Finding Success in Productive Failure

by Thao K. Huynh,  PGY2 Oncology Pharmacy Practice Resident, University of Maryland School of Pharmacy

Learning is an activity that requires our undivided attention. Learning a novel or unfamiliar concept often involves struggle, which is a necessary component to critical thinking. A teaching method that causes a greater amount of struggle will lead to more learning as exemplified by productive failure.

According to Manu Kapur, productive failure engages students in differentiating prior knowledge from features of the new concept.¹ In using productive failure as a teaching strategy, there are two phases. The first phase requires students to generate and explore, while the second phase encourages them (with the help of the teacher) to consolidate and assemble new knowledge. In the first phase, teachers must discern what students know about a novel concept that hasn’t yet been formally taught. During this phase, a complex problem is presented collaboratively amongst students. During the second phase, the novel concept must be consolidated and presented with direct instruction in a structured way.²

In order to test this method, Kapur conducted a classroom-based research experiment comparing direct instruction with productive failure. In an all-boys school in Singapore, seventy four ninth grade students were taught about the mathematical concept variance in two different ways. Students were tested prior to the intervention to evaluate their pre-existing knowledge and it was similar at baseline in both groups. In one classroom students were provided direct instruction (DI).  In the other classroom, students in the productive failure (PF) group were provided no instruction or guidance. The two classes differed in that the DI class of 35 students participated in four 55 minute periods of instruction where the teacher explained the concept of variance, modeled the application of the concept by working through problems with the class, and highlighted pitfalls and misconceptions. Students in this class were able to work through problems in triads followed by a discussion of the solutions by the teacher. DI students were assigned homework after each class to reinforce the topic. In contrast, the PF class was not provided direct instruction during the first two periods. Instead, students worked in triads to solve one of the problems on their own. In periods three and four, PF students were provided the same direct instruction but the teacher also compared and contrasted student solutions and explained the solution in a manner similar to the DI class.  The PF class were given and solved fewer problems than did students in the DI class.   Moreover, students in the PF class were not given homework. Following this period of instruction, both groups of students took the same exam.  All students performed similarly overall – however, students in the PF group outperformed students in the DI group on the conceptual insight and data analysis portions of the exam.¹

Although, it can be argued that students in the productive failure group outperformed the direct instruction group due to a greater amount of time devoted to to studying the concepts; their performance is likely related to the amount of time and energy put forth into delving critically into the topic.

Productive failure can be used in the pharmacy curriculum and clinical pharmacy practice to help enhance learning experiences.  Incorporating productive failure may work in a group setting such as Abilities or Practice Lab, case discussions, and problem-based learning courses.  Students in these small learning groups naturally (without direct instruction) have some prior knowledge on the topic.  Productive failure will facilitate development of their critical thinking skills.  Allowing student to struggle is critical to the process.  This sets the stage for the instructor to introduce new concepts to students. This environment, according to Collins, leads “metacognitive scaffolding, contrasting cases, peer-interaction scripts, mixed-ability groups, and perhaps representational scaffolding in the intervention phase will enhance student learning in the [direct] instruction phase.”³ Furthermore, a lack of struggle can lead to detrimental outcomes.  For example, in a study that observed nurses on a patient care unit, when problems arose, instead of analyzing the situation, they opted for quick (but less than optimal) solutions.⁴

I believe productive failure is most useful when there is sufficient time to explore different solutions to a complex problem. If time is critical, productive failure may be detrimental. In addition, for productive failure to have the greatest benefit, learners should have some baseline knowledge on the topic.  In other words, the new concepts must be related in someway to the students prior knowledge.  If the new concepts are too foreign, students might not be able see the connections between their prior knowledge and the new knowledge.  I’ve encountered a recent situation where productive failure would not have worked well. An oncologist wanted to use equine thymoglobulin for a critically ill patient with steroid refractory acute graft-versus-host disease post stem cell transplant.  The drug needed to be administered as soon as possible. Not being an expert in this topic and having never seen horse thymoglobulin used before, I needed to clarify this with my preceptor. I found a review article on the topic but my preceptor needed to provide guidance. In this situation, using productive failure as a strategy to learn a novel concept without any instruction would have caused further delay in patient care as well as the potential for patient harm. On the other hand, productive failure worked with two students I asked to lead a topic discussion. The topic was acute myeloid leukemia, which neither student had any previous knowledge. I began with few patient cases for the students to read and discuss with one another.  After the students had an opportunity to struggled through the patient cases, I provided a formal one-hour interactive presentation. I measured the success of how well the students applied the information to direct patient care situations.  I saw a dramatic increase in knowledge from both students and they were able to provide appropriate recommendations to the patient care team.

Looking for ways to improve the way you teach?  Embracing the productive failure method will facilitate learning by encouraging learners to differentiate new concepts from prior knowledge and to be analytical in applying these new concepts.

References

1. Kapur M. Productive failure. Cognition and Instruction. 2008 Jul;26(3):379-424.
2. Kapur M, Toh PLL. Educational design research – Part B: Illustrative cases. Enschede, the Netherlands: T. Plomp, & N. Nieveen (Eds.); c2013. Chapter 17, Productive failure: From an experimental effect to a learning design; p. 341-355.
3. Collins A. What is the most effective way to teach problem solving? A commentary on productive failure as a method of teaching. Instr Sci 2012;40:731-735.
4. Edmondson AC. Learning from failure in health care: frequent opportunities, pervasive barriers. Qual Saf Health Care 2004;13(Suppl II):ii3-ii9.