gusl: (Default)
I would like to see some neuroscience research as to why nervousness/anxiety seems to cause cluttered speech / stuttering. Does attending to others' perceptions of you use up resources, like a musician who can't keep up with the fast tempo? Does an analogous thing happen with musical performances? Do people improve when they don't have to look at the audience? (I know that many stutterers have no problem singing, probably because it involves no improvisation)

This must be related to speech production being one of those things that isn't 100% deliberate.
gusl: (Default)
In the last 2 hours, I created a little shell app in Python, which presents flashcards for vocabulary learning. I'm happy because I never touched Python code before.
* you call the app passing your username and the vocabulary file
* it presents a random word in L1, and then asks if you remember the corresponding word in L2
* it records your responses and reaction time
* when you quit, it serializes this information to a file (rather than using a text file... I'm wondering if this will matter)

* respond to single keystrokes (spare my poor Enter key). DONE
* use machine learning to model forgetting, and the spacing effect
* optimize the flashcard schedule accordingly, to maximize retention
* eventually, use information like POS, semantic relations, word length

I suspect that an optimal active learning schedule (to minimize uncertainty about the user model) might be pretty close to the schedule that maximizes retention.

Here's a neat reflective feature of Python: by using 'input', you can let the user ask the value of any variable by naming it.
gusl: (Default)
Can blind people perceive race? What about those who were born blind (who learn about race in a mostly unsupervised way)?

How does their conceptions of race affect their attitudes?

Are there cases of former racists who lost their prejudice because of going blind, thereby realizing that they really couldn't tell the difference afterall? "OMG! My best friend is a X!"
gusl: (Default)
Yesterday, after the advice-seeking, I presented David with my ideas for automatic model induction.

He drew a line on the whiteboard. On the leftmost end, he drew a bag with different cognitive-model-types (neural networks, ACT-R, Bayesian networks); at the right end, fully instantiated models.

His question was: how far to the left are you willing to push this idea?

This was a good question, to which I did not have an answer.

He thinks my idea is very exciting, because human expert modelers prune the search space too much (often to a single model, i.e. single hypothesis), and stick with it as long as the data isn't enough to reject it. Given the huge underdetermination in psychology, this methodology is pretty far from optimal. Also, scientists tend to stick to their paradigm, always using the same kinds of model. By searching through many kinds of models, my idea has the potential to improve methodology.

However, this model search is computationally a very hard problem, unless I specify more constraints. My usual answer to this would be: let's do some task analysis, and copy what humans do. However, in this case, he would say that we don't gain anything by having computers do the work. But I think a halfway is possible: yes, by using heuristics, we do lose some generality, but since computers can crunch more data than humans, they can look through a wider range of models, and this way improve the quality of models that are being proposed today.

Another issue was how experts select among the multiple well-fitting models. A lot of tacit knowledge goes into this (sometimes, you need to have read an obscure paper in order to prefer or disprefer a certain model). There was no proposal to automate this.
gusl: (Default)
My view of natural language: there is a huge communicative gap between any two human beings: our thoughts need to be serialized in order to fit through the information bottleneck that are our communicative channels. This channel is 1D, slow and noisy. Things are not so bad, because one can usually (but not reliably) convey the concepts that one is thinking of, given enough time and a collaborative communication partner. But we can and should overcome these limits.

Consciousness is a spotlight. So is speech. I can give you (or myself) a tour of the structures in my mind (associations, explanations, complex theories1 2) by means of a narrative. But wouldn't it be great to do it in parallel, as happens in "The Matrix"3? This communicative limitation of ours is analogous to having PCs that can only copy CDs the way VCRs used to copy VHS tapes: by playing the whole thing.4

I have a lot of concepts and background knowledge in common with the rest of humanity, but unless they speak one of a handful of languages, we will not be able to communicate. It's like trying to look something up in a book where the index is gibberish.

But even when two people speak the same language, they are restricted to using words and constructs that are available in their common languages. More fundamentally, this 1D channel doesn't allow one to convey associations and feelings that appear briefly in one's consciousness (thought is faster than speech), although filmmakers sometimes do a pretty good (if extremely expensive) job of it.

But why does the text medium copy the auditory medium? Legacy. Text has the potential to be much more. I believe that we now have the potential of inventing visual languages that provide much more natural representations of our thoughts, for the sake of better human-human interaction.

1- someone once said that finishing a complex mathematical proof is like arriving at the end of long and winding road: even if you can retrieve each step as requested, you can't see it all at once.

2- as represented by argument maps.

3- To be nitpicky, Neo's auto-lessons taught him procedural, not declarative knowledge.

4- this analog analogy also models the introduction of noise. This is also analogous to the way that revisited memories get distorted.
gusl: (Default)
The advertisement for the Symposium on the Synergy between Implicit and Explicit Learning Processes, like all good Symposia, asks a few very general questions for people to think about. I would say it is the most "intellectual" and more interesting part of a conferece, where people try to compare and integrate different viewpoints and ideas from disparate fields.

I *really* should read the references listed on this one. It seems like a good basis to stand on.
gusl: (Default)
Schunn, Klahr - Self vs Other Generated Hypotheses in Scientific Discovery suggests that people are more skeptical towards others' ideas than their own. They also mention microworld discovery tasks, which is a lovely idea.
gusl: (Default)
Human communication is really lossy... even when the two people use the same logic and the same architecture. Maybe the goal of my formalization dreams could be couched in terms of bridging this gap: even if we both understand and use the same "logic of common sense", neither of us speaks a language that can express it easily. Hopefully, one day, natural language will seamlessly use metaphors from mathematics & programming languages (see Sussman's "The Legacy of Computer Science"). Not the mathematics & programming languages of today, mind you, but formal structures matching the common sense logic that we use in everyday life (I think that planning formalisms come close to what I want).

Why is it so hard to express oneself musically? I can hear beautiful music in my head, but it takes lots of training to communicate it to others, and even then there's a bottleneck. I can easily "see" a picture that I can't paint in my mind's eye. I can automatically recognize a known person's face, but I can't easily give this information to someone else.
I believe that this bottleneck lies in the brain itself: it's what happens when we convert information from parallel to serial. Since our communication channels are serial, communicating such "parallel" information with others requires us to first convert it to serial.

New media can do a lot to relieve many of these constraints, but I think that some of these constraints are fundamental.

Could one make a business out of creating software to let people express themselves and/or communicate better? What about software for people who have communication disorders?

Btw, has anyone modeled the tip-of-the-tongue effect? This seems exactly like the kind of thing that would not exist if our brains were purely serial. While in some examples of recognition-is-easier-than-production tasks (see also one-way-functions), one may accept several possible false matches, this does not seem to be the case with the tip-of-the-tongue effect (only the right word will satisfy the person).

See also: Thinking the Unthinkable


Related to this issue of self-expression, I will soon become an emacs wiz.
gusl: (Default)
The Distorted Tunes Test (via [ profile] patrissimo) is probably useful in music cognition research.

I got 26/26, unsurprisingly. I'd probably have some trouble with a rhythm test, though.

I've met people who call themselves tone deaf. I wonder how they would score here. So maybe people who can't sing in key aren't being reckless: maybe they just have bad pitch perception.
gusl: (Default)
I tend to be overly critical of myself.

I have been looking at evidence from my past (videotapes, old notes, printouts of my previous website), and they don't seem nearly as silly or embarassing as I remember them. In fact, I'm continuously delighted at the interesting ideas this kid had (I'm biased, of course)... in fact, I have to go back to freshman year of college to find something I now consider really silly. But maybe this could be a selection effect? I'll tend to remember better memories that I exaggerated?

Anyway, it's funny how we distort our memories... I think Alex Ramonsky has written about why you shouldn't revisit old memories when you're in a negative mood: you end up seeing them in a negative light, and making them all a bit darker before putting them back...

Why do I tend to downplay my qualities and emphasize my shortcomings? I really should be more self-confident. Can I blame a perceived cultural pressure to be "humble"?

Btw, I hate people who give me ego boosters. It's like a drug, and I could become dependent on it, you know? Don't some people get narcissism that way? ...I think I've mostly made myself immune from it, though.
gusl: (Default)
Will a cognitive linguist criticize this, please?


Some dialectal features can be viewed as production rules. Individuals in certain areas tend to have these production rules, while outsiders don't. This account also explains second-language hypercorrection effects.


to be written


/haus/ ("Haus") -plural-> /heuzer/ ("Häuser") -> [hoizer]

Portuguese in Brazil

In Brazil there are two independent, regional phonological phenomena, that interact with each other. This divides the country in 4 parts, providing us with a perfect combinatorial design for testing our hypothesis.

Phonemes in Brazil:

Raising is the dominant rule, in Portugal as well as in Brazil. Raising makes:
/bate/ -> /báti/

Other regions lost raising, due to Italian influence:
/bate/ -> /báte/

Affrication: /ti/ -> [tshi], /di/->/dzhi/
e.g. /tia/ -> [tshia]

Affrication did not come from Africa, but from local Indian languages. The plosives /t/ and /d/ become the affricates [tsh] and [dzh], when followed by an [i]. This is similar to what is observed in Japanese ESL students, and possible something in history of US English (the word "Acadian" somehow became "Cajun").

This rule probably had the majority of Brazil by 1920, despite resistance in the Northeast. The Italian immigration to the South of the country probably erased the affrication rule in some places in São Paulo and Rio Grande do Sul.

The results I expect:

Looking at different regional pronunciations:

Raising +, Affrication +: [bátshi] (Rio, Salvador, most of SP)
Raising +, Affrication -: [báti] (Recife)
Raising -, Affrication +: [báte] (some speakers in SP, RS)
Raising -, Affrication -: [báte] (some speakers in SP, RS)

When there is no raising, the affrication rule does not apply to the word "bate". We conclude from this that the affrication production only gets executed after the raising production.

Of course, it might be the case that these phonemes are incommensurable. Maybe speakers of different dialects are not thinking of the same phoneme. But I'd like to avoid going too deep into philosophy here.
gusl: (Default)
Highly recommended: Michael Huemer - Why People Are Irrational about Politics (even if I disagree with his moral objectivism at first sight)
My highlights:
Read more... )


I have an idea for a system for collaborative knowledge construction by skeptics, meant to avoid bias.
Read more... )


Here's a scarier link about techniques of persuasion, manipulation, hypnosis, etc. He characterizes the US Marines and revivalist churches as "brainwashing cults". Persuasion and Brainwashing Techniques Being Used On The Public Today


finally, via Google Ads:
The Theseus Learning System. Maybe I can make some money this way: selling software for critical-thinking education / idea refinement / writing. But my real interest is to create systems to enlighten real debates.

Let me be almost original and invent the phrase "epistemic hygiene".
gusl: (Default)
Björne, P., and Balkenius, C. (2003). Autism as an attentional disorder. In Proceedings of the XIth European Conference on Developmental Psychology. (apparently only the abstract is available)


Attention is probably one of the most fundamental abilities for learning. It has been proposed to encompass parts such as disengagement from current focus, orienting to new target, reengagement at the new target, as well as selective and sustained attention. An early attentional deficit will possibly account for many of the impairments associated with autism. Research has not yielded consistent results as to which part of attention is specifically impaired in autism. One way to catch the qualitative attentional impairment is to analyze videotapes of persons with autism performing daily tasks, correlating these results with instruments designed to measure their general cognitive abilities, which display a characteristic, uneven profile. Then it is possible to design experiments that can properly tap the attentional abilities and deficits in persons with autism. We expect to find a deficit in voluntary attentional shift to expected target, perhaps due to a lower level of attentional engagement. We also expect to find difficulties in orienting and reengaging attention, which could in part account for the fragmentized understanding of the world and the difficulties in generalizing knowledge across domains. Besides direct experimental evidence, these hypotheses can be tested in computational simulations of a model of context processing.

Are autists less able to control what they pay attention to? How much autistic behavior is the consequence of being concentrated all the time?

Are autists similar to absent-minded mathematicians who never stop thinking about a problem, but since birth? I can imagine that an attentional problem in early development would be enough to cause most autistic symptoms.

I wonder if teaching math to young children could cause autistic-like behavior. Just imagine their wild imaginations applied to problems in topology...
gusl: (Default)
One thing that's weird in this country is how long it takes some people to update their beliefs.

For example, many people will pigeonhole me as a foreigner (whether by my appearance or by seeing me speaking English), and insist on speaking English to me, and will not believe that I speak Dutch *even*after* I speak several sentences of good Dutch. Maybe this is due to confirmation bias: people only hear what they want to hear.

On the other hand, if people pigeonhole me as a Dutch person (by hearing my accentless speech), they won't believe me when I say that I don't understand something... and will have no pity on my poor understanding, and speak too fast.

Why don't people have an in-between category?
gusl: (Default)
I'm constantly playing music in my head, mostly Bluegrass, Celtic or other happy catchy tunes. Often, I'll hum it.

This was recently brought to my attention by my housemate's girlfriend, who said it was keeping her up (she must be more sensitive than me!). I was completely unaware of my vocalization, and didn't imagine that hear it could bother anyone (I'm aware that I do it, but not when I'm doing it). Now I have to consciously monitor myself.

I think this is more pronounced in the early mornings and at night. Sometimes I wake up singing... maybe it's related to feeling lazy. Or is it a way of distracting myself from the real world?
It is especially common when I'm thinking, programming (although not too hard).
Repetitive (rhythmic?) things, like cycling, put me in the mood for the music to start playing in my head.
Maybe this music is what keeps my attention inside my head. But why would I try to focus my attention inside my head?

I also think this is strongly correlated to jiggling my legs restlessly. But if I'm watching TV, I won't hum or hear music.

What sucks is when you hate the tune, but it's stuck playing in your head. But this is not often the case.

When does "hearing music" become hallucination?
gusl: (Default)
'Foreign accent syndrome' explained . Do you know anyone who sounds foreign even though they're completely local?
I knew a girl in college who sounded like she was from the South, despite the fact that she had spent her whole life in Connecticut. It annoyed her when even the foreign guy (me) noticed. I am of course *not* implying that people in the South are brain-damaged.

It would be very interesting to make cognitive models of speech production (which is harder than perception) for second-language speakers. Why do Indians render [w] as [v], while Brazilians render it as [u]? (both languages have both phonemes)
This research might even help actors.


By the way, some knowledge is neither declarative nor procedural: for instance the knowledge that recognize a face.
It's not declarative because you don't know how you do it, and you can't pass it on to someone else.
But it's not procedural because it's about perception, not action.
gusl: (Default)
Robin Hanson - Is Fairness About Clear Fitness Signals?

I've often struggled with the concept of "fairness" (distinct from "justice" in the sense of "being wronged", which is relatively unproblematic, and can be formalized in terms of (social) contracts). "Fairness" here is about what feels right, absent any agreements.

The concept of "blame"/"responsibility" is similarly problematic (e.g. "when can you blame someone? what about the environment where he grew up? his genes?"), but at least it has a reason to exist: assigning blame can prevent future problems.

"Fairness", OTOH, seems like a quirky concept with no clear purpose. Why is it unfair to compete against a disabled person, while it's fair to compete with (and merciless beat) someone who is chronically lazy? We say this is the case because the lazy person chooses to be lazy. So humans have a folk theory of free will. (never mind the free-will / determinism debates).

Anyway, Robin Hanson provides an interesting answer to the above, without going into this question of attributing "choice": instead he focuses on the hypothesis that the outcomes of unfair dealings are poor signals of genetic fitness... although this doesn't explain why unfair games can be repulsive. Is this because they waste people's time, not satisfying their immense curiosity about others' genetic fitness? or because unfair games could give deceptive signals? Could it be that people are repulsed by the sight/knowledge of others struggling helplessly?

What about people who say that sweatshops are "unfair", even when they are a good deal for everyone involved? Are they framing the situation as a competition between boss and worker? Does it bother them that rich and poor are interacting in a capitalistic way?
(I'm reminded of a link about this: a theory of how things get "morally contamined" by association. Maybe it was a response to Amitai Etzioni)
gusl: (Default)
Why do some silly problems really stump people?

The Psychology of the Monty Hall Problem: Discovering Psychological Mechanisms for Solving a Tenacious Brain Teaser

What about this one? What the heck are people thinking?
Three friends, bought a recorder with $30. each of them participated with $10. The shop assistance realized after they left that the recorder is $25.. He sent after them a guy with the $5. The guy gave each of the friends a dollar, so it means that each of them paid $9. Two dollars are left with the guy who took the $5. Now the three friends paid each $9, so we have now $27 adding to them the two dollars which are left with the guy. we find that we have $29.
We know that they paid $30, so where the dollar disappeared.
[16:13:08] Gustavo Lacerda says: each of them has $1, the thief has $2, and the shop owner $25
[16:13:17] Gustavo Lacerda says: nothing disappeared
[16:16:33] Gustavo Lacerda says: " Now the three friends paid each $9, so we have now $27 adding to them the two dollars which are left with the guy. we find that we have $29."
This makes no sense. I would like to know why this argument is tempting to a non-mathematician.
gusl: (Default)
Robert Hecht-Nielsen presents his theory of cognition

For him, the fundamental mechanism of cognition is what he calls "cogent confabulation":
Hecht-Nielsen noted that the common method used in search engines, data mining and drug trial analysis -- maximum a posteriori probability -- is not the mechanism of cognition. "Humans and animals don't do this," he argued. "Instead, animal cognition maximizes cogency, and in a non-logic environment, cogency maximization implements what I call the 'duck test': if a small animal waddles like a duck, swims like a duck, quacks like a duck and flies like a duck, we conclude that it is a duck because that is the conclusion which most strongly supports the probability of the assumed facts being true."

Here is a nice summary about him and his theories.


gusl: (Default)

December 2016

18 192021222324


RSS Atom

Most Popular Tags

Style Credit

Expand Cut Tags

No cut tags