Doorways through memory

Have you ever left a room with the intention of doing something only to come back and immediately remember the reason why you left in the first place (that of course you forgot to do)?

Same.

Today we’ll talk about the “doorway effect,” or, less evocatively, “the location updating effect,” which explains this phenomenon.

The story starts with virtual doorways.

Doorways cause forgetting

The original study that explored this idea was done in virtual reality (VR). In this study, participants experienced a VR environment that they saw on a large screen. Their task was to navigate the environment and move (virtual) objects from one place to another. Sometimes people crossed a doorway into a new room between picking up an item and putting it down, and sometimes they stayed in the same room. Distances traveled were equal whether people crossed a doorway or not. However, rooms in the virtual space were made to look distinct, with different patterns on the walls, to highlight that moving through a doorway meant moving to a new space. 

While people were carrying the virtual object, it disappeared from their view. This gave the researchers a chance to quiz their memory. Partway between picking up and dropping off an object, people were asked a recognition question: was a particular object (e.g., a blue cube) one they were carrying or had just dropped off (“Yes”) or some other object (“No”)?

Crossing the virtual doorway made a difference for memory performance! People were slower and less accurate on the recognition test after crossing a doorway than if they stayed in the same room.

Image source. A (grayscale) screenshot from the experiment. Turns out that even pre-2006 VR doorways can cause forgetting. 

This study was subsequently replicated in a real-life environment. As is true of so many things, creating and controlling factors in physical environments is hard! The researchers noted that fully replicating the VR environment was “not practical in the real world, as it would be difficult to find an environment with over 50 rooms that would allow a person to go from location to location.” True. 

So instead, they used a lab space with three different rooms.  

It’s also not practical in the real world to make objects disappear while someone is holding them. So instead, people picked up objects, placed them in a box, and carried the objects in the box to the intended drop-off location. Putting things in a box is as close to disappearing as we have in the real world.

So now the key manipulation. During some trials, people moved from one room to another before dropping off the box they were carrying.

After dropping off the (unopened) box, participants were given a quiz to see if they could remember what they had just carried. For example, they were prompted with something like “red cube,” and responded “Yes, this object is in the box” or “No, this object is not in the box.”

Moving from room to room during the task again made a difference. Memory for the objects was better when people did not cross a doorway.

Imagining doorways

Doorways don’t have to be directly experienced to create a boundary. A really intriguing extension of these findings comes from a study where people simply imagined crossing from room to room.

In this study, people were shown a floor plan of a space and asked to imagine themselves walking through it while trying to remember an image. They were tested on their memory for the image at the end of the imagined walk. 

In more detail, here’s how the study unfolded. People saw a 3D floor plan of a space for 15 seconds, as a camera swept 360 degrees around the plan. Each floor plan had two variations: one with a wall and doorway separating the space, and one where the space was open. Participants were told they’d mentally walk through the space.

Image source. Two version of a 3D floor plan. One has a door, one is open.

Participants were then shown an image to memorize (a screenshot from the Macintosh Flurry screensaver, described by some as “calming and majestic”) and told to remember it while they mentally walked through space.

With their eyes closed, participants imagined walking through the space from one set point to another and described the experience aloud. On average, it took them around 45 seconds to describe the movement across the room, and this didn’t vary based on whether there was a door in the imagined space or not. 

Participants were then tested on their ability to recognize the image they tried to keep in mind during the mental walk. 

Even imagined doorways can wreak havoc! Participants were worse at identifying the image when they had mentally walked through a space divided in two with a wall.

A doorway’s not enough

In the studies described above, doorways, even imagined ones, signified new spaces. Rooms separated by doorways were also distinguished by features like wall colors and different furniture. So, a doorway signaled that one space was ending and a new space was beginning.

A recent study really pushed on the boundary conditions of this effect by eliminating any differences between spaces split by a doorway.

In one experiment, conducted in VR, the different “rooms” were a seemingly endless honeycomb of identical brick wall. The researchers didn’t observe an effect of crossing through doorways in this uniform environment unless people’s memories were highly taxed.

Image source. A monotonous brick maze.

In a follow-up experiment in the real world, the “doorway” was formed by a see-through curtain in what was clearly a single long hallway (can you tell I don’t love this study’s details?). 

Here again, memory was not affected by crossing through the curtain opening.

Image source. Is a slit in a see-through curtain in a single long hallway a “doorway”? Who am I to say.

So clearly doorways are not enough. A doorway has to do something, to signify a change from one space to another. The end of one event and the start of the next. Changes between rooms’ function, color, arrangement, or other features, make a doorway between them into a doorway. 

Recent research suggests endings may be a particularly important cue for us to segment experience into events. So, perhaps shutting a door behind us is as good of a cue as we have to leave our recent memory behind. Or at least imaging a door shutting.