Where the idea comes from
The observation is nearly a century old. In 1927, Bluma Zeigarnik found that interrupted tasks were remembered roughly twice as well as completed ones — unfinished work stayed active in memory as an unresolved tension (Zeigarnik, 1927, English translation; history and replication, University of Waterloo). That is the Zeigarnik effect, and it is the load-bearing wall under the phrase "open loop."
Signal density and the freeze
Recent work formalizes what happens when open loops multiply. In a 2026 framework for what the author calls Open Loop Overwhelm, the mechanism is signal density: each unresolved commitment in the environment is a signal competing for salience, and past a threshold the salience-ranking system collapses — attention scans across competing loops instead of committing to one, and task initiation fails (Pines, 2026). The environment does the remembering. A desk covered in half-open states is not laziness; it is a distributed, overloaded external memory system.
The problem is not that you cannot start. It is that every possible start carries the debt of all the others.
Why "write it down" stops helping
If capturing every loop into a list solved everything, the planner graveyard would not exist. The research on prospective memory in ADHD shows the intended action itself — "at 4pm, take the medication" — is held by the same central-executive resources that are already overloaded, so intentions sit in the system most likely to drop them (Altgassen et al., 2014; mechanism summarized by The ADHD Desk). A list is external storage, which helps — but only if a second mechanism also exists to resurface the item at the right moment.
The evidence-backed fix: offloading, done actively
Cognitive offloading — using the body and environment to reduce cognitive demand — improves performance across memory, perception, and planning (Risko & Gilbert, 2016), and the authors note that people with impaired unaided cognition benefit most. For ADHD specifically, the strongest behavioral interventions put externalization first: a randomized trial of cognitive-behavioral therapy for adults with ADHD built its first modules around externalizing structure and follow-through, and the gains held at 12 months (Safren et al., 2010).
Two refinements make the difference between a list and a closed loop:
- Externalization of the cue, not just the item. The reminder to act must also live outside the brain — which is why implementation-intention encoding ("if X, then Y"), which offloads the trigger, measurably improves prospective memory performance in ADHD samples (medRxiv, 2026).
- A presence that notices silence. Interview research with adults with ADHD finds task management is socially scaffolded in practice — people lean on check-ins, body doubles, and a felt sense of accountability — while most apps treat the user as a solitary agent (arXiv, 2026).
The practical translation
- Contain, do not clear. Removing reminders without replacing tracking just lets the loops reaccumulate (Pines, 2026).
- Keep visibility, reduce density: one reliable external home per loop, not thirty surfaces.
- If the tracking requires you to remember to update it, it is still internal load wearing app clothes.
If you want the tool layer of this argument, score the candidates for yourself in the friction directory — proactivity is the column that separates a system from a partner.