No guidance, no memory
The null baseline: movement without the directed policy or history-dependent updates.
Material Memory in Physical Collectives
Jolt simulations in which materials update a local memory, compared against memory-free and damping-only controls.
When the push stops,
what does the material remember?
A connected assembly can respond to a force because of its current shape, its motion, or changes left by earlier interactions. We use a rigid-body simulation to separate those possibilities.
Local rules update friction, stiffness, and plasticity from interaction history. The experiment asks whether those changes leave a persistent trace, and whether that trace helps or hinders the next response.
Apply a temporary driving force. Record how the bodies move and how their local material properties change.
A lasting offset is only the start of the question
A system can keep moving or settle slowly without storing a useful trace of its history. Every campaign therefore pairs the memory rule with controls that separate changes from interaction history from guidance and extra damping.
Guidance alone
Keep the directed policy, but turn material memory off. This isolates what guidance already accomplishes.
Guidance with memory
Enable local changes from interaction history. This is the treatment being tested.
Damping control
Retain the directed policy with the inertial control. This tests whether apparent retention can be explained by extra damping.
Three ways to ask what remains
After a pulse
Remove guidance and measure the remaining displacement: does the assembly retain an imprint?
After a reversal
Ramp the drive back and forth. Does the response depend on the path taken to the current input?
After damage
Disturb an imprinted assembly and let it continue. Does stored history help it recover, or hold it in an unhelpful configuration?
The dossier currently presents the protocol and implementation. A results page needs the recorded trajectories and control comparisons alongside the conclusions.