By Dr. Heidi Day
This portfolio describes an upper division neuroscience class focused on critically thinking about data presented in journal articles.
Background
NRSC 4062 Neurobiology of Stress is an upper division elective in neuroscience. The course is a traditional lecture class that covers the fundamental responses of the body and brain to stress; how various systems of the body are impacted by stress, and how individuals may be vulnerable versus resilient to stressful experiences. As part of the class, students are required to read research articles. Over several semesters of teaching this class, it became clear that students were primarily reading the text of the articles, and mostly ignoring the data. I was surprised to learn that many students have significant difficulty interpreting graphical data, and significant gaps in comprehension of experimental design. Thus I modified the course to include the following course goal: Develop critical thinking skills specifically as they pertain to assessment and interpretation of primary scientific research literature.
Implementation
When reading primary scientific literature, ideally, students will be able to: 1. Assess and interpret data in a primary research article without relying on the author’s interpretation of that data. 2. Evaluate the limitations of the data. 3. Formulate additional research questions. 4. Construct a simple experimental design to answer one of those research questions. Assessment of students in these areas was ongoing, and culminated with a “Journal Article exam”, where students were provided with an original primary scientific research article, without title, abstract or discussion. They were required to provide a title and abstract for the article, answer various questions about the data presented in the article, and provide follow-up research questions and an experimental design to answer one of those questions. To prepare for this assessment, published experiments with their accompanying graphs were chosen to illustrate specific points in regular lectures. These experiments and graphs were completely deconstructed during class discussions, including explanations of the experimental design and group by group interpretation of the data. Clicker questions were asked in class, and “low stakes” homework assignments included multiple choice questions based on interpretation of graphical data explained in lecture, and aspects of experimental design. Journal article assignments were reduced to just four, but more class time was devoted to discussion of the data, including student presentations of the data. Interim assessments included free response exam questions to describe and interpret graphical data previously seen in lecture, and to design experiments based on a hypothesis related to concepts learned in class.
Student Work
Questions based on either interpretation of graphical data or experimental design (with or without showing hypothetical data) were asked in 3 consecutive exams. The grading of these questions revealed good student growth in both of these areas over the course of the semester, especially for students initially scoring in the lower tertile. The mean grades for the Journal Article exam were consistent across semesters: 77% (fall 2016), 77% (spring 2017) and 78% (summer 2017). Survey results showed that students recognize the importance of understanding how to interpret graphical data, and felt their ability to do so had improved. Similarly, they reported their ability to think critically about journal articles had improved. Students generally reported they were more likely to think of additional scientific questions and would be better able to design an experiment to answer those questions. Finally, students felt better equipped to critically assess scientific claims made by the media.
Reflections
Overall I was very pleased with the growth shown by students in the area of graphical interpretation and experimental design. I was also happy the survey data were largely very positive, with students generally feeling they had improved in these specific areas. The journal exam grades were lower than I would prefer, and individual answers revealed specific areas of weakness that will be addressed more specifically in future semesters. I would like to provide additional real-world relevance by expanding this critical thinking exercise to include assessment of a stress-related claim made by the media.