The catch-up or fallback illusion: Why we accidentally slow-down people on the motorway and lose football matches

Movement and orientation in space is an everyday experience. But it is also a fundamental theoretical problem for science. How does a moving observer perceive the relative movement between two other persons or objects? Despite its everyday relevance, this "perceptual three-body problem" is largely unexplored. Dr. Tobias Meilinger from the department of Prof. Bülthoff at the Max Planck Institute (MPI) for Biological Cybernetics and Dr. Bärbel Garsoffky and Prof. Dr. Stephan Schwan from the Leibniz Institute for Knowledge Media (IWM) examined under which circumstances the perception of movement is distorted - and thus an illusion arises.


1. Short introductory paragraph explaining your study
Example: two people run after one another
Sometimes the observer also moves in addition to the people in the video.Sometimes the observer also moves in addition to the people in the video.

Now participants judge whether the persecutor is catching up with the persecuted. Here people do not give a correct judgement, but they are trapped into a movement illusion: If two people have the same distance and retreat from the observer (=video), a catch up illusion occurs: it looks like as if the pursuer is catching up. When the runners approach the observer the illusion reverses: it looks like the persecutor is falling behind. This new motion illusion has never been described before. Where does it come from?Now participants judge whether the persecutor is catching up with the persecuted. Here people do not give a correct judgement, but they are trapped into a movement illusion: If two people have the same distance and retreat from the observer (=video), a catch up illusion occurs: it looks like as if the pursuer is catching up. When the runners approach the observer the illusion reverses: it looks like the persecutor is falling behind. This new motion illusion has never been described before. Where does it come from?

There are different theories from the perception of two moving bodies that can be transferred to the present three moving bodies' case. For example, always keep the angle between the two runners constant, keep their relative distance (e. g. concealment) constant, keep the change in size constant, or determine the proper movement speeds and subtract them. None of these theories fit to the observed data. The theory that can explain the illusion is based on distance compression: the center lines of a road look shorter when they are further away. The distance between the runners is like a median strip: if the runners run away from the observer, the distance gets perceptually shorter and shorter and thus gives the impression of a relative movement. The magnitude of the measured movement distortion corresponds exactly to the predictions of this distance compression, but not to the existing theories. There are different theories from the perception of two moving bodies that can be transferred to the present three moving bodies' case. For example, always keep the angle between the two runners constant, keep their relative distance (e. g. concealment) constant, keep the change in size constant, or determine the proper movement speeds and subtract them. None of these theories fit to the observed data. The theory that can explain the illusion is based on distance compression: the center lines of a road look shorter when they are further away. The distance between the runners is like a median strip: if the runners run away from the observer, the distance gets perceptually shorter and shorter and thus gives the impression of a relative movement. The magnitude of the measured movement distortion corresponds exactly to the predictions of this distance compression, but not to the existing theories.

2. Why were you interested in this topic?
The original idea to this work came from Stephan Schwan. He is interested in movies and is examining implicit rules for camera movements. One of this rules is that if there is a wild chase and the camera from behind catches up, this psychologically suggests that the chaser will catch the chased one - and vice versa for the camera falling behind. We wanted to test whether we can find this tendency also with strict psychological measures. The original idea to this work came from Stephan Schwan. He is interested in movies and is examining implicit rules for camera movements. One of this rules is that if there is a wild chase and the camera from behind catches up, this psychologically suggests that the chaser will catch the chased one - and vice versa for the camera falling behind. We wanted to test whether we can find this tendency also with strict psychological measures.

3. What should the average person take away from your study?
How we see the world is not necessarily how the world really is. But how does the illusion affect everyday life? Let's take two examples of perception from driving a car and football games. On the motorway: I'm driving on the left-hand lane of the motorway. Two cars are driving on the right-hand lane. The question: Can I foresee that the rear one wants to change to my lane? My assessment is influenced by the fact that I drive fast myself. I don't notice that the rear car accelerates, because the fallback illusion hides that. The driver, however, starts to overtake, comes onto my track and slows me down. The distortion of her/his distance to the front car causes me to underestimate its true speed. A similar case on the football field: The striker from my team dashes to the opponent's goal, followed by an opposing player. I myself follow slower and want to play her/him the ball, but I overestimate the speed of the catching-up opponent due to the illusion and therefore I don't pass the ball on, missing the chance to score. How we see the world is not necessarily how the world really is. But how does the illusion affect everyday life? Let's take two examples of perception from driving a car and football games. On the motorway: I'm driving on the left-hand lane of the motorway. Two cars are driving on the right-hand lane. The question: Can I foresee that the rear one wants to change to my lane? My assessment is influenced by the fact that I drive fast myself. I don't notice that the rear car accelerates, because the fallback illusion hides that. The driver, however, starts to overtake, comes onto my track and slows me down. The distortion of her/his distance to the front car causes me to underestimate its true speed. A similar case on the football field: The striker from my team dashes to the opponent's goal, followed by an opposing player. I myself follow slower and want to play her/him the ball, but I overestimate the speed of the catching-up opponent due to the illusion and therefore I don't pass the ball on, missing the chance to score.

4. What is your study/paper contributing to the added value for the society?
If the study can help to prevent accidents by teaching automated driving systems such a bias, or make highway drivers more relaxed when somebody changes to their lane, this would help. If the study can help to prevent accidents by teaching automated driving systems such a bias, or make highway drivers more relaxed when somebody changes to their lane, this would help.

5. Are there any major caveats?
What questions still need to be addressed?We do not know so far, how strong the illusion is in real traffic situation. Stereovision might reduce the illusion, but conditions like driving during the night or in fog might also strengthen it.

Zur Redakteursansicht