Integration Dynamics: Contraction, Expansion, and Capacity

Integration Dynamics: Contraction, Expansion, and Capacity
(A cross-scale model of how systems process change)

Paul Stevens
27^th March 2026

There’s a quiet frustration that sits beneath most conversations about change.

We assume that if something is seen clearly enough — if the right idea lands, if the right insight clicks — then change should follow. That understanding naturally leads to transformation. And when it doesn’t, we tend to turn inward: questioning motivation, discipline, or even our own sincerity.

But what if change was never about exposure in the first place? What if the real question isn’t what we encounter, but what we are able to hold when we do?

This piece circles that question through a simple lens — contraction, expansion, and capacity — not as a theory to adopt, but as a way of noticing.

A rhythm, perhaps.

The way we narrow to find stability, widen to meet something new, and slowly grow into the space between the two. Not always smoothly. Not always successfully. But often enough to reshape what feels possible.

And maybe that’s where change actually lives — not in what arrives, but in what can be stayed with, long enough to take root.

Downloadable .pdf option

I. Introduction — The Problem of Change

Systems encounter novelty continuously. New information, new conditions, and new interactions arise at every scale — from physical systems adjusting to environmental shifts, to individuals navigating changing relationships and internal states.

Yet exposure to novelty does not guarantee change.

In practice, most systems do not change, even when presented with viable alternatives. Some adapt quickly, incorporating new information into stable patterns. Others resist, returning to familiar structures even when those structures are no longer effective. Still others destabilise entirely, fragmenting under conditions they cannot process.

The same input can therefore produce entirely different outcomes.

This raises a more precise question:

What determines whether change is successfully integrated?

Existing approaches tend to describe either structure (how systems are organised) or behaviour (how they respond). Less attention is given to the conditions under which new input becomes part of a system’s ongoing operation — or fails to.

This paper proposes that change is governed by three interrelated dynamics: contraction, expansion, and capacity.

Rather than occurring as a linear progression, integration unfolds through cycles. Systems narrow to stabilise, widen to incorporate novelty, and, over time, increase their ability to process complexity. Crucially, these processes are not guaranteed to succeed. At any point, systems may reject, absorb, or fragment in response to the same conditions.

Understanding these dynamics helps explain why change can feel effortless in one moment and impossible in the next — not because the world has changed, but because the system encountering it has.

II. The Constraint — Why Change Is Not Automatic

It is often assumed that exposure to new information leads naturally to adaptation. In practice, this is not the case.

Systems do not change simply because something new appears. They change only when that novelty can be sustained long enough to alter existing structure.

New input introduces deviation — a direction not already contained within the system’s current organisation. This deviation creates the possibility of change, but not its guarantee.

Most deviation is not integrated. It is filtered.

When the difference is too great, too abrupt, or insufficiently supported, the system cannot stabilise around it. The input is ignored, resisted, or rapidly overwritten by existing patterns. The system returns to what it already knows.

This is not a failure of awareness or intention. It is a structural constraint.

Systems preserve coherence by default. Anything that cannot be incorporated without destabilisation is rejected.

For change to occur, deviation must not only be introduced, but held.

Even brief stabilisation can be sufficient. If the system can maintain coherence while accommodating the new pattern — however temporarily — reorganisation becomes possible. The structure may then extend to include what was previously external to it.

This leads to a more precise formulation:

Change requires not only new input, but the capacity to sustain deviation without loss of coherence.

Without this capacity, novelty does not transform the system. It is filtered out by it.

III. Contraction — Focused Integration

Contraction refers to a narrowing of scope in which processing becomes more focused and internally directed.

During contraction:

• attention concentrates on a limited set of variables
• existing pathways are reinforced
• noise is reduced
• precision increases

This mode enables systems to stabilise. By limiting the range of input, the system can refine its internal organisation, strengthening coherence and reducing internal conflict.

In this sense, contraction is not a withdrawal from change, but a condition for it.

Without sufficient contraction, systems cannot maintain the stability required to integrate new structure. Any introduced deviation would immediately destabilise the system and be rejected.

Contraction is therefore essential for:

• consolidating learning
• maintaining identity
• preserving functionality under load
• preparing the system to hold greater complexity

However, contraction also carries risks.

When prolonged or excessive, it can lead to:

• repetition of existing patterns
• rigidity
• reduced responsiveness to new conditions

In this state, the system becomes increasingly efficient at reproducing what it already knows, while losing the flexibility required to incorporate what it does not.

Contraction can therefore be understood as focused integration — a mode that stabilises and refines structure, but does not by itself extend it.

IV. Expansion — Distributed Integration

Expansion refers to a widening of scope in which processing becomes more distributed and open to variation.

During expansion:

• attention broadens
• new inputs are admitted
• connections form across previously separate domains
• patterns are reorganised

This mode enables systems to encounter and incorporate novelty. By increasing the range of input, the system can access structures and relationships that are not available within its existing organisation.

If contraction stabilises what is known, expansion introduces what is not.

Expansion is therefore essential for:

• learning
• adaptation
• creativity
• exploration

However, expansion is not inherently stabilising.

Without sufficient underlying coherence, increased exposure to variation does not lead to integration, but to instability. The system admits more deviation than it can hold, and coherence begins to break down.

When unregulated, expansion can lead to:

• overload
• fragmentation
• loss of stable reference

In this state, the system becomes sensitive to fluctuation, unable to sustain consistent structure across inputs. New patterns appear, but do not stabilise.

Expansion therefore depends on prior contraction.

The ability to widen scope without fragmentation is determined by the coherence already established. Systems that have not sufficiently stabilised cannot effectively integrate what they encounter.

Expansion can therefore be understood as distributed integration — a mode that introduces new structure, but requires sufficient stability to incorporate it.

V. Capacity — The Governing Variable

Contraction and expansion describe modes of processing. Capacity determines whether those modes result in stable integration.

Capacity can be defined as:

the system’s ability to integrate novelty without losing coherence.

It governs:

• how much deviation can be tolerated
• how long new input can be held
• whether new patterns stabilise or collapse

Two systems can encounter identical conditions and diverge completely based on capacity alone.

One may incorporate the new structure and extend. Another may reject it and return to prior patterns. A third may fragment under the same input.

The difference is not the novelty itself, but the system’s ability to hold it.

A system with low capacity:

• rejects novelty quickly
• relies on repetition
• becomes rigid under pressure

A system with high capacity:

• tolerates greater variation
• sustains deviation for longer
• reorganises more effectively

Capacity is not fixed. It changes over time.

It increases through successful integration. Each stabilised pattern expands the system’s ability to process future novelty. It decreases under load, fragmentation, or prolonged instability.

Capacity is also not purely individual.

In interacting systems, the ability to hold deviation can be distributed. Multiple systems, each maintaining coherence, can collectively stabilise patterns that would not be sustainable in isolation. In such cases, capacity emerges across the interaction, rather than residing in any single system.

This extends the scope of integration beyond the individual. What cannot be held alone may, under the right conditions, be held together.

This makes capacity the central variable in change dynamics.

Novelty introduces the possibility of change. Capacity determines whether that possibility is realised.

VI. The Cycle — Integration as Oscillation

Change does not occur in a single step. It emerges through cycles of contraction and expansion.

A typical sequence may be described as:

contraction → expansion → stabilisation

• Contraction builds internal coherence and precision
• Expansion introduces new inputs and relationships
• Stabilisation integrates these into a new structure

These phases are not independent. Each conditions the next.

Contraction increases the system’s ability to maintain coherence under load. This prepares it to encounter and hold greater deviation during expansion. Expansion then introduces new structure that challenges the system’s existing organisation. If that structure can be sustained, stabilisation extends the system’s capacity.

In this sense, expansion is often only possible because of prior contraction, and stabilisation is only possible if expansion has not exceeded capacity.

These dynamics form a continuous oscillation. Systems narrow to stabilise, widen to incorporate, and reorganise at a new level of coherence.

When this cycle is interrupted, predictable outcomes follow:

• Contraction without expansion leads to repetition and stagnation
• Expansion without contraction leads to overload and fragmentation

Growth therefore depends not on remaining in one mode, but on moving effectively between them.

Capacity increases through this process.

Each successful cycle — in which deviation is introduced, sustained, and stabilised — expands the system’s ability to process future novelty. Over time, this allows the system to operate across a wider range of conditions without loss of coherence.

What appears as sudden insight or rapid change is often the visible phase of this cycle — the moment at which previously accumulated structure allows a new pattern to stabilise.

The transition may feel abrupt. The process that enabled it is not.

Systems evolve not by avoiding instability, but by learning to stabilise it.

VII. Implications for Human Living

At the human level, these dynamics are experienced directly.

Periods of contraction may appear as:

• focus on routine
• repetition of familiar patterns
• reduced openness to new input

Periods of expansion may appear as:

• increased curiosity
• exposure to new environments or ideas
• shifts in perspective

Neither state is inherently preferable. Both serve necessary functions.

Difficulties arise when the balance is disrupted:

• excessive contraction leads to rigidity and stagnation
• excessive expansion leads to overwhelm and instability

Understanding this reframes a common assumption. The problem is not a lack of change, but a mismatch between exposure to novelty and the capacity to integrate it.

Managing this relationship becomes central.

Introducing change at a rate that can be sustained allows growth without fragmentation. This often requires regulating input — not avoiding novelty entirely, but shaping its timing, intensity, and duration.

Boundaries therefore play a functional role. By limiting or structuring exposure, systems maintain the conditions necessary for integration.

Relational environments are equally significant.

Systems do not stabilise in isolation. Interaction with others can either increase noise or reduce friction, directly affecting how much novelty can be processed. In this sense, presence and stability are not passive qualities. They alter the conditions under which both oneself and others can adapt.

Capacity can also emerge collectively.

A small number of highly coherent individuals, interacting consistently, can stabilise patterns that would not be maintainable in isolation. In such cases, the ability to hold deviation is distributed across the interaction. What cannot be integrated alone may, under the right conditions, be integrated together.

This helps explain why certain changes appear to spread rapidly once a threshold is reached. The underlying dynamics remain the same; what shifts is where capacity resides.

These dynamics also apply to prolonged instability.

Some patterns persist not because they cannot be resolved, but because they exceed the system’s current capacity to integrate them. In such cases, systems may cycle repeatedly around the same material, building partial structure over time. When capacity increases — through internal stabilisation, external support, or both — rapid reorganisation can occur.

What appears as sudden change is often the visible resolution of a longer process.

Taken together, these observations suggest a practical formulation:

Growth depends less on increasing exposure, and more on developing the capacity to sustain and integrate what is encountered.

VIII. Conclusion — From Change to Integration

Change is often treated as a function of exposure — the more a system encounters, the more it evolves.

This paper suggests otherwise.

Systems do not evolve simply by encountering more. They evolve by becoming able to integrate more.

Contraction, expansion, and capacity describe the dynamics through which this occurs. Together, they explain why change is uneven, why growth requires cycles of stability and disruption, and why the same conditions can produce entirely different outcomes depending on the state of the system.

These dynamics are not limited to any single domain. They apply across scales — from physical systems adjusting to environmental constraints, to biological systems adapting through feedback, to human systems navigating experience, relationships, and meaning.

In each case, the pattern holds:

New structure is encountered.
It is either rejected, destabilises the system, or becomes integrated.

The determining factor is not the novelty itself, but the system’s ability to hold it.

This reframes the problem of change.

Rather than asking how to increase exposure, the more useful question becomes:

What conditions allow a system to sustain and integrate what it encounters?

From this perspective, contraction is not avoidance, but preparation. Expansion is not progress in itself, but opportunity. Capacity is what determines whether either results in transformation.

This shifts the emphasis from forcing change to enabling integration.

Systems do not change when they see something new.
They change when they can remain coherent while seeing something new.

And over time:

Systems evolve not by encountering more, but by becoming able to hold more — within themselves, and, where necessary, together.