Students' Book Reviews

This week's project... to write a book review. 

Here are the links to the pages of the students who have done it:


Julia          Bejansu          Hosung          Laureen          Minji   

Jet             Chloe              Veronika  

Book Review - On Intelligence

This week's class assignment is to read a book and write a critical review of it, so here is mine.



←　This is the book I'm reading at the moment.

It is a science book written by Jeff Hawkins, a theoretical neuroscientist who first worked in Silicon Valley developing personal computing software and hardware, then later returned to his first love - neuroscience - to seek answers to the question of how that lump of matter in our heads determines who we are.

Since I am a cognitive scientist with particular interest in theoretical neuroscience, this book is especially relevant to me.  However, I would recommend it to anyone who is interested in understanding the details of how the brain works at at the neuronal level.

Hawkins opens by explaining how in his early years he came to work for a number of computing companies because he had been unable to find any place where he could pursue theoretical neuroscience.  There were people studying neuroscience, but he thought they were too focussed on the cells rather than the functions that they perform (cognition, intelligence, etc), and there were computer scientists and artificial intelligence researchers, but they were too focussed on the functions without paying any attention to how these functions could be occurring in an animal (via neurons).  Because there was no field of study that suited him, he applied for a job and ended up designing several pieces of pattern recognition software that were very popular and made his name.

With this under his belt, he went back to study neuroscience and eventually founded his own non-profit research facility, now called the Redwood Centre for Theoretical Neuroscience, where he continued to develop the theory of neocortex information processing that he has laid out in this book.  He now has his own company that aims to implement his theories of the neocortex in useful artificial intelligence applications - Numenta.

The theoretical part of the book starts by explaining why he believes the neocortex to be the seat of intelligence.  As far as our nervous systems are concerned, it is the cortex that differentiates us (the mammals) from the other animals (lizards, birds, fish, amphibians, insects...).  Animals with and without a neocortex are much the same in terms of their repertoire of motor behaviours; they run, jump, bite, chew, swim, etc.  However, Hawkins argues that it is not this kind of behaviour that exemplifies intelligence, but rather the more abstract and unobservable behaviours, like remembering the layout of the environment, planning actions that may not receive a reward for some time, or developing complex social relationships (not to mention language).  The neocortex does this by discovering patterns in the things that we experience, and sequences in these patterns, and then by using these sequences of patterns to make predictions about what we will experience.

The most fascinating part of the book is the middle few chapters, in which Hawkins goes into detail about how the pattern and sequence detecting could occur in the columns that make up our cortex.  It all comes down to hierarchies.  The environment is sensed by our eyes, ears, etc, and simple patterns are detected by low levels of the hierarchy (early processing in the cortex).  Then the next level of the cortical hierarchy looks at these and finds more complex patterns, and then the next layer does the same to find even more complex patterns.

Above you can see how one of these hierarchies could be understood in terms of function, while to the right you can see how neocortical columns can implement these hierarchies.

I actually read about this book a few years ago and wanted to get my hands on a copy at the time.  I am glad that I didn't though, because I have recently been doing a lot of research into hierarchical processing and learning in different parts of the brain (cortex, basal ganglia and cerebellum), and reading this book at this time in particular has provided answers to many questions that I have only just started asking, as well as guiding me towards many more questions that I would like to find the answers to from now.

Right now I'm sitting in the classroom trying to motivate my students to write the second instalment of their blogs and thinking about MOTIVATION...

It seems that none of them is really wanting to write their blog, and one or two are looking like I'm asking them throw a baby out of the window!  But when we started this project on Monday and followed it up over the next two days, they all quite enjoyed it.

However, it does take effort, and I think this is where the problem lies.
Though everyone wants to learn English and has demonstrated their motivation by travelling all the way to England at no small expense to themselves or their family, when it comes to putting a certain type of effort into a task, the motivation fizzles.

What is this 'certain type of effort'?

I've noticed a big difference between motivation to do closed tasks and motivation to do open tasks.

By closed tasks, I mean small, simple, easily achievable tasks such as choosing the correct word to go in a space in the text, choosing which category something goes into, matching the things on the left with the things on the right, etc.  In English as a second language learning, these kinds of tasks abound.  They are quick to complete, easy to understand and can be easily found or created for any grammar point of vocabulary exercise.

An open task, on the other hand, is one where the learners have to generate everything themselves.  For example, creating a role-play, practising, and performing it in front of the class.  There is a lot more freedom involved in this kind of exercise since you must create a story, make it interesting, consider the kind of language you will use, if it is correctly constructed, if the vocabulary is appropriate, if you are interacting suitably, and then do the performance at the end.



These two kinds of task are approached very differently by students, whether they be a highly self-motivated group of adult learners or a bunch of teenagers more interested in chatting with their peers than learning anything.  When you put a sheet of writing with some gaps in in front of someone, they have a very well-defined task to do which can be marked at the end for an immediate assessment of performance.  People dive into this kind of task.  A few weeks ago I had a class of Italian teenagers who were mostly out of control, but if I gave them a gap-fill exercise, they would knuckle down for about 10 minutes of frantic concentration.  Which was great!!  The problem was that the task is so narrow and short lived that it required very little imagination and engagement.  You have to give very little of yourself to a gap-fill - it's all already there on the paper.

I got to thinking about this seriously a few years ago when one of my students who had become a good friend started at university.  While studying English at my school he had worked hard and done many exercises as well as being very sociable and using English with lots of different friends.  However, when he started at university his English ability suddenly soared.  The reason for this was that he was being given projects to complete which required him to research thoroughly before compiling the information he had gathered into his own piece of writing or presentation.  This required a great deal of continuous and focussed effort - he had to immerse himself in the task, often for over a week.  He was experiencing new vocabulary that he had to remember in order to understand what he was reading and hearing, and which he would use in his work.  The grammar that he had studied before but never had much chance to use was required in order to express complex ideas (it is predicted that sales will have exceeded 1 million units by January).

While closed tasks do require effort, and students may spend hours and hours at home doing task after task, it is a lazy effort!  But this desire to put in effort in the easiest way possible is not just a problem for students.  I know that I do it too, in all areas of my life.  It all boils down to the path of least resistance.  Imagine you decide to go hiking in some mountains, like the ones shown in this picture.  You want to get from one side of the mountains to the other - do you go over the peak of one of the mountains, or do you go through the valley between the two?  Of course, the valley.  You have decided to go walking in the mountains, but you plan the route to be as easy as possible.  (It may be that your plan is to go to the top of the mountain, but even in this case, you would plan the easiest route that got you to the top). This is all well and good; it would be stupid to plan to use the most difficult route!  But what about if your goal is to improve your hill climbing skill and power?  In that case, selecting the path of least resistance is not the most beneficial strategy.  This is a lot like studying English - it's not so much about the time you put in, or even the effort, but the kind of effort.  It should be via tasks that challenge you to encounter and generate as much (new) language as possible.

To this end, I've recently been trying to set my students weekly projects to complete, from presentations to story writing to creating a blog about something they are interested in.  I'm finishing writing this in the cafe near my house now.  I asked them to finish writing their blogs for homework this evening... we'll see tomorrow whether they have done it!

My students' blogs...



Being a mean and horrible teacher, I've forced my students to each create their own blog about something interesting for them.







Here they all are:

Ana               Julia               Bejansu               Hosung               Tanya   
 
Laureen         Minji              Veronika             Patrick                Valseta

Jet                Chloe             Haneul                 Marie

There's some good stuff here.  Have a look :)

For this first post I'm going to talk about something that came up in the staff room at work today...
weird past tenses.

The example: glide

What is the past of glide?   

Possibilities:  glided, glid, glode, glad.

Each of these is supported in some way.  A quick look around the internet shows that 'glided' is the most commonly accepted one at the moment, but it used to be 'glode' and 'glad' at different times.  'glid' seems to sound right to many people.

The point of this is not really to find out about this verb in particular, but to look at other weird past forms.  Fortunately, English is weird, so if we just think of a couple of syllables that are found in verb endings, such as '-aid' and '-eik', then generate verbs from here and conjugate to the past, we should find some interesting variety:

slide-->slid                           make-->made

hide-->hid                            take-->took

ride-->rode                          shake-->shook

bide-->bided                        brake-->braked

glide-->glided                       break-->broke

side-->sided                         wake-->woke

vie-->vied                             bake-->baked

confide-->confided               fake-->faked

divide-->divided                   rake-->raked

"This morning, after I wook, I tade a shower and mook breakfast.  I rid my bide to work.  It tade about half an hour because of the heavy traffic.  In the classroom we brook at 10 o'clock, and I drought a cup of coffee.  This wook me up and energose me."

Each of the verbs in the paragraph above has been changed in a way that is similar to the other verbs in its group.  But think about what you did when you tried to understand what I meant by "wook", "tade" and "mook".  A first glance (I glence at the words(?)) at the words alone doesn't help us to understand the intended verb.  We then have to try to think what other possibilities the word is similar to: wook, took, snuck (see bottom of page*), cook, f!!k.  The position of the word in the sentence tells us it's a verb, so we can instantly eliminate many possibilities.  The first letter, w, gives us a strong clue.
But, actually, it would be almost impossible to guess without the context.  This highlights that we do not work out irregular past verbs by applying a rule (like the +ed of regular verbs - talk+ed), but instead have a very specific memory of each one.

To show how this works in your brain, I've drawn a little diagram:

You have already selected the word that you want to use, and now you need to find the past tense of it.  I've used capital letters to show that this is a variable, so WORD can be the present tense of any verb you are looking for.  This is sent to a kind of look-up table containing all of the irregular verbs along with their past tenses.  If WORD is found in the list, its corresponding PAST is used.  Alternatively, if WORD is not found in the list, we deduce that this must be a regular verb and simply stick "ed" on the end (or just "d", or maybe replace "y" with "ied").  For example, if the word is eat, we send it to the look-up table, find that eat appears, and has the corresponding past tense of "ate", so we use this.  If our word is "plonk", however, after searching through the look-up table and finding that it is not there, we just at "ed" to the end to get "plonked".

We can do a little test to see an example of this working in our heads.  Make up a new verb and think about what its past form would be:
What is the past of "gibbysnum"?
It feels natural to say "gibbysnummed" (doubling the "m" because the "u" has a short sound).  This is because we don't find this verb in our internal look-up table.



So now the question of why, as an English speaker, we feel unsure about whether the past of "glide" is "glided" or "glid".  I'm not fully sure about this, but I would guess it feels to have something to do with the similarity in both meaning and pronunciation of "glide" to "slide".  The past of "slide" is, without a doubt, "slid".  For this reason, it feels as if the past of "glide" should follow suit and be "glid".  But remember that our brain is not made of look-up tables; it's made of neurons, and those neurons can be strongly or weakly connected to other neurons.  A word we are very sure of will form a very strongly connected network, while one we are not so sure of will form a much more weakly connected network, and a feeling of uncertainty or discomfort when it is used.  For me, it doesn't really matter whether I say "glided" or "glid" - both of them feel slightly odd.  I suggest that our brain finds a very weak entry for "glide --> glid" in the look-up table, and thus is confused about whether to use this or to disregard it and go with the regular +ed rule.

Another question that stems from this is why we even have irregular verbs (or nouns, adjectives, phrases, etc) at all.  It clearly takes more memory to remember a whole set of irregular past forms, and this memory could be put to another use if we just standardised all English verbs to be regular ones.  Stephen Pinker talks about this here.  What it boils down to is that verbs that are used more often have a habit of evolving particular (irregular) past forms.  This makes sense.  If you use it often enough, it kind of deserves to be given special treatment!
Some other examples of different kinds of words that behave in the same way are:

• ordinal numbers - first, second (not onest and twost)
• fractions - half, quarter (which would also be onest and twost)
• plurals - men, women, children (not mans, womans, childs)
• adjectives - better, worse, elder (not gooder, badder, older (though older also exists with a different meaning))

https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEhsk0ejPskCwu_LCqS-54e4sOavbHl5kZZ5T4Dl8s01QxPY-DMP5ijFh-qqj30Mc536aRFvqTzNwjbpjBIHmb-7T7Px_m-I23hIgm4Jz8lXYbiG8k0oIZ-2MXOfPZ3jvdeO7jAlzHVgavf_/s1600/Verb+tenses-+Irregular+verbs.png

**Snuck is one option for the past tense of sneak.  Apparently, it is not the correct option.  "Sneaked" is officially more correct.  However, 'official correctness' is not necessarily the best way to judge what most people use.  googlefight.com allows you to compare the number of times two phrases appear on websites, and if you put t"sneaked" vs "snuck" into googlefight.com, "snuck" easily comes out on top.  Moreover, I read that it is one of the few verbs that is changing from a regular "-ed" past tense form to an irregular form.  In modern English, this change usually happens the other way (from irregular to regular).