Mental Rotation Of Images
The idea of mental imagery has always been a controversial subject in the
field of psychology. Many psychologists have argued that such a concept
is impossible to measure because it can not be directly observed.
Though they are right about this, it is not impossible to measure how quickly
mental rotations of images are processed in our brains. Subjects in
this experiment were presented two shapes simultaneously, via computer
screen, and asked to make judgement, as quickly as possible, as to whether
the two shapes presented were the same or mirror images. Two different
shapes were used in this experiment, each given as often as the other.
During each trial one shape remained stationary and the other was rotated
with varying amounts of 0, 60, 120, and 180 degrees. As the angle of
rotation increased reaction times were calculated to see if this had any
baring on the speed of the reaction. As predicted, reaction times
increased along with angular disparity.
Rotation of Mental Images: Measured by Reaction Times
There are a limited number of ways to discover and understand how the human
mind works and reacts to things. One can not sit and directly observe the
brain and eye working together (James, Schneider & Rodgers, 1994).
The concept behind mental rotation of images tries to do this by measuring
reaction times as the angular disparity of an object increases. Thus,
demonstrating the time it takes for the eye and brain to make a connection
when presented with a stimulus. Though our experiment was solely
limited to calculating reaction times to mental rotations of images,
Wohlschlager and Wohlschlager (1998) took this concept one step further to
see if mental object rotation and manual object rotation shared a common thought
process in our brain.
Wohlschlager & Wohlschlager (1998) based their ideas for this study on a
theory, most impressively demonstrated by Cooper (1976), stating that the
resemblance of mental rotation to external physical rotation, calls for a
mental process that mimics external physical rotation. However, it is
pointed out that there is a principal difference between motion perception
and mental rotation. Whereas motion perception is a rather automatic
process, mental rotation is strategic and shares some characteristics with
voluntary actions (Wohlschlager & Wohlschlager, 1998).
In conducting their experiment Wohlschlager & Wohlschlager (1998) used 66
right-handed psychology students who took part in this experiment to fulfill
their course requirement. Of the 66 students 30 took part in the mental
rotation condition and 30 in the manual rotation condition.
As expected, findings showed that there was in fact a strong
relationship between the reaction times of the mental and manual
rotations. As the angular disparity of both the mental and manual
rotations increased so did reaction times. In almost all cases manual
and mental reaction times matched each other. It was also observed by
Wohlschlager & Wohlschlager (1998), that manual object rotation did not
always follow the shortest path. Findings showed that in some trials students
rotated images the longer way around. Thus, it was concluded that
mental processes also rotate things the long way around, since reaction times
did not differ much between the two groups, ultimately supporting their
Desrocher, Smith & Taylor (1995) conducted a similar experiment with
intentions of measuring reaction times. Only in this experiment the
independent variable was not mental vs. manual rotation, but gender. They
were interested in seeing if gender differences played a role in regard to
reaction times, when presented with either a letter stimuli or picture
stimuli. It has been noted that up to the present time, there have been
no major findings that show any significant difference in the way men and
woman process letter images. In tests, both men and woman performed
equally well (Desrocher, smith & Taylor, 1995). However, in several
experiments conducted measuring reaction times
when presented with a picture stimulus, men have been said to out perform
woman in almost all cases. Overall, when comparing the reaction times
of pictures to letters, picture reaction times are said to be generally
In conducting this experiment 20 right-handed subjects were used, 10 male and
10 female. Subjects participated in this study either to earn course
credit, or for payment of ten dollars per hour.
Three things were concluded from this experiment. First, picture
rotations took longer to perform than letter rotations. Second, there
were no sex differences in reaction times for either stimulus presented, and
third, there was a linear increase in reaction times as angular disparity of
both stimuli sets increased (Desrocher, smith & Taylor, 1995).
As predicted this study did replicate some previous findings. However, based
on the data from this experiment sex differences did not yield a significant
difference in reaction times when presented with the different stimuli.
Concurrent with our data, both of the previously mentioned experiments
concluded the same thing; reaction times increased as angular disparity
increased. Although the objectives in each experiment were somewhat
different, there has been a sufficient amount of evidence to support the hypothesis,
that reaction times will increase as the angular disparity of objects
Ten St. Johns University undergraduate students participated in this
experiment to fulfill their Research Method Lab Requirement.
Ten IBM computers with color monitors were used in this experiment. To
conduct the experiment, students used custom made Mel Lab Software, designed
for experiments in perception, cognition, social psychology and human
The experiment consisted of 10 practice trials and 128 actual trials.
In order to move on to the actual test students had to get 8 out of the 10
practice trials correct. Once an 80% accuracy rate was obtained in the
practice trial, students were then permitted to move on to the actual test.
This experiment used two different images, each appearing randomly, and as
often as the next. During each trial an image was presented on the
computer screen along side another. The images were always similar; the
two different images were never mixed. As each image appeared students
had to decide as quickly as possible if the two images were the same or
mirror images. Answers were recorded by either hitting the (1) key, if
the images were the same, or hitting the (2) key, if they were mirror
images. On each trial one image was rotated in respect to the other in
varring degrees of 0, 60, 120, and 180, each rotation occurring randomly and
as often as the rest.
A One Way ANOVA was conducted to look into the findings presented from this
experiment. An F(3,36)=9.392 p < .001 was obtained, indicating a
statistical significant difference in the data collected.
This study was conducted to see how quickly mental images are processed in
our minds by measuring reaction times. Our finding, as well as the
findings in the two experiments discussed support the hypothesis that
reaction times will increase as the angle of rotation increases.
However, when comparing the reaction times between the four groups of 0, 60, 120,
and 180 degrees we see that the statistically significant difference we
arrived at did not come from all of the groups. In examining the
numbers, we see a strong statistical significance between groups 1 & 2, 1
& 3, and 1 & 4. This would be because of the large difference
in the number of degrees rotated. When looking at the difference
between groups 2 & 3, 2 & 4, and 3 & 4, we see no statistical
significance at all. The difference in rotational degrees was not as
large between these groups.
Another aspect of the results that I would like to point out is that our
experiment only used 10 subjects, who were not randomly picked. Only 20
subjects were used in the experiment conducted by Desrocher, smith &
Taylor (1995). I don’t feel that these numbers of subjects (20 and 10)
are enough to have conclusive results. In order to get an accurate
sample of the population more participants should have been used in both
experiments. This one fact may have been the reason why part of the
results of Desrocher, smith & Taylor (1995) were not concurrent with
Although the objectives of the two experiments I discussed were different,
result were the same regardless as to what the differences might have
been. In all experiment including the one I did, results were
concurrent. Findings supported the hypothesis that was set forth: As angular
rotation increased, reaction time will also increase. Thus, showing reaction
times do serve as an appropriate method for analyzing how quickly the brain
reacts to mental rotations of images.