Experiments as the gold standard for research: A new twist

Worldwide, there were more than 1.56 billion people who used Facebook daily in March 2019. There were “only” half as many daily active users in 2014, but typical adults in the U.S. already spent 40 minutes daily on Facebook, much of that using News Feed. In June 2014, researchers at Facebook published the results of an experiment conducted in 2012. For one week, they removed 10 to 90% of posts containing positive-sounding words from News Feed streams of about 155,000 randomly selected users and 10 to 90% of posts containing negative-sounding words from News Feed streams of another 155,000 users. For control groups, posts were randomly removed from News Feed streams. The researchers then assessed whether these treatments were associated with changes in use of positive and negative words in subsequent posts sent by the subjects of the experiment. The dating website OkCupid did a similar experiment at about the same time, by informing some pairs of users that they would be good matches when the site’s algorithm predicted that they wouldn’t or vice versa. These experiments generated much outrage among social media users and the commentariat. I learned this story from new research about attitudes of people toward experimentation, so I decided to provide some background about the role of experiments in science and then describe this new research. I’ll return to the Facebook and OkCupid experiments at the end.

In Chapter 4 of Tools for Critical Thinking in Biology (TCTB), I described experiments as the gold standard for research, in keeping with common practice. The dictionary definition of experiment is quite general, e.g., “a scientific procedure undertaken to make a discovery, test a hypothesis, or demonstrate a known fact.” I used a much more specific definition in explaining why experiments are the gold standard for research – experiments as randomized, controlled trials. My main examples in Chapter 4 of TCTB were two randomized, controlled trials of the effects of smoking marijuana on pain in human subjects. The methodology of these experiments was typical of a wide variety of clinical trials in medicine, from tests of new drugs to comparisons of different surgical procedures and much more. In a randomized, controlled trial in clinical medicine, the experimental subjects are people who volunteer to be randomly assigned to one of two or more treatment groups. For example, all individuals might suffer from psoriasis. Those assigned to one group would receive a standard treatment for psoriasis while those in a second group would receive a new drug. In this case, the first group would be a control group and the experiment would ask whether the new drug better alleviates the symptoms of psoriasis. The Facebook and OkCupid experiments were also randomized, controlled trials, except that subjects didn’t volunteer to participate.

The main purpose of randomly assigning treatments to volunteers in medical experiments is to protect against conscious and unconscious bias. If there were such bias, we couldn’t confidently attribute any difference in outcome to the difference in treatments that the subjects received because the treatments would be confounded with the source of bias. For an example of unconscious bias, suppose researchers assigned the first 25 volunteers to the new drug and the next 25 to the standard psoriasis treatment. It could be that individuals with more itching and pain would be more motivated to volunteer for the experiment and therefore predominate among the first 25 volunteers. If so, volunteers with more severe psoriasis would get the new drug while those with less severe psoriasis would get the standard treatment. The new drug might be effective, but there might be little or no measurable difference in outcome because the subjects treated with the new drug were in worse condition to start with.

Besides randomization of treatments to subjects and comparison to a control group, clinical trials are usually also “double blinded”, meaning that both volunteer subjects and researchers don’t know the treatment that each subject received until after the results are recorded and analyzed. This eliminates another potential source of bias: if the researchers wanted to show that the new drug was more effective, and if they knew which subjects got the new drug, they might be tempted to minimize the symptoms in their evaluation of these subjects.

Experiments are considered the gold standard for research because they can give relatively unambiguous answers to scientific questions, including questions of practical importance like how best to treat a certain disease. But experiments may not be feasible for some questions; for example, questions about processes that occur at large spatial or temporal scales. Why do more species of birds live in the tropics than at higher latitudes? How did flight evolve in the transition from dinosaurs to birds? In TCTB, I discussed several examples of research based largely on correlational and comparative data rather than experiments to illustrate how scientists tackle the complexity of causation. I used two main examples in Chapter 5, “Correlations, comparisons, and causation.” These examples asked the questions: Does use of cell phones increase risk of brain cancer? Does exposure to lead early in childhood contribute to increased crime when exposed children become young adults?

Experiments may also raise ethical questions, as illustrated by the 2014-2016 epidemic of Ebola in West Africa. Ebola is extremely contagious and this epidemic killed 11,000 people, about 40% of those infected. Several potential vaccines were under development when the epidemic began, but none had been rigorously tested for efficacy. How do you design a randomized, controlled trial of a vaccination that might, or might not, save individuals from bleeding to death within a few weeks after symptoms first appear? Is it ethical to randomly assign some people in an outbreak area to a control group that doesn’t receive the vaccine? In this case, health workers used a procedure called ring vaccination, in which first appearance of the disease in a village led to vaccination of all direct contacts of the initially afflicted individual, then to all contacts of this group. For some randomly selected villages, this ring vaccination began immediately; for others, it was delayed by three weeks. Health workers tested a vaccine developed by Merck in this experiment, and found that vaccination started immediately was more effective than vaccination delayed by three weeks. This demonstrated convincingly that the Merck vaccine protected against Ebola, but to the detriment of some people in the control group, who might have benefitted from being vaccinated sooner. However, suppose the health workers had instead given everyone the Merck vaccine as soon as possible. The epidemic might have ended anyway due to a change in the weather or the fact that most susceptible people had already recovered or died, so the epidemic had simply run its course. Without an experiment, there’s no way to distinguish these possibilities from success of the vaccine, therefore no way to know if the vaccine would work if used in a future outbreak.

This background sets the stage for a new twist on the idea that experiments are the gold standard for research. In May 2019, Michelle Meyer and 6 coauthors reported results of 16 studies under the provocative title “Objecting to experiments that compare two unobjectionable policies or treatments.” The studies consisted of brief online surveys that 5,873 people volunteered to complete. For most of the studies, volunteers were randomly assigned to one of three treatments. In other words, the studies were themselves experiments – psychological experiments to assess how people respond to the process of experimentation itself. Many scientists consider experiments to be the gold standard for research; how about people in general?

As an example of the research by Meyer and her colleagues, subjects in treatment A of study 4 were informed that a doctor decided to prescribe a particular FDA-approved blood pressure medication (A) to all of her patients with hypertension. Subjects in treatment B were informed that the doctor decided to prescribe medication B, also approved by the FDA. Subjects in treatment C were informed that the doctor decided to randomly assign her patients to receive either medication A or medication B. The subjects were simply asked to rate the doctor’s decision on a 5-point scale, with 1 being very inappropriate and 5 being very appropriate. About 35% of the subjects in treatment C rated the doctor’s experimental approach either somewhat or very inappropriate, while fewer than 10% of the subjects in treatments A and B rated the doctor’s approach as inappropriate. In other words, the participants in this experiment were more willing to accept a doctor simply making a decision that her hypertensive patients should use drug A (for participants in the A group) or drug B (for those in the B group) than to accept that a doctor should randomly assign her patients to either drug A or drug B (for those in the C group), when both A and B had already been approved by the FDA to treat hypertension.

Meyer and her colleagues studied a wide range of scenarios in these experiments – direct-to-consumer genetic testing, design of autonomous vehicles, recruitment of health workers in developing countries, and more. In almost all cases, subjects were “objecting to experiments that compare two unobjectionable policies or treatments.” The researchers considered several possible reasons for this result, concluding that “Regardless of the reasons, the unfortunate lesson for those who care about evidence-based practice is that implementation of an untested policy based on intuition about what works may be less likely to invite objection than rigorous evaluation of two or more otherwise unobjectionable policies.”

Meyer and her colleagues don’t think this attitude makes sense. In the example I described in detail, the doctors who were prescribing medication A or medication B for hypertension were in effect doing an uncontrolled experiment. But this isn’t a very interesting or useful experiment compared to the randomized, controlled trial done by the doctors in group C, who would learn which drug worked better for their patients. Comparing different alternatives like this is the heart of the scientific method, which is why experiments are the gold standard for research.

How does this conclusion translate to the Facebook and OkCupid experiments that I described at the beginning? Those cases are different in one important respect: the users of Facebook and OkCupid were not informed that they were subjects of an experiment. But Meyer and Chabris argued that informed consent would have been impossible in these social media experiments; that in any case the users ultimately benefitted from the experiments; and, fundamentally, that the introduction of a new product like Facebook’s News Feed or OkCupid’s algorithm to predict successful matches for those seeking dates online are themselves experiments that users are subjected to. They’re just not very informative experiments. Does OkCupid’s algorithm really identify someone who will be compatible with you based on interests that the two of you share? Before this experiment, the company had found that pairs identified as good matches were more likely to have four-message exchanges than pairs not so identified. By altering the results of the algorithm reported to some users (changing a computed 30% probability of compatibility to 90%, or vice versa), OkCupid was trying to test whether their algorithm really measured things that contributed to compatibility, compared to the alternative hypothesis that users simply responded to the power of suggestion, e.g., pursuing email conversations with others reported to be 90% compatible even when the OkCupid algorithm implied that they were actually only 30% compatible. The researchers found that both the algorithm and the power of suggestion influenced subjects to pursue email conversations with people identified as good matches, although the simple power of suggestion had a somewhat larger effect.

Scientists in medicine, agriculture, ecology, psychology, sociology, education, economics, and many other fields rely on experiments. I taught about experiments in college classes ranging from introductory biology for nonmajors to graduate seminars. The new paper by Meyer’s group makes me wonder how receptive my students were to learning about experimentation during my 37 years of college teaching.

Posted in Correlation & Causation, Ethics, Experimentation, Experimentation, Medicine, Psychology, Teaching, Vaccination | Tagged , , , , , | Comments Off on Experiments as the gold standard for research: A new twist

Vaccination is in the news again (original posting of April 12, 2019, updated with new information on June 4, 2019)

I’ve written two blog postings about vaccination. One outlined two factors that may have contributed to recent outbreaks of whooping cough (pertussis) in the United States: increased rejection by parents of vaccination of their children against whooping cough, and replacement of a vaccine containing killed pertussis cells by an acellular preparation with fewer side effects. Members of the anti-vaccination movement argued that increased prevalence of whooping cough was due solely to the new vaccine being less effective than the original vaccine, and not at all to vaccine rejection by parents. I disputed this argument, on grounds that causation is complex and that both vaccine effectiveness and rejection likely contributed to whooping cough outbreaks

My second posting described the 1977 success of a worldwide vaccination program to eliminate smallpox and the current status of attempts to eliminate polio. In that posting, I reported that there were 359 cases of polio worldwide in 2014 compared to 350,000 twenty-six years earlier. In 2018, there were only about 30 identified cases of infection with wild polio virus, all in Pakistan and Afghanistan.

Vaccination is in the news again because of a rash of outbreaks of measles in the United States. The year 1963 was a turning point for measles in the U.S. Before 1963, almost all children got measles by age 15 and there were at least 400 to 500 deaths from measles annually (400 to 500 deaths were reported each year to the Center for Disease Control and Prevention [CDC]; the total would have been greater because not all deaths were reported). Scientists began developing a measles vaccine in the 1950s; this was first deployed in 1963 and replaced by an improved version in 1968. Cases of measles in the U.S. dropped rapidly, and the CDC declared in 2000 that measles was no longer continuously present in the U.S., although cases can be brought into the U.S. by travelers each year.

In most years since 2010, the CDC reported fewer than 200 cases, but there were 667 in 2014, 372 in 2018, and 981 already in the first five months of 2019, more than in any full year since 1992. Many of these were not isolated cases, but occurred in outbreaks in which the disease was apparently introduced by a traveler from abroad and then spread through contacts between the traveler and unvaccinated people and between those who got the disease from the traveler and additional unvaccinated people. For example, 550 cases occurred in Brooklyn and Queens, New York, between the beginning of the outbreak there in September 2018 and the end of May. On the other side of the country, 71 cases occurred in the Vancouver area in southern Washington between January and May 2019, although this outbreak is now over. Most cases of measles in these outbreaks occurred in unvaccinated children.

Public health departments in the various states of the U.S. are responsible for setting guidelines for vaccination of residents. Vaccines for measles, mumps, and rubella are combined in the MMR vaccine, which the CDC recommends for children in two doses, at ages of 12-15 months and 4-6 years. All states require that children attending public schools be vaccinated, and many also require vaccination for children in daycare centers and private schools. All states also allow exemptions for medical reasons, and some allow exemptions for religious or philosophical reasons.

Medical exemptions to vaccination requirements are non-controversial, since some children may not be able to tolerate vaccination because they have compromised immune systems or allergic reactions to some constituent of a vaccine. Religious and philosophical exemptions are more problematic, and some states are eliminating one or both of these options. There is a vigorous anti-vaccination movement in the U.S., and anti-vaxxer parents who don’t wish to home-school their children may claim religious or philosophical exemptions to vaccination requirements so they can send their kids to school with other children. Indeed, legislators in 19 states have introduced bills to expand options for exemption from vaccination, although no state has passed such legislation since 2003.

Suppose you have a 3-year-old daughter who can’t be vaccinated against measles because she is allergic to one of the components of the MMR vaccine. You wish to send her to a preschool where some children have received their first MMR dose while others have not been vaccinated because their parents claim a religious or philosophical objection. How dangerous is this for your daughter? The answer depends on how infectious the measles virus is and on what percentage of people in the area where you live are able to transmit the disease because they haven’t been vaccinated. In fact, measles is extremely infectious – before the introduction of vaccination in 1963, a person infected with measles would transmit the disease to about 12 new individuals. This means that measles will spread in a population unless more than about 92% of the population is vaccinated. Conversely, if more than 92% of the population is vaccinated, measles likely won’t spread because the disease will have run its course in the initial infected person (making him no longer infectious) before he has a chance to encounter a vulnerable, unvaccinated person.

The process outlined in the last paragraph is called herd immunity – unvaccinated individuals may be effectively immune if a large enough percentage of the population is vaccinated that the disease can’t spread. The percentage that must be vaccinated depends on the infectiousness of the disease; smallpox is much less infectious than measles and whooping cough, which is why public health workers were able to eliminate smallpox by a worldwide vaccination campaign, but we still have outbreaks of measles and whooping cough. In Tools for Critical Thinking in Biology (TCTB), I outlined a simple algebraic model of herd immunity. News reports of the current measles outbreaks in the U.S. often mention herd immunity as a protection for children and others who can’t be vaccinated, sometimes without explaining how it works. If you don’t have access to TCTB, Pam Belluck and Adeel Hassan provided an excellent overview of measles, including the process of herd immunity, in a recent column in the New York Times.

Returning to your hypothetical 3-year-old about to enter preschool, thinking about herd immunity will help you assess how vulnerable she is to contracting measles if you live in a place where a measles outbreak is underway, like Brooklyn or Vancouver Washington in spring 2019. Such an outbreak likely means that fewer than about 92% of the children in the epicenter of the outbreak have not been vaccinated and that measles is spreading among these unvaccinated children. This may include unvaccinated children in your chosen preschool, who won’t be able to transmit measles to their vaccinated playmates in the preschool, but will be able to transmit the disease to your daughter who is unable to be vaccinated.

In TCTB, I explained a moral question that arises from a basic mathematical model of herd immunity. As a parent, you have a moral obligation to protect your children. Anti-vaxxers argue that this justifies refusing vaccination for their children, but this argument ignores an overwhelming amount of scientific evidence for the benefits of vaccination. As members of society, we have an additional moral obligation to avoid harming others. This includes children who can’t be vaccinated for medical reasons, which these days is only a problem if such children live in a community where an outbreak occurs and many parents have refused to vaccinate their children on religious or philosophical grounds.

Anti-vaxxers resist having their children vaccinated for various reasons, some with a modicum of rational justification, others that ignore or deny abundant scientific evidence. One common reason for refusing vaccination against measles is the claim that this can cause autism in children. This claim is based on fraudulent research reported by Andrew Wakefield and his colleagues in 1998. Even ignoring the fraud involved in Wakefield’s research, the hypothesis of a causal association between childhood vaccination and autism has been repeatedly refuted in subsequent research, most recently in a study of 657,461 children born in Denmark between 1999 and 2010. Some anti-vaxxers argue that children don’t need to be vaccinated against measles because they can be treated with antibiotics if they get the disease, ignoring the fact that measles is a virus and that antibiotics don’t affect viruses. Other anti-vaxxers argue that natural immunity from getting a disease is somehow better than “artificial” immunity from vaccination. The best example to refute this claim comes from considering chicken pox. The virus that causes chicken pox, often in childhood, persists throughout life in nerve cells in the brain, and can cause a very painful skin disease called shingles in adults. Older adults are especially vulnerable to shingles, which can cause blindness or hearing loss or contribute to cardiovascular disease, in addition to producing a painful rash. In addition, recent evidence shows that measles can impair the immune system, making individuals more vulnerable to other infections by viruses and bacteria that cause pneumonia, ear infections, diarrheal diseases, and other conditions.

Frank Bruni published an opinion piece in the New York Times on March 9, 2019, in response to news of measles outbreaks in preceding weeks. This is how he summarized his main point: “The parents who are worried or sure about grave risks from vaccines reflect a broader horror that has flickered or flared in everything from the birther movement to “Pizzagate,” that nonsense about children as Democratic sex slaves in the imagined basement of a Washington pizza joint. Their recklessness and the attendant re-emergence of measles aren’t just a public health crisis. They’re a public sanity one, emblematic of too many people’s willful disregard of evidence, proud suspicion of expertise and estrangement from reason.” Use and abuse of social media exacerbates the problem; a recent analysis of Twitter postings between July 2014 and September 2017 presented evidence that “Twitter Bots and Russian Trolls Amplify the Vaccine Debate“. Just last week (April 2, 2019), the tweeter-in-chief added another example of estrangement from reason in his claim that noise made by windmills causes cancer.

Posted in Evaluating evidence, Medicine, Vaccination | Tagged , , , | Comments Off on Vaccination is in the news again (original posting of April 12, 2019, updated with new information on June 4, 2019)

Becoming an ecologist: A detour makes a difference

Note: I started this blog to update stories in my 2015 book, Tools for Critical Thinking in Biology, as new research was reported. This posting doesn’t serve that purpose, but instead is a reflection on the circuitous route that led me to a career in ecology.

Watching Ken Burns’ and Lynn Novick’s 2017 documentary on the Vietnam War brought back a flood of memories of that formative time in my life as a scientist. I was a senior biology major at Dartmouth College in 1967-1968. My advisor was a fungal geneticist who had gotten me a job in a colleague’s lab at Stanford the previous summer. In spring 1968, I was awarded an NSF fellowship for graduate school and accepted by 3 prestigious PhD programs in genetics around the country. As required in those days, I had registered for the draft at age 18, but college students were granted deferments from military service for the duration of their undergraduate and graduate programs. In 1968, however, as more and more soldiers were needed for the growing war in Vietnam, deferments for graduate school were suspended.

My classmates had a variety of responses to the war and the change in rules about educational deferments. One of my best friends enlisted in the Air Force. A few of my classmates emigrated to Canada. Many Dartmouth seniors, including me, decided to apply for K-12 teaching jobs to avoid being drafted. Of course we hadn’t been trained or certified as teachers to meet requirements of public school districts so we taught at private schools or public schools in rural areas or inner cities where recruitment of certified teachers was difficult. I took a job at a private school near Detroit, where I taught 5th through 12th grade math and biology.

After teaching for two years, I decided to apply for conscientious objector status. This was granted by my draft board, which meant that I would be assigned to alternative service such as working in a hospital if I was drafted. I also decided to reapply to graduate school, was successful, and began a PhD program at Harvard in fall 1970. I was called for a physical shortly after the semester started, but given a medical deferment by a sympathetic eye doctor whose daughter was a graduate student in biochemistry in California.

The result of my physical exam for the draft enabled me to continue my graduate program. Rather than genetics, which I would have studied if I had entered graduate school directly after completing my BA degree, I was now interested in studying animal behavior and ecology. Three things contributed to this change of direction. During my senior year at Dartmouth, I had taken a course in animal behavior taught by a new faculty member. With some distance from my senior genetics project, I thought about how much I had enjoyed the ideas and field work in this animal behavior course. I also remembered the expectation of my supervisors in the windowless basement lab at Stanford where I worked in summer 1967 that we should remain during lunch for conversation and mini-seminars. I much preferred to enjoy the California sunshine during that break. Finally, I was teaching teenagers in April 1970 when the first Earth Day took place. I was excited by their enthusiasm about environmental issues, but felt poorly prepared to teach them about natural history and ecological principles as the foundation for acting to protect the environment. This led to a graduate program in which I did dissertation research on foraging behavior and ecology of beavers.

After completing my PhD, I accepted a teaching position in biology at the University of Nevada, Reno where I taught from 1974 until my retirement in 2011. UNR was not a high-powered research institution during my early years on the faculty. Teaching loads were high and research expectations were modest. We had a handful of students in a new PhD program, with most of our graduate students supported by teaching assistantships. There were wonderful opportunities for ecologists, however, since Reno sits between the Sierra Nevada on the west and the Great Basin on the east, with short commutes to both desert and montane field sites. Reno has expanded so commutes are longer now, but the Great Basin remains the least well-studied North American desert.

Events during the height of the Vietnam War triggered a shift in my research interests that led to a very satisfying career of teaching and research. Although I had no major administrative responsibilities at UNR, I helped build our department and the university by serving on key search committees. I was a member of our Faculty Senate during the recession of 2008-2009, when loss of state funds caused contraction of several programs on campus. We rebounded from that period with growth in the size and especially the quality of biology and other programs. One of the greatest pleasures of my retirement has been getting to know our new, young faculty as they establish exciting, well-funded research programs while developing valuable outreach projects and showing genuine commitment to good teaching.

Posted in Teaching | Tagged , | Comments Off on Becoming an ecologist: A detour makes a difference

Why I worry about climate change

In recent discussions, several friends have described why current events make them deeply pessimistic about the future. I usually respond that I’m an optimist; for example, I believe that our democratic institutions are strong enough to withstand assault from partisans who make policy through abuse of social media. What about climate change? I have to admit that it’s no longer possible to be very optimistic about our lasting impacts on Earth’s climate from continuing use of fossil fuels.

I wrote about climate change in the last chapter of my recent book, Tools for Critical Thinking in Biology, because this is the most important issue facing humanity today. After describing multiple lines of evidence about the nature of climate change, I introduced three general principles necessary for understanding future effects of climate change. These principles are inertia, feedback, and tipping points. Carbon dioxide that we add to the atmosphere today will contribute to global warming for centuries because of inertia. Various processes that affect the climate are self-reinforcing (positive feedback) or self-damping (negative feedback, see page 247 of Tools for Critical Thinking in Biology for a potential example of negative feedback in climatology). And some changes triggered by continued burning of fossil fuels will cause major changes that are irreversible for millennia; for example, melting of the Greenland ice sheet will initiate a tipping point by raising sea level up to 23 feet, changing the very geography of Earth’s continents and causing major disruption of human society.

I taught ecology for many years at the University of Nevada, Reno, and did research on beavers and other mammals. I spent about a week discussing climate change at the end of the ecology course, emphasizing the fundamental roles of inertia, feedback, and tipping points in thinking about future changes in climate caused by human activity. New research on beavers (Figure 1) by Dr. Ken Tape and colleagues at the University of Alaska provides a beautiful illustration of how positive feedback affects climate conditions. Sadly, this new research adds another reason to be pessimistic about our ability to avert the long-term effects of climate change.

Massachusetts beaver.

North American beaver (Castor canadensis) in central Massachusetts. Photo by Steve Jenkins.

On December 28, 2017, President Trump posted one of his infamous tweets: “In the East, it could be the COLDEST New Year’s Eve on record. Perhaps we could use a little bit of that good old Global Warming …” Eastern North America did experience a prolonged bout of unusually cold weather this winter, but climate change is about long-term trends in global conditions, not day-to-day changes in local weather. In fact, much of the Earth saw higher than average temperatures in December 2017, and the last December that was colder than the 20th century average occurred in 1985. Alaska in particular was much hotter than normal in December 2017 (Figure 2).

Average temperatures across Alaska in December 2017.

In December 2017, Alaska was hotter than every previously recorded, almost 16 degrees Fahrenheit above the 20th century average.

Much of the Arctic, including Alaska, has a thin layer of soil that freezes in winter and melts in summer and a deeper layer of permanently frozen soil, called permafrost. Dead vegetation in temperate and tropical environments decomposes, releasing greenhouse gases. In the Arctic, much dead vegetation is incorporated in permafrost, trapping organic matter that would otherwise release carbon dioxide and methane to the atmosphere. As temperatures increase, permafrost closest to the surface melts, decomposition of long-dead plant material proceeds, and these greenhouse gases are added to the atmosphere. This is a positive feedback loop because the greenhouse gases contribute to further temperature increases, leading to more melting of permafrost, leading to further temperature increases in a runaway process. It’s particularly insidious because methane is a more potent greenhouse gas that carbon dioxide.

How do beavers enter the picture? Dr. Tape’s research has not yet been published, but was summarized by Kendra Pierre-Louis in the New York Times based on a presentation by Tape at a scientific meeting in December 2017. Beavers are widely distributed in North America, but haven’t been known to occur north of the vast boreal forest that extends across much of Canada and Alaska. These far northern areas have no trees and support only low-growing plants comprising tundra vegetation. With climate change, willows and other shrubs have spread into the tundra, followed by beavers. Beavers cut these woody plants, using their bark for food and their stems and branches to build dams that create ponds where they live (Figure 2). Their hydrological engineering produces new channels for water to flow across the permafrost, causing additional melting of the permafrost and exacerbating the positive feedback process that leads to additional release of greenhouse gases. Tape and his colleagues used aerial photography and satellite imagery to document the dramatic spread of beavers in the North Slope of Alaska since 1999.

Tundra beaver dam in the Yukon Territory, Canada.

Figure 2. A beaver dam in the tundra in Ivvavik National Park, Yukon Territory, Canada. Photo by Jay Frandsen/Parks Canada, 2015.

Human civilization developed during the last 11,000 years of relative climate stability. Recent temperatures on Earth have exceeded average temperatures during each of the 110 centuries in which civilization developed and flowered. We are committed to further changes in climate because of the greenhouse gases we’ve already added to the atmosphere, where they will remain for centuries (19% of the molecules of carbon dioxide emitted today by burning fossil fuels will remain in the atmosphere 1000 years from now).

Virtually all nations on Earth took a small step toward slowing the rate of climate change by agreeing to the Paris climate accord in December 2015. The United States under President Obama played a major role in facilitating this agreement. Unfortunately, President Trump gave notice of his intention to withdraw the United States from the Paris accord shortly after his inauguration as Obama’s successor in January 2017. This would make the U.S. the only nation on Earth not participating in the agreement – a true global pariah.
I used part of a tweet by President Trump to introduce Ken Tape’s new study of beavers in the Arctic. Here is the entirety of Trump’s tweet: “In the East, it could be the COLDEST New Year’s Eve on record. Perhaps we could use a little bit of that good old Global Warming that our Country, but not other countries, was going to pay TRILLIONS OF DOLLARS to protect against. Bundle up!” Trump is objecting to Obama’s commitment to the Green Climate Fund, a United Nations program to help poor countries deal with climate change. In fact, the U.S. pledged to contribute $3 billion to this fund, more than any other country in total, but far less than several other countries per capita; for example, $9.41 for the U.S. compared to $18.77 for Great Britain. Britain was fifth of 20 wealthy countries in per capita contribution to this fund and the U.S. would have been 11th, except that Trump stopped payment after $1 billion of our pledge had been delivered, leaving us at 19th out of 20.

I worry about climate change because the evidence that it’s happening is all around us – warmer temperatures, increased sea level, melting glaciers, harsher storms – and because inertia, positive feedback, and tipping points will exacerbate future changes that are already in the pipeline. It’s especially disheartening because of how quickly the U.S. has abdicated its leadership role in addressing climate change. We’ve been the most powerful nation on Earth for several decades; part of our importance on the global stage came from the example we set on issues such as environmental protection, conservation, and reliance on science for decision making. We now have national leaders who would call themselves climate skeptics but in fact are willfully ignorant of the science of climate change. By nature, I’m an optimist. Unfortunately, recent political history has turned me into a pessimist – at least about climate change.

Posted in Causation, Climate change, Observations | Tagged , , , | Comments Off on Why I worry about climate change

On climate science, the New York Times blows it

Bret Stephens is a right-wing journalist and climate-science denier who spent most of his career writing for the Wall Street Journal. He was hired as an opinion writer by the New York Times in early 2017, and wrote his first column on April 28.

I first learned about Mr. Stephens’s new job at the New York Times from a blog posting by Joe Romm, a physicist who writes extensively about climate science and energy economics. Romm pointed out an inconsistency between the current advertising campaign of the Times – “Truth. It’s more important now than ever” – and Stephens’s history of writing about climate change and other issues. The Times received more than 1500 comments and 550 letters to the editor about Stephens’s first column within 48 hours of its appearance on April 29. Romm, Susan Matthews, David Roberts, and others wrote thorough rebuttals of Stephens’s argument, such as it was. I won’t reiterate these rebuttals, but wish to make a few additional points about denial of climate science in Stephens’s New York times column.

Mr. Stephens claims to accept the fact that the average temperature on Earth has increased in the last 150 years due in part to human activity. This gives his column a patina of plausibility compared to the claims of others who deny climate science. But this apparent plausibility is really only a pretense. He ignores all of the other evidence of recent, rapid climate change – melting glaciers, shrinking ice caps on Greenland, decreased ice in the Arctic Ocean, melting permafrost, rising sea levels, more extreme weather events. He fails to acknowledge that our addition of greenhouse gases to the atmosphere is the primary cause of recent climate change.

Stephens’s biggest problem is with the mathematical models that are used to project future climate conditions under various scenarios. He incorrectly implies that climatologists claim certainty about these projections, and constructs an analogy between this purported certainty of climatologists and the supposed certainty of the Hillary Clinton presidential campaign that she would win the election last November. According to Stephens, the Clinton campaign was wrong, so the climatologists might also be wrong. This analogy is phony at several levels. The politicians weren’t as certain as Stephens implies. The climatologists include uncertainty in all of their model projections. And the climate models are based on detailed knowledge of atmospheric chemistry and physics, unlike the statistical models of polling data that politicians and the news media use, which depend on extrapolating from surveys of a small number of voters to the voting population as a whole. In addition, there are various ways of testing global climate models by comparing observed data to predictions of the models. The figure below shows one example illustrating that these models can account for measured average temperature of Earth’s surface between 1860 and 2000 only if models include both emission of greenhouse gases by burning fossil fuels and natural factors such as variation in solar radiation and volcanic activity.

Average annual temperature of Earth’s surface from 1860 to 2000 (red lines) compared to projections of three sets of global climate models: models with only natural forcings such as changes in solar radiation and sunspot activity (a), models with only anthropogenic forcings such as release of greenhouse gases into the atmosphere (b). and models with both natural and anthropogenic forcings (c). The gray bands show the range of projections for four models in case. The temperature anomaly shown on the vertical axes of these graphs is the deviation from the average temperature between 1880 and 1920 in degrees Celsius (multiply these values by 1.8 for degrees Fahrenheit). Projections shown in these graphs are hindcasts, tests of how well models fit existing data, rather than forecasts, or predictions of future data.

For Stephens, the uncertainty surrounding predictions of climate models appears to justify a wait-and-see approach to dealing with climate change. This ignores three fundamental principles of climate change that I discuss in my book “Tools for Critical Thinking in Biology”: inertia, positive feedback, and tipping points. About 20% of the carbon dioxide that we add to the atmosphere by burning fossil fuels today will remain in the atmosphere for 1000 years (inertia). Increased temperatures in the arctic cause less ice in the Arctic Ocean, making the surface darker, meaning more heat is absorbed, warming the arctic region further, melting more ice, and so forth (positive feedback). Finally, some effects of anthropogenic climate change may cause transitions to dramatically different conditions (tipping points) that will be irreversible for hundreds of human generations. For example, sea level was 23 feet higher during the Eemian interglacial period that ended, 125,000 years ago, but temperature was only a fraction of 1oF higher than today. As global temperatures continue to increase in response to continued addition of greenhouse gases to the atmosphere, with arctic temperatures rising much faster than those at lower latitudes, we may reach a tipping point in which the Greenland ice cap melts, causing sea levels similar to those in the Eemian that will remain elevated for many generations. Inertia, positive feedback, and tipping points in the global climate mean that Stephens’s wait-and-see approach is indefensible.

Stephens seems to be fixated on trying to undermine the overwhelming consensus among climate scientists in the basic facts of climate change and the resulting projections of what we face if we continue business as usual in our use of fossil fuels. Contrary to his claim, this consensus doesn’t mean that the science is settled. Far from it: I’m not a climatologist, but climatology is one of the most active and exciting areas of research today, with new findings reported almost weekly (most of which should make us even more worried about the future).

The heroes of Stephens’s story are ordinary citizens, who may be “indifferent to … the prospect of planetary calamity”, but “have a right to be skeptical of an overweening scientism.” Stephens implies that climate scientists are often villains, but I see climate scientists who forge ahead with their research in the face of abuse from politicians beholden to fossil fuel interests as heroic. Ordinary citizens can also be heroic if they nurture their curiosity about science, develop real skepticism instead of the faux skepticism of science deniers, and above all demand honesty in themselves and others.

Posted in Climate change, Evaluating evidence, Journalism, Modeling, Science and politics | Comments Off on On climate science, the New York Times blows it

“This is a sad day for the children of the U.S.”

Howard Mielke is a toxicologist at the Medical School of Tulane University in New Orleans. He described March 21, 2017 as “a sad day for the children of the U.S.” in response to news that the Environmental Protection Agency (EPA) plans to eliminate two programs that protect children from exposure to lead in paint. One program supports public education on risks of lead-based paint and trains contractors to do remodeling projects without releasing lead into the atmosphere from old paint; the second provides grants to states and Indian tribes to deal with these risks in their jurisdictions. These programs currently cost $16 million and have 73 full-time employees.

Tools for Critical Thinking in Biology is about various kinds of evidence that scientists use to answer questions and test hypotheses. Evidence can take many forms, from systematic observations to results of experiments to comparisons and correlations. Some questions that scientists ask have implications for public policy, including questions about the effects of lead exposure on health. In Chapter 5, I use two main questions to illustrate the possibilities and pitfalls of inferring causation from comparisons and correlations: Do cell phones cause brain cancer? Does lead exposure in childhood make individuals more likely to commit crimes later in life? In the first case, a correlation between cell phone use and brain cancer is not convincing evidence that cell phone use causes brain cancer (see my book for an explanation). In the second case, scientists have found a host of correlations between lead exposure and rates of violent crime about 20 years later: in different countries of the world, different states and cities of the U.S., and different neighborhoods of New Orleans. Together with detailed studies of the impact of lead exposure on brain development, these correlations imply that early exposure to lead has adverse, longterm consequences for the health of children. In particular, children exposed to lead in childhood have lower IQs, a greater risk of attention deficit / hyperactivity disorder and other behavioral problems, and a greater likelihood of being arrested for violent crimes later in life.

Children in the U.S. have been exposed to lead from two main sources: paint and gasoline. Use of lead in gasoline was phased out in the 1970s. Lead-based paint was outlawed for residential use in 1978, although its use had declined substantially after a peak in the 1920s. Nevertheless, many older homes in poor neighborhoods of cities have flaking and chipping paint containing lead. This is why the EPA has supported the removal of paint from older homes in a careful process that doesn’t produce lead-filled dust that can be inhaled by children, as well as adults and pets.

President Trump released a budget on March 15, 2017 that proposed cuts across the government with the exception of the departments of Defense, Homeland Security, and Veterans Affairs. The EPA was targeted for the largest percentage cut of 31.4%, for a total decrease of $2.6 billion. In this context, elimination of two programs to protect children from exposure to lead-based paint seems like small potatoes: $16 million is less than 1% of the total cut proposed for the EPA budget. Trump’s plans will harm the environment in many ways, but this small part of his plans seems particularly petty and mean-spirited. The evidence is strong that lead exposure is harmful to kids and society; this has been recognized for a long time and led to elimination of lead from paint and gasoline 40 years ago; the cost to continue remediation efforts is minuscule. Perhaps Congress will recognize these facts and reject this and other parts of Trump’s agenda that adversely affect the environment and human health.

Posted in Causation, Ethics, Medicine, Science and politics | Comments Off on “This is a sad day for the children of the U.S.”

Monarch butterflies, milkweed, and migration: The law of unintended consequences

I described the magic of migration by monarch butterflies in Chapter 1 of Tools for Critical Thinking in Biology. These tiny animals fly from breeding areas in the northern US and Canada to overwintering sites in high-elevation forests in central Mexico each fall. After a period of semi-dormancy until spring, the adult monarchs mate, return to the southern US to lay eggs on milkweed plants, and then die. Larvae hatch from these eggs to feed on milkweed foliage, then pupate on milkweed (or nearby lawn furniture) until adult butterflies emerge. Some of these adults fly further north to reproduce, and these successive journeys north continue for 3 or 4 generations until monarchs have reached southern Canada again. In fall, the process repeats itself, with Canadian monarchs flying 2,500 miles to Mexican overwintering sites that they have never seen.

Tagged monarch butterfly

Monarch butterfly wearing an ID (photo by Jim Gagnon)

This story of migration by monarch butterflies is magical because it took 40 years and the help of thousands of volunteers for two Canadian scientists, Fred and Norah Urquhart, to discover the wintering area in the mountains of central Mexico. The Urquharts developed a method of tagging butterflies; volunteers throughout North America reported their captures of tagged butterflies to the Urquharts and did tagging of their own to gradually extend the known range of movement of the animals. The story is also magical because scientists have used observations and experiments to learn what triggers migration in the fall, how monarchs from as far north as Canada find their way to Mexico, and why migration may benefit monarchs. I describe some of this research in my book, but the research continues and important new discoveries have extended the story since the book was published in April 2015. One of these stories links migration and milkweeds, with potential unintended consequences for conservation of monarchs, so I need to give a little background about milkweeds and conservation before telling this story.

Why is milkweed important for monarch butterflies?

Adult monarchs eat nectar from a large variety of flowers, helping to pollinate the flowers in the process. Females lay eggs only on milkweed, however, and larvae eat only milkweed as they grow and develop. Milkweeds contain compounds called cardenolides that are toxic to many vertebrates; these compounds ingested by monarch larvae remain in the tissues of adults that develop from the larvae. Cardenolides make adult monarchs poisonous to predators such as blue jays, which can learn to avoid monarchs from trying to eat a single individual. This is another fascinating story that you can learn about here.

What is the conservation status of monarch butterflies?

Monarch butterflies face many risks, both during their breeding season in the United Sates and southern Canada and in their overwintering sites in central Mexico (and along the Pacific coast of California where monarchs breeding west of the Rocky Mountains spend the winter). Breeding habitat is lost to growth of cities, shopping malls, roads, and agriculture. The milkweed that monarchs depend on grows in and adjacent to agricultural fields; farmers now plant seeds of corn and other crops that are genetically engineered to be resistant to a herbicide called Roundup. This enables farmers to use Roundup to control weeds without killing their crops. These weeds include milkweed, so this process is detrimental to monarchs.

Some overwintering sites along the Pacific coast of California are in towns and cities, making them vulnerable to further development and loss of habitat, although local governments have set aside parks to protect monarchs that winter there. Most monarchs overwinter in the highlands of central Mexico, which have a history of disturbance by logging and cattle grazing. Monarchs in Mexico occupy a handful of sites with a total area of 1 to 18 hectares (2.5 to 45 acres; an acre is about the size of an American football field without the end zones). The Mexican overwintering population fluctuates from year-to-year depending on how successful breeding was the previous summer, how much mortality occurred during migration, and weather conditions in the overwintering sites, but the general trend has been downward since 1995. The lowest estimate of the Mexican overwintering population was 25 million in 2014, although the population rebounded to 150 million in 2016. These may seem like very large numbers, but the Mexican overwintering population has decreased by more than 25 million monarchs eight times since 1995. If this had happened between 2014 and 2015, monarch butterflies east of the Rocky Mountains would be extinct.

Monarch Butterfly Overwintering Populations.

Estimates of number of monarch butterflies overwintering in Mexico, 1995-2016. From Center for Biological Diversity.

What can citizens do to help protect this charismatic insect? One strategy is to plant milkweed in our yards to compensate for the loss of milkweed in agricultural areas. There are about 100 species of milkweeds native to North America, most of which are used by monarch butterflies. However, the most common milkweed available to gardeners in plant nurseries is an exotic species called tropical milkweed. Unlike North American milkweeds, tropical milkweed remains green and continues to flower in fall and winter in places with mild climates, like the Gulf coast states. Because of this newly available food supply, some monarchs migrating from farther north remain through the winter in areas along the Gulf coast rather than continuing to Mexico. These monarchs breed through the winter and have established non-migratory populations that remain along the Gulf coast all year.

Common milkweed, native to North America, and tropical milkweed, often sold in plant nurseries and planted by gardeners in North America, to the possible detriment of monarch butterflies.

Does this sound like a problem? Probably not, until I give you one more piece of information. As I described above, monarchs have an effective chemical defense against many vertebrate predators. They are also attacked by several parasites, the most common of which is a protozoan called Ophryocystis elektroscirrha (OE). There are populations of monarchs that are year-round residents in Hawaii and Florida, and OE is much more prevalent in these populations than in migrating populations, perhaps because migration culls infected butterflies, who can’t fly as well as healthy ones, or because migration enables butterflies to escape areas where they can become infected. In a recent study, Dara Satterfield and colleagues found that the new, non-migratory populations along the Gulf coast have higher infection rates than migratory monarchs. Satterfield’s group suggests that these sedentary monarchs along the Gulf coast could act as a reservoir for OE, causing infection of migrants as they pass through this area on their way to or from the wintering sites in Mexico. In trying to help monarchs by planting exotic tropical milkweed in our gardens, we may instead be putting them at greater risk. If you’re interested in contributing to the conservation of monarch butterflies, why not share this story with your local nurseryman, and ask him to sell native milkweed instead of the exotic tropical species that contributes to the unintended consequence described here?

Monarch migration

Migration of monarch butterflies in eastern North America. Green diamonds show the main locations of overwintering monarchs in the highlands of central Mexico. Red squares show alternative sites where small numbers of migrating monarchs spend the winter. Blue dots show locations of new sedentary (non-migratory) populations along the Gulf Coast. From “Loss of migratory behaviour increases infection risk for a butterfly host”, by D. A. Satterfield et al., 2015.

Posted in Causation, Citizen Science, Conservation, Ecology, Migration, Observations | Comments Off on Monarch butterflies, milkweed, and migration: The law of unintended consequences

Medical use of marijuana revisited

I described two experimental studies of the use of marijuana to reduce pain in Tools for Critical Thinking in Biology and discussed two technical reviews of the medical value of marijuana in a subsequent blog posting. The gist of these reviews was that there was limited but adequate evidence that marijuana could help patients suffering from chronic pain, but little evidence for other beneficial health effects. Additional research reinforces the evidence supporting use of marijuana for pain control, especially as an alternative to opioid drugs, which are widely abused in the US. This research has an important flaw, however, that was ignored in recent news stories.

In 2014, Michael Bachhuber and colleagues at the Philadelphia Veterans Affairs Medical Center compared death rates caused by overdoses of opioid painkillers such as oxycontin in US states with and without laws allowing medical use of marijuana. They used data through 2010 when 13 states had passed such legislation. Bachhuber’s group found that annual death rate due to opioid overdoses was 25% less in these states than in the remaining 37 states that didn’t allow medical use of marijuana.

In a follow-up study published in July 2016, Ashley Bradford and her father David, graduate student and professor respectively at the University of Georgia, compared Medicare data on prescription drug use in states with and without medical marijuana legislation. On average, individual doctors wrote almost 2000 fewer prescriptions for opioids and other typical painkillers in states in which medical marijuana was legal.

Bachhuber’s group and the Bradfords infer from their comparisons of states with and without legalization of medical marijuana that individuals might benefit by having marijuana available to control pain. For example, in a news story in Science magazine, Greg Miller quotes David Bradford describing his results as “suggestive evidence that medical marijuana might help divert people away from the path where they would start using [an opioid drug], and of course if they don’t start, they’re not on that path to misuse and potentially death”.

These analyses illustrate a classic breakdown of logic called the ecological fallacy: reaching conclusions about individuals from aggregated data for groups to which the individuals belong. The groups are states where individuals live; in Bachhuber’s study the data are death rates in each state due to opioid overdoses, and in the Bradfords’ study the data are numbers of prescriptions for opioid painkillers written by doctors in each state. Average death rates and opioid prescription rates are lower in US states with legal use of medical marijuana than in the remaining states. Paradoxically, individuals who use medical marijuana might also use more opioids or have a greater likelihood of dying than individuals who don’t use medical marijuana, despite the opposite effect in comparing average rates of opioid prescription and death between states with and without laws for medical use of marijuana.

Bachhuber’s group was fairly cautious in interpreting their results, although news reports such as “Could pot help solve the U.S. opioid epidemic?” in the November 4, 2016, issue of Science ignored the caveats discussed by Bachhuber and colleagues. For example, Bachhuber’s group mentioned the alternative hypotheses that “increased access to medical cannabis may reduce opioid analgesic use by patients with chronic pain” or act as a “’gateway’ or ‘stepping stone’ leading to further . . . opioid analgesic overdoses”. However, the researchers weren’t able to discriminate between these hypotheses because they only used statewide averages, not data for individual patients.

How could it be possible for medical marijuana to be a stepping stone for individuals to overdose on opioids and at the same time for states without legal access to medical marijuana to have more deaths from opioid overdoses? One reason is that other differences between states with and without laws allowing medical use of marijuana might account for greater death rates from opioid overdoses in the latter states. These other potential differences are called confounding variables. Bachhuber’s team considered unemployment rate as one possible confounding variable, on grounds that higher unemployment might cause more drug abuse, but they found no differences in unemployment rates between states with and without medical marijuana. Unemployment rate is only one of many possible confounding variables, however; states without legal access to medical marijuana might have older residents, with more pain, or younger residents, with more propensity to abuse drugs, or lower educational levels, or more military veterans suffering from PTSD, than states with legal access to medical marijuana. Indeed, one of the most dramatic differences between states that do and don’t allow medical use of marijuana is that all of Hillary Clinton’s 232 electoral votes in the 2016 presidential election came from states with laws allowing medical marijuana while only 23 of Donald Trump’s electoral votes came from such states. This certainly doesn’t mean that having medical marijuana available caused people to vote for Clinton and not having it available caused them to vote for Trump, but it illustrates the pitfalls of drawing conclusions about causation from purely correlational data. I illustrate this point in more detail in Chapter 5 of Tools for Critical Thinking in Biology.

Map of US states with and without legalization of medical marijuana.

US states with operational legalization of medical marijuana as of December 28, 2016. This excludes Arkansas, Florida, and North Dakota that legalized medical marijuana in November 2016 but haven’t yet established a process for patients to obtain marijuana. It also excludes Texas that legalized medical marijuana in 2015 with a doctor’s prescription, ignoring the fact that federal law forbids doctors from prescribing marijuana. Instead, doctors may issue a letter to the patient recommending marijuana for certain medical conditions, as occurs in the states where legalization is operational.

To test the alternative hypotheses proposed by Bachhuber’s group that medical marijuana may be a safer alternative to opioid drugs for treating pain or that medical marijuana may be a stepping stone to opioid abuse, we really need data on individual people who do and don’t have access to medical marijuana. Even better would be experimental data, although it’s hard to imagine an ethical study that could test these hypotheses. However, experiments are possible to test the efficacy of marijuana compared to opioids for pain control. For example, Colorado recently legalized marijuana for recreational use as well as medical use, and the state is using $9 million of the taxes collected from sales of marijuana to support experimental studies, much as California did in the early days of marijuana research (see Chapter 4 of Tools for Critical Thinking in Biology). Emily Lindley is in charge of one of the Colorado studies; her group will use 50 volunteer patients with back or neck pain who will be assigned randomly to receive either vaporized marijuana, an opioid drug, or a placebo on three visits to the hospital, after which they will report any pain relief they experience.

So, does legalization of marijuana for pain control reduce the likelihood of patients overdosing on opioids? We don’t know the answer at this time, so we shouldn’t use this possibility as an argument in support of legalization of marijuana. Uncertainties like this can be frustrating to those interested in promoting or opposing particular policies, but such uncertainties are embraced by scientists as challenges to be overcome. Be on the lookout for further research that may help answer this question!

Posted in Causation, Correlation & Causation, Evaluating evidence, Experimentation, Medicine | Comments Off on Medical use of marijuana revisited

Storytelling in science: What can we learn about critical thinking from stories about mountain lions and sea otters?

Tools for Critical Thinking in Biology is a book of stories about doing science. Most of these stories are incomplete because science is a work in progress, therefore one of my great pleasures is learning about new research that extends the stories in the book. Two new books tell exciting stories that complement stories in my book, and I encourage you to read these new books to learn more about how science works. The new books are Heart of a Lion: A Lone Cat’s Walk across America, by William Stolzenburg, and Serendipity: An Ecologist’s Quest to Understand Nature, by James A. Estes. Heart of a Lion describes a remarkable journey by a young male mountain lion from South Dakota to Connecticut and illustrates what we can learn from purely observational evidence, and Serendipity is Jim Estes memoir of his lifetime studying sea otters and killer whales in the Aleutian Islands.

Scientists use many different tools to learn how the world works. These tools include ways of studying nature such as comparative studies and experiments. Some studies use fancy equipment such as DNA sequencers; others use complex statistical methods. Regardless of the approach used and the technology employed, all science depends on reliable observations as the basic source of information. I illustrate this idea in my book with two brief stories about the evaluation of observational evidence in studies of ivory-billed woodpeckers and wolverines. In 2004, a birder in Arkansas reported the first sighting of an ivory-billed woodpecker since 1944, but extensive follow-up work didn’t confirm this observation, so ivory-billed woodpeckers are still considered extinct. In 2008, a graduate student set up remotely activated cameras to study pine martens in northern California. One of her cameras recorded a wolverine, the first definitive evidence of this rare and elusive carnivore in California since 1922. The graduate student and her team collected hair and fecal samples which yielded DNA confirming that the subject of their photo was a wolverine, probably a migrant from 650 kilometers away in Idaho.

Mountain lion

Mountain lion in Yellowstone National Park (photo by K. Fink).

Mountain lions once occupied most of the western hemisphere, but were extirpated east of the Rocky Mountains following European settlement. Except for a small, isolated population in Florida, the easternmost population in North America is now found in the Black Hills of South Dakota. Mountain lions are solitary and territorial, and young disperse from their natal areas at 12 to 18 months of age. By dispersing, young males avoid being killed by adult males that defend territories encompassing the ranges of several females.

Biologists documented dispersal of mountain lions from the Black Hills in the 1990s by attaching radio transmitters to young animals. Some dispersers went west, looking for openings in the matrix of mountain lion territories in the Rockies where they could establish territories of their own. Other dispersers went east, where opportunities to establish territories were virtually limitless but opportunities to mate were scarce or nonexistent, though the dispersers didn’t know this. Will Stolzenburg tells the story of one such disperser in Heart of a Lion. The hero of this story is a juvenile male that left its birthplace in the Black Hills in summer 2009. After a journey of more than 2000 miles through Minnesota, Wisconsin, Michigan, probably Canada, and New York, the lion was killed by a car in southwestern Connecticut in June 2011. Unlike other dispersers that traveled shorter distances, this lion did not carry a radio transmitter that enabled researchers to follow him day-by-day. His story as told by Stolzenburg is about the use of bits and pieces of observational evidence to reconstruct his route. This evidence includes sightings of tracks, photographs from remotely activated cameras, video recordings, hair samples, and scat. Researchers extracted DNA from the hair and scat that the lion left along his route, confirming that a single male had left these samples in Minnesota, Wisconsin, and New York and that this same male was the one killed by a car in Connecticut. This is similar to the wolverine story I told in my book, but much more detailed since the mountain lion’s journey was traced from beginning to end.

The scientific consensus is that mountain lions no longer live in eastern North America, but many locals believe that mountain lions were never completely extirpated from this region. As explained by Stolzenburg, this belief is based on numerous reported sightings over many years in many locations. Some of these observations were supported by photographs that turned out to be domestic cats, golden retrievers, and various other animals. In other cases, people really did see mountain lions, but ones that had escaped from captivity. In 2011, the US Fish and Wildlife Service reviewed the evidence for persistence of mountain lions in eastern North America and concluded that none of it was credible. Besides giving us an important history of our evolving relationship with wildlife, Stolzenburg shows how observational evidence can be evaluated for reliability and can contribute to our understanding of nature.

Sea otter

Sea otter at Morro Bay (photo by Mike Baird).

Serendipity: An Ecologist’s Quest to Understand Nature by Jim Estes is the second book I encourage you to read. I discuss the complexity of causation in Chapter 8 of my book using the ecology of sea otters in the Aleutian Islands as a case study. Jim Estes started research on sea otters in in the Aleutians in 1970 when he was a graduate student and has studied them ever since. Estes discovered that sea otters are a keystone species that has dramatic impacts on nearshore environments in the Pacific Ocean. Sea otters eat sea urchins, which graze on large marine algae called kelp. Where sea otters are absent, the ocean floor near shore is carpeted by sea urchins; where sea otters are present, sea urchin populations are much smaller and forests of kelp develop. These kelp forests provide habitat for fish, which provide food for eagles and other predators.

After working in the Aleutians for 20 years, Estes and his coworkers observed a sharp drop in the population of sea otters during the 1990s, which they eventually attributed to predation on otters by killer whales. With a declining sea otter population, sea urchins began to increase and kelp was reduced, changing many nearshore environments from kelp forests to sea urchin barrens. This is an example of a trophic cascade, in which a top predator influences the abundance of species below it in the food chain, including the vegetation at the base of the food chain.

Serendipity: An Ecologist’s Quest to Understand Nature is a wonderful memoir of Estes’ long career in science, including not only the details of what he did and learned but also an explanation of the thinking that led him from one project to another during his career and discussion of the broader implications of his work. Many of us can identify chance events that changed the direction of our lives; Estes entitled his book Serendipity because it shows how he was able to capitalize on these events in his own life to make fundamental contributions to our understanding of ecology. Read his book if you want to learn more about sea otters and killer whales and, more generally, about how science works.

Ecologists have identified several examples of trophic cascades driven by top predators in recent years. I discuss wolves in Yellowstone National Park as well as killer whales in the Aleutians in Chapter 8 of my book, while Will Stolzenburg devoted an entire book called Where the Wild Things Were: Life, Death, and Ecological Wreckage in a Land of Vanishing Predators to this topic. Stolzenburg emphasizes implications for conservation of new knowledge about trophic cascades.

While writing this posting, I learned about a new study that nicely illustrates an implication of a trophic cascade that may be surprising to you. Several ecologists used a mathematical model to predict the effects of recolonization of eastern North America by mountain lions. According to the model, control of eastern deer populations by mountain lions would reduce collisions of vehicles with deer, resulting in 155 fewer human deaths, 21,400 fewer human injuries, and savings of $2.1 billion over a period of 30 years. Between 1890 and 2008, mountain lions attacked humans 153 times and killed humans 21 times in the entire US and Canada, so occasional attacks and deaths would occur in the east if mountain lions were reintroduced, but these would almost certainly be far less than the number of injuries and deaths avoided due to the reduction in deer-vehicle collisions.

Posted in Ecology, Evaluating evidence, General material, Long-term study, Observations | Tagged , , , , , | 1 Comment

Careful readers are a writer’s best friends

Writing for publication entails many drafts and many reviews of these drafts by friends, colleagues, copyeditors, and critics. The credibility of science depends on peer review, as explained by Frederick Grinnell in The Everyday Practice of Science and by me in Chapter 9 of Tools for Critical Thinking in Biology, where I give personal examples, historical examples, and contemporary examples that have been prominent in the news media.

Two readers of the published version of my book made several suggestions for corrections. I wish to thank them here and explain what they found in their careful reading. Don’t get bogged down in the first explanation dealing with a problem that appears earliest in the book, but return to it after reading some of the later explanations that are more straightforward.

1) In Chapter 5, I use two main examples to show how researchers use, and sometimes abuse, comparative and correlational data to answer questions about causation. For instance, people have wondered whether microwave radiation emitted by cell phones can cause brain cancer. One source of evidence to address this question comes from comparison of past cell phone use by people with and without brain cancer; this is called a retrospective case-control study. Cases are people with brain cancer, controls are people without brain cancer, and the study is retrospective because the data come from estimates by cases and controls of how much they used cell phones in the past. Case-control studies often use odds ratios to report results: what are the odds of brain cancer for a cell phone user compared to those for a nonuser? More specifically, what is the ratio of these odds? If this ratio is greater than 1, a cell phone user is more likely to get brain cancer than a nonuser, although we also have to consider the possibility that a ratio greater than 1 could simply be due to chance.

Odds ratios are a little tricky to understand, so I use a completely different example to show how they are calculated. When the Titanic sank in 1912, 650 adult males died and 132 survived, while only 102 adult females died and 300 survived. Therefore the odds of dying for males were 650/132 = 4.92 while the odds of dying for females were 102/300 = 0.34. These translate to an odds ratio of 4.92/0.34 = 14.5. I then wrote “adult male passengers were almost 15 times as likely to die when the Titanic sank as adult female passengers.” As one post-publication reviewer noted, this is incorrect; the odds ratio really means that the odds of dying were 15 times greater for males than females. We can also express the difference in mortality of male and female passengers on the Titanic as a relative risk. The likelihood of death for males was 650/(650+132) = 0.83; the likelihood of death for females was 102/(102+300) = 0.25. Therefore males were 3.32 times as likely to die as females (relative risk = 0.83/0.25). This is a pretty dramatic difference, but not as dramatic as I implied in the book.

Using odds to express probabilities of events has a long history that is intertwined with the history of gambling. According to FanGraphs on March 8, 2016, the Chicago Cubs have an 18.6% chance of winning the 2016 World Series. This means that the odds against the Cubs winning are 81.4%/18.6% = 4.38 to 1. If you bet $1 on the Cubs today, your friend should be willing to pay you $4.38 if they win in exchange for collecting $1 from you if they lose. Although odds and odds ratios are somewhat indirect measures of relative probabilities, they have some advantages for statistical analysis of medical questions like the relationship of cell phone use to brain cancer.

2) I describe how interactions between genes and environments influence human traits in Chapter 7, using studies of twins to develop this theme. I introduce this topic with a story about two genetically identical twins that were adopted at an early age and raised by separate families. These twins were reunited at age 39 and discovered that they shared a remarkable set of traits, starting with the fact that both of their adoptive families named them Jim. In addition, as I wrote, “both Jims smoked the same brands of cigarettes and beer – amazing!” As one post-publication reviewer noted, smoking beer is pretty amazing.

3) In Chapter 8, I used Hurricane Katrina to introduce the idea that events often result from complex webs of causation. About 1800 deaths and $81 billion in property damage in New Orleans were attributed to Katrina, which hit the Gulf Coast on August 29, 2005, but several other factors also contributed to this devastation. My diagram of this web of causation implies that George W. Bush’s praise of his director of the Federal Emergency Management Agency (“Brownie, you’re doing a heckuva job”) was one of these contributing factors. Instead, I should have described this praise as a symptom of an underlying cause, appointments of unqualified people to lead agencies such as FEMA.

4) Jim Estes of the University of California, Santa Cruz and the US Geological Survey has studied sea otters and other marine mammals in the Aleutian Islands for more than 40 years. Chapter 8 describes some of their research, including their discovery that predation by killer whales caused the population of sea otters to crash in the 1990s. This is a fascinating, multidimensional story involving questions such as these: The researchers first saw an attack by a killer whale on a sea otter in 1991 – why did killer whales switch from larger mammalian prey to sea otters at this time? The researchers saw six total attacks in the early 1990s, yet estimated that the sea otter population declined from 53,000 in 1991 to 12,000 in 1997 – could predation by killer whales account for such a large decline without more attacks being seen by humans? Sea otters are much smaller than other mammalian prey of killer whales – how many sea otters would supply the daily energy needs of a killer whale? I address this last question in Box 5.2. Killer whales require about 247,500 kilocalories of energy per day, and five sea otters would supply this amount of energy. One kilocalorie equals 1 Calorie in human nutrition (not 1000 Calories, as I state in the book). We use 2,000 to 3,000 Calories per day, only about 1% of the energy a killer whale uses. Of course, we’re much closer in size to a sea otter than a killer whale.

5. Chapter 8 features several predators – sea otters that eat sea urchins, killer whales that eat sea otters, and wolves that eat elk. It describes the big effects that these predators can have on their prey and on the habitats where both live. I emphasize different kinds of evidence that researchers have gathered to understand these relationships between predators and prey in ecological communities, but I also ask readers to consider how humans influence these communities. These aren’t purely scientific issues, but entail decisions about management – commercial fishing in the Aleutians, hunting in the Rockies, removal of wolves that kill livestock near Yellowstone National Park. These decisions involve not only science but ethics. In closing Chapter 8, I quote A Sand County Almanac by the conservationist Aldo Leopold: “I was young then, and full of trigger itch; I thought that because fewer wolves meant more deer, that no wolves would mean hunter’s paradise. But after seeing the green fire die, I sensed that neither the wolf nor the mountain agreed with such a view.” As one of my post-publication reviewers reminded me, the “green fire” refers to the wolf’s eyes.

6. I use a rubric called FiLCHeRS to summarize six key tools for critical thinking: evaluation of claims by assessing their Falsifiability, Logic, Comprehensiveness, Honesty, Replicability, and Sufficiency. James Lett proposed this rubric in 1990 and illustrated its use in evaluating paranormal beliefs such as astrology and extrasensory perception. I apply FiLCHeRS to claims by those who dispute the scientific evidence for human impacts on climate change. In discussing sufficiency, I quote Lett: “extraordinary claims demand extraordinary evidence.” This idea actually has a long history, starting with the philosopher David Hume in 1748, continuing with a restatement by Laplace in 1812, another restatement by Carl Sagan in 1980, and finally Lett’s version in 1990 that changed Sagan’s verb from “require” to “demand”.

7. Finally, I use the classic case of industrial melanism to explain how evolution works in Appendix 2. Peppered moths fly at night and rest on tree trunks during the day. All peppered moths collected in England before 1948 were light-colored, providing effective camouflage against predation by birds when resting on patches of whitish lichens growing on tree trunks. People began to collect dark-colored moths in the mid-1800s, and virtually all moths collected in industrial areas were dark by 1898. By this time, soot from factories had killed the lichens and coated the tree trunks, so dark moths were better camouflaged, as illustrated in Figure A2.1. In short, pollution had two effects that changed the environment for peppered moths – it killed lichens on tree trunks, exposing the naturally darker color of the trunks, and the particulate components of the pollution (soot) made the trunks even darker. With pollution controls starting in England in the 1950s, soot on trees was gradually washed away by rain, lichens returned to the trees, and light-colored moths were again favored by natural selection. H. B. D. Kettlewell did several experiments in polluted and unpolluted woods to test the hypothesis that predation by birds was the agent of selection on moth populations, with light moths being better camouflaged in unpolluted woods and dark moths being better camouflaged in polluted woods.

Light and dark peppered moths resting on a lichen-covered tree trunk (left panel) and on a soot-covered trunk (right panel). Look carefully to see the light moth on the lichen-covered trunk in the left panel. From Ecological Genetics, by E. B. Ford (1964, Methuen & Co.)

Light and dark peppered moths resting on a lichen-covered tree trunk (left panel) and on a soot-covered trunk (right panel). Look carefully to see the light moth on the lichen-covered trunk in the left panel. From Ecological Genetics, by E. B. Ford (1964, Methuen & Co.)

Posted in Causation, Correlation & Causation, Ecology, Ethics, Evaluating evidence, Evolution, General material, Probability & Statistics, Twins | Comments Off on Careful readers are a writer’s best friends