An Analysis of Web-Based Instruction in a Neurosciences Course*

Chris Candler, MD and Robert Blair, PhD
University of Oklahoma College of Medicine.

Introduction

    During the spring semester of 1997, 150 first-year medical students at The University of Oklahoma College of Medicine were introduced to web-based instruction in a newly offered neurosciences course. This course was selected as a pilot project for the medical school's new computer-aided instruction initiative. This initiative, which grew out of a student-lead computer interest group, sought to enhance the educational process while developing student's basic computer and information retrieval skills. The project began with several planning sessions, which led to a massive computer and network installation in the medical school education building. Concurrently, students and software developers began construction of a web-based syllabi, atlas, and lab manual for the Neurosciences course.

    The new course consisted of approximately eight hours of traditional neurosciences lectures per week supplemented with 10 laboratory sessions throughout the semester. During their laboratory sessions, students were responsible for identifying traditional gross brain structures in addition to 614 structures on digital cross-sectional images. Each image highlighted one structure at a particular section level of the CNS. The image bank and other ancillary course material were accessible via the college's curriculum web site. Students could access the Neurosciences Web Site via the school's computer labs or from workstations at home. Although mandatory computer ownership was not in place at that time, it was strongly encouraged by the administration.

    Proper interface design was an important consideration in the development of this online course. The quantity and complexity of the material necessitated a logical interface and intuitive site navigation to create a successful learning environment. In the previous neuroanatomy course, students memorized structures using an atlas of neuroanatomy and monochrome 2x2 slides. The black and white atlas sections contained many numerically labeled structures that crowded the image, obscuring many features. Conversely, the online course covered 40 CNS cross-sections, which outlined each structure on a separate image. Thus, when presented with an unlabeled image of a particular CNS section, the students could click through a list of 15 or so structures which, when selected, loaded that same image section with the structure clearly outlined. The entire web site included over 600 images, each highlighting a separate structure.

    The web user interface was designed to clearly display images and the list of relevant structures. HTML frames were used to simultaneously present the list of structures and the CNS cross-section image on the same screen. Through the use of frames, students could select a structure from the text frame which loaded the new image into the image frame. A third navigation frame was added so students could quickly jump to various parts of the online course.

    In addition to the atlas of CNS structures, the Neurosciences Web Site contained specific information on various tracts, a detailed tutorial on basal ganglia disorders, and several lesion localization problems. The localization problems were popular with students, as they required students to integrate their knowledge of neuroanatomy with the appropriate clinical manifestation. Using the aforementioned frameset, students were presented with a clinical case, which required them to select from one of six images of CNS lesions. If they choose the incorrect lesion image then the text frame displayed the neuroanatomical manifestations of the image they selected.

    A faculty member provided the CNS section images in kodachrome slide format. After all the images were acquired, medical students digitized them with a slide scanner, which was capable of digitizing slides in high resolution. These images were then carefully labeled with the popular graphic-editing application Adobe Photoshop where they were subsequently saved in a compressed JPEG format. This widely used format allowed high quality images to be reduced to 50-150 Kilobytes in size. Although many image formats are available to instructors, the JPEG format is presently the de facto standard for displaying full color images on the web.

    To provide medical students with the highest degree of availability, the web site was designed to work in either of the two common web browsers (Microsoft's Internet Explorer and Netscape's Navigator) and on either a Mac or a PC computer. Such flexibility allowed students to view the online course from a variety of hardware and software configurations. The only technical requirement involved was that the machine had to be capable of displaying graphics in at least 16-bit color (thousands of colors).

    Students could access the online course through a variety of means. They could use the standard computer labs on campus or use their home machines by dialing into the campus modem pool. At that time, the campus modem pool consisted of 20 36.6K modems connected to the campus network through which the students could then access the web server. Several months after the course the modem bank was expanded to roughly 200 modems.

    Web-based resources that are merely accessory to existing coursework are used little by students.⁸ To maximize the educational value and student use of these web resources, practical examinations were partially comprised of the online material.⁹ Although students used pen and paper to record their answers, they had to identify structures from computer sections and gross brain sections. During this practical exam the class scored an average of 96% in identifying structures from digital computer images and 88% in identifying the traditional gross brain structures. Predictably, during the three days preceding the exam, the web server registered over 200,000 "hits" from the entire class (approximately 1300 per student). Simply defined, a "hit" is recorded any time a student requests a web page or a new image from the web server.

    Although true comparative studies are needed, the above finding suggests that online web-aided instruction is a valid learning and testing modality. Informal evaluations from both students and faculty were favorable. It was clear that students wanted more of their curriculum online. Students also requested more of the lesion localization problems and other problem-based learning exercises to increase their exposure to the clinical sciences.

    While the computer lab machines downloaded the images almost instantly due to the speed of the campus network, students at home often complained that their modem connection was prohibitively slow for the larger images. In subsequent courses, CD-ROM's were created that contained all the images from the online course site. Students are allowed to check out a CD-ROM overnight and transfer the images to their home computer's hard drive. After this, students may speedily review the images from their hard drive without the bandwidth constraints of a modem. Schools looking to implement web-assisted instruction to existing coursework should consider this CD-ROM solution and construct their web site in such a way that it may easily be transferred to a CD-ROM.

    Despite careful design and development of the online course interface, students said it felt backwards to them. Instead of selecting the name of the structure, which brought up a labeled image, students have asked that the interface be revised so that it would present them with an image outlining a structure that they must identify. Witnessing how pairs of students used the material reinforced the need for this change. One student would be the examiner and cover up the structure he was selecting while the examinee, after reviewing the labeled image would have to identify the structure. Students felt this was a more natural way to study for the exam.

    Student and faculty feedback will guide future improvements in this and other online courses. Presently, the online component of the Neurosciences course is under revision for the upcoming spring semester. Many of the lab sessions have been reorganized, requiring changes in the online format. Gross brain images and radiographs have been added to the cross-section images, which should present students with a more holistic sense of neuroanatomy.

    The portable and accessible nature of web-based materials may eliminate the problems often seen in traditional computer-aided instruction. Advantages to web-assisted instruction include increased availability of materials, ease of web course construction, and the potential for a more interactive learning environment. Disadvantages include hardware and software costs of web servers and development equipment, instructor's time, and the pressure for students to own computers. It is difficult to quantify the exact costs required for this project since it necessitated the installation of a massive computer and networking infrastructure, which was needed for other reasons.

    Clearly, the experience with the online Neurosciences course was successful, both in terms of objective outcomes and subjective feedback. More comparative studies are needed to evaluate the efficacy and cost-effectiveness of this exciting new medium. It should not concluded that web-assisted instruction is a replacement for instructors or lectures. However, careful design and planning can make such materials an excellent complement to existing courses.

1. Faughnan, JG, Elson, R. Information technology and the clinical curriculum: Some predictions and their implications for the class of 2003. Acad Med 198;73:766-9
2. McAuley, RJ. Requiring students to have computers: Questions for consideration. Acad Med 1998;73:669-73
3. Keane, DR, Geoffrey, RN, Vickers, J. The inadequacy of recent research on computer-assisted instruction. Acad Med 1991;66:444-8
4. Cohen, PA, Dacanay, LS. Computer-based instruction and health professions education. Eval Health Prof 1992;l 15:259-81.
5. Kulik, JA, Kulik, CC. Review of recent research literature on computer-based instruction. Contemporary Educ Psychol 1987;12:222-30.
6. Kulik, CC, Kulik, JA, Shwalb, BJ. The effectiveness of computer-based adult education: A meta-analysis. J. Educ Computing Res 1986;2:235-52.
7. Niemiec, R. Walberg, HJ. Comparitive effects of computer-assisted instruction: A synthesis of reviews. J. Educ Computing Res 1987;3:19-37.
8. Bahr, JP. Computer-aided instruction in perinatal education. Am J Perinatology 1986;3:147-150.
9. Friedman RB. Top ten reasons the world wide web may fail to change medical education. Acad Med 1996;71:979-81.

Candler, C, Blair R. An analysis of web-based instruction in a neuroscience course. Med Educ Online [serial online] 1998;3:3. Available from URL http://www.utmb.edu/meo/