The Origin of Life and Intelligence: Mutual Falling in Complex Systems
I am conducting personal research on the origin of life from a systems perspective.
In previous articles, I have discussed the principles of self-organizing systems from the perspective of mutual falling and the three elements of complex systems.
In this article, I aim to reorganize these principles and theoretically bridge them to the formation of life and intelligence.
Mutual Falling
Mutual falling refers to a process in which multiple parameters adjust over time to achieve energy stability.
The term “falling” is used, inspired by the way two objects attract each other through gravity. However, the force at play is not limited to gravity.
This concept does not only explain physical processes such as star formation or atomic formation but also chemical processes like molecular synthesis.
Mutually falling parameters possess homeostasis, meaning they restore themselves over time even when acted upon by external forces.
Furthermore, a specific mutual falling state can be maintained as long as energy or other substances are continuously supplied from the outside. This is a property akin to metabolism.
Additionally, a stable state achieved through mutual falling can influence its surroundings, akin to catalytic action.
In some cases, this influence spreads similar states around it, which can be seen as autocatalysis or self-replication.
Thus, even in a simple and low-precision form, mutual falling can exhibit characteristics resembling life.
From this perspective, it can be considered that a biological entity is a state where numerous mutual falling processes are intricately integrated. In other words, life can be viewed as a process composed of mutual falling.
The Three Elements of Complex Systems
Science struggles to understand complex systems because it can only analyze them using models broken down into parameters and static equations.
To analyze complex systems, it is necessary to decompose them into three elements: links between parameters, iterative processes, and conditional branching.
This becomes evident when considering computer systems or simulation models.
In programming, links between parameters or objects model data structures. Iterative processes are represented by “for” loops, and conditional branching by “if” statements.
The parameters and static equations used in scientific analysis represent only partial aspects of these data and processing structures.
Computer systems and simulation models are built upon these data and processing structures.
Thus, with these elements, it is possible to represent and analyze complex systems.
Mutual Falling and the Three Elements of Complex Systems
Mutual falling integrates the three elements of complex systems: links, iterative processes, and conditional branching.
The “mutual” aspect of mutual falling represents the links themselves. The process of falling embodies iterative processing. Moreover, the phenomenon where falling stops upon getting too close signifies conditional branching.
Therefore, identifying and incorporating mutual falling into analysis should enable a better understanding of complex systems.
Mutual Falling Networks
When there are multiple parameters and mutual falling relationships exist between them, a mutual falling network is formed.
Through the process of mutual falling, the structure of a mutual falling network evolves to achieve energy stability.
Evolutionary Adaptation of Mutual Falling
The structure of a mutual falling network affects its response to external inputs.
In some cases, the structure of the mutual falling network changes in response to external inputs. Even then, the result is an evolution towards energy stability.
When similar external inputs are repeatedly applied, the structure of the mutual falling network eventually stabilizes in a form that adapts to those input patterns.
In this case, the network as a whole, including the external input, settles into a structure that is energetically stable.
This can be considered as the mutual falling network evolving adaptively.
Learning-Based Adaptation of Mutual Falling
Mutual falling involves several coefficients, such as the speed and strength of falling or thresholds for conditional branching.
These coefficients may change in response to external inputs. When similar inputs are repeatedly applied, the coefficients stabilize to adapt to those input patterns.
In this process, the coefficients stabilize into an energetically stable configuration, taking into account the external input.
This indicates that mutual falling has adapted in a learning-based manner to external input patterns.
Learning-based adaptation is possible not only for a single mutual falling process but also for the entire mutual falling network.
Internal Adaptation and External Adaptation
In summary, when the structure or coefficients of a mutual falling network change flexibly, the network may adapt to external input patterns in the forms of evolution or learning.
Adaptation implies achieving energy stability, including the external input.
Adaptation where the mutual falling network changes in response to external inputs is termed internal adaptation.
On the other hand, an adapted mutual falling network may eventually fix its structure or coefficients, marking the establishment of evolution or learning.
Once fixed, the network may respond to changes in external input patterns without altering its structure or coefficients.
In such cases, the network may influence external inputs, potentially reverting them to their original patterns.
This type of adaptation, where external input patterns are adjusted to maintain an energetically stable state, can be referred to as external adaptation.
Passive Adaptation and Active Adaptation
Adaptation triggered by external inputs, involving changes in structure, coefficients, or external inputs, is termed passive adaptation.
Conversely, creating an energetically stable state through spontaneous changes in structure, coefficients, or external inputs, without being triggered by external inputs, can be termed active adaptation.
Synchronization Phenomenon of Pendulums
This principle is exemplified in the synchronization phenomenon observed in two pendulums suspended from the same bar.
From the perspective of Pendulum A, the vibrations it receives from Pendulum B are external inputs. Pendulum A internally adapts by adjusting its coefficients, such as its amplitude and phase, to these external inputs. Simultaneously, Pendulum A influences Pendulum B by imparting vibrations, thereby causing external adaptation in Pendulum B.
Furthermore, Pendulum A is not merely passively adapting; its own oscillations serve as a trigger, actively adapting both its own oscillation and that of Pendulum B.
Life and Intelligence
A highly adapted mutual falling network, developed through learning and evolution as forms of passive adaptation, eventually achieves self-reinforcing learning and evolution through active adaptation.
This process aptly explains the properties of life and intelligence.
Moreover, the same principle operates in the synchronization of two pendulums.
In other words, from the perspective of adaptation through mutual interaction between a mutual falling network and the external environment, the properties of life and intelligence can be explained.
The properties of life and intelligence can also be described as emerging from simple mechanisms such as the mutual falling of two objects, changes and stabilization of structures and coefficients, and pendulum synchronization.
In Conclusion: Decomposing into Elements of Complex Systems
The adaptation of a mutual falling network can be fully represented by incorporating the previously discussed three elements of complex systems, in addition to parameters and static equations.
This representation involves a combination of parameters, static equations, links, iterative processes, and conditional branching.
While it does not yet fully replicate the complexity of actual life or human intelligence, algorithms like the Game of Life and neural networks used in artificial intelligence are essentially simple combinations of parameters, static equations, links, iterative processes, and conditional branching.
These examples demonstrate how extremely complex phenomena can be decomposed into simple components by considering the three elements of complex systems.
Moreover, they strongly suggest that the properties of life and intelligence can be sufficiently explained from the perspective of adaptation within mutual falling networks.
