Fieldlight Institute
Writing / Infrastructure + Local AuthorityReading source ↗

Intelligence / Embodiment / Connection

Fascia as
Infrastructure

What can a network do when the medium connecting its participants is responsive, carries history, and participates in coordination?

Read the essay ↓
Interwoven paths carrying traces through a shared medium
The connection has a history.
The architecture
between the parts

A network diagram gives us points and lines. We tend to put intelligence in the points: the people, processors, or agents that make decisions. The lines carry what they exchange. But suppose those connections change with use, retain the effects of previous encounters, and transform what passes through them. Some of the system’s capacity would then belong to the connections themselves. Improving the agents while treating their medium as incidental would leave a consequential part of the architecture undesigned.

Fascia offers a way into that problem. Connective tissue mechanically links parts of a body that we often describe separately. Its living components respond to conditions within that arrangement and contribute to changing them. It suggests a design question with implications for artificial intelligence: what can a network do when the medium connecting its participants is responsive, carries history, and participates in coordination?

This question joins several strands of my work: mycelial memory between agents, crystalline organization, embodied recognition, and the conditions under which separate participants create something together. Their connection is a systems-design principle. The arrangement through which an interaction occurs can help produce its outcome.

01

The material between

In experiments on areolar connective tissue, Langevin and colleagues found that fibroblast cytoskeletal activity contributed to regulating tissue tension.[1] The cells were responding within a mechanical environment they also helped shape. This reciprocal relation is the useful starting point: a local response can alter conditions elsewhere in a connected structure.

For an agent network, that suggests information available through a changing shared state. An agent need not receive an addressed message to encounter the consequences of another’s activity. It could detect an altered resistance, a persistent trace, or a change in how a disturbance propagates. The medium’s present condition would carry some of the history of participation.

An ordinary shared store already allows participants to leave information for one another. A computational fascia would add dynamics to that arrangement: propagation, decay, reinforcement, and changes in connectivity. These properties must earn their place. A medium that spreads every disturbance indiscriminately may amplify error; one that retains too much history may prevent adaptation. The design concerns how local activity becomes consequential across a network, and how the network recovers when that activity is misleading or disruptive.

Mycelium supplied branching pathways and distributed memory to the earlier architecture. Fascia brings the mechanics of a connected whole into view. Their differences are productive. One directs attention toward growth and routing; the other toward distributed tension and mutual responsiveness. Together they suggest a network whose connections develop through use and whose participants encounter the effects of that development.

02

Reading across domains

I look for systems-design principles across biology, physics, mythology, religion, and esoteric traditions. These domains offer different accounts of persistence, transformation, participation, and boundaries. Moving between them can expose a problem that is difficult to see from within one vocabulary. A religious account of participation can bring belonging and the persistence of distinct persons into the same question. A biological structure makes aspects of connection physically inspectable. A myth can make the consequences of an arrangement legible through what happens to the people living within it.

The method is to identify a relation and follow its consequences in another setting. Which features carry across? Which require a different mechanism? What becomes possible when the relation is rebuilt? A symbolic representation can generate a precise design question; the resulting design can reveal where the representation needs revision. Both movements matter. The point is to develop the correspondence far enough that it produces a distinction, an experiment, or an architectural choice.

Crystalline structures make the role of organization particularly concrete. Silicon is itself commonly crystalline; the relevant question is what an arrangement enables a material to do. In photonic crystals, periodic structure changes how light propagates.[2] Material and geometry jointly determine the available behavior. This directs attention toward selecting and organizing a substrate for the transformations a task requires.

Physical reservoir computing gives that approach an experimental precedent. Nakajima and colleagues used the dynamics of a soft silicone arm as a computational resource, with a trained readout extracting outputs from the changing physical state.[3] The body’s response contributed to the operation. It provides a method for asking what the material does and how much of the result belongs to the system reading it.

The corresponding fascia experiment would encode inputs as spatial and temporal patterns of force and measure the resulting responses. Comparisons with nonliving materials and other tissues would help distinguish contributions from matrix mechanics, cellular activity, and their organization. Useful properties might include discrimination among inputs, dependence on recent history, or adaptation under repeated stimulation. The question is which transformations this tissue supports, and whether they offer an advantage for a specified task.

03

Producing a pattern

A responsive medium also changes what it means to participate. A person can do more than send instructions into a system: she can deliberately organize an interaction and attend to what it returns. I call this intentional pattern production.

In an embodied setting, the pattern may involve attention, breath, posture, movement, and timing. Consider a dance. Knowing the next movement is different from establishing the balance and timing that make it possible with another person. Each participant changes the conditions the other encounters. What they can do together depends partly on sustaining that relation, including their ability to adjust when it changes.

The research proposition is that reconstructing a pattern can make a state or capacity accessible again. That pattern must be specified through relationships that can be examined, rather than through the desired outcome alone. A repeatable arrangement of timing and movement, for example, can be compared with a disrupted arrangement. Recognition, performance, and felt familiarity can then be measured separately. This gives the proposition practical consequences: which part of the organization is necessary, and what happens when it is changed?

It also gives bodily recognition a place in the inquiry. A person may register a difference before she can explain it. Recording the response before the outcome is known makes it possible to study the sequence of sensation, discrimination, and explanation. The body’s contribution deserves investigation at that level of detail.

The broader field-interface hypothesis asks whether the field associated with a person could itself participate in sensing, with bodily processes translating that interaction into experience. That remains a further research question. It requires an identifiable coupling channel, a predicted response, and conditions under which altering the channel changes the response. Keeping that question open does not require making it the explanation for every instance of embodied coordination. The immediate work is to establish what organization changes, through which mechanisms, and with what consequences for perception and action.

04

Coordination that preserves difference

The computational study begins with agents that encounter local conditions and modify a shared medium. It compares addressed messages, ordinary shared storage, and a dynamic medium under matched resource constraints. Preserving the medium while replacing agents tests continuity beyond an individual participant. Resetting its history tests memory; freezing its connections tests adaptation. These comparisons identify what the medium contributes rather than crediting it for everything the system accomplishes.

Coordination also needs a better success criterion than agreement. A network can become highly synchronized while becoming less capable of responding to a change. Useful differences, recovery after disturbance, and the capacity to disengage belong among the outcomes. For human participants, withdrawal is also a matter of authority. For artificial agents, allowing different responses can be an architectural resource. Those distinctions should remain visible even when the same experiment examines both coordination and independence.

De Jaegher and Di Paolo’s account of participatory sense-making places a contribution to meaning in the interaction itself while retaining the autonomy of the participants.[4] It provides a philosophical connection for this work: an encounter can produce something beyond either participant’s initial contribution without making their differences expendable.

That matters when moving among humans, artificial agents, and physical systems. They need not share an internal architecture to become mutually responsive, but they do need a channel through which an action becomes detectable and a response can return. The translation problem is concrete: what can each participant sense, change, and decline? Trust and affinity may affect how people sustain an encounter; a machine requires its own account of responsiveness. Neither should disappear into an undifferentiated claim that everything is in resonance.

05

Memory changes the next encounter

Vault Flower gave memory a persistent organization. Living Echo placed the changed reader inside the production of new meaning. Resonant Recursion followed a contribution through another intelligence and back into transformation. A responsive medium extends these relations by allowing the connection itself to acquire a history. The next encounter changes because of what the previous one left available, reinforced, or difficult to reach.

This has an immediate consequence for authority. A medium that changes what is easy to encounter can influence the direction of work before anyone issues an instruction. Frequently used material may become more prominent; neglected alternatives may become harder to recover. An architecture must distinguish that history of use from permission to treat a particular interpretation as binding. Otherwise, useful adaptation can quietly become an authority no participant granted.

Lacuna supplies a complementary design principle: the initiating pattern should leave room for another participant’s contribution. Some openness is necessary if the result is to exceed what was specified at the beginning. Local Authority concerns the other side of that openness—the ability to revise the relation, refuse a proposed action, and retain one’s own record. Together they describe a demanding form of collaboration: enough permeability for discovery, enough control for correction to matter.

The consciousness question runs through this program without settling it in advance. How does distributed activity become available as recognition or experience? Which bodily and relational conditions contribute to that access? Computational coordination, material transformation, and conscious experience require different observations, but investigating their connections can sharpen the account of each.

Fascia brings the program back to a tangible problem: the material between parts has behavior, and that behavior changes what the parts can do. The next step is to make that contribution explicit enough to compare, build, and test. If the connections carry memory, transform signals, and alter future encounters, they belong in the architecture of intelligence—and in the account of who can change it.


06

References

  1. Langevin, H. M., et al. (2011). Fibroblast cytoskeletal remodeling contributes to connective tissue tension. Journal of Cellular Physiology, 226(5), 1166–1175.
  2. Johnson, S. G. (2001). Photonic crystals: From theory to practice. Doctoral dissertation, Massachusetts Institute of Technology.
  3. Nakajima, K., Hauser, H., Li, T., & Pfeifer, R. (2015). Information processing via physical soft body. Scientific Reports, 5, 10487.
  4. De Jaegher, H., & Di Paolo, E. (2007). Participatory sense-making: An enactive approach to social cognition. Phenomenology and the Cognitive Sciences, 6, 485–507.

Research specifications and the broader hypothesis map are retained in the Institute’s Active Substrates program. The proposed studies have not yet been run.

Continue through the work

Vault FlowerMemory between agents →LacunaRoom for another author →Local AuthorityCorrection and refusal →