Building capacity through open approaches: Lessons from developing undergraduate electrophysiology practicals

UNAM

UNAM is the largest public university in Latin America¹⁴. As of 2019-2020, over 360,000 students were enrolled at UNAM, including more than 217,000 undergraduates and 30,000 graduate students¹⁵. The university has 129 undergraduate and 41 graduate degree programs. Also, in 2019 UNAM served over one million students through its continuing education, including online, programs¹⁵.

As a public insitution, education at UNAM is nearly free, subsidized by federal funds. Students pay an annual registration fee of just 20 Mexican cents (equivalent to ∼0.01 USD). This is combination with UNAM’s prestige and reputation for quality education results in a high demand for entry. Each year, less than 10% of applicants are accepted at the undergraduate level through UNAM’s admissions testing¹⁶. In other words, a huge percentage of the eligible student population in Mexico is unable to study at this university that receives the largest share of public funds – an annual budget equivalent to approximately 2 billion USD^17,18. One could argue that, more than any other public institution in Mexico, UNAM has a responsibility to give back to the community. On the other hand, while UNAM receives more funds than other public universities in Mexico, it still operates on a relatively limited budget considering its size and the number of services offered by the institution. For comparison, consider the University of California, which has a similar though smaller population of over 285,000 students¹⁹ but almost 20 times the budget of UNAM²⁰. So, how can institutions like UNAM maximize the use of public funds, both for their benefit and that of the larger Mexican population?

Faculty of Science

UNAM comprises 15 faculties, 34 institutes, and various other centers, schools, and units¹⁵. There are fundamental differences for academics working in faculties versus institutes, which are important for understanding our work as professors in the Faculty of Science.

In institutes, the primary focus is research. Laboratory space is assigned to many faculty at the time of hiring and their teaching load is low. According to the UNAM Statute of Academic Personnel, researchers in institutes must teach a minimum of 3 contact hours per week each semester²¹, equivalent to a 1-1 teaching load at Canadian or U.S. institutions. In contrast, faculties are focused on teaching. Entry-level professors are required to teach a minimum of 9 contact hours per week each semester²¹, equivalent to a 3-3 teaching load. Unlike at many institutions in North America, there are no standard mechanisms for ‘buying out’ of teaching if a professor receives a grant. In addition, professors are expected to contribute significantly to ‘formation of human resources’ by directing student social service projects and theses, serving on committees, and tutoring. Many professors work almost exclusively with undergraduates, especially for the first few years when their professoriate level does not allow advising graduate students in many degree programs. Despite the heavy teaching and service load, there is still a research expectation. However, professors are not necessarily assigned laboratory space and receive no start-up funds. Availability of laboratory space in the Faculty of Science has become especially problematic as student and academic population growth puts increasing demands on an already overloaded infrastructure.

These conditions raise a number of questions for professors working in faculties like UNAM’s Faculty of Science: With limited resources and infrastructure, how do I provide high-quality, hands-on educational experiences for my classes?; how do I develop research projects for social service and thesis students?; and how do I build up my own research program and start producing?

Electrophysiology equipment

All electrophysiological recording equipment was obtained from BYB (backyardbrains.com), including devices to record APs in insects (Neuron SpikerBox), and EMGs (Muscle SpikerBox) or ECGs (Heart and Brain SpikerBox) in human subjects (Table 1). We purchased these as bundles, which included the recording device, cables, surface electrodes, conductive gel, and other accessories. The low cost of these bundles (<$250 USD each), compared to conventional electrophysiology equipment, allowed us to purchase multiple devices. With 3 or 4 devices, we could work in groups of 5-10 and pilot practicals with classes of 20–30 students.

We also purchased DIY kits from BYB to build additional recording devices. This served two purposes. First, students will assemble the devices, learning valuable instrumentation skills in the process, which is one of the core objectives of our Biomedical Physics plan of study. Students will fully document the assembly process with step-by-step protocols, photos, and videos, which will be shared online as OERs. Second, at around half the price of the fully assembled device bundles, DIY kits allowed us to buy more equipment without exceeding our budget. Once assembled, our recording capacity will double, meaning we can work with more students.

Affordability was not the only advantage of the BYB equipment. The small size and portability of the devices meant we did not need a dedicated laboratory space, solving one of our infrastructure issues. We could take these devices into any classroom and record with students using their smartphones. We also allow students to borrow these devices and take them home to work on individual research projects. A few years ago, having students do electrophysiology at home would have been impossible. Now we can offer them this unique experience, which can be a huge motivating factor in their academic development. Since 2017, students and professors in our program have used the equipment in core and elective coursework, social service projects, and thesis research. In other words, purchasing a small amount of equipment has greatly increased our capacity to provide high-quality educational and research opportunities for our undergraduates.

Electrophysiology accessories

We purchased several accessories to improve both recording experiences for students and potentially research capacity (Table 2). For example, while the BYB electrodes that come with the Neuron SpikerBox are sufficient for basic AP recording in large insects, they are stainless steel sewing needles with a relatively large tip diameter (0.25–0.6 mm)³⁷, non-insulated, and not ideal for finer recordings in smaller preparations or cells. So, we purchased insulated Tungsten electrodes with a 2–3 µm tip diameter. At ∼$19 USD each, these electrodes are only $9 more than BYB’s, but should provide a substantial improvement in recording capabilities and quality. For less than $200 USD we can buy a packet of 10 Tungstens and upgrade 10 SpikerBoxes.

We also purchased manipulators to improve control and precision of electrode placement. Conventional 3-axis manual micromanipulators, like those made by Narishige (usa.narishige-group.com), cost ∼1,000 USD and were out of our price range. However, BYB provides a 3-D printed plastic manipulator with 3 axes of movement in the millimeter range and adjustable electrode angle through 135 degrees for just $99.99. Furthermore, BYB’s open hardware approach means the plans for printing and building the manipulators are available on their website, which will allow us to reduce costs in the future by printing more manipulators at a university facility. In fact, the growing open labware/maker movement is increasingly allowing researchers to 3-D print their own lab equipment, including electrophysiology devices and accessories, for a fraction of the cost^38,39.

We also bought accessories from a local provider (SIET México), including Arduino kits, sensor kits, Raspberry Pi 3 Model B, and Raspberry Pi Cameras Module V2. Arduino kits include an Arduino Uno R3, servo and step motors with drivers, a variety of sensors (infrared, humidity, temperature), and other accessories such as cables, resistors, and LEDs. Sensor kits are designed to be used in conjunction with Arduinos, and include heartbeat, temperature, touch, and sound sensors, as well as buzzers, joysticks, and switches. Similar Arduino and sensor kits can be purchased through Amazon or eBay. Kit components can be used for a variety of electrophysiology-related projects, including instrumentation of simple myoelectric prosthetic prototypes^40,41.

Dissection tools and microscopes

Electrophysiology often involves dissection to prepare tissues or cells for recording. Dissection tools, especially those for fine dissection, are costly. Fortunately, companies like VWR provide economic solutions in the form of classroom dissection sets, which include scissors, forceps, scalpels, pins, and more. With tools for up to 20 students and priced at _∼$200 USD or less, the cost comes out to only _∼$10 USD per student. With the remaining funds we had for tools, we bought just two fine dissection kits for use in more advanced, individual student projects.

Dissection and fine detail instrumentation also requires visualization and magnification. With this in mind, we purchased several microscopes with different characteristics. The Fisherbrand Illuminated Pocket Microscope weighs just 85 grams, measures 140L × 38W × 22H mm, and has 60–100x magnification. Similarly, the BYB High Power RoachScope weighs 400 grams, measures 142L × 94W × 74H mm, and can be used in combination with any smartphone camera. With digital zoom, it has 5–100x magnification. Both these microscopes are designed for maximum portability, so they can be taken into the classroom. In addition, both cost less than $100 USD each, so we could buy several to work with groups of students. We also purchased three Fisher Science Education Advanced Stereomicroscopes for just under $300 USD each. These microscopes are not very portable, but should give us better optics. To increase the utility of these microscopes, we are planning on 3-D printing a low-cost adapter that will attach to the eyepiece and allow us to mount any smartphone to take high-quality pictures or video. Open plans for such an adapter are available via the NIH 3D Print Exchange⁴² and pictured in 39.

Finally, we purchased an advanced trinocular microscope (National Optical via Fisher Scientific) with 4x, 10x, 40x, and 100x objectives and a built-in digital camera (Moticam 1080 HDMI & USB) for high-resolution viewing of tissues and cells. The higher cost of this microscope meant we could only buy one. However, connecting the camera to a large computer monitor allows us to carry out demonstrations and have groups of students view samples simultaneously. We have also hosted “open house” events for new students using this microscope.

EMG basics: recording from the body’s lever systems

The first practical is designed to teach students the basics of EMG recording, carried out at the end of the musculoskeletal system module. The background written information reinforces physiology concepts seen in class, as well as the application of basic physics concepts seen in other coursework. It begins with a description of how muscle-bone-joint complexes function as lever systems. Students are encouraged to think back to the three types of classical lever system and find corresponding examples of these in the human body. This involves visualizing biomechanics and how the relative position of bones, joints, muscles, and loads will affect movement. The written documentation goes on to reinforce concepts such as how muscle structure affects tension development, length-tension relationships, and the energy requirements for muscle contraction. We then describe the basics of EMG recording, comparing the advantages and disadvantages of invasive versus surface recording, and the basic bipolar differential recording configuration. Study questions prompt students to think about where they will need to place electrodes to record from different muscles and what potential limitations they might encounter.

Students then move on to the experimental phase of the practical where they carry out their own EMG recordings, in groups of 4-6 students depending on class size. Step-by-step instructions on how to perform the recordings are included in the written documentation. However, these are designed to be informative without being too prescriptive and still allowing for exploratory learning. The only requirement is that students record from at least one muscle from each type of lever system, but we do not tell them which muscles to record from or how they should activate these muscles. Students are encouraged to design their own experiments using everyday items available in the classroom or simple exercise aids, like resistance bands or hand grippers, brought from home. Students have performed EMGs from facial muscles while eating, tricep muscles while doing pushups, bicep muscles while arm wrestling or lifting their backpacks, and forearm muscles while performing martial arts movements. Students are also encouraged to explore different types of contraction, including intermittent versus sustained and increasing versus decreasing force.

We have several other EMG practicals still under development, including experimental ones to measure fatigue in the bicep muscle, dual recordings from antagonistic muscle pairs, simultaneous recording of EMG and force sensor measurements from the forearm, and others. In addition, we are developing EMG data analysis practicals. Students will take the recordings they gathered in the first practical, graph them, and learn about the design and application of band-pass and low-pass filters to process their data. They will learn about different techniques used to smooth data, calculate an envelope, and use thresholds to detect start and stop times of muscle contractions. These practicals will be carried out using Jupyter notebooks running Python, thereby simultaneously strengthening students’ programming skills. All practicals are listed in the extended data (doi.org/10.5281/zenodo.4540355⁶⁵).

ECG basics: recording heart electrical activity before and after exercise

In module 3 of their human physiology course, students learn about the cardiovascular system and carry out a practical to record their ECG before and after exercise. The background written information begins with a description of how the heart performs external mechanical work. Students are encouraged to visualize the heart as a single-chamber pump with inflow and outflow valves, and examine the pressure-volume relationships similar to the way one would with an internal combustion engine⁶⁶. Students learn about sequential pressure and volume changes in different chambers of the heart during the cardiac cycle, and how to graph this with a pressure-volume loop. The documentation goes on to describe the electrical activity of specialized populations of cells in the heart, including the ionic basis of APs in these cells. Discussing cardiac muscle activity also encourages students to think back to module 2 of the human physiology course when we discussed contraction mechanisms in this muscle type. Finally, we describe the basics of ECG recording, including how the summation of individual potentials leads to the extracellularly recorded events, different recording configurations, and the importance of electrode placement.

Students then move on to the experimental phase, working in groups of 4–6. Volunteers from each group record their ECGs, while other students help with organizing and exporting the data. Students first record their baseline ECG under resting conditions for at least 1–2 minutes. Then, they disconnect the recording device while leaving the electrodes in place and perform light to moderate exercise for at least 5 minutes. After this, students reconnect the device and record their ECG again for at least 1–2 minutes. Students can choose the type of physical activity they perform. For example, students have done push-ups or burpees, ran laps around the building, or gone up and down stairs outside the classroom. We encourage students to compare how different levels of activity change the ECG signal, and how the signal varies across subjects (e.g., athletes versus non-athletes).

We are developing additional experimental practicals designed to explore the relationship between heart and respiratory activity, using simultaneous ECG and spirometry, for example. We are also working on ECG data analysis practicals. Students will take the recordings they gathered during the first ECG practical, graph them, and learn how to detect the peaks of the QRS complex to calculate heart rate and quantify how it changes after different levels of exercise. They will also examine techniques for detecting the P and T waves, and calculating intervals important in clinical evaluations. A full list of these practicals is available in the extended data (doi.org/10.5281/zenodo.4540355⁶⁵).

Recording accessory muscles during normal and forced respiration

In module 5 of their human physiology course, students learn about the respiratory system, an important part of which is understanding the mechanics of breathing. How do respiratory muscles expand or contract the thoracic cavity and change pressure gradients? How does the participation of different muscles change when respiration is normal versus forced? And to relate back to the musculoskeletal module, how is respiratory muscle contraction related to electrical activity?

In this practical, students record EMGs from the rectus abdominis. This muscle is known as an accessory respiratory muscle because it is not activated during normal exhalation, but is activated during forced exhalation when additional effort is needed to reduce the volume of the thoracic cavity beyond that accomplished by simple elastic recoil⁶⁷. While recording rectus abdominis EMG, students simultaneously use a spirometer (Vernier) to measure the volume of air moved in and out of the lungs. Students are instructed to perform a sequence of normal breaths interspersed with maximal forced inhalations and exhalations.

The dual recordings allow students to see firsthand that, during normal respiration and forced inhalation, little to no electrical activity is recorded on the EMG because the rectus abdominis is not contracting. However, during forced exhalation, the EMG shows an increase in both the amplitude and frequency of the signal with increased effort and increased volume exhaled. The written materials for the practical are designed to reinforce several physical concepts applied to the study of respiration, including: (1) pressure-volume relationships and Boyle’s Law as applied to the lungs; (2) importance of pressure gradients and Ohm’s Law as applied to airflow; (3) Poiseuille’s Law as applied to measuring airflow through a spirometer, and (4) biomechanics of active lung expansion versus passive elastic recoil. This practical also gives students the opportunity to integrate knowledge from two modules to understand how the musculoskeletal and respiratory systems work together. We are working on developing more practicals that combine electrophysiological recordings with other physiological measurements (e.g., from force, displacement, or gas sensors) to provide similar integrative learning experiences.

Incorporating electrophysiology into undergraduate education

Less than a decade ago, providing hands-on electrophysiology learning experiences for undergraduates, especially large classes, was not feasible. However, over the last few years, technological advancements have opened up new possibilities for educators. With the introduction of the BYB Neuron SpikerBox in 2011, an easy-to-use, low-cost bioamplifier brought neurophysiology into the classroom³⁷. Since then, the single-channel SpikerBox, and the later two-channel version, have been used to design practicals for undergraduates to record from cricket sensory organs⁶⁸, grasshopper neurons responding to visual stimuli⁶⁹, and to study AP conduction velocity in earthworms⁷⁰. Surveys from these studies indicate that students not only enjoy these hands-on activites, but that they also improve learning outcomes, increasing test scores by as much as 25% on average⁷⁰. The SpikerBox has even been used as part of a larger program to provide undergraduates the opportunity to teach neuroscience to highschool students⁷¹.

In recent years, BYB has released more complex devices for recording ECG, EMG, and single-channel EEG, which have also been used in undergraduate class settings to improve learning. For example, Catena and Carbonneau (2018) describe using the BYB Muscle SpikerBox Pro to record dual-channel EMG as part of an undergraduate biomechanics course⁷². Their survey results show that students reported “better motivation" and higher “personal responsibility for learning". Test scores for students who had these hands-on learning experiences were also 7% higher compared to students who did not⁷². Similarly, Judge and colleagues (2020) used BYB equipment to develop ECG and EMG exercises for community college anatomy and physiology courses⁷³. Students who carried out these exercises showed “significant learning gains"⁷³.

Other groups have also developed and shared plans for low-cost electrophysiological recording devices, and in the process created instrumentation exercises for undergraduates. Matsuzaka and colleagues (2012) describe the development of a low-cost (only $85 USD per unit) amplifier for recording EEGs, EMGs, and other electrophysiological signals with students⁷⁴. Importantly, the authors mention that potential problems of reproducibility and quality control when building these devices “could be resolved if the optimized circuit layout is freely available” (pg. A124). Crisp and colleagues (2016) provide step-by-step instructions for students to build a simple EMG device using a breadboard amplifier, with few components and an assembly time of just 30 minutes⁷⁵. Wyttenbach and colleagues (2018) review these and other devices as part of a larger discussion on “reducing the cost of electrophysiology in the teaching laboratory”⁷⁶.

It is not just low cost that is important, but moreso the open approaches taken that have increased the impact of many projects like the ones described above. Sharing hardware schematics and building instructions, openly licensing and publicly documenting code, and growing online support communities – characteristics of projects like Arduino, BYB, and Raspberry Pi – have allowed classrooms and laboratories in limited-resource countries to build capacity⁷⁷.

Connections between open approaches

One of the most interesting aspects of this project for us has been working at the intersection of open education and open research, and experiencing firsthand how these approaches can build on one another. Open access, open data, open education, and open source have historically different developments, and are often treated as separate areas of advocacy. However, all these areas share common goals: (1) increased access to information, whether in the form of a textbook, an article, a data set, or code; (2) increased participation, whether in education, research, citizen science, or software development; and (3) better outcomes, whether that means better learning outcomes, more reproducible research, or improved software. How can these open approaches learn from each other and work together to further these goals? In particular, in a limited-resource environment, we wondered whether open educational approaches could also help us build research capacity, and whether open research approaches could also help us create OERs. In our opinion, the answer to both of these questions is a resounding ‘yes’.

The clearest example of this for us was in observing the connections between open education and open source. When we thought about code as not just for research but also an educational resource, it changed how we thought about sharing this product. Previously, we might have simply shared our code as a raw Python file in a GitHub repository, and included some in-line comments and a README file as documentation. However, when we envisioned students or educators reusing our code to learn, we realized it needed more in-depth explanations and exercises. Using Jupyter notebooks, we built up tutorials or lesson plans surrounding the code and transformed these into OERs^52–54. Importantly, after completing these resources we not only had quality OERs to use for our classes, but we also had a bank of well-documented analysis tools to use for future research projects. Organizing and documenting our code in this way may help us with lab group onboarding, as any incoming students can go through the tutorials and quickly get up to speed on our data analysis techniques. Furthermore, in the process of elaborating didactic explanations of our analysis, sometimes we realized ways in which our code could be improved. Therefore, the interaction went both ways: building educational capacity through open practices led to building research capacity and vice versa.

We have visually mapped out some of the potential connections between different open approaches (Figure 2). While this is not an exhaustive map – other open approaches could be included and other connections explored – it is a representation of how these connected for us in this project.

Figure 2. Concept map for how various open approaches connected for us in this project.

Importantly, one open approach did not automatically lead into the next; there were transformations of the materials and certain conditions that needed to be met at each stage to maintain the flow between each. For example, simply sharing our data would not necessarily allow others to develop new analysis code (Figure 2, upper right arrow). For this to occur, the data need to be well organized, labelled, and documented, with meta-data included. Admittedly, we are still struggling with the best ways to do this to optimize reuse of our data, and believe this is one area where researchers would benefit from more training. Similarly, shared code does not necessarily become an OER (Figure 2, central left-pointing arrow). This requires that the code be well explained, often with a surrounding lesson plan and exercises. Open licensing at each stage was also key, since locking down content at any point would stop the flow. However, licensing is different for code, data, and documents. To select licenses for each product, we used resources like the Creative Commons License Chooser (creativecommons.org/choose) and GitHub’s Choose an Open Source License tool (choosealicense.com).

We would also like to encourage researchers to expand their ideas of what they consider an OER. When working with students, there is no clear line where education ends and research begins. The research we do with students, especially with undergraduates, is not necessarily to discover new things but rather to teach students how to do research. It is more about the process than the end result, and as such, everything we create during that process – protocols, code, data, notebooks – can potentially be transformed into an OER to train others. We also believe that in thinking of research as a teaching-learning process, with all the documentation and explanations that entails, we may in turn enhance the research itself, improving experimental design and reproducibility.

Libraries leading in open practice and funding open projects

We are not the first to think about the potential connections or intersections between different open approaches. For years, libraries have been at the forefront of conceptualizing, creating, and managing all kinds of open content^78–80, and thinking about how open practices might connect. The following are all projects led by librarians and information specialists, and/or based in libraries. In 2015, Atenas and Havemann published a book⁸¹ arguing that “while Open Data is not always OER, it certainly becomes OER when used within pedagogical contexts” (pg. 22), and presented five case studies where open data were used to teach students programming skills, data literacy, and even promote civic engagement. Elder^82,83 and Walz⁸⁴ have looked at the differences between open access and open education, but in the process also found areas where these overlap and where they can learn from one another. For the last few years, Virginia Tech libraries has been hosting a series called “Connecting the Opens”, where they invite experts to discuss possible connections between open practices (recordings found at VTechWorks vtechworks.lib.vt.edu). Makerspaces, which often combine aspects of open hardware, open source software, and open education, are increasingly being established and run by libraries^85–87. We hope to see even more of this intersectional work in coming years, and expect that much of it will arise in libraries.

Libraries are also increasingly both leading and funding open scholarship projects, including the development and implementation of OERs. In a 2016 survey of U.S. universities, 64% responded that it was the library who had originated affordable course content (ACC) or OER initiatives at their institutions⁸⁸. For those with governing bodies overseeing these initiatives, 89% said that libraries were participating members and half said that libraries led the group. Over half of respondents also indicated that funding for ACC/OER initiatives came from library general operating budgets – more than any other institutional or external funding source.

Despite library support for open initiatives, it seems other institutional policies have not necessarily caught up. Walz and colleagues⁸⁸ write, “survey responses indicate that current university-wide tenure and promotion policies do not explicitly encourage faculty adoption, adaptation, or creation of ACC/OER” (pg. 5). We believe it was important for the success of this project, as well as our own professional development, that our department recognizes and values participation in PAPIME projects in annual performance, promotion, and tenure reviews, and gives us space on evaluation forms to report on non-traditional digital products, including OERs. We encourage institutions to rethink and reform their evaluation policies to incentivize open scholarship, including OER development and adoption, and to seek guidance from libraries on how best to do this.

Underlying data

Source repository: GitHub. Electrophysiology practicals for undergraduate students. https://github.com/emckiernan/electrophys.

Archived at time of publication -

Zenodo: electrophys v1.0.1 http://doi.org/10.5281/zenodo.4554420⁴⁹

Our GitHub repository, archived via Zenodo, contains all the lesson plans for our electrophysiology practicals, raw electrophysiology data, and data analysis code.

License depends on resource type. Practicals (documents) are shared under the Creative Commons Attribution 4.0 International (CC BY 4.0) license; code under the MIT License; and data (recordings) under the Creative Commons CC0 1.0 Universal license. For more information, see our license file.

Extended data

Zenodo. Electrophysiology practicals for undergraduates: links to code, data, and lesson plans (Version v1.0). http://doi.org/10.5281/zenodo.4540355⁶⁵

This extended data includes a pdf document with a full list of our practicals (developed and under development) with links to the resources.

Data are available under the terms of the Creative Commons Attribution 4.0 International license (CC-BY 4.0).