Mental Rotation
12,679 Views
•
•
概述
Source: Laboratory of Jonathan Flombaum—Johns Hopkins University
Visual mental imagery refers to the ability to conjure images in one’s mind’s eye. This allows people to process visual material above and beyond the constraints of a current point-of-view; for example, a person could imagine, using their mind’s eye, how something might look in a different color, or what it would look like if it were made from a different material or rotated and seen from a different perspective. Mental imagery seems to support important human behaviors in many contexts. For example, people report visualizing routes and maps when planning a route or giving directions. They report visualizing movements, such as swinging a bat, to prepare for an actual action. They also report the mental rotation of objects in order to consider how an object might fit into a receptacle or clear a barrier.
This video demonstrates how to use the mental rotation procedure in order to investigate visual mental imagery.
Procedure
1. Stimulus design.
- Present the stimuli for this experiment—letters and numbers, along with their mirror images (Figure 1)—in various rotations.
Figure 1. Sample stimuli for a mental rotation task. The top rows show two characters (R and 4) and a copy of each rotated. The bottom row shows a mirror image of each (on the right side of each pair) and a rotated version of each, as well. The mental rotation task involves identifying rotated versions of familiar characters and distinguishing them from rotated versions of their mirror images. - Use PowerPoint, Keynote, or a similar program to generate the stimuli.
- For this experiment, use the uppercase letter ‘R’ and the lowercase letter ‘g,’ along with the numbers ‘4’ and ‘7.’
- In PowerPoint, make one copy of each of the four characters. Use the Helvetica Light font for this experiment.
- Make a second copy of each character and flip them to produce a mirror image of each.
- Make 12 copies of each of the four characters and their mirror images.
- Rotate each of the 12 copies by an increasing increment of 15°. The original copy of each is 0°, the next is 15°, then 30°, and so on, until 180°. Reference Figure 2 to see the full set of stimuli used for this experiment.
Figure 2. A complete set of stimuli for the mental rotation task. Four characters (3, R, 4, and g), each paired with their mirror images and rotated in 15° increments between 0 and 180. - From the 13 rotated copies of each stimulus and its mirror image, generate the stimuli for each individual trial. A trial consists of one of the four non-mirror images printed at the top of a page. The bottom of the page includes that letter or number and its mirror image at one of the 12 rotations (Figures 3 and 4).
- Make a page for each trial of the experiment. Each of the four numbers/letters have 13 trial pages, so make sure there are 52 pages total and print them out.
- Number the backs of the pages 1-52. The number on each page is the number tag. The number a page is labelled with does not matter. However, when numbering the pages, also make a key, a table that reports the nature of the trial on each page and the correct answer (left or right), so responses can be associated with individual trials later. Create the key to look something like an Excel spreadsheet labelled “Mental Rotation Key” (Table 1).
| Number Tag | Character | Angle | Correct Answer |
| 1 | 3 | 60 | RIGHT |
| 2 | g | 75 | RIGHT |
| 3 | g | 30 | RIGHT |
| 4 | g | 60 | LEFT |
| 5 | g | 165 | RIGHT |
| 6 | 4 | 105 | LEFT |
| 7 | 3 | 15 | LEFT |
| 8 | 3 | 165 | LEFT |
| 9 | 4 | 180 | LEFT |
| 10 | R | 15 | RIGHT |
| 11 | g | 180 | RIGHT |
| 12 | g | 45 | RIGHT |
| 13 | g | 105 | RIGHT |
| 14 | 3 | 45 | RIGHT |
| 15 | 4 | 15 | LEFT |
| 16 | R | 60 | LEFT |
| 17 | R | 45 | LEFT |
| 18 | R | 150 | LEFT |
| 19 | g | 0 | RIGHT |
| 20 | R | 30 | LEFT |
| 21 | 3 | 120 | LEFT |
| 22 | 4 | 90 | LEFT |
| 23 | R | 75 | LEFT |
| 24 | 4 | 135 | RIGHT |
| 25 | 3 | 180 | LEFT |
| 26 | 4 | 45 | LEFT |
| 27 | R | 90 | RIGHT |
| 28 | 4 | 0 | LEFT |
| 29 | 4 | 120 | LEFT |
| 30 | 3 | 135 | RIGHT |
| 31 | R | 135 | LEFT |
| 32 | 3 | 30 | LEFT |
| 33 | 4 | 75 | LEFT |
| 34 | 3 | 105 | LEFT |
| 35 | 3 | 150 | LEFT |
| 36 | R | 105 | RIGHT |
| 37 | 4 | 60 | RIGHT |
| 38 | 4 | 30 | LEFT |
| 39 | R | 120 | RIGHT |
| 40 | R | 180 | RIGHT |
| 41 | g | 135 | RIGHT |
| 42 | 3 | 0 | LEFT |
| 43 | 3 | 90 | LEFT |
| 44 | 4 | 150 | RIGHT |
| 45 | 4 | 165 | LEFT |
| 46 | 3 | 75 | RIGHT |
| 47 | R | 165 | LEFT |
| 48 | g | 90 | RIGHT |
| 49 | g | 150 | RIGHT |
| 50 | g | 15 | LEFT |
| 51 | R | 0 | RIGHT |
| 52 | g | 120 | RIGHT |
Table 1. An example of a key that reports the nature of the trial on each page and the correct answer (left or right).
Figure 3. A sample test page for one trial of the mental rotation experiment. The participant must report whether the character on the left or the right (below the line) is the rotated ‘R’ (as opposed to its mirror image rotation).
Figure 4. A sample test page for one trial of the mental rotation experiment. The participant must report whether the character on the left or the right (below the line) is the rotated ‘4’ (as opposed to its mirror image rotation).
- In order to associate results with the content of each trial, a place to record the results is needed, so make another spreadsheet with a column for the trial number, a column for the number tag of that trial, a column for the response given (right or left), and a column for the time it takes the participant to produce a response. Print out this response sheet.
- In addition, make a demo page to use in order to explain the instructions. Have it show an example of one of the characters, its mirror image, and some examples of the character at one of the rotations. Label it clearly (Figure 5).
Figure 5. An instruction sheet. This sheet presents one of the characters, its mirror image, and a rotation of each in order to facilitate explanation of the procedure to the participant.
- In addition to printed versions of the trial and demo pages, acquire a stopwatch for the experiment, and it is also useful to have an assistant to help with timing and recording responses.
2. Procedure.
- Use the demo page to explain the instructions to the participant: “This experiment will investigate your ability to rotate objects using visual imagery. On each trial, you will be asked to decide which one of two images is a rotated version of a familiar number or letter. The alternative will be a mirror image rotation of the same number or letter. Here is one example: the number 3. And here is its mirror image. In a trial, you will see the original 3, along with something like the two examples on the bottom of the page, the rotated 3 and a rotation of its mirror image. You will say “left” or “right” to indicate which one you think is the original. You should respond as quickly as possible without sacrificing accuracy. In other words, respond as soon as you are sure of the answer, but not sooner.”
- Shuffle the 52 test pages to randomize the order. After shuffling, enter the number tags of the trials in the order they will be presented in on the response sheet.
- Start the experiment. Place the stack of pages face-down between the experimenter and the participant, the number tags facing up.
- Have the assistant hold the stopwatch.
- On each trial, have the assistant say, “Go,” when they start the timer. At this signal, flip over the page for that trial. Have the participant study the page and say “right” or “left” to indicate which character is the answer for that trial.
- When the participant reports a response, have the assistant stop the timer. Then, record the response time and the answer given on the response sheet.
- Repeat this procedure for all 52 trials.
3. Analysis.
- When the experiment is done, analyze the data.
- Enter the handwritten number tags, responses, and response times into a digital copy of the response sheet, and save it with a new name, for example, “Participant 1 MR responses” (Table 2).
| Trial # | Number Tag | Response Given | Response Time | Number Tag2 | Character | Angle | Correct Answer | Answer Correct? |
| 4 | 1 | RIGHT | 4876 | 1 | 3 | 60 | RIGHT | 1 |
| 38 | 2 | RIGHT | 6758 | 2 | g | 75 | RIGHT | 1 |
| 40 | 3 | RIGHT | 3579 | 3 | g | 30 | RIGHT | 1 |
| 26 | 4 | LEFT | 8752 | 4 | g | 60 | LEFT | 1 |
| 10 | 5 | RIGHT | 6494 | 5 | g | 165 | RIGHT | 1 |
| 49 | 6 | LEFT | 6587 | 6 | 4 | 105 | LEFT | 1 |
| 16 | 7 | LEFT | 3434 | 7 | 3 | 15 | LEFT | 1 |
| 45 | 8 | LEFT | 9172 | 8 | 3 | 165 | LEFT | 1 |
| 35 | 9 | LEFT | 1856 | 9 | 4 | 180 | LEFT | 1 |
| 17 | 10 | RIGHT | 6818 | 10 | R | 15 | RIGHT | 1 |
| 12 | 11 | RIGHT | 4797 | 11 | g | 180 | RIGHT | 1 |
| 5 | 12 | RIGHT | 5378 | 12 | g | 45 | RIGHT | 1 |
| 21 | 13 | RIGHT | 3301 | 13 | g | 105 | RIGHT | 1 |
| 25 | 14 | RIGHT | 1393 | 14 | 3 | 45 | RIGHT | 1 |
| 33 | 15 | LEFT | 3937 | 15 | 4 | 15 | LEFT | 1 |
| 42 | 16 | LEFT | 5827 | 16 | R | 60 | LEFT | 1 |
| 31 | 17 | LEFT | 6004 | 17 | R | 45 | LEFT | 1 |
| 9 | 18 | LEFT | 6174 | 18 | R | 150 | LEFT | 1 |
| 46 | 19 | RIGHT | 6619 | 19 | g | 0 | RIGHT | 1 |
| 3 | 20 | LEFT | 2276 | 20 | R | 30 | LEFT | 1 |
| 18 | 21 | LEFT | 4176 | 21 | 3 | 120 | LEFT | 1 |
| 28 | 22 | LEFT | 7819 | 22 | 4 | 90 | LEFT | 1 |
| 24 | 23 | LEFT | 7368 | 23 | R | 75 | LEFT | 1 |
| 6 | 24 | RIGHT | 4984 | 24 | 4 | 135 | RIGHT | 1 |
| 47 | 25 | LEFT | 4495 | 25 | 3 | 180 | LEFT | 1 |
| 7 | 26 | LEFT | 5476 | 26 | 4 | 45 | LEFT | 1 |
| 50 | 27 | RIGHT | 7919 | 27 | R | 90 | RIGHT | 1 |
| 27 | 28 | LEFT | 7182 | 28 | 4 | 0 | LEFT | 1 |
| 48 | 29 | LEFT | 5793 | 29 | 4 | 120 | LEFT | 1 |
| 13 | 30 | RIGHT | 8986 | 30 | 3 | 135 | RIGHT | 1 |
| 36 | 31 | LEFT | 9457 | 31 | R | 135 | LEFT | 1 |
| 11 | 32 | LEFT | 7903 | 32 | 3 | 30 | LEFT | 1 |
| 29 | 33 | LEFT | 9703 | 33 | 4 | 75 | LEFT | 1 |
| 51 | 34 | LEFT | 9565 | 34 | 3 | 105 | LEFT | 1 |
| 1 | 35 | LEFT | 9341 | 35 | 3 | 150 | LEFT | 1 |
| 8 | 36 | RIGHT | 2849 | 36 | R | 105 | RIGHT | 1 |
| 52 | 37 | RIGHT | 2355 | 37 | 4 | 60 | RIGHT | 1 |
| 2 | 38 | LEFT | 2094 | 38 | 4 | 30 | LEFT | 1 |
| 32 | 39 | RIGHT | 7338 | 39 | R | 120 | RIGHT | 1 |
| 43 | 40 | RIGHT | 5431 | 40 | R | 180 | RIGHT | 1 |
| 37 | 41 | RIGHT | 2734 | 41 | g | 135 | RIGHT | 1 |
| 19 | 42 | LEFT | 5978 | 42 | 3 | 0 | LEFT | 1 |
| 14 | 43 | LEFT | 3305 | 43 | 3 | 90 | LEFT | 1 |
| 22 | 44 | RIGHT | 5273 | 44 | 4 | 150 | RIGHT | 1 |
| 41 | 45 | LEFT | 4472 | 45 | 4 | 165 | LEFT | 1 |
| 23 | 46 | RIGHT | 2353 | 46 | 3 | 75 | RIGHT | 1 |
| 34 | 47 | LEFT | 8211 | 47 | R | 165 | LEFT | 1 |
| 20 | 48 | RIGHT | 2049 | 48 | g | 90 | RIGHT | 1 |
| 44 | 49 | RIGHT | 9719 | 49 | g | 150 | RIGHT | 1 |
| 39 | 50 | LEFT | 9562 | 50 | g | 15 | LEFT | 1 |
| 15 | 51 | RIGHT | 1282 | 51 | R | 0 | RIGHT | 1 |
| 30 | 52 | RIGHT | 3548 | 52 | g | 120 | RIGHT | 1 |
Table 2. An example of a completed response sheet.
- Open the response key and sort by number tag. Cut and paste the contents of the response key into the sheet with the participant responses, so now there are columns reporting both the responses and response times of the participant, and the details of each trial.
- Add one more column to the end of the sheet, labelled “Response Correct?” Compare the response given with the correct response in each row, and enter a 1 in the response correct row if the answer given was correct or a 0 if it was not.
- Summarize the results. Average the “Response Correct?” column to see the overall proportion of trials with correct responses. This should be very high or perfect for most participants. Average the response time column to get a sense of the overall response times. That column can also be averaged while filtering by the letter on the trial and/or the orientation of the letter.
Individuals must rely on visual mental imagery—the ability to conjure images in one’s mind’s eye—to accurately perceive the world and guide actions.
For example, mental imagery is used to visualize a route when planning directions to particular location, or what a house might look like if it were remodeled.
Experimental psychologists can measure a person’s visual mental imagery through the use of a mental rotation paradigm, which involves identifying rotated versions of familiar characters and distinguishing them from rotated versions of their mirror images.
Using the mental rotation procedure, this video will demonstrate how to design stimuli and conduct an experiment, as well as how to analyze and interpret results investigating visual mental imagery.
In this experiment, participants are presented with stimuli and asked to distinguish whether subsequent stimuli are rotations of the original item or of its mirror image.
In this case, the task stimuli consist of letters, such as R and g, as well as numbers, like 4 and 7, all printed in Helvetica Light font.
Two versions of the letters and numbers are produced: the original and a flipped, mirror image. The characters are then manipulated, such that each one is rotated by an increasing increment of 15°, starting at 0° and ending at 180°.
During each trial, participants are presented with one of the four manipulated characters and then asked to decide from two possible choices which one is the rotated version of the original item as quickly and accurately as possible.
Thus, the dependent variable is response time—how long it takes for the participant to make a response.
It is hypothesized that response times will be faster for characters that have little rotation, compared to those that are rotated the most. In other words, the response times are longer the more a character is rotated from its canonical orientation.
To begin the experiment, gather stimuli sheets that have been created for each individual trial. For each trial, note that one of the four non-mirror images is printed at the top, and the two choices are located on the bottom of the page.
Number the back of each page from 1–52, which is called the ‘number tag.’ To randomize the order, shuffle the test pages.
To more easily associate the results with the content of each trial, create a response sheet that includes the trial number, number tag in the order of presentation, response given, and response time.
As the last preparation step, gather a stopwatch and an assistant.
When the participant arrives, explain the instructions to them using a demo page. Note that one of the characters, its mirror image, and a couple of examples of the character at one of the rotations is shown.
Next, place the test pages facedown between the experimenter and the participant.
During each trial, once the assistant starts the timer and says, “Go,” flip over a page for the participant.
When the participant reports a response, stop the timer. Then, record the response time and the answer on the response sheet. Repeat this procedure for all 52 trials.
Once the experiment is complete, create a digital copy of the response sheet, including the number tags in numerical order, responses given, response times, and correct answers.
Mark whether the responses given were correct by entering a 1 in the ‘Answer Correct?’ column or a 0 if incorrect.
For the correct trials, plot the average response times for each character shown as a function of rotation. Note that the response times increase proportionally with the degree of rotation. These results suggest that the brain simulates the physical transformations.
Now that you are familiar with designing a mental rotation experiment, you can apply this approach to answer specific questions about visual mental imagery.
Practically speaking, people who are especially good at visual thinking about physical spaces can be identified through the mental rotation task. Certain individuals are exceptionally good at using mental imagery to guide their actions—like architects and mechanical engineers.
In addition, researchers use functional magnetic resonance imaging to investigate brain regions involved in mental rotation.
When people mentally rotate objects without looking at them, there is an enormous amount of brain activity in the visual cortex in particular, and in regions such as the parietal lobe—brain areas generally thought to be involved in seeing. In other words, the brain systems used to actually see visual stimuli are also used to imagine visual stimuli.
Finally, researchers examine mental rotation in virtual reality to study how mental imagery is involved in navigating through different spatial environments and obstacles.
You’ve just watched JoVE’s introduction to conducting a mental rotation experiment. Now you should have a good understanding of how to design and conduct the experiment, and finally how to analyze and interpret the results.
Thanks for watching!
Results
A common way to graph the results is to plot the response time for each character as a function of the rotation of the character (and its mirror image; Figure 6).
Figure 6. Results from the mental rotation task. Response times are plotted for each of the characters as a function of the amount of rotation in a given trial. Generally, response times are longer the more a character is rotated from its canonical orientation, suggesting that brain mechanisms simulate physical transformations.
One of the most interesting common results associated with mental rotation tasks is that the amount of time it takes to produce a response is proportional to the degree of rotation distinguishing the target character and its rotated pair. In other words, the time it takes to rotate an object mentally seems proportional to the time it would take to actually rotate physical objects in order to place them at the same orientation. This suggests that mental rotation relies on mechanisms that really try to simulate physical space in the brain, even though no pieces of the brain rotate.
Applications and Summary
One of the main practical applications for mental rotation tasks is to identify people who are especially good at visual thinking about physical spaces. Think about the skills it takes to be a good architect, mechanical engineer, an expert carpenter, or welder. Some people are really good at using mental imagery to guide their actions, and some people are not very good at all, reporting that they don’t even really see pictures in their mind’s eye the way most people do. The mental rotation test is a good way to identify exceptionally good and exceptionally bad visualizers in order to help people find the best uses of their abilities.
Mental rotation has also been an important part of neuroscience research aimed at understanding the parts of the occipital and parietal lobes involved in human vision. One of the most surprising findings is that when people mentally rotate objects without looking at them, there is an enormous amount of brain activity in visual cortex and brain areas generally thought to be involved in seeing. In other words, the brain systems used to actually see visual stimuli are also used to imagine visual stimuli.
成績單
Individuals must rely on visual mental imagery—the ability to conjure images in one’s mind’s eye—to accurately perceive the world and guide actions.
For example, mental imagery is used to visualize a route when planning directions to particular location, or what a house might look like if it were remodeled.
Experimental psychologists can measure a person’s visual mental imagery through the use of a mental rotation paradigm, which involves identifying rotated versions of familiar characters and distinguishing them from rotated versions of their mirror images.
Using the mental rotation procedure, this video will demonstrate how to design stimuli and conduct an experiment, as well as how to analyze and interpret results investigating visual mental imagery.
In this experiment, participants are presented with stimuli and asked to distinguish whether subsequent stimuli are rotations of the original item or of its mirror image.
In this case, the task stimuli consist of letters, such as R and g, as well as numbers, like 4 and 7, all printed in Helvetica Light font.
Two versions of the letters and numbers are produced: the original and a flipped, mirror image. The characters are then manipulated, such that each one is rotated by an increasing increment of 15°, starting at 0° and ending at 180°.
During each trial, participants are presented with one of the four manipulated characters and then asked to decide from two possible choices which one is the rotated version of the original item as quickly and accurately as possible.
Thus, the dependent variable is response time—how long it takes for the participant to make a response.
It is hypothesized that response times will be faster for characters that have little rotation, compared to those that are rotated the most. In other words, the response times are longer the more a character is rotated from its canonical orientation.
To begin the experiment, gather stimuli sheets that have been created for each individual trial. For each trial, note that one of the four non-mirror images is printed at the top, and the two choices are located on the bottom of the page.
Number the back of each page from 1–52, which is called the ‘number tag.’ To randomize the order, shuffle the test pages.
To more easily associate the results with the content of each trial, create a response sheet that includes the trial number, number tag in the order of presentation, response given, and response time.
As the last preparation step, gather a stopwatch and an assistant.
When the participant arrives, explain the instructions to them using a demo page. Note that one of the characters, its mirror image, and a couple of examples of the character at one of the rotations is shown.
Next, place the test pages facedown between the experimenter and the participant.
During each trial, once the assistant starts the timer and says, “Go,” flip over a page for the participant.
When the participant reports a response, stop the timer. Then, record the response time and the answer on the response sheet. Repeat this procedure for all 52 trials.
Once the experiment is complete, create a digital copy of the response sheet, including the number tags in numerical order, responses given, response times, and correct answers.
Mark whether the responses given were correct by entering a 1 in the ‘Answer Correct?’ column or a 0 if incorrect.
For the correct trials, plot the average response times for each character shown as a function of rotation. Note that the response times increase proportionally with the degree of rotation. These results suggest that the brain simulates the physical transformations.
Now that you are familiar with designing a mental rotation experiment, you can apply this approach to answer specific questions about visual mental imagery.
Practically speaking, people who are especially good at visual thinking about physical spaces can be identified through the mental rotation task. Certain individuals are exceptionally good at using mental imagery to guide their actions—like architects and mechanical engineers.
In addition, researchers use functional magnetic resonance imaging to investigate brain regions involved in mental rotation.
When people mentally rotate objects without looking at them, there is an enormous amount of brain activity in the visual cortex in particular, and in regions such as the parietal lobe—brain areas generally thought to be involved in seeing. In other words, the brain systems used to actually see visual stimuli are also used to imagine visual stimuli.
Finally, researchers examine mental rotation in virtual reality to study how mental imagery is involved in navigating through different spatial environments and obstacles.
You’ve just watched JoVE’s introduction to conducting a mental rotation experiment. Now you should have a good understanding of how to design and conduct the experiment, and finally how to analyze and interpret the results.
Thanks for watching!