A post-fact world: Part III -what is a scientist to do?

I started this 3-part series noting that a lot of scientists (including myself) are very dismayed to be living in a post-fact world. I think the instinctive reaction to this that I have heard over and over again is basically “I have to do more outreach, talk more to the public, explain my science in a more understandable fashion and just get them to understand”. This is in many ways an unsurprising response. It is playing to our natural tendencies and strengths. It is in many ways doubling down on what we already do. Its also more than a little elitist (we need to educate them who don’t know as much as we). It is also empirically rejected – this is the knowledge deficit model (if people only knew more science they would behave differently) which has been thoroughly studied and resoundingly rejected (can I say trashed?) by social scientists (e.g. did you know among the general public the more scientifically literate people are, the LESS likely they are to perceive serious risks in climate change and the more likely their political affiliation is to predict their views on climate change?).  The knowledge deficit model (tell people smoking is dangerous) didn’t work to stop people from smoking. And its not working on climate change. More generally, it is not ever going to work. The literature on this is extensive.

Just to be clear, I have this “knowledge deficit” response too – I’ve spent much of the last semester working with three middle schools helping them understand climate change and exploring what they can do about it. And doing this certainly cannot hurt. So I’m not arguing against doing it or criticizing those who have these inclinations. But I am wondering if it is the best response or just the easiest and most comfortable response?

So I spent the previous two posts in this series trying to get outside of my own little scientist head and see what history and social science can tell us about how we got to a post-fact world. Namely, I argued that:

1. Our current post-fact world has been coming for half a century and is part of broad brush societal trends
2. Humans are not particularly prone to careful abstract thinking about cause-and-effects and largely choose beliefs and make decisions based on a mix of social-thinking, emotions and fast-thinking.

Or to put it succinctly, the human brain never worked by the knowledge deficit model (adding knowledge=changed beliefs and behavior) and societal trends for the last 50 years have only moved us further away from that non-existent ideal. So where does this leave us as scientists in dealing with a post-fact world?

A post-fact world: Part II – what our social scientist colleagues already know about human thoughts and behavior

This is the second post in a three part series on being a scientist in a post-fact world. The first post explored the history of how we got here. This post focuses on the fact that social scientists have pretty much known for a long time that most humans decisions are not taken based on facts and helped by increased understanding. The third will attempt to look at what scientists living in a post-fact world should do.

I am by no means a social scientist. But I most definitely recognize their existence and the validity of the work. Therefore as an honest scientist, I should first look at what is in the literature and rigorously studied before leaping to and giving primacy to my own intuitive ideas about how people’s minds work.

If I were to summarize the findings in a few words its that humans don’t base their thinking and behavioral decisions on fact-informed logic. Computers do. Spock does (did?). Academics pretend we do. Scientists arguably do in the aggregate across all scientists, but demonstrably don’t as individuals. And most humans don’t even pretend to be logical-factual in their decision making processes.

Here is a blitzkreig summary of social science literature on human decision making (so short as to be almost insulting to the complexities of the field, but hopefully digestible):

• Humans are short term thinkers – economists have formalized this in the notion of the discount rate. Businesses often use a discount rate of about 8%. That means that $100 today is worth about twice as much as $100 given to me 9 years from now, and four times as much as $100 given to me 18 years from now. Note this has nothing to do with inflation. It is a statement about how much humans are willing to defer gratification. Economists treat this as a rational behavior. You may or may not see this as rational. But it sure explains a lot about why it is really really hard to get people to care about graphs of the impacts of climate change that have a title with the year 2080 in it. Problems in 2080 press on me about 0.52% or roughly 1/200th as much as the problems I have today. Climate change is going to have to to be 200x more impactful on my life than finishing my dissertation, get tenure or raising my kids are this year.
• Humans have diverse, ordered needs – Maslow famously identified a hierarchy of needs. The idea is that humans only worry about higher level needs after lower level ones are addressed. The first priorities are physiological (food+warmth), then security (physical safety). Then comes social belonging/love followed by esteem/prestige. At the top of the list is self-actualization. Now, to be sure Maslow’s hierarchy is a simplification, and it has been corrected and amended to death, but the original idea remains compelling for capturing some core ideas. Where does keeping the planet protected come in? Where does appreciating a cool butterfly? or appreciating biodiversity?  They’re certainly pretty high in the list (i.e. low priority), quite probably only at the tippy-top actualization level. UNLESS they become part of social belonging and self-esteem, which leads immediately to …
• Humans make a lot of choices as expressions of identity and belonging – A great deal of human behavior is made as a signal of what group one belongs to more than a carefully thought out story. A great example is climate change. Circa 2000 polls showed that the best demographic predictor of belief in climate change was education level (more education made one more likely to believe in climate change). But over the 2000s climate change became simply a predictor of political affiliation (in the US Democrats were more likely to believe in climate change than Republicans). Climate change become a badge of identity and belonging rather than a fact evaluated based on education. In short, social calculus explains much of our thinking and behavior and explains many things that would otherwise seem irrational.
• Humans have dual circuits for thinking – Daniel Kahneman won a Nobel prize for his research on this topic. He calls them fast and slow thinking. Fast thinking can do amazingly complicated things including reading a billboard at high speed. But it is not based on logic or abstract thinking. And it is subject to many biases and fallacies – in short to many errors. Slow thinking is basically our Spock and the only place that applies logic and facts to arrive at novel conclusions. But the point is that a surprising number of our decisions are taken by the error-prone fast thinking part of our brain.
• Human emotions drive much decision making – This comes as a shock to classical economists but not to psychologists nor advertising executives. This simple idea has led to the far more effective campaigns against smoking that involve television campaigns and pictures on cigarette packages that involve disgusting teeth, people recoiling from the smoker etc vs the older campaigns that just put a black-and-white print message on cigarette packs saying “Warning Surgeon General has determined that cigarette smoking is dangerous to your health”

Before summarizing the implications of this for a post-fact world, let me briefly comment on how this relates to scientists:

1. Scientists are a very unusual subset of human personalities – in a nice paper by Weiler et al 2012 they use the Myers-Briggs personality assessment*. Myers-Briggs identifies four roughly independent axes of personality variation. So each individual’s personality is located in a 4-dimensional space. They compared US climate scientist’s personalities (and other types of scientists) to the general public. What they found is that on one axis (extrovert vs introvert) scientists were no different than the general public (about a 50/50 split in both cases). But on the other 3 axes, scientists were statistically significantly biased towards one end. Most importantly, scientists are more intuiters while the general public is more sensers. These words can be a little misleading but basically intuiters work with abstract thinking, while sensers work with concrete sensory-driven thinking. Scientists are also more thinkers (analytical and looking for cause and effect) than feelers (focused on empathy and personal relationships) (this is the weakest distinction of the three significant axes). Finally scientists are more judgers (prefer linear processes leading to crisp outcomes) than perceivers (happy to follow non-linear processes retaining a cloud of ambiguity). Long before this study, psychologists took the 2^4 (=16) corners of the 4-D space and identified jobs that people with that personality type were typically found in. One personality INTJ (introvert, intuiter, thinker, judger)  was actually labelled the scientist box. But INTJ is a rare corner – it is about 1.5% of the population (note that by default each corner should have 6.25% of the population). And ENTJs (extroverts but sharing the other 3 personality traits with INTJ and most scientists) are another 4% of the population So if you read the first part of this post and said “the people I hang out with aren’t this irrational”, it is because you are hanging out with a very weird outlying 5% of the population who are demonstrably unusually prone to linear thinking about cause-effect and abstract processes.
2. Scientists work by belief too – But it would be a serious mistake to take this as a badge of exclusionary elitism. All that data above about irrational thought processes applies to scientists too (we are humans before we are scientists). Get really honest with yourself about why you believe in climate change. Is it because you understand the details. Do you know the laws and equations of black-body radiation. It only requires high school algebra and geometry to calculate a rough heat balance of the earth. Have you done it? Can you name the key experiments and observations confirming this theory? Or are you using social processes to decide who you trust and believing in climate change because others say so? And what is the source of your knowledge about CO2 being a greenhouse gas – can you explain why? have you looked at empirical data? or do you “know” it because somebody you trust told you its true? Don’t worry everybody thinks using the “social computer” of many people communicating to each other. We would be absolutely paralyzed if we had to deeply know everything ourselves – even as scientists we have to trust other people to build our cognitive world view  (see an interesting-looking book on The Knowledge Illusion – Why we are never thinking alone). We differ from non-scientists only in our criteria for picking who to trust, not in avoiding “knowing” by “trusting”. Or to take a different line of argument, do you know of scientists who have a pursued an idea when it seems hopeless? sometimes they turn out right (and then become famous). One example is the story of Barry Marshall who discovered and proved that stomach ulcers are caused by a bacteria. Starting only with correlational evidence (most ulcers had the same bacteria), her persisted not only against expert opinion but through a series of experiments that would appear to reject his idea, before later experiments confirmed his idea. Lakatos recognizes this aspect very clearly – the hardened core assumptions are chosen by belief and cannot be rejected or accepted. And Kuhn’s idea of scientific revolutions clearly showed that scientists belief systems play an important role. So scientists work by beliefs and trusting others even within science! What is special about science is that we have an adversarial system in which the rules of logic win over majorities of scientists, not because individual scientists are more Spock-like than the rest of humanity.

Summary

People decide what to believe and take decisions on how to behave through a complex process in which emotions, social considerations, and error-prone “fast” thinking circuits all play a large role. The logical, analytical, abstract, fact-driven slow-thinking circuits are used fairly rarely. Scientists tend to use their slow-thinking circuits (I’m making a few leaps here) more than most people and so it is bad to generalize from how we think. But even scientists tend to primarily think with emotions and social considerations and “know” things that would probably better be described as “trust” or “believe”. And even within the realm of logical thinking it is not obvious how things far in the future or high up in the Maslow hierarchy of needs should be weighed in. The New Yorker made a similar argument recently in an article entitled “Why facts don’t change our minds?”

So what do you think? Am I overstating the irrationality of human behavior? Of scientists? Should every human think completely rationally and be data-driven about whether to smoke or worry about climate change? Whether they should or not, do they think purely rationally? What does this mean for scientists who want the world to change based on facts we discover?

 

---

*Again social scientists have moved beyond Myers-Briggs and lean more to the Big Five Personality traits which has more empirical justification. But Myers-Briggs has been used for decades, many people know how to interpret it, and it has close relationships to the Big Five. Just my opinion, but if you’ve never taken a Myers-Briggs test, it is worth the 20 minutes to see where you come out. You can find lots of decent tests online these days and it may give you some insights about yourself, but mostly it will give you some insights that not everybody else thinks the same way you do.

A post-fact world: Part I – on how we got here

Like many scientists I have been dismayed of late to observe just how much “post-fact” or “alternative-truth” the world I live in has become. Like many scientists I have been reflecting a lot lately on what it means to be a scientist living in a post-fact world. Has my job and expertise become completely irrelevant? As regular readers know when I say I’ve been reflecting, you should expect a long post (or series of posts). I have broken my thoughts up into a 3 part series. In the first I take a historical perspective and argue that this post-fact world is really the result of trends in society going on for decades. In the second part, I turn to our social scientist colleagues who have been studying how people think and choose behavior and highlight some of the most salient points that they have learned to explain why we are in a post-fact world. In the third part, I attempt (and attempt is still a pretty strong word, struggle might be better) to draw conclusions from this about what a scientist should do in a post-fact world.

This is of course an ambitious undertaking. Even an arrogant undertaking. Many caveats apply. I am neither a historian of science nor a psychologist nor risk behavior specialist. For sure, almost none of my ideas are original, but despite that, I am not going to carefully footnote each source (this being an informal blog). And I am indubitably biased in having a primarily American perspective on this (although I try to include global examples and believe the general trends are global). In short, this is just another crackpot musing on the internet. Warning: my thoughts run a little contrary to the directions I have heard many scientists heading. But the primary value of this blog to me is the chance to throw ideas out there and hear other peoples thoughts on them. So here goes …

As a peer-reviewer are you gatekeeping or editing?

Dynamic Ecology has had a couple of recent posts relating to peer review roles (reviewer, associate editor) that seem to have struck a nerve. I want to  provide some thoughts on the two fundamental roles of peer-review: gatekeeping and editing.

Michael Rosenzweig: an appreciation

I am currently attending a Festschrift this week for Michael Rosenzweig. Make no mistake, he is still actively doing science, but with 50+ years of scientific career, it seems like a good time to reflect on what an impressive career he has had. Just for full disclosure upfront, he was my PhD adviser, so I’m hardly the most unbiased reporter, but of course that gives me a close perspective.

Mike was awarded the Ecological Society of America’s Eminent Ecologist award in 2008 and he has well over 100 papers, many massively cited, and three books, so I imagine many are familiar with his published work, and it would take too much space to summarize it anyway. I want to offer several more reflective and in some cases more personal thoughts. Take them as a reflection of my respect and appreciation for Mike or my musings on the ingredients of a good scientific career as you wish.

#ibstucson – International Biogeography Society 2017 meeting

I returned this weekend from the IBS 2017 meeting in Tucson. It was a great meeting. The organizers moved it on fairly short notice from Brazil to Tucson due to concerns about Zika. This resulted in a lot of extra work for the organizers, but it didn’t show. It was a well-run meeting. And it was my favorite type of a meeting a few hundred people organized around a fairly specific topic.

I’m not going to repeat individual talks – check out the twitter feed for many great talks (#ibstucson). As is usual with me, such meetings inspire big-picture musings. This one probably more than most, since the last time I was able to attend IBS was the inaugural meeting in Mesquite Nevada in 2003. I noticed a lot of differences in the 14 year gap.

Is basic vs applied research a single dimension?

Jeremy had a post on Monday musing on a propensity for researchers that start out doing basic research and end up mixing applied research in later in their careers. I think the core observation is, on average of course, not by individual, correct. And there were a lot of spirited explanations of why this is in the comments. His framing of a single trade-off dimension between basic and applied is extremely common, and embedded in the funding of many nations’s scientific agencies (e.g. in the US, NSF only funds basic research while the US Department of Agriculture funds applied research).

But I’ve always found that trade-off limiting. Among other things, it implies something cannot be both basic and applied, something which I reject (and Don S gave a pretty spirited rebuttal of in the comments as well). I have found the notion of two trade-off axes put forth by Donald Stokes, in his book Pasteur’s Quadrant: Basic Science and Technological Innovation to be a more useful framing (also see a decent summary of the book in Wikipedia).

Serial bullies: an academic failing and the need for crowd-sourced truthtelling

I define a serial bully as somebody who repeatedly bullies new victims and never gets caught or stopped*. I don’t have exact statistics at my fingertips, but it is a definite 90/10 scenario (90% of the bullying is done by 10% of the people) – and it is that small fraction that are the serial bullies. Every campus has a PhD adviser (or three) who repeatedly abuses and victimizes his/her students. And you might have a senior colleague in your department who bullies everybody junior to her/him just because they can. Or you may have met a researcher who will do anything, ethical or not, to “win” at research, leaving behind a trail of people feeling used or abused. And although there are many unique aspects to sexual harassment, it most certainly involves bullying-like abuse of power against someone and it most certainly shares the trait that most offenders repeat over and over without getting called on it (as recent shameful cases to make the news show – just e.g. the Marcy case).You may or may not apply the word bully to all of these cases. But what all these have in common is somebody who is harming other people over and over again with little regard for the consequences, because, well, there usually are no consequences. And that is what I want to talk about.

Sample R code for yesterday’s 10 commandments post

Yesterday I presented what I tongue-in-cheek (or arrogantly – take your pick) called “10 commandments for good data management”. In that post I laid out what I believe to be best practices for managing and analyzing scientific data. Key points were to separate a data entry copy from an analysis copy of the data and to organize them differently, to use row-column organization of raw data, to use a star schema, and to denormalize data before analysis.

Here I present a worked example. It is from a hypothetical survey of raptors (data actually generated by a computer simulation). It records abundances for a number of species of raptors at a number of sites and on a number of days. The sites are unimaginatively named alpha, beta, gamma, etc. Dates are American (mm/dd/yyyy) format. Species names are real species names for raptors in North America. Abundances are made up. There is also data on temperature for each of those sites for each of those days. And some ancillary information on sites (including lat/lon coordinates). It is a constellation schema in the terms of yesterdays post. One fact table is abundance with dimensions of time, site, and taxa. The other fact table is measure with dimensions of time and site. It also has a number of errors in the data entry of the types typically seen.

Ten commandments for good data management

Usually when I am asked to give a few words to describe myself I say macroecologist or large-scale-ecologist. And I might on other days say biodiversity scientist or global change scientist. But a lot of days I would say “ecoinformatician”. Ecoinformatics is the subset of bioinformatics that applies to ecology – that is to say informatic (data) techniques applied to ecology. Some of you may know that I spent 9 years in business before returning to my PhD. But not many know that most of what I was doing was business informatics. Helping companies understand their data. It wasn’t planned. I just have always liked seeing what the data has to tell me. But it turned out to be great training as ecology dived into informatics just as I hit graduate school.

Not surprisingly given my background, I spend a lot of time being asked to make recommendations on how to work with data. I’ve also been involved in some very large data projects like BIEN. Here I don’t want to focus on the large (often social) issues of really big projects (my slides from ESA 2015 on the next 100 years of ecoinformatics are on figshare if you’re interested). Here I want to focus on the much smaller single person or lab-scale project. This post is attempts to summarize what I have learned to be best practices over both my business informatics and ecoinformatics careers. I am intentionally going to stay tool or software agnostic. In this post I really want to emphasize a frame of mind and mental approach that can be implemented in literally dozens of different software packages. In a second post tomorrow, I will give a worked example in R since I know that has the highest popularity in ecology.