You walked into the kitchen with a clear plan. By the time you crossed the threshold, the plan was gone. This is the doorway effect, and it is not a quirk of your brain breaking. It is a side effect of how the brain organizes experience into discrete events. Crossing a doorway is a strong cue that one event has ended and another has begun, and information bound to the previous event becomes harder to retrieve once you are inside the new one.
The first formal demonstration was Gabriel Radvansky and David Copeland's 2006 Memory & Cognition paper, in which participants carrying virtual objects between rooms remembered the objects worse than participants carrying the same objects the same distance within a single room. The effect has been replicated in immersive virtual reality, in real walks through real buildings, and across age groups. It is not subtle, and it is not folk wisdom.
The short answer: Your brain treats each room as a separate "event." Crossing the doorway closes the previous event in working memory, and the goal you were holding ("get scissors") is now stored across an event boundary instead of inside the active model. The retrieval cue you were using disappeared at the threshold.
What is actually happening
Cognitive science has converged on event segmentation theory, formalized by Jeff Zacks and colleagues, as the best account. The brain does not experience life as a continuous stream of seconds. It carves activity into discrete, meaningful events, with boundaries triggered when prediction breaks down: a scene shift, a goal change, a physical transition. The deeper background on how working memory holds and loses information lives in memory 101.
Radvansky's event horizon model applies this directly to space. Each event is held in a working "event model" that includes the goal, the location, the people present, and the relevant objects. When you cross a threshold, the brain takes that as a cue to close the current event and open a new one. The information bound to the prior model, including what you walked in here for, now sits across a boundary rather than inside the currently active model. It hasn't been deleted. It is harder to reach.
Pettijohn and Radvansky's 2016 follow-up showed the deficit reflects interference from the new event model competing with the old one, not a momentary updating glitch. That is why returning to the original room sometimes restores the memory and sometimes does not. You are reinstating cues, but you are also creating yet another event boundary.
Is the effect always there?
Not always, and the nuance matters. The most important caveat comes from McFadyen and colleagues' 2021 multimodal study in BMC Psychology, which ran four experiments in immersive VR, in passive video, and in real-world walks. Without a concurrent working-memory load, they found no doorway effect. Only when participants had to count backwards while walking did the effect appear, and even then it showed up as more false alarms rather than fewer correct recalls.
The honest summary: doorways disrupt memory most reliably when cognitive load is already high. That fits the everyday experience perfectly. People do not usually walk into the kitchen meditating on the goal "scissors." They are holding the goal while navigating, while thinking about a conversation, while tracking a child or a phone notification. The boundary lands on a system already at capacity.
What helps
There is no peer-reviewed clinical fix because there is nothing pathological to fix. But the underlying science suggests a few low-stakes habits.
- Verbalize the goal before you move. Saying "scissors, scissors" out loud, or even subvocally, rebuilds the goal as a stronger phonological trace that survives the boundary. It also forces deeper encoding in the first place.
- Reduce concurrent load during transitions. This is the single most evidence-supported lever. If you finish reading the email, then get the scissors, the doorway has very little to do.
- Keep the target in hand or in view. A grocery list works for the same reason: it externalizes the working-memory load so the boundary can't disrupt it.
- Retrace your steps. Returning to the original location reinstates the event model that held the goal. It works inconsistently, per Pettijohn and Radvansky, but it costs almost nothing to try.
For broader context on how working memory and consolidation interact, see BrightYears' brain health guide, our piece on sleep and memory consolidation, and why brief daily sessions beat long ones.
When to see a doctor
This is the part that matters. Forgetting why you walked into a room is normal, even frequent, in healthy adults of every age. It is not a sign of dementia. The patterns that warrant a clinician's evaluation are different:
- You're losing words for objects you use daily, repeatedly, for longer than a few weeks.
- Family members are noticing memory changes that you aren't.
- You're getting lost in familiar places.
- Memory loss is sudden, not gradual.
- Lapses are accompanied by confusion, personality change, or trouble with sequenced tasks.
If any of those describe you, talk to your primary-care physician. The Lancet Commission's 2024 report identified 14 modifiable dementia risk factors that can shape your trajectory, and early evaluation matters. Walking into the kitchen and forgetting the scissors does not.