A view to the gallery of my mind

Peter Singer on Effective Altruism

2013-05-20T18:04:23Z

I don’t usually do link posts, but I think that this video on Effective Altruism is one that everyone should watch.

xuenay @ 2013-05-15T12:01:00

2013-05-15T09:01:07Z

Gah, in case you were wondering about the old entries popping up on your friends list / RSS: apparently fixing a few broken links on the WordPress copy of those posts made my LJ crossposter plugin repost them here. Tried deleting some of them and hitting the "do not add to friends list / RSS" button on a few others, hopefully that should fix it...

Jasen Murray on tranquility meditation

2013-05-15T07:44:22Z

Since the page that I previously used to link to for a description of how to do tranquility meditation has died, I’m reposting the instructions here. I found them very useful in getting started with meditation, and they seemed to work better for me than any other instructions. Original credit for writing them goes to Jasen Murray.

—-

Very brief summary:

Use either the breath or metta as your object of meditation. Do not focus all of your attention on the object, merely maintain constant awareness of it while also experiencing your entire physical body. You will experience tension in the body (particularly in the head) especially when distractions such as thoughts arise. Let the distraction go, neither following it nor trying to suppress it, while releasing any tension that you notice. Release tension by maintaining uninvolved awareness of the tension, reminding yourself that it is ‘happening on its own’ (I explain this in the detailed instructions below). Keep on doing this. You will pass through the jhanas. Move through them by the same process of releasing tension while maintaining awareness of you object. Eventually, after hanging out in the 8th jhana for a while, a complete cessation of perception and feeling will occur. When perception and feeling return, you will clearly see how attachment is produced and thus be able to release it. The first time you do this is ‘stream entry.’ Repeat until fully enlightened.

More detailed Summary:

You pick an object of meditation. Bhante V. prefers metta (loving-kindness) first and the breath second. He says both he and his students have found metta to produce the fastest results.

If you choose the breath:

Be aware of your breathing. Do not lock your attention on a particular subset of body sensations such as those at the nostril or abdomen, just be aware of whether you are inhaling or exhaling and the length of each inhalation and exhalations.

Now, as you breath in, experience your entire physical body. As you breath out, experience your entire physical body.

If you choose metta:

Start by remembering a time when you were happy until you can feel that happiness, perhaps as a warmth in your chest. Once you can feel it, wish yourself happiness, perhaps in the following manner: “May I be happy. May my mind be peaceful and calm. May my mind be filled with joy. May my mind be clear, and alert.” Really feel the wish, radiating loving-kindness toward yourself. Use this feeling as your object of meditation. If it starts to fade, make the wish again. If you choose this object, the feeling will transform into the other Brahma Viharas as you pass through the jhanas. That’s fine, let it. (more below).

Either way, you will notice tensions, particularly in your head when a distraction (such as a thought) arises.

These tensions arise whenever there is attachment to a sensation. So long as there are such tensions, there is attachment.

In the normal state, there are too many layers of mental activity to see the low-level process of attachment with sufficient clarity such that it can be released.

The purpose of this meditation is to gradually relax your body and mind while maintaining clear, alert mindful attention until all perception ceases in a moment of cessation. When perception returns, you will get a clear glimpse of ‘dependent origination’ and thus see how attachment occurs so that you can stop doing it. I don’t have a good model of this yet, but I’m working on it.

People seem to have a difficult time describing how they relax these tensions. They often say things like “Relaxing this tension is not really a matter of ‘doing’ anything. It is the ‘doing’ that is the source of the tension. Let go of all doing.” There’s something to that, but it is easy to misinterpret. The confusion comes from the mistaken belief that the feeling of ‘effort’ or ‘control’ is produced by the processes responsible for generating the relevant behavior in the same way that the experience of color is produced by the processes responsible for sight. Those feelings are actually just the result of more attachment to sensations. They are produced by the same processes that resist information (in this case, my guess is the resistance is to the fact that experience is happening on it’s own and thus cannot be controlled and that there is no solid permanent ‘you’).

So, maintain awareness on the breath, remind yourself that all experience is happening on its own and cannot be controlled and simply be aware of the tension while leaving it be. It will eventually feel like there is an outer layer to the tension that is softening, breaking up and melting away, leaving a smaller, lighter tension behind. Repeat the process.

If thoughts arise, tension will arise along with them. Let go of the thought, even mid sentence and release the corresponding tension in this way.

If you keep this up, you will get more and more relaxed and pleasant feelings will begin to arise in your body. These signal the beginning of 1st jhana and will grow into an intense joy.

The different levels of relaxation are called ‘tranquility’ jhanas. I do not know if or how these correspond to the ‘absorption’ jhanas or the ‘vipassana’ jhanas. You move through them by continuing the processes of letting go of any tension that you notice. It goes something like this:

1st jhana – intense joy throughout the mind and body, maintaining attention on meditation object feels effortful. Remember that the feeling of effort is just tension and let go of it.

2nd jhana – more intense joy throughout the mind and body, effortless attention on meditation object. Eventually the intensity of the joy will feel a bit too coarse and you will notice some attachment to it. Release this tension.

3rd jhana – less intense comfort/happiness throughout the mind and body. Eventually the feeling of comfort/happiness will seem to coarse and you will notice attachment to it. Release this tension.

4th jhana – equanimity, very peaceful and still, even unpleasant sensations do not seem to be a problem. The next tension to release comes from attachment to distinctions/diversity.

5th jhana – base of infinite space. ‘physical’ sensations take on a formless character, distinctions are not held on to and the feeling of the body seems to dissolve out into the space surrounding ‘you.’ If metta was your object, it transforms into Karuna (compassion) here. This is experienced as radiating compassion in all directions into infinite space (hence the ‘infinite compassion’ of a buddha). Something like continuity of ‘consciousness’ is still being held on to.

6th jhana – base of infinite consciousness. The illusion of a separate, continuous ‘observer’ consciousness breaks down and each seems to be aware of itself. This is difficult to describe, but very cool. It seems as if everything in your sense fields is a tiny bit of ‘you’ looking back at itself. Karuna now transforms into mudita (sympathetic joy). Something like ‘form’ or consciousness is still being held on to…I bit shakier on the next transition as I’ve only experienced it a few times.

7th jhana – nothingness. The black or blank space around sensations becomes more prominent than the sensations themselves. Very peaceful. Mudita now turns into upekkha (equanimity). Perception, if only of nothingness, is still being held on to.

8th jhana – neither perception/feeling nor yet non-perception/feeling. I’m not sure about this one. I may not have experienced it yet. People describe it as a moving back and forth between minimal perception and very minimal perception in which there is still awareness of some kind. This is the same regardless of the object you started with. Some say that you can only tell that you were in the 8th jhana rather than asleep by looking back on your memory of the time spent meditating.

There really can’t be any further instruction at this point because there’s too little going on. You just continue practicing. Eventually, perception and feeling cease completely for some amount of time. When they return, you get a glimpse of what Bhante V. calls ‘dependent origination’ and ‘nibanna.’ This is ‘path’. ‘Fruition’ in this model is apparently something different (though I’m not yet sure what) that comes a bit later after more practice. There are various levels of enlightenment (the 4 paths) that correspond to the number of times you’ve experienced cessation followed by fruition.

Although releasing tension is an important part of the instructions, it is critical that you don’t get carried away and go looking for tension. The instruction to ‘look for’ some aspect of your experience usually leads people to carry out the same kind of operation that produces tension – trying to force your experience to conform to your expectations. Just stay with your object of meditation ( but not too tightly) and let go/allow any other sensations to happen.

Originally published at Kaj Sotala. You can comment here or there.

Dissecting edugames: iCivics.org

2013-05-08T08:37:13Z

The Serious Games Market blog showcases a number of interesting edugames, and I thought that I should try some. One of the posts linked to an interesting-sounding site called iCivics.org, which has a number of educational games that are designed to teach kids about the way the US government works.

Some of the games were relatively good. Others were dreadful enough that even with a designed playtime of half an hour or even less, I couldn’t bring myself to play them to the end. One in particular reminded me of old point-and-click adventure games in the worst possible ways: not allowing the kinds of actions that would have felt the most logical, ultimately leaving “try using everything on everything” as the only way to proceed. Few were anywhere near as pleasing to play as a good entertainment game. But still, there were some entertaining ones, and several that would be interesting to analyze.

I have to keep running around this map to get laws executed. The distance between places matters.

The first game that I tried was Executive Decision, in which you take the role of the President of the United States. In practice, this involves running around a minimap of Washington DC, where letters keep popping up in your mailbox at the White House. Mostly, the letters contain new laws that Congress has sent for you to sign. If they’re good laws, you should sign them and then take them to the relevant government building to be implemented. If they contain bad parts, you should flag those parts as bad and veto the bill, in which case they will soon come back without the bad parts. Occasionally you’ll be asked to fly to a meeting in a foreign country, to go over to the Pentagon to choose the most appropriate branch of the military to deal with a conflict, or to go to the Congress to hold a speech to bolster support for your pet issue.

All of this requires running around the map, and running around the map takes up time, which you have a limited amount of. So Executive Decision is basically a resource-management game where you need to plan your moves as efficiently as possible, in order to maximize the amount of actions you can take. This kept me moderately entertained as I played it, and it was short enough that I didn’t have the time to get bored, though I wouldn’t call it a crowning moment of fun.

Now, what was the game intended to teach, and what does it actually teach?

Fortunately for us, iCivics gives us teacher aids to be used in conjunction with the games. There’s an “Executive Command Post-game Powerpoint” with a a number of questions which one is supposed to know the answers to after having played the game:

What does the Chief of Staff do? (He aids you in your duties as President)
What is the purpose of the State of the Union address? (To identify key issues to focus on (set the agenda))
Why must the President go back and speak to the Congress again? (To raise support for the issues on the agenda)
What do you do if you approve of a bill and want it to become law? (Sign it)
If you disagree with a bill you should… (…veto it)
Can you sign only part of a bill into law? (No, you must sign the whole bill or veto the whole bill)
What is it called when you deliver a law to someone else to carry out? (Delegating it)
When war breaks out, what must you do as President? (Command the armed forces)
When the President acts as our representative to other countries, it is called… (…diplomacy)
Bonus question: What is the name of the President’s plane? (Air Force One)

The game probably does an adequate job of teaching most of those. I’d expect the items 3-6 and 8 to be remembered the best, since they are things that you spend the most time doing. The others require somewhat more attention – you might forget that the guy giving you advice on what to do was called the Chief of Staff, or that the part at the very start of the game where you choose your pet issue was the State of the Union speech.

But the game also teaches a number of other things, which were probably not intended by designers. To quote Vaniver on Less Wrong:

People should be expected to learn the game, not the reality, and that will especially be the case when the game diverges from reality to make it more fun/interesting/memorable. If you decide that the most interesting way to get people to play an interactive version of Charles Darwin collecting specimens is to make him be a trainer that battles those specimens, then it’s likely they will remember best the battles, because those are the most interesting part.

One of the research projects I got to see up close was an educational game about the Chesapeake; if I remember correctly, children got to play as a fish that swum around and ate other fish (and all were species that actually lived in the Chesapeake). If you ate enough other fish, you changed species upwards; if you got eaten, you changed species downwards. In the testing they did afterwards, they discovered that many of the children had incorporated that into their model of how the Chesapeake worked; if a trout eats enough, it becomes a shark.

Here are some unintended lessons from the game:

The President needs to physically visit different government departments in order to delegate the task actually of implementing various laws. While doing this, it is important for the President to plan his route in a way that lets him visit many departments in a very brief time.
It’s better to be an Education President than a Security President, because the Department of Education is physically closer to the White House than the Department of Homeland Security is, so it takes less time to run between the White House and the Department Education.
Some of the laws that the President gets to sign are obviously stupid. The President may choose to veto these, in which case he’ll soon get to sign a new version of the law without the stupid parts. Bad laws are always obviously bad, and Congress never overrides the veto of a bad law. (I don’t know what would have happened if I’d tried vetoing a good law, though the Teacher’s Guide says that Congress is likely to override you if they try to veto your declaration of war.)
Good laws have noble-sounding goals, and the President never needs to worry about unintended consequences. Nor is the monetary cost of a law an issue. Even though some laws are titled “deficit reduction” laws, the only difference is in their name, and in the fact that they belong to the Treasury’s jurisdiction.
Congress will randomly invite the President to hold a speech for them, and the President can win over their support for his pet issue by holding a speech filled with platitudes each time, until they start sending him nothing but bills related to that issue.
Having a foreign country declare war on the US is inconvenient, because then you have to keep running over to the Pentagon to tell your generals how to deal with the constant acts of aggression, when you’d rather be promoting your pet issue.
The President can end wars by waiting to be invited to meetings in other countries and then flying over to those meetings sufficiently many times.

You may think that I’m being facetious here. But these really are things that one learns when playing the game, because you need to learn them to play the game well. For the same reason, they’re things which are quite likely to stick to the player’s mind the most. Of course, the player also remembers the context of the game, and may be able to use other knowledge to figure out that which parts are only specific to this game and untrue in real life. So hopefully nobody learns many untrue things from Executive Command. Also, many of those points, such as the possibility of Congress overriding a Presidential veto, are addressed in the other games.

But the point is that the game mechanics are a large part of what the player’s focus and attention are on. If the mechanics are divorced from the actual educational content of the game, that means that part of the game’s educational potential is wasted, since part of what the players learn while playing the game is useless. On the other hand, if mastering the game mechanics is the same thing as learning the educational content, then a much larger part of what the game teaches is the thing that you actually want to teach. As a loose analogy to physics, you could talk about the efficiency of an edugame: how much of the “learning energy” that goes into a game is converted into “useful learning” and how much gets wasted? In other words, of all the things that a player learns while playing the game, how much is actually the kind of learning that we want them to learn, and how much does the player need to discard as an artificial quirk of the mechanics? As in physics, we can probably never get a 100% efficiency, but we can try to get a pretty good ratio. Of course, if you are happy with your players only learning simple things like “the President can veto bills he doesn’t like”, then you might be content with accepting even a large amount of wasted learning.

A good example of a game where even a moderately high amount of “wasted learning” is probably fine was Do I Have a Right? Together with Branches of Power (which I’ll cover in a moment), this game was one of my favorites. It has you running your own law firm, and in many respects it’s similar to various real management games such as Theme Park. In the beginning you only employ one lawyer, but as you proceed in the game, you can hire more, level up your hires, buy them better equipment, purchase various extra furniture to your office to make customers happier, and place newspaper advertisements to attract more customers. The game is divided to a number of days, and after each day, your achievements are chronicled in a newspaper with a style of humor that reminds me of the writing in the various SimCity games. If not for the fact that it’s quite short, and that it won’t take very long to acquire all the upgrades, this could have been a real entertainment game.

Do I Have a Right?

The learning component comes from various customers walking into your office and explaining their case. For example, one customer said the following: “I was found guilty of littering and paid an $80 fine. Now they want to put me on trial and fine me again for the same littering. Do I have the right to stop this trial?” You are provided a list of various civil rights as outlined in the US Constitution, and consulting it, you see that the Fifth Amendment prohibits double jeopardy. If you have a lawyer who specializes in that right, you lead the customer to that lawyer and have the case taken care of. If you don’t, you can tell the customer to come back later. Sometimes customers also think that they have rights which the Constitution doesn’t actually give them, in which case you can turn them away.

This is a quite nice way of incorporating a learning element into the gameplay in a way that feels natural and uncontrived, and is actually effective at teaching the player to recognize what the different rights are and which of them are relevant for various cases. This is much more fun than rote memorization would be, and doesn’t even feel like you’re working to learn. Also, you need to level up your lawyers and hire more of them in order to have every possible Constitutional right covered by your company, which gives you another in-game reason to spend a lot of attention looking at the various rights.

One thing that I also found clever was that you can earn a small number of extra points by clicking on the important ideas in the client’s story – in this case, the game considered “put me on trial and fine me again” an important idea, and awarded five points for clicking on that part of the text. Clicking on unimportant parts produced pictures of unhappy faces, and might have cost some points. Since identifying the right in question involves recognizing the key elements of the story and disregarding the irrelevant ones (such as this being about littering in particular), and a young player might initially be at a loss about what the relevant parts are, it’s a nice touch to put in an extra feedback mechanism that provides immediate assistance on that in particular.

Another game which I enjoyed and thought was interesting was Branches of Power, in which you are trying to push different agendas into laws that are accepted by each of the three branches of government, in the following order: the executive branch promotes an idea, the legislative branch makes the idea into a bill, the executive branch signs it into law, and the judicial branch resolves any court cases that challenge it. There are ten different issues that you can try to promote, and if you can make each of them into a law that survives judicial review, you win.

The interesting thing is that you can control the actions of each of the branches – but you can only control one of them at a time, and the rest keep acting independently in the meanwhile. So while you’re running the legislative branch and crafting bills in a way that will pass both Congress and Senate, the President is running around deciding whether to support your bills, and the Supreme Court is running around deciding the legality of your bills. If you want to be sure that the President will actually sign your bills, you can either engineer them in such a way that it’s in his best interests to sign them, or you can jump to take control of him and make sure that he does sign them. (At least I think that’s how it works, since I never actually crafted a bill which wouldn’t have been in the President’s interests to sign.) And of course, if you do need to take control of the president, that means that in the meanwhile, the legislative branch might craft the bills in ways that you wouldn’t want them to.

In addition to being a pretty novel and interesting mechanic, which I don’t remember having seen anywhere else, this is also quite educational. Not only does it teach people about the different and partially opposing incentives that the various parts of government have, it also helps convey a more generally useful lesson: that people are more likely to do what you want them to do if you manufacture situations where their interests align with yours. That’s a very general rule about politics and human interaction overall… subtly taught in a simple Flash game in a manner which, again, does not even make the learner realize that they’re being taught!

So overall, I was pretty impressed with several of these games, and felt that even some of the ones that were perhaps less successful (like Executive Decision) still had many useful lessons for edugame design in general. They’re still not very great games in terms of entertainment value, but they did give hints of how one could make an edugame that had great entertainment value. I still haven’t played all the games at iCivics, so I may do another post on them if I find more interesting ones.

Originally published at Kaj Sotala. You can comment here or there.

Videogames will revolutionize school (not necessarily the way you think)

2013-04-01T15:42:52Z

A lot of the hype around educational games centers around “gamification”, and using game techniques to make the boring drilling of facts into something more fun. Which would be a definite improvement, but I don’t think that it’s ambitious enough.

Instead, let’s start by considering the question: what kind of things should education teach, and why?

Typically, school has taught facts. Bad school systems only focus on teaching facts and testing the extent to which they have been memorized, good school systems also make at least some effort to test the ability to apply them. Unfortunately, it is hard to test the ability to apply something, but easy to test whether it has been memorized. But the ability to memorize something says nothing about whether it was understood, so we get laments like the following:

For example, consider college freshmen who have taken their first college-level physics class, passed it with good grades, and can write down Newton’s laws of motion. [...] Lots of studies have shown that many such students, students who can write down Newton’s laws of motion, if asked so simple a question as “How many forces are acting on a coin when it has been thrown up into the air?” (the answer to which can actually be deduced from Newton’s laws) get the answer wrong. Leaving aside friction, they claim that two forces are operating on the coin, gravity and “impetus,” the force the hand has transferred to the coin. Gravity exists as a force and, according to Newton’s laws, is the sole force acting on the coin when it is in the air (aside from air friction). Impetus, in the sense above, however, does not exist, though Aristotle thought it did and people in their everyday lives tend to view force and motion in such terms quite naturally.

So these students have entered the semiotic domain of physics as passive content but not as something in terms of which they can actually see and operate on their world in new ways. There may be nothing essentially wrong with this, since their knowledge of such passive content might help them know, at some level, what physics, an important enterprise in modern life, is “about.” I tend to doubt this, however. Be that as it may, these students cannot produce meanings in physics or understand them in producerlike ways.

They have not learned to experience the world in a new way. They have not learned to experience the world in a way in which the natural inclination to think in terms of the hand transmitting a force to the coin, a force that the coin stores up and uses up (“impetus”), is not part of one’s way of seeing and operating on the world (for a time and place, i.e., when doing modern physics). — James Paul Gee, What Video Games Have to Teach Us about Learning and Literacy, pp. 22-23

The issue that Gee is really highlighting is the fact that although students have learned some words, the mental model of physics that they have is one of folk physics, not scientific physics. A mental model, as I’m using the term, is a mental simulation of some set of laws of cause and effect that exist in the world. If you have a well-formed model of physics, you can ask yourself questions like “how would this object behave under the influence of these forces” or “what forces are acting on this object in this situation”, study your model, and get an answer back.

A mental model doesn’t need to be about a formal and easily-defined domain such as physics: most aren’t. Whenever you hear somebody make a claim that makes you think “that doesn’t sound quite right”, the claim has violated the predictions of your existing models. Models can be very extensive or very limited: a young child might know that on ordinary days of the week, mother will return from work at 5 PM, but have no other idea of what “work” means.

But the important thing about mental models is that they simulate parts of reality. And reality is a dynamic process, where things are constantly changing in ways that we wish to predict. Simulations of reality, in order to be accurate, must then be processes as well.

For example, simply knowing that an object in free fall on Earth will accelerate at 9.81 m/s per second isn’t very useful if one only understands it as a string of English words. Physics students need to understand that it is actually a description of a dynamic process, a characterization of the way that a falling object behaves over time. They haven’t really learned the meaning of this before they can use the information to imagine and predict what happens if they drop a rock from their balcony. Although we want our students to learn dynamic models and to understand processes, for the most part we have been forced to communicate those models via static representations (writing, pictures) that require highly non-trivial mental effort to translate into dynamic models.

But we now have computer programs, which can actually function as dynamic representations. A computer program is a process by its very nature, and it can in principle made to represent almost arbitrary other processes. We’re no longer just forced to use a static representation of a dynamic process when we can instead give a student something dynamic to play with. This should hopefully make it much easier to turn the learned content into a dynamic mental model from the start.

This also suggests that we should reconsider the very things that we are teaching in school – today’s curriculum has been shaped by what’s possible or easy to teach and test using only static representations, but computer programs allow for much more dynamic teaching and testing. Instead of telling a student, “you’ll pass if you can write an essay that lists the reasons why the Roman Empire fell”, a teacher could instead say, “you’ll pass if you play this computer game where you’re the ruler of Rome and succeed in preventing its fall”. The essay basically only tests memorization, while the game – if it has been well-designed – tests the ability to actually apply the knowledge, to correctly identify the reasons for the empire’s fall and to then counteract them.

But the reasons for the fall of the Roman Empire, too, are something that’s taken from the current curriculum, which we might want to reconsider entirely. What kinds of models do we want to teach our children, and why? Perhaps what we’re really after is a more general notion of why different societies might collapse, and what kinds of dynamics are in play, using the Roman Empire as a case study that we start out from. Or maybe we decide that this isn’t valuable enough in comparison to the other things that we could be teaching, and we decide to throw away the whole topic.

Why do so many children (and adults!) dislike school? Probably because static representations are often bad at teaching dynamic models, and many teachers might not even realize that that’s what they’re supposed to be teaching. This creates the feeling that school learning is boring, unless the student is already talented at turning the static explanations into dynamic models. Which isn’t to say that writing is all bad: it’s much easier and faster to create, and if the learner can connect the writing to content that’s already in the learner’s head, it can be a very effective way of deepening and broadening one’s understanding. When you already know have a good model of the domain in question, even static materials can be easy to translate into dynamic components that you can add to and integrate with your existing model. The problem only occurs when there isn’t anything that the learner could connect the material to. James Paul Gee compares reading game manuals with reading science texts:

But, in any case, the problem with the texts associated with video game—the instruction booklets, walkthroughs, and strategy guides—is that they do not make a lot of sense unless one has already experienced and lived in the game world for a while. Of course, this lack of lucidity can be made up for if the player has read similar texts before, but at some point these texts originally made sense because the player had an embodied world of experience in terms of which to situate and spell out their meanings.

The same thing is most certainly true of the sorts of texts that show up in learning content areas like science and math in school, especially in the later grades, high school, and college. A biology textbook does not make a lot of sense unless and until one has experienced and lived in the world of biology as practice for a while. And again, this lack of lucidity is mitigated if the student has already read a good many similar texts. However, at some point these texts also originally made sense because the student had an embodied world of experience (in reality or, at least, simulated in his or her mind) in terms of which to situate and spell out their meanings.

When I give talks on video games to teachers, I often show them a manual or strategy guide and ask them how much they understand. Very often they are frustrated. They have no experience in which to situate the words and phrases of the texts. All they get is verbal information, which they understand at some literal level, but which does not really hang together. They cannot visualize this verbal information in any way that makes sense or makes them want to read on. I tell them that that is how their students often feel when confronted with a text or textbook in science or some other academic area if they have had no experiences in terms of which they can situate the meanings of the words and phrases. It’s all “just words,” words the “good” students can repeat on tests and the “bad” ones can’t. (What Video Games Have to Teach Us about Learning and Literacy, pp. 102-103.)

To be fair, there are many school teachers who really do focus on teaching a genuine understanding of content, and make a good job of it – the traditional school system isn’t all bad. I had such teachers on all grades, and often they were successful at their pursuits. But they were still required to assign grades, and it is hard to genuinely and fairly test a student’s understanding in some domain if you cannot actually place them in that domain. Ultimately, they too assigned grades based on things like tests and projects, which are fundamentally static measures of understanding and have a hard time measuring dynamic understanding. The need to assign grades, and to measure performance by some fair (and thus, in a pre-computer era, mostly static) method, crucially handicapped efforts aimed at really improving the understanding of the students.

Of course, current edugames aren’t really set up to deliver a new kind of educational experience. Rather, many are designed as aids for teaching the informational content of the existing curriculum – which is rather backwards, when you think of it. We’d really want our students to learn dynamic models but we can’t teach or test those directly, so we teach and test them on static facts instead – and when we finally do get an instructional aid that could teach and test dynamic models, we try to fit it into the mold of teaching facts, because that’s what they’ll be tested on!

It probably isn’t a coincidence that so many edugames are about mathematics, because math is the subject that’s the closest to being tested in a dynamic way, and is thus the most naturally suited for computer instruction.

Another issue that we aren’t yet very good at making games that teach dynamic models. Ian Bogort has coined a term for the teaching of dynamic models: procedural rhetoric. Just as verbal rhetoric is the art of persuading and teaching by using spoken words, while visual rhetoric does the same using pictures, procedural rhetoric persuades and educates by using a dynamic model. Let’s look at his argument in a little more detail.

One way of defining a game is as a collection of rules that define various consequences for the actions that a player takes. Shoot at the alien, the alien loses hit points and gets angry at you. Thus, when somebody plays a game, they are placed in a microcosm where the laws of cause and effect have been defined by the designer of the game, and they need to learn and internalize those laws in order to succeed at the game. In effect, the game designer can be seen as making a statement about the kinds of causal laws that exist, and the player comes to understand that position via their own experience, having discovered and experienced the laws for themselves.

Now the causal laws of many video games are mostly only applicable within the video game itself, and few people think of applying them in any other context. But games could present broader arguments. One of Bogort’s examples is The McDonald’s Videogame, in which

The player controls four separate aspects of the McDonald’s production environment, each of which he has to manage simultaneously: the third-world pasture where cattle are raised as cheaply as possible; the slaughterhouse where cattle are fattened for slaughter; the restaurant where burgers are sold; and the corporate offices where lobbying, public relations, and marketing are managed. In each sector, the player must make difficult business choices, but more importantly he must make difficult moral choices. In the pasture, the player must create enough cattle-grazing land and soy crops to produce the meat required to run the business. But only a limited number of fields are available; to acquire more land, the player must bribe the local governor for rights to convert his people’s crops into corporate ones. More extreme tactics are also available: the player can bulldoze rainforest or dismantle indigenous settlements to clear space for grazing (see figure 1.1). (Ian Bogort, Persuasive Games: The Expressive Power of Videogames, Kindle Locations 721-728.)

Presumably, the game designers hope that by playing the game, the player comes to see the laws of cause and effect that push corporations towards unethical behavior by sometimes making it more profitable than ethical behavior. Having personally experienced a situation where those laws operated, the player can apply their experience more generally, and start to be more suspicious about the behavior of not only McDonald’s, but any corporation which is operating under similar laws of cause and effect. Of course, the player may reject the argument and feel that the position that the game designers are advocating is a flawed one – but that is the case with all rhetoric.

The ultimate goal of procedural rhetoric in the service of education is to give the player a genuine understanding of the laws operating in a game, in a way that allows for that understanding to be generalized to similar situations in real life, while also being fun. That’s a very tough challenge, and we don’t really know how to do it yet. On the other hand, there are already games that can be used for a similar purpose despite not being explicitly educational, such as by having students try to evolve a humanity-eradicating plague in Plague Inc. and then talking about the lessons about evolution that this teaches [1 2]. Such an approach is probably the most effective one for now, but it could be much improved if we had games designed expressly for the task.

If we did, we could truly revolutionize schooling. Throw away exams and grades, and just give kids games to play with, and have discussions about the games afterwards. If we wanted to have some measure of how far the students had progressed, just look at how much they had achieved in the game. Of course, massive changes of this kind are going to face a lot of resistance, so for now edugame designers who agree with these goals should be working towards more gradually shifting the system in this direction.

Another important skill, which both Gee and Bogort emphasize, is the ability to study models critically. It’s not enough that we teach students different models of how the world works – they also need to learn to evaluate the merits and plausibility of different models. What simplifying assumptions are being made? How does this model mesh together with others? How can one validate the claims made by a model? And so on.

Some of this can be taught by simple means, such as having the students play a model and then ask them to look for differences between it and reality. But there’s also a certain beauty in the discovery that the process by which models are created, evaluated, and argued for is itself a process, and can thus be modeled as a game whose laws and caveats can be learned by playing it. My Fundamental Question game, still in early planning stages, is one attempt to teach critical evaluation of models by showing some of the ways by which information can be unreliable.

And then, of course, the students will be asked to critically evaluate the model about critically evaluating models. Maybe we’ll even have a game about that.

Originally published at Kaj Sotala. You can comment here or there.

Book review: What Video Games Have to Teach Us about Learning and Literacy

2013-03-26T14:56:21Z

What Video Games Have to Teach Us about Learning and Literacy. James Paul Gee. Palgrave Macmillan.

(This review is based on the first edition of the book.)

This book was a very nice discussion about video games in light of various academic theories of learning. I particularly liked this point:

“The fact that human learning is a practice effect can create a good deal of difficulty for learning in school. Children cannot learn in a deep way if they have no opportunities to practice what they are learning. They cannot learn deeply only by being told things outside the context of embodied actions. Yet at the same time, children must be motivated to engage in a good deal of practice if they are to master what is to be learned. However, if this practice is boring, they will resist it.

“Good video games involve the player in a compelling world of action and interaction, a world to which the learner has made an identity commitment, in the sense of engaging in the sort of play with identities we have discussed. Thanks to this fact, the player practices a myriad of skills, over and over again, relevant to playing the game, often without realizing that he or she is engaging in such extended practice sessions. For example, the six-year-old we discussed in the last chapter has grouped and regrouped his Pikmin a thousand times. And I have practiced, in the midst of battle, switching Bead Bead to a magic spell and away from her sword in a timely fashion a good many times. The player’s sights are set on his or her aspirations and goals in the virtual world of the game, not on the level of practicing skills outside meaningful, goal-driven contexts.

“Educators often bemoan the fact that video games are compelling and school is not. They say that children must learn to practice skills (“skill and drill”) outside of meaningful contexts and outside their own goals: It’s too bad, but that’s just the way school and, indeed, life is, they claim. Unfortunately, if human learning works best in a certain way, given the sorts of biological creatures we are, then it is not going to work well in another way just because educators, policymakers, and politicians want it to.

“The fact is that there are some children who learn well in skill-and-drill contexts. However, in my experience, these children do find this sort of instruction meaningful and compelling, usually because they trust that it will lead them to accomplish their goals and have success later in life. In turn, they believe this thanks to their trust in various authority figures around them (family and teachers) who have told them this. Other children have no such trust. Nor do I.” (pp. 68-69)

This part struck a particular chord in me since I had just read an opinion piece making exactly such an argument: that not all parts of education can be made to be fun, and that “it’s important to realize early on that mastery often requires persevering through tedious, repetitive tasks and hard-to-grasp subject matter”. I found myself somewhat annoyed with that position, but couldn’t formulate my exact reasons for why.

After reading What Video Games Have to Teach Us about Learning and Literacy, things became much clearer in my head: part of the value of video games is that they can make a subject feel interesting and meaningful on its own. Once a person has encountered a topic in an interesting context, they will be much more likely to find the topic interesting in other contexts as well. Personal example: when we were first taught probabilities in high school, me having read The Hitchhiker’s Guide to the Galaxy made the subject matter feel more interesting, even though our exercises made no mention of the Infinite Improbability Drive.

Yes, children should learn that mastering valuable skills often requires repetitive practice… but if we want them to actually learn, we should also be teaching them how to experience that practice as interesting and meaningful, and as something that is helping them get better in a field they care about. What we should not teach children is the attitude that much of learning is dull, pointless and tedious, detached from anything that would have any real-world significance, and something that you only do because the people in power force you to. Unfortunately, many traditional school systems are very successful at teaching exactly this attitude, and only the kids who have sufficient trust in various authority figures to make the learning feel meaningful manage to avoid it – and even they only succeed partially.

What Video Games Have to Teach Us about Learning and Literacy also talks about the impact of identities on learning, and by associating school with games and school success with success in fun games, we could help learners more easily develop identities as good students, helping make the learning process feel more meaningful – even when they had to tackle tasks that weren’t as inherently fun.

I also liked the discussion of the fact that if a person reads a text that covers a topic the person doesn’t have much experience of, it can be very hard to understand exactly what it is that the text is saying. The words aren’t clearly connected to the concepts that they are discussing. And much of school learning does consist of having the students read elaborate discussions of concepts that they don’t necessarily have much experience of. Even when the students do successfully memorize the rough content of the writing, they are not likely to understand it or be able to apply it very well.

In contrast, somebody playing a video game is actively engaged in the content of the game, free to experiment around with it. Well-designed video games also involve a gradual and natural progression where the players naturally obtain various skills required for playing the game. Once they have beaten the game, it is certain that they have acquired those skills to a far greater extent than if they had just read and memorized the game manual. Games provide for active learning, and the way that a game proceeds from easy initial levels to challenging late-game levels forces a player to constantly acquire additional skills while also practicing the basic skills, in an organic and natural fashion.

The main flaw of the book is that while it provides an excellent discussion of academic theory on learning, its discussion of the way the theory relates to games is at times somewhat superficial. A more detailed analysis of the content of some games in light of the theory would have been nice.

Originally published at Kaj Sotala. You can comment here or there.

Fan fiction libraries

2013-03-20T17:58:55Z

Today’s analogy: a fan fiction writer sets their story in a world created by someone else, and thus has the opportunity to use both characters and world/story elements that were originally created by others. Especially for novice writers, this can be a boon, as they can focus on some sub-area of fiction writing without needing to create everything from scratch. That experience will help them later understand how to create their own elements and how such elements need to fit together with everything else.

But even experienced writers might prefer to just focus on telling some particular kind of story, without needing to design the whole world and characters from the ground up. In that case, using an existing world can make things much easier. Of course, often a writer will want to tell a story which isn’t quite a perfect fit for an existing world. If that’s the case, the borrowed elements will need to be tweaked, with the creator replacing them with altered versions which inherit most of the elements’ original properties but change some things. If this would require too many changes, it can be simpler to just create an entirely new world, instead of spending a lot of effort forcing existing elements into a purpose they’re not really a good match for.

Still, there are advantages with using existing elements. If the writer doesn’t modify them, and the elements behave as others expect them to behave, the story becomes compatible with a vastly larger set of stories, all taking place in the same universe. Other stories can in turn easily build on the contributions that this story made. It also becomes easier for others to read the story, as those others will already be familiar with the expected behavior and properties of the reused elements and can take advantage of their existing knowledge.

In other words, a writer who’s writing fan fiction is like a programmer who’s using existing libraries.

(And although I’ve only spoken about fan fiction so far, obviously the real world is the biggest standard library of them all. Fanfic writers sometimes get flak for being uncreative and just playing in someone else’s world: but at the same time mainstream writers have no shame in recycling the ideas of others, such as when they brazenly use concepts like “people” and “cars” without bothering to come up with their own objects.)

Originally published at Kaj Sotala. You can comment here or there.

Why I’m considering a career in educational games

2013-03-12T17:27:36Z

Your honor, the prosecution would like to argue that the way the world is currently organized with regard to education vs. entertainment doesn’t really make any sense.

Exhibit #1: the award-winning strategy game XCOM: Enemy Unknown, a 2012 reboot of an old strategy game franchise. As of this writing, I have logged 94 hours of play on this game, much of that due to getting so addicted that I couldn’t quit even when I wanted to. As a result of playing, I have learned numerous pieces of utterly useless trivia. For example, I know that the easiest enemies in the game (sectoids) have three hit points on difficulties “Easy”, “Normal” and “Classic”, meaning that they can be killed with a single grenade, but they upgrade to having four hit points on “Impossible”. I also know that soldiers who are assigned to the “Sniper” class initially begin with the “Headshot” special ability. At the next level, one may choose between the “Snap Shot” and “Squadsight” special abilities, out of which the “Squadshot” special ability is clearly far superior. And so on.

Exhibit #2: the classical education system. Even when I have a genuine interest in the topic that I’m supposed to be studying, it often involves an active expenditure of willpower to get myself to do so. The human brain is most strongly motivated by frequent and rapid feedback, but traditional education tends to involve rather long feedback cycles. Maybe there are exercises that are due once a week, but it can also be the case that you’re required to spend a considerable time reading a book and listening to lectures before you’ll get a single piece of feedback in the form of your exam grade. Much of the education is delivered in a form that keeps the learner passive: lectures (a terrible way of learning) or books, rather than the kind of interactivity that would really be engaging. When there are exercises, they often feel pointless, boring and unfun.

Thus, games are doing a far better job of teaching things than the education system is. The defense is about to present witnesses who will argue that traditional education is slowly but surely reforming, shifting towards better methods of teaching. The defense will no doubt point out that the prosecutor himself is currently taking a university course based on the problem-based learning paradigm. The prosecutor hastens to grant these points. However, they do not alter his point, which is that such reforms aren’t taking things far enough. All, or at least most, of education could be done via games that were as addictive and enjoyable as traditional games.

Next, the defense will present arguments that educational games are all bad, and that you can’t really make a good one. I request that the honorable judge dismiss this argument as sheer nonsense. We have already shown that enjoyable games can teach quite a lot of things, such as the statistics of various aliens. I would also point out that I began being taught English in school around (I think) the third grade, but I never learned much in school that I wouldn’t already have learned from other sources, computer games being some of the most notable ones. Finally, part of my understanding of history comes from playing games such as Civilization, Colonization, and Europa Universalis. Games are already teaching us countless of things: it’s just that we might want to adjust the things that they are teaching us.

And let us not forget exhibit #3: DragonBox. About a month ago, I witnessed a kid who was around eight years old blaze through ~80 levels of the thing in just a few hours and have a lot of fun doing so, and afterwards she had no trouble solving the equation ax/5=a/b on pen and paper. Also, her older brother was complaining that he wanted to play, too, which was the first time that I’ve ever seen kids argue over who gets to solve first-degree equations. Before this, I also witnessed a four-year old solve about a hundred of such equations playing the game, though with considerably more coaching. This was the game that really opened my eyes for the possibilities of educational games.

But DragonBox, as fantastic it is for teaching the rules of algebra, does nothing to teach the reasons for the rules. It doesn’t impart a deep understanding of why math works the way it does. Because of that, it remains a useful tool for teaching algebra, but only a tool – it doesn’t work as a stand-alone teaching method. You can’t learn math from only playing DragonBox, the way I pretty much learned the basics of English from only playing video games.

What kind of a game would let you learn math only from playing it? Let’s cast away all modesty for a while, and think big instead. Why was math invented in the first place? Part out of intellectual curiosity, part for solving practical problems. Geometry was created to help with things such as planting the crops and building houses. A game which was really good at teaching math might put you in an imaginary world where no abstract math existed yet, and would task you with inventing ways for improving the world. You would invent math from the first foundations, for the same reasons people originally invented it – to solve the concrete problems threatening the kingdom. You’d see your people living in caves or primitive huts, start thinking about how it’d be better if they had some better homes, and then invent geometry for that purpose. Then, based on how well you did, people would start building better houses and you could walk around your kingdom looking for new problems to solve or new improvements to make.

How would that work in practice, given that math is fundamentally an act of creativity? How could there be a game that let players doodle around with math, experimenting with ideas, until they finally discovered the foundations of first geometry, and then the other subfields of mathematics? I don’t really know, but I do think that it could be done. For one, you’d want to equip the game with some sort of a theorem-prover, so that the players could experiment around with putting together various kinds of axioms and lemmas and see whether they produced interesting-looking theorems. Maybe an architect would suggest that it would be useful if you could prove some property about triangles, and then you could play around until you produced a statement that the game deemed to be logically equivalent with the wanted property. If you were running low on ideas, you would be given hints – perhaps in the form of taking a walk around your kingdom, until you saw something in nature that gave you an idea of an intermediate step or useful additional lemma, and the game would then give that to you as an intermediate goal.

Of course, there’s no reason for why this would need to be restricted to just mathematics. Inventing biology, physics, chemistry, medicine, economics, political science, and so on would certainly also be useful for your kingdom. The sciences are the easiest, since they have clear-cut correct answers that can be tested automatically, but one could also think about ways of teaching humanities in this way. History, for one, would be a natural fit, and the students could practice writing skills by composing essays and stories about what happened in the game.

School, then, would become a place where you went to play a fun game and talk about it with your friends and teachers afterwards. (I’m much inspired by the way an Australian teacher had his students by Plague Inc., after which they talked about the game in the light of the theory of evolution.) We could do away with the stressful and unfun exams this way – it’s obvious that we need exams for as long as school is stupid and boring and students won’t study unless they’re tested on the material, but with a game, you’re constantly proving your talents in-game. If we insist on giving kids grades – and I’m not sure that we should – we can do it by scoring their progress in the game.

Your honor, I submit that this kind of an organization would make far more sense. Maybe people would still need to spend some willpower to start playing the educational games rather than the entertainment games – after all, games that are optimized only for fun are likely to win in that department – but they wouldn’t need much willpower. And once they got started, they’d be hooked for a good while.

The prosecution rests. But not for long, because there’s still a lot of work to be done.

Originally published at Kaj Sotala. You can comment here or there.

Book review: A Mind Forever Voyaging: A History of Storytelling in Video Games

2013-01-22T18:43:46Z

A Mind Forever Voyaging: A History of Storytelling in Video Games. Dylan Holmes. CreateSpace Independent Publishing Platform.

As a form of storytelling, what makes video games distinct from other forms of storytelling, such as movies or books? What are the strengths and weaknesses of this form, what techniques has it borrowed from other media, and what untapped potential does it still have?

A Mind Forever Voyaging: A History of Storytelling in Video Games is a book that is essentially doing two things at once. It provides a history of thirteen games that have made important contributions to the art of video game storytelling, and on the side, it also provides some commentary on more general questions like the ones above. Doing two things at once is always harder than just doing one thing, but A Mind Forever Voyaging pulls it off pretty well. One or two early transitions between the specific and the general felt a little jarring, but then I either got used to them or the shifts became more natural.

The book is an interesting read in both senses. I had thought myself relatively knowledgeable about the history of video games, but until now, I hadn’t known what 1983 title had been possibly the first video game in history that had managed to make its players cry. And as there several games that I had heard a lot about but never played, it was interesting to hear exactly why Half-Life, for example, had been so popular.

The games that get a full chapter devoted to them are: The Secret of Monkey Island, Planetfall, Ultima IV: Quest of the Avatar, System Shock, Final Fantasy VII, Metal Gear Solid, Half-Life, Shenmue, Deus Ex, Metal Gear Solid 2: Sons of Libery, Façade, Dear Esther, and Heavy Rain. A number of others also get a couple of paragraphs worth of coverage each. As the author readily admits, this necessarily leaves out many games that would have deserved to be included, and the selection of which ones to include is a somewhat subjective one. In one case, a game was excluded from getting the full treatment because it was too good: Planescape: Torment was left out because ”there was too much to talk about: it begs for in-depth literary analysis, which was beyond the scope of what I was doing”.

Still, although one can always quibble about particular games that should have also been included, overall the selection strikes me as a good one: I’m pretty sure that if I’d had to pick thirteen games for such a project, I would have done worse. Before reading the relevant chapters, I felt a little dubious about whether it was really necessary to include two games from the same series – Metal Gear Solid 1 and 2, neither of which I had played. But when I did read the chapter about MGS2, I became apparent that the game had been quite innovative in the way that it exploited its nature as a sequel, and deserved a mention because of that fact. The nature of video game sequels is also somewhat special – as the author points out, video games are exceptional in that the sequels are often better than the original games, which isn’t the case with most other forms of media. That alone merited some discussion.

The titles have basically been picked on the basis of their novelty: whether they contributed new innovations to the art of video game storytelling. As such, the book can also be read as a collection of different storytelling techniques and considerations as applied to video games, which makes for a fascinating read.

How can game mechanics and storytelling aspects be integrated so that they support each other in building a more immersive experience? How much does immersion suffer from the game being so difficult that the player must keep reloading earlier saves? If it is exceedingly hard to make conversations with other characters feel like conversations with real people, is it sometimes better to not include any other characters at all? When can a game get away with addressing the player directly, potentially breaking the fourth wall? What techniques can be used to create the illusion that the player’s choices actually matter and have consequences? Such are some of the questions which are touched upon in the book, and seeing the intricacies behind some of the games I had liked made me appreciate them, and video games storytelling in general, more than I had before. If I were running my own video game studio, this book would probably be required reading for all my employees.

Some of those questions get relatively superficial coverage: they’re raised when discussing a single game, in the context of how that game did things, and then they’re never touched upon again. Others feel like recurring themes. The book will discuss a theoretical aspect of one game, and then move on and return to the same topic from another angle when discussing an entirely different game. These interwoven threads are not always pointed out explicitly, and it remains up to the reader to notice them and put the pieces together.

For example, one recurring theme in the book is the notion that video games are made distinct by the need to develop the whole world beforehand: a strength of video games is that the player can freely explore a world on their own, but fully exploiting this strength also requires the game designer to prepare interesting content that maintains that illusion of freedom and being able to do anything. If the game has many interactive or simulationist elements – an environment that actually gets damaged when it’s shot at, NPCs that display signs of intelligence, a system of moral consequence – it also becomes more likely that the player will be disappointed when the cracks in the illusion show up. Examples of such cracks include there being indestructible parts of the environment, the NPCs being clearly revealed as just scripted pieces of dialogue, or when the player’s actions don’t actually matter or morality is reduced to just another score to be maximized. The designer can avoid this problem by just making things more tightly akin to a movie, where the player is just a passive observer who’s along for the ride – but to do so means neglecting some of the unique potential that video games have.

Another solution is to try to use artificial intelligence techniques and machine-generated content in order to avoid needing to specify all the content by hand – but again, this can easily fail as the successes of the technique make its failures ever more obvious. This is clearly highlighted in the book’s discussion about Façade, an indie game about the breakdown of a couple’s marriage which uses a natural language parser to let the player converse with the couple and try to save their marriage. Sometimes the game gets lucky at interpreting the player’s writing and delivers a strongly compelling experience, and at other times, it performs… considerably less well.

A sort of meta-theme in the book, uniting many other themes, is the sense of game designers being engaged in a constant struggle to overcome the limitations of their format – both technical and financial. In a sense, it is a study of human ingenuity, of many people over many decades throwing themselves into a novel domain and gradually accumulating new ways of handling that domain, each building on the previous accomplishments of the others.

Although the book draws heavily upon the academic study of games, it never comes off as dry and boring: instead, it is a fast and enjoyable read. When I first started reading it, I thought that I’d read it for about half an hour before going to bed – I finally managed to force myself to put it away two and a half hours later. While this is a common occurrence with fiction, a non-fiction book that pulls this off is far more rare.

When not chronicling and analyzing specific games, the style of theoretical analysis is more tilted towards breadth than depth – which is fine, especially given that the book is mainly focused on providing a history of video game storytelling, not building a grand theory of said storytelling. Still, one gets a clear feeling that the author would have been capable of discussing each of the issues in far more detail than he does now. In any case, while the theoretical analysis does occasionally feel somewhat superficial, and never gets to the point of giving off a similar sense of brilliance as reading someone like Henry Jenkins does, it remains fascinating throughout. Reading it, I felt myself wanting to give it to some friends of mine to read, so that we could discuss its analyses together.

As is often the case, possibly the biggest failing of the book is that even at 250 pages, it feels too short. I would have gladly read a version of the book that was twice or even thrice the lenght, and covered that many more games. Right now, the book feels more like a snapshot of the history of storytelling in video games, rather than a history of it.

Perhaps the thing that I like the most about the book is that after reading it, I was left with a clear feeling of the very greatest video games still being ahead of us. Video games remain a young art form, and while game designers have experimented with many techniques for better storytelling, the full potential of those techniques remains untapped, waiting for someone to perfect them. We have only began to glimpse at just how good games could be.

(Full disclosure: the author is a long-time online friend of mine, which has probably biased this review a little, but I wouldn’t have written this in the first place if I hadn’t liked the book on its own merits.)

Originally published at Kaj Sotala. You can comment here or there.

Living books

2013-01-12T20:33:28Z

Do you feel like your books are static, passive objects, just sitting on a shelf and waiting for you to turn them alive? Think again.

As long as there is a light source in your room, then light is constantly being reflected off any exposed books in the room – from their covers if they’re closed, their pages if they’re open. That light hits the surface of the book, in a constant stream, and the surface transforms it, encoding the information contained within the surface into a signal, as the surface selectively absorbs some of the light and reflects the rest of it away.

The form and shape of the book’s letters is now contained within the light that gets reflected off, broadcast all across the room. If you are in a room with many books, they are all constantly bombarding you with their message, all the different waves of information hitting you countless of times per second. Like a radio station that’s sending whether or not one tunes into it, those signals keep coming even if you don’t pay attention to them. When you finally do, your eyes transform one of the patterns of light into a pattern of electricity, the raw signal undergoing a series of further transformations as your visual cortex extracts the information the light contained. Like a truck containing boxed goods, from which one first unloads the boxes and then opens the boxes to reveal their content, the signal of light first gives up the information about the letters, and then the information about the shapes and forms of the letters gives way to reveal the semantic content of the writing, the actual meaning of the words. You might never even consciously see the physical form of the writing as that meaning comes to life within your brain, igniting intricate networks of memories and associations, plunging you into a different world.

Our ancestors – both humans and the early creatures which eventually evolved into humans – lived a life of predator and prey, a life where some objects in our environment were dangerous or at least capable of running away, requiring us to take immediate action. It is because of the need to instantly know whether we should consider acting that we automatically classify everything as either alive or dead, animated or static.

But “animated or static” is just an abstraction that our brain imposes on its model of reality, a classification scheme that has often been useful for our purposes. Look closer, at atomic and subatomic levels, and everything is in perpetual motion: the universe is constantly recomputing itself, as the laws of physics dictate. Tiny particles are dancing and vibrating, information is being transmitted, received and transformed. The world around us, even so-called dead matter, is ever alive.

Originally published at Kaj Sotala. You can comment here or there.

Book review: Unfit for the Future

2012-11-20T07:39:53Z

Unfit for the Future: The Need for Moral Enhancement. Ingmar Persson & Julian Savulescu. Oxford University Press.

The core thesis of Unfit for the Future is that human morality evolved to allow cooperation and altruism in small groups, but that we today face challenges requiring extensive global coordination. Challenges such as weapons of mass destruction and climate change require both individual humans and nation-states to make various kinds of sacrifices for the benefit of all, but it is currently very unlikely to get everyone to actually make such sacrifices. Humans do have moral emotions such as a sense of justice and fairness that cause them to willingly make sacrifices in order to benefit those they know, but international cooperation requires trusting and helping faceless strangers – and humans have also evolved to be naturally suspicious or even xenophobic towards people outside their tribe. Since traditional moral education isn’t enough to overcome these challenges, we need to engage in “moral enhancement” and alter our biological moral dispositions.

The tone of the book is very academic and rational: there are few if any appeals to emotion, and logical reasoning from first principles is almost purely the style of argument. This makes the authors’ train of thought relatively clear to follow, though it also makes for a rather dry reading, and things are occasionally expressed in needlessly convoluted ways.

The best part of the book is the explanation of the coordination challenges involved with international cooperation, of why rational self-interest isn’t enough to overcome the challenges, and how our commonsense morality has evolved to solve some of these problems. The reader is assumed to already be mostly on board with the notion of risks from climate change and WMDs: some time is spent on explaining these risks, but probably not enough to sell the topic as a really extreme one for someone new to it.

Surprisingly, the book spends relatively little time (one chapter) talking about actual moral enhancement, and few concrete enhancement methods are proposed. Rather, there are a few examples of developing technologies that could be useful for moral enhancement, and a suggestion that more research be dedicated to developing more enhancement methods. Some criticisms of moral enhancement are discussed and argued against. The book concentrates on establishing the need for moral enhancement, not on proposing specific enhancement methods.

The main weakness of the book is that it does not always seem to engage with the strongest possible opposing arguments. A minor thesis that’s offered is that we should be ready to give up our privacy in order to prevent terrorists with WMDs, because of the untold damage that those terrorists could cause. The authors move to dismiss people having any moral right to privacy in only four (!) pages, and do so by considering two possible defenses for privacy: that violating privacy requires violating property rights, and that having one’s privacy violated makes one uncomfortable. The former is rendered irrelevant by the possibility of privacy violations that do not violate property rights (such as mind-reading devices or scanners that could see and hear through walls). The latter is rejected on the grounds that if you could forbid people from knowing something about you simply because it made you feel uncomfortable, “you could acquire very extensive rights against others just by being extremely sensitive about what others think of you”.

Leaving aside the fact that the latter argument is excessively simplistic, there is no absolutely no discussion of the fact that privacy gives people the chance to do harmless things for which they might nonetheless be discriminated against. Homosexual acts are the classic example, but even if one made the (false) assumption that liberal democracies – in the context of which the authors mostly frame their discussion – no longer exhibited homophobia, there are plenty of more examples to be found. Perhaps a person became sexually aroused by looking at (entirely non-sexual) pictures of children or animals, or enjoyed violent pornography, but would nevertheless never harm a soul. Ironically, a major part of the authors’ argument is that it is easier to destroy than to create, and that we find potential harms to be more pressing than an equivalent amount of potential gain. It is exactly because of such reasons that people who were thought to be possibly dangerous would be harshly and unfairly discriminated against – because even if the risk of them actually harming anyone would be small, few people would be willing to take that risk.

Nor do the authors discuss the fact that a lack of privacy could lead to excessive self-censorship, with even people who wouldn’t be discriminated against for acting according to their desires restricting their behavior just in case (again, the potential for harms outweighing the potential for gains). And once people could perfectly observe the behavior of everyone else, and see that everybody was acting conservatively, then even behavior that was previously within normal bounds might be come to be seen as suspicious, leading to an ever-more conformist, cautious, and unhappy society. The human suffering of such a development gives reason to believe in a strong moral right to privacy, and the suffering in question might easily outweigh the suffering from even several nuclear terrorist attacks. But aside for briefly mentioning that a fear of terrorism might cause some ethnic minorities to be unfairly discriminated against, the authors consider none of this.

It might also be somewhat distracting for some that the authors are clearly left-wing, which leads them to occasionally make ideological claims which are not very well-defended. For example, the authors briefly mention prevailing economic inequality as an example of one of humanity’s moral failings, citing differences between the poorest and richest nations as well as the poorest and richest people within some Western countries. None of the arguments for economic inequality of this form not necessarily being a bad thing are addressed. Fortunately, for the most part the left-wing digressions are minor points, and possible disagreement with them does not detract from the book’s major theses.

Overall, the book makes a nice argument for its core thesis, but could have been made much stronger by improving the strawmannish discussion of privacy, removing or better supporting ideologically contentious points, making the risk from WMDs better argued for, and by spending more time discussing moral enhancement itself, not just the need for it.

Originally published at Kaj Sotala. You can comment here or there.

Idea Stories

2012-10-21T10:34:52Z

When it comes to “idea” fiction, I think there are three main types of stories:

1. Twist Idea Stories. These have a single idea, which may or may not be hinted at during the story. It’s finally revealed on the last page or so, making for a twist ending. (E.g. many classic sci-fi short stories, from authors such as Asimov and Clarke.)
2. Big Idea Stories. These take a single big idea, or at most a couple of them, and spend the whole story examining the consequences of that from every possible angle. (E.g. “what if a virus made everyone infertile”, “what if you could statistically predict future history”, “what if you could make people immortal”.)
3. Idea Barrage Stories. These are full of small, interesting ideas, and the stories just keep throwing more and more ideas at the reader on every damn page. (E.g. some of Greg Egan’s works.)

And of course there can be intermediate variants of these, and there are novels which are arguably both Big Idea Stories and Idea Barrage Stories, in that exploring the single big idea spawns off smaller ideas which are its logical consequences. If we gave somebody living in 1950 the history of the years 1980-2012, they might think of it as a sci-fi story that started exploring the Big Idea of “what if there was this network of computers that spanned the world and which everyone could use” and then turned to an Idea Barrage with all the e-mails, instant messengers, YouTubes, 4chans, Facebooks, BitCoins, and so forth.

I like all three, though I note that it’s often hard to pull off a Big Idea Story well – often the author doesn’t actually manage to spend very much time examining the idea itself, and it gets relegated to be just a backdrop in an otherwise ordinary story. Which isn’t to say that it couldn’t be a good story nevertheless, it just isn’t a good Idea Story. But I just love a well-written Twist Idea or Idea Barrage Story.

What are your tastes?

(Cross-posted: Facebook, Google Plus.)

Originally published at Kaj Sotala. You can comment here or there.

Technology will destroy human nature

2012-10-05T13:34:58Z

squid314 recently made two posts [1 2] about some of the dangers of technology, and of becoming too powerful for yourself. Now, I’ll admit that I didn’t entirely understand his concern. As far as I could tell, he was worried that at some point, we might perfectly know the best possible strategy for pursuing all of our desires, and have the willpower to do so. Then, in a sense, one could say that we’d no longer experience having a free will. There would always be only one reasonable action in any situation, and we would always pick that one.

Well, I’m not too concerned about that. But the post hilighted one possible way that technology could damage something that we consider dear and essential, by removing essential constraints. That’s actually a rather major worry, and a far broader one than just one example suggests. (This essay was also influenced by a recent comment by Randal Koene.)

First, though, let’s review a bit of history.

In 1967, the biologist Sol Spiegelman took a strand of viral RNA, and placed it on a dish containing various raw materials that the RNA could use to build new copies of itself. After the RNA strands had replicated on the dish, Spiegelman extracted some of them and put them on another dish, again with raw materials that the strands could use to replicate themselves. He then kept repeating this process.

No longer burdened with the constraints of needing to work for a living, produce protein coats, or to do anything but reproduce, the RNA evolved to match its new environment. The RNA mutated, and the strands which could copy themselves the fastest won out. Everything in those strands that wasn’t needed for reproduction had just become an unnecessary liability. After just 74 generations, the original 4,500 nucleotide bases had been reduced to a mere 220. Useless parts of the genome had been discarded; the viral RNA had now become a pure replicator, dubbed “Spiegelman’s monster”. (Source.)

What happens in evolution is that organisms adapt themselves to exploit, and protect themselves from, the various regularities of the environment. Light reflects off distant objects in a predictable manner, so creatures have evolved eyes that they can use to see. If the environment ceases to possess some regularities, it will necessarily change the organisms. Put a fish with eyes in a cave with no light, and it will lose its sight over a few thousand years at most. Even humans have kept evolving as our environment has changed. Sickle-cell disease is more common in people whose ancestors are from regions with malaria. A single sickle-cell gene makes you more resistant to malaria, but two give you the disease. That’s an acceptable tradeoff in an environment with a lot of malaria, but a burden outside that environment.

You could say that the environment constrains the kind of organisms that can exist there. Now, those constraints aren’t immediate: that cave fish won’t lose its eyes right away. But over enough time, as different kinds of fish compete for survival, the ones which don’t waste their energy on growing useless eyes will win out.

Humans, as I was suggesting before, have also evolved to meet some very specific environmental constraints. As our environment has changed – either by our own doing, or due to reasons that have nothing to do with us – those constraints have changed somewhat, and we have changed with them. But many things about our nature, things that we might consider fundamental, have not changed. We still tell stories, enjoy the company of others, and are distinct individuals. Sure, the exact forms that those things take have changed over time. Today we are more likely to watch a story on TV than to hear one over a campfire – but both are still recognizable forms of story-telling. Countless of human universals are found in cultures all over the planet:

aesthetics; affection expressed and felt; age grades; body adornments; childhood fears; classification of kin; cooking; cooperation; customary greetings; daily routines; dance; distinguishing right and wrong; dreams; emotions; empathy; envy; family (or household); folklore; generosity admired; gossip; hope; hospitality; imagery; jokes; judging others; leaders; likes and dislikes; manipulating social relations; marriage; meal times; mourning; music …

Individuals may disagree about which of those things really are fundamental – whether losing some specific universal would really be a loss – but most people are likely to say that at least some of those things are important and worth keeping.

But as technology keeps evolving, it will make it easier and easier to overcome various constraints in our environment, our bodies, and in our minds. And then it will become increasing tempting to become a Spiegelman’s monster: to rid yourself of the things that the loosened constraints have made unnecessary, to become something that is no longer even remotely human. If you don’t do it, then someone else will. With enough time, they may end up ruling the world, outcompeting you like Spiegelman’s monster outcompeted the original, umutated RNA strands.

Exactly what kinds of constraints am I talking about, here? Well, there are several, in a roughly increasing order of severity:

Not being too powerful for yourself. Scott’s concern: that at some point, we might perfectly know the best possible strategy for pursuing all of our desires, and have the willpower to do so. Then, in a sense, one could say that we no longer experienced having a free will – there would always only be one reasonable action in any situation, and we would always pick that one.
Having distinct minds. We might not be too far away from having the ability to directly connect brains with each other. I think about something, and the thought crosses over to your brain, merging with your stream of consciousness. With time, this technology could be perfected so that large groups of people could join together into a single entity, coordinating and doing everything much better than any “traditional” human. Combined with an ability to copy memories, the concept of “personal identity” might cease to have any meaning at all – there would be no persons, just an amorphous mass of consciousnesses all sharing most of the same memories.
Unmodifiable desires: as desire modification becomes possible, anyone could reprogram their brains to be constantly perfectly satisfied and never do anything else (except possibly the bare minimum needed for survival). Sure, the possibility feels unappealing now… but maybe you’re having a bad day, and you choose to modify your brain to feel just a little better, all the time. And then the thought of being permanently blissed out doesn’t feel so bad after all, and you modify your brain just a little more… how could you not envy the folks who are never unhappy, especially since the option to self-modify is always there?
Inability to design superintelligent AGIs. We are constantly investing in ever-improving AI, for the obvious economic reasons: it allows for ever-more work to be automated. It may indeed prove impossible to regulate AI development in order to stop super-intelligent AGIs (artificial general intelligences) from arising. If so, then it might also prove impossible to ensure that safe and human-friendly AGIs prevail: like with Spiegelman’s monsters, the AGIs not burdened with the constraints of respecting human life and property may end up winning the AGIs that wish to protect humanity, after which they’ll recycle human settlements into their raw materials.
An inability to become mindless outsourcers. Nick Bostrom suggests a scenario where we learn to offload all of our thought to non-conscious external programs. To quote: “Why do I need to know arithmetic when I can buy time on Arithmetic-Modules Inc. whenever I need to do my accounts? Why do I need to be good with language when I can hire a professional language module to articulate my thoughts? Why do I need to bother with making decisions about my personal life when there are certified executive-modules that can scan my goal structure and manage my assets so as best to fulfill my goals?” And so, we give in to the temptation to cut away more and more parts of our brains, letting computer programs run those tasks… until there is no conscious experience left.
An inability to copy the best workers, choosing only the ones best fit for their tasks. If we could upload brains to computers, it could also become possible to copy minds. This could be far quicker than ordinary reproduction, making copying the primary method by which humans multiplied – and one’s ability to acquire and retain more hardware to run one’s copies on, would become the main criteria that evolution selected for. As Nick Bostrom writes: Much of human life’s meaning arguably depends on the enjoyment, for its own sake, of humor, love, game-playing, art, sex, dancing, social conversation, philosophy, literature, scientific discovery, food and drink, friendship, parenting, and sport. We have preferences and capabilities that make us engage in such activities, and these predispositions were adaptive in our species’ evolutionary past; but what ground do we have for being confident that these or similar activities will continue to be adaptive in the future? Perhaps what will maximize fitness in the future will be nothing but non-stop high-intensity drudgery, work of a drab and repetitive nature, aimed at improving the eighth decimal of some economic output measure. Even if the workers selected for in this scenario were conscious, the resulting world would still be radically impoverished in terms of the qualities that give value to life.

To rephrase what I have been saying:

“Humans” inhabit a narrow region in a multidimensional space of possibilities, and various constraints currently keep everyone stuck in that tiny space. If any of those constraints were to be relaxed – the space of possible minds stretched in any direction – then the new kinds of minds, no longer burdened with the constraints that make our fundamental values so adaptive, would be free to expand in entirely new directions. And it seems inevitable that, given a broader space of possible adaptations, evolutionary pressures would eventually lead to the dominance of minds – or at least replicators – which were very different from what most of us would value as “human”.

Get enough of one constraint, and you might still recognize the outcome as having once been human. Get rid of enough constraints, and you’ll get the equivalent of a Spiegelman’s monster, no longer even remotely human.

There have been some suggestions of how to avoid this. Nick Bostrom has suggested [1 2] that we create a “singleton”, a world-order with a single decision-making agency at the highest level, capable of controlling evolution. The singleton could be an appropriately-programmed AGI, the right group of uploads, or something else. Maybe this will work: but I doubt it. I expect all such efforts to fail, and humanity to eventually vanish. Possibly within my lifetime, if we’re unlucky.

I’ll conclude this essay with the immortal words of H.P. Lovecraft:

The most merciful thing in the world, I think, is the inability of the human mind to correlate all its contents. We live on a placid island of ignorance in the midst of black seas of infinity, and it was not meant that we should voyage far. The sciences, each straining in its own direction, have hitherto harmed us little; but some day the piecing together of dissociated knowledge will open up such terrifying vistas of reality, and of our frightful position therein, that we shall either go mad from the revelation or flee from the deadly light into the peace and safety of a new dark age.

Needless to say, Lovecraft was being too optimistic.

Originally published at Kaj Sotala. You can comment here or there.

Introduction to Connectionist Modelling of Cognitive Processes: a chapter by chapter review

2012-09-29T15:39:53Z

This chapter by chapter review was inspired by Vaniver’s recent chapter by chapter review of Causality on Less Wrong. Like with that review, the intention is not so much to summarize but to help readers determine whether or not they should read the book. Reading the review is in no way a substitute for reading the book.

I first read Introduction to Connectionist Modelling of Cognitive Processes (ICMCP) as part of an undergraduate course on cognitive modelling. We were assigned one half of the book to read: I ended up reading every page. Recently I felt like I should read it again, so I bought a used copy off Amazon. That was money well spent: the book was just as good as I remembered.

By their nature, artificial neural networks (referred to as connectionist networks in the book) are a very mathy topic, and it would be easy to write a textbook that was nothing but formulas and very hard to understand. And while ICMCP also spends a lot of time talking about the math behind the various kinds of neural nets, it does its best to explain things as intuitively as possible, sticking to elementary mathematics and elaborating on the reasons of why the equations are what they are. At this, it succeeds – it can be easily understood by someone knowing only high school math. I haven’t personally studied ANNs at a more advanced level, but I would imagine that anybody who intended to do so would greatly benefit from the strong conceptual and historical understanding ICMCP provided.

The book also comes with a floppy disk containing a tlearn simulator which can be used to run various exercises given in the book. I haven’t tried using this program, so I won’t comment on it, nor on the exercises.

The book has 15 chapters, and it is divided into two sections: principles and applications.

Principles

1: ”The basics of connectionist information processing” provides a general overview of how ANNs work. The chapter begins by providing a verbal summary of five assumptions of connectionist modelling: that 1) neurons integrate information, 2) neurons pass information about the level of their input, 3) brain structure is layered, 4) the influence of one neuron on another depends on the strength of the connection between them, and 5) learning is achieved by changing the strengths of connections between neurons. After this verbal introduction, the basic symbols and equations relating to ANNs are introduced simultaneously with an explanation of how the ”neurons” in an ANN model work.

2: ”The attraction of parallel distributed processing for modelling cognition” explains why we’re supposed to be interested in these kinds of models in the first place. It elaborates on some interesting characteristics of ANNs: the representation of knowledge is distributed over the whole network, they are damage resistant and fault tolerant, and they allow memory access by content. All of these properties show up in the human brain, but not in classical computer programs. After briefly explaining these properties, there is an extended example of an ANN-based distributed database storing information about various gang members. In addition to being content addressable, it also shows typicality effects – it can be asked a question like ”what are the members of the gang ‘Sharks’ like”, and it will naturally retrieve information about their typical ages, occupations, educational backgrounds, and so forth. Likewise, if asked to return the name of a pusher, it will suggest the name of the most typical pusher. In addition to explaining what the model is like, this chapter also explains the reasons for why it works the way it does.

3: ”Pattern association” describes a specific kind of an ANN, a pattern associator, and a particular kind of learning rule, the Hebb rule. Pattern associators are networks which are presented with a certain kind of pattern as input and a certain kind of pattern as output, after which they will learn to transform the input pattern to the output pattern. They are capable of generalization: if they encounter an input which is similar to ones they have encountered before, they will produce a similar output. They are also fault tolerant, in that they can produce good results even if parts of the network are destroyed. They also automatically perform prototype extraction. Suppose that there is a prototypical ”average” apple, and all other apples are noisy versions of the prototype. Pattern associators presented with several different patterns representing apples will learn to react the most strongly to an apple which is closest to the prototype, even if they have never actually seen the prototype itself.

4: ”Autoassociation” deals with autoassociator networks, and explains how the Delta learning rule works. Autoassociators are a special case of pattern associators – they are taught to reproduce the same pattern at output that was present at input. While this may seem pointless at first, autoassociators are an effective way of implementing a kind of memory: once trained, they can reproduce a complex pattern merely from seeing a small fragment of the original pattern. This has an obvious connection to the human brain, which can e.g. recall a complicated past memory from simply picking up a smell that formed a minor part of the original memory. Autoassociators are also capable of forming categories and prototypes from individual experiences, such as forming a category corresponding to the concept of a dog from seeing several dogs, without explicitly being told that they all belong to the same category. (Or, to put it in Less Wrong jargon, they learn to recognize clusters in thingspace.)

5: ”Training a multi-layer network with an error signal: hidden units and backpropagation” deals with the limitations of single-layered networks and how those limitations can be overcome by using more complex networks that require new kinds of training rules.

6: ”Competitive networks” differ from previous networks in that they can carry out unsupervised learning: while the previous nets had an explicit teacher signal, competitive networks learn to categorize input patterns into related sets on their own. They can perform both categorization, transforming related inputs into more similar outputs, and orthogonalization, transforming similar inputs into less similar outputs.

7: ”Recurrent networks” are capable of doing more than just simple transformations: they have feedback loops and more complicated internal states than non-recurrent networks can. This can be used to create sequences of actions, or to do things like predicting the next letter in a string of words and to identify word boundaries.

Applications

8: ”Reading aloud” can be difficult, especially in a highly irregular language like English, where most rules of how to transform a spelling to sounds have frequent exceptions. A child has to try to discover the regularities in an environment where there are both regularities and many exceptions. This chapter first briefly discusses traditional ”2-route” models of reading aloud, which presume that the brain has one route that uses pronounciation rules to read aloud regular words, and another route which memorizes specific knowledge about the pronounciation of exception words. These are then contrasted with connectionist models, in which there is no distinction between specific information and general rules. ”There is only one kind of knowledge – the weights of the connections which the model has acquired as a result of its experiences during training – and this is all stored in a common network.” The chapter then discusses several connectionist models which are successful in reading words aloud correctly, and which produce novel predictions and close matches to experimental psychological data.

9: ”Language acquisition” ”examines three aspects of language learning by children – learning the past tense, the sudden growth in vocabulary which occurs towards the end of the second year, and the acquisition of syntactic rules”. It describes and discusses various connectionist models which reproduce various pecularities of children’s language learning. For example, some young children initially correctly learn to produce the past tense of the word ”go” as ”went”, then later on overgeneralize and treat it as a regular verb, saying ”goed”, until they finally re-learn the correct form of ”went”. As in the previous chapter, models are discussed which are capable of reproducing this and other pecularities, as well as providing novel predictions of human language learning, at least some of which were later confirmed in psychological studies. The various reasons for such peculiarities are also discussed.

This chapter discusses many interesting issues, among others the fact that vocabulary spurts – dramatic increases in a child’s vocabulary that commonly happen around the end of age two – have been taken as evidence of the emergence of a new kind of cognitive mechanism. Experiments with connectionist models show that this isn’t necessarily the case – vocabulary spurts can also be produced without new mechanisms, as learning in an old mechanism reaches a threshold level which allows it to integrate information from different sources better than before.

10: ”Connectionism and cognitive development” elaborates on the issue of new mechanisms, discussing the fact that children’s learning appears to advance in stages. Traditionally, such qualitative changes in behavior have been presumed to be due to qualitative changes in the brain’s architecture. This chapter discusses connectionist models simulating the apperance of object permanence – the realization that objects continue to exist even when you don’t see them – and the balance beam problem, in which children are asked to judge the direction in which a balance beam loaded with various weights will tilt. It is shown that as the models are trained, they undergo stage-like change like a child, even though their basic architecture remains constant and the same training rule is used.

11: ”Connectionist neuropsychology – lesioning networks” shows that selectively damaging trained networks can closely mimic the performance of various brain damaged patients. Models of damaged performance are examined in the fields of reading, semantic memory, and attention.

12: ”Mental representation: rules, symbols and connectionist networks” discusses and counters claims of connectionist networks never being able to learn some kinds of rules which require one to use rules or symbols, due to having no explicit representation of them.

13: ”Network models of brain function” discusses two models which attempt to replicate brain functionality and experimental data about the brain: a model of the hippocampus, and a model of the visual cortex. Both are shown to be effective. The hippocampus model in good in storing and recalling patterns of data. The visual cortex model, on the other hand, succeeds in identifying faces regardless of their position in the visual field, and regardless of whether the face is seen from the front or from the side.

14: ”Evolutionary connectionism” has a brief discussion of connectionist networks that are trained using evolutionary algorithms. 15: ”A selective history of connectionism before 1986” is pretty much what it sounds like.

———–

In addition to providing a nice analysis of various models originally published in journal papers, the book also provides a bit of a historical perspective. Several of the chapters start with analyzing an early model that produced promising but imperfect results, and then move on to a later model which built on the previous work and developed more realistic results. References are provided to all the original papers, and readers who are interested in learning more about the various topics discussed in the book are frequently referred to sources which discuss them in more depth.

I would recommend this book to anyone who has an interest in psychology, can handle a bit of math, and who isn’t yet familar with all the various and fascinating things that connectionist models are capable of doing. Although the models discussed in the book are all very simple as compared to the ones in the brain, they do make the idea of the brain being successfully reverse-engineered during this century feel a lot more plausible.

Originally published at Kaj Sotala. You can comment here or there.

Political logic 101

2012-08-16T16:05:01Z

Liberal Logic 101 is an interesting site. As with any political site of this nature, a lot of the pictures and comparisons are just dumb: comparisons picked to make the enemy look bad with no context provided, or outright just stating “liberals boo, conservatives yay” without even trying to offer any kind of justification. But seeing somebody of a different ideology pass these kinds of pictures around as self-evident does help realize how many of the pictures you agree with and see passed around with are similarly vacuous and serving no purpose for any useful debate.

The idealist in me wishes that everyone was made to study lots of these kinds of pictures from a variety of ideologies, to see how silly they all were and to stop passing them around. The realist thinks that most people would just end up thinking “wow, people of [any group I don't agree with] are really stupid”.

And then there are some pictures that genuinely are clever and funny, in that they point out something inconsistent – or at least something that seems inconsistent if you’re inclined to accept certain premises, such as fetuses being people. Maybe if people looked at these kinds of pictures, they’d see how much of our reasoning depends on accepting various premises (or moral intuitions), and how an argument that seems silly if you reject one premise becomes insightful once you accept it, and there’s no objective reason for accepting or rejecting any particular one. Then again, maybe not…

Also interesting are the “Furious Liberal Friday” pictures. I don’t know whether these are genuine quotes from people or just made up, but many of them do sound like ones that people would make in real life. E.g. here’s the text of one of the Furious Liberal Friday pictures on that page:

“‘I don’t tolerate intolerance. I know that’s a hard concept to grasp, but you can’t have your cake and eat it too.’
(Actual comment left on LiberalLogic101 Facebook page)
(I suppose it would be funny if it weren’t so sad that the person who wrote it still does not get it.)”

I’m not sure how well one can emphasize with that without being part of a frequently misunderstood (political?) movement, but having a lot of experience with being told that political pirates just hate artists and went to get everything for free, this certainly struck a chord. The feeling of frustration-combined-with-hopeless-amusement when somebody makes a claim about you that you feel is completely off-base is certainly very familiar. And I can certainly imagine that most conservatives won’t feel like they’re intolerant, and instead feel that they’re motivated primarily by altruism and rational thought. Just like most liberals feel like they are motivated primarily by altruism and rational thought.

Originally published at Kaj Sotala. You can comment here or there.

Influences

2012-08-15T18:40:28Z

Obviously, many things have had an influence on my thought. Here are some of the major ones. Unfortunately, the focus is on the most recent ones, since I’ve mostly forgotten many of the earlier works.

I expect to add more as I remember them.

Books

Against Intellectual Monopoly (free online version), by Michele Boldrin and David K. Levine. This book completely changed my thinking on copyright and patents, and opened my eyes to the vast amounts of harm that they do in their current form.
No Boundary: Eastern and Western Approaches to Personal Growth by Ken Wilber. Unfortunately, this book needs to be read with a grain of salt, as there’s lots of mystical mumbo-jumbo mixed in with the insightful stuff. Still, if you can manage to read it with the right reductionist mindset, you can extract lots of useful philosophy about the nature of emotions, the self and the mind. Together with Feeling Good: The New Mood Therapy by David D. Burns, it did a lot to make me become happier.
Righteous Mind: Why Good People Are Divided by Politics and Religion by Jonathan Haidt. While there are a number of reasonable criticisms that can be made against Haidt’s work, he still does an excellent job of exploring the foundations of conservative and liberal morality, and showing in how they differ. His explanation of the ways by which human moral reasoning is to a large extent rationalization is also excellent.
Stein On Writing, by Sol Stein. The best book on writing that I’ve ever read, useful and inspiring at the same time. Provides advice on both fiction and non-fiction writing.
Why Everyone (Else) Is a Hypocrite: Evolution and the Modular Mind (free first chapter; my unfinished summary), by Robert Kurzban. Does an amazing job of exploring the modular nature of the human mind, and its many implications.

Online resources

The Sequences, by Eliezer Yudkowsky. Huge and of somewhat varying quality, but overall excellent. No short description will do justice to these posts; they helped to completely transform my thinking.

Bloggers

Ferrett Steinmetz writes on a variety of topics, mostly relating to human relationships, with humbling amounts of wisdom. His experiences with depression gave part of me the insight I needed in order to write Avoid misinterpreting your emotions. If I had to pick one post of his that sticks most clearly to my mind, it would be “It’s that breakup season“, with its elegant decomposition of relationships to their “love”, “trust”, and “like” components.
Robin Hanson seems to often oversimplify things a little too much for my liking, but every now and then he produces some great cynical insights about how society really works, as opposed to how people claim it works. For some examples, see Politics isn’t about Policy, Academia’s Function, and Academics as Warriors. I also really like This is the Dream Time, and strongly suspect that some form of his Futarchy proposal might in fact be the best possible form of government.

Originally published at Kaj Sotala. You can comment here or there.

On unhealthy relationships

2012-08-13T20:56:50Z

Clarisse Thorn: How my life wasn’t always Happy Fun Boundaries Are Perfect Land.

“Here is the strange part, for me, in remembering him: I don’t think he consciously wanted me to hurt myself like that. If he had been deliberately abusive, if he had really wanted to tear me apart, if he’d been physically abusive [...] Maybe then I would never have gotten involved? Maybe then I would have walked away sooner? But maybe not.”

“Can I teach other people to set boundaries in situations like that? I don’t know. The feminist ideas and gender analysis I was exposed to as a kid didn’t prevent that experience (although, again, maybe those things would have helped if the situation had been more obvious: if he’d been physically abusive, for example, or more overtly controlling).”

I recommend the above article particularly for those with little experience with relationships. There’s a lot about this text and situation that seems familiar to me: in my first relationship, I too should have been better at setting my borders and policing them. And when my partner didn’t properly respect my borders, it wasn’t out of malice either: I have no doubt that she really did love me, but rather just didn’t realize what she was doing, or just couldn’t help being needy when she did.

It was exactly that which made it so hard for me to say no when I should have: had the relationship been openly abusive, I would have realized it pretty quickly, but when I was already in the relationship and my partner needed me and clearly cared about me, how could I have said no? (At least I would have left it had it been openly abusive from the start. It didn’t seem dysfunctional at first, either.) And even if the requests seemed unreasonable, wasn’t it reasonable that I who was better off compromised on what I wanted? And if it was impossible to even raise the issues of what I experienced as unfair without her pretty much breaking down and starting to hate herself, and me being forced to patch her back together without us ever really getting to the point of talking about those issues… then sometimes, that resentment had few other places to go than to turn inward, and I might wonder whether I was the one who should have tried harder.

Some people will know or guess who I’m talking about, so let me emphasize again that I don’t blame her, nor bear her any ill will. Again, I’m sure that she really did care about me, but was just undergoing a really hard time, and was in a really bad shape. She’s better now. We’ve talked about it, she’s sorry about it, and I’ve forgiven her. (I also let her read this text beforehand and made sure that she was okay with me posting it.) And she did teach me to be far more sensitive about my borders, and I think that I’ve done a much better job of setting my limits since then. It’s rare, but sometimes that which doesn’t kill you really does make you stronger.

Would I have done things any differently if I had read this article beforehand? As the author of the article says, when asking herself whether she would have acted differently in case her lover would have been more abusive – maybe not. I was a lonely teenager, being really in love for the first time in his life, after having had experienced many unrequited crushes before. It’s possible, and perhaps even likely, that I would have regardless just tried to do everything to make the relationship work, just as I did back then. But maybe this will help someone else instead.

Of course, none of this is to say that people should dump their partners if the partner is having a difficult time, or that you shouldn’t ever compromise on your desires if you’re clearly better off than your partner is. That’s what makes these issues so hard – there are no clear lines of what to do when. But at least make sure that you really are helping because you genuinely want to help… not because you’re being guilted into it.

Originally published at Kaj Sotala. You can comment here or there.

Anticipation and meditation

2012-08-06T16:38:08Z

Germund Hesslow’s paper Conscious thought as simulation of behaviour and perception, which I first read maybe three months back, has an interesting discussion about anticipations.

I was previously familiar with the idea of conscious thought involving simulation of behavior. Briefly, the idea was that when you plan an action, you are simulating (imagining) various courses of action and evaluating their possible outcomes in your head. So you imagine bringing your boyfriend some flowers, think of how he’d react to that, and then maybe decide to buy him chocolate instead. Imagining things is a process of constructing a simulation of them. Nothing too surprising in that idea. Here’s how Hesslow puts it:

What we perceive is quite often determined by our own behaviour: visual input is changed when we move our head or eyes; tactile stimulation is generated by manipulating objects in the hands. The sensory consequences of behaviour are to a large extent predictable (Fig. 2a). The simulation hypothesis postulates the existence of an associative mechanism that enables the preparatory stages of an action to elicit sensory activity that resembles the activity normally caused by the completed overt behaviour (Fig. 2b). A plausible neural substrate for such a mechanism is the extensive fibre projection from the frontal lobe to all parts of sensory cortex. Very little is known about the function of these pathways, but there is physiological evidence from monkeys that neurons in polysensory cortex can be modulated by movement[33].

But the “buy flowers or chocolate?” example concerns relatively long-term decision-making. We also simulate the short-term consequences of our actions (or at least try to). And what I had not consciously realized before, but what was implied in the excerpt above, was that very immediate consequences will be simulated as well.

Discussing this paper with a friend, we considered the subjective experience of such anticipatory simulations. Suppose that I want to open a door, and start pushing down the handle. Even before I’ve pushed it all the way down, I seem to already experience a mild foretaste of what having pushed it down feels like. I know what it will feel like to have completed the action, a fraction of a second before actually having completed that action, and it feels faintly pleasing when that anticipation is realized.

Which was interesting to realize, but not particularly earth-shattering by itself. But the real discovery came soon after reading the paper. I was doing some vipassana-style meditation, focusing on the feeling of discomfort that came from wanting to swallow as there was excess saliva gathering in my mouth. I realized that what I thought of as “discomfort” was actually a denied anticipation. I wanted to swallow, and there was already in my mind a simulation of what swallowing would feel like. I was already experiencing some of the pleasure that I would get from swallowing, and my discomfort came from the fact that I wanted to experience the rest of that pleasure. When I realized this, I focused on that anticipated pleasure, trying to either make it stop feeling pleasant, or alternatively, strengthen the pleasure so that I could enjoy it without actually swallowing. My clock rang before I could fully succeed in either, but I did notice that it made it considerably easier to resist the urge.

On my way to town, I started observing my mental processes and noticed that that tiny anticipation of pleasure was everywhere. Coming to the train station, there was an anticipation of not needing to wait for long. Using a machine to buy more time on my train card, there was an anticipation of the machine working. Waiting for the train, there was an anticipation of seeing the train arrive and getting to board it. And each time that I experienced discomfort, it was from that subtle anticipation being denied. Anticipating the experience of seeing the train being there on time could have led to frustration if it was running late. Anticipating the experience of boarding the train led to impatience as the train wasn’t there yet, and that sequence of planned action that had already been partially initiated couldn’t finish. Suddenly I was seeing the anticipatory component in every feeling of discomfort I had.

When I realized that, I started writing an early draft of this post, which contained the following rather excited paragraph:

That’s what “letting go of attachments” refers to. That’s what “living in the moment” refers to. Letting go of the attachment to all predictions and anticipations, even ones that extend only seconds into the future. If one doesn’t do that, they will constantly be awaiting what happens in some future moment, and will experience constant frustrations. On some intellectual level I already understood that, but I needed to develop the skill for actually noticing all my split-second anticipations before I could really get it.

Unfortunately, what often happens with insights gained from meditation is that one simply forgets to apply them. Or if one does, in principle, remember that they should apply the insights, they’ll have forgotten how. Being able to isolate the anticipation from the general feeling of frustration, and then knowing how to let go of the attachment to it, is a tricky skill. And I ended up mostly just forgetting about it, especially once my established routine of meditating once per day got interrupted for a month or so.

I did some meditation today, and finally remembered to try out this technique again. I started looking for such anticipations whenever I experienced a feeling of discomfort, and when I found any, I just observed them and let go of them. And it worked – I was capable of meditating for a total of 70 minutes in one sitting, and got myself to a pleasant state of mind where everything felt good. That feeling persisted for most of the rest of the day.

But after that session, it feels like my earlier characterization of the technique as “a cessation of attachments to predictions” would be a little off. That description feels clunky, and like it doesn’t properly describe the experience. “Letting go of a desire for sensations to feel different” sounds more like it, but I’m not sure of what exactly the difference is.

This probably also relates to another meditation experience, which I had about two months back. I was concentrating on my breath, and again, I noticed that the sensation of saliva in my mouth was bothering me. At first I tried to just ignore it and keep my attention on my breath; or alternatively, to let go of the feeling of distraction so that the sensation of saliva wouldn’t bother me anymore. When neither worked, I essentially just thought “oh, screw it” and accepted the sensation just as it was, as well as accepting the fact that it would continue to bother me. And then, once I had accepted that it would bother me… the feeling of it bothering me melted away, and vanished from my consciousness entirely. I was left with a warm, strongly pleasant feeling that lasted for many hours after I’d stopped meditating.

I haven’t been able to put myself back into that exact state, because as far as I can tell it, getting into it requires you to genuinely accept the fact that you’re feeling uncomfortable. In other words, you cannot use the acceptance as a means to an end, thinking that “I’ll now accept this unpleasantness so that I’ll get back to that nice state where it doesn’t feel unpleasant anymore”. That’s not genuine acceptance anymore, and therefore it doesn’t work.

Anyway, it feels like the “isolate anticipations and let go of them” and “accept your feelings and discomforts exactly as they are” techniques would be two different ways of achieving the same end. The feeling of pleasure I got today wasn’t as strong as the feeling of pleasure I got when I managed to accept my discomforts as they were, but it seemed to have much of the same character.

Some – though not all – meditators report a lack of achievement after reaching high levels of skill. They’re just happy with doing whatever, with no need to accomplish more things. And after meditating today, I too felt happy with whatever would happen, with no urgency to accomplish (nor avoid!) any of the things that I had planned for today. There seems to be a fine line between “use meditation to get rid of your disinclination for doing the things you want to do” and “use meditation and get rid of your inclination to do anything”.

In any case, I will have to try to remember this technique from now on, and keep experimenting with it. Hopefully, having written this post will help.

Originally published at Kaj Sotala. You can comment here or there.

New website opened

2012-07-22T09:01:15Z

I revamped my website and finally brought it a little more to the 21st century: no longer handmade HTML for each page! I'll also be cross-posting my future LJ posts on the site's blog section in the future, so I now finally have a Real Blog (TM) instead of just a LiveJournal account. Though everything will still be posted to LJ as well, don't worry.

I'll also be cross-posting any posts I write in my Finnish Uusi Suomi blog at the new site. There's also an option to only show the English posts, for those who can't read Finnish.

Opinions have been divided on the logo graphic, so I'll have to see if I find something that's a little more universally liked. It's not a huge priority, though.

There was a bunch of content on the old site that I wasn't happy with and thus didn't move over, or just didn't feel was valuable enough to bother with. If you want to see something that I had on the old site, please use the Wayback Machine archive.

(Also, the fancy plugin that I installed was supposed to automatically cross-post this entry here, but it looks like it didn't, so I had to do it manually. Let's hope that I figure out how to make it work.)

Thoughts on moral intuitions

2012-06-28T11:04:51Z

Our moral reasoning is ultimately grounded in our moral intuitions: instinctive "black box" judgements of what is right and wrong. For example, most people would think that needlessly hurting somebody else is wrong, just because. The claim doesn't need further elaboration, and in fact the reasons for it can't be explained, though people can and do construct elaborate rationalizations for why everyone should accept the claim. This makes things interesting when people with different moral intuitions try to debate morality with each other.

---

Why do modern-day liberals (for example) generally consider it okay to say "I think everyone should be happy" without offering an explanation, but not okay to say "I think I should be free to keep slaves", regardless of the explanation offered? In an earlier age, the second statement might have been considered acceptable, while the first one would have required an explanation.

In general, people accept their favorite intuitions as given and require people to justify any intuitions which contradict those. If people have strongly left-wing intuitions, they tend to consider right-wing intuitions arbitrary and unacceptable, while considering left-wing intuitions so obvious as to not need any explanation. And vice versa.

Of course, you will notice that in some cultures specific moral intuitions tend to dominate, while other intuitions dominate in other cultures. People tend to pick up the moral intuitions of their environment: some claims go so strongly against the prevailing moral intuitions of my social environment that if I were to even hypothetically raise the possibility of them being correct, I would be loudly condemned and feel bad for even thinking that way. (Related: Paul Graham's What you can't say.) "Culture" here is to be understood as being considerably more fine-grained than just "the culture in Finland" or the "culture in India" - there are countless of subcultures even within a single country.

---

Social psychologists distinguish between two kinds of moral rules: ones which people consider absolute, and ones which people consider to be social conventions. For example, if a group of people all bullied and picked on one of them, this would usually be considered wrong, even if everyone in the group (including the bullied person) thought it was okay. But if there's a rule that you should wear a specific kind of clothing while at work, then it's considered okay not to wear those clothes if you get special permission from your boss, or if you switch to another job without that rule.

The funny thing is that many people don't realize that the distinction of which is which is by itself a moral intuition which varies from people to people, and from culture to culture. Jonathan Haidt writes in The Righteous Mind: Why Good People Are Divided by Politics and Religion of his finding that while the upper classes in both Brazil and USA were likely to find violations of harmless taboos to be violations of social convention, lower classes in both countries were more likely to find them violations of absolute moral codes. At the time, moral psychology had mistakenly thought that "moving on" to a conception of right and wrong that was only grounded in concrete harms would be the way that children's morality naturally develops, and that children discover morality by themselves instead of learning it from others.

So moral psychologists had mistakenly been thinking about some moral intuitions as absolute instead of relative. But we can hardly blame them, for it's common to fail to notice that the distinction between "social convention" and "moral fact" is variable. Sometimes this is probably done for purpose, for rhetorical reasons - it's a much more convincing speech if you can appeal to ultimate moral truths rather than to social conventions. But just as often people simply don't seem to realize the distinction.

(Note to international readers: I have been corrupted by the American blogosphere and literature, and will therefore be using "liberal" and "conservative" mostly to denote their American meanings. I apologize profusely to my European readers for this terrible misuse of language and for not using the correct terminology like God intended it to be used.)

For example, social conservatives sometimes complain that liberals are pushing their morality on them, by requiring things such as not condemning homosexuality. To liberals, this is obviously absurd - nobody is saying that the conservatives should be gay, people are just saying that people shouldn’t be denied equal rights simply because of their sexual orientation. From the liberal point of view, it is the conservatives who are pushing their beliefs on others, not vice versa.

But let's contrast "oppressing gays" to "banning polluting factories". Few liberals would be willing to accept the claim that if somebody wants to build a factory that causes a lot of harm to the environment, he should be allowed to do so, and to ban him from doing it would be to push the liberal ideals on the factory-owner. They might, however, protest that to prevent them from banning the factory would be pushing (e.g.) pro-capitalism ideals on them. So, in other words:

Conservatives want to prevent people from being gay. They think that this just means upholding morality. They think that if somebody wants to prevent them from doing so, that somebody is pushing their own ideals on them.

Liberals want to prevent people from polluting their environment. They think that this just means upholding morality. They think that if somebody wants to prevent them from doing so, that somebody is pushing their own ideals on them.

Now my liberal readers (do I even have any socially conservative readers?) will no doubt be rushing to point out the differences in these two examples. Most obviously the fact that pollution hurts other people than just the factory owner, like people on their nearby summer cottages who like seeing nature in a pristine and pure state, so it's justified to do something about it. But conservatives might also argue that openly gay behavior encourages being openly gay, and that this hurts those in nearby suburbs who like seeing people act properly, so it's justified to do something about it.

It's easy to say that "anything that doesn't harm others should be allowed", but it's much harder to rigorously define harm, and liberals and conservatives differ in when they think it's okay to cause somebody else harm. And even this is probably conceding too much to the liberal point of view, as it accepts a position where the morality of an act is judged primarily in the form of the harms it causes. Some conservatives would be likely to argue that homosexuality just is wrong, the way that killing somebody just is wrong.

My point isn't that we should accept the conservative argument. Of course we should reject it - my liberal moral intuitions say so. But we can't in all honestly claim an objective moral high ground. If we are to be honest to ourselves, we will accept that yes, we are pushing our moral beliefs on them - just as they are pushing their moral beliefs on us. And we will hope that our moral beliefs win.

Here's another example of "failing to notice the subjectivity of what counts as social convention". Many people are annoyed by aggressive vegetarians, who think anyone who eats meat is a bad person, or by religious people who are actively trying to convert others. People often say that it's fine to be vegetarian or religious if that's what you like, but you shouldn't push your ideology to others and require them to act the same.

Compare this to saying that it's fine to refuse to send Jews to concentration camps, or to let people die in horrible ways when they could have been saved, but you shouldn't push your ideology to others and require them to act the same. I expect that would sound absurd to most of us. But if you accept a certain vegetarian point of view, then killing animals for food is exactly equivalent to the Holocaust. And if you accept a certain religious view saying that unconverted people will go to Hell for an eternity, then not trying to convert them is even worse than letting people die in horrible ways. To say that these groups shouldn't push their morality to others is to already push your own ideology - which says that decisions about what to eat and what to believe are just social conventions, while decisions about whether to kill humans and save lives are moral facts - on them.

So what use is there in debating morality, if we have so divergent moral intuitions? In some cases, people have such widely differing intuitions that there is no point. In other cases, their intuitions are similar enough that they can find common ground, and in that case discussion can be useful. Intuitions can clearly be affected by words, and sometimes people do shift their intuitions as a result of having debated them. But this usually requires appealing to, or at least starting out from, some moral intuition that they already accept. There are inferential distances involved in moral claims, just as there are inferential distances involved in factual claims.

So what about the cases when the distance is too large, when the gap simply cannot be bridged? Well in those cases, we will simply have to fight to keep pushing our own moral intuitions to as many people as possible, and hope that they will end up having more influence than the unacceptable intuitions. Many liberals probably don't want to admit to themselves that this is what we should do, in order to beat the conservatives - it goes so badly against the liberal rhetoric. It would be much nicer to pretend that we are simply letting everyone live the way they want to, and that we are fighting to defend everyone's right for that.

But it would be more honest to admit that we actually want to let everyone live the way they want to, as long as they don't things we consider "really wrong", such as discriminating against gays. And that in this regard we're no different from the conservatives, who would likewise let everyone live the way they wanted to, as long as they don't do things the conservatives consider "really wrong".

Of course, whether or not you'll want to be that honest depends on what your moral intuitions have to say about honesty.

A Date with Darin

2012-06-17T23:11:16Z

Been a while since I wrote any fiction. Just a random piece whose beginning popped into my head and demanded to be written just as I was about to fall asleep.

Conflicted characters are fun.

----

"Look, I'm not really the best person to have a relationship with. I can be selfish as fuck, I break my promises more often than not, and I suck at making compromises." I played with the fork I'd been using to eat, but kept my eyes on my date as I spoke. "You'd probably be happier if we kept this as just sex, to be honest."

"I know. But I'm willing to give it a shot."

I sighed. "Don't give up easily, do you." They never did, or they wouldn't have had the courage to ask in the first place. Suddenly feeling reluctant to keep looking at him in the eyes, I glanced around me, studying the other customers of the Indian restaurant we were eating in.

"Give up my only chance of ending up together with you? Not in a million years."

I couldn't help smiling. It was a little corny and cliché, but that just made it cute. Darin was cute - which was the reason I'd been having sex with him.

"Well", I said slowly. "I do like you."

Which was kinda the problem. If I hadn't liked him, I could have just told him no. I did want to be in a real relationship one day, and there was no reason why it couldn't have been with Darin. He was interested, and willing to try despite the risks, and I liked him - so I wanted to try, too.

But since I liked him, I didn't want him to get hurt, either.

So I continued. "Still, like I said, I'm difficult. If we have a real relationship, it's often going to be ugly. So to start off, I'm going to explain to you all the ways by which I'm a selfish asshole who's going to hurt you, and you can then decide whether or not you really want this."

He nodded, and the eagerness in that nod told me everything I needed to know. We'd finish our meal, go somewhere more private, and I'd explain in detail all my faults. I'd tell him not to make a decision now, but rather think about it for a couple of days and then decide. He'd say okay, and then we'd need something else to do for the rest of the evening, so we'd move on to the fucking.

And then he'd be a good boy and go think about it for a couple of days like I told him to, and then he'd say that he understood the risks, and I'd look at him and see that he didn't. And then I'd know that I should tell him no, but I wouldn't. He'd be so disappointed if I turned him down after all, and it could work out and it would be wonderful if it did, and to be honest, if it'd let me find Mr. Right I'd go ahead and break Darin's heart many times over. For I really did want a happy relationship, and because I was selfish enough of an asshole to try it if there was the slightest chance of it working out.

I sighed, again, and picked at my rice. "Alright, then. We'll talk about it at my place."

This was going to be such a train wreck.

Movie madness

2012-05-13T08:08:20Z

A possibly entertaining pastime I just came up with, for a bunch of people who can think at their feet:

Find a movie none of you has seen before, spoken in a language none of you knows. If it has subtitles in your language, turn them off. Each time when a character in the movie speaks, somebody improvises a "translation" of what they're saying. Optionally, each character in the movie may be "adopted" by the first person to translate them, who then translates all of their lines from that point on. Or not, in which case all characters remain free-for-all. See for how long you can keep the story consistent.

To make things more interesting, you could decide to change the genre of the movie. For instance, if you know that it's a romantic comedy, you could decide beforehand that it's really a sci-fi drama, or vice versa.

On fantasy and discount rates

2012-05-12T11:52:37Z

"...some method for discounting future and distant consequences is necessary. It is possible, perhaps, that the degree of discounting would exactly correspond to the increasing degree of uncertainty that goes with predicting remote events. But there is no simple formula that relates time or distance to uncertainty—some events a year from now or 5,000 kilometers from here may be much more predictable than other events only one week from now or 100 meters away." (Wallach & Allen, Moral Machines)

This bit made me think. Wallach & Allen state correctly that the relation between (physical or temporal) distance and uncertainty is not a simple one, and some things which happen far away are more predictable than some things that happen nearby. But in what kind of a universe would that statement be incorrect? If it always made sense to consider events happening a week from now 20% more predictable than events happening a year from now, or events happening 100 meters away 20% more predictable than 5,000 kilometers away, regardless of the type of the event, what would that imply of the world one lived in?

It seems to me that this would have to be a world where the laws of nature were highly local. Things such as the speed of light and the boiling point of water would have to vary smoothly, depending on where and when you were. Actually, whether the universe had things such as "light", "water", or "boiling" in the first place would also vary by location. There are some things that we probably need to keep constant in order to avoid a logical contradiction, though. For instance, since we stipulated that events happening 100 meters away should be 20% more predictable than events happening 5,000 kilometers away, the geometry of the universe should not change as to make the concept of distance meaningless, nor should the axioms of arithmetic be changed at random.

Would it be possible to live in such a universe? Certainly if the discount rate was high enough, the universe would be so chaotic as to make all advance planning pointless. Not to mention the fact that organisms wouldn't live for very long if their blood might literally begin to boil at any time. But let's stipulate that the rate of change was slow, and that the organisms living in the universe were generally changed in such ways as to not outright kill them or drive them insane. Note that we are now again forced to introduce some predictability that is not a direct function of distance.

Things would perhaps get easier if we dropped the bit about change over time. The laws of nature would be spatially local, so traveling 5,000 kilometers would get you to a place where things worked quite differently, but things wouldn't change if you stayed put. For this, we'll obviously have to limit ourselves to laws of physics which allow for time and space to be separated in such a way. Now we don't need to protect the organisms living in this universe from its changes anymore - organisms in different regions will simply evolve to exploit their local laws of nature, and to avoid going into places where they cannot survive anymore.

Some of the regions in such a universe would be teeming with life (though whether we'd recognize it as life is another matter), while other regions would be desolate, incapable of supporting any kind of complex structure. Journeying far from your home would let you see things that were literally impossible back at your place of birth, but to travel far enough would mean a certain death. Although you could never directly witness the wonders of the regions that were too different from yours, you might find creatures that lived at the borders of such regions. They could travel farther away than you, although they could not come to the place you were from; and if you could find a way to communicate, the two of you might be able to swap tales. You could tell them of the things you had seen, and in turn, be told of wonders you could imagine, but never quite comprehend.

Personal achievement reports

2012-05-05T09:52:23Z

I haven't posted my personal achievement reports for March and April, and in fact I think I might discontinue the habit for now. The reason is simple: for the last two months I've been doing writing for the Singularity Institute - on a per-project basis since early March, and on a full-time basis since early April. The job takes up most of my energy, so most of my reports would consist of little more than "worked X hours, meditated Y hours this month". And I don't want to speak much about the projects I'm doing before they're done.

So expect to see the occasional meditation log, "hey I completed this" update, or whatever, but not necessarily more achievement reports. Though I might still change my mind about this.

Neuroinformatics 4 seminar, session III - GWT/meditation, neural correlates of consciousness

2012-03-19T16:07:08Z

Yes, I know that I'm way behind on my reports: session III was over a month ago. Better late than never.

I've been thinking about global workspace theory on and off in the context of meditation. Haven't come up with anything particularly insightful, basically just a repetition of the argument in the Dietrich paper: in meditation, attentional resources are used to actively amplify a particular event such as a mantra until it becomes the exclusive content in the working memory buffer. This intentional, concentrated effort selectively disengages all other cognitive capacities of the prefrontal cortex.

Put into GWT terminology, normally sensory systems and "thought systems" within our brain generate a number of (bottom-up) inputs that compete for control of the global neuronal workspace (GNW), and some process of top-down attention picks inputs that get strenghtened until they dominate the neuronal workspace. In meditation, the practicioner seems to train their attentional network into only choosing a specific set of stimuli (e.g. their breath, a mantra, the sensations of their body, etc.) and ignoring all the others. As they concentrate on these stimuli, those get transmitted into all the brain regions that receive input from the GNW. Since this is an abnormal input that most of the systems can't do anything with, they gradually get turned off - especially since the it doesn't matter what output they produce in response, as the successful meditation practicioner pays no attention to it. Of course, it will take a lot of practice for a practicioner to get this far, since the brain is practically built to "get sidetracked" from meditation and concentrate on something more important.

It's interesting to ask why this would lead to perceptual changes, such as an increased tolerance for pain. A straighforward guess would be that if the GW/GNW gets taken over by a very simple stimulus, and that stimulus gets broadcast into all the different systems in the brain, then there are systems related to learning that can't help but to analyze the stimulus. If a meditation practicioner consciously begins to break a sensation into smaller and smaller components, or begins to note and name individual sensations, then the implict learning systems will pick up on this and learn how to do it better. Also, as the meditator forces his brain to analyze very simple inputs, the brain allocates disproportionate computational resources into analyzing them and begins to find in them increasingly subtle hidden details - which the meditator then dismisses, forcing his brain to go to even more extreme lengths to find something. Over time and with enough practice, he learns to feel and notice these subtle sensations even when not meditating.

Of course, it's a bit of a misnomer to talk about the brain "finding" subtler sensations, since those sensations are themselves also generated by the brain. Rather what's happening is that there is a hierarchical process in which simpler inputs get increasingly complex layers of interpretation applied on them, and meditation strips away those layers of interpretation. Thus information that's usually thrown away during earlier processing stages becomes revealed and accessible to the conscious mind. That'd my guess, anyway. It's also interesting to note that savant abilities are also hypothesized to be created via having access to lower-level brain processing, but so far I haven't heard of anyone becoming a genius savant through meditation, even if it should be theoretically possible.

As I noted the last time, there's still the puzzle of how the attentional networks find out about an input that might be worth promoting into the GNW, if the GNW is already dominated by another input. A hypothesis that might make sense is that we're actually rapidly cycling a lot of content into and out of consciousness, and the attentional networks decide which stuff gets the most "clock cycles" (here's an obvious analogy to operating systems and multiprogramming). E.g. this text gets processed within the GNW, then I hear a sound coming from outside and that input pushes its way to the GNW for a brief moment, and then an attentional system decides that it isn't important and gets back to the task of writing this text. While the outside noise has pushed the text out of the GNW, it's still locally active in the brain regions that were most heavily involved in processing it, and the attentional network can home into the activation in those regions and strenghten it again.

Alternatively, this whole hypothesis of swapping stuff in and out might be unnecessarily complicated, and there could just be cross-region communication that wsan't conscious. There are a number of results saying that cross-modality integration of sense data can happen without consciousness. E.g. in ventriloquism we see a talking puppet mouth and hear sound coming from the puppeteer's closed mouth. Somehow this conflict gets resolved into us hearing the sound as if it were coming from the puppeteer's mouth, without us being consciously aware of the process. Also the results of the paper below, which suggest that attention and consciousness can both occur without each other, would support that hypothesis.

None of that actually has anything to do with the third session, though - it's just stuff that occurred to me while thinking about some of the seminar papers in general. So let's get to the actual topic...

----

The third Neuroinformatics presentation covered Giulio Tononi & Christof Koch (2008) The Neural Correlates of Consciousness: An Update. Annals of the New York Academy of Sciences. The paper was pretty packed with information, and there was a lot of interesting stuff mentioned. I won't try to cover all of it, but will rather concentrate on some of the most interesting bits.

In particular, the previous Neuroinformatics papers seemed to come to close to equating consciousness and attention. If input from our senses (or from internal sources like e.g. memory) becomes conscious if it is chosen to be promoted to consciousness by attentional processes, does that mean that we are conscious of the things that we pay attention to? Subjectively, I'm often conscious of experiences that I try to direct my attention away from, though that might just mean that a top-down attentional mechanism is competing with a bottom-up one. Introspection is notoriously unreliable, anyway.

Tononi & Koch argue that the two are not the same, and there can be both attention without consciousness and consciousness without attention. Let's first look at attention without consciousness. Among the studies that they cite, Naccache et al. (2002) is probably the easiest to explain.

The experimental subjects were shown ("target") numbers ranging from 1 to 9, and had to say whether the number they saw was smaller or larger than 5. (They were not shown any fives.) Unknown to them, each number was preceded by another ("priming") number, hidden by a geometric masking shape. In some versions of the experiment, the subjects knew when they were going to see the number, and could pay attention around that time. In other version, they did not, and could not focus their attention specifically at the right window in time. When the subjects were paying attention at the right time (and therefore also paying attention to the priming number), there was what's called a priming effect. Their reaction times were faster when the prime number was congruent with the target number, i.e. either both were smaller than 5 or both were larger. When the numbers were incongruent, the reaction times were faster. When the subjects couldn't focus their attention on the right time period, the priming effect didn't occur. Tononi & Koch interpret these results to mean that there can be attention without consciousness: the priming numbers were always seen too quickly to enter conscious awareness, but they caused a priming effect depending on whether or not the subjects paid attention to them.

The opposite case is consciousness without attention. There are experiments in which the subjects are made to focus their attention to the middle of their visual field, and something else is then briefly flashed in their peripheral field of vision. Subjects are often capable of reporting on the contents of the peripheral image and performing some quite complex discrimination tasks. They can tell male faces from female ones, or distinguish between famous and non-famous people, even though the image was (probably) flashed too briefly for top-down attention to kick in. At the same time, they cannot perform some much easier tasks, such as discriminating a rotated letter "L" from a rotated letter "T". So at least some kinds of consciousness-requiring tasks seem to be possible in the absence of directed attention, while others aren't.

Tononi & Koch conclude this section by summarizing their view of the differences between attention and consciousness, and by citing Baars and saying something akin to his Global Workspace Theory:

Attention is a set of mechanisms whereby the brain selects a subset of the incoming sensory information for higher level processing, while the nonattended portions of the input are analyzed at a lower band width. For example, in primates, about one million fibers leave each eye and carry on the order of one megabyte per second of raw information. One way to deal with this deluge of data is to select a small fraction and process this reduced input in real time, while the nonattended data suffer from benign neglect. Attention can be directed by bottom-up, exogenous cues or by top-down endogenous features and can be applied to a spatially restricted part of the image (focal, spotlight of attention), an attribute (e.g., all red objects), or to an entire object. By contrast, consciousness appears to be involved in providing a kind of “executive summary” of the current situation that is useful for decision making, planning, and learning (Baars).

As has often been the case lately, I wonder how much weight I should actually put on these results. A study that has not been replicated is little better than an anecdote, and while Tononi & Koch do cite several studies with similar results, there have been previous cases where the initial replications all seemed to support a theory but then stopped doing so. So for all that I know, everything in the paper (and the previous papers, of course) might turn out to be wrong within a few years. Still, it's the best that we have so far.

Like some of the GWS/GNS papers, this one also suggested that non-dreaming sleep involves reduced connectivity between cortical regions, and the regions communicate in a more local manner. That's also interesting.