Mental rotation: Part 1

Another classic experiment in cognitive psychology gave rise to a phenomenon known as mental rotation. We begin with just one example. On every trial in the experiment we present a single letter and we wait for the participant to make a key press response. The letter itself can either be a capital R or its left-right ‘mirror’ reversal. Participants are instructed to respond as quickly and accurately as they can. Is it a normal R or a mirror-reversed R? Press one key for R and a different key for mirror reversed R. So the measures we are taking are reaction times and measures of accuracy. The reasoning here is simple.

As long as participants are following these instructions – rather than just pressing any key to get the experiment over with – then the following argument is made. We take it that the time to make the physical key press responses directly reflects the time it takes to complete the mental processes upon which the task judgment depends. Cast your minds back to the very first week. What we need to do is set up the experiment so that we keep everything the same but vary one thing. And the one thing that we vary is the angle of rotation of the letter that we present.

We don’t physically rotate the letter during the trial but simply present a rotated version of the letter on the trial. Across the trials the letter is in a sense rotated around an invisible clock face. The upright the letter is at 12 o’clock. We also test letters at 2 o’clock, 4 o’clock, 6 o’clock, 8 o’clock, 10 o’clock. Across trials we vary the angle of rotation of the letter and on a random half of the trials the letter is a normal R and on the remaining trials it is a mirror-reversed R. What we are interested in is how response times change as different rotated versions of the letters are presented. The data from the experiment are clear cut.

Response times were quickest when the letter is upright (12 o’clock) and slowest when it is upside down (6 o’clock). More striking perhaps is that the response times gradually lengthen as the letter was rotated between these two extreme values. We might therefore conclude that the data quite clearly show that in making the letter judgment people conjure up an image of the letter and mentally rotate this to an upright position. Once the image is upright then they can readily see if the R is normal or mirror reversed. But if you don’t believe in the pictures in the head view how else can these data be explained?