Abstract
Evidence on mental imagery is seen in this replication of Cooper and Shepard’s (1973) first experiment on mental rotation through the correlation between the orientation of the stimuli and the reaction time of participants. The within-subjects design was used and the participants investigated in this study were first year Psychology students from the University of Nottingham Malaysia Campus. The stimuli in this experiment was a normal and reversed “G” and “R” rotated from range of 0 to 180 degrees and presented on computers, using the software PsychoPy. The results of this experiment were similar to the study by Cooper and Shepard (1973) whereby there was a positive correlation between the orientation and the reaction time. As the amount of angular rotation increased, the response time to the presented stimulus increased. So, mental rotation occurs and is compared to the internal representation of the normal upright letter that is stored in the long-term memory.
Mental Rotation: The Correlation between the Angle of Rotation of a Stimuli and Reaction Time
Visual mental imagery is described as “seeing” a sensory input that is not currently and physically present but is perceived through internal information rather than external (Kosslyn & Osherson, 1995). Galton (1883) first measured mental imagery by asking participants to picture what is set on their breakfast table and their descriptions were vivid, well-defined and illustrative (Galton, 1883). Further research on this was by Paivio (1971) in his Dual-Coding Hypothesis, which works through a system that activates concepts, stated that people can either encode information verbally or through mental images or both (Kosslyn, Holyoak, & Huffman, 1976). However, this theory was argued by Pylyshyn (1973) that internal representation is more abstract rather than analogical (Kosslyn, Holyoak, & Huffman, 1976).
With the concept of mental imagery in mind, Shepard & Metzler (1971) conducted experiments on the rotation of mental images with two-dimensional pictures of three-dimensional shapes rotated from 0 to 180 degrees. Results found that there was a relationship between the angles of rotation and the response time (Vandenberg & Kuse, 1978). Extended research on the rotation of mental imagery by Cooper and Shepard (1973) was conducted in experiments using alphanumeric stimuli: normal representation of the stimuli, that was already stored in their long-term memory, and the reversed or “mirror” representation of the stimuli (Cooper & Shepard, 1973).
In their first experiment, participants were presented with a stimulus that was rotated in different angles and requested to determine if the stimulus was normal or reversed. The findings showed an increase in reaction time as the amount of rotation increased (Cooper & Shepard, 1973). A conclusion was drawn after analysing these results that participants were mentally rotating an internal representation of that test character into its normal upright form and subsequently comparing this with its permanently stored internal representation of this character’s standard upright position. The only difference of the second experiment from the first, was that the participants were provided with auditory information regarding the identity and orientation of the test stimulus before being presented with the following stimulus.
Another experiment has also used Cooper and Shepard’s (1973) study on mental rotation as a validated literature to test event-related potentials. The aim was to recognize and confine an electrophysiological correlate of the mental rotation process. (Peronnet & Farah, 1989). In this current study, the first experiment of Cooper and Shepard’s (1971) Mental Rotation study was replicated but with a few alterations made. Participants were Psychology students in the University of Nottingham Malaysia Campus, letters “G” and R” were used instead of letters and numerals and after obtaining the data, only the normal version of the letter was analysed. Moreover, the range of orientations was from 0 to 180 degrees for both normal and reversed stimuli and the stimulus was continuously presented in the middle of the screen.
Based on Cooper and Shepard’s (1973) first experiment on mental rotation, the findings predicted to be obtained would be an increase in the participant’s response time to the stimulus the more it is angularly rotated. This could be due to how people manipulate their own mental images by mentally rotating to fit an internal representation in the long-term memory store.
Method
Design
The design used in this experiment was a within-subjects design. The independent variable was the orientation of the “normal” “G” and “R” stimuli and the dependent variable was the reaction time (in seconds) it took participants to respond to the stimuli. Participants were exposed to all the conditions of the independent variable which was the different angles of rotation from 0 to 180 degrees. The results that were expected to be obtained was that as the angle of rotation increased, the reaction time would also increase.
Participants
The 51 participants (7 males and 44 females), that were 1st Year Psychology students, took part in all the conditions of the experiment. They were all from the University of Nottingham Malaysia Campus and obtained through an opportunity sampling as they all attended the same Practical Methods in Psychology module. The participants had normal vision or corrected vision and showed no signs of any other sensory impairment.
Apparatus & Materials
The experiment was conducted using computers that were running the open-source software, PsychoPy. It is used for psychology experiments to run experiments, display the stimuli and collect the data for results. The stimuli presented were letters that were either “normal” or reversed as shown in Figure 1. The width and height of each letter was 400 pixels. The duration of the stimuli being displayed was 2 seconds and its position was at the centre of the screen.
To ensure that participants’ focus was at the centre and to ready the participants to the stimuli displayed after, a fixation cross was put in prior to each stimulus. The fixation cross was set to the colour black and the font was Arial. The response keys that were to be pressed as quickly and accurately as possible was the “left” and “right” arrows on the computer’s keyboard.
Figure 1: The stimuli of the “normal” “G” and “R” and the reversed “G” and “R” used in the experiment (not to scale)
Procedure
At the start of the experiment, an on-screen dialog box was required to be filled up with personal information about the participant (participant identification, age and gender). Shortly after completing that, there were instructions on the screen which briefly told participants how to respond to the following stimuli. The instructions were the “left” arrow key was to notify that the letter was normal and “right” arrow when the letter was reversed. Also, advice was given to respond as quickly and accurately as possible and once participants understood what to do and were ready to begin the main trial, any key could be pressed.
A fixation cross was used to guide the visual field to prepare participants for the following stimulus. The cross lasted for 0.5 s and was positioned at the centre of the screen [0, 0]. The main stimulus was the “normal” “G” and “R or the “mirror” (or reversed) image of “G” and “R” with the angles of rotation ranging from 0 to 180 degrees. There were 76 trial types of the stimulus and one of the four images along with one of the angles of rotation was presented and would change on every repeat of the routines. There was a total of one hundred and fifty-two trials as the experiment was set to 2 repeats of each trial type. The letter size was [400, 400] pixels, the position was in the middle of the screen, [0, 0] and it displayed on the screen for 2 s. The participant could respond to the stimulus by pressing either the “left” or “right” keys (the 4 arrows under the return key) that were located on the computer’s keyboard. The correct answer the participant gave would prompt the continuation of the following stimulus.
Once all the trials were completed, a short text was shown to inform participants of the end of the experiment and to press any key to exit. The data of the accuracy and response time of the participants from PsychoPy was compiled and allocated to a saved folder in the computer.
Results
The strength of relationship between the orientation of the “normal” stimuli and the reaction time of each participant was measured after obtaining the data. This was accomplished by firstly, creating a new copy of the data in the Microsoft Excel file (but not the whole data folder) to analyse the results and keep the original file unchanged to be used. The data was later analysed through deleting the columns of the reactions times of reversed “G” and reversed “R” and filtering out the incorrect trials.
The average reaction time for the 4 remaining trials (2 trials with “G” and 2 trials with “R”) of each orientation were then measured. Lastly, the relationship between the orientations of the stimuli and the average reaction time for the trials were assessed and there was an overall positive correlation between the two variables, r = 0.669. A scatterplot summarized the results of this participant (Figure 2)
Figure 2: Relationship between the orientation and the mean reaction time of one of the participant’s results in a replica of the Cooper & Shepard (1973) experiment. Pearson’s r = 0.669
Once all the participants had calculated the correlation of their individual data, they recorded their gender and r into a piece of paper which was keyed into SPSS to analyse the data. A one-sample t-test was used to determine the mean correlation of the relationship between orientations and the reaction time. The average correlation (M = 0.55, SD = 0.26) was greater than 0 and was statistically significant, (t (50) = 14.92, p < .001). There was a positive correlation so the reaction time increases as the angles of rotation increases.
Discussion
In one of the participant’s results, there was a positive correlation between the angles of rotation of the letters and the response time of the participants to identify whether the letter was normal or reversed. The overall results obtained from this current experiment were similar to the Cooper and Shepard (1973) and Shepard Metzler (1971) studies. They showed that participants were significantly slower in responding to the stimuli the more the amount of its angular rotation. Therefore, this supported the notion that people rotate the image mentally and the more the letter was angularly rotated, the more time it took to rotate it to its standard upright position and identify if it was normal or reversed.
Mental rotation is a mental transformation process that states that the shape of the presented stimulus moves through the intermediate orientations prior to arriving at the target orientation (Tarr & Pinker, 1989). Results from this current experiment was consistent to the previous study as reaction time was monotonically longer for the letters that were farther away from its standard upright position. The tilted stimulus was mentally rotated to be in congruence to its normal representation recognized in the long-term memory store as proposed by Cooper and Shepard (1973) (Cooper, 1975).
The data for the reversed stimuli was omitted from the analysis as the degree of rotations of the reversed might have differed from the normal stimuli for a few reasons. Firstly, it was a confounding variable as it might take relatively more time to identify it than the normal stimuli. Secondly, participants may not have been used to a “mirror” image and therefore, it is unclear whether reaction time was due to the stimulus being reversed or the angular rotation. It was still included in this study to increase accuracy as there was an option to click the “left” or “right” key instead of pressing a single button throughout.
However, there were some participants that had a negative correlation instead. The variables could have been that when it came to 180 degrees, participants had mentally flipped the image instead and this would result in a faster reaction time (Cooper & Shepard, 1973). Furthermore, there was no test on whether the conflicting individual results could be attributable to the disparity between the number of female and male participants. Also, the ages of participants were not recorded in this study and hence, it is unknown if the ages of participants may have varied the results. There could be future research testing the presence of gender differences on this matter and gathering the ages of participants and calculating the mean and standard deviation of this variable.
Mental imagery is manifested in people as seen in this study of mental rotation. With the angles of rotation of the character increased, the slower it takes the participant to respond and identify it. The results evidently supported mental imagery due to how people mentally rotated a stimulus for it to match its internal representation of its standard upright position stored in the permanent memory.
References
Cooper, L. A. (1975). Mental rotation of random two-dimensional shapes. Cognitive psychology, 7(1), 20-43.
Cooper, L. A., & Shepard, R.N. (1973). Chronometric studies of the rotation of mental images. Visual Information Processing, 72-121
Cooper, L. A., & Shepard, R.N. (1973). The time required to prepare for a rotated stimulus. Memory & Cognition, 1, 246-250
Galton, F. (1883). Inquiries into the human faculty & its development. JM Dent and Company.
Kosslyn, S. M., Holyoak, K. J., & Huffman, C. S. (1976). A processing approach to the dual coding hypothesis. Journal of Experimental Psychology: Human learning and memory, 2(3), 223-233.
Kosslyn, S. M., & Osherson, D. N. (1995). An invitation to cognitive science: Visual cognition (Vol. 2). MIT Press.
Peronnet, F., & Farah, M. J. (1989). Mental rotation: An event-related potential study with a validated mental rotation task. Brain and cognition, 9(2), 279-288.
Tarr, M. J., & Pinker, S. (1989). Mental rotation and orientation-dependence in shape recognition. Cognitive psychology, 21(2), 233-282.
Vandenberg, S. G., & Kuse, A. R. (1978). Mental rotations, a group test of three-dimensional spatial visualization. Perceptual and motor skills, 47(2), 599-604.