Saskia Klink is a PhD student at the University of Bayreuth, who is collaborating with Adrienne Keller. Recently, Adrienne traveled to Germany to work with Saskia on several research projects in her lab. This post is an outgrowth of their discussions together during that time. Check out their previous post to learn more about their collaborative research!
“I have not failed. I’ve just found 10,000 ways that won’t work.” – Thomas A. Edison
Thomas Edison recognized that failures, setbacks, and missteps are central to the scientific process. Creative and thoughtful scientists learn how to reshape such moments to become stepping stones to bigger successes down the road. Yet, when final successes are splashed across mainstream media outlets and dominate peer-reviewed journals, the multitude of failures that breed the flashy headline are rarely acknowledged. Even within scientific circles, scientists tend to dance around discussions of failures and setbacks. This mentality is not only disingenuous to the scientific process, but also slows scientific progress when the wheel to success is reinvented time and again across labs using similar techniques. Here, we provide a glimpse under the hood of how we have experienced the scientific process in all its glory (or lack thereof?).
Recently, we experienced a series of “mini-failures” while troubleshooting a frequently-cited method for staining tissues of fungi, a technique that allows us to see these tissues clearly under a microscope. While the published protocol appears to be simple and straightforward, we quickly learned that successful staining is just as much of an art as it is a science, requiring more nuance than described in the published literature. The protocol shows textbook-quality images that clearly differentiate between fungal and plant cells. Meanwhile, our initial attempts yielded samples that had drastically too much or too little stain; our best luck at finding Goldilocks was a bluish blur of fungal tissues that was vaguely distinct from the host plant cells. After talking with other researchers who are experienced with these approaches, we learned to adapt the protocol based on the plant species we were using and its root characteristics. Methodological setbacks can be particularly challenging and disheartening, especially for students just starting out in scientific research. We think one way to better support and retain new scientists is better communication among ourselves – if we more thoroughly document and share the nuances and challenges of specific techniques with each other in formalized communication channels, we will all benefit. Students will gain confidence and knowledge, and we can spend our time banging our heads to solve new methodological challenges rather than reinventing the wheel.
Setbacks in field research requiring multiple years of data collection can be particularly stressful. Several years ago, we installed hundreds of tubes in a forest near Indiana University to quantify how much plant-derived carbon was being transferred to the soil and used by microbes to fuel soil processes. We then waited more than a year to give nature time to run its course and provide meaningful data. When we analyzed a subset of our sample tubes in the field last May, we gained confidence that the method was working and excitedly planned for a full sampling in the fall. After a month of field work and two more months of lab work (not to mention those years of waiting), we finally had a full data set to analyze! This is usually one of the most exciting and anticipatory stages of a project for a field ecologist. Enter stage left: failure. Contrary to our smaller testing data set, the full data set showed too much variation within any given treatment to be able to detect meaningful soil carbon patterns. While the method appeared to work well in our test sampling, it didn’t hold up during our larger sampling campaign. Although disappointing, this wasn’t a dead-end, but rather– in the spirit of Thomas Edison – the discovery of one of 10,000 ways that won’t work. We developed some hypotheses of what went wrong (e.g., the method may be highly sensitive to environmental variables, such as precipitation or the length of time the tubes were in the ground before sampling), and we’ll likely be out in the field again soon to test these hypotheses further.
These examples from our experience working together underline the fact that the scientific process is a spiral – a succession of trial and error, failure, slow progress, and baby-steps to bigger successes. For students, it’s important to recognize that trial and error is a natural part of the scientific process. We’ve found that supporting each other through our failures (exchanging witty GIFs depicting failure can be an effective antidote) and celebrating small successes makes the process spiral forward more effectively, rather than spiral downwards towards a sense of doom.
Edited by Kat Munley and Evan Arnet
