Online Muller-Lyer Illusion Experiment

Home Page

Lexical Decision Task
Recognising words.

Serial Position Curve
A memory experiment.

Back
to Illusions & Experiments.

Muller-Lyer Illusion
Online Experiment

This experiment is concerned with your judgement of where the midpoint of a horizontal line is. You will be presented with 20 arrows, one after another. For each you must try to click on the exact center of the horizontal line section. Try to imagine the arrowhead is not there and just click on the centre of the line. Sounds easy? give it a go! Afterwards you will be shown your results in a chart (explained below). Here it is...

There should be an applet here!
Either your browser doesn't support Java or it's disabled in your Preferences

When you have finished the experiment please click here to submit your results to our database (and use Back to get back to here):

You can see the current online results combining all the data that has been submitted so far (this will appear in the chart above):

Explanation of Results

X-axis: the angle of the arrow head (as according to the diagrams at the bottom).

Y-axis: your error in judging the centre. Zero means you were exactly right. A posotive (+) error means you were too close to the arrowhead, whereas a negative (-) means you were too far away.

Red line: this is the best way to fit a line to the data - it shows the basic trend, if the experiment has worked it won't be horizontal!

Statistical Significance: This basically tells us if the experiment has worked. If it says "Significant at the 5% level" this means there is only a 5% chance (1 in 20) that the data looks good by accident, that is, without some real effect causing it to look the way it does. if it says "Not Statistically Significant" then probably there is too much variation to determine if there really is an effect.

Discussion

For most people, if the experiment works, the result should be a upward sloping line which crosses the zero mark near the middle (where the arrow looks like a sideways 'T'). Essentially, when the arrowhead is pointed like a conventional 'arrow', we tend to underestimate the length of the line on the arrow head end. When we take a guess at the centre that side is actually longer than we perceive it to be, so our 'centre' is too far from that end. When the arrow is inverted the opposite happens. You could say our perception of the figure as a whole biases our ability to isolate an individual feature.

Collecting the data from a large number of participants (as we are doing when you submit your data) could increase the significance of the experimental result. However, some factors may cause the data to become less clear, for example: everyone uses different types of computers and monitors; we can't be sure they have understood the instructions properly and we have no idea who is actually doing the experiments. Running experiments online is a hot topic in Psychology at the moment. This is just a fun example, but nevertheless it will be interesting to see how it works out

New results are collated approximately every two weeks, so come back to see the latest updates!