activity increases, since it facilitates traversing the search tree for solutions through alternative paths.
- Semantic memory is essential for the creative process. Concepts are created through a cross-modal structure that integrates and refines the "raw information" generated by the sensorimotor and verbal regions, and transforms it into coherent, complete, and refined cross-modal semantic representations. In this way, the information is structured at various levels of abstraction and with a large number of attributes, and this is a basic condition to allow the creative process.
- The emotional mechanisms of the brain are capable of representing and storing entities, objects or concepts in a richer and more detailed way. The availability of a greater amount of structured information at various levels of abstraction, and with a greater number of relationships, allows traversing the search tree through more varied paths, being able to reach very creative solutions.
- Motivation increases the robustness, integrity and overall efficiency of large-scale neural networks, and therefore also enhances the neurocognitive processes involved in human creativity.
- Changes in the information filtering neuronal system of the thalamus, based on the alteration of D2 receptors, allow a greater amount of information to be processed. The availability of a greater amount of information (in some cases of medium or low relevance) to be processed makes it possible to traverse the search tree for solutions through unusual paths.
- Large-scale networks can dynamically cooperate, overlap, and reconfigure themselves, allowing unexpected paths to be traversed through the search tree.
- The hierarchical structure of the brain, based on linked sequences of top-down and bottom-up processes, facilitates top-down control of bottom-up emerging thoughts when performing creative tasks.
Chapter 6.
Neurocognitive paradigm of human creativity The ordered and intertwined set of the neurocognitive bases of human creativity constitutes a neurocognitive paradigm of human creativity. This paradigm shows that the creative process is much more complex, flexible and effective than can be deduced from analyzing the current conceptual models of the creative process.
In fact, in this doctoral thesis, and based on this paradigm, a neurocognitive model of the creative process has been proposed, which restructures, completes and details all the conceptual models proposed so far.
This neurocognitive model of the creative process has great utility in four complementary fields of research: 1. It provides a general framework within which future research on the neurocognitive bases of human creativity can be focused. This framework shows both the most important neurocognitive factors and the relationships between them. In fact, certain specific experiments have been suggested, which would allow completing and detailing the proposed neurocognitive model.
2. It allows improving and complementing the conceptual models on the creative process made up to now.
3. Provides a conceptual framework in which current methodologies to enhance human creativity can be evaluated and improved, as well as new, more efficient methodologies designed. In fact, new methodologies have been proposed to amplify human creativity.
4. It allows implementing educational strategies in order to increase human creativity. In fact, some have been proposed, among which the following stand out: - Promote general knowledge of any discipline and in-depth knowledge of preferred disciplines.
- Learn to detect parallels and common patterns between the different fields of knowledge.
- Avoid premature specialization, before having adequate generalized training.
- Promote abstract learning about the common components between the different fields of knowledge.
- Encourage hard work and thorough documentation of the problem to be solved.
- Foster motivation, even obsession, to solve a problem.
- Increase the level of self-demand.
- Force periods of mental relaxation after long periods of hard and obsessive work.
- Increase the number of personal experiences.
- Do not accept any social rules.
- Question and analyze any type of information.
- Have strong emotional experiences (both positive and negative).
- Mix the information of the problem with the sensory information of the environment.
- Avoid taking drugs that inhibit any creative neurocognitive process (for example, avoid taking Ritalin, since by enhancing attention, working memory capacity is increased, and this induces a decrease in the activity of creative neurocognitive processes).
In this Doctoral Thesis different practical exercises have been proposed to stimulate the creative capacity of children.
5. Finally, knowledge of the neurocognitive bases of human creativity opens new paths to stimulate it with the help of chemical substances, bionic systems, and above all artificial intelligence.
Several proposals have been made, and among them the creation of "increased creativity systems" stands out, which are electronic devices connected by sensors to the human brain, which allow the excitation of certain brain areas and the inhibition of others, in order to stimulate creative creativity of the brain. These systems would also be connected to computer systems to increase efficiency in the generation and evaluation of ideas, having at each moment a gigantic amount of relevant information.
Chapter 7.
Computational paradigm of creativity based on the human neurocognitive structure Analyzing the different neurocognitive bases that support human creativity, computational parallels have been established and different suggestions have been made for the design of a creative computational system. By grouping these suggestions, it has been possible to make an adequate estimate of the comparative importance of each one, and its relative particular contribution in the overall design of the computational system. The suggestions are of various types, and correspond to different aspects of the computational system (system structure, operation, limitations, information structure, information manipulation strategies, etc.).
As a result, a multi-agent computing system (MAS) has been designed. This system is based on three groups of agents that interact with each other in various ways, based on the type of problem to be solved. The different agents of each group can assume different roles, depending on the design stage in which the system is. The system includes an evaluation system, with variable criteria, initially established by the users, but that the system can change, by itself, throughout its operation. Similarly, a knowledge representation system has been proposed that facilitates the generation of creative ideas.
It should be noted that the designed multi-agent system does not intend to emulate human creative capacity. Instead, what has been done is the design of a creative computational system based on the human neurocognitive structure. The operation of this system may provide similar results to the human brain on some cases, although on other cases it could lead to very different results. Nevertheless, the design of this computational system is only a starting point, capable of inspiring the implementation of more evolved future systems, with a computational structure capable of evolving by itself, moving away from the human creative neurocognitive structure.
The results and content of this doctoral thesis may be of special importance for neurobiologists, since it allows them to have a global vision of the neurocognitive bases of human creativity, and this allows them to establish new research strategies in order to effectively complete studies on the neurocognitive bases of human creativity. It is also important for psychologists and educators since it allows the implementation of different strategies to develop human creative capacity. It is important for bionic engineers, as it provides the basis for the implementation of new bionic technologies, capable of increasing human creativity. It is important for the pharmaceutical industry, since it provides important clues for the creation of chemical substances capable of increasing human creativity. Finally, it is also important for artificial intelligence researchers, as it allows them to develop new paths of research, and implementation, of new creative computational systems.
