6.1: Introduction

Austin Lim, PhD (DePaul University)

Claire Sun (Mount Sinai)

Editor: Katie L. Willis, Ph.D. (University of Oklahoma)

---

Most of the information you learn in neuroscience textbooks were discovered by people who applied the scientific method to systematically answer a research question. In order to come to their conclusions, these neuroscientists used a variety of techniques, borrowed from biology, medicine, chemistry, physics, engineering, psychology, and many other academic disciplines. Neuroscience researchers use the wide range of scientific methodological strategies to answer: “How does the nervous system work?”

In this chapter, we will describe a few of the methods that are used to ask and answer questions in neuroscience. These methods can be divided roughly into four major categories: Imaging anatomy, imaging function, imaging cells, and finally manipulating the nervous system. Within each of these categories, the techniques will start with the “big picture” view, and by the end of each section we zoom down to the level of molecules and genes.

Just a minor caveat about these techniques. Many of these strategies can be used independently, but some may be used simultaneously.

Figure 6.1 Being able to zoom in allows us to see small details, but information from the big picture is missing.

There are three major components that can be described for some of the techniques:

1. A description of how the method works. Each of these methodological strategies rely on using the knowledge from other fields of study and applying that knowledge to a question about the nervous system. Knowing how the methods work allow us to decide what types of data can be collected and when each method will be appropriate to use.

2. Some of the main advantages of the technique. Not every method is applicable for studying every question. Instead, it would be beneficial for neuroscientists to choose the most appropriate strategy for a particular research question given the circumstances.

3. Some shortcomings or limitations of the technique. As above, no strategy is perfect. Methods that look at the “big picture” aren’t able to look at microscopic level detail. There are limitations for every technique and being aware of these limitations inform us about caveats in interpreting data.

Although almost all these techniques can be used in either humans or non-humans, some techniques are best adapted for humans and others for non-humans. Using a technique on an inappropriate species can lead to poor results. For example, humans can lay still for tens of minutes at a time during an imaging scan session, while rats or mice will not be able to do this simple behavior (without restraint or anesthesia). Before getting started with our list of research methods, we should ask ourselves “Is it better to study the question in humans or in non-human animals?”

About resolution

Many of the imaging techniques are concerned with resolution. Resolution in neuroscience is very similar to resolution in digital photography or computer monitors. A picture taken in high resolution contains more information than one in low resolution, and can therefore be more valuable. Imaging strategies consider two different types of resolution, spatial and temporal.

Spatial resolution refers to the ability to differentiate two points in space from each other. An imaging method with high spatial resolution means that two signals very close together can be identified as two different signals instead of one. Spatial resolution is usually measured in units of distance or volume. Electron microscopy (described in 6.3.1) offers the highest spatial resolution of the imaging techniques.

Figure 6.2 Methods with higher spatial resolution are better for identifying the location of different parts of the image. Precisely identifying the parrot’s beak in the leftmost image (low resolution) is difficult, but is easy in the rightmost image (high resolution).

Temporal resolution refers to the ability to distinguish two events in time from one another. Temporal resolution is measured in units of time. Functional imaging techniques with high temporal resolution, such as electrophysiology, can differentiate two signals as close together in the hundreds of microseconds range.

Figure 6.3 Methods with higher temporal resolution are better for precisely detecting when a signal occurs. With low temporal resolution techniques (top of detected signals), two discrete events spaced apart by time appear as one event.

The case for human participants

The main reason we conduct studies in humans is to understand the human nervous system. It is knowledge of the human condition that leads to the development of therapies that prevent, cure, or otherwise improve the quality of patient’s lives. And while curing diseases in non-human animals is the purpose of veterinary medicine, curing humans is the primary end goal of biomedical research, and that goal is only truly met by testing on humans. A cure developed for rodents might give us a clue about the human case, but without seeing how well it works in humans, the therapy won’t save human lives.

Humans are great at following directions without much training, unlike non-humans - as anyone who had tried to house train a puppy can attest. We can follow simple sets of directions that untrained non-humans find difficult, such as “lie still”. During functional activity scans, we can perform extremely complex higher-order cognitive tasks, such as “imagine that we give you 20 dollars, and we will take away a dollar when you answer a question incorrectly” (rats and chimpanzees have no concept of currency). We can describe our feelings in writing and self-report on our symptoms, both of which are tasks that are impossible in non-humans.

Believe it or not, recruiting human participants for neuroscience or psychology studies is usually cheaper than studying that same question in nonhuman animal models. Collecting data from undergraduates is sometimes free, since many are required to participate in some number of hours of psychology studies to pass introductory level classes. Although the cost per patient differs depending on the nature of the study, many patients get paid as little as $10 an hour. On the other hand, housing colonies of mice, rats, or monkeys gets very expensive, very quickly.

The case for non-human subjects

But, it is often impossible to test every question in humans. Instead of always studying humans, scientists often use non-human model organisms, the most common organisms being the worm C. elegans, fruit flies (Drosophila melanogaster), zebrafish (Danio rerio), song birds, mice, rats, and macaque monkeys. The closer we move towards the human, the more similarities the model organism shares with us. Of the commonly used model organisms, macaque monkeys are the non-humans that are most similar to humans. We share 93% of our genetic material with macaques, but we still have different metabolic and physiological processes, and our behaviors are much different from theirs.

Figure 6.4 A short list of animal models that are often used in neuroscience research and their genetic similarity to humans. Humans are 99.9% genetically identical to other humans.

Ethical constraints prevent us from performing experiments that may cause physical or psychological harm if performed in humans. We would never conduct a test on humans to assess what concentration of neurotoxin leads to brain damage (these experiments aren’t done very frequently in nonhumans anyway.) Invertebrates, such as worms and fruit flies are not heavily regulated by ethics oversight committees, allowing scientists to conduct a wider set of experiments on these animals.

Experimental preparations

Performing an experiment in an intact, living organism, whether human or nonhuman, is described as an in vivo (Latin meaning “within life”) preparation. The main strength of this strategy is that the data collected here are more predictive of the human condition, which is one of the main goals of biomedical research. However, the in vivo preparation has challenges, because thousands of variables within a living system are uncontrolled or still unknown. There are also very strict ethical limitations on the nature of experiments that can be done in vivo.

On the other hand, an in vitro (Latin meaning “within glass”) preparation is an experiment performed on cultured cells or isolated molecules of DNA, RNA, or protein. These preparations have the opposite strengths and weaknesses of in vivo preparations. They allow for extremely good control over variables, but the results are less reliable in translating to a therapy. The regulations on these experiments are much more lax compared to in vivo experiments; most of the regulatory guidelines are to protect the experimenter rather than the patient or the experimental subject.

Falling in between these two preparations is an ex vivo experiment. In this kind of experiment, a section of the living organism is taken, such as a slice of brain, a tissue biopsy, or a detached frog leg. The strengths and limitations of these experiments are somewhere in between that of the other two preparations.

Nonhuman animals perform all kinds of behaviors that humans never experience. These behaviors are unique to nonhumans and can only be studied in nonhumans of course. We will never be able to understand flying (birds) or slithering (snakes) as a means of locomotion by studying humans.

One thing that makes human studies difficult is how weird we are. We all come from very different backgrounds, bringing our own set of experiences, flaws, and biases into every research study. Your current mood and mind state can alter behaviors profoundly - just ask any one who gets “hangry.” In nonhuman studies, many of these variables are closely monitored. When we raise model organisms in the lab, almost every aspect of their life is controlled, from their food source, living conditions, and their daynight cycles, which eliminates several sources of variability that affect human studies.

Another factor that makes human studies challenging is that we are WEIRD: The acronym meaning that most participants in psychology studies are Western, Educated, and from Industrialized, Rich, and Democratic countries. One study estimates that about 96% of psychology studies use WEIRD subjects, whereas these people are only 12% of the total global population. This creates a bias in sample selection, since WEIRD people perform differently on behavioral tasks compared to others. Even our susceptibility to visual illusions is affected by our WEIRD-ness.

Figure 6.5 Many WEIRD people are susceptible to the Muller-Lyer illusion, and we perceive the lower line to be longer than the top line even though they are the same length.