Saturday, June 23, 2012

Automaticity, the Stroop, and Human Behavior

Think back to when you learned to tie your shoelaces; you needed to think carefully through each step of the process. Probably, you made errors over the course of multiple attempts. Now you probably do not even think about the steps, but simply initiate a series of movements that proceed without any further influence. There are many behaviors that followed this same pattern of learning: reading, writing, typing, bicycling, piano playing, driving, etc. When a behavior no longer requires direct, deliberate thinking to perform, the behavior is automatized.

The amazing thing about automatized behaviors is that many people state that they do not consciously know how the behavior is performed, they just can do it. This is typically referred to as unconscious memory or implicit memory. For example, try explaining how to properly maintain your balance while riding a bike. Difficult, right?

In some instances, a behavior can be so over learned that we do it automatically. Or, in other words, we do it without thinking. A great example of this is reading. For literate people, it's impossible to not read. What I mean by that is, if a random word were to flash on the screen, you could not stop your self from reading it. Even if the word were flashed on the screen for a fraction of  second so that it only appeared as a blip, you would have a better than chance probability of identifying the correct word if it were presented again alongside another random word.

Not only can behaviors become automatized, some behaviors can be more automatized than others. And when you have automatized behaviors compete against each other, you get interference, or the slowing down of a behavior. This concept was best and most famously demonstrated in what is known as the Stroop effect named at J.R. Stroop.  In Stroop's classic experiment, participants were slower to properly identify the ink color when the ink was used to produce color names different from the color of the ink. That is, participants were slower to identify red ink when it spelled the blue. What makes this finding interesting is that participants are specifically instructed not to pay attention to the word names and simply report the color of the ink. However, this seems to be a nearly impossible task, as the word name seems to interfere with the participant's ability to report only the ink color.

Below is an example of the Stroop task. First read the color-matched words (trial 1) and then name the colors of the X's (trial 2) as quickly as you can. Note how quickly you're able to complete each task. For the third trial, try NOT read the word, but rather name the ink color in which the word is printed.

The difficulty you experience completing the third trial relative to the other two is interference. The automaticity of reading is eliciting a response that is different from the one that is required (naming the ink color) and therefore competes with the response that your suppose to make.

Aside from this being a cool demonstration of automaticity, did you know that you can use automatic behaviors to influences the behavior of others? In an experiment done by Langer and colleagues, it was demonstrated how compliant people will be with a small request as long as they heard what sounded as if they are being given a sound reason, even though no actual reason is given. In the experiment, an individual approached people standing in line to use a copy machine and presented one of three requests:

1. Excuse me. I have 5 pages. May I use the Xerox machine because I am in a rush? (sound reason)
2. Excuse me. I have 5 pages. May I use the Xerox machine? (no reason)
3. Excuse me. I have 5 pages. May I use the Xerox machine because I need to make some copies? (non-sound reason).

When given a request plus a sound reason (#1), 94% of subjects complied with the request. However, when given a request with no reason (#2), only 60% of subjects complied. But when given a request plus what sounds like a reason (#3), but actually wasn't a reason, the compliance rate jumped back to 93%!

What people were responding to was not sound reasoning, but to syntactical structure: a request followed by what should be a sound reason. Most of our behavior, it turns out, is automatic. That includes speaking and thinking. In other words, when we are doing something, we are doing it without occupying our minds. This means that our minds are often free to engage in other, higher order, and more meaningful activities; like watching TV.

Wednesday, June 20, 2012

Saccadic Masking, Chronostasis, and Concsiousness

Time. What is it? To some, Time is a kid-friendly news magazine characterized by info-graphics and colorful pictures. To others, Time represents a continuum of continued progress of events from the past, present, and future. While Time is indefinitely continuous, its rate can appear quite variable, at least to humans.

Below is a clock face with an active seconds hand. Focus your gaze on something nearby (e.g., something just above your CPU monitor) then shift your gaze to the clock and watch time elapse for 5 seconds. What you should notice is that the first second to elapse appears to take longer than the subsequent 4 seconds (It's subtle, you may have to try this a few times to notice the illusion).
If you did notice time lag, then you just experienced chronostasis. Chronostasis is the experience or illusion that occurs immediately after a saccade that appears to stop time momentarily. This occurs so much that every day we experience 40 minutes of chronostasis, though it goes entirely unnoticed.

So what's going here? When our eyes move, the image reflected on to the retina is also in motion. This creates 
motion blur (see picture at beginning of this post). A blurred image being utterly incomprehensible (and of no use) to us sighted humans, our brains have a mechanism to circumvent the blur an create a comprehensible image. This phenomenon is known as saccadic masking. During saccadic masking, the blur is suppressed, along with visual processing, and the gap in visual processing that should be experienced as your eyes move from on side to another. The brain then replaces the blur with an image of the very next thing that your eyes fixate on.  

This explains chronostasis experienced in the clock illusion. As you shift your gaze to fixate on the clock, instead of seeing a incomprehensible blur, the blur is supplanted with an image of the clock that your currently fixated. That's why that first second to elapse appears longer than all other subsequent seconds.

The process is depicted below:

The first figure depicts what actually occurs, but only our subconscious brain perceives. Figure two shows what what we actually see. If you imagine that point two is the image of the clock, our brain then fills in the time that had motion blur with the same image. 

If the brain is replacing a past image with a current image, does that mean what I'm seeing is not really in the present but the past? Yes, in fact, human awareness or what we experience as the "present" is actually the very recent past; more specifically our consciousness lags 80 milliseconds behind actual events. This is how saccadic masking and chronostasis are possible; before we become aware, our brain has to make sense of stimuli first, which takes just about 80 milliseconds.

As it turns out, saccadic masking and the 80 millisecond lag in awareness explain many of the visual illusions that many may already be aware of such as the moving snake illusion. As you move your eyes looking at the image below, the illusion of motion is apparent. If you fix your gaze on a black dot, the illusion of motion will cease. 
There are other phenomenon as it relates to time perception: Time passing more quickly as we age and time slowing down as we become scared. Of course, the latter phenomenon should not be confused with time dilation, which is an actual difference of elapsed time (as opposed to perceived difference) between two events as measured by observers moving relative to each other. 

Friday, January 27, 2012

Al Sharpton is an Idiot

America's Panacea for racial Freudian slips, Al Sharpton, had a "heated clash" with a Republican congressman whose identity isn't important since, if he actually agreed with Sharpton, he wouldn't have been on his program in the first place. 

What caused Sharpton to contort his face in such unnatural manner was the following exchange: 
Sharpton: "Is it fair that billionaires pay a lower tax rate than their own secretaries?" To which the Republican replied, "Well they actually don't, according to the IRS." 

Now, if two reasonable people were having this discussion, the conversation might proceed like this, Sharpton:
"Will you please show me the IRS data about which you speak? So I may correct myself"; Republican: "Sure Al; and would you like some motherfucking ice tea?" 

Reality turn out differently, "
Sharpton started to raise his voice and asked the question again. 'Is it fair? Is it fair?' Huelskamp continued to deny the accuracy of Sharpton's question. 'It's not true, Al.' Sharpton then said 'on the basis of the report—if the report is inaccurate, fine, you stipulate that—I'm asking you, is it fair? Is the arrangement fair, in your opinion?'" 

There is only one answer that Sharpton is willing to hear: NO! He's not interested in facts; notice how he continues to seek an answer even after granting that the statement "billionaires pay less than their secretaries" might be false. 


1. Warren Buffett's "secretary" is an idiot.
 Some time ago, Mr. Buffett claimed that his secretary made $60,000 and paid a 30% tax rate. Here's the problem with that: our income tax system is progressive, meaning that we pay higher tax rates as our income gets higher. As the table below shows, there are six tax brackets.

Let's say our income (husband and wife jointly filing) is $112,000. This would put us in the 25% bracket. If the U.S. had a flat rate tax system (and no deduction), we would pay 25 percent of $112,000 in income tax, or $28,000 
(a common misunderstanding of how the tax system functions). Under our tax system we pay 10 percent on the first $17,000, or $1,700. We then pay 15 percent on the next band of income up to $69,000, or $7,800. We then pay 25 percent on the marginal amount over $69,000, for another $10,750 in taxes. When we total the taxes paid on these three bands of income it comes to $20,250, for an average (or effective) tax rate of 18.8 percent. The national average is 11% 
(Note that effective tax rate is what was actually paid. It will always be lower than the marginal tax rate).

According to the bracket for single filers 
(let's assume Mr. Buffett's "secretary" is single), she would have paid $830 on the first $8,300 of her income (10%) and then she would have paid $3,255 for the remaining $21,700 of her income ($30,000 - $8,300). A total of $4,085 would have been paid for an effective tax rate of 13.6% (This does not include deductions and credits, which would have likely lowered her rate to 7%). Clearly Mr. Buffett needs to educate his "secretary" on how to file tax returns; after which, he should fire her.  

2. Who decides what is fair?

How Much Should the Top 1% Pay Before It's Considered Fair?


The top 1% of income earners (those with incomes of
 $343,927 in 2009) paid a larger percentage of the income tax (36.73%) than the bottom 90% combined (29.5%). Also in 2009, the top 1% earned 16.9% of the adjusted gross income (that's down from 20% before the recession). 

On the other side of the coin are the non-payers. 
A non-payer tax return is one filed by an individual or couple who, thanks to legal credits and deductions, owes nothing. In 2009, a record number of individuals (58 million) paid no income tax. That is, 42% of tax filers were non-payers. Only 0.3% (2009 data, table 1.1) of those non-payers earn more than $100,000. 

I'll repeat that since it's mildly important: 42% of the bottom 85% of income earners (i.e., <$100,000) paid no income tax.


Moreover, many of these non-payers also receive "refundable" tax credits even though they have no income tax liability. For millions of these non-payers, these refundable credits exceed their total payroll tax (i.e., social security, medicare, unemployment) contributions. In
 2008, more than a third of all tax returns resulted in complete nonpayment; that is, people got back every dollar that was withheld from their paychecks during the year. 

3. Raising Taxes on Wealthy Won't Fix Anything

Earlier, congress wanted to impose a 5.6% surtax on those with incomes above $1,000,000. According to the CBO, that tax would raise $450 billion over the next 10 years. That's 45 billion a year. Last year's deficit (fiscal year 2011) was 1.3 trillion, the second largest deficit of any nation in the history of the world. After you subtract 450 billion from 1.3 trillion you get 850 billion. If you cut out all non-defense discretionary spending you save $530 billion, leaving our deficit at $320 billion. Only after halving securing spending, will last year's deficit go away.


4. Stop Blaming the Bush Tax Cuts

The CBO broke "down the major components of the $11.8 trillion swing from surpluses to deficits over the ten year period 2002 to 2011."  Higher spending turns out to be the largest factor erasing those surplus projections.


As the chart above shows, the Bush Tax cuts have been declining in significance as a contributing factor to the annual deficits while increased spending and recent tax cuts (e.g., payroll tax cut) are larger contributing factors to annual deficits.

5. The Elephant in the Room
Mandatory spending is consumes 60% of the annual budget. That Social Security (761 billion), Medicare (468 billion), and Medicaid (269 billion). And last*, but certainly not least is Defense spending ($660 billion), which is not actually mandatory spending, although it should be since no one wants to touch it. 

Until people start focusing on reforming these programs and these programs alone, all other talk is a waist of time.   

The information is out there, all you have to do is let it in...
    

*All other mandatory spending programs combined (e.g., food stamps, unemployment) total over 600 billion.