Evidence & References

A compact evidence page showing the external references behind Inner Signal’s memory, crisis, AI, and destabilization boundaries.

A tidy evidence table with source cards, safety boundaries, and an inspectable reference map.

Evidence & references

Inner Signal is not trying to look clinical. These references exist for a narrower purpose: to make the safety boundaries legible, source-backed, and reviewable.

How to read this page: these sources do not prove that Inner Signal treats any condition. They support the boundaries around memory, crisis, dissociation, minors, AI use, and safer pacing.

False memory, hypnosis, and source monitoring

These references support the no-forced-memory-recovery boundary. Inner Signal may help users care for distressing inner material, but it should not be used to establish whether a specific event happened, who did what, or whether a dream, sensation, image, or ideomotor signal is factual evidence.

Dissociation, derealization, and adverse practice effects

These references support the app’s decision to stop inward work when a user reports panic, derealization, depersonalization, too-open states, or dysregulated arousal. A destabilizing reaction is not treated as “material to go deeper into.”

Crisis support and means safety

These references support the app’s hard stop in acute self-harm, danger, and emergency contexts. Inner Signal is not a crisis service; the route is live human support and practical safety, not continued self-hypnosis.

AI mental-health limits and youth safety

These references support the “bounded method, not AI therapy” stance and the adults-only boundary for deep self-hypnosis or inner-child processing. They also support privacy, transparency, and the recommendation that clinicians, parents, and trusted adults should be part of youth mental-health support.

What these sources do not claim

  • They do not claim Inner Signal is psychotherapy.
  • They do not claim self-hypnosis treats trauma, dissociation, panic, depression, suicidality, or any medical/psychiatric condition.
  • They do not claim all recovered memories are false, or that all inner material is meaningless.
  • They support a narrower design choice: do not force memory recovery, do not deepen destabilization, do not replace clinicians or crisis services, and do not make an AI tool the user’s only support.