NUS Home


Last Updated: Thursday, 07. August 2008
For comments, suggestions, or to report typo/grammatical errors : please email stevenhalim at gmail.com
This document is going to evolve over time. So please come back for more interesting stuffs later. :)

My Findings
Interesting Things that I've 'Found' From My Research

People said that once you have your PhD title (and even during your PhD candidature), others (will) start listening to what you said..., especially if you put a label "I have researched that... bla bla bla". I believe this is quite true... please read the articles in this page and see whether you agree with me after reading them. Of course, please counter these findings if you happen to disagree :).

My research involves the usage of visualization, and visualization itself requires human. I have explored some stuffs about how human visual perception and brain cognitive system works... For the majority of this page, I share with you several interesting things that I've found so far about this "human factor" issue, written in a down-to-earth manner... I also have observed several other stuffs that many people find surprising or even magical but actually can be explained scientifically.

If you find out that what I said is true and you want to use the findings, please feel free to do so. If you want to suggest/add something to this page, my e-mail address is listed above, send me an e-mail :).

Disclaimer:
1. This list is not exhaustive!!
2. There are still so many things that scientists don't understand yet...
3. I don't list down stuffs that are deemed as too "high-level"...
4. The correctness of some parts of my claims below are still debate-able.
5. The 'findings' presented here are works done by other researchers, I'm just rephrasing them in writing...
6. I admit that are some/rare individuals who can do more than what normal people can do.
   e.g. computing long digit numbers with their eyes closed...

The Topics:

Human Memory
1. Why do we easily forget something that has just been told recently??
2. Once we have memorized something, is it true that we can't really "forget" it??
3. Why it is difficult to compute numbers with our eyes closed??
4. Why do we remember more things that we see than what we hear/smell/etc??

Decision Making
5. Is it true that human cannot make optimal decision??

11. When is the best time to make a decision?

Human Capabilities
6. Is it true that (now) ALL human's capabilities can be emulated/bettered by machines??

7. How to maximize human visual perception strength??

Probability
8. Is it true that the world is small??
9. Why over assumptions are dangerous?

10. Want to bet that you can find someone with same birthday as you in a room of 23 people?

Aviation
12. Why a commercial aircraft takes so long to descent for landing?

13. Why the aircraft runway is only one directional and what is the consequence?

1. Why do we easily forget something that has just been told recently??

Do you ever encounter a situation where you say: "erm... I forgot", even though you have just been told about it recently?

For example, your friend Andy introduced you with his friend Brian, and then after talking for a while, you accidentally forget that the name of Andy's friend (which is Brian)...

I bet you have experience similar like this one...

Human have several types of memory, psychologist classify them as short term memory (volatile, limited size), working memory (to process computation/to do cognition, limited size), and long term memory (more persistent, very big size, can be as big as the capacity of thousand books). There are many other proposals of types of human's memory, but that is beyond the scope of this writing.

When we first receive an information, we will store it (subconsciously) in our volatile short term memory. In fact, our short term memory can only store up a "queue" of 7
± 2 chunks of items at one time. If we receive another information, we will need to "delete" some information from our short term memory before we can store the new information (usually older one first, but sometimes random... so I can't say that our short term memory works like a perfect queue). Once the information is lost from the short term memory, it is lost forever.

The information stored in our short term memory decay over time. 10 minutes after listening to a long sermon/presentation, we will only be able to recall roughly 10 points from the talk. 1 hour later, perhaps 5 points... 1 week later, maybe only 1 point or you may already forgot the whole thing...

It is only after we internalize, process, memorize, relate this new information into our long term memory, then recalling this information in the future is possible.

This is why, you "forget" the name of Andy's friend. You have internalized the name "Andy", but you haven't store "Brian" in your long term memory. During the conversation with Andy and Brian, many information request for a place in your short term memory. By chance, the term "Brian" is deleted.

Knowing this fact, what can I do?

-.
If the information is important, write it down before you lost it.
   Later, revise your notes and integrate it into your long term memory.

-.
After getting new information, quickly integrate it to your long term memory.
   e.g. by repeating your friend's name in your heart, focus your concentration on the topic so that
   the information is not "overwritten" by other non-important information, etc.

-. If you need to temporarily store information, store it in chunks.
   Don't store a hand phone number 94220587 as 8 digits, but as chunks, e.g. 9422  -  0587 (2 chunks).
   You can remember more temporary information by doing this...

-. If you know you are going to receive a stream of information (e.g. a sermon, a presentation, etc),
   record/note it down/use tape recorder. You won't be able to remember all!!

-. For the presenter, "purposely say a wrong thing and then correct it" is a trick to force his/her audience
   to "integrate" it to his/her long term memory. But don't do this trick too often otherwise it will be
   annoying... :). Just repeat/emphasize the important points frequently throughout your talk...

-. Expecting someone to remember the outline of our talk is "almost impossible"
   If possible, give the audience the handout/outline of our talk (hardcopy) or
   repeat the outline page during the presentation to assist the audience or
   Upload your presentation files to Internet so that your audience can download and review it again.

-. When you want to convey a message to others, try to make it short and simple...
   Complex information which requires more storage than what your audience's short term memory
   can handle is very confusing.

2. Once we have memorized something, is it true that we can't really "forget" it??

If you have memorized something in the past..., then unless your brain cells that stored the long term memory are damaged, it is likely that when you say you "forget" an information, it is just because you "don't know where you stored the information in your brain".

Want some example?

Assume that you attended a Sunday service (a presentation if you are not attending churches) last week. The pastor (or presenter) gave a very catchy illustration about an old married couple, who keeps calling each other "honey" despite their old age. You think that this illustration is interesting and you discuss it with your friends over lunch.

Now (1 week later), Your friend ask you:
"Hey, do you remember the interesting illustration that our pastor told us last week? I forgot"...

You said:
"Erm... I forgot too..."...

Then, your friend mention:
"Is it about 'old couple' doing something?, but I'm not sure what is that thing"...

Then, out of nowhere, the information pops up again in your brain and you said:
"Oh yeah yeah... Our pastor told us about old married couple that keeps calling each other "honey" because they actually forget the name of each other... hehe that was a funny illustration."


We can recall information from our brain by using proper cues. This cue can be a 'keyword', or a 'situation', or an 'environmental setting', etc...

We have seen many cases where the witness of a crime scene, who is usually traumatized and 'forgot' the event, is brought back to the actual crime scene to help him/her reconstruct what is actually happening on the scene. The police are giving cues to his/her brain to recall the "forgotten" memory.

We don't actually "forget" something...
With proper cues, we can recall it...

Knowing this fact, what can I do?

-.
If you sure you have remember that information before and now you "forget" it...
   (e.g. during exam, especially the memorizing-type exams), try to recall that information
   by asking/suggesting several keywords/cues to your brain...
   It is there in your brain, but you just can't recall it yet...

-.
Go to the same place/create the same environmental settings to revive old memories.
   (e.g. reconstructing your romantic first date with your wife/husband/girlfriend/boyfriend
   to re-strengthen your relationship)

-. Use learning techniques that have been proposed by human-brain researchers,
   e.g. Mind Map (popularized by Tony Buzan).

3. Why it is difficult to compute numbers with our eyes closed??

Try it, memorize this: 3248734221*23443232/432342-23439 (store it into your long term memory by repeating the numbers many times), then close your eyes and finally, try to calculate it...

Use a calculator to verify your answer later... if you have one... :)

I bet that you will have a lot of difficulty trying to do the computation and give up before you do. You will be able to do it anyway, but with a lot of effort. This is because our working memory is also limited. We need to store temporary calculations (which is required for computing the example above).

To alleviate this problem, we need cognitive aids/tools such as: pencil and paper to write down these temporary values. With cognitive tools, we can do more.

In fact, my research on "Information Visualization" is about studying how to maximize the human ability of visual perception and cognition so that we can do/infer more information than what we can achieve now.

Knowing this fact, what can I do?

-. Many intellectual works are done with cognitive tools: pencils, papers, calculators, computer...
   We need to go beyond our limited working memory...
   Use these tools, don't be afraid to use them :)

4. Why do we remember more things that we see than what we hear/smell/etc??

Our vision has the biggest bandwidth to our brain. 20 billion++ neurons (they works in parallel) of the brain are devoted for analyzing visual information. Therefore, we acquire much more information through our vision than through all of our other senses combined.

Since more information is coming from vision than from other senses, it is quite logical that more visual information will be recorded in our long term memory.

Human also has something called "visual memory". Try to recall back a beautiful image that you've seen from your holiday trip to... for example, to Vienna last year. You can do it... (try other images if you never travel anywhere, e.g. try recalling the face of your father/mother/wife/husband, etc.)

Now, can you re-construct what is the flavor of a very delicious cake that your mother cooked for you last year?

Knowing this fact, what can I do?

-. When you want to convince/persuade people of some ideas...
   Use MORE pictures/graphs/visualizations rather than adding more verbal words...
   They will "accidentally" remember it more :)

-. Be careful of what you see, e.g. pornography...
   Human visual system is quite powerful and we know that we can't actually erase this information
   from our long term memory (we can only choose not to recall it).

5. Is it true that human cannot make optimal decision??

Human can differentiate good options with the obviously bad options easily, but human will have difficulty when presented with a task to choose the optimal one, especially if the problem is inherently complex.

Human is not good at measuring an absolute goodness of something. But if given the baseline B, human can use it to compare and to say whether the new item X is better or worse than B, provided that the differences are clear enough.

I don't know the actual scientific reason why this is so, but here are some possible reasons:
-. Human's common sense/preference can actually bias his/her decision.
-. Human is in general risk-averse, taking a decision which is 'safer' but not 'optimal'.
-. Human is inconsistent, when tired, he/she just choose a decision without thinking further.

Knowing this fact, what can I do?

-. Accept the fact that some human judgments are not optimal...
   e.g. soccer referees often make mistakes, etc

-. As human can't really judge optimally, give human a means for comparison (baseline).
   Highlight the differences between the given option and the baseline.
   e.g. when presenting your company's improvement to your boss...
   don't just give him/her current year's numbers...,
   but show him/her last year's numbers and compare it with this year's numbers...
   it is much more convincing that way.

6. Is it true that (now) ALL human's capabilities can be emulated/bettered by machines??

With the advent of machines, many human's task are transferred to machines. Computer/machine is undoubtedly powerful, faster, never tired, consistent, etc... Machines are much better in these aspects than human... Some people generalize this fact and saying that one day everything that human can do will be bettered by machines...

At least for now, NOT ALL capabilities of computer can be emulated/bettered by machines yet. In fact, there are still many examples where human are still superior than TODAY's machines!! I can't prove what will happen in the future, but at least for now, the question in this section is not yet valid.

Here are two examples (there are more) that cannot (or possibly never) be done by computer yet:

1. How our brain works:

Olny srmat poelpe can raed tihs. cdnuolt blveiee taht I cluod aulaclty uesdnatnrd waht I was rdanieg. The phaonmneal pweor of the hmuan mnid, aoccdrnig to a rscheearch at Cmabrigde Uinervtisy, it deosn't mttaer in waht oredr the ltteers in a wrod are, the olny iprmoatnt tihng is taht the frist and lsat ltteer be in the rghit pclae. The rset can be a taotl mses and you can sitll raed it wouthit a porbelm. Tihs is bcuseae the huamn mnid deos not raed ervey lteter by istlef, but the wrod as a wlohe. Amzanig huh? yaeh and I awlyas tghuhot slpeling was ipmorantt! if you can raed tihs psas it on !! (--- credit to someone who created this paragraph, perhaps from Cambridge University)

Current Artificial Intelligence (AI) and Machine Learning techniques is trying to push the capability of computer programs to emulate human's intelligence. However, we can say that today, we are still far from achieving what our brain can do now.

2. How our visual perception works:

Our visual perception system is a very strong pattern-finding machine. Pattern seeker of enormous power and subtlety. Here are some examples:


In reading distorted words --- (Gimpy, credit to CAPTCHA project)
Computer Optical Character Recognition (OCR) technologies are often confused.


In relating information from disparate pictures --- (Pix, credit to CAPTCHA project)
The common object from the 4 pictures above is: worm.
No known computer algorithm to do this yet...


In finding patterns in the figures (TSP tour crossings)
Computer needs to do a lot of n^2 line intersection tests to do so.

Knowing this fact, what can I do?

-. Combine the strengths of human and machine in many aspects :)
   Even though it is acknowledged that human is the weakest link in human-machine system, human is
   also the most important player in this system. Proper combination may yield a powerful system.

   This idea of "human computation" or "human in the loop" or "human cycles" is used by others
   researchers too. In particular, I would like to highlight the works done by Luis von Ahn et al.,
   who have shaped the way we used Internet (remember the CAPTCHA questions when you register
   for something in the Internet??) and in enhancing image labeling (ESP Game, Peekaboom).

7. How to maximize human visual perception strength??

Human visual perception system is intelligently designed (by the Creator) to be able to perceive patterns if presented in a correct manner (e.g. using too many colors --- convoluted). If the information is not presented carefully, it will 'disappear'! The question now: what is the correct way to present information?

One way is to use highlighting to make some things stands out from the rest. This will assist our visual perception system to focus on things that are interesting. If presenting a list, e.g. a list of names, product ID, etc... do it in a sorted order, etc.

Knowing this fact, what can I do?

-. Present information in a way that maximizes human visual perception strength.

8. Is it true that the world is small??

You meet someone that you don't know yet. You introduce your name and vice versa. Then, both of you start explaining each other's background... Suddenly you (or him/her) jump out and say: "Hei.., so you are my friend's cousin?", or "Oh, you are my friend's friend", etc...

Therefore, both of you conclude that "Hm... so the world is indeed small"...

Is that true??
I beg to differ :)
Follow my reasoning below:

If you have 100 friends (the number varies, but to simplify my explanation, let's stick with 100).
And each of your friend has another 100 friends.
Then you have 10.000 2nd-degree friends (there will be overlaps, but to simplify, ignore the overlaps).
If your 2nd degree friend has another 100 friends.
Then you have 1.000.000 3rd-degree friends...

In a population of this 1.000.000 people, any random pair of people A and B is either:
1st level friend: A and B are friend of each other (A knows B),
2nd level friend: Both know someone called C (A knows C, B knows C), or
3rd level friend: A knows C, B know D and C knows D.

Of course, this 1.000.000 people are scattered all over the world but mostly concentrated in the place where you live. Therefore, if you met and introduce yourself to any stranger (not your 1st level friend, otherwise you won't call him/her stranger) within this population, you will know that this stranger is either your 2nd level friend or 3rd level friend.

In Singapore with 4.000.000 people, this kind of situation is quite probable to happen..., roughly there is 25% chance that any stranger that you met is your friends's friend's friend (3rd level friend).

Moreover, people tends to group with other people with similar characteristics: race, religion, families, friends, etc... This reduce the circle of people that we interact with, thus increasing the chance of encountering 'the world is small' paradox.

Even more, when you make friend with this stranger (making your 2nd level or 3rd level friend) as your friend (1st level friend), you have just expanding the range of your 3rd level friend from 1.000.000 to approximately 1.010.000 people, thus further increasing the probability of that you will encounter this issue again in the future.

Knowing this fact, what can I do?

-. The next time you meet a stranger who knows your friend/relative or your friend's friend,
    do not be too surprised. It is quite probable...

9. Why over assumptions are dangerous?

Every assumption that we add into a problem carries a probability of error p (and thus the probability of correctness if we adopt this assumption is 1-p). The more assumptions that you make, the more incorrect your understanding about the problem is.

For example, in  your exam, you are given a question. There are several possible additional interpretations for it, but the question doesn't clearly state whether that interpretation is true/false. In that case, I suggest that you don't introduce that assumption in your understanding.

If adding assumption A carries a 10% probability of error, adopting assumption A will make the probability of correctness of our understanding of the problem to be 90%. If we add more assumption B, again B carries a 10% probability of error, having both assumptions A and B will cause the probability of correctness of our understanding to be 90%*90% = 81%. The more assumptions, the more likely we are making wrong understanding...

Therefore, we should avoid making unnecessary assumptions. In Occam's Razor principle, we learn that among all possibilities to interpret something, pick the simplest that works, shaving away unnecessary assumptions.

Knowing this fact, what can I do?

-. In interpreting exam questions... Don't assume funny stuffs if it is not stated.
-. In relationship, don't assume too many things about your boyfriend/girlfriend.

10. Want to bet that you can find a pair with same birthday in a room of 23 people?

Don't bet "you can't", but bet "I can". Actually, the chance of finding a pair with same birthday in a room of 23 people is greater than 50%!! This may seems a paradox... You thought it should be smaller than that right?

1 year is 365 days. To have someone that has the same birthday as me. The room should contains 365... This is what people commonly believe. And it is correct... as long as the statement is "same birthday as me". When you change this statement into "any pair of people" in that room, the situation changes. It has been shown that with only 23 random people in a room, the chance is greater than 50%.

Knowing this fact, what can I do?

-. Within your circle of friends of size > 23, say... you have 50 friends, don't be too surprised
   if there are several pairs with same birthday... Want to test this? Check your friend's birthday list now!

11. When is the best time to make a decision?

Some people tends to make hasty decisions (including me), but actually this is not optimal. The best time to make a decision is not just the time you can think slowly and carefully. The best time to make a decision is the last possible moment before the options are gone/change...

The reason is simple. After making/committing a decision, it is usually hard for us to change it anymore. Over the time, new options may come and if they are better, you will regret your past decisions if you commit too early... This, unfortunately, happens many times.

If you know you are not ready to marry someone before 25 years old, it is not 'logical' to commit a relationship with someone when you are just 20 years old. It is more likely that in the next 5 years you will meet someone better than not. The 'mistakes' done by many couple in this 'logical' concept have caused 'unnecessary' pains due to break-ups.

If you know that you are going to stay in a new city for 4 days and you want to buy some souvenirs for your family/friends, don't quickly buy souvenirs during your first day of stay. You'll regret your decision if you find another souvenir shops that sell cheaper or better goods. Wait until the 4th day before committing your decisions.

Of course, you can't delay your decision forever. At certain point... at the last possible moment... you need to commit your decision.

Knowing this fact, what can I do?

-. No haste in major decision making, including in finding soul mate, buying house/cars/souvenirs, etc...

12. Why a commercial aircraft takes so long to descent for landing?

When you board an aircraft, it usually rises up to 10.000 feet (30.000m) --- I like to watch the flight information, far above the clouds so as not to be affected by bad weather (below the clouds), etc.

I still don't know why an aircraft have to rise up to 30km and not just 10km or 20km... Here I want to explain why it takes so long to descent..., at least it is much longer than the time required for it to ascent to 30km.

Common scenario:

When the aircraft ascent, within 4-5 minutes, the aircraft will already reach high enough altitude and from there the aircraft ascent slowly to reach its desired flight altitude (at this point, the aircraft is just 'cruising' at constant speed along this 'airways'). But, 30 minutes before landing or more, the flight captain already announces everyone to fasten their seatbelt again as the aircraft is preparing for landing. If you observe the flight information, the aircraft altitude decrease slowly...

A question pops up in my mind: Why do we need 30 minutes or even more to descent if during ascent, it only takes 4-5 minutes to go up to altitude of 30km?

The answer is simple, and I guess most of you can figure it out before reading this... it is just because for the sake of comfort of the aircraft's passengers.

During ascent, all passengers already expect sudden change of altitude, anyway, moving upwards for short period of time is not as scary as going downwards for the same period of time, so the flight captain uses this short period of time to go up as high as possible by pointing the aircraft's nose upward.

In the contrary, human don't like to endure free-fall or even faster-than-free-fall situation... Maybe it is more to psychological effect, since going downwards, we are very aware of potential collision with the ground but when we go upwards, we don't see potential collision with the clouds...

Just some calculation, if we do free fall from 30km, the fastest to go from 30km to 0m (crashing to land) is about 77 seconds (using the simple physics formula h = h0 + v0.t + 0.5*g*t^2 and gravity constant 10m/s^2). Obviously, no one want to descend that fast, it will be a terrific nightmare if your aircraft goes down from 30km to landing in 77 seconds -_-'. (note that most people is okay if the aircraft goes up from 0m to 30km in 77 seconds).

The aircraft also cannot point its nose downwards too much as it will even be more nightmarish for the passengers. Imagine that now your aircraft propels you to ground from 30km in 20 seconds...

To reduce the complaints of discomfort by the passengers, the aircraft must descent slowly (the aviation term for this is 'approaching' airport). This can be achieved by altering the wings and/or slowing down the aircraft so that the lifting force from the wings are slightly lower than the overall weight of the aircraft and thus according to simple law of physics, the aircraft will slowly go down (note: the overall weight of the aircraft during take off and landing is different!! The aircraft will be lighter during landing because most of the fuel have been consumed during the flight...)

Knowing this fact, what can I do?

-. On board your flight, don't complain if the aircraft descent slowly. You will prefer it stays that way...

13. Why the aircraft runway is only one directional and what is the consequence?

Aircraft runway is a long, wide, and straight road of about 3-4 kilometers. There are two ends A and B. Why in every airport, all aircraft can only take off or landing in one direction only? e.g. from A to B only and never from B to A!

Answer: you don't want to see a head-on collision right?

By adopting one direction only for the airport's runway(s), we minimize the probability of fatal collisions (if any). However, follow up question pops up: if all airports have only one direction and the aircrafts can actually come from any directions, what should the aircrafts do because of this convention?

They simply have to travel in circles just before landing if they happen to arrive from wrong sides... Again, for the sake of the comfort of passengers, commercial airlines will make a big U-turn (rather than a sharp U-turn), aligning its nose straight to the runway, and then descent slowly. For bigger and busier airports, e.g. London Heathrow, this circling phase is more noticeable since now the aircraft not only have to align itself to the direction of runway, but it must wait for its landing time... The only way to do it is to travel in circle...

Knowing this fact, what can I do?

-. On board your flight, again don't complain if your aircraft seems already so close to the destination airport, but it suddenly turns away and make a big looping U-turn...

14. Other Ideas...

The speed of the car's left and right wheel differ when it makes a left or right turn. Why?
(the technical term: 'differential gear'/'gear train').
It is quite interesting to study 'gears' :)

References

Every scientific writing will not be complete or even debate-able if it doesn't include proper reference list. This list of references is here so that the interested readers can probe further on this interesting research results :). I copy several book cover images/other symbols from amazon.com, edwardtufte.com, and other sources in Internet as images are more easier to memorize than text.

  1. Sanders, M.S. and McCormick, E.J. 1993. Human Factors in Engineering and Design. McGraw Hill.

  2. Ware, C. 2004. Information Visualization: Perception for Design. Morgan Kauffman.

  3. Ahn, L.v, Blum, M, Hopper, N.J, and Langford, J. 2003. CAPTCHA: Telling humans and computers apart: http://www.captcha.net; http://www.espgame.org; http://www.peekaboom.org.


  4. Buzan, T. 2006. The ultimate book of mind maps. HarperCollins Canada.

  5. Schneiderman, B. and Plaisant, C. 2004. Designing the User Interface. Addison Wesley.

  6. Tufte, E. 1983, 1990, 1997, 2006. The Visual Display of Quantitative Information, Envisioning Information, Visual Explanations, Beautiful Evidence. Graphic Press.


This document, 21_my_findings.html, has been accessed 8882 times since 13-Nov-06 22:50:42 SGT. This is the 1st time it has been accessed today.

A total of 4387 different hosts have accessed this document in the last 6347 days; your host, ec2-54-81-58-140.compute-1.amazonaws.com, has accessed it 1 times.

If you're interested, complete statistics for this document are also available, including breakdowns by top-level domain, host name, and date.