Personality preference influences medical student use of specific computer-aided instruction (CAI)
© McNulty et al; licensee BioMed Central Ltd. 2006
Received: 10 June 2005
Accepted: 01 February 2006
Published: 01 February 2006
The objective of this study was to test the hypothesis that personality preference, which can be related to learning style, influences individual utilization of CAI applications developed specifically for the undergraduate medical curriculum.
Personality preferences of students were obtained using the Myers-Briggs Type Indicator (MBTI) test. CAI utilization for individual students was collected from entry logs for two different web-based applications (a discussion forum and a tutorial) used in the basic science course on human anatomy. Individual login data were sorted by personality preference and the data statistically analyzed by 2-way mixed ANOVA and correlation.
There was a wide discrepancy in the level and pattern of student use of both CAI. Although individual use of both CAI was positively correlated irrespective of MBTI preference, students with a "Sensing" preference tended to use both CAI applications more than the "iNtuitives". Differences in the level of use of these CAI applications (i.e., higher use of discussion forum vs. a tutorial) were also found for the "Perceiving/Judging" dimension.
We conclude that personality/learning preferences of individual students influence their use of CAI in the medical curriculum.
Computer-aided instruction (CAI) has become an increasingly important component of the medical curriculum due in large part to the development of Internet applications and the ease with which curricular content is distributed through networks [1–4]. The rapid increase in development of educational software and the more recent explosion in information databases available through the Internet have provided easy access to educational materials that have enhanced a student's abilities to learn either in small groups or individually with increased efficiency, better outcomes, and at reduced costs .
However, our data indicate that students are not uniformly making use of computer resources. Recent studies have revealed a wide disparity in utilization by individual students [6, 7], which was attributed to differences in student attitudes toward computer technology . Evidence in favor of this hypothesis was provided by a follow-up study showing that the degree to which individual medical students accessed the medical school network was related to their personality preferences as measured by the Myers-Briggs Type Indicator (MBTI) test .
The MBTI has been used extensively to measure the personality profiles of medical students [9–12] and describes eight preferences within four separate dimensions [13, 14]. "Introversion" vs. "Extroversion" is the dimension that describes a person's focus of attention and source of energy, whether from within or from the outside world. "INtuition" vs. "Sensing" is the dimension that describes how an individual processes information either by focusing on the relationships between facts or the facts themselves. "Feeling" vs. "Thinking" is the dimension that describes whether decisions are made subjectively and personally or objectively and logically. The final dimension, "Perceiving" vs. "Judging", determines whether an individual's preference is to be spontaneous and flexible or decisive and orderly. The MBTI test is now generally accepted as a useful tool to help predict learning styles as well .
This study was designed to further test the hypothesis that personality preference is an important factor in the utilization of CAI in a medical curriculum. The specific objective of the study was to evaluate the extent to which personality preference influenced utilization of two different Web-based CAI applications developed for the M1 (first year) course in human anatomy. The two CAI applications differed in the level of user interactivity and the degree to which each application was directly applicable to learning objectives in the course.
First year undergraduate medical students who were subjects in the present study took the MBTI test through the Office of Learning Assistance. The median age was 23 (21–31 range) and 48% of the students were female.
One of the CAI applications tested was the "LUMEN Forum", which allowed asynchronous communications among course faculty and students utilizing WebBoard conferencing server software (http://Chatspace.com). The software included a variety of administrative tools for content management allowing delivery of multiple conferences that were individually tailored to address specific threaded discussions of general concepts and topics covered in the course. This application was highly interactive because the students contributed information whether in the form of text, or links to multimedia (images, videos, etc) as part of threaded discussions. The discussions were not always directly applicable to learning objectives in the course.
The second CAI application was "LUMEN Flash", a tutorial that provided a series of questions on a specific subject matter for review. This CAI was developed using ColdFusion (Allaire/Macromedia) and SQL (Structured Query Language) environments for delivering database driven applications through the Web. The applications created tables after each student access to record which "stack" of cards the student requested and which "cards" the student marked as correct or which cards/questions the student missed. This tracking allowed the student, upon returning to "LUMEN Flash" at a later time, to select only those cards s/he had previously not seen or had marked as being incorrect. The application included 7 general subject categories from which the students could select. The number of questions ("cards") in the categories ranged between 30 and 48 (mean = 39). The "LUMEN Flash" application was less interactive than the "LUMEN Forum" since students did not contribute to the program, but the scope of the categories were directly applicable to the learning objectives in the course.
Students were required to login to both CAI applications before use. Login data were stored for individual students in entry logs. The frequency with which students posted on the "LUMEN Forum" was determined directly from the postings by individual students in each conference. For the "LUMEN Flash" application, logins in which a student attempted to answer less than 10 questions in any session were excluded as incomplete sessions based on discussions with those students who explained that these sessions were usually terminated due to distractions or other time demands. All data were entered into Excel spreadsheets. Once the data for individual students were entered, the names of students were deleted from the database prior to further analyses in order to maintain the confidentiality of individual students. The study was exempted by the IRB.
Means and standard errors are reported for all of the grouped data. A 2-factor mixed ANOVA was used for data analyses. Pearson correlations (r) were also used to describe associations between continuous variables. The MBTI yields 16 separate personality groupings, but statistical analyses of these groupings were not conducted because of the very small sample sizes in many of the groups.
Frequency of logins for each of the four MBTI dimensions. Means (± sem) for utilization of the specific CAI applications tested in this study. MBTI preferences exhibiting higher values for each CAI are bolded to highlight trends.
ANOVA Main Effects
Introvert n = 37
13.2 (± 2.0)
9.4 (± 1.3)
p < 0.001
p = 0.831
Extrovert n = 79
13.8 (± 1.2)
9.6 (± 1.0)
iNtuition n = 64
11.1 (± 1.1)
8.2 (± 0.9)
p < 0.0005
p = 0.006
p = 0.240
Sensing n = 52
16.6 (± 1.8)
11.2 (± 1.3)
Feeling n = 77
13.5 (± 1.2)
9.7 (± 1.0)
p < 0.0005
p = 0.885
p = 0.768
Thinking n = 39
13.6 (± 2.1)
9.1 (± 1.3)
Perceiving n = 49
11.5 (± 1.4)
10.2 (± 1.3)
p < 0.001
p = 0.443
p = 0.032
Judging n = 67
15.1 (± 1.5)
9.1 (± 1.0)
There was a wide range in the frequency of logins among individual students ranging from 0–49 for the "LUMEN Forum" and 0–38 for the "LUMEN Flash". The overall frequencies of logins for each of the CAI applications were significantly different with the average access of the "LUMEN Forum" being higher (13.6 ± 1.0 sem) compared to the "LUMEN Flash" (9.5 ± 0.7 sem). Use of the "LUMEN Forum" was significantly higher regardless of the MBTI dimension (Table 1).
Correlation of individual logins to "Forum" and "Flash" CAI.Coefficients (r) and p values for correlations between the frequency of logins to the "LUMEN Forum" vs. the "LUMEN Flash" when individual students were sorted by personality preference.
When the data were sorted by personality preferences, 2-way ANOVA revealed an effect of MBTI preference for the "iNtuitive/Sensing" dimension. Students with "Sensing" preferences utilized the "LUMEN Forum" and the "LUMEN Flash" more frequently than those with "iNtuitive" preferences (Table 1). An effect of personality preference on student utilization of specific CAI was further indicated by interaction effects for the "Perceiving/Judging" dimension. In this case, "Judgers" tended to prefer the "LUMEN Forum" application to a much greater degree than the "LUMEN Flash" compared to "Perceivers" (Table 1).
Correlation between logins and postings to the "LUMEN Forum".Coefficients (r) and p values for correlations between the frequency of logins vs. postings to the "LUMEN Forum" when individual students were sorted by personality preference.
Three general observations are gleaned from the results of our study. First, there was a wide disparity in student use of CAI that was specifically developed to facilitate instruction of the subject matter. Second, there was a generally strong correlation in the level of individual use of respective CAI and this association was not dependent on MBTI personality/learning preference (with the possible exception of the "Thinking" preference). Third, the level of CAI use was influenced by MBTI personality/learning preferences.
To our knowledge, this is the first study that has examined effects of personality/learning preferences on utilization of specific CAI in the context of the medical curriculum. The finding that students with a "Sensing" preference tended to log in more frequently to both CAI is noteworthy because "Sensing" individuals characteristically focus on facts and conceptualizing through practical applications, which are traits consistent with the nature of both applications. The "LUMEN Flash" was mostly factual and the "LUMEN Forum" frequently provided practical applications of important anatomical concepts. Further effects of MBTI personality/learning preferences on the level of use of the respective CAI were found when comparisons were made between CAI (e.g., "Judgers" tended to use the "LUMEN Forum" more frequently than "LUMEN Flash").
Our observation that the "Sensing/iNtuitive" dimension influenced use of CAI is consistent with a report by Ahn  who found that students with "Sensing" preferences were more satisfied with CAI used in distance education. Friend and Cole  also reported that "Sensing/Thinking" individuals responded best to CAI compared to "iNtuitive/Feeling" students. In another study, the "Thinking/Feeling" preference was the most significant dimension affecting use of CAI and the development of declarative knowledge . Smith et al.  reported that "iNtuitive/Thinking" types of teachers were more likely to use technology than the "Sensing/Feeling" types. The "iNtuitive/Thinking" types also learned significantly better using CAI designed to teach language . Students with "Feeling" preferences had poorer attitudes toward technology and made more mistakes using CAI than those with a "Thinking" preference . In an earlier study, we also reported that students who accessed the medical education network more frequently tended to have a "Thinking" preference . The results of the present study cannot be compared directly with our previous study because that study did not examine utilization of any specific CAI; it only measured general use of the computer. Much of the variability reported in the literature probably relates to differences in the paradigms used in the evaluations.
With the advent of Web-based instruction, which promotes non-linear interactions with most CAI, research has begun to focus on how individual learner differences influence the use of these instructional paradigms (cf. ). The wide disparity in frequency of CAI use among individual students was found in an earlier study in which computer use by individual medical students was quantified . Coates and Humphrey  also reported that students in an economics class exhibited similar variability in use of on-line practice quizzes and discussion boards. In this case, 18% never attempted any of the practice quizzes and only 25% of students read more than a quarter of all postings in the discussion forum. Although our study suggests that innate personality/learning preferences may explain differences in the level of use of CAI by students in the medical curriculum, it does not exclude many other factors (e.g., background, prior experiences. cognitive styles, etc), which presumably also have an influence.
Our study did not specifically address important questions related to summative evaluations, but the findings are relevant assuming that the degree of utilization of specific CAI applications is correlated with outcomes on in-course examinations as demonstrated in a previous study . An important advantage of the experimental paradigm for future studies is that measurements of effectiveness of CAI as related to performance in class (knowledge retention) are fully integrated into the day-to-day real-life curriculum of the students.
We conclude that personality/learning preferences affect the level of use of CAI specifically developed for use in the medical curriculum. More broadly, our results are linked to the overall theory that instruction is most effective when it "fits" with the individual student's needs. Because a student's approach to learning predicts academic achievement, it is of obvious importance to tailor computer applications to the individual student's intellectual and psychological profile using CAI that is founded in basic principles of learning and instructional design. This concept was emphasized by Pocius , who stated that, "Understanding the effects of personality variables on computer use can be used to improve the quality of human-computer interaction. The awareness of which personality traits introduce individual differences in human-computer interactions can illuminate ways in which a particular human-computer interaction task can be altered to accommodate different users."
We thank Cathy Lai for programming and Lora Grzywacz for secretarial help. Supported by the NIH/NLM (1 G08 LM06823-01) to JAMcN.
- Candler C, Blair R: An Analysis of Web-Based Instruction in a Neurosciences Course. Med Ed Online. 1998, 3: 3-6.Google Scholar
- Chodorow S: Educators must take the electronic revolution seriously. Acad Med. 1996, 71: 221-226. 10.1097/00001888-199603000-00009.View ArticleGoogle Scholar
- Daniel J: Computer-Aided Instruction on the World Wide Web: The Third Generation. J of Econ Ed. 1999, 163-174.Google Scholar
- Ward JP, Gordon J, Field MJ, Lehmann HP: Communication and information technology in medical education. The Lancet. 2001, 357: 792-796. 10.1016/S0140-6736(00)04173-8.View ArticleGoogle Scholar
- Fletcher JD: Evidence for learning from technology-assisted instruction. Technology Applications in Education: A Learning View. Edited by: O'Neil HFJ and Perez RS. 2002, Hillsdale, NJ, Lawrence Erlbaum Assoc., Inc., 370.Google Scholar
- Coates D, Humphreys. BR: Evaluation of computer-assisted instruction in principles of economics. Ed Tech Soc. 2001, 4: 133-144.Google Scholar
- McNulty JA, Halama J, Dauzvardis MF, Espiritu B: Evaluation of web-based computer-aided instruction in a basic science course. Acad Med. 2000, 75: 59-65.View ArticleGoogle Scholar
- McNulty JA, Espiritu B, Halsey M: Medical student use of computers is correlated with personality. JAIMSE. 2002, 12: 9-13.Google Scholar
- Brown F, Peppler RD: Changes in medical students' Myers-Briggs "preferences" between their first and fourth years of school. Acad Med. 1994, 69: 244-10.1097/00001888-199403000-00024.View ArticleGoogle Scholar
- Rezler AG, Buckley JM: A comparison of personality types among female student health professionals. J of Med Ed. 1977, 52: 475-477.Google Scholar
- Stillwell NA, Wallick MM, Thal SE, Burleson JA: Myers-Briggs type and medical specialty choice: a new look at an old question. Teach and Learn Med. 2000, 12: 14-20. 10.1207/S15328015TLM1201_3.View ArticleGoogle Scholar
- Wallick MM, Cambre KM: Personality types in academic medicine. J of Louis St Med Soc. 1999, 151: 378-382.Google Scholar
- Briggs Myers I: Introduction to Type. 1998, Palo Alto, CA., Consulting Psychologists Press, Inc., 43-sixthGoogle Scholar
- Keirsey D: Please Understand Me 3rd Ed. Albany NY: Prometheus Nemesis Book Co. 1978, 396.Google Scholar
- Campbell K: Learner characteristics and instructional design. [http://webxtc.extension.ualberta.ca/resources/resource_detail.cfm?Resource_Name=Learner%20Characteristics%20and%20Instructional%20Design]
- Ahn C: Relationship of personality types and learners' performance in computer-mediated distance education. Diss Abs Int. 2000, 60 (11-A).Google Scholar
- Friend CL, Cole CL: Learner control in computer-based instruction: a current literature review. Ed Tech. 1990, November: 47-49.Google Scholar
- Odom M, Pourjalali H: Knowledge Transfer From Expert Systems vs. Traditional Instruction: Do Personality Traits Make a Difference?. J of End User Comp. 1996, 8: 14-20.Google Scholar
- Smith B, Munday R, Windha R: Prediction of teachers' use of technology based on personality type. J of Instruct Psych. 1995, 22: 281-285.Google Scholar
- Meunier L: Human Factors in a Computer Assisted Foreign Language Environment: The Effects of Gender, Personality and Keyboard Control. CALICO J. 1996, 13: 47-72.Google Scholar
- Alberty MC: An analysis of the relationships among selected aspectes of an individual's psychological type, attitude toward technology, and ability to learn computer programming. Education. 1987, Pittsburg, University of Pittsburg, 270.Google Scholar
- Chen SY, Paul RJ: Editorial: individual differences in web-bsed instruction - an overview. Br J Ed Tech. 2003, 34: 385-392. 10.1111/1467-8535.00336.View ArticleGoogle Scholar
- McNulty JA, Halama J, Espiritu B: Evaluation of computer-aided instruction in the medical gross anatomy curriculum. Clin Anat. 2004, 17: 73-78. 10.1002/ca.10188.View ArticleGoogle Scholar
- Pocius K: Personality Factors in Human-Computer Interaction: A Review of the Literature. Comp in Human Behavior. 1991, 7: 103-135. 10.1016/0747-5632(91)90002-I.View ArticleGoogle Scholar
- The pre-publication history for this paper can be accessed here:http://www.biomedcentral.com/1472-6920/6/7/prepub
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.