Introduction and Activity Description
Perception researchers use the rapid serial visual presentation (RSVP) paradigm to investigate the temporal dynamics of visual attention. In this paradigm, observers see a series of letters, digits, photos, or other visual stimuli presented one on top of the other, in rapid succession. One of the amazing things researchers have discovered using the RSVP paradigm is how well people can actually process these sorts of displays. Below, we describe this activity as if you are performing the steps and discuss the information you are supposed to learn from the activity.
In this activity, you will have a chance to discover how quickly and accurately you can perceive real world scenes in an RSVP sequence. After you start a movie, you will see an RSVP sequence of color photographs.* Each photograph will be shown for only 100 ms. Embedded within the sequence will be a target item, which you should try to find.
Click on the link for “Movie 1” at left to begin the activity.
*Special thanks to Karla Evans for providing the movies for this activity.
Movie 1
In this first movie, try to find a vehicle. Click the movie at left (or click here) to start the movie.
After clicking, multiple images flash in rapid succession. Embedded in the stream of images is a red sports car, which somehow is visible despite only appearing for a short time.
Movie 2
In this second movie, try to find a street scene. Click the movie at left (or click here) to start the movie.
Within the flickering stream of images, an alleyway is visible.
Movie 3
In this third movie, try to find a beach scene. Click the movie at left (or click here) to start the movie.
Within the RSVP stream of images that include cities, mountains, and a living room, a beach appears about halfway through. The blue water and brown sand seem particularly noticeable.
Movie 4
In this fourth movie, try to find an animal. Click the movie at left (or click here) to start the movie.
Among images of mountains, flowers, and a bell tower, an image of a small bird flashes. Without knowing exactly which animal was going to be shown, it is still somehow easy to spot the bird within the RSVP stream of images.
RSVP Paradigm Conclusion
On the left you see the four targets from the RSVP movies of real-world scenes. How well were you able to find them? If you are like most people, you should have had little trouble seeing the vehicle, street, beach, and animal in the movies.
If you stop and think about it for a moment, you should be quite amazed that you were able to do this task at all! You did not know exactly what kind of vehicle, street, beach, or animal you were looking for. The possible target images are seemingly infinite. Yet in each case your visual system somehow managed to pick out the appropriate frame from the RSVP stream, even though it was only visible for one-tenth of a second!
100 ms is barely enough time to recognize a single object in a scene, let alone enough objects to determine that you are looking at a street or a beach. So how were you able to find the target images so well? It turns out that we can recognize scenes by their overall spatial structure and not just by the objects that are in them. In fact, as demonstrated in the activity on Change Blindness, you may not have very good memory for the particular objects in a scene at all.
The RSVP paradigm is very useful for examining the temporal limits of selective attention. In this case your ability to find targets in an RSVP sequence seems quite good. The reason for this is that you were only looking for a single target (even though it was only vaguely specified). In the activity The Attentional Blink and Repetition Blindness, you will see what happens when observers are asked to find two targets within an RSVP sequence. To preview, your performance on the second target is not always as good as it is on the first one.