How are monkeys and humans different (I mean, besides the tail)

Marc Hauser, one of a handful of professors to be tenured by Harvard University (most senior faculty come from other universities), has spent much of his career showing that non-human primates are smart. It is very dangerous to say "Only humans can do X," because Hauser will come along and prove that the cotton-top tamarin can do X as well. Newborn babies can tell Dutch from Japanese? Well, so can the tamarins.

For this reason, I have wondered what Hauser thinks really separates human cognition from that of other animals. He is well-known for a hypothesis that recursion is the crucial adaptation for language, but I'm never sure how wedded he is to that hypothesis, and certainly he can't think the ability to think recursively is all that separates human thought from tamarin thought.

Luckily for me, he gave a speech on just that topic at one of the weekly departmental lunches. Hopefully, he'll write a theory paper on this subject in the near future, if he hasn't already. In the meantime, I'll try to sketch the main point as best I understood it.

Hauser is interested in a paradox. In many ways, non-human primates look quite smart -- even the lowly tamarin. Cotton-top tamarins have been able to recognize fairly complex grammatical structures, yet they do not seem to use those abilities in the same ways we do -- for instance, they certainly don't use grammar.

In some situations, non-human primates seem to have a theory of mind (an understanding of the contents of another's mind). For instance, if a low-ranking primate (I forget the species, but I think this was with Chimpanzeees) sees two pieces of good food hidden and also sees that a high-ranking member of the troop can see where one piece was hidden but not the other, the low-ranking primate will high-tail it to the piece of food only he can see. That might seem reasonable. But contrast it with this situation: these primates also know how to beg for food from the researchers. What if primate is confronted with two researchers, one who has a cloth over her eyes and one who has a cloth over her ears. Does the primate know to beg only from the one who can see? No.

Similarly, certain birds can use deception to lure a predator away from their nest, but they never use that deceptive behavior in other contexts where it might seem very useful.

These are just three examples where various primates seem to be able to perform certain tasks, but only in certain contexts or modalities. Hauser proposes that part of what makes humans so smart are the interfaces between different parts of our brains. We can not only recognize statistical and rule-based regularities in our environment -- just like tamarins -- but we can also use that information to produce behavior with these same statistical and rule-based regularities. That is, we can learn and produce grammatical language. We can take something we learn in one context and use it in another. To use an analogy he didn't, our brains are an office full of computers after they have been efficiently networked. Monkey computer networks barely even have modems.

This same theory may also explain great deal of strange human infant behavior. More about that in the future.

Do ballplayers really hit in the clutch?

If you've been watching the playoffs on FOX, you'll notice that rather than present a given player's regular-season statistics, they've been mostly showing us their statistics either for all playoff games in their career, or just for the 2007 post-season. Is that trivia, or is it an actual statistic? For instance, David Ortiz hits better in the post-season than during the regular season. OK, one number is higher than the other, but that could just be random variation. Does he really hit better during the playoffs?

Why does this even matter? There is conventional wisdom in baseball that certain players hit better in clutch situations -- for instance, when men on base. This is why RBIs (runs-batted-in) are treated as a statistic, rather than as trivia. Some young Turks (i.e., Billy Beane of the Oakland A's) have argued vigorously that RBIs don't tell you anything about the batter -- they tell you about the people who bat in front of him (that is, they are good at getting on base). Statistically, it is said, few to no ballplayers hit better with men on and 2 outs.

So what about in the post-season?

I couldn't find Ortiz's lifetime post-season stats, so I compared this post-season, during which he's been phenomenally hot (.773 on-base percentage through the weekend -- I did this math last night during the game, so I didn't include last night's game), compared with the 2007 regular season, during which he was just hot (.445 on-base percentage).

There are probably several ways to do the math. I used a formula to compare two independent proportions (see the math below). I found that his OBP is significantly better this post-season than during the regular season. So that's at least one example...

Here's the math.

You need to calculate a t statistic, which is the difference between the two means (.773 and .445) divided by the standard deviation of the difference between those two means. The first part is easy, but the latter part is complicated by the fact that we're dealing with ratios. That formula is:

square root of: (P1*(1-P1)/N1 + P2*(1-P2)/N2)
where P1 = .773, P2 = .445, N1 = 659 (regular season at-bats - 1), N2 = 22 (post-season at-bats - 1).

t = 2.99, which gives a p value of less than .01.

I was also considering checking just how unusual Colorado's winning streak is, but that's where my knowledge of statistics broke down (maybe we'll learn how to do that next semester). If anybody has comments or corrections on the stats above or can produce other MBL-related math, please post it in the comments.

New Harvard president to be installed today

Drew Faust will be installed today as Harvard's 28th president. That's right -- Harvard has only had 28 presidents since Henry Dunster was named in 1640. That's not counting some acting presidents, like Samuel Willard (1701-1707) or Nathaniel Eaton, who was "schoolmaster" from 1637 to 1639.

Faust is of course the first female president of Harvard. She is also the first since Charles Chauncy (1654-1672) to have neither an undergraduate nor graduate degree from Harvard (From 1672 to 1971, all Harvard presidents had done their undergraduate work at Harvard. The last three -- Derek Bok, Neil Rudenstine and Larry Summers -- had graduate degrees from Harvard).

The ceremony will be outdoors on Harvard Yard at 2pm. It rained heavily overnight, but it seems to be clearing up now. If the weather is decent, I'll check it out and report back. I hear tell that Harvard ceremonies have all the pomp and splendor you would expect, but I have yet to see one.

People who can't count

Babies can't count. Adults can. When I say that babies don't count, I don't mean that they don't know the words "one," "two," "three," or "four." That's obvious. What I mean is that if you give an infant the choice between 5 graham crackers or 7, the baby doesn't know which to pick.

Does that mean we have to learn numbers, or does it mean that the number system simply comes online as we mature. Babies also have bad vision, but that doesn't mean the learn vision. One of the reasons we might assume that number is innate rather than learned is that all reasonably intelligent children learn to count around the same time...or do they? This is where a few cultures, such as the Piraha, become very important.

I believe that I have heard that there are some languages that only have words for "one," "two," and "many," but I'm not sure, so if you know, please make comments. I am fairly certain that the Piraha are the only known culture not to even have a word for "one."

How does one know whether they have a word for "one?" Your first impulse might be to check a bilingual dictionary, but that begs the question. How did the dictionary-maker know? The way a few people have done it (like Peter Gordon at Columbia and Ted Gibson & Mike Frank from MIT) is to show the Piraha a few objects and see what they say. According to Mike Frank's talk at our lab a couple weeks ago, they never found a word that was used consistently to describe "one" anything. Instead, there was a number that was used for small numbers of object (1, 2, etc.), another word for slightly larger numbers, and third word that seems to be used the way we use "many."

Well, maybe they just weren't using number words in this task. Did they really even understand what was being required of them? Who knows. But there are other ways to do the experiment. For instance, you can test them the same way we test babies. You show them two boxes. You put 5 pieces of candy into one box. Then you put 7 pieces of candy into the other box. Then you ask them which box they want. Remember that they never see both groups of candy at the same time, so they have to remember the groups of candy in order to compare them.

Well, Mike Frank tried this. This is an example of the responses he got:

"Can I have both boxes?"

No. You have to choose.

"Oh, is that what this game is about? I don't want to play this game. Who needs candy? Can we do spools of thread? My wife needs those. Or how about some shotgun shells?"

This experiment was a failure. Instead, they tried a matching task. You show them a row of, say, 5 spools of thread. Then you ask them to put down the same number of balloons as there are spools of thread. They can do this. Now, you change the game. You show them some number of spools of thread, then cover those spools. You then ask them to put down the same number of balloons. Since they can't see the thread, they have to do this by memory.

The Piraha fail at this and related tasks. People who can count do not.

Of course, they might not have understood the task. This is very hard to prove one way or another. I have been running a study in my lab that involves recent Chinese immigrants. I designed the study and tested it in English with Harvard undergrads. They found the task challenging, but they quickly figured out what I needed them to do. Some of my immigrant participants do so as well, but many of them find it impossibly difficulty -- literally. Some of them have to give up.

It's not that they aren't smart. Most of them are Harvard graduate students or even faculty. What seems to be going on is a culture clash. For one thing, they aren't usually familiar with psychology experiments, since very few are done in China. I suspect that some of the things I ask them to do (repeat a word out loud over and over, read as fast as possible, etc.) may seem perfectly normal requests to my American undergraduates but very odd to my Chinese participants, just as the Piraha discussed above didn't want to choose boxes of candy. So it is always possible that the Piraha act differently in these experiments because they have different cultural expectations and have trouble figuring out what exactly is required of them.

That said, it seems pretty unlikely at this point that the Piraha have number words or count. This suggests counting must be learned. In fact, it suggests counting must be taught. This contrasts with language itself, which often seems to spring up spontaneously even when the people involved have had little exposure to an existing language. (Click here for a really interesting take on why the Piraha don't seem particularly interested in learning how to count.)

Brazil issues warrant for scientist's arrest

The government of Brazil recently ordered the arrest of the well-known linguist and anthropologist, Dan Everett.

Everett has been both famous and infamous for his study of the Piraha people, a small tribe in Brazil. He has made a number of extraordinary claims about their culture and language, such as that they do not have number words or myths.

These claims are important because they undermine a great deal of current linguistic and psychological theory, and so they have been hotly debated. Some of these debates, however, have spilled over from arguments about data and method to personal attacks.

Just before Everett spoke at MIT last fall, a local linguist sent out an email to what amounted to much of Boston's scientific community involved in language and thought. It looked like the sort of email that one means to send to a close friend and accidentally broadcasts. Language Log describes it better than I can, but the gist was that Everett is a liar who exploits the poor Piraha for his own fame and glory.

This was just one instance in a series of ad hominem attacks on Everett over the last few years. I am not going to weigh in on whether Everett is exploiting anybody, because I simply don't feel I know enough. I've never met a Piraha -- not that any of Everett's detractors have either, to my knowledge. If this means anything, I am told by friends who have visited the Piraha that they really like Everett.

A couple weeks ago, I heard from a friend who has collaborated with Everett that all further research on Piraha language and has been essentially banned. A warrant is out for Everett's arrest on charges of, essentially, exploiting the Piraha. I have no idea how much this has to do with the controversy the aforementioned linguist has been raising, but I suspect that it is not unrelated.

On the topic of language, my Web-based study of how people interpret sentences is still ongoing, and I could use more participants. Not to exploit this post to further my own academic fame and glory...

Scientists create mice with human language gene

Scientists at the Max Plank institute in Germany recently announced that they had successfully knocked the human variant of the FOXP2 "language" gene into mice.

The FOXP2 gene, discovered in 2001, is the most famous gene known to be associated with human language. There has been some debate about what exactly it does, but a point mutation in the gene is known to cause speech and language disorders.

Part of the interest in FOXP2 stems from the fact that it is found in a wide range of species, including songbirds, fish and reptiles with only slight variations. Also, FOXP2 is expressed in many parts of the body, not just the brain. Previous research had found that removing the gene from mice decreased their vocalizations...and ultimately killed the mice.

In the new study, scientists created a new mouse "chimera" with the human variant of the FOXP2 gene. This time, the only differences they could find between the transgenic mice and typical mice was in their vocalizations.

Read more about FOXP2.

(Disclosure: This research does not appear to have been published yet. I heard about it from Marc Hauser of Harvard University, who heard about it this summer from a conference talk by Svante Paabo of Max Plank, one of the researchers involved in the project.)

GoogleScience

A Google search can help you find cutting-edge research. A Google search can also be cutting-edge research.

Many questions in linguistics (the formal study of language) and psycholinguistics (the study of language as human behavior) are answered by turning to a corpus. A corpus is a large selection of texts and/or transcripts. Just a few years ago, they were difficult and expensive to create. Arguably the most popular word frequency corpus in English -- the verable Brown Corpus -- was based off of one million words of text. One million words sounds like a lot, but so many of those are "the" and "of" that in fact many words do not appear in the corpus at all.

The Google corpus contains billions of pages of text.

So what does one do with a corpus? One obvious thing is to figure out which words are more common than others. The most common words in English are short function words like "a" (found on 4.95 billion web pages) and "of" (3.61 billion pages). Google, of course, doesn't tell you how many times a words appears, but only on how many pages it appears...which may actually be an advantage, since we're typically more interested in words that show up on many websites than a word that shows up many, many times on just one website.

You might be interested in what the most common noun is. Is it "time," "man," "city," "boy," or "Internet?" You might check to see whether verbs or nouns are more common in English by comparing a large sample of verbs and nouns. You can also compare across languages. Children learn verbs more slowly than nouns in English but not in Chinese. Is that because verbs are more common in Chinese than English?

OK, those are fun experiments, but none of them sound very cutting-edge. If you want to see the Google corpus in action, check out Language Log. The writers there regularly turn to the Google corpus to answer their questions. Google is probably less-commonly used in more formal contexts, but the PsychInfo database turned up 76 hits for "Google." Many were studies about how people use Google, but some were specifically using the Google corpus, such as "Building a customised Google-based collocation collector to enhance language learning," by Shesen Guo and Ganzhou Zhang. Another -- "Nine psychologists: mapping the collective mind with Google" by Jack Arnold -- looked at the organization of conceptual knowledge. At a recent conference, I saw a presentation by vision scientists using Google Image to explore the organization of visual memory. I expect to see more and more of this type of research in the near future.

Of course, there's nothing specific to Google about this. It's just what everybody seems to use.

Free will

In “Self is Magic,” a chapter from Psychology and the Free Will, which is unfortunately not yet out, Daniel Wegner presents some fascinating data showing how easy it is to trick ourselves into believing we are in control of events when in fact we are not. Most of us are in some sense familiar with this illusion. Perhaps we believe our favorite team won partly because we were watching the game (or, if we’re pessimists, because we weren’t).

In one of the many experiments he describes, “the participant was attired in a robe and positioned in front of a mirror such that the arms of a second person standing behind the participant could be extended through the robe to look as though they were the arms of the participant.” The helper’s arms moved through a series of positions. If the participant heard through the headphones a description of each movement before it happened, they were reported “enhanced feeling of control over the arm movements” than if they had not heard those commands in advance. (read the paper here.)

Ultimately, he tries to use these data to explain the “illusion” of free will. In the following paragraphs, I am going to try to unpack the claim and see where it is strong and where it may be weak.

First, what exactly is he claiming? He does not argue that free will does not exist – he assumes it does not exist. That’s not a criticism– you can’t fit everything into one article – but it means I won’t be able to evaluate his argument against free will, though I will discuss free will in terms of his data and hypotheses below. Instead, he is interested in why we think we have free will. In all, he claims that (1) we can be mistaken about whether our thoughts cause events in the world, (2) this is because when we think about something happening and then it happens, we’re biased to believe we’ve caused it, and (3) this illusion that our conscious thoughts lead to actions is useful and adaptive – that is, evolution gave it to us for a reason.

That’s what he says. First, he isn’t really arguing that we don’t have a conscious will. Clearly, we will things to happen all the time. Some of the time he seems to be arguing that our will is simply impotent. The rest of the time he appears to think that the contents of our will are actually caused by something else. That is, our arm decides to move and tells our conscious thoughts to decide to want a cookie (more on this later).

Otherwise, I think claim #1 is pretty straightforward. Sometimes we have an illusion of conscious control when in fact we have none. Wegner compares this to visual illusions, of which there are plenty. Just like with visual illusions, just because you know it’s an illusion (watching the ballgame isn’t going to affect the score) you can’t help not feel it anyway. In fact, given that illusions exist in sight, sound and probably many other senses, so it’s not surprising that the sense of conscious will also is subject to illusions.

It’s important to point out that this itself is not an argument against the potency of conscious will. The fact that we are sometimes mistaken does not mean we are always mistaken. Otherwise, we’d have to claim that because vision is sometimes mistaken, we are all blind. Of course, Wegner isn’t actually making this argument. He already assumes that free will is an illusion. He is just interested in this article in showing how that illusion might operate, which is point #2.

Although he describes the illusion of conscious choice as a magic show we put on for ourselves, this does not mean that he thinks conscious thought has no effect on behavior – that’s point #3 above. He simply doesn’t that deciding to pick up a cookie leads to your hand reaching out to pick up a cookie. In fact, it would be a pretty extraordinary claim that conscious thought (and the underlying brain processes) have no purpose, effect or use whatsoever. In that case, why did we evolve them? This question is answerable, but it would be hard to answer. The claim, then, is simply that the conscious decision to perform a behavior does not cause that behavior.

Suppose we agree that picking up a cookie is not caused by the conscious decision to pick up a cookie – which, just to be clear, I don’t – what does cause that conscious decision? Wegner does not get into this question, at least not in this chapter, which is a shame. In these last paragraphs I’ll try to describe what he might mean and what the consequences would be.

What would it mean if the conscious mind did cause cookie-picking-up? That depends on what the conscious mind is. Perhaps it’s an ethereal, non-corporeal presence that makes a decision, then reaches down and pulls a lever and the hand reaches out to grab a cookie. That would be similar to what Descartes argued for many centuries ago, but it’s not something many cognitive scientists take seriously now. The basic assumption – for which there is no proof but plenty of good evidence – is that the mind is the brain. Activity in your brain doesn’t cause your conscious mind to want a cookie, nor does your conscious mind cause brain activity. Your conscious mind is brain activity. If we assume that this is how Wegner thinks about the mind, then his hypothesis can be restated:

The part of the brain that is consciously deciding to pick up a cookie does not give orders to the part of the brain that actually gives the motor commands to your hand to pick up the cookie. The motor cortex gets its marching orders from somewhere else.

This is an interesting hypothesis, and I’m not going to discuss it in too much detail right now. What I am interested in is what does this hypothesis have to say about free will? I would argue: maybe nothing. If your decisions are made in your brain by a non-conscious part of your mind (of which there are many) and the conscious part of your mind turned out to simply be an echo chamber where you tell yourself what you’ve decided to do, would you say that you have no free will?

The real question becomes: what decides to pick up the cookie? Where is the ultimate cause? Lack of free will means that the ultimate cause is external to the person. They picked up the cookie because of events that occurred out there in the world. Free will means that the ultimate cause was internal to the person. Nothing in Wegner’s article is really relevant to distinguishing between these possibilities (again, this is not a criticism. That wasn’t what his article was about. It’s what my article is about).

The loss of a belief in a non-corporeal mind has left us with a dilemma. Nothing we know about physics or chemistry allows for causes to be internal to a person in the sense that we mean when we say “free will”. This makes many people feel that free will can only exist if there is a non-corporeal mind operating outside the constraints of physics. On the other hand, nothing we know about physics or chemistry allows for consciousness to exist, yet essentially all cognitive scientists – including, probably, Wegner – are reasonably comfortable believing in consciousness without believing in a non-corporeal mind.

In the 19th century, physicists said that the sun could not be millions of years old, much less billions of years old, because there was no known mechanism in physics or chemistry that would allow the sun to burn that bright that long. Although entire fields of thought – such as evolution or geology – required an old, old sun in order to make any sense of their own data, the physicists said “Impossible! There must be another explanation for your data.” Later, they discovered the mechanism: fusion.

In the early 20th century, there were chemists who said that the notion of a “gene” was hogwash, because there was no known chemical mechanism for inheritance in the form of a gene. The fact that mountains of experimental data could not be explained without reference to “genes” didn’t bother them. Then Watson and Crick found the mechanism in the structure of DNA.

We may be in the same situation now. We have an incredible amount of data that only makes sense with reference to internal causation – free will. Evolution, Wegner says, built the belief in free will into us. Liz Spelke and others have run fantastic experiments showing that even infants only a few months old believe in something akin to free will. The world makes very little sense if we don’t believe that our friends, colleagues and random people on the street are causing their own behavior. Or maybe we’re not in the same situation, and free will is truly a figment of our imagination.

Physicists were right about one thing: the sun hasn’t been burning for billions of years. It doesn’t burn. It does something else entirely. The real answer to the question of free will may look like rote, dumb physical causation – a snowball rolling down a hill. It may look very similar to Descartes’ non-corporeal soul. Or it may look very different from both.

Note: Wegner is a very engaging writer. If you are interested, most of his articles are available on his website.

Scientists prove that if money could buy happiness, you wouldn't know what to buy

Humans are miserable at predicting how happy or how unhappy a given thing will make them, according to Daniel Gilbert, professor of psychology at Harvard University and author of Stumbling on Happiness.

In many realms of knowledge, people aren't very good at predicting the future. It turns out that fund managers rarely beat the market, experts are poor predictors of future events, and just about everybody is impaired at predicting how those events will make them feel. It's disappointing that neither mutual fund managers nor talking heads are really earning their salaries, but it is astonishing that most people can't predict how winning the lottery or losing their job will make them feel.

You might be tempted not to believe it, but there are dozens of carefully-contrived studies that show just that, many of them authored by Gilbert. In one study that is particularly relevant to me, researchers surveyed junior professors, asking them to predict how happy would they be in the next few years if they didn't make tenure. Not surprisingly, they expected to be pretty unhappy, and this was true of both people who did eventually make tenure and those who did not. However, when actually surveyed in the first five years after the tenure decision, the self-described level of happiness of those who did not get tenure was the same as those who did get tenure. This is not to say that being denied tenure didn't make those junior professors temporarily unhappy, but they got over it quickly enough.

(As a side-note, Gilbert mentions that both professors who got tenure and those who did not were happier than junior professors. "Being denied tenure makes you happier," he joked.)

The Big Question, then, is why are we so terrible at predicting future levels of happiness? One possibility is that we're just not that smart. Either there is no selective pressure for this ability -- and thus it never evolved -- or evolution just hasn't quite gotten there yet. Rhesus monkeys can't figure out whether they want 2 bananas or 4 bananas, and humans don't know how happy 2 or 4 bananas would make them feel.

Another possibility is that there's actually an advantage buried in this strange behavior. One possibility is that it's important to be motivated to have children (happy!) or protect your children from harm (unhappy!) but once you've had children, there's not actual advantage to increased happiness and if your children die, there's no advantage in being unable to recover from it. (In fact, parents are typically less happy because of having children, and they live shorter lives as well. I don't know about happiness levels of parents who have lost children.)

Gilbert is unsure about this argument. For every evolutionary argument you can give me in favor of poor predictions of happiness, he argued, I can give you one against. For instance, you also predict that being rejected by a potential mate will decrease your happiness more unhappy than it actually does. Thus you might not approach potential mates and thus not have children. It's ultimately a hard question to test scientifically.

(If you are wondering where I got these Gilbert quotes, it's from his lecture to the first-year graduate students in the psychology program last Thursday.)

New study shows that music and language depend on some of the same brain systems

Music and language depend on the some of the same neural substrates according to researchers at Georgetown University.

The quick summary is that the authors, Robbin Miranda and Michael Ullman of Georgetown University, found that memory for musical melodies uses the same part of the brain as memory for words, and that "rules" for music use the same part of the brain as rules (grammar) for language. In the case of this particular experiment, musical "rules" means following the key structure of a song.

Why is this interesting? To the extent that anything about the mind and brain is well understood, music is particularly not well understood. I suspect this is probably partly because it took a long time for anybody to figure out how to study it empirically. Language, on the other hand, is fairly well understood. That is, it's maybe like physics in the 1600s, whereas the study of music isn't even that advanced.

If researchers are able to tie aspects of music processing to parts of the brain that we already know something about, suddenly we know a whole lot more about music. That's one exciting outcome.

The other exciting outcome is that, as I said, language has been studied scientifically for some time. This means that psychologists and neuroscientists have a whole battery of empirical methods for probing different aspects of language. To the extent that music and language overlap, that same arsenal can be set loose on music.


This shouldn't be taken as implying that nobody else has ever studied the connection between language and music before. That's been going on for a long time. What's important here is that these aspects of music were tied to one of the most complete and best-specified models of how the brain understands and produces language -- the Declarative/Procedural model.

Unfortunately, the paper isn't yet available on the Ullman website, but you can read a press release here.

Full disclosure: I was working in the Ullman lab when Robbin joined as a graduate student. You can read about some of my research with Ullman here.

Your brain knows when you should be afraid, even if you don't

I just got back to my desk after an excellent talk by Paul Whalen of Dartmouth College. Whalen studies the amygdala, an almond-shaped region buried deep in the brain. Scientists have long known that the amygdala is involved in emotional processing. For instance, when you look at a person whose facial expression is fearful, your amygdala gets activated. People with damage to their amygdalas have difficulty telling if a given facial expression is "fear" as opposed to just "neutral."

It was an action-packed talk, and I recommend that anybody interested in the topic visit his website and read his latest work. What I'm going to write about here are some of his recent results -- some of which I don't think have been published yet -- investigating whether you have to be consciously aware of seeing a fearful face in order for your amygdala to become activated.

The short answer is "no." What Whalen and his colleagues did was use an old trick called "masking." If you present one stimulus (say, a fearful face) very quickly (say, 1/20 of a second) and then immediately present another stimulus (say, a neutral face) immediately afterwards, the viewer typically reports only having seen the second stimulus. Whalen used fMRI to scan the brains of people while they viewed emotional faces (fearful or happy) that were masked by neutral faces. The participants said they only saw neutral faces, but the brain scans showed that their amygdalas knew different.

One question that has been on researcher's minds for a while is what information does the amygdala care about? Is it the whole face? The color of the face? The eyes? Whalen ran a second experiment which was almost exactly the same, but he erased everything from the emotional faces except the eyes. The amygdala could still tell the fearful faces from the happy faces.

You could be wondering, "Does it even matter if the amygdala can recognize happy and fearful eyes or faces that the person doesn't remember seeing? If the person didn't see the face, what effect can it have?"

Quite possibly plenty. In one experiment, the participants were told about the masking and asked to guess whether they were seeing fearful or happy eyes. Note that the participants still claimed to be unable to see the emotional eyes. Still, they were able to guess correctly -- not often, but more often than if they were guessing randomly. So the information must be available on some level.

There are several ways this might be possible. In ongoing research in Whalen's lab, he has found that people who view fearful faces are more alert and more able to remember what they see than people who view happy faces. Experiments in animals show that when you stimulate the amygdala, various things happen to your body such as your eyes dilating. Whalen interprets this in the following way: when you see somebody being fearful, it's probably a clue that there is something dangerous in the area, so you better pay attention and look around. It's possible that subjects who guessed correctly [this is my hypothesis, not his] were tapping into the physiological changes in their bodies in order to make these guesses. "I feel a little fearful. Maybe I just saw a fearful face."

For previous posts about the dissociation between what you are consciously aware of from what your brain is aware of, click here, here and here.

Monkeys know their plurals

Anybody who reads this blog knows that I am deeply skeptical of claims about animal language. Some of the best work on animal language has come from Marc Hauser's lab at Harvard. Recently they reported that rhesus monkeys have the cognitive machinery to understand the singular/plural distinction.

First, a little background. Many if not most scientists who study language are essentially reverse-engineers. They/we are in the business of figuring out what all the parts are and how they work. This turns out to be difficult, because there are many parts and we don't really have the option of taking apart the brains of random people since they usually object. So the task is something like reverse-engineering a Boeing 747 while it's in flight.

There are many different ways you could approach the task. Hauser tries to get at language by looking at evolution. Obviously, rhesus monkeys can't speak English. Just as obviously, they can do some of the tasks that are necessary to speak English (like recognizing objects -- you have to recognize something before you can learn its name). Any necessary components of language that non-human animals can successfully perform must not be abilities that evolved for the purpose of language. If you can figure out what they did evolve for, you can better understand their structure and function. So the next step is perhaps to figure out why those particular abilities evolved and what non-human animals use them for. This ultimately leads to a better understanding of these components of language.

That is one reason to study language evolution in this manner, but there are many others (including the fact that it's just damn cool). If you are interested, I suggest you read one this manifesto on the subject.

Back to the result. Nouns in many languages such as English can either be singular or plural. You couldn't learn to use "apple" and "apples" correctly correctly if you couldn't distinguish between "one apple" and "more than one apple". This may seem trivial to you, but no non-human animals can distinguish between 7 apples and 8 apples -- seriously, they can't. In fact, some human groups seemingly cannot distinguish between 7 apples and 8 apples, either (more on that in a future post).

So can rhesus monkeys? Hauser and his colleagues tested wild rhesus monkeys on the beautiful monkey haven of Cayo Santiago in Puerto Rico. The monkeys were shown two boxes. The experimenters then put some number of apples into each box. The monkeys were then allowed to approach one box to eat the contents. Rhesus monkeys like apples, so presumably they would go to the box that they think has more apples.

If one box had 1 apple and the other had 2 apples, the monkeys went with the two apples. If one box had 1 apple and the other had 5, the monkeys picked the 5 apple box. But they chose at random between 2 and 4 apples or 2 and 5 apples. (For those who are familiar with this type of literature, there are some nuances. The 2, 4 or 5 apples had to be presented to the monkeys in a way that encouraged the monkeys to view them as a set of 2, 4 or 5 apples. Presenting them in a way that encourages the monkeys to think of each apple as an individual leads to different results.)

This suggests that when the monkeys saw one box with "apple" and one with "apples," they knew which box to choose. But when both boxes had "apples," they were at a loss. Unlike humans, they couldn't count the apples and use that as a basis to make their decision.



Full disclosure: I considered applying to his lab as a graduate student. I am currently a student in a different lab at the same school.

Caveat: These results have not been formally published. The paper I link to above is a theory paper that mentions these results, saying that the paper is under review.

How do universities choose professors? The survey is in.

Several studies have looked at university hiring practices. Search committees (committees in charge of filling vacant positions) around the country were surveyed, and the results are in.

The studies, published in Teaching of Psychology, looked specifically at psychology departments, so it may not generalize well to other departments. However, as a PhD student in psychology, it's the department I care most about.

The older of the two, written in 1998 by Eugene Sheehan, Teresa McDevitt & Heather Ross, all then at the University of Northern Colorado, had several interesting results. One was that teaching was valued more highly than research, which was surprising to me. I would like to know how this broke down by type of institution. Some schools are said to highly value teaching (i.e., Oberlin, Swarthmore) while others are said to highly value research (i.e., Harvard, Georgetown). Since they only got back 90 complete surveys, they probably couldn't do do an analysis breaking down by "teaching schools" and "research schools."

Luckily, R. Eric Landrum & Michael A. Clump from Boise State University wondered the same thing and published an answer in 2004,. They compared public and private universities as well as undergraduate-only departments against programs with graduate programs as well. Private schools were significantly more likely to care about teaching experience, whereas public institutions were significantly more likely to care about research-related issues and the ability to get grants. Undergraduate-only departments were similarly much more conserned with teaching-related issues, whereas programs with graduate students cared more about research- and grant-related issues.

Another interesting result was that the Sheehan study found that the job interview was the most important factor in deciding between interviewed candidates. This is not surprising in the sense that we all know that the interview is very important. It is surprising because it's well-known that job interviews are very poor indicators of future performance, and you would think a university psychology department would know that. The later study did not consider interviews vs. CVs vs. letters of recommendation.


Now for the data.


Sheehan et al. 1998:

The factors that the search committees considered when deciding who to interview are listed below, in order of most important to least:

Letters of recommendation
Fit between applicant's research interest and department needs
Experience teaching courses related to the position description
General teaching experience
Quality of course evaluations
Quality of journals in which the applicant has published
Number of publications
Potential for future research
Quality of applicant's doctoral granting institution
Awards for teaching


The factors considered when deciding among interviewed candidates were, in order of most important to least:
Performance at interview with search committee
Performance during colloquium (i.e., the "job talk")
Fit between applicants research interests and department needs
Experience teaching courses related to the position description
Performance during undergraduate lecture
Candidate's ability to get along with other faculty
General teaching experience
Letters of recommendation
Candidate's personality
Performance at interview with chair



Landrum & Clump, 2004:

Factors more imporant to private schools vs. public:
It is important that applicant publications be from APA journals only.
Teaching experience at the undergraduate level is important for applicants.
Research experience utilizing undergraduate undergraduates is important for our applicants.
Experience in academic advising is important for the successful job applicant.
It hurts an applicant if he or she does not address specific courses listed in the job adertisement.
In our department, teaching is more important that research.
Teaching experience.
Previous work with undergraduates.

Factors more important to public schools vs. private:
Our department has an expectation of grant productivity.
Faculty in our department need to receive grants in order to be successful.
In our department, research is more important than teaching.
Quality of publications.
Potential for successful grant activity.


The comparison between undergraduate-only and undergrad/grad programs revealed many more significant differences, so I refer you to the original paper.

Another non-human first

First ever Economist obituary for a non-human:

http://www.economist.com/obituary/displaystory.cfm?story_id=9828615

Not your granddaddy's subconscious mind

To the average person, the paired associate for "psychology," for better or worse, is "Sigmund Freud." Freud is probably best known for his study of the "unconscious" or "subconscious". Although Freudian defense mechanisms have long since retired to the history books and Hollywood movies, along with the ego, superego and id, Freud was largely right in his claim that much of human behavior has its roots outside of conscious thought and perception. Scientists are continually discovering new roles for nonconscious activities. In this post, I'll try to go through a few major aspects of the nonconscious mind.

A lab recently reported that they were able to alter people's opinions through a cup of coffee. This was not an effect of caffeine, since the cup of coffee was not actually drunk. Instead, study participants were asked to hold a cup of coffee momentarily. The cup was either hot or cold. Those who held the hot cup judged other people to be warmer and more sociable than those who held the cold cup.

This is one in a series of similar experiments. People are more competitive if a briefcase (a symbol of success) is in sight. They do better in a trivia contest immediately after thinking about their mothers (someone who wants you to succeed). These are all examples of what is called "social priming" -- where a socially-relevant cue affects your behavior.

Social priming is an example of a broader phenomenon (priming) that is a classic example of nonconscious processing. One simple experiment is to have somebody read a list of words presenting one at a time on a computer. The participant is faster if the words are all related (dog, cat, bear, mouse) than if they are relatively unrelated (dog, table, mountain, car). The idea is that thinking about dogs also makes other concepts related to dogs (i.e., other animals) more accessible to your conscious thought. In fact, if you briefly present the word "dog" on the screen so fast that the participant isn't even aware of having seen it, they will still be faster at reading "cat" immediately afterwards than if "mountain" had flashed on the screen.

Mahzarin Banaji has made a career around the Implicit Association Test. In this test, you press a key (say "g") when you see a white face or a positive word (like "good" or "special" or "happy") and a different key (say "b") when you see a black face or a negative word (like "bad" or "dangerous"). You do this as fast as you can. Then the groupings switch -- good words with black faces and bad words with white faces. The latter condition is typically harder for white Americans, even those who self-report being free of racial prejudice. Similar versions of the test have been used in different cultures (i.e., Japan) and have generally found that people are better able to associate good words with their own in-group than a non-favored out-group. I didn't describe the methodology in detail here, but trust me when I say it is rock-solid. The interpretation that this is a measure of implicit, nonconscious prejudice is up for debate. For the purposes of my post here, though, this is clearly a nonconscious prejudice. (Try it for yourself here.)

Vision turns out to be divided into conscious vision and nonconscious vision. Yes, you read that correctly: nonconscious vision. The easiest way to tell this for yourself is to blindfold a single eye. You probably know that you need two eyes for depth perception, but with one eye blindfolded, the world doesn't suddenly look flat. (At least, it doesn't for me.) You may notice some small differences, but to get a real sense of what you have lost, try playing tennis. The ball becomes nearly impossible to find. This is because the part of your vision that you use to orient in space is largely inaccessible to your conscious mind.

An even more interesting case study of this -- though not one you can try at home -- is blindsight. People with blindsight report being blind. As far as they can tell, they can't see a thing. However, if you show them a picture and ask them to guess what the picture is of, they can "guess" correctly. They can also reach out and take the picture. They are unaware of being able to see, but clearly on some level they are able to do so.

It is also possible to learn something without being aware of learning it. My old mentor studies contextual cueing. The experiment works like this: You see a bunch of letters on the screen. You are looking for the letter T. Once you find it, you press an arrow key to report which direction the "T" faces. This repeats many hundreds of times. Some of the displays repeat over and over (the letters are all in the same places). Although you aren't aware of the repetition -- if asked, you would be unable to tell a repeated display from a new display -- you are faster at finding the T on repeated displays than new displays.

In similar experiments about language learning, you listed to nonsense sentences made of nonsense words. Unknown to you, the sentences all conform to a grammar. If asked to explain the grammar, you would probably just say "huh?" but if asked to pick between two sentences, one of which is grammatical and one of which is not, you can do so successfully.

Actually, an experiment isn't needed to prove this last point. Most native speakers are completely ignorant of the grammar rules governing their language. Nobody knows all the grammar rules of their language. Yet we are perfectly capable of following those grammar rules. When presented with an ungrammatical sentence, you may not be able to explain why it's ungrammatical (compare "Human being is important" with "The empathy is important"), yet you still instinctively know there is a problem.

And the list goes on. If people can think of other broad areas of subconscious processing, please comment away. These are simply the aspects of the unconscious I have studied.

You'll notice I haven't talked about defense mechanisms or repressed memories. These Freudian ideas have fallen out of the mainstream. But the fact remains that conscious thought and perception are just one corner of our minds.