Smart People Things

Lectures and lecture notes are at the bottom of the page.

I didn't sleep at all Sunday night. The room was insanely cold and all I have is a thin white sheet and a fuzzy blue paperweight to sleep under. Nonetheless I got up early and was ready for a day full of lectures and entirely too much food.

The breakfast was catered, which was awesome. I had like six bowls of honey nut cheerios and some fruit before I couldn't make myself eat any more. Never put large amounts of food in front of me, because I don't have any idea when to stop. We had a safety talk given to us by a legit-looking chief of police or something. He talked a lot about where to go to get legit Philly Cheesesteak. Apparently, we have to say "cheesesteak wit whiz" to make it legit. It will have fried onions and... cheez whiz. For real. This should be interesting. Also, we got rape whistles and lip balm. It was pretty rad. 

LECTURE TIME: SEE LECTURE NOTES AT BOTTOM. They are long. 

David Brainard then gave us a 2 hour demonstration of various optical illusions and used them to teach us about how the mind perceives things. We used our experiences to discuss how we thought the brain worked and why it worked that way (evolutionarily, etc). I got a video of one of them. Try and close one eye when you watch this and see if you can see the window oscillating. It is actually turning around in circles. With a tube attached to it.

We also did a lot of stuff involving how we perceive color in a relative rather than absolute manner. The same shade of yellow will look totally different on a background of gray than on a background of bright yellow. However, the mind is really good at tuning our perception of color if there are other colors to make it relative to, which is pretty cool. That concept is called color constancy. 

Alright! So after all our talks we had a catered dinner which was, once again, too delicious for my own good. Instead of talking about normal people things and taking a break after all the lectures, we decided it would be a great idea (at our table of about 8 people) to talk about the complexities of grammar, how to measure the complexity of a given language (we decided it was a fruitless undertaking), the definition of what a word is (this question is the bane of all morphologists and especially of myself), and the possibility of free will (I am a staunch determinist and base my beliefs on what little I know about particle and quantum physics, and yes, those things are relevant to the discussion of free will). After dinner, we all went on a campus tour. Photo time!

Oldest building on campus. So pretty!

That thin in the middle of the books? That's a dolphin.
Yep. I thought the same thing. 

THESE ARE THE FRESHMEN DORMS. WTF. 

The pond in the middle of a beautiful garden-esque area on campus.

After doing some sight-seeing around campus, and visiting a gelato place near campus (I got BEER flavored gelato, it blew my mind) me and a few other people in the conference decided to try and hit up a couple bars before calling it a night. We ended up walking for like... an hour, and then finally made a decision on someone's GPS about where to go. The bar we were trying to get to was located on top of a restaurant, except we had no idea how to get up. Someone in the group tried to open a door that was on the sidewalk level, but it turned out to be someone's apartment, and someone yelled down furiously, "HEY!" and then we all picked up the pace and walked to the corner without looking back. We finally found a door which led into the restaurant but clearly had a staircase leading up to the second floor. The bar was called... Fuzzy's? Funky's... something. I can't remember. I'll look it up later. Anyway, it was super tiny but extremely busy. It was about as local as you could get. 

There was a bluegrass band playing and the bartender had little homemade cardboard signs for all the beers he served, so naturally a couple people got PBRs while the Russian girl in our group ordered a local beer that tasted like soy sauce. We moved into the empty room on the other side of the... apartment complex? And we hung out for a few minutes talking about neuroscience, school, and getting to know everyone while they finished their beers. 

Finally we headed out to look for a decent bar. Long story short, we didn't find any. We walked into about 5 separate bars, most of which were either full/full of angry-looking people, or just scary looking in general. We ended up going back before we'd actually accomplished anything but decided the next time we wanted to go out to a bar, we would stay away from the area north of campus and just go downtown to South Philly. 

On to the smart people things!

Communication, cognition, and the evolution of language: Field studies of monkeys and apes. 

Robert Seyfarth

The first lecture was fascinating. Dr. Seyfarth began by showing an evolutionary tree of the primates with a time frame for how long ago certain species split off from others. According to the normal rate of gene variation and the current variance in the genome of several species, we know that chimps and gorillas split off about 5 mya. Language, however, didn't develop until what Seyfarth believes to be a maximum of 400,000 years ago. He is interested in studying other ape species for evidence of social behavior and how language might have originated in primitive ancient hominid species or ancient human species. Additionally, he is also interested in finding out how non-language-using animals see the world. It has been thought that language shapes and dictates the way in which humans can think, and that without language, many higher function cognitive processes are not possible. However, Seyfarth (and I as well) believes that even without language, representational thought it possible. The problem, then, is how to prove it. 

Seyfarth has been doing research in Africa for quite some time, in the range of 15-20 years, on baboons. He has collected shit tons of data (not scientific enough? How's 'metric shit tons of data'? Putting things in metric makes them scientifically better, yeah?) on how they interact with each other both inter- and intra-familially. Additionally, he has followed one particular group for the years he's been doing research and has recordings of each one making different calls. 

Baboons, vervet monkeys, macaques, and Diana monkeys all have something in common. They have specific calls for different predators. Baboons have calls for eagles, snakes, and leopards. The calls are also usually binary between males and females. That is, since primates are sexually dimorphic, their calls are subsequently different when compared between genders. Regardless, males can understand the female calls and vice versa. He played the audio clips for the various calls and noted that no matter what, if they play one of those calls, the baboons always react in the same way. For the eagle call, they all call back to the original caller, look overhead, and try to hide in bushes and jump out of the trees to get on the ground. For the snake call, they also call back, check the ground, gather to where the snake has been spotted, and stare it down. For the leopard call, they call back and run up into the trees. Mind you, the calls are invariable between members of the same gender. For the eagle, it is always a very specifically timed and pitched grunt, etc. In this way, the other baboons can recognize what the spotter has seen and is trying to communicate. But how do we know it's not just instinct instead of representational thought?

There is something called the habituation/discrimination paradigm. In humans, 2-day-old infants respond to speech sounds through various facets of their body language. One of the ways to measure these responses is to measure how many times the infant suckles on a bottle while listening to speech as opposed to silence. When infants aren't hearing any human speech, the suckling rate is low. When they hear a new sound, they increase the rate at which they suckle. For example, if you say "ba ba ba," they will suckle faster. However, if you repeat it too many times, they get bored and stop responding because they are already aware of the stimulus and don't find it interesting, or something. But if you alternate insert a "pa" in the middle of a bunch of "pa"s, the suckling rate goes back up because of the new stimulus. (One extra thing we know from that is that 2-day-old infants are aware of voicing in stops, which is fascinating.) In the same way, he wanted to test if this happened to baboons, but with their calls. 

Presumably the calling back feature was instinctive. That is, the baboons had essentially no choice in whether or not they called back after a predator call, or at least, that was the idea. So Seyfarth played the call over and over again at 2 minute intervals or something along those lines, and after only a few calls, the baboons stopped calling back. This is because they have representational thought, as the theory goes. They already were aware that there were eagles in the area due to the first call, and after several calls from the speaker they knew they no longer needed to communicate that idea since they were all aware of it. I thought that was pretty cool. It also lines up with my ideas about animal cognition and the origins of language. I don't think at all that language came before thought, and I think many would agree with me. Without thought, there's not really much to communicate beyond basic ideas for survival and there would then be no pressure to develop complex language.

The second part of the lecture covered the social aspects of baboon life. Baboons are very highly aware of how high up on the social chain each member of their group is. Members of families are always grouped together. Actually, hang on. This only applies to females. Males haven't been studied yet because they actually move between groups to mate with everything they see. Anyway, females are highly aware of the social order. If an alpha female's status is somehow surpassed by one of the family members of the beta family, the entire beta family become the alpha family. There are specific calls that are only used to communicate with members of a higher social status family, and some calls which are only used to communicate to members of a family of lower status, and when status changes, the baboons change how they call to those members and vice versa. Also, when the wrong calls are played on speakers (as in the call of a low status family using an inappropriate call on a member of a family of higher status), the whole group looks around in confusion and moves around to investigate. Also, they have calls for reconciliation after fights and are context specific. These things are obviously not language, but they are nonetheless interesting and point to how language may have started before humans were around. Additionally, the baboons not only recognize the predator calls from members of their species, but also the calls that birds and meercats use to signal the presence of those predators. How cool is that?

Also, I read a paper last semester about macaques and how they process their species-specific calls in their brains. Evidently, there is a homologue to Wernicke's area in their brains that lights up like a Christmas tree on an fMRI when presented with their species-specific calls. Whoah! The message to take away from all this is that thought must have come before language and it doesn't make sense to hypothesize the other way around. However, the new question is how complicated that representational thought can be without language to give it parameters and structure. Apparently, complex social relationships can be encoded just fine without language. So, where's the limit?

The time course of speech comprehension

Delphine Dahan

Dr. Dahan talked about how the brain begins interpreting spoken words and how we use the initial sounds in a word to make predictions on what the rest of the word will be. Using eye tracking technology, she presents her subjects with a grid of four objects and plays sound files to them to see where they will look. For instance, there will be a picture of a boob, a boot, a block, and a tractor. Boob and Boot have the same initial consonant-vowel combination, which block only has the same first sound, and tractor has none in common with them. She plays the sound file for "boob" and waits to see where the subject will look. In a nutshell, what she found is that the initial sounds of a word make us hypothesize about the next few sounds and prime us for the understanding of the words in those hypotheses. It is not until well after the word has been finished that we can reevaluate what we heard in the case of an incorrect hypothesis. In the case of not being able to hear part of a word, context in the next few hundred milliseconds and sometimes even for the duration of several new words is taken into account to retroactively fill in the parts that were missed. She also noted that our understanding of spoken language is very heavily influenced by the goal the listener has. That is, we can sometimes hear what we want to hear. 

Secondly, the goal you have affects how you use information in real time. If you are trying to choose between two pictures that represent similar words after a word is played back to you, the frequency of those words in the language has no effect on how fast you will recognize those words. For example, if you heard the word 'ant,' your brain will actually recognize it much faster than 'aardvark' because you have heard 'ant' many more times than you've heard 'aardvark.' When you are asked to evaluate sensory information quickly, though, word frequency has huge effects on how quickly you are able to recognize words. I don't know if I explained that very well, but it's a blog, so whatever. 

Takeaway message: There is inertia present in speech perception - activation continues to last according to word frequency even after disambiguation occurs. 

I don't know if I'm allowed to link to the pdf's we were given but if you want to read any of these studies please leave a comment or send me your e-mail or something.