Malaysian Online Journal of Instructional Technology

ISSN: 1823-1144

Vol. 1, No. 2, December 2004

The Improvement of Mental Rotation

through Computer Based Multimedia Tutor

Khairul Anuar Samsudin & Azniah Ismail

Faculty of Information Technology and Communication

Universiti Pendidikan Sultan Idris

35900 Tanjung Malim, Perak, Malaysia

kanuar@upsi.edu.my & azniah@upsi.edu.my

Abstract.

Mental rotation is identified as one of the three critical components of spatial ability. Its importance is pronounced in the fields of engineering, science, architecture and graphical studies. The biological and environmental issues confronting its acquisition and improvement are dealt in the first section of the paper. Second section of the paper outlines the pre-post test quasi-experimental design procedure engaging a convenient sample involving 58 undergraduates from two intact classes that trained in the Computer Based Multimedia Tutor (CBMT) for 5 weeks to improve performance in mental rotation in terms of accuracy and speed. Subjects were pre-tested with an online version of mental rotation test prior to intervention and post-testing was done immediately after treatment. An ANCOVA procedure was carried out to analyse the data gathered. Final section of the paper discusses the findings and its impact on such interventional programme. The efficacy of CBMT was demonstrated where subjects in experimental condition were more accurate than the control group. However, no similar gain in CBMT was replicated for speed performance as both groups were statistically equivalent in their reaction time when solving the mental rotation tasks. Differences attributed to gender and spatial experience factors were also observed favouring males and subjects with high spatial experience in the study.

INTRODUCTION

The ability of individuals to visualise and manipulate mental images has been recognised as an important cognitive ability instrumental in both mundane activities and academic endeavours. Numerous meanings and terms have been proposed by cognitive scientists and psychologists to explicate this human trait. Defining spatial ability, its acquisition and enhancement, and factors influencing its development are intensely discussed in the visual- and spatial–related literatures. The generic term of spatial ability has been associated with spatial cognition, spatial reasoning, spatial intelligence, and spatial sense. These diverse definitions make it difficult be precise about its meaning (Eliot & Smith, 1983) which further adds more intricacies and sophistication to its discussion. The enhancement of this skill and best method of training has been a significant concern of educators, psychologists and scientists (Pellegrino et al., 1983). Despite all the debates, one important consensus manifested is that without developed spatial ability students are often hindered in the learning environments and ultimately within their chosen field (Bertoline, 1998). Spatial ability of an individual often refers to the ability to manipulate or transform the image of spatial patterns into other arrangements (Eckstrom et al., 1976). Many researchers have used spatial ability to benchmark performance of mathematics (Gallagher, 1989), engineering drawing and graphics (Baartmans & Shery, 1996; Ferguson, 1977), science education (Pribyl & Bodner, 1987; Lord & Rupert, 1995), physical education (Meeker, 1991) and educational therapies (Crano & Johnson, 1991). Its importance is further emphasised in domains of highly visuo-spatial nature such as engineering, architectecture, pilot training, and graphical studies.

Bertoline (1998) defines spatial cognition as the mental process used to perceive, store, recall, create, edit, and communicate spatial images. Gardners (1983) identifies spatial intelligence as one of the seven distinct types of intelligence where his account of spatial intelligence touches upon the ability to convey a sense of the "whole" of a subject, a "gestalt" organisation, different from a logical-mathematical kind of organisation. Clements & Michael (1992) states that spatial reasoning comprising cognitive processes by which mental representations of spatial objects, relationships and transformation are constructed and manipulated. Similarly, Linn & Petersen (1985) define spatial reasoning as “skills in representing, transforming, generating and recalling symbolic, nonlinguistic information”. The National Council of Teachers of Mathematics Spatial (NCTM, 1989) defines spatial sense as an intuitive feel for one’s surroundings and objects in them. In fact the council members believe that students development of spatial skills require experiences that focus on geometric relationships; the direction, orientation, and perspectives of object in space; the relative shapes and sizes of figures and objects; and how a change in shape relates to a change in size. Regardless of these definitions, it is suggested that the ability to think quickly and to construct complex mental models is a sign of intelligence and an important pre-requisite to learning. In an effort to better understand its cognitive attributes and processing, Linn & Petersen (1985) categorised this ability into three components: spatial perception, spatial visualisation and mental rotation. Among these components mental rotation was found to produce significant gender differences that have persisted in most studies prior to instructional intervention (Khairul et al., 2003; Lehmann, 2000; Leopold et al., 2001; Turos & Ervin; 2000). Mental rotation is the ability to quickly and accurately rotate two- (2D) or three-dimensional (3D) objects in one’s mind.

Several psychological studies have identified various factors that can impact on individuals’ development of spatial ability. Factors identified to have major influences are individuals’ gender, age and spatial-related experience (Miller, 1996). Masters & Sanders (1993) found a substantial male advantage in 14 out of 14 studies assessed in a meta-analysis of sex differences on the Mental Rotation Test. In terms of age factor, individuals’ spatial ability seems to reach maturation stage at adolescence and will gradually decline in the late twenties in the general populations due to aging effect even among individuals who are using these abilities in their profession (Salthouse et al., 1990). The optimal age for acquiring spatial ability is between eleven and twelve years old as suggested by Ben-Chaim (1989) which implies instructional strategy for teaching spatial tasks should be given higher priority in middle schools. Students’ early experiences in activities involving manipulatives, paper-cutting and, folding and unfolding of nets prove to be a contributing factor in improving their spatial ability. A study by Deno (1995) found out that the non-academic activities such as model building, sketching, and assembly of parts during the high school years has substantially increased the positive correlation to spatial visualisation.

There are several theories that attempt to explain gender differences in spatial ability from the biological and environmental perspectives. Earl’s & Silverman’s (1994) Hunter-Gatherer theory asserts that human-evolution has impacted the gender differences on a socio-biological perspective. Engaged in hunting males developed the spatial skills for these tasks whilst females gained better peripheral perception and greater spatial memory due to activities related to gathering or foraging. At the physiological level men showed an increase in right hemispheric activation while they process spatial information (Gur et al., 2000). Conversely, no such increase was found in female subjects suggesting bilaterality of women may underlie the sex differences in spatial performance because the right unilateral activation of men is associated with greater spatial performance (Gur et al., 2000). The extent to which nature and nurture is involved in the development of spatial skills remains controversial (Turos & Ervin, 2000). Gender-based types of environment and activities engaged by individuals may also influence the development of spatial skill. Baenniger & Newcombe (1989) argue that males normally have the environments that provide them with more opportunities to engage in spatial task and are encouraged to participate in more activities that require the use of spatial skill than women.

Notwithstanding, the issues of lack of spatial skill and gender differences it is agreed that spatial ability can be acquired through proper instruction and training. To what extent spatial ability can be trained is an important point of discussion. Practice effects have also been detected in previous research for spatial training both for general and specific skills. Both genders were found to benefit from such training as demonstrated by improvement in spatial performance. Although both gender groups achieved considerable performance gain but the magnitude or level of gain was not comparably equal. Empirical evidence shows females through practice improved at a significantly higher rate because according to Baenniger & Newcombe (1989) men might be operating closer to their maximum potential. On the other hand Sorby’s (1998) study suggests that there is significant room for improvement of spatial skill for both men and women depending on the difficulty of the task.

PURPOSE OF STUDY

The purpose of this research was to examine the effectiveness of using novel method of training namely the Computer-Based Multimedia Tutor (CBMT) in enhancing the mental rotation involving undergraduates that enrolled in Information Technology major when compared to conventional instructional approach. Differences due to gender and prior spatial experience among the two main groups were also investigated. Possible interactions between various factors that may influence the effectiveness of spatial training programme were dealt with in this study.

The following predictions were formulated to address the issues as discussed in the main body of the preceding section namely (i) spatial training in the computer-based multimedia tutor (CBMT) will be more effective in terms of performance criteria of accuracy and speed than the conventional instructional method, (ii) a significant gender difference will be observed on the mental rotation measure both for accuracy and speed favouring males, and (iii) a significant difference attributed to prior spatial experience will be observed on the mental rotation measure both for accuracy and speed favouring participants with high prior spatial experience?

METHODS

Participants

A total of 58 undergraduates students majoring in Information Technology and Communication, Universiti Pendidikan Sultan Idris (UPSI), volunteered to participate in the experiment comprising 39 females and 19 males. Subjects deemed having high spatial experience were 23 (17 females and 6 males) and those with low spatial experience were 35 (22 females and 13 males). The participants were not randomly assigned as the experiment was conducted using two existing intact classes. Thirty-one and twenty-seven students from the first class and second class respectively volunteered and were given course credits for their participation. Albeit the constraints both the experimental and control groups were statistically equivalent in terms of age, ethnic composition, prior spatial experience and gender.

Materials

Individual differences in mental rotation were measured online at the PsychLab Website that contained the electronic mental rotation test developed by Chay (2000) based on the Vandenberg & Kuse (1978) mental rotation test. The electronic test version comprised 30 evaluation items each composed of a target figure and a comparison figure. The comparison figure could be either a congruent figure or an incongruent figure that was positioned at various degrees of orientation. All the 30-item questions were made up with one-half having congruent figures and the remaining half consisting incongruent figures as the correct answers. The scoring procedure yielded maximum possible correct scores of 30 and all questions were randomly presented to avoid order effect.

The mental rotation training was carried out by each participant on each personal computer with installed CBMT programme freely downloaded from the Visualization and Spatial Reasoning website (Blasko & Holiday-Darr, 2000). The CBMT was developed to specifically train in mental rotation where problems were presented in fashion normally found in the test. The interactive feature of the computerised tutor enabled feedback for trainees’ responses aided by animation of spatial solution. The categorisation of participants into high spatial experience and low spatial experience levels was achieved via the Spatial Experience Questionnaire (SEQ) designed by McDaniel et al., (1978). The SEQ comprised a list of twenty-five spatial activities such as drawing, map reading, and playing chess.

Procedure

This study employed a factorial quasi-experimental 2 x 2 x 2 design with pretest covariate and posttest measurement and three factor -- treatment group (two levels: experimental and control), gender (two levels: females and males) and prior spatial experience (two levels: high and low). This quasi-experimental design was chosen due to the non-randomised assignment to the treatment and control groups. The SEQ comprises a list of twenty-five spatial activities such as drawing, map reading, and playing chess. Each item has a four-point scale ranging from two extremes: very often and never to be rated by participants on how much they have participated in that activity. The two levels of spatial experience were categorised based on the total scores of the SEQ such that low experience was 42 and below, and high experience was 43 and above.

Figure 1 The online mental rotation test

At pretest, participants from both groups were evaluated in separate sessions to determine their entry level of the variables under study namely mental rotation (MR). All subjects were explicitly explained that their performance in the spatial test depend on both accuracy and speed factors. Prior to testing each individual was allowed to practice a few sample test items to familiarise with the instructions and interface. All subjects were allowed 3 minutes to complete the test. Raw data for the number of correct attempts and response time were automatically generated and can be retrieved for analysis in statistical programme.

During the intervention, training in CBMT to improve performance in mental rotation was applied to the experimental group whilst the control group received similar materials in conventional instructional approach. The latter group received equivalent materials in printed form for students to study and attempt mental rotation exercises. Participants in CBMT performed the tasks of mental rotation on varying degrees of difficulty and object complexity. Simple rotation tasks involved making decision on the correct rotational paths for simple objects that were orientated only one of the three principal axes. More complex tasks comprised problems dealing with objects of intricate shapes and rotational trajectory involving multi-axial rotations. Participants who encountered difficulties could interrogate the animation feature in this package to see the demonstration of the rotational movements of objects to reach the desired positions. Both authors acted as instructors for the experimental group and control group in two sessions throughout the interventional activities. The programme was carried out for 5 weeks where each group entered their classes for the treatment once per week with each session lasting for one-and-half hours.

Figure 2 A snapshot of the Computer Based Multimedia Tutor

Posttest was administered on the fifth week utilising the same electronic mental rotation test to reveal any performance gain resulting from the spatial training. The five-week gap between pre-testing and post-testing will ensure that any improvement for the measured performance is not attributed to individuals’ re-collection of the earlier administration but is due to the intended effect of spatial training.

FINDINGS

Accuracy of Mental Rotation Test

An analysis of covariance (ANCOVA) was used to adjust for differences in pretest scores. The analysis of the data was performed for independence, normality, homogeneity of variance, linearity, and parallelism as set by the conditions for using ANCOVA. All of the assumptions were appropriately met. The model utilised to conduct the ANCOVA was the General Linear Model (GLM). The first prediction of this study was that participants that engaged in spatial activities in multimedia tutor would perform significantly accurate than their counterparts using the conventional instructional approach. Data from the analysis of the 2 x 2 x 2 ANCOVA on the mental rotation posttest for accuracy (maximum possible score of 30) using the mental rotation pretest as the covariate are shown in Table1 revealing significant main effects for group treatment, F(1,49)=12.44, p<.01 favouring those who received training in CBMT.

The second predictions that participant with superior prior spatial experience would achieve better result than those that were less experienced in spatial activities. There was a main effect of prior spatial experience in mental rotation test for accuracy, F(1,49)=36.68, p<.01 where those with high prior spatial experience performed better than those with less.

Table 1 Group, spatial experience and gender analysis including

covariate adjustment for accuracy in mental rotation.

T

The third prediction was that males will outperform females in this spatial test of mental rotation. Predictably, a significant main effect of gender was observed revealing males to be more accurate than females by achieving higher total items correct in the mental rotation test, F(1,49)=10.77, p<.01. No two-way and three-way interactions among the three factors were found to be statistically significant.

Response Time of Mental Rotation Test

Data from the analysis of the 2 x 2 x 2 ANCOVA on the mental rotation posttest for response time (in seconds) using the mental rotation pretest as the covariate are shown in Table 2 revealing no significant main effect for group treatment, F(1,49)=.486, p=.489. The first prediction of this study was that participants that engaged in spatial activities in multimedia tutor would perform significantly faster than their counterparts using the conventional instructional approach did not materialise as predicted. Both groups were statistically equal for the gain of performance achieved through training.

Participants with superior prior spatial experience were predicted to achieve better response time than those that were categorised as possessing low prior spatial experience. There was a main effect of prior spatial experience in mental rotation test for speed, F(1,49)=10.52, p<.01 where those with high prior spatial experience performed quicker than those with less.

Table 2 Group, spatial experience and gender analysis including

covariate adjustment for speed in mental rotation.

Female participants were assumed to be outperformed by male participants in the third prediction by performing slower. As predicted, a significant main effect of gender was observed favoring males where they were much faster than females, F(1,49)=19.11, p<.01. Interaction between method of instructions and prior spatial experience was found to be significant, F(1,49)=4.87, P<.05. and interaction between method of instructions and gender was bordering significance, p=.06. No other two-way and three-way interactions were found to be statistically significant.

DISCUSSION

The findings insofar suggest that spatial training through novel method used in the study namely CBMT was more effective than conventional classroom practices only for accuracy criterion not speed criterion. It implies that the participants became more accurate in the performance of mental rotation tasks through training using CBMT than conventional method of instructions. Participants from the latter method have also improved but their performance gain was less than the former method. However, the superiority of CBMT was not replicated for speed factor as both groups using different training programmes have achieved performance gain whose difference was not statistically significant. In other words, participants from both groups beneffited equally from training regardless the method of instructions used. This finding concurs partly with the first prediction for accuracy but not speed performance of mental rotation using the innovative method.

The finding of gender difference indicates there was evident that males outperformed females in mental rotation both for accuracy and speed. Males were generally more accurate and faster after the interventional treatment. Female participants lagged behind male counterparts where the former were less accurate and slower in performing the spatial test thus confirming with the second prediction. No interactions were found for the three factors in the analysis of performance gain for accuracy criterion that infers training with CBMT was not implicated by influences attributed to gender and prior spatial experience. Both male and female subjects could achieve positive gain through training in CBMT by becoming more accurate compared to training in conventional method regardless of their spatial experience background.

As expected, prior spatial experience factor seemed to influence the performance of mental rotation where the participants with high spatial experience gained superior performance both for accuracy and speed. Also present were an interaction between group and prior spatial experience and to a certain extent an interaction between group and gender for speed performance in mental rotation. It was observed that females with high prior spatial experience in experimental group were faster than females with similar level of spatial experience in the control group. However, females with low prior spatial experience in experimental group were slower than females with similar level of spatial experience in the control group. Surprisingly, males in experimental group were slower than males in control group irrespective of their level of spatial experience. These observations further implicate the design and implementation of spatial training programme in mental rotation as intended improvement of performance especially in making them to solve spatial problems quicker depends on interrelated variables such as method of instruction, gender and background of previous experience. Looking at the summative pattern of these results compels educators and trainers to carefully select the best instructional strategy that learners can best adapt to using it effectively.

As demonstrated by the findings the improvement of one’s performance of mental rotation for accuracy was not quite the same with making the participants quicker. It appears that by engaging in the mental rotation activities in CBMT participants may have developed the cognitive strategy to solve mental rotation tasks where they were able to infer the correct rotations of objects with better precision. This ability was utilised in mental rotation test which enabled the subject to mentally rotate an object until the mental image was in congruence with the target object to determine a match or mismatch. Their understanding was further enhanced by simulation capability of the CBMT that allowed trainees to view a stepwise demonstration of the correct rotation for the desired position. This feature was very helpful to participants that were initially overwhelmed by the complex tasks involving intricate object shapes and multiple rotations. In general CBMT has made spatial training effective in enhancing participants’ spatial skills by becoming more accurate. The result supports Dixon’s (1995) finding that learning environment which utilises computer-based dynamic instructions can bring positive effect as students explore the geometric concepts of reflection and rotation. The same effect did not materialise for improving participants’ response time in mental rotation. It seems that the 5 week training programme of CBMT may not be sufficient to make them quicker in solving spatial problem related to mental rotation than those who received conventional mode of training. Two propositions are suggested that either the duration of training be extended (more contact hours per week added or extra weeks added) or another spatial training activities dedicated to improve speed performance is undertaken.

In summary the spatial training conducted in the CBMT has been very effective in improving the performance of students in mental rotation task in terms of accuracy. The implementation of an interventional programme needs to address the delicate background of its potential users. The diversity of potential users dictates that for any interventional activities to have an impact its design and implementation entails addressing issues related to gender, prior experience and other socio-cultural factors. The former factor is getting more emphasis as our society is becoming more egalitarian of late. These findings must be taken with caution as the sample size used in the research was relatively small and the nature of the quasi experimental design employed will preclude any generalisation. Notwithstanding these constraints the experience learned is beneficial for future research to refine methods or strategies to improve students spatial abilities that are important in both mundane activities and academic pursuit.

ACKNOWLEDGEMENT

The authors would like to thank the laboratory technicians for their help in the installation of the training modules and software plug-ins to run the programme.

REFERENCES

Baartmans, B.G. & Sheryl S. ( 1996). Making connections: Spatial skills and engineering drawing. The Mathematics Teachers, 89, 348-353.

Baenniger, M. & Newcombe, N. (1989). The role of experience in spatial test performance: A meta analysis. Sex Roles, 20, 327-345.

Ben-Chaim, D. (1989). Adolescents' ability to communicate spatial information: Analyzing and effecting students' performance. Educational Studies in Mathematics, 20, 121-146.

Bertoline, G.R. (1998). Visual science: An emerging discipline. Journal for Geometry and Graphics, 2(2), 181 – 187.

Blasko, D. & Holiday-Darr, K. (2000). The visualization assessment and training homepage. http://viz.bd.psu.edu/viz/.

Chay, J.C. (2000). 3D mental rotation test. http://www.uwm.edu/People/johnchay/mrp.htm.

Clements, D. H. & Michael T. B. (1992). Geometry and spatial reasoning. In D. A. Grouws (Ed.). Handbook of Research on Mathematics Teaching and Learning. New York: Macmillan Publishing Company, pp. 420-464.

Crano, W.D. & Johnson, C.D. (1991). Facilitating reading comprehension through spatial skills training. Journal of Experimental Education, 59(2), 113-127.

Deno, J.A. (1995). The relationship of previous spatial visualization ability. Engineering Design Graphics Journal, 59(3), 5-17.

Dixon, J.K. (1995). Limited English proficiency and spatial visualization in middle school students’ construction of the concepts of reflection and rotation. The Bilingual Research Journal, 19(2), 221-247.

Eckstrom, R.B., French, J.W., Harman, H.H. & Derman, D. (1976). Kit of Factor-referenced Cognitive Tests. Princeton, NJ: Educational Testing Service.

Eals, M. & Silverman, I. (1994). The hunter-gatherer theory of spatial sex differences: Factors mediating the female advantage in location memory. Ethology and Sociobiology, 15, 95-105.

Eliot, J. & Smith, I. M. (1983). An International Directory of Spatial Tests.Slough, England: NFER-Nelson.

Ferguson, E.S. (1977). The mind's eye: Nonverbal thought in technology. Science 197(4306), 827-836.

Gallagher, S.A. (1989). Pedictors of SAT mathematics scores in a causal model of mathematics achievement. Journal for Research in Mathematics Education, 15(5), 361-377.

Gardner, H. (1983). Frames of Mind: The Theory of Multiple Intelligence. New York: Basic.

Gur, R., Alsop, D., Glan, D., Petty, R., Swanson, C., Maldjian, J., Turetsky, B., Detre, J. & Gee, Gur, R.(2000). An Fmri study of sex differences in regional activation to a verbal and spatial task. Brain and Language, 74, 157-170.

Khairul Anuar Samsudin, Abdul Samad Haniff & Maizatul Mohd Yatim. (2003). Improving pre-service teachers’ spatial ability using Web-based virtual environment. Proceedings of the Malaysia Educational Research Association. Five Decades of Educational Research and Development: Taking Stock and Moving Forward. Bangi, Selangor, Malaysia, 11-13 September.

Lehmann, W. (2000). Group differences in mental rotation. Magdeburger Arbeiten zur Psychologie, 2, 1-18. http://www-e.uni-magdeburg.de/methpsy/

Leopold, C., Renata A.G. & Sorby, S.A. (2001). International experiences in developing the spatial visualization abilities of engineering students, Journal for Geometry and Graphics Volume, 5 (1), 81–91.

Linn, M.C. & Petersen, A.C. (1985). Emergence and characterization of sex differences in spatial ability: A meta-analysis. Child Development, 56, 1479-1498.

Lord, T. & J. Rupert. (1995). Visual-spatial aptitude in elementary education majors in science and math tracks. Journal of Elementary Science Education, 7(2), 47-59.

Masters, M.S. & Sanders, B. (1993). Is the gender difference in mental rotation disappearing?. Behavior Genetics, 23, 337-41.

McDaniel, E., Guay, R., Ball, L., & Kolloff, M. (1978). A spatial experience questionnaire and some preliminary findings. Paper presented at the Annual Meeting of the American Psychological Association, Toronto, Ontario.

Meeker, M. (1991). How do arithmetics and maths differ? Gifted Child Today, 14(1), 6-7.

Miller, C.L. (1996). A historical review of applied and theoretical spatial visualization publications in engineering graphics. The Engineering Design Graphics Journal, 60(3), 12-33.

NCTM (National Council of Teachers of Mathematics). (1989). Curriculum and Evaluation Standards for School Mathematics. Reston, VA:NCTM.

Pellegrino, J.W., Alderton, D.L. & Shute, V.J. (1983). Understanding spatial ability. Educational Psychologist ,19, 239-253.

Pribyl, J.R. & Bodner, G. M. (1987). Spatial ability and its role in organic chemistry: A study of four organic courses. Journal for Reasearch in Science Teaching, 24(3), 229-240.

Salthouse, T.A., Renee, L., Babcock, E., Skovronek, D., Mitchel, R.D. & Palmon, R. (1990). Age and experience effects in spatial visualization. Developmental Psychology, 26, 128-136.

Sorby, S. (1998). Developing 3-D visualization skills. Paper presented at the Annual Meeting of the Engineering Design Graphics Division, Madison, Wisconsin.

Turos, J. & Ervin, A. (2000). Training and gender differences on a Web-based mental rotation task. The Penn State Behrend Psychology Journal, 4(2), 3-12.

http://viz.bd.psu.edu/viz/conferences/Training-and-Gender.pdf

Vandenberg, S.G. & Kuse, A.R. (1978). Mental rotation, a group test of three-dimensional spatial visualization. Perceptual & Motor Skills, 47, 599-604.