2017 : WHAT SCIENTIFIC TERM OR CONCEPT OUGHT TO BE MORE WIDELY KNOWN?

There's a saying that there are no cultural relativists at thirty thousand feet. The laws of aerodynamics work regardless of political or social prejudices, and they are indisputably true. Yes, you can discuss to what extent they are an approximation, what are their limits of validity, do they take into account such niceties as quantum entanglement or unified field theory (of course they don't). But the most basic scientific concept that is clearly and disturbingly missing from today's social and political discourse is the concept that some questions have correct and clear answers. Such questions can be called "scientific" and their answers represent truth. Scientific questions are not easy to ask. Their answers can be verified by experiment or observation, and they can be used to improve your life, create jobs and technologies, save the planet. You don't need pollsters or randomized trials to determine if a parachute works. You need an understanding of the facts of aerodynamics and the methodology to do experiments.

Science's main goal is to find the answers to questions. And the rate of advance of science is determined by how well we can ask sharp, scientific questions, not by the rate at which we answer them. In the field of science with which I identify, condensed matter physics, important new discoveries and new questions arise on the scale of about once every five years. They are mostly answered and worked through on a time scale much less than that. Science is also driven by luck and serendipitous discovery. That can also be amplified by asking good questions. For example, the evolutionary biologist Carl Woese discovered a third domain of life by asking how to map out the history of life using molecular sequences of RNA rather than fossils and superficial appearances of organisms. The widely publicized ennui of fundamental physics is a result of the failure to find a sharp scientific question.

It ought to be more widely known that the truth is indeed out there, but only if one knows how to ask sharp and good questions. This is the unifying aspect of the scientific method and perhaps its most enduring contribution.

Having been born in the tiny Welsh village of Llanbadoc 141 years after Alfred Russel Wallace, I have always had a sneaking sympathy with people eager to give Wallace joint billing with Darwin for “The Best Idea Anyone Ever Had.” 

Wildlife populations are most threatened when their numbers become reduced to the point that their genetic diversity is lost. Their narrowing gene pool can accelerate into what is called an “extinction vortex.” With ever fewer gene variants (alleles), the ability to adapt and evolve declines. As inbreeding increases, deleterious genes accumulate, and fitness plummets. The creatures typically have fewer offspring, many of them physically or behaviorally impaired, susceptible to disease, increasingly incapable of thriving. Most people assume they are doomed, but that no longer has to be what happens.

“Genetic rescue” restores genetic diversity. Conservation biologists are warming to its use with growing proof of its effectiveness. One study of 156 cases of genetic rescue showed that 93% had remarkable success. The most famous case was a dramatic turnaround for the nearly extinct Florida panther. By the mid-1990s only 26 were left, and they were in bad shape. In desperation, conservationists brought in 8 female Texas cougars (which are closely related to the Florida cats). Five of the females reproduced. The result of the outcrossing was a rapid increase in litter success—424 panther kittens born in the next 12 years. The previous population decline of 5% a year reversed to population growth of 4% a year. Signs of inbreeding went away, and signs of increasing fitness grew. Scientists noticed, among other things, that the genetically enriched panthers were becoming harder to capture.

Often genetic diversity can be restored by means as straightforward as connecting isolated populations with wildlife corridors or larger protected areas, but new technological capabilities are broadening the options for genetic rescue. Advanced reproductive technology offers an alternative to transporting whole genetically-distinct parents—artificial insemination has brought genetic refreshment to cheetahs, pandas, elephants, whooping cranes, and black-footed ferrets. With the cost of genome sequencing and analysis coming down, it is becoming possible to examine each stage of genetic rescue at the gene level instead of having to wait for external traits to show improvement. This has already been done with Rocky Mountain bighorn sheep.

Another strategy being considered is “facilitated adaptation.” Different populations of a species face different local challenges. When a particular population can’t adapt fast enough to keep up with climate change, for example, it may be desirable to import the alleles from a population that has already adapted. With gene editing becoming so efficient (CRISPR etc.), the desired genes could be introduced to the gene pool directly. If necessary, the needed genes could even come from a different species entirely. That is exactly what has been done to save the American chestnut from the fungus blight that killed four billion trees early in the 20th century and reduced the species to functional extinction. Two fungus-resistant genes were added from wheat, and the trees were made blight-proof. They are now gradually in the process of being returned to their keystone role in America’s great eastern forest.

One further reservoir of genetic variability has yet to be employed. In museums throughout the world there are vast collections of specimens of species that have been reduced to genetically-impoverished remnant populations in the wild or in captive breeding programs. Those museum specimens are replete with “extinct alleles” in their preserved (though fragmented) DNA. Ancient-DNA sequencing and analysis is becoming so precise, the needed alleles can be identified, reproduced, and reintroduced to the gene pool of the current population, restoring its original genetic diversity. The long-dead can help rescue the needful living.

There's a concept from computer security known as a class break. It's a particular security vulnerability that breaks not just one system, but an entire class of systems. Examples might be a vulnerability in a particular operating system that allows an attacker to take remote control of every computer that runs on that system's software. Or a vulnerability in Internet-enabled digital video recorders and webcams that allow an attacker to recruit those devices into a massive botnet.

It's a particular way computer systems can fail, exacerbated by the characteristics of computers and software. It only takes one smart person to figure out how to attack the system. Once he does that, he can write software that automates his attack. He can do it over the Internet, so he doesn't have to be near his victim. He can automate his attack so it works while he sleeps. And then he can pass the ability to someone—or to lots of people—without the skill. This changes the nature of security failures, and completely upends how we need to defend against them.

An example: Picking a mechanical door lock requires both skill and time. Each lock is a new job, and success at one lock doesn't guarantee success with another of the same design. Electronic door locks, like the ones you now find in hotel rooms, have different vulnerabilities. An attacker can find a flaw in the design that allows him to create a key card that opens every door. If he publishes his attack software, not just the attacker, but anyone can now open every lock. And if those locks are connected to the Internet, attackers could potentially open door locks remotely—they could open every door lock remotely at the same time. That's a class break.

It's how computer systems fail, but it's not how we think about failures. We still think about automobile security in terms of individual car thieves manually stealing cars. We don't think of hackers remotely taking control of cars over the Internet. Or, remotely disabling every car over the Internet. We think about voting fraud as unauthorized individuals trying to vote. We don't think about a single person or organization remotely manipulating thousands of Internet-connected voting machines.

In a sense, class breaks are not a new concept in risk management. It's the difference between home burglaries and fires, which happen occasionally to different houses in a neighborhood over the course of the year, and floods and earthquakes, which either happen to everyone in the neighborhood or no one. Insurance companies can handle both types of risk, but they are inherently different. The increasing computerization of everything is moving us from a burglary/fire risk model to a flood/earthquake model, which a given threat either affects everyone in town or doesn't happen at all.

But there's a key difference between floods/ earthquakes and class breaks in computer systems: the former are random natural phenomena, while the latter is human-directed. Floods don't change their behavior to maximize their damage based on the types of defenses we build. Attackers do that to computer systems. Attackers examine our systems, looking for class breaks. And once one of them finds one, they'll exploit it again and again until the vulnerability is fixed.

As we move into the world of the Internet of Things, where computers permeate our lives at every level, class breaks will become increasingly important. The combination of automation and action at a distance will give attackers more power and leverage than they have ever had before. Security notions like the precautionary principle—where the potential of harm is so great that we err on the side of not deploying a new technology without proofs of security—will become more important in a world where an attacker can open all of the door locks or hack all of the power plants. It's not an inherently less secure world, but it's a differently secure world. It's a world where driverless cars are much safer than people-driven cars, until suddenly they're not. We need to build systems that assume the possibility of class breaks—and maintain security despite them.

Wherever we look, we find otherwise sane men and women making extraordinary efforts to avoid changing their minds.

Of course, many people are reluctant to be seen changing their minds, even though they might be willing to change them in private, seemingly on their own terms—perhaps while reading a book. This fear of losing face is a sign of fundamental confusion. Here it is useful to take the audience’s perspective: Tenaciously clinging to your beliefs past the point where their falsity has been clearly demonstrated does not make you look good. We have all witnessed men and women of great reputation embarrass themselves in this way. I know at least one eminent scholar who wouldn’t admit to any trouble on his side of a debate stage were he to be suddenly engulfed in flames.

If the facts are not on your side, or your argument is flawed, any attempt to save face is to lose it twice over. And yet many of us find this lesson hard to learn. To the extent that we can learn it, we acquire a superpower of sorts. In fact, a person who surrenders immediately when shown to be in error will appear not to have lost the argument at all. Rather, he will merely afford others the pleasure of having educated him on certain points.

Intellectual honesty allows us to stand outside ourselves and to think in ways that others can (and should) find compelling. It rests on the understanding that wanting something to be true isn’t a reason to believe that it is true—rather, it is further cause to worry that we might be out of touch with reality in the first place. In this sense, intellectual honesty makes real knowledge possible.

Our scientific, cultural, and moral progress is almost entirely the product of successful acts of persuasion. Therefore, an inability (or refusal) to reason honestly is a social problem. Indeed, to defy the logical expectations of others—to disregard the very standards of reasonableness that you demand of them—is a form of hostility. And when the stakes are high, it becomes an invitation to violence.

In fact, we live in a perpetual choice between conversation and violence. Consequently, few things are more important than a willingness to follow evidence and argument wherever they lead. The ability to change our minds, even on important points—especially on important points—is the only basis for hope that the human causes of human misery can be finally overcome. 

Time is a scientific concept that deserves greater thought and study, though, despite advancements in the mathematical behaviors of time over long horizons under what I would call extreme conditions, it is a concept we will never be positioned to properly understand. 

The word "meme" denotes a rapidly spreading idea, behavior, or concept. Richard Dawkins originally coined the term to explain the action of natural selection on cultural information. For example, good parenting practices lead to children who survive to pass on those good parenting practices to their own children. Ask someone under twenty-five what a meme is, however, and chances are you will get a different definition: typing "meme" into Google Images yields page after page of photos of celebrities, babies, and kittens, overlaid with somewhat humorous text. Memes spread only as rapidly as they can reproduce. Parenting is a long-term and arduous task that takes decades to reproduce itself. A kitten photo reproduces in the few seconds it takes to resend. Consequently, the vast majority of memes are now digital, and the digital meaning of meme has crowded out its social and evolutionary meaning. 

Some memes are fortunate at birth: they represent clear new concepts, are blessed with a memorable name, and have prominent intellectual "parents" who ably shepherd them through the crucial initial process of dissemination, clarification and acceptance. "Meme" itself is one of those lucky memes. Many other memes, however, are less fortunate in one or more of these respects, and through no fault of their own languish, for decades or even centuries, in the shadows of their highborn competitors. 

Why, in an age in which we have the world’s information easily accessible at our fingertips, is there still so much widespread disagreement between people about basic facts? Why is it so hard to change people’s minds about truth even in the face of overwhelming evidence?  

Humans have fewer than two legs... 
...on average. 

Carl Sagan spoke too soon when he spoke about demons. Modern science, he told his numerous followers, banished witches, demons and other such creatures from this world. Simply spread flour on the floor and check for suspicious footprints—this kind of reasoning, he claimed in The Demon-Haunted World, characterizes sound, scientific thinking.

So why is Maxwell’s demon still on the frontlines of science? Since he was first conjured in 1874, modern day inquisitors have valiantly chased after him with math and physics instead of holy water. But rather than going the way of the phlogiston and the ether, he has emerged unscathed and is now a fixture in standard physics textbooks. Lest you think he is real, let me tell you he is not. But a demon he is in spades. He sorts, and sorting is at the origin of sorcery.

Neat-fingered and vigilant, he can reverse time, momentarily violate the second law of thermodynamics, power a perpetual motion machine, and generate pockets of hellish heat in substances that should otherwise reach temperature equilibrium. His smarts are debatable, matching those of a virtuoso piano player or those of a humble switchman on railway tracks. No offense is taken if he is compared to a simple valve. On the contrary, it means he can do a lot of work with minimum effort. Frequently portrayed as holding a cricket bat (to send molecules to-and-fro), manning a trapdoor (to let them pass or keep them out) and holding a torch, a flashlight or a photocell (to be able to see in the dark), this miniscule leviathan can wreak havoc.           

He was named after the Scottish scientist James-Clerk Maxwell, known for his theory of electromagnetism, by his colleague William Thomson (aka Lord Kelvin). Almost immediately after his public debut, he was exorcized. Sightings of his demonic activity were brushed away as statistical anomalies or insignificantly tiny—no need to worry, we were told.          

But science can, and often does, turn imaginary beings into real things. Maxwell’s demon is a case in point. Since he was first conjured, scientists have tried to bring him to life repurposing twisted metal, ratchets and gears, molecules, enzymes and cells, and even electronics and software. In 1929 the physicist Leo Szilard published a paper about him in the prestigious Annalen der Physik (which reached fame as “Szilard’s exorcism”) before pairing up with Einstein to patent a refrigerator that would momentarily and locally reverse entropy. Another milestone took place in 2007, when an article in Nature titled “A Demon of a Device” described some of the first successful molecular nanomotors. At that time, Sir Fraser Stoddart considered it as auguring an entirely new approach to chemistry; he went on to win the Nobel Prize for Chemistry in 2016. Artificial intelligence is another technology with a connection to Maxwell’s being.

Maxwell’s demon is neither the first nor the last, but he is certainly the most well known of all science’s demons. Descartes’s demon preceded him by more than two centuries. Like a master illusionist who can take over your sense of reality by throwing a cloak over your head, this genie can intercept your sense-impressions and take over from there. He remains the patron saint of virtual reality. Laplace’s demon comes next: the master calculator who can relate every particle to the laws of motion and who can know the past and the future has inspired advances in supercomputing and Big Data. The so-called “colleague of Maxwell’s demon,” appeared a few decades after the original one. His arriviste career included travelling faster than light and helping to explain quantum entanglement. By sharing with biblical shedim the power of instantaneous locomotion, he can intercept messages, mess with causality, and fiddle with time. These shape-shifting masters of disguise are getting smarter and more powerful. By taking on the name of the scientists who conjured them, they dutifully fulfill their patronymic destiny, reappearing regularly as spick-and-span newborns with the sagacity of old men.

At the turn of the nineteenth century the French mathematician Henri Poincaré claimed he could almost see Maxwell’s demon through his microscope, calculating that with a little patience (of millions of millions of centuries) his mischief would be evident for everyone.  While some deemed him too small to really matter for us, others noted that small causes were known to produce great effects, like the spark or the pebble starting an avalanche. Other scientists argued that the universe as a whole was so large that he could confidently reign like a master in a confined territory. For Norbert Wiener, founder of cybernetics, it was clear that “there is no reason to suppose that metastable demons do not in fact exist.” The physicist Richard Feynman wrote an entire article explaining why, when and how he would eventually tire. According to Isaac Asimov, everything that appeared to us as if arriving by chance was really because “it is a drunken Maxwell’s demon we are dealing with.”

It is perhaps most ironic that the philosopher Sir Karl Popper, known for describing the process of scientific progress as one based on hypothesis-creation and falsification, admired his ability to survive every assassination attempt against him: “Although innumerable attempts have been made on his life, almost from the day he was born, and although his non-existence has frequently been proved, he will no doubt soon celebrate his hundredth birthday in perfect health and vigor.”

Demons are here to stay. The reassuring predictions of scientists such as Carl Sagan have proved to be overly premature. In our hurry to make our world modern and base it on purely secular reason, we have failed to see the demons in our midst.

The term, “media richness,” was first described in the context of media richness theory (MRT) by Richard Daft and Robert Lengel in 1986. Media richness describes the density of learning that can be conveyed through a specified communications medium. Face-to-face communication is the richest medium according to MRT because it allows for the simultaneous interpersonal exchange of cues from linguistic content, tone of voice, facial expressions, direction of gaze, gestures, and postures. MRT was developed prior to the rise of electronic communication media in order to help managers in business contexts decide which medium was most effective for communicating a message. Rich media like conversations and phone calls were deemed best for non-routine messages, while lean media like unaddressed memoranda were considered acceptable for routine messages. In the last two decades, media richness has been extended to describe the strengths and weaknesses of new media from email to websites, video conferencing, voicemail, and instant messaging. Media richness deserves to be more widely known because people make choices throughout a day about communications media often without considering the consequences of the choice of medium, and the goodness of fit between the content of a message and the medium through which it is being communicated.

Humans evolved in media-rich contexts. Living in stable, tightly knit social groups, face-to-face communication was the only mode of communication for hundreds of thousands of years. Until about 5,000 years ago, the concept of media choice didn’t exist because—apart from smoke signals—it was face-to-face or nothing. Articulate speech and language complemented the rich repertoire of vocalizations, facial expressions, glances, stares, gestures, and postures, and upon which our ancestors relied, thus creating a rich and potentially highly nuanced communications repertoire. Within small groups, people attended closely to what was said, who said it, and how it was said. Pleasantries were exchanged, advice was given, loving whispers traded, and admonitions delivered with a full sensory armada of verbal content, tone of voice, measured eye contact, gesture, and posture. People were mostly bathed in conversation, reassured by touch, verbally upbraided for unreasonableness, publicly shamed by calculated stares, and physically reprimanded for antisocial behavior. Although communication between people has probably never been without the potential for guile and social manipulation, deception was hard to pull off because information flowed through multiple visual, auditory, and even tactile and olfactory channels. Communication had immediate effects and consequences. One-to-many communication reached only as far as the human voice could carry.

The consequences of media richness and the concept of media choice became relevant for people only with the introduction of writing in early agricultural societies. Initially developed to facilitate clerical and payroll functions, writing was soon marshaled in support of military, political and religious causes, and much later for the exchange of personal information and the composition of poetry and philosophical treatises. Communication through writing was augmented early in the 20^th century by modes of remote voice communication (telephone, microphone, and radio), and later by combined visual and auditory modes such as movies, television, and websites which provided unprecedented scope for unidirectional communication. Historians and scholars of communication theory talk about tradeoffs between the richest face-to-face channels and the leaner modalities of email, voicemail, and text messaging that provide fewer cues, slower feedback, and limited scope for retribution. Media richness has been criticized in recent years because it fails to predict why people choose lean over rich media, especially in situations where a richer medium clearly would be more effective. Many will have experienced the shock of an email informing about the death of a loved one or have been anguished by a misplaced comma or inappropriate emoticon in a text message rejecting an overture of friendship or love. This is exactly why media richness is important and interesting from an evolutionary perspective.

In a world of unconstrained media choice, people often choose leaner and functionally unidirectional modalities because they want to make a point, or at least think that they want to make a point. A need for incessant and immediate connection (or just the need to save money) can provoke the blurting of something through a cheap, low-grade channel rather than waiting for the chance to use a richer one. Leaner media also carry lower risk of rejection or immediate retribution. Regardless of the reason, we now live in a world where people are opting for leaner modes of communication because they have been socialized inadequately in richer ones and are functionally ignorant of the concept of media richness. The scope for misunderstanding has never been greater, while the opportunities for providing physical comfort and solace or for exacting meaningful and appropriate retribution have never been more limited. We still yearn to see one another, but contacts often consist more of broadcast performances of static faces and less of breathing exchanges of shared wonder, love, tribulation, and loss.

Like all primates, humans have nurtured harmonious relationships and maintained social cohesion by being intensely good, high-bandwidth communicators. Media richness is a concept worthy of wider propagation because it will help insure the future of individuals and societies in times of increasing individual social isolation, electronic bullying, touch aversion, personal anxiety, and social estrangement.

Here is a trivial fact about our mental lives. So trivial, it is rarely even noticed. And it is hard to talk about without sounding sophisticated. The post-modernists have explored versions of it, but the notion I mean to promote—the notion of “construal”—is painfully obvious. It refers to the fact that our attitudes and opinions and choices pertain to things not as they are in the world, but as they are represented in our minds.

Economic theorizing presumes that people choose between options in the world: Job A versus Job B, or Car A versus Car B. From the point of view of a psychologist, however, that presumption is really quite radical: When a person is presented with a choice between options A and B, she chooses not between A and B as they are in the world, but rather as they are represented by the 3-pound machine she carries behind the eyes and between the ears. And that representation is not a complete and neutral summary, but rather a selective and constructed rendering—a construal.

There is, of course, no way around it. The behaviorists tried to avoid it by positing that behaviors were direct responses to stimuli, that mental life didn’t interfere in relevant ways. But, clearly, that’s not the case. We now know a lot about our rich mental lives, which shape and mold what we experience, making construal not neutral. A food when it is 10% fat is less appetizing than when it is 90% fat free. A risky venture that entails some lives saved and others lost is a lot more appealing when our attention is directed towards the lives saved than the lives lost. Our attempts to elicit empathy for global catastrophes are ineffective—a phenomenon referred to as “psychic numbing”—partly because our construal processes are not able to trigger differential indignation for outcomes as a function of their gravity.

Visual illusions provide a compelling illustration, where our experience of the object simply does not conform to the actual object in the world. Susan Sontag famously observed that “to photograph is to frame, and to frame is to exclude.” In fact, the mind is a lot messier than a camera. We don’t merely choose where to look—our minds influence what we see. And they influence what we see both when we think fast and when we think slow; both when we respond impulsively, without conscious thought, and when we deliberately choose what to take seriously and what to ignore.

Construal lies at the core of behavioral economics. Violations of standard rationality assumptions arise not from stupidity, computational limitations or inattention, but from the simple fact that things in the world, depending on how they are described or interpreted, get construed differently, yielding inconsistent judgments and preferences.

Real world options, like automobiles, houses, job offers, potential spouses, all come in multiple attributes. How much weight we give each attribute is largely a function of where our attention is directed, our pet theories, what we expect or wish to see, the associations that come to mind. One rule of construal is that things are judged in comparative rather than absolute terms. How water feels to the hand depends on whether the hand had previously been in colder or warmer water. In the delivery room, a doctor’s decision of whether to perform a caesarian section depends on the gravity of immediately preceding cases.

Knowledge in the form of scripts, schemas, and heuristics, serves to make sense of stimuli in ways that transcend what is given. What we experience is determined not simply by the objective building blocks of the situation, but by what we know, care about, attend to, understand, and remember. And what we care about, attend to, and remember is malleable. In one study, participants were invited to play a Prisoner’s Dilemma game, referred to as either the Wall Street or the Community game. While the payoffs and set-up were identical, the mere label altered participants’ construal, changing their tendency to cooperate or to defect.

Psychological costs and subsidies also enter people’s construal, and are quite different from the financial costs and subsidies policy makers are typically concerned with. In one well-known study, when fines were introduced for picking up children late from daycare, parents were more likely to pick up their children late. Parents who had previously felt bad—had incurred a psychic cost—for showing up late now construed the fine as a contract—paying a fee entitled them to late pick-up.

Psychologists see it as an integral feature of human cognition, but if your aim is to influence behavior, construal presents a difficult challenge. The difference between success and failure often boils down to how things are construed. Although similar from an accounting point of view, Earned Income Tax Credit (EITC)—in contrast with TANF and other forms of welfare—has been an effective form of government assistance. This is attributed to the construal of EITC as just reward for labor, delivered in the form of a tax refund check, rather than a separate assistance payment. It is seen as an entitlement rather than welfare, designating beneficiaries as taxpaying workers, rather than “on the dole.”

Construal needs to be more widely appreciated because so much thinking and intuition, in policy and in the social sciences, tends to focus on actual circumstances as opposed to how they are construed. The words that make up this essay are just words. It’s partly their construal that will make some readers think they’re useful and others think they’re of little use. 

Five years from now, what will your future self think of your current self? Will you even be the same person?

Over a century before the birth of Christ, Bactrian King Milinda challenged the Buddhist sage Nagasena to define identity. The Buddhist responded by inquiring about the identity of the king’s chariot: “…is it the axle? Or the wheels, or the chassis, or reins, or yoke that is in the chariot? Is it all of these combined, or is it apart from them?” The king was forced to concede that his chariot’s identity could not be reduced to its’ pieces. Later, Greek scholar Plutarch noted that the passage of time further complicates definitions of identity. For example, if a ship is restored piece by piece over time, does it retain its’ original identity? As with Theseus’ reconstructed ship, the paradox of identity applies to our constantly regenerating bodies (and their resident brains). Flipping in time from the past to the future, to what extent can we expect our identity to change over the next five years?

Considering the future self to be an entirely different person could have serious consequences. Philosopher Derek Parfit worries that people who regard the future self as distinct should logically have no more reason to care about that future self than a stranger. By implication, they should have no reason to save money, maintain their health, or cultivate relationships. But perhaps there is a middle ground between self and stranger. To the extent that someone imagines their future self to be similar to their present self, this sense of “future self-continuity” might predict their willingness to at least consider the interests of the future self.

Here, I argue that beyond rekindling philosophical debates, future self-continuity is a critical (and timely) scientific concept, for a number of reasons. First, future self-continuity can be measured. Remarkably, neuroimaging research suggests that a medial part of the frontal cortex shows greater activity when we think about ourselves versus strangers. When we think about the future self, the activity falls somewhere in between. The closer that activity is to the current self, the more willing individuals are to wait for future rewards (or to show less “temporal discounting” of future rewards). More conveniently, researchers can also simply ask people to rate how similar or connected they feel to their future selves (e.g., in five years). As with neural measures, people who endorse future self-continuity show less temporal discounting, and have more money stashed in their savings accounts.

Second, future self-continuity matters. As noted, individuals with greater future self-continuity are more willing to wait for future rewards—not just in the laboratory, but also in the real world. Applied research by Hal Hershfield and others suggests that adolescents with greater future self-continuity show less delinquent behavior, and that adults with greater future self-continuity act more ethically in business transactions. Future self-continuity may even operate at the group level, since cultures that value respect for elders tend to save more, while nations with longer histories tend to have cleaner environments.

Third, and most importantly, future self-continuity can be manipulated. Simple manipulations include writing a letter to one’s future self, whereas more sophisticated interventions involve interacting with digital renderings of future selves in virtual reality. These interventions can change behavior. For example, adolescents who write a letter to their future selves make fewer subsequent delinquent choices, and adults who interact with an age-progressed avatar later allocate more available cash towards retirement plans. While the active ingredients of these manipulations remain to be isolated, enhancing the similarity and vividness of future self representations seems to help. Scalable future self-continuity interventions may open up new channels for enhancing health, education, and wealth.

The need for future self-continuity continues to grow. On the resource front, people are living longer while job stability is decreasing. In the face of increasing automation, institutional social safety nets are shrinking, forcing individuals to bear the full burden of saving for their futures. And yet, in the United States, saving has decreased to the point where nearly half of the population would have difficulty finding $400.00 to cover an emergency expense. On the environmental front, global temperatures continue to rise to unprecedented levels—along with attendant droughts, increases in sea levels, and damage to vulnerable ecologies. Human choice has a hand in these problems. Perhaps increased future self-continuity—in individuals as well as policymakers—could generate solutions.

The dawn of a new year is as good a time as any to take the perspective of your future self. Imagine yourself in five years. Did you do everything you could today to make the world a better place—both for your present and future selves? If not, what can you change?

For all its fiber-enabled, live-video-streaming, 24/7-connected promise, our information network encapsulates a fundamental flaw: it’s proving a suboptimal system for keeping the world informed. While it embraces nodes dedicated to propagating a rich seam of information, because the internet’s governing algorithms are optimized to connect us to what they believe we are already looking for, we tend to retreat into familiar and comfortably self-reinforcing silos: idea chambers whose feeds, tweets and updates inevitably echo our pre-existing prejudices and limitations. The wider conversation, a precondition for a healthy intellectual culture, isn’t getting through. The signals are being blocked. The algorithmic filter is building ever-higher walls. Facts are being invalidated by something called “post-truth.” And that’s just not healthy for the quality of informed public debate that Edge has always celebrated.

Thankfully a solution is suggested by neural networks of a biological kind. Inside our brains, no neuron ever makes direct contact with another neuron; these billions of disconnected cells pursue their own individual agendas without directly communicating with their neighbors. But the reason we are able to form memories, or sustain reasoned debate, is that the very gaps between these neurons are programmed to build connections between them. These gaps—called synapses—connect individual neurons using chemical or electrical signals, and thus unite isolated brain cells into a healthy central and peripheral nervous system. The synapses transfer instructions between neurons, link our sense receptors to other parts of our nervous system, and carry messages destined for our muscles and glands. Truth be told, without these unheralded gap-bridging entities called synapses, our disconnected brain cells would be pretty irrelevant.

We need to celebrate the synapse for its vital role in making connections, and indeed to extend the metaphor to the wider worlds of business, media and politics. In an ever-more atomized culture, it’s the connectors of silos, the bridgers of worlds, that accrue the greatest value. And so we need to promote the intellectual synapses, the journalistic synapses, the political synapses—the rare individuals who pull down walls, who connect divergent ideas, who dare to link two mutually incompatible fixed ideas in order to promote understanding.

Synaptic transfer in its scientific sense can be excitatory (encouraging the receiving neuron to forward the signal), or inhibitory (blocking the receiving neuron from further communicating the message). Combined, these approaches ensure a coherent and healthy brain-body ecosystem.  But as we promote the metaphorical sense of synaptic transfer, we can afford to be looser in our definition. Today we need synapse-builders who break down filter bubbles and constrained world-views by making connections wherever possible. These are the people who further healthy signalling by making unsolicited introductions between those who might mutually benefit; who convene dinner salons and conferences where the divergent may unexpectedly converge; who, in the Bay Area habit, “pay it forward” by performing favors that transform a business ecosystem from one of hostile competitiveness to one based on hope, optimism and mutual respect and understanding.  

So let’s re-cast the synapse, coined a century ago from the Greek words for “fasten together,” and promote the term to celebrate the gap-bridgers. Be the neurotransmitter in your world. Diffuse ideas and human connections. And help move us all beyond constrained thinking.

What exactly do brains do? The usual answer is that they form mental representations of the world on the far side of the skull. Brains, that is, create internal virtual worlds—their best or most useful simulations of the real external world, one which exists independently of any of them but within which they all reside.

The problem, however, is that fundamental physics denies the existence of this observer-independent world. From quantum physics in the early 20^th century to the black hole firewall debate that rages today, physicists have found that we tangle ourselves in paradox and violate laws of physics when we attempt to compile multiple viewpoints into a single spacetime. The state of a physical system, we’ve learned, can only be defined relative to a given observer. (Here “observer” does not mean consciousness, but a physical system capable of acting as a measuring device—yet one that itself must enjoy only a relational existence.) Slices of spacetime accessed by different observers cannot be considered broken shards of a single, shared world, but rather as self-contained and incommensurable versions of reality, each a universe unto itself.

In other words, there’s no third-person view of the world. There is one world per observer, and no more than one at a time.

What happens, then, to the concept of representation? What is it that brains are doing if there is no observer-independent world out there for them to represent?

One possibility is that instead of representing the world, brains enact one.

The term “enactivism” was introduced by Francisco Varela and colleagues in the 1990s, but it’s taken on fresh significance in light of an emerging set of ideas at the forefront of cognitive science today, including embodied cognition, the Bayesian brain, active inference and the free energy principle—ideas that emphasize top-down, generative perception wherein observers actively shape the worlds they perceive. The old passive view of perception is giving way to an active one, just as the Newtonian observer was replaced by the participator of modern physics. Suddenly the words of physicist John Archibald Wheeler apply to cognitive science: “We used to think that the world exists out there, independent of us, we the observer safely hidden behind a one-foot thick slab of plate glass, not getting involved, only observing. However, we’ve concluded that that isn’t the way the world works. We have to smash the glass, reach in.”

According to enactivism, observer and world co-evolve, hoisting one another up by their bootstraps through reciprocal interaction. Perception and action are inextricably and cyclically linked: our perceptions guide our actions and our actions determine what we perceive. Their symmetry ensures an even more essential symmetry, that between observer and world. The observer’s actions form the world’s perceptions, and the world’s actions form the observer’s perceptions. Labels such as “observer” and “observed,” “inside” and “outside,” thus become profoundly interchangeable, removing any need to resort to mystical, magical or hopelessly vague talk about consciousness as something over and above the physical world.

What the enactivist perspective leaves us with, ontologically speaking, is an observer and world that exist relative to one another and not in any absolute way. The enacted world is not one that can be described from a third person perspective. It is observer-dependent, rendered in first person, no more than one at a time. Which, of course, is exactly the ontology prescribed by fundamental physics.

Does it really matter if cognitive science aligns with fundamental physics? For the everyday purposes of research and practice in neuroscience, representation works well enough, just as our belief in a single, shared physical world mostly suffices, whether we’re driving to work or launching rockets to the moon. It’s only when we push to the outermost edges of either discipline—when dealing with tiny distances or intense gravity in physics, or when asking about the fundamental nature of consciousness in cognitive science—that the cracks in the third person perspective begin to show and a more fundamental theory is needed.

Fundamental physics and cognitive science have long been embroiled in a perpetual game of chicken or egg: the brain arises from the physical world and yet everything we know of the physical world we know only through the brain. Each serves as foundation for the other. So long as we restrict our inquiry to one at a time, something critical is left unexplained. If we want to understand reality as a whole, we need to understand both sides of the coin and how they are fused together. Physics finds that the world is observer-dependent but remains silent on the nature of the observer. Cognitive science finds that the observer is an active participant but never questions the nature of the physical world. Enactivism might just be the concept we need to begin to piece the two sides together. 

When we measure a value that is consistent with the prediction our tendency is to trust the result.  When we get an unexpected answer we apply greater scrutiny, trying to determine if we’ve made an error before believing the surprise. This tilts us toward revealing errors in one category of measurements and leaving them unexposed in another. It is particularly dangerous when our assumptions are flawed—and if there is anything we should bet on, given the history of progress within physics, it’s that our underlying assumptions are in some way flawed. Bias can creep into the scientific process in predictable and unpredictable ways.

Blind analyses are employed as a protection against bias. The idea is to fully establish procedures for a measurement before we look at the data so we can’t be swayed by intermediate results. They require rigorous tests along the way to convince ourselves that the procedures we develop are robust and that we understand our equipment and techniques. We can’t “unsee” the data once we’ve taken a look.

There are options when it comes to performing a blind analysis.  If you are measuring a particular number, you can apply a random offset to the number that is stored but not revealed to the analyzers. You complete the full analysis and reveal the offset and true result only when the work is done. Another method is to designate a sensitive segment of the data, the “signal,” as off limits. You don’t look at the signal until you’re convinced that you understand the remaining data, the “background.”  You can fully develop your analysis using the background and a simulated fake signal. Only when the analysis is fully developed do you look at the signal and obtain the result, a process known as “opening the box.” Another flavor of blind analysis was employed by the LIGO experiment in the discovery of gravitational waves. Fake signals were periodically inserted into the data so that full analyses were undertaken without analyzers knowing if the signals they were seeing were real. They carried the analyses all the way to the point of preparing the corresponding publication before they learned if they were analyzing real or fake data. 

Blind analyses force scientists to approach their work with humility, acknowledging the potential for bias to influence the process. They require creativity and rigor as we establish an understanding of the data without direct access to it. They enforce good stewardship of the data, which can represent significant investments into experiments that are not easily repeatable. They highlight the mystery and anticipation inherent in the discovery process where opening the box has the potential to reveal a surprise. Humility, rigor, stewardship, and mystery are the essential ingredients of blind analyses and represent the best that science has to offer.    

Given the operation of squaring a number, there are two numbers that are special because the operation doesn't change them: 0 and 1. 0 squared is 0, and 1 squared is 1, but square any other number and the output will differ from the input. These special numbers are called fixpoints (of squaring). In general, a fixpoint is a value (or state of a system) that is left unchanged by a particular operation. This concept, easily definable by the brief equation x=f(x), is at the essence of several other ideas with great practical significance, from Nash equilibria (used in economics and social sciences, to model failures of cooperation) to stability (used in control theory to model systems ranging from aircraft to chemical plants) to PageRank (the foremost web search algorithm). It even appears in work at the foundations of logic, to give definition to truth itself!

A prevalent example in everyday life is the occurrence of fixpoints in any strategic context ("game") with multiple players. Suppose each player can independently revise their strategy so as to best respond to the others' strategies. If we consider this revision process as an operation, then its fixpoints are those combinations of strategies for which the revision operation yields exactly the same combination of strategies: that is, no player is incentivized to change their behavior in any way. This is the concept of a Nash equilibrium, a kind of fixpoint that directly applies to many real-world situations—ranging in importance from roommates leaving dirty dishes in the sink to nuclear arsenals. Moving to a different Nash equilibrium (such as disarmament) requires changing the revision operator (e.g., with an agreement that binds multiple players to change their strategies at the same time).

Fixpoints are also an extraordinarily useful framework to consider ideas that seem to be defined circularly. As a concrete example, the golden ratio phi can be defined as phi=1+1/phi, which actually means phi is the unique fixpoint of the operation that takes the reciprocal then adds one. In general, self-referential definitions can be classified according to the number of fixpoints of the corresponding operation: more than one, and it's imprecise but potentially serviceable; one, and it's a bona fide definition; lacking any, and it's a paradox (like "this sentence is false"). In fact, it is operations that lack fixpoints which underly a whole host of deeply-related paradoxes in mathematics, and led to Gödel's discovery that the early-20th-century logicians' quest to completely formalize mathematics is impossible.

Mathematics is not the only reasoning system we might use; in daily life we use ideologies and belief systems that aren't complete—in the sense that they can't express arbitrarily large natural numbers (nor do we care for them to). We can consider persistent ideologies and belief systems as fixpoints of the revision operation of "changing one's mind"—subject to the types of questions, evidence, and methods of reasoning that the ideology judges as acceptable. If there's a fixpoint here, we're unable to change our minds in some cases, even if presented with overwhelming evidence. As with Nash equilibria, the key to escaping such a belief system is to modify the revision operation: considering just one new question (such as "what questions am I allowed to ask?") can sometimes be enough to abolish ideological fixpoints forever. 

One remarkable feature of the scientific belief system is its non-fixedness: new beliefs are constantly integrated, and old beliefs are not uncommonly discarded. Ideally, science would be complete in the limit of infinite time and experiments without losing its openness, analogous to how programs that produce never-ending lists (such as the digits of pi or the prime numbers) are formally given meaning by infinitely large fixpoints which may only be successively approximated. While an absence of fixpoints in the logical foundations of arithmetic dooms it to "incompleteness," fixpoint theorems have recently been used to show that if we relax our notion of completeness to the almost-equally-satisfying concept of "coherence," there is a revision operation which is guaranteed to have a coherent fixpoint, and can even be approximated by computable algorithms!

Perhaps science, too, aspires to an unreachable, infinite fixpoint in which all knowable facts are known and all provable consequences are proven, such that there would be no more room to change one's mind—and we hope that with each passing year, our current state of knowledge more closely approximates that ultimate fixpoint.

The word scale, when it refers to an object, may refer to a simple ruler. However, the act of measurement is the start of a deep relationship between geometry, physics and many important endeavors of humanity.

Scaling, in the geometrical context, is the act of resizing an object while preserving certain essential characteristics such as its shape. It indicates a relationship that is robust under certain transformations.

One uncovers scaling relationships by measuring and plotting key variables against each other. The simplest scaling relation is a linear one—yielding a straight line on a X-Y plot—denoting proportionality. The number of miles you can travel before refueling scales roughly linearly with the amount of fuel left in your gas tank. Twice the gas—double the miles that can be driven!

More complicated scaling functions include power-laws, exponentials and so on each of which are reflective of the underlying spatial geometry or dynamics of the system.

The circumference of a sphere scales proportionally with its radius. The surface area and volume scale, on the other hand, as the square and cube of the radius respectively. These power-law relationships show up as straight lines on a log-log plot. The slopes reflect the dimension of the object being measured demonstrating that length is one-dimensional, area is two-dimensional and volume is three-dimensional. This is true regardless of the shape of these solids—spherical, cubical and so on - and so plots using other solids will show the same robust scaling relationships and scaling exponents.

But sometimes length does not scale with an exponent of one—what then?

Richardson’s pioneering measurements of the length of coastlines across the world showed that measuring with increasing precision gave ever-increasing answers for their lengths. The various data sets (measurements with different precision) for each coastline scaled using an exponent greater than one.

Benoit Mandelbrot, in his classic paper “How long is the coast of Britain?” provided the insight needed to understand this result. As one measures the coastline with ever-increasing precision, one measures and adds the lengths of ever-smaller fjords, nooks and crannies. The coastline is not a smooth object; over a certain range of length-scales it is rough and undulating enough that it behaves like a self-similar object with a dimension between one and two. This insight led to the development of fractal geometry—the geometry of irregular, fractured objects: coastlines, mountains…broccoli. Systems that possess the property of self-similarity often show power-law scaling—alluding to a beautiful link between geometry and physics. Incorporating fractal geometry also turns out to be important for computer algorithms to generate the realistic-looking artificial worlds in movies and video games that we now take for granted.

More generally, a variety of natural phenomena appear scale-invariant (or look the same) when an important underlying physical parameter is changed (or rescaled) in a specific way. This physical parameter in many cases is a carefully constructed dimensionless quantity e.g. the ratio of two key length scales in the system being studied. The equations describing the phenomena must then obey and thus reflect the observed scale invariance.

Richardson’s delightful ditty captures the scale invariance of the Navier-Stokes equations of fluid turbulence—which remain unsolved to this day.

Big whirls have little whirls that feed on their velocity,
and little whirls have lesser whirls and so on to viscosity.

The laws of physics embody scaling. Newton’s law of gravitation states that the attractive gravitational force between two bodies scales linearly with each of their masses, and inversely with the square of the distance between them.

If you get large enough or small enough, almost any scaling law observed in the real world breaks down. This makes us think about the range of applicability and what causes the break down. The scaling relations implied by Newton’s laws break down at very small distances—giving way to quantum mechanics. The same happens at very large velocities - giving way to Einstein’s theory of relativity.

Thus it is often useful to refer to a scale to specify the range over which a theory or observed phenomenon is applicable: one refers to the quantum or sub-atomic scale, the human scale, the astronomical scale and so on.

Scaling concepts have found broad applicability in unexpected areas such as the study of social networks.

Technology companies are often not constrained by geography and can easily scale up their user base. Often early stage technology startups show exponential increases in usage that are then reflected in exploding valuations. Supply-demand problems that require resources to scale exponentially are simply not sustainable. As companies mature the exponential scaling and valuation must level off. On the other hand, power-law relationships and solutions indicate a business that may be scalable for a sustained period and are sought after by investors with a longer time horizon.

During the first internet bubble (circa 2000), companies rushed to get users and valuations were based on the number of people using their web-sites.

Today, social network companies value users on the strength and quality of their interactions with other users. One can argue that valuations of these companies should scale quadratically with the number of users. It remains to be seen if such scaling arguments used to justify higher valuations hold up, compared to more traditional measures such as revenue and profitability of a company.

Scaling relationships inevitably break down beyond a certain range, providing an important clue that other effects either ignored or not yet considered are now important. One should not ignore these deviations. While solving for persistent social problems that appear at all scales one should remember—what works for a family may not work for a business. What works for a business may not work for a nation.

Humans all over the world share the same genome, the same neural architecture and the same behavioral niche (three-generational system of resource provisioning, long-term pair-bonding between men and women, high levels of cooperation between kin and non-kin). At the same time, human cultures are highly variable. Some societies see revenge as a duty, others as a sin; some regard sex as a pleasure, others as a danger, and some reward innovations while others prefer traditions. In many instances, these cultural differences are robust, and can last for millennia despite cultural contacts, political assimilation or linguistic replacement.

How can we account for such variability? Traditionally, it is assumed that cultural variability cannot be explained by species-specific evolved mechanisms and that it must be the product of socially transmitted norms in the forms of religious beliefs, informal enforcement or political conquest.

This assumption is based on a common misconception about natural selection, which is wrongly thought to select mechanisms that systematically produce universal, uniform and unchanging behaviors. But all evolved mechanisms, physiological or psychological, actually come with a certain level of flexibility in response to local contexts. This is called phenotypic plasticity. The genotype codes for a mechanism that is able to express different phenotypes (organs, behaviors) in response to detectable and recurring changes in the environment.

Tanning is a case in point. While the mechanism of skin pigmentation is a universal adaptation to protect human cells from ultraviolet damage and to synthesize vitamin D, it responds differently to different contexts, making skin darker in low latitudes (and in the summer) and lighter in high latitudes (and in the winter).

Phenotypic plasticity has been shown to be evolutionary advantageous when there are different optimal phenotypes in different environments. This is of course the case for tanning: the optimal level of tanning differs whether an individual lives in high or low latitudes and whether it is winter or summer. Natural selection must therefore be able to maintain the right skin color despite variations in the environment. One solution is to select for a certain level of plasticity (skin pigmentation is not completely plastic) and to build a mechanism that can detect the amount of light, and adjust the level of melatonin accordingly.

The study of phenotypic plasticity has long been limited to physical traits (such as skin pigmentation). However, recent research in neuroscience, ecology and psychology has shown that phenotypic plasticity extends to behaviors. For instance, in a harsh and unpredictable environment where the future is dreary, organisms tend to adopt a short-term life strategy: maturing and reproducing earlier, investing less in offspring and in pair-bonding, being more impulsive. On the contrary, in a more favorable and predictable environment, organisms switch to a long-term strategy: maturing and reproducing later, investing in offspring and in pair-bonding, and being more patient. Importantly, these switches between present-oriented and future-oriented behaviors can affect all kind of behaviors: reproduction and growth of course, but also attitudes toward consumption, investment in learning or health, trust in others, political opinions, technological innovation, etc. In fact, every behavior (and neural structure) for which time and risk are relevant dimensions is likely to involve a certain degree of plasticity.

Could phenotypic plasticity be relevant to explain cultural differences?

When we observe cultural differences between two societies, we can’t help but think that the difference has its roots in different cultural heritages (religious, legal, literary, etc.). This is because we have no alternative mechanism to explain such differences. Suggesting that the two groups differ in terms of psychological mechanisms seems to mean going back to the 19^th century "national character" studies or to mysterious "cultural mindset" that re-describe the phenomenon rather than explain it. By contrast, phenotypic plasticity offers a plausible (and not mutually incompatible) mechanism to explain why people have different mindsets in different societies, why they are more impulsive, why they trust others less or why they are afraid of innovations. Even better, it makes some predictions as to how the environment should impact people’s psychology. For instance, a common sense idea is that people should innovate when they are in danger and need an urgent solution. Evolutionary theory suggests the opposite: when resources are scarce and unpredictable, innovation is too risky: Better stick to what you know than jeopardize everything. And indeed, this is what people do.

Finally, phenotypic plasticity has the potential to solve several limitations of the standard culture-as-transmission-of-information paradigm: Why does the same cultural background (language, religion, ethnicity) give rise to radically different behaviors according to whether an individual was born in a low vs. a high social class or in an old vs. recent generation? Or how does an old and apparently robust cultural phenomenon crumble in a few generations, sometimes in a few years, and without any cultural external input? This may be because different environments triggered different "behavioral strategies" in people, transforming a common cultural heritage into diverging new cultures.

To sum up, phenotypic plasticity is key in the study of human behavior. It provides a framework to account for the fact that the same genome and the same neural architecture can give rise to cultural variability in humans.

A hundred years ago, in 1917, Albert Einstein had a problem. He had just come up with a beautiful new theory of gravity called general relativity. But the theory predicted that the universe should either expand or contract. Even for Einstein, this was a bridge too far. The universe, though it might have had a beginning, certainly did not appear to be changing.

General relativity is a rigid theory; little of it can be changed without destroying its elegant mathematical structure. The only wiggle room was a single quantity, which Einstein called the cosmological constant. If he assumed this quantity was zero, the equations looked particularly simple, but they required the universe to be dynamical.

The gravity of ordinary matter (such as galaxies) is attractive. Introducing a positive cosmological constant adds a repulsive counterforce. By setting the cosmological constant to a particular nonzero value at which the two tendencies cancel each other, Einstein thought he could get his theory to spit out a static, unchanging universe.

Einstein later called this idea his “biggest blunder,” an accurate assessment. First, his move doesn’t actually accomplish the task: Einstein’s static universe is unstable, like a pencil balanced on its tip. Because matter is distributed unevenly, the opposing forces couldn’t possibly be arranged to balance out everywhere. Individual regions would soon begin to expand or contract. Worse, Einstein missed out on a spectacular prediction. Had he believed his own equations, he could have anticipated the 1929 discovery that galaxies are in fact receding from one another.

This dramatic turn of events proved that the universe was not static. But an expanding universe didn’t imply that the cosmological constant was necessarily zero! As quantum field theory triumphed in the second half of the 20th century as a description of elementary particles, physicists recognized that the cosmological constant “wants to be there.” The very theories predicting with unprecedented accuracy the behavior of small particles also implied that empty space should have some weight, or “vacuum energy.” This kind of energy happens to be indistinguishable from a cosmological constant, as far as the equations of general relativity are concerned.

So it was no longer an option to set the cosmological constant to zero. Rather, it had to be calculated. Estimates indicated an enormous value—save for some unlikely precise cancellation between large positive and negative contributions from different particles. But a huge cosmological constant would show up as a repulsive force that would blow up the entire universe in a split second (or, alternatively, would cause it to collapse instantly, if the constant happened to come out negative). Evidently this was not what the universe was doing.

This drastic conflict between theory and observation is the “cosmological constant problem.” It remains the most serious problem in theoretical physics. It has contributed to the development of revolutionary ideas, particularly the multiverse and the “landscape” of string theory. 

The string landscape solves the cosmological constant problem by a strategy similar to throwing many darts randomly: Some will hit the bullseye by accident. In string theory there are many different ways of making empty space, and they would all get realized as vast regions in different parts of the universe. In some regions, the universe “hits the bullseye”—the cosmological constant is accidentally small. There, spacetime does not quickly explode or collapse, so structure and observers are more likely to evolve in these lucky regions.

A crucial prediction of this approach was pointed out by Steven Weinberg in 1987: If an accidental near-cancellation is the reason the cosmological constant is so small, then there’s no particular reason for it to be exactly zero. Rather, it should be large enough to have a just-noticeable effect.

In 1998, astronomers did find such an effect. By observing distant supernovae, we can tell that the universe is not just expanding but accelerating (expanding ever more rapidly). Other observations, such as the history of galaxy formation and the present rate of expansion, have since provided independent evidence for the same conclusion. Empty space is filled with vacuum energy. In other words, there’s a positive cosmological constant of a particular value, which we have now measured.

The cause of the acceleration is sometimes described more dramatically as a “mysterious dark energy.” But in science we shouldn’t embrace mystery where there is none. If it walks like a duck and quacks like a duck, we call it a duck. In this case, it accelerates the expansion and affects galaxy formation precisely like a cosmological constant, so we should call it by its name.

One of the most fascinating consequences of a positive cosmological constant is that we’ll never see much more than the present visible universe. In fact, billions of years from today the most distant galaxies will begin to disappear, accelerated out of sight too far for light from them to reach us. Eventually our local group of galaxies will hover alone in a vast emptiness filled with nothing but vacuum energy.

Fallibilism is the idea that we can never be 100% certain that we are right, and must therefore always be open to the possibility that we are wrong. This might seem a pessimistic notion, but it is not. Ironically, this apparent weakness is a strength; for admitting one’s mistakes is the first step to learning from them, and overcoming them, in science and society.

Fallibilism lies at the heart of the scientific enterprise. Even science’s most well-established findings—the "laws" of nature—are but hypotheses that have withstood scrutiny and testing thus far. The possibility that they may be wrong, or be superseded, is what spurs the generation of new alternative hypotheses, and the search for further evidence that enables us to choose between them. This is what scientific progress is made of. Science rightly champions winning ideas, ideas that have been tested and have passed, and dispenses with those that have been tested and have failed, or are too vague to be tested at all.

Fallibilism is also the guiding principle of free, open, liberal, secular societies. If the "laws" of nature can be wrong, then think how much more fallible are our social and political arrangements. Even our morals, for example, do not reflect some absolute truth—god-given or otherwise. They too are hypotheses—biological and cultural attempts to solve the problems of cooperation and conflict inherent in human social life. They are tentative, provisional, and capable of improvement; and they can be, and have been, improved upon. The awareness of this possibility—allied to the ambition to seize the opportunity it represents, and the scientific ability to do so—is precisely what has driven the tremendous social, moral, legal and political progress of the past few centuries.

Fallibilism—the notion that we may not be right—does not mean that we must be entirely wrong. So it is not a license to tear everything up and start again. We should respect tried and tested ideas and institutions, and recognize that our attempts to improve on them are equally fallible. Nor does fallibilism lead to "anything goes" relativism—the notion that there is no way to distinguish good ideas from bad. On the contrary, fallibilism tells us that our methods for distinguishing better ideas from worse work, and urges us to use them to quantify our uncertainty, and work to resolve it. Better still, to work together. Your opponent is no doubt mistaken, but in all likelihood so are you; so why not see what you can learn from each other, and collaborate to see further?

Anyone wishing to understand the world, or change it for the better, should embrace this fundamental truth.

Thanks to online shopping, buying things now is much easier than before. You do not like something—you can always return it back for a full refund. A physicist might say that one can turn the arrow of time back for non-used purchases. A cosmologist may comment that we use a similar time reversal method in our research.

Indeed, to understand the origin of the universe we may study its present evolution, and then solve the Einstein equations back in time. What we find is very similar to what happens then we play a movie back. At present, galaxies move away from each other. Playing the movie back shows them moving closer together. Going further back in time, we see that at some point the density of matter becomes infinitely large. This is the cosmological singularity.

Solving the same equations forward in time, starting from the singularity, one finds that all matter appears from the singularity in a huge explosion, called the Big Bang. When the original cosmic fire cools down, matter condenses into galaxies, and they fly away from each other.

The possibility to go back and forth in time is a powerful tool, which helps us to visualize the evolution of the universe. The resulting picture is so convincing that many of us are using it even now in our lectures. It tells us, essentially, that our universe is a gift sent to us from the cosmological online retailer 14 billion years ago. We can track its delivery from its origin to the present time, running our calculations all the way back to the Big Bang.

Of course, we know that one cannot turn the arrow of time back because of the second law of thermodynamics. But the standard lore of the Big Bang theory was that the expansion of the universe was nearly adiabatic, and therefore approximately reversible. In particular, it was assumed that the total number of elementary particles in the universe did not change much during the cosmological evolution.

Some parts of this picture, however, were problematic. One may wonder, for example, who paid the bill for sending us more than 10⁹⁰ elementary particles populating our universe, and who made the universe uniform and suitable for life?

In the beginning of the '80s, it was found that one can solve these problems if soon after the Big Bang there was a short stage of an exponentially fast expansion of the universe, called inflation. This idea experienced numerous modifications, and now we do not even think that the universe was born in fire. Instead, it could be created from a tiny vacuum-like speck of matter with special properties, weighting less than a gram and containing no elementary particles at all. Normal matter emerged only later, when the universe became exponentially large and its original vacuum-like state decayed.

During the last 35 years, many predictions of this theory have been verified by cosmological observations, but nevertheless it is somewhat difficult to get used to it, in part because it does not match the conventional picture of the time-reversible universe.

Indeed, let us try to follow the cosmological evolution back in time in this scenario. At first, galaxies move towards each other and the universe becomes very dense, just as in the standard Big Bang theory. But then something really weird happens. Suddenly, at the moment corresponding to the end of inflation, played back in time, all 10⁹⁰ elementary particles populating our part of the universe completely disappear!... And then—not a flash, not a sound, nothing, except the exponentially shrinking and disappearing empty universe.

Thus, in this scenario, the movie played back in time has a nonsensical ending. Making 10⁹⁰ elementary particles instantly vanish is absolutely impossible!

But there is nothing wrong about it, if one moves in the proper time direction. According to inflationary theory, all elementary particles in the universe were created by conversion of vacuum energy to matter during the decay of the original vacuum-like state. The theory of this process is well developed, but this process is irreversible: once particles are created, they cannot be un-created.

This is a part of the mechanism that makes the new theory work: We do not need anybody to send us a huge container with more than 10⁹⁰ elementary particles; a tiny package with one gram of matter is more than enough. On its way to us, this package started growing up, unwrapping itself, producing billions of galaxies containing hundreds of billions of stars. We cannot, even in our imagination, follow this package back to its origin and watch all matter dissolving into nothingness. Our universe is unreturnable, so our only choice is to accept this gift and use it to the best of our abilities.

Every day you play with the light of the universe. 
Subtle visitor, you arrive in the flower and the water… 

You might not think of humility as a scientific concept, but the special brand of humility that is enshrined in scientific culture is deserving of special recognition for its unique heuristic transformative power. 

Suppose we choose an American woman at random. Given just a couple of numbers that we know characterize American women's heights, we can be 95% certain (meaning, we will be wrong only once out of every 20 times we do this) that her height will be between 4 feet 10 inches and 5 feet 10 inches. It is the statistical concept of standard deviation which allows us to say this. To show how, let us take such a data set: the heights of 1,000 randomly chosen American women, and let us plot this data as points on a graph where the x-axis shows height from 0 to 100 inches, and the y-axis shows the number of women who are of that height (we use only whole numbers of inches). If we then connect all these points into a smooth line, we will get a curve that is bell-shaped. Some data sets that have this characteristic shape, including this one, are said to have what is called a normal (also known as Gaussian) distribution, and it is the way in which a great many kinds of data are distributed.

For data that is strictly normally distributed, the highest point on our graph (in this example, the height in inches which occurs most frequently, and which is called the mode) would be at the average (mean) height for American women, which happens to be 64 inches, but with real-world data the mean and the mode can differ slightly. As we move rightward from that peak to greater height along the x-axis, the curve will start sloping down, become steeper, and then gradually become less steep and peter out to zero as it hits the x-axis just after where the tallest woman (or women) in our sample happens to be. The same exact thing happens on the other side as we go to smaller heights. And this is how we get the familiar symmetrical bell-shape. Suppose that the height of Japanese women is also 64 inches on average but it varies less than that of American women because Japan is less ethnically and racially diverse than America. In this case, the bell-shaped curve will be thinner and higher and will fall to zero more quickly on either side. Standard deviation is a measure of how spread out the bell-curve of a normal distribution is. The more the data is spread out, the greater the standard deviation.

The standard deviation is half the distance from one side of the bell curve to the other (so it has the same units as the x-axis), where the curve is about 60% of its maximum height. And it can be shown that about 68% of our data points will fall within plus or minus one standard deviation around the mean value. So in our example of the heights of American women, if the standard deviation is 3 inches, then 68% of American women will have a height between 61 and 67 inches. Similarly, 95% of our data points will fall within two standard deviations around the mean, so in our case, 95% of women will have heights between 58 and 70 inches. And similarly, it can be calculated that 99.7% of our data points will fall within three standard deviations around the mean, and so on for even greater degrees of certainty.

The reason that standard deviation is so important in science is that random errors in measurement usually follow a normal distribution. And every measurement has some random error associated with it. For example, even with something as simple as just weighing a small object with a scale, if we weigh it 100 times, we may get many slightly different values. Suppose the mean of all our observations of its weight comes out to 1352 grams with a standard deviation of 5 grams. Then we can be 95% certain that the object's actual weight is between 1342 and 1362 grams (mean weight plus or minus two standard deviations). You may have heard reports before the discovery of the Higgs Boson at CERN in 2012 that they have a "3 sigma" result showing a new particle.  (The lower case Greek letter sigma is the conventional notation for standard deviation, hence it is often also just called "sigma.") The "3 sigma" meant that we can be 99.7% certain the signal is real and not a random error. Eventually a "5 sigma" result was announced for the Higgs particle on July 4th, 2012 at CERN, and that corresponds to a 1 in 3.5 million chance that what they detected was due to random error.

It is interesting that measurement error (or uncertainty in observations) is such a fundamental part of science now but it is only in the 19th century that scientists started incorporating this idea routinely in their measurements. The ancient Greeks, for example, while very sophisticated in some parts of their mathematical and conceptual apparatus, almost always reported observations with much greater precision than was actually warranted, and this often got amplified into major errors. A quick example of this is Aristarchus's impressive method of measuring the distance between the Earth and the Sun by measuring the angle between the line of sight to the Moon when it is exactly half full, and the line of sight to the Sun. (The line between the Earth and the Moon would then be at 90 degrees to the line between the Moon and the Sun, and along with the line from the Earth to the Sun, this would form a right triangle.) He measured this angle as 87 degrees which told him the distance from the Earth to the Sun is 20 times greater than the distance from the Earth to the Moon.

We are told that we are entitled to our own opinions but not our own facts. This leaves out the observation that the war for our minds and attention is now increasingly being waged over neither facts nor opinions but feelings.

In an era in which anyone can publish anything, the quest to control information has largely been lost by institutions, with a race on to weaponize empathy by understanding its basis in linguistics and tweaking the social media algorithms which now present our world to us accordingly. As the theory goes, it is not that we don’t have our own opinions so much as that we have too many contradictory ones, and it is generally our emotional state alone which determines on which ones we will predicate action or inaction.

Russell Conjugation (or “emotive conjugation”) is a presently obscure construction from linguistics, psychology and rhetoric which demonstrates how our rational minds are shielded from understanding the junior role factual information generally plays relative to empathy in our formation of opinions. I frequently suggest it as perhaps the most important idea with which almost no one seems to be familiar, as it showed me just how easily my opinions could be manipulated without any need to falsify facts. Historically, the idea is not new and seems to have been first defined by several examples given by Bertrand Russell in 1948 on the BBC without much follow up work, until it was later rediscovered in the internet age and developed into a near data-driven science by pollster Frank Luntz beginning in the early 1990s.

In order to understand the concept properly you have to appreciate that most words and phrases are actually defined not by a single dictionary description, but rather two distinct attributes:

I) The factual content of the word or phrase.
II) The emotional content of the construction.

Where words can be considered “synonyms” if they carry the same factual content (I) regardless of the emotional content (II). This however leads to the peculiar effect that the synonyms for a positive word like “whistle-blower” cannot be used in its place as they are almost universally negative (with “snitch,” “fink,” “tattletale” being representative examples). This is our first clue that something is wrong, or at least incomplete with our concept of synonym requiring an upgrade to distinguish words that may be content synonyms but emotional antonyms.

The basic principle of Russell Conjugation is that the human mind is constantly looking ahead well beyond what is true or false to ask “What is the social consequence of accepting the facts as they are?”  While this line of thinking is obviously self-serving, we are descended from social creatures who could not safely form opinions around pure facts so much as around how those facts are presented to us by those we ape, trust or fear. Thus, as listeners and readers our minds generally mirror the emotional state of the source, while in our roles as authoritative narrators presenting the facts, we maintain an arsenal of language to subliminally instruct our listeners and readers on how we expect them to color their perceptions. Russell discussed this by putting three such presentations of a common underlying fact in the form in which a verb is typically conjugated:

I am firm. [Positive empathy]
You are obstinate. [Neutral to mildly negative empathy]
He/She/It is pigheaded.  [Very negative empathy]

In all three cases, Russell was describing people who did not readily change their minds. Yet by putting these descriptions so close together and without further factual information to separate the individual cases, we were forced to confront the fact that most of us feel positively towards the steadfast narrator and negatively towards the pigheaded fool, all without any basis in fact.

Years later, the data-driven pollster Frank Luntz stumbled on much the same concept unaware of Russell’s earlier construction.  By holding focus-groups with new real time technology that let participants share emotional responses to changes in authoritative language, Luntz was lead to make a stunning discovery that pushed Russell’s construction out of the realm of linguistics and into the realm of applied psychology. What he found was extraordinary: many if not most people form their opinions based solely on whatever Russell conjugation is presented to them and not on the underlying facts. That is, the very same person will oppose a “death tax” while having supported an “estate tax” seconds earlier even though these taxes are two descriptions of the exact same underlying object. Further, such is the power of emotive conjugation that we are generally not even aware that we hold such contradictory opinions. Thus “Illegal aliens” and “undocumented immigrants” may be the same people, but the former label leads to calls for deportation while the latter one instantly causes many of us to consider amnesty programs and paths to citizenship.

If we accept that Russell Conjugation keeps us from even seeing that we do not hold consistent opinions on facts, we see a possible new answer to a puzzle that dates from the birth of the web: “If the internet democratized information, why has its social impact been so much slower than many of us expected?” Assuming that our actions are based not on what we know but upon how we feel about what we know, we see that traditional media has all but lost control of gate-keeping our information, but not yet how it is emotively shaded. In fact, it is relatively simple to write a computer program to crawl factually accurate news stories against a look-up table of Russell conjugates to see the exact bias of every supposedly objective story.

Thus the answer to the puzzle of our inaction it seems may be that we built an information superhighway for all, but neglected to build an empathy network alongside it to democratize what we feel. We currently get our information from more sources than ever before, but, at least until recently, we have turned to traditional institutions to guide our empathy. Information, as the saying goes, wants to be free. But we fear authentic emotions will get us into trouble with our social group, and so continue to look to others to tell us what is safe to feel.

In what ways are you related to the contents of your salad? Or to the ingredients of a slice of pepperoni pizza? Or whatever your next meal might be? Of course, consumption is an ecological relationship. Your body digests and absorbs the nutrients from your food, which provide energy for your metabolism and material components for your cells. But another fundamental kind of relationship is more cryptic, and in many ways more profound.  

Middle class Americans don’t live like kings—they live better than kings.   

At the heart of scientific thinking is the systematic evaluation of alternative possibilities. The idea is so foundational that it’s woven into the practice of science itself.  

The eradication of smallpox was one of the most significant achievements of modern medicine. It was possible due to an effective vaccine, coupled with global vaccination programs. Theoretically, it is possible to eradicate other diseases such as measles or polio in a similar manner; if enough of the global population could be vaccinated, then these diseases would cease to exist. We have come tantalizingly close to eradication in some cases: In 2000, the Center for Disease Control and Prevention declared that measles had been eliminated from the United States. Sixteen years later, the Pan American Health Organization announced that measles had been eradicated from the Americas. Polio is now endemic in only three countries in the world. Infectious diseases that routinely killed young children are now preventable thanks to childhood vaccination programs. Yet despite these milestones, there have been several outbreaks of vaccine-preventable diseases in recent times. How can this be? 

For societies to survive and thrive, some significant proportion of their members must engage in reciprocal altruism. All sorts of animals, including humans, will pay high individual costs to provide benefits for a non-intimate other. Indeed, this kind of altruism plays a critical role in producing cooperative cultures that improve a group’s welfare, survival, and fitness. 

Every human—not excepting scientists—bears the whole stamp of the human condition. This includes evolved neural programs specialized for navigating the world of coalitions—teams, not groups. (Although the concept of coalitional instincts has emerged over recent decades, there is no mutually-agreed-upon term for this concept yet.) These programs enable us and induce us to form, maintain, join, support, recognize, defend, defect from, factionalize, exploit, resist, subordinate, distrust, dislike, oppose, and attack coalitions. Coalitions are sets of individuals interpreted by their members and/or by others as sharing a common abstract identity (including propensities to act as a unit, to defend joint interests, and to have shared mental states and other properties of a single human agent, such as status and prerogatives).  

Why do we see the world this way? Most species do not and cannot. Even those that have linear hierarchies do not. Among elephant seals, for example, an alpha can reproductively exclude other males, even though beta and gamma are physically capable of beating alpha—if only they could cognitively coordinate. The fitness payoff is enormous for solving the thorny array of cognitive and motivational computational problems inherent in acting in groups: Two can beat one, three can beat two, and so on, propelling an arms race of numbers, effective mobilization, coordination, and cohesion.  

When Dirac formulated the postulates of quantum theory, he required Hermiticity to be the fundamental symmetry for his equations. For Dirac, the requirement of Hermiticity was the mathematical device that he needed to ensure that all predictions for the outcomes of real-world measurements of quantum systems resulted in a real number. This is important since we only observe real outcomes in actual experimental observations. Dirac’s choice of Hermiticity as the fundamental symmetry of quantum theory was not seriously challenged for around seventy years.  

No, homophily has nothing to do with sexual orientation. In the 1950s a pair of sociologists, Paul Lazarsfeld and Robert Merton, coined the term homophily to refer to the pervasive tendency of humans to associate with others who are similar to themselves.    

The concept of opportunity costs—the loss of potential gains from alternatives not chosen when a mutually exclusive choice must be made—is one of the most important concepts in the field of economics. But the concept is not well appreciated in the field of psychology.           

One reason for its absence is the sheer difficulty of calculating opportunity costs that occur in metrics other than money. Consider mate choice. Choosing one long-term mate means forgoing the benefits of choosing an available and interested alternative. But how are non-monetary benefits calculated psychologically?           

The complexities are multiple. The benefit-bestowing qualities of passed-over mates are many in number and disparate in nature. And there are inevitable tradeoffs among competing and incommensurate alternatives. Sometimes the choice is between a humorless mate with excellent future job prospects and a fun-loving mate destined for a low-status occupation; or between an attractive mate who carries the costs of incessant attention from others versus a mate who garners little external attention but with whom you have less sexual chemistry. Another intangible quality also factors into the equation—the degree to which competing alternatives appreciate your unique assets, which renders you more irreplaceably valuable to one than the other.           

Uncertainty of assessment surrounds each benefit-conferring quality. It is difficult to determine how emotionally stable someone is without sustained observation through times bad and good—events experienced with a chosen mate but unknown with a foregone alternative. Another complication centers on infidelity and breakups. There is no guarantee that you will receive the benefits of a chosen mate over the long run. Mates higher in desirability are more likely to defect. Whereas less desirable mates are sure bets, more desirable partners represent tempting gambles. How do these mating opportunity costs enter into the complex calculus of mating decisions?           

Despite the difficulties involved in computing non-monetary opportunity costs, probabilistic cues to their recurrent reality over evolutionary time must have forged a psychology designed to assess them, however approximate these computations may be. Although mating decisions provide clear illustrations, the psychology of opportunity costs is more pervasive. Humans surely have evolved a complex multifaceted psychology of opportunity costs, since every behavioral decision at every moment precludes potential benefits from alternative courses of action.

Many of these are trivial—sipping a cappuccino precludes downing a latte. But some are profound and produce post-decision regret, such as missed sexual opportunities or lamenting a true love that got away. The penalties of incorrectly calculating mating opportunity costs can last a lifetime.

The world is governed by scientific principles that are fairly well taught and understood. A concept that resonates throughout physical science is rheology, and yet there are so few rheologists in academia that an international symposium fills only a small conference room. Rheo, coming from the Greek to flow, is primarily the study of how non-Newtonian matter flows. In practice, most rheologists look at how nanoparticles behave in complex compounds, often filled with graphene, silica, or carbon nanotubes. Rheology was key to the creation of tires and other polymer systems, but as new technologies incorporate flexible and stretchable devices for electronics, medical implants, regenerative and haptic garments for virtual reality experiences, rheologists will need to be as common as chemists.

The primary reason for the relative obscurity of such an important topic is the hidden complexity involved. The number of connections in the human brain has been discussed by neuroscientists and often invites a sense of awe and inspiration to reduce and unify. Despite some well-known and simple equations, a complex nano-filled elastomer is also nearly impossible to model. Before it is possible to scale the dreams of today, experiments and theoretical models will need to converge in ways that they do not now. Through rheology, van der waal forces, semiconductivity, superconductivity, Quantum tunneling, and other such properties of composite materials can be harnessed and exploited, rather than hinder. Rheology is the old, newly relevant transdisciplinary science.

Included Middle is an idea proposed by Stéphane Lupasco (in The Principle of Antagonism and the Logic of Energy in 1951), further developed by Joseph E. Brenner and Basarab Nicolescu, and also supported by Werner Heisenberg. The notion pertains to physics and quantum mechanics, and may have wider application in other domains such as information theory and computing, epistemology, and theories of consciousness. The Included Middle is a theory proposing that logic has a three-part structure. The three parts are the positions of asserting something, the negation of this assertion, and a third position that is neither or both. Lupasco labeled these states A, not-A, and T. The Included Middle stands in opposition to classical logic stemming from Aristotle. In classical logic, the Principle of Non-contradiction specifically proposes an Excluded Middle, that no middle position exists, tertium non datur (there is no third option). In traditional logic, for any proposition, either that proposition is true, or its negation is true (there is either A or not-A). While this could be true for circumscribed domains that contain only A and not-A, there may also be a larger position not captured by these two claims, and that is articulated by the Included Middle.

Heisenberg noticed that there are cases where the straightforward classical logic of A and not-A does not hold. He pointed out how the traditional law of Excluded Middle has to be modified in Quantum Mechanics. In general cases at the macro scale, the law of Excluded Middle would seem to hold. Either there is a table here, or there is not a table here. There is no third position. But in the Quantum Mechanical realm, there are the ideas of superposition and possibility, where both states could be true. Consider Schrödinger’s cat being possibly either dead or alive, until an observer checks and possibility collapses into a reality state. Thus a term of logic is needed to describe this third possible situation, hence the Included Middle. It is not “middle” in the sense of being between A and not-A, that there is a partial table here, but rather in the sense that there is a third position, another state of reality, that contains both A and not-A. This can be conceptualized by appealing to levels of reality. A and not-A exist at one level of reality, and the third position at another. At the level of A and not-A, there are only the two contradictory possibilities. At a higher level of reality, however, there is a larger domain, where both elements could be possible; both elements are members of a larger set of possibilities.

Included Middle is a concept already deployed in a variety of scientific domains and could benefit from a wider application in being promoted to “meme” status. This is because beyond its uses in science, Included Middle is a model for thinking. The Included Middle is a conceptual model that overcomes dualism and opens a frame that is complex and multi-dimensional, not merely one of binary elements and simple linear causality. We have now come to comprehend and address our world as one that is complex as opposed to basic, and formal tools that support this investigation are crucial. The Included Middle helps to expose how our thinking process unfolds. When attempting to grasp anything new, a basic “A, not-A” logic could be the first step in understanding the situation. However, the idea is then to progress to the next step which is another level of thinking that holds both A and not-A. The Included Middle is a more robust model that has properties of both determinacy and indeterminacy, the universal and the particular, the part and the whole, and actuality and possibility. The Included Middle is a position of greater complexity and possibility for addressing any situation. Conceiving of a third space that holds two apparent contradictions of a problem is what the Included Middle might bring to contemporary challenges in consciousness, artificial intelligence, disease pathologies, and unified theories in physics and cosmology.

A few months ago I sequenced my genes. It is a 700 Megabyte text file that looks something like this:

AGCCCCTCAGGAGTCCGGCCACATGGAAACTCCTCATTCCGGAGGTCAGTCAGATTTACCCTTGAGTTCAAACTTCAGGGTCCAGAGGCTGATAATCTACTTACCCAAACATAGGGCTCACCTTGGCGTCGCGTCCGGCGGCAAACTAAGAACACGTCGTCTAAATGACTTCTTAAAGTAGAATAGCGTGTTCTCTCCTTCCAGCCTCCGAAAAACTCGGACCAAAGATCAGGCTTGTCCGTTCTTCGCTAGTGATGAGACTGCGCCTCTGTTCGTACAACCAATTTAGG

Each individual A, C, G and T are organic molecules that form the building blocks of what makes me "me": my DNA. It is approximately 3.3 billion pairs of nucleotides organized in around 24 thousand genes. 

The information of every living being is codified in this manner. Our shape, our capacities, abilities, needs and even predisposition to disease are determined largely by our genes. 

But this information is only a small percentage (less than 2%) of what can be found in the DNA that each of the cells of my body carry. That is the percentage of the DNA that encodes proteins, the molecules that carry out all the functions that are necessary for life. The other 98% is known as non-coding DNA and as of 2016 we believe only an extra 10% to 15% have a biological function that show complex patterns of expression and regulation while the rest is still largely referred to as "junk DNA". This does not necessarily mean that the majority of our DNA is junk, it only means that we still do not know why it is there nor what it does. The human genome still has many tricks up its sleeves.

The story of how our DNA is expressed and regulated is the story of the transcriptome, and despite all our technological advances, its study is still in its early stages but already showing enormous potential to better diagnose, treat and cure disease.

In order for our DNA to be expressed and produce a specific protein, the code must be "copied" (transcribed) into RNA. These gene readouts are called transcripts, and the transcriptome is the collection of all the RNA molecules, or transcripts, present in a cell. 

In contrast with the genome, which is characterized by its stability, the transcriptome is constantly changing and can reflect in real time, at the molecular level, the physiology of a person depending on many factors, including stage of development and environmental conditions.

The transcriptome can tell us when and where each gene is turned on or off in the cells of tissues and organs of an individual. It functions like a dimmer switch, setting whether a gene is 10% active, or 70% active, and therefore enabling a much more intricate fine-tuning of gene expression. By comparing the transcriptome of different types of cells we can understand what makes a specific cell from a specific organ unique, how does that cell look when working normally and healthy and how its gene activity may reflect or contribute to certain diseases.

The transcriptome may hold the key to the breakthrough we have been waiting for over the last 30 years in gene therapy. There are today two complementary yet different approaches: the replacement or editing of genes within the genome (such as the widely known CRISPR-Cas9 technique) and the inhibition or enhancement of gene expression.

On the latter approach, RNA based cancer vaccines that activate an individual’s innate immune system are already in clinical trials with promising results in diseases such as lung or prostate cancer. The vaccination with RNA molecules is a promising and safe approach to let the patient’s body produce its own vaccines. By introducing a specific synthetic RNA, the protein synthesis can be controlled without intervening in the human genome and by letting the cell's own protein building machinery work without altering the physiological state of the cell.

This concept will unlock a path to a prosperous future in terms of aging prevention, brain functioning and stem cell health, as well as the eradication of cancer, hepatitis B, HIV or even high cholesterol. The transcriptome has opened our eyes to the mind-staggering complexity of the cell and when fully fathomed, it will finally enable us to truly start conquering our genetic destiny. 

John McCarthy, the late co-founder of the field of artificial intelligence, wrote, "He who refuses to do arithmetic is doomed to talk nonsense." It seemed incongruous that a professor who worked with esoteric high level math would be touting simple arithmetic, but he was right; in fact in many cases all we need to avoid nonsense is the simplest form of arithmetic: counting.  

In 2008, the US government approved a $700 billion bank bailout package. A search for the phrase "$700 million bailout" reveals hundreds of writers who were eager to debate whether this was a prudent or rash, but who couldn't distinguish the difference between $2 per citizen and $2,000 per citizen. Knowing the difference is crucial to understanding the efficacy of the deal, and is just a matter of counting.

Consider the case of a patient who undergoes a routine medical screening and tests positive for a disease that affects 1% of the population. The screening test is known to be 90% accurate. What is the chance that the patient actually has the disease? When a group of trained physicians were asked, their average answer was 75%. They reasoned that it should be somewhat less than the 90% accuracy of the test, because the disease is rare. But if they had bothered to count, they would have reasoned like this: on average, out of every 100 people, 1 will have the disease, and 99 won't. The 1 will have a positive test result, and so will about 10 of the 99 (because the test is 10% inaccurate). So we have a pool of 11 people who test positive, of which only 1 actually has the disease, so the chance is about 1/11 or 9%. That means that highly trained physicians are reasoning very poorly, scaring patients with an estimate that is much too high, all because they didn't bother to count. The physicians might say they are trained to do medicine, not probability theory, but as Pierre Laplace said in 1812, "Probability theory is nothing but common sense reduced to calculation," and the basis of probability theory is just counting: "The probability of an event is the ratio of the number of cases favorable to it, to the number of all cases possible."

In their report on climate change, the IPCC stated the consensus view that "most of the observed warming over the last 50 years is likely to have been due to the increase in greenhouse gas concentrations." Yet some criticized this consensus, saying scientists are still uncertain. Who's right? Naomi Oreskes took it upon herself to resolve the question, by counting. She searched a scientific database using the keywords "climate change" and scrutinized the 928 abstracts that matched. She found that 25% of the articles did not address the consensus (because, for example, they were about regional climate rather than global), but that none of the 928 rejected the consensus. This is a powerful form of counting. What's more, I don't need to take Prof. Oreskes's word for it: I did my own experiment, sampling 25 abstracts (I didn't have the patience to do 928) and I too found that none of them rejected the consensus.

This is a powerful tool. When faced with a complex issue, you have the capability to resolve the question, not by examining your political predispositions and arguing for whatever agrees with them, but by examining the evidence and counting the number of cases that are favorable, and comparing to those that aren't. You don't need to be a mathematical wizard; just apply what you learned as a toddler: count!

Thought, passion, love... this internal world we experience, including all the meaning and purpose in our lives, arises naturally from the interactions of elementary particles. This sounds absurd, but it is so. Scientists have attained a full understanding of all fundamental interactions that can happen at scales ranging from subatomic particles to the size of our solar system. There is magic in our world, but it is not from external forces that act on us or through us. Our fates are not guided by mystical energies or the motions of the planets against the stars. We know better now. We know that the magic of life comes from emergence.

It is the unimaginably large numbers of interactions that make this magic possible. To describe romantic love as the timely mutual squirt of oxytocin trivializes the concerted dance of more molecules than there are stars in the observable universe. The numbers are beyond astronomical. There are approximately 100 trillion atoms in each human cell, and about 100 trillion cells in each human. And the number of possible interactions rises exponentially with the number of atoms. It is the emergent qualities of this vast cosmos of interacting entities that make us us. In principle, it would be possible to use a sufficiently powerful computer to simulate the interactions of this myriad of atoms and reproduce all our perceptions, experiences, and emotions. But to simulate something does not mean you understand the thing—it only means you understand a thing’s parts and their interactions well enough to simulate it. This is the triumph and tragedy of our most ancient and powerful method of science: analysis—understanding a thing as the sum of its parts and their actions. We have learned and benefitted from this method, but we have also learned its limits. When the number of parts becomes huge, such as for atoms making up a human, analysis is practically useless for understanding the system—even though the system does emerge from its parts and their interactions. We can more effectively understand an entity using principles deduced from experiments at or near its own level of distance scale—its own stratum.

The emergent strata of the world are roughly recapitulated by the hierarchy of our major scientific subjects. Atomic physics emerges from particle physics and quantum field theory, chemistry emerges from atomic physics, biochemistry from chemistry, biology from biochemistry, neuroscience from biology, cognitive science from neuroscience, psychology from cognitive science, sociology from psychology, economics from sociology, and so on. This hierarchical sequence of strata, from low to high, is not exact or linear—other fields, such as computer science and environmental science, branch in and out depending on their relevance, and mathematics and the constraints of physics apply throughout. But the general pattern of emergence in a sequence is clear: at each higher level, new behavior and properties appear which are not obvious from the interactions of the constituent entities in the level below, but do arise from them. The chemical properties of collections of molecules, such as acidity, can be described and modeled, inefficiently, using particle physics (two levels below), but it is much more practical to describe chemistry, including acidity, using principles derived within its own contextual level, and perhaps one level down, with principles of atomic physics. One would almost never think about acidity in terms of particle physics, because it is too far removed. And emergence is not just the converse of reduction. With each climb up the ladder of emergence to a higher level in the hierarchy, it is the cumulative side-effects of interactions of large numbers of constituents that result in qualitatively new properties that are best understood within the context of the new level.

Every step up the ladder to a new stratum is usually associated with an increase in complexity. And the complexities compound. Thermodynamically, this compounding of complexity—and activity at a higher level—requires a readily available source of energy to drive it, and a place to dump the resulting heat. If the energy source disappears, or if the heat cannot be expelled, complexity necessarily decays into entropy. Within a viable environment, at every high level of emergence, complexity and behavior is shaped by evolution through natural selection. For example, human goals, meaning, and purposes exist as emergent aspects in psychology favored by natural selection. The ladder of emergence precludes the necessity for any supernatural influence in our world; natural emergence is all it takes to create all the magic of life from building blocks of simple inanimate matter. Once we think we understand things at a high level in the hierarchy of emergence, we often ignore the ladder we used to get there from much lower levels. But we should never forget the ladder is there—that we and everything in our inner and outer world are emergent structures arising in many strata from a comprehensible scientific foundation. And we also should not forget an important question this raises: is there an ultimate fundamental level of this hierarchy, and are we close to knowing it, or is it emergence all the way down?

The great promise of the Internet was that more information would automatically yield better decisions. The great disappointment is that more information actually yields more possibilities to confirm what you already believed anyway. 

I.J. Good, a collaborator of Alan Turing’s during World War II, once wrote that a machine more intelligent than humans would be the “last invention” we make. Many believe that our society is about to reach this point, as smart machines increasingly invade our daily lives and imaginations. This invasion is proceeding along two fronts. Such machines are increasingly to blame for economic change, leading some people to question the necessity of work altogether. But, more deeply, we wonder whether such ultraintelligent machines will be “like us”: they may think, but will they feel?

To help anchor these latter kinds of questions, Stevan Harnad formulated the symbol grounding problem. For a machine, it’s problematic to learn what the symbols that make up human languages mean by simply transforming heard words into spoken words. It’s like trying to learn Egyptian hieroglyphs using a dictionary in which each hieroglyph is defined using other hieroglyphs. Consider this exchange I just had with the most high-powered chatbot on the web:

Me:   Hi there.
It:               Hello, how are you?
Me:   Not too bad. You?
It:               Fine thank you.
Me:   I've been thinking lately.
It:               Thinking about what?
Me:   The symbol grounding problem.
It:               Can you tell me a happy story?

The solution to the symbol grounding problem is to ground symbols in the rich soil of physical experience. For example, a robot may learn that the symbol “jump” correlates with sudden loss of pressure of the soles of its feet. This robot may begin to empathize with humans by asking one of them to jump, observing her feet leaving the ground, and inferring that she feels the same things that it feels when it jumps. Finally, such a machine may recursively ground increasingly abstract symbols in these already-grounded symbols, until eventually it understands the English idiom “don’t jump to conclusions” the first time it hears it.

The public should care about this seemingly esoteric corner of cognitive science because machines that do not ground symbols in this way will be dangerous. Consider my exchange with the chatbot. Not only did it not know that its final comment was “wrong,” but it failed to predict that I would be frustrated or amused by the comment. Similarly, another machine may fail to predict my terrified response to its actions.

Current machines can now, after receiving a million photographs containing a human and another million that do not, tell you whether or not a new photograph contains a human, without having to ground symbols in experience. But consider another data set, composed of two million conversations: In the first million, the speakers are discussing how best to help Bob; in the second million, they are conspiring to harm him. Current state-of-the-art machines cannot tell you whether the speakers in a new conversation intend to help or harm Bob.

Most humans can listen to a conversation and predict whether the person being discussed is in danger. It may be that we can do so because we have heard enough discussions in real life, books, and movies to be able to generalize to the current conversation, not unlike computers that recognize humans in previously-unseen photographs. However, we can also empathize by connecting words, images, and physical experience: We can put ourselves in the shoes of the people talking about Bob, or into the shoes of Bob himself. If one speaker says “one good turn deserves another” and follows it with a sarcastic sneer, we can take those verbal symbols (“one,” “good,” …), combine them with the visual cue, and do some mental simulation.

First, we can go back in time to inhabit Bob’s body mentally, and imagine him/us acting in a way that lessens the speaker’s hunger or assuages another of her physical or emotional pains. We can then return to the present as ourselves and imagine saying what she said. We will not follow up the statement with a sneer, as she did. Our prediction has failed.

So, our brain will return to the past, inhabit Bob’s body again, but this time mentally simulate hurting the speaker in some way. During the act we transfer into the speaker’s body and suffer her pain. Back in the present, we would imagine ourselves saying the same words. Also, feelings of anticipated revenge would be bubbling up inside us, bringing a sneer to our lips, thus matching the speaker’s sneer. So: we predict that the speakers wish to harm Bob.

Growing evidence from neuroscience indicates that heard words light up most parts of the brain, not just some localized language module. Could this indicate a person twisting words, actions, their own former felt experiences, and mental body snatching into sensory/action/experiential braided cables? Might these cables support a bridge from the actions and feelings of others to our own actions and feelings, and back again?

These machines may be useful and even empathetic. But would they be conscious? Consciousness is currently beyond the reach of science, but one can wonder. If I “feel” your pain, the subject and the object are clear: I am the subject and you are the object. But if I feel the pain of my own stubbed toe, the subject and object are not as obvious. Or are they? If two humans can connect by empathizing with each other, cannot two parts of my brain empathize with each other when I hurt myself? Perhaps feelings are verbs instead of nouns: they may be specific exchanges between cell clusters. May consciousness then not simply be a fractal arrangement of ever smaller sensory/motor/experiential braids grounding the ones above them? If myths tell us that the Earth is flat and rests on the back of a giant turtle, we might ask what holds up the turtle. The answer, of course, is that it’s turtles all the way down. Perhaps consciousness is simply empathy between cell clusters, all the way down.

Over the 7 million years since humans and chimpanzees shared a common ancestor, the inventory of human tools has gone from a handful of stone implements to a technological repertoire capable of replicating DNA, splitting atoms, and interplanetary travel. In the same evolutionary timespan, the chimpanzee toolkit has remained relatively rudimentary. It was "tool innovation"—constructing new tools or using old tools in new ways—that proved crucial in driving increasing technological complexity over the course of human history.

How can we explain this wide divergence in technological complexity between such closely related primate species? One possibility is that humans are unique among primate species in our capacity to innovate. If so, we might expect that innovation would be early-developing like walking or language acquisition. And yet there is little evidence for precocious innovation in early childhood. Although young children are inquisitive and keen to explore the world around them, they are astonishingly poor at solitary tool innovation. New Caledonian crows and great apes outperform young children in tool innovation tasks. This is particularly striking given the dazzling technological and social innovations associated with human culture. How does a species with offspring so bad at innovation become so good at it? 

Technological complexity is the outcome of our species’ remarkable capacity for cumulative culture; innovations build on each other and are progressively incorporated into a population's stock of skills and knowledge, generating ever more sophisticated repertoires. Innovation is necessary to ensure cultural and individual adaptation to novel and changing challenges, as humans spread to every corner of the planet. Cultural evolution makes individuals more innovative by allowing for the accumulation of prefabricated solutions to problems that can be recombined to create new technologies. The subcomponents of technology are typically too complex for individuals to develop from scratch. The cultural inheritance of the technologies of previous generations allows for the explosive growth of cultural complexity.

Children are cultural novices. Much of their time is spent trying to become like those around them—to do what they do, speak like they do, play and reason like they do. Their motivation to learn from, and imitate, others allows children to benefit from and build upon cumulative cultural transmission. Cumulative culture requires the high fidelity transmission of two qualitatively different abilities—instrumental skills (e.g., how to keep warm during winter) and social conventions (e.g., how to perform a ceremonial dance). Children acquire these skills through high fidelity imitation and behavioral conformity. These abilities afford the rapid acquisition of behavior more complex than could ever otherwise be learned exclusively through individual discovery or trial-and-error learning.

Children often copy when uncertain. This proclivity is enormously useful given that a vast amount of behavior that we engage in is opaque from the perspective of physical causality. High-fidelity imitation is an adaptive human strategy facilitating more rapid social learning of instrumental skills than would be possible if copying required a full causal representation of an event. So adaptive, in fact, that it is often employed at the expense of efficiency, as seen when kids "over-imitate" behavior that is not causally relevant to accomplishing a particular task.

The unique demands of acquiring instrumental skills and social conventions such as rituals provide insight into when children imitate, when they innovate, and to what degree. Instrumental behavior is outcome-oriented. Innovation often improves the efficiency of solving defined problems. When learning instrumental skills, with an increase in experience, high-fidelity imitation decreases. In contrast, conventional behavior is process-oriented. The goals are affiliation and group inclusion. When learning social conventions, imitative fidelity stays high, regardless of experience, and innovation stays low. Indeed, innovation impedes learning well-prescribed social conventions. Imitation and innovation work in tandem, deployed at different times for different purposes, to support learning group-specific skills and practices. The distinct goals of instrumental skills and social conventions drive cumulative culture and provide insight into human cognitive architecture.

Cumulative culture affords collective insights of previous generations to be harnessed for future discoveries in ways that are more powerful than the solitary brainpower of even the most intelligent individuals. Our capacity to build upon the innovations of others within and across generations drives our technological success. The capacity for cumulative culture has set our genus Homo on an evolutionary pathway remarkably distinct from the one traversed by all other species.

If I offered to give you $20 today or $100 in a year, which would you choose? It’s a pretty straightforward question, and, truth be told, one that has a logical answer. Unless you need that $20 to ensure your near-term survival, why not wait for the bigger prize if it were sure to come. After all, when was the last time that any reputable financial institution offered an investment vehicle guaranteed to quintuple your money in 365 days? Yet, if you pose this question to the average person, you’ll be surprised to find that most will opt to take the $20 and run. Why? To understand that, and the implications it holds for decisions in many domains of life, we first have to put the framework of the decision in context.

This type of decision—one where the consequences of choices change over time—is known as an “intertemporal choice.” It’s a type of dilemma well-studied by economists and psychologists, who often facepalm at the seemingly irrational decisions people make when it comes to investing for the future, but it’s a less familiar one, at least in name, to many outside those fields. The “irrational” part of intertemporal decisions derives from the fact that humans tend to discount the value of future rewards excessively, making it difficult to part with money that could offer pleasure in the moment in order to allow it to grow and, thereby, secure greater satisfaction and prosperity in the future.

Troubling as this situation might be for your 401(k), it’s essential to recognize that the origin of intertemporal choices, and as a result, the domains to which this framework can profitably be applied, aren’t limited to financial ones. In truth, much of human social life—our morality, our relationships—revolves around challenges posed by intertemporal choice. Do I pay back a favor I owe? If not, I’m certainly ahead in the moment, but over time I’ll likely lose any future opportunities for cooperation, not only with the person I left hanging but also with any others who learn of my reputation. Should you cheat on your spouse? Although it might lead to pleasure in the short term, the long-term losses to satisfaction, assuming your marriage was a good one, are likely to be devastating. Should you spend long hours to hone a skill that would make you valuable to your team or group rather than spending a summer’s day enjoying the weather? Here again, it’s the sacrifice of pleasure in the short term than can pave the way for greater success in the long one.

It’s challenges like these—ones involving cooperation, honesty, loyalty, perseverance, and the like—that were the original intertemporal dilemmas our ancestors faced. If they weren’t willing to accept smaller benefits in the short term by being less selfish, they weren’t going to have many friends or partners with whom to cooperate and sustain themselves in the long one. To thrive, they needed to demonstrate good character, and that meant they needed self-control to put off immediate gratification.

Today, when we think about self-control, we think about marshmallows. But what is the marshmallow test really? In point of fact, it’s just a child-friendly version of a dilemma of intertemporal choice: one sweet now, or two later? And as Walter Mischel’s work showed, an ability to have patience–to solve an intertemporal choice by being future-oriented–predicted success in realms ranging from investing, to academics, to health, to social life. In short, being able to delay gratification is a marker of character. People who can will be more loyal, more generous, more diligent, and more fair. In truth, it’s because dilemmas of intertemporal choice underlie social living that they can so easily be applied to economics, not the other way around. After all, self-control didn’t evolve to help us manage economic capital; it came about to help us manage social capital.

Recognition of this fact offers two important benefits. First, it provides a framework with which to unify the study of many types of decisions. For example, the dynamics and, as a consequence, the psychological mechanisms that underlie cheating and compassion will overlap with those that underlie saving and investing. After all, sacrificing time or energy to help another will build long-term capital just as does saving money for retirement. What’s more, this decision framework is a scalable one. For example, the dilemmas posed by climate change, overfishing, and related problems of sustainability are nothing if not intertemporal at base. Solving them requires a collective willingness to forgo immediate profits (or to pay higher prices) in the short term to reap larger, communal gains in the long term.

The second benefit that comes from recognizing the broad reach of intertemporal choice is the expansion of the tool set that can be used to solve its associated dilemmas. While economists and self-control researchers traditionally emphasize using reason, willpower, and the like to overcome our inherent impatience for pleasure, realizing the intertemporal nature posed by many moral dilemmas suggests an alternate route: the moral emotions. Gratitude makes us repay our debts. Compassion makes us willing to help others. Guilt prevents us from acting in selfish ways. These moral emotions—ones intrinsically linked to social living—lead people, directly or indirectly, to value the future. They enhance our character, which when translated to behavior, means they help us to share, to persevere, to be patient, and to be diligent.

So as the new year dawns, remember that most resolutions people will make for the next 365 days and beyond will have an intertemporal aspect. Whether it’s to save more, to eat less, to be kind, or to reduce a carbon footprint, it will likely require some forbearance. And helping people to keep that forbearance going will necessitate all of us—scientists and nonscientists alike—to continue exploring the mind’s inclination for selfish, short-term temptations and its many mechanisms to overcome them in a multidisciplinary manner.

Leibniz was famously satirized by Voltaire, in his play Candide, as saying that ours is the best of all possible worlds. While that played well on the stage, what Leibniz actually wrote, in 1714, in his Monodology, was a good deal more interesting. He did argue that God chose the one real world from an infinitude of possible worlds, by requiring it to have “as much perfection as possible.” But what is often missed is how he characterized degrees of perfection. Leibniz defined a world with “as much perfection as possible” to be one having “as much variety as possible, but with the greatest order possible.”

I believe that Leibniz’s insight of a world that optimizes variety, subject to “the greatest order possible”, is a powerful concept that could be helpful for current work in biology, computer science, neuroscience, physics and numerous other domains including social and political theory and urban planning.

To explain why, I have to define variety. I believe we ought to see variety as a measure of complexity which applies to systems of relationships. These are systems of individual units, which each have a unique set of interactions or relationships with the other units in the system. Leibniz saw the universe as just such a system of relationships. In a Leibnizian world, an object’s properties are not intrinsic to it-rather they reflect the relationships or interactions that object has with other objects.

Systems of relationships are often visualized as graphs or networks. Each element is represented by a node and two nodes are related when they are connected by a line. We know of a great many systems, natural and artificial, that can be represented by such a network. These include ecosystems, economies, the internet, social networks etc.

What does it mean for such a relational world to have a high variety. I would argue that variety is a measure of how unique is each element’s role in the network. In a Liebnizian world, each element has a view of the rest which summarizes how it is related to the other elements. An element’s view tells us what the whole system looks like from its point of view. Variety is a measure of how distinct these different views are.

Leibniz expressed this almost poetically.

"This interconnection (or accommodation) of all created things to each other, brings it about that each simple substance has relations that express all the others, and consequently, that each simple substance is a perpetual, living mirror of the universe.”

He then reaches for a striking metaphor, of a city.

“Just as the same city viewed from different directions, appears entirely different and, as it were, multiplied perspectively, in just the same way it happens that, because of the infinite multitude of simple substances, there are, as it were, just as many different universes, which are, nevertheless, only perspectives on a single one,”

One can almost hear Jane Jacobs in this, when she praises a good city as one with many eyes on the street.

A system of relations can, I believe, be said to have its maximal variety when the different views are maximally distinct from each other.

A city has low variety if the views from many of the houses are similar. A city has high variety if it is easy to tell, just by looking out the window, which street you are on.

An ecosystem is a system of relations, such as who eats who. A niche is a situation characterized by what you eat and who eats you. An economy is a system of relations including who buys what from who. The variety of an ecosystem is a measure of the extent which each species has a unique niche. The variety of an economy measures the uniqueness of each firm’s role in the market.

It is commonly asserted that ecosystems and economies evolve to higher degrees of complexity. But to develop these ideas we need a precise notion of complexity. Negative entropy does not suffice to measure complexity because the network of chemical reactions in our bodies and a regular lattice both have low entropy. We want a measure of complexity that recognizes that chemical reaction networks are far more complex than either random graphs or regular lattices.

I would suggest that variety is a very helpful notion of complexity, because it distinguishes the truly complex from the regular. In a high variety network it is easy to know where you are from looking around at your neighbourhood. In other words, the less information you need about the neighbourhoods to distinguish each node from the rest, the higher the variety. This captures a notion of complexity distinct from and, perhaps, more useful than, negative entropy.

Having defined variety, we can go back and try to imagine what Leibniz meant by maximizing variety but “with the greatest order possible.” Order can mean subject to law. Can there then be a law of maximal variety?

Such a law might be emergent in a complex system such as an economy or ecology. This might arise as follows: when the variety is maximal, the network is most efficient because there is a maximum amount of cooperation and a minimum of redundancy. Entities compete not to dominate a single niche, but to invent new ways to cooperate by inventing new niches, which have a novel interrelation to the rest.

George Boole, the son of a shoemaker, left school at sixteen, and ended up a Professor of Mathematics at Queens College in Cork in Ireland. As the only breadwinner in his family, he became a teacher at age sixteen, opened his own school in Lincoln in England by age nineteen, and fifteen years later was a mathematician and philosopher of logic. Why did Google celebrate his 200^th birthday on the 2^nd November 2015? And why should we remember his contributions?

In 1854 he wrote a book called An Investigation of the Laws of Thought which distilled the essence of logical thought down to terms like AND (e.g., x AND y), OR (e.g., x OR y), and NOT (e.g., NOT x), or combinations of these. This took Aristotelian logic (the syllogism, expressed in words) and insisted these be formulated as equations, which was a revolutionary step. Equally important, Boolean logic is today seen as the foundations of the "information age," or what we also call the "computer age." This is because each "value" in these logical statements or equations reduces down to either being true or false, with zero ambiguity. The logic is binary. No wonder they could be applied, more than a century later, in the design of electronic circuits in computers.

For me, the importance of Boolean logic even goes beyond its contribution to the logic of computers. Each of his terms is an "operation," like the familiar mathematical ones of addition and multiplication. These terms describe what happens if you take something as input, and perform an operation on it. You end up with an output. This is at the core of what I call "systemizing": take input, perform an operation, and observe the output. Boole, without realizing it, was describing how the human mind systemizes. In this way, he anticipated how to describe a uniquely human aspect of cognition, one that enables humans to do engineering (designing a system) and to innovate (changing a system).

Consider a much discussed example: Vinton Cerf, co-inventor of the Internet, was pouring peppercorns into a funnel and found if he dropped handfuls of peppercorns into it, the funnel got blocked or congested. Nothing came out. But if he poured the peppercorns in one a time, they didn’t get stuck, they flowed out smoothly. In system 1, the input is a handful of peppercorns, the operation is pouring them into a funnel, and the output is disappointingly nothing! In system 2, the input is one-peppercorn-at-a-time, the operation remains the same, but now the output is a pleasing flow of peppercorns. There are lessons here for how we as humans design not just pepper grinders, but also traffic systems (that either cause or avoid congestion), or how we design the post office, to cope with a volume of letters. Indeed, Boolean logic allows us to describe not just how an engineer designs any system, but how we humans systemize.

Here’s just one more legacy of Boolean logic. We know that people with autism have very logical minds, and have a strong drive to systemize. If we can generalize, they have a preference for information that can be systemized (such as factual information, or repeating, lawful patterns of information that doesn’t change unlawfully). They don’t cope well with information that is hard to systemize, because it contains ambiguity, or because it changes unexpectedly (such as social interaction, where what people do or say is rarely the same, except in highly ritualized contexts). Our modern understanding of autism as a hyper-systemizing mind owes a huge debt to Boolean logic that enables us to characterize the beauty and the extraordinary power of binary thinking, and also where such thinking is best used.

I’m voting for this concept, one that is so central to the scientific process, so much a given, that hardly any scientist ever actually speaks those words.

Scientists present their work—say, “We manipulated Variable X, and observed that this caused Z to happen,” or “We measured this and found that it takes Z amount of time to happen.” And when they do, most of the time what they’re actually saying is, “We manipulated Variable X, and observed that, on the average, this caused Z to happen.” “We measured this and found that, on the average, it takes Z amount of time to happen.” Everyone knows this.

Of course. Everyone in a population doesn’t have the exact same levels of something or other in their bloodstream. A causes B to happen most of the time, but not every single time. There’s variability.

When scientists present their data, they typically display the average—the mean, the X on a graph, the bar of a particular height in the figure. And it always comes with an “error term—a measure of how much variability there was around that average, a measure of how much confidence there is in saying “on the average.” Measure something or other in three people, observe values of 99, 100, and 101, producing an average of 100; measure that something in three people and observe values of 50, 100, and 150, producing an average of 100. This are two very different circumstance; “the average was 100,” tells you a lot more about how some sliver of the universe works in the first case than the second.

This is how scientists in most disciplines go about their business, with the recognition that you’re always seeing how things work on the average. So why is it important that this be more widely known? I can think of at least three reasons, of increasing importance.

First, this should constitute a big shout-out to scientists and the scientific process. Perhaps counter to the general perception, scientists don’t pronounce upon some fact that they have discovered; they pronounce upon the temporary way station of statistical confidence that they have discovered. “On the average” means that there’s stuff you can’t explain yet, and there’s even the possibility that you’re entirely wrong. It’s a badge of the humility that defines science, when things are working right. And it sure couldn’t hurt if that sort of humility became more commonplace in lots of other settings.

The second reason is that the variability around an average is usually much more interesting than the average itself. “More interesting” in the scientific sense; if variability means that there’s stuff you can’t explain yet, it’s also the guide to where to look to understand things more, to identifying previously unappreciated factors that give rise to the variability.

For example, “On the average, having a particular variant of a gene produces a particular behavior in people….unless, as it then turned out with additional research, someone had a particular type of childhood.” Pursuing the question of why there were exceptions revealed all sorts of things about environmental regulation of gene transcription, child development, gene/environment interactions, and so on.

Moreover, variability is often more interesting in the human sense as well. All things considered, it’s not that exciting that humans average a score of 100 on this thing called an IQ test. It’s the variation that interests us. Or that, on the average, adult humans males can run 100 meters in, say, 25 seconds. It’s Usain Bolt that gets our attention. Or that there’s an average life expectancy—it’s what you and your loved ones are destined for that matters. Crowds of protestors don’t gather in some nation’s capital because of the average income in that country; it’s because of the magnitude of the variance, the extent of inequality.

The third reason is the most important and subtle, and ultimately has little to do with science. Take a population of people. Figure out their average height. Their average weight. Average IQ, shoe size, number of friends, hip/waist ratio, radiant brightness of smile, symmetry of face, athletic skill, sex drive, scores on psych instruments that measure perkiness or optimism or gumption. Define your average human across these parameters. And then good luck trying to find such a person. Because they don’t exist. Even if someone seems to be, say, the average weight, they won’t really be if you look closely enough, measuring things out to the level of grams, or milligrams, or micrograms, or…. Nothing and no one is precisely average, because “averageness” is an emergent property of populations, an artificial construct. It’s like a strange attractor in chaotic systems, which oscillates around a singular point, a hypothetical average that, no matter how closely you look, is never actually achieved. Oh, it does “on the average,” but never in reality.

This matters because, psychologically, we tend to morph “average” into “the norm” and then into “normal” or “ideal.” And what that means is that we all always come up short in achieving what we have labeled as normal and ideal; we’re all a little too heavy, or too tall, with a nose that’s a little too much this, a personality that’s a bit too little that. We all deviate from the norm, from something that is an artificial, statistical construct that does not really exist. We all are “abnormal,” in a sense that’s more pejorative than statistical. And thus feel badly about who we are. What “on the average” truly means in populations is liberating.

Over the centuries, reflective individuals have speculated about the causes of differences among individuals (who is talented and why, what makes certain persons influential) and among societies (why have certain cultures thrived and others vanished; which societies are bellicose and why). In the 19^th century, scholars began systematic study of such issues; work in this vein has continued and expanded. This line of study has been dubbed historiometrics or, alternatively, cliometrics. These terms, while literally and etymologically appropriate, are hardly transparent or snappy—among the reasons that historiometric scholars and approaches have yet to receive due credit.

Among continental scholars, the Belgian statistician Adolphe Quetelet is often credited with opening up this line of work. But in the Anglo-American scholarly community, the British polymath Francis Galton is generally seen as the patron saint of historiometric studies. As a member of the distinguished Darwin family, Galton was particularly interested in the nature and incidence of genius. Using statistical methods, he demonstrated how genius thrived in certain families and attributed this distribution to hereditary factors. Yet Galton was also sensitive to the possible confounds between hereditary and environmental contributions; and so he pioneered in comparisons of identical twins, fraternal twins, and other members of the same family. On both sides of the English channel, historiometric studies had been launched!

In the 20^th century, these lines of work, whether or not officially labeled as historiometry, continued in many locales. Deserving special mention is the American psychologist, Dean Keith Simonton, who has devoted several decades to historiometric study—carrying out dozens of intriguing studies as well as explicitly laying out the methods available in the armamentarium of the historiometric scholar. If you are speculating about the kinds of issues alluded to above—for example, during which decade of life do scholars in specific domains carry out their most influential research, during which decade of life do artists create their most enduring works—chances are that Simonton has already carried out relevant research; and, if not, he will readily suggest how such studies might be conducted and their results interpreted.

I’ve often wondered why Simonton’s work is not more widely known and appreciated. I suspect it’s because the work spans standard social science (psychology, sociology) and humanistic studies (history, the arts)—and most scholars work comfortably within, rather than across, these two cultures. It’s also notable that Simonton has worked largely alone—with neither a big staff nor a large research budget—and that is an unusual research profile in our time.

Enter big data. Neither Quetelet nor Galton had significant computational aids—pencils and paper on one’s desk were the media of choice. When Simonton began his studies in the 1970s, we were in the era of large mainframes, punch cards, and limited computing power. To be sure, Simonton (and other self-styled cliometricians like Charles Murray) has kept up with advances in technology. But only in the last decade or so has it become possible, indeed easy, to pursue historiometric puzzles, drawing on vast amounts of data that are sitting on one’s lap, or, more precisely, on one’s laptop.

Occasionally, historiometric findings have found their way into the mainstream media. A few months ago, a team of researchers led by Roberta Sinatra introduced the Q phenomenon. Curious about the distribution of influential work over the investigative lifespan of productive scientists, the researchers examined publication records of scientists drawn from seven disciplines. And they discovered—presumably to their and others’ surprise—that “the highest-impact work in a scientist’s career is randomly distributed within her body of work.” Many commentators, prominent among them Simonton, have reacted to this claim and it’s safe to say that we have not heard the last of the Q phenomenon.

Has the moment for historiometry finally arrived? Given the fascination of historiometric questions, and the relative ease these days of researching them, what was once an exotic exercise of eccentric European savants can now become a regular part of the disciplinary terrain.

I am not quite persuaded that the moment has arrived. To be sure, the availability of vast sources of data and powerful data mining techniques have greatly enhanced the "metric" part of historiometrics. Yet, the "historio" part is equally important. Historians should be judged by the quality of the questions that they raise, and the sense that they are able to make of what they have uncovered. These are issues of judgment, not merely issues of measurement. (As has long been quipped, "garbage in, garbage out.") And so, to return to the Q phenomenon, the unexpected findings of the Sinatra team open up a slew of possible explanations and interpretations. But the available data themselves will never tell us which issues to pursue next—one needs a solid historical sense as well as a dollop of historical humility. And whether the Sinatra team—or some other team or individual—will itself significantly raise the stock of historiometry depends on its historical wisdom as well as its data analytic prowess.

There may be broader lessons here. Scholarship—whether scientific or humanistic—always entails a dialectic between issues worth pursuing and the methods available for pursuing them.

Historiometric curiosity dates back to Classical times; but it has been the advent of measurement techniques (statistics, data analytics) that has permitted this curiosity to be pursued with increasing power and elegance. It’s desirable to maintain a balance between questions/curiosity/judgment, on the one hand, and analytic measures, on the other: when either becomes dominant, the pursuit itself can be compromised.

And for extra credit: when you want to explain to others what you are up to, find a succinct and memorable descriptor!

An astonishing concept has entered mainstream cosmological thought: it deserves to be more widely known. Physical reality could be hugely more extensive than the patch of space and time traditionally called "the universe." Our cosmic environment could be richly textured, but on scales so vast that our astronomical purview is restricted to a tiny fraction: we’re not aware of the "grand design," any more than a plancton whose "universe" was a spoonful of water would be aware of the world’s topography and biosphere. We may inhabit a "multiverse."

However powerful our telescopes are, our vision is bounded by a horizon: a shell around us, delineating the distance light can have travelled since the big bang. But this shell has no more physical significance than the circle that delineates your horizon if you're in the middle of the ocean. There are billions of galaxies within our horizon. But we expect far more galaxies lying beyond. We can’t tell just how many. If space stretched far enough, then all combinatorial possibilities would be repeated. Far beyond the horizon, we could all have avatars. And it may be some consolation that when we make a bad decision, there’s another one of us, far beyond our horizon, who has made a better one.

So the aftermath of "our" big bang could encompass a stupendous volume. But that’s not all. "Our" big bang could be just one island of space-time in an unbounded cosmic archipelago. A challenge for 21st-century physics is to answer two questions. First, are there many "big bangs" rather than just one? Second—and this is even more interesting—if there are many, are they all governed by the same physics? Or is there a huge number of different vacuum states—each the arena for different microphysics, and therefore offering differing propensities for spawning life?

If the answer to this latter question is "yes" there will still be overarching laws governing the multiverse—maybe a version of string theory. But what we’ve traditionally called the laws of nature will be just local bylaws. Even though it makes some physicists foam at the mouth, we then can’t avoid the A-word—“anthropic." Many domains could be still-born or sterile: the laws prevailing in them might not allow any kind of complexity. We therefore wouldn’t expect to find ourselves in a typical universe. Ours would belong to the unusual subset where there was a "lucky draw" of cosmic numbers conducive to the emergence of complexity and consciousness. Its seemingly designed or fine-tuned features wouldn't be surprising.

Some claim that unobservable entities aren’t part of science. But it’s hard to defend that hard-line view. For instance, (unless we are in some special central position and our universe has an "edge" just beyond the present horizon) there will be some galaxies lying beyond our horizon—and if the cosmic acceleration continues they will remain beyond forever. Not even the most conservative astronomer would deny that these never-observable galaxies (which, as I’ve already mentioned, could hugely outnumber those we can see) are part of physical reality. These galaxies are part of the aftermath of our own big bang. But why should they be accorded higher epistemological status than unobservable objects that are the aftermath of other big bangs? So it’s surely a genuine scientific question to ask whether there’s one big bang or many.

Fifty years ago we weren’t sure whether there was a big bang at all (Fred Hoyle and other "steady statesmen" still contested the idea). Now we can confidently describe cosmic history back to the ultra-dense first nanosecond. So in fifty more years, it’s not overoptimistic to hope that we may have a "unified" physical theory, corroborated by experiment and observation in the everyday world, that tells us what happened in the first trillionth of a trillionth of a trillionth of a second, when inflation is postulated to have occurred. If that theory predicts multiple big bangs we should take that prediction seriously even though it can’t be directly verified (just as we take seriously general relativity’s predictions for the unobservable insides of black holes, because the theory has survived many tests in domains we can observe).

Some physicists don’t like the multiverse: they’d be disappointed if some of the key numbers they are trying to explain turn out to be mere environmental contingencies governing our local space-time patch—no more truly "fundamental" than the parameters of the Earth’s orbit round the Sun. But that disappointment would surely be outweighed by the revelation that physical reality was grander and richer than hitherto envisioned. In any case our preferences are irrelevant to the way physical reality actually is—so we should surely be open-minded.

Indeed, there’s an intellectual and aesthetic upside. If we’re in a multiverse, it would imply a fourth and grandest Copernican revolution; we’ve had the Copernican revolution itself, then the realization that there are billions of planetary systems in our galaxy; then that there are billions of galaxies in our observable universe. But now that’s not all. The entire panorama that astronomers can observe could be a tiny part of the aftermath of "our" big bang, which is itself just one bang among a perhaps-infinite ensemble.

We may, by the end of this century, be able to say, with confidence, whether we live in a multiverse, and how much variety its constituent "universes" display. The answer to this question will determine how we should interpret the "bio friendly" universe in which we live.

In physics, math, and computer science, the state of a system is an encapsulation of all the information you'd ever need to predict what it will do, or at least its probabilities to do one thing versus another, in response to any possible prodding of it. In a sense, then, the state is a system’s “hidden reality,” which determines its behavior beneath surface appearances. But in another sense, there’s nothing hidden about a state—for any part of the state that never mattered for observations could be sliced off with Occam’s Razor, to yield a simpler and better description.

When put that way, the notion of “state” seems obvious. So then why did Einstein, Turing, and others struggle for years with the notion, along the way to some of humankind’s hardest-won intellectual triumphs?

Consider a few puzzles:

1. To add two numbers, a computer clearly needs an adding unit, with instructions for how to add. But then it also needs instructions for how to interpret the instructions. And it needs instructions for interpreting those instructions, and so on ad infinitum. We conclude that adding numbers is impossible by any finite machine.
2. According to modern ideas about quantum gravity, space might not be fundamental, but rather emergent from networks of qubits describing degrees of freedom at the Planck scale. I was once asked: if the universe is a network of qubits, then where are these qubits? Isn’t it meaningless, for example, for two qubits to be “neighbors,” if there’s no pre-existing space for the qubits to be neighboring in?
3. According to special relativity, nothing can travel faster than light. But suppose I flip a coin, write the outcome in two identical envelopes, then put one envelope on earth and the other on Pluto. I open the envelope on earth. Instantaneously, I’ve changed the state of the envelope on Pluto, from “equally likely to say heads or tails” to “definitely heads” or “definitely tails”! (This is normally phrased in terms of quantum entanglement, but as we see, there’s even a puzzle classically.)

The puzzle about the computer is a stand-in for countless debates I’ve gotten into with non-scientist intellectuals. The resolution, I think, is to specify a state for the computer, involving the numbers to be added (encoded, say, in binary), and a finite control unit that moves across the digits adding and carrying, governed by Boolean logic operations, and ultimately by the laws of physics.  It might be asked: what underlies the laws of physics themselves? And whatever the answer, what underlies that? But those are questions for us. In the meantime, the computer works; everything it needs is contained in its state.

This question about the qubits is a cousin of many others: for example, if the universe is expanding, then what’s it expanding into? These aren’t necessarily bad questions. But from a scientific standpoint, one is perfectly justified to respond: “you’re proposing we tack something new onto the state of the world, such as a second space for ‘our’ space to live in or expand into. So would this second space make a difference to observation? If it never would, why not cut it out?”

The question about the envelopes can be resolved by noticing that your decision on earth, to open your envelope or not, doesn’t affect the probability distribution over envelope contents that would be perceived by an observer on Pluto. One can prove a theorem stating that an analogous fact holds even in the quantum case, and even if there’s quantum entanglement between earth and Pluto: nothing you choose to do here changes the local quantum state (the so-called density matrix) over there. This is why, contrary to Einstein’s worries, quantum mechanics is consistent with special relativity.

The central insight here—of equal importance to relativity, quantum mechanics, gauge theory, cryptography, artificial intelligence, and probably 500 other fields—could be summarized as “a difference that makes no difference is not a difference at all.” This slogan might remind some readers of the early 20^th-century doctrine of logical positivism, or of Popper’s insistence that a theory that never ventures any falsifiable prediction is unscientific. For our purposes, though, there’s no need to venture into the complicated debates about what exactly the positivists or Popper got right or wrong (or whether positivism is itself positivistic, or falsifiability falsifiable).

It suffices to concentrate on a simpler lesson: that yes, there’s a real world, external to our sense impressions, but we don’t get to dictate from our armchairs what its state consists of. Our job is to craft an ontology around our best scientific theories, rather than the opposite. That is, our conception of “what’s actually out there” always has to be open to revision, both to include new things that we’ve discovered can be distinguished by observation and to exclude things that we’ve realized can’t be.

Some people seem to find it impoverishing to restrict their ontology to the state, to that which suffices to explain observations. But consider the alternatives. Charlatans, racists, and bigots of every persuasion are constantly urging us to look beyond a system’s state to its hidden essences, to make distinctions where none are called for.

Lack of clarity about the notion of “state” is even behind many confusions over free will.  Many people stress the fact that, according to physics, your future choices are “determined” by the current state of the universe.  But this ignores the fact that, regardless of what physics had to say on the subject, the universe’s current state could always be defined in such a way that it secretly determined future choices—and indeed, that’s exactly what so-called hidden-variable interpretations of quantum mechanics, such as Bohmian mechanics, do.  To me, this makes “determination” an almost vacuous concept in these discussions, and actual predictability much more important.

State is my choice for a scientific concept that should be more widely known, because buried inside it, I think, is the whole scientific worldview.

You can never understand one language until you understand at least two.

This statement by the English writer, Geoffrey Willans, feels intuitive to anyone who has studied a second language. The idea is that learning to speak a foreign language inescapably conveys deeper understanding of one’s native language. Goethe, in fact, found this such a powerful concept that he felt moved to make a similar, but more extreme, assertion:

He who does not know foreign languages does not know anything about his own.

As compelling as this may be, what is perhaps surprising is that the essence of this idea—that learning or improvement in one skill or mental function can positively influence another one—is present not only in human intelligence, but also in machine intelligence. The effect is called transfer learning, and besides being an area of fundamental research in machine learning, it has potentially wide-ranging practical applications.

Today, the field of machine learning, which is the scientific study of algorithms whose capabilities improve with experience, has been making startling advances. Some of these advances have led to computing systems that are competent in skills that are associated with human intelligence, sometimes to levels that not only approach man’s capabilities but, in some cases, exceeds it. This includes, for example, the ability to understand, process, and even translate languages. In recent years, much of the research in machine learning has focused on the algorithmic concept of deep neural networks, or DNNs, which learn essentially by inferring patterns—often patterns of remarkable complexity—from large amounts of data. For example, a DNN-based machine can be fed many thousands of snippets of recorded English utterances, each one paired with its text transcription, and from this discern the patterns of correlation between the speech recordings and the paired transcriptions. These inferred correlation patterns get precise enough that, eventually, the system can “understand” English speech. In fact, today’s DNNs are so good that, when given enough training examples and a powerful enough computer, they can listen to a person speaking and make fewer transcription errors than would any human.

What may be surprising to some is that computerized learning machines exhibit transfer learning. For example, let’s consider an experiment involving two machine-learning systems, which for the sake of simplicity we’ll refer to as machines A and B. Machine A uses a brand-new DNN, whereas machine B uses a DNN that has been trained previously to understand English. Now, suppose we train both A and B on identical sets of recorded Mandarin utterances, along with their transcriptions. What happens? Remarkably, machine B (the previously English-trained one) ends up with better Mandarin capabilities than machine A. In effect, the system’s prior training on English ends up transferring capabilities to the related task of understanding Mandarin.

But there is an even more astonishing outcome of this experiment. Machine B not only ends up better on Mandarin, but B’s ability to understand English is also improved! It seems that Willans and Goethe were onto something—learning a second language enables deeper learning about both languages, even for a machine.

The idea of transfer learning is still the subject of basic research, and as such, many fundamental questions remain open. For example, not all “transfers” are useful because, at a minimum, for transfer to work well, there appears to be a need for the learned tasks to be “related” in ways that still elude precise definition or scientific analysis. There are connections to related concepts in other fields, such as cognitive science and learning theory, still to be elucidated. And while it is intellectually dangerous for any computer scientist to engage in “anthropomorphizing” computer systems, we cannot avoid acknowledging that transfer learning creates a powerful, alluring analogy between learning in humans and machines; surely, if general artificial intelligence is ever to become real, transfer learning would seem likely to be one of the fundamental factors in its creation. For the more philosophically minded, formal models of transfer learning may contribute to new insights and taxonomies for knowledge and knowledge transfer.

There is also exceptionally high potential for applications of transfer learning. So much of the practical value of machine learning, for example in search and information retrieval, has traditionally focused on systems that learn from the massive datasets and people available on the World-Wide Web. But what can web-trained systems learn about smaller communities, organizations, or even individuals? Can we foresee a future where intelligent machines are able to learn useful tasks that are highly specialized to a specific individual or small organization? Transfer learning opens the possibility that all the intelligence of the web can form the foundation of machine-learned systems, from which more individualized intelligence is learned, through transfer learning. Achieving this would amount to another step towards the democratization of machine intelligence.

When he was sixteen years old, Albert Einstein imagined that he was chasing after a beam of light and observed what he “saw.” And this vision launched him on the path to developing his theory of special relativity. Einstein often engaged in such thinking; he reported: "…The psychical entities which seem to serve as elements in thought are certain signs and more or less clear images which can be ‘voluntarily’ reproduced and combined... this combinatory play seems to be the essential feature in productive thought before there is any connection with logical construction in words or other kinds of signs which can be communicated to others..."

Einstein was relying on mental emulation, a kind of thought that many of us use and we all probably could use more frequently—and more productively—if we become aware of it. Mental emulations range from the sublime to the ordinary, such as imagining currently impossible events (including chasing after a beam of light), or imagining the best route to take when climbing up a hill, or visualizing the best way to pack your refrigerator before starting to take cans, boxes and jugs out of your shopping bag.

A mental emulation is a way to simulate what you would expect to happen in a specific situation. Just as Einstein reported, most mental emulations appear to involve mental imagery. By definition, mental imagery is an internal representation that mimics perception. You can visualize “running” an event in your head, and “seeing” how it unfolds. Mental emulations are partly a way of allowing you to access knowledge that you have only implicitly, to allow such knowledge to affect your mental images.

My colleague Sam Moulton and I characterize mental emulation as follows: Each step of what you imagine represents the corresponding step in the real world, and the transitions between steps represent the corresponding transitions between steps in the event. Moreover, the processes that transition from state to state in the representation mimic the actual processes that transition from state-to-state in the world. If you imagine kicking a ball, you will actually activate motor circuits in the brain, and neural states will specify the transitional states of how your leg is positioned as you kick. And as you imagine tracking the flying ball, your mental images will transition through the intermediate states in ways analogous to what would happen in the real world.

Mental emulation is a fundamental form of thought, and should be recognized as such.

You may have worked so closely with a partner that you reached a point where each of you could finish the other’s sentences. You had a pretty good idea of how your partner would respond to an event. What’s behind such a skill?  

Decentering is the activity of taking the perspective of another entity. When we look into the past, we try to explain why that entity behaved in a way that might have surprised us. Peering into the future, we decenter in order to anticipate what that entity is likely to do.

Skilled decentering comes into play, not just with partners, but also with strangers and even with adversaries. It gives us an edge in combat, to get ready to ward off an adversary’s attack. It helps authors write more clearly as they anticipate what might confuse a reader or what a reader expects to find out next. Police officers who are good at decentering can de-escalate a tense situation by moving an encounter into less volatile directions. Teachers rely on decentering to keep the attention of students by posing questions or creating suspense, anticipating what will catch the attention of their students. Team members who can quickly decenter can predict how each will react to unexpected changes in conditions, increasing their coordination.

Decentering is not about empathy—intuiting how others might be feeling. Rather, it is about intuiting what others are thinking. It is about imagining what is going through another person’s mind. It is about getting inside someone else’s head.

Despite its importance, particularly for social interaction, decentering has received very little attention. Military researchers have struggled to find ways to study decentering and to train commanders to take the perspective of an adversary.  Social psychologists have not made much progress in unpacking decentering. Part of the problem might be that researchers examine the average decentering accuracy of observers whereas they should be investigating those observers whose accuracy is consistently above average. What are their secrets? What is behind their expertise?

It could be even more valuable to study decentering outside the laboratory—in natural settings. I suspect that some people don’t even try to decenter, others may try but aren’t particularly effective, and still others may be endowed with an uncanny ability to take another person’s perspective.

Decentering also comes into play when we interact with inanimate objects such as intelligent technologies that are intended to help us make decisions. That’s why the definition at the beginning of this essay refers to entities rather than people. The human factors psychologist Earl Wiener once described the three typical questions people ask when interacting with information technology: What is it doing? Why is it doing that? What will it do next? We are trying to take the perspective of the decision aid.

It would be nice if the decision aid could help us decenter, the way a teammate could. After all, when we interact with a person, it is natural to ask that person to explain his or her choice. But intelligent systems struggle to explain their reasons. Sometimes they just recite all the factors that went into the choice, which isn’t particularly helpful. We want to hear less, not more. We want the minimum necessary information. And the minimum necessary information depends on decentering.

For example, if you are using a GPS system to drive to a location, and the device tells you to turn left whereas you expected to turn right, you would NOT want the system to explain itself by showing the logic it followed. Rather, the system should understand why you expected a right-hand turn at that juncture and should provide the central data element (e.g., an accident up ahead is creating thirty-minute delays on that route) you need to understand the choice.

If you were driving with a human navigator who unexpectedly advised you to turn left, all you would have to do is say, “Left?” Most navigators instantly determine from your intonation that you want an explanation, not a louder repetition. Good navigators grasp that you, the driver, didn’t expect a left-hand turn and would quickly (remember, there is traffic getting ready to honk at you) explain, “Heavy traffic ahead.” Your navigator would convey the minimum necessary information, the gist of the explanation. But the gist depends on your own beliefs and expectations—it depends on the navigator’s ability to decenter and get inside your head. Smart technology will never be really smart until it can decenter and anticipate. That’s when it will become a truly smart partner.

When we are uncomfortable with the recommendations offered by a decision aid we don’t have any easy ways to enter into a dialog. In contrast, when we dislike a person’s suggestions we can examine his or her reasons. Being able to take someone else’s perspective lets people disagree without escalating into conflicts. It allows people to be more decent in their interactions with each other. If they can decenter, then they can become decenter.

If you asked one hundred people on the street if they understand how a refrigerator works, most would respond, yes, they do. But ask them to then produce a detailed, step-by-step explanation of how exactly a refrigerator works and you would likely hear silence or stammering. This powerful but inaccurate feeling of knowing is what Leonid Rozenblit and Frank Keil in 2002 termed, the illusion of explanatory depth (IOED), stating, “Most people feel they understand the world with far greater detail, coherence, and depth than they really do.”

Rozenblit and Keil initially demonstrated the IOED through multi-phase studies. In a first phase, they asked participants to rate how well they understood artifacts such as a sewing machine, crossbow, or cell phone. In a second phase, they asked participants to write a detailed explanation of how each artifact works, and afterwards asked them re-rate how well they understand each one. Study after study showed that ratings of self-knowledge dropped dramatically from phase one to phase two, after participants were faced with their inability to explain how the artifact in question operates. Of course, the IOED extends well beyond artifacts, to how we think about scientific fields, mental illnesses, economic markets and virtually anything we are capable of (mis)understanding.

At present, the IOED is profoundly pervasive given that we have infinite access to information, but consume information in a largely superficial fashion. A 2014 survey found that approximately six in ten Americans read news headlines and nothing more. Major geopolitical issues from civil wars in the Middle East to the latest climate change research advances are distilled into tweets, viral videos, memes, “explainer” websites, soundbites on comedy news shows, and daily e-newsletters that get inadvertently re-routed to the spam folder. We consume knowledge widely, but not deeply.

Understanding the IOED allows us to combat political extremism. In 2013, Philip Fernbach and colleagues demonstrated that the IOED underlies people’s policy positions on issues like single-payer health care, a national flat tax, and a cap-and-trade system for carbon emissions. As in Rozenbilt and Keil’s studies, Fernbach and colleagues first asked people to rate how well they understood these issues, and then asked them to explain how each issue works and subsequently re-rate their understanding of each issue. In addition, participants rated the extremity of their attitudes on these issues both before and after offering an explanation. Both self-reported understanding of the issue and attitude extremity dropped significantly after explaining the issue—people who strongly supported or opposed an issue became more moderate. What is more, reduced extremity also reduced willingness to donate money to a group advocating for the issue. These studies suggest the IOED is a powerful tool for cooling off heated political disagreements.

The IOED provides us much-needed humility. In any domain of knowledge, often the most ignorant are the most overconfident in their understanding of that domain. Justin Kruger and David Dunning famously showed that the lowest performers on tests of logical reasoning, grammar, and humor are most likely to overestimate their test scores. Only through gaining expertise in a topic do people recognize its complexity and calibrate their confidence accordingly. Having to explain a phenomenon forces us to confront this complexity and realize our ignorance. At a time where political polarization, income inequality, and urban-rural separation have deeply fractured us over social and economic issues, recognizing our only modest understanding of these issues is a first step to bridging these divides.