Week 7--Interpreting Ambiguity

Just noting that, for the second time, Serendip has logged me out and returned me to the home page after clicking "preview comment."  Has this happened to anyone else?

 For my survey, I asked residents of Rockefeller 1st-1st and 1st-2nd to answer three questions using this image:

1.) Which of the three soldiers looks the tallest?

All six subjects answered that the soldier on the right looks the tallest, which did not surprise me because that was my first impression (and the reason I made it the first question).

2.) Briefly explain your answer.

Although all six subjects said something different, all of them mentioned perspective, the spacing between the lines, the direction of the lines, and the angle at which each soldier stands.  These results did not surprise me, either, because it is clear from how the lines slant that there is an impression that there's a vanishing point and that the third soldier is placed towards that vanishing point, when he's really just placed the highest in the image.

3.) "All three are the same height."  Do you agree with this statement?"

Five of the subjects circled "yes" and one wrote in "maybe" (yes and no were the only choices I gave).  I am curious to know what makes that subject unsure, since you can easily measure each soldier or even remove the lines from the drawing.  But I do see how she could be uncertain without measuring, because, even though I know they're all the same height, I can't quite get my eyes to see that, too.

I did my survey on peoples reaction (the volleyball team) to the skull image. I wanted to see if there was a trend in what they saw. In other words I wanted to see what interpretation of the ambiguous figure the brain prefered, or if it even prefered one over the other.

I gave each player an image of the ambiguous figure and asked them four questions about it:

1. How many images do you see? and what are they?

2. What image did u see first?

3. What image did you like better?

4. Why do you think your brain chose the image you saw first?

Everyone saw two images (one person saw a woman playing chess). Everyone agreed that the first image they saw was the skull, and majority said they preferred the image of the lady in the mirror. Overall, they thought that the reason they saw the skull first was because it was the dominant image, in addition to being a white image surrounded by black.

It turns out that although the individuals preferred the softer image (more feminie-> all girls) the brain selected the other image. I interpreted this information as follows: the persons personal preference doesn't impact the choice of the brain. The brain chooses based on what is seen first; the dominant image. I wouldn't say that the brain has a preference as in it "likes" one image over the other, but it does "prefer" images more easily detectable.

The picture I used for the survey was the one with the man's face as one figure and a woman in a winter coat over with her back showing, standing by a tree. I asked 15 people in my hall to explain to me what is happening in the picture. 8 people said that saw a lady in a fur coat standing by a tree first. Afterwards, they saw the man's face. 5 people said that they saw the face first, and the woman second. 1 person said, "well I see two things: a man's face and the back of a woman". 1 person said that they only saw a woman in a fur coat and nothing else.

It was interesting to see that people either saw the man's face first or the woman first. It was even more interesting that to some people, they can only see one image and nothing else. 

A factor that could have affected the results is the distance the person held the paper with the picture on it. I saw that if the person held the paper far away, they saw the man's face first. However, if they brought the paper close to their face and studied it very closely, they saw the woman first.  

I asked seventeen people to look at the optical illusion and answer the five questions represented in the graphs. I was mostly interested in how people felt in regards to optical illusions: my last graph. However, I started off by asking people what they first saw. I was suprised that more than half of the people saw the person walking first because I see the face first. I realize that how close or how big the image is greatly affects what image one sees first. I first saw the image on the computer full size, which made the face more apparent. There was a discrepency in the results between question three and four because one person answered that she could not see both images, but every respondent answered that they could voluntarily switch between the face and the person walking. The results to the last question were the most interesting. I find optical illusions very frustrating, but most people responded that they find them fascinating. One of the reasons the results to this question might have turned out this way is because the majority of the respondents could see both images and switch between both images, which makes it less frustrating. If the respondent group had a harder time seeing both images, more of them would have been frustrated.

After the class with Paul Grobstein the other day I was left wondering if our brains locate and give meaning to randomness does context affect the meaning we perceive?  I used the images below to test this.

Picture 1

Picture 2

There are two images hidden in each picture.  The middle color block can be seen as a goblet/candlestick/vase and the outer color block can be seen as two profile faces. 

I gave half of my participants Picture 1 and half Picture 2 and asked the following questions:

1) Look at the figure below for 1 second (basically aglance).

2) What do you see?

3) Look at the figure below again for 10 seconds.

4) What do you see?

I was hoping to see that the background (the context) would affect a person's ability to see both images.  I hypothesized that more people would be able to see both  images in Picture 2 because the outer blue border calls more attention to there being two images, whereas in Picture 1 the image of the profile faces is lost into the white of the background.

However, I found that the background didn't actually made a difference.  Of the twelve people that I surveyed (six with each picture) all six people saw both images in Picture 1 and five people saw both images in Picture 2.  (Interestingly, the person who did not see both images in Picture 2 only saw the goblet/candlestick/vase image.)  While this information does support my hypothesis, five of the six still saw both images.  So, from my results I would say that context does not affect the meaning perceived from an image.

However, that being said, my sample size was excessively small and many of the people who I surveyed had already seen this picture before in some form or another and already knew about the two images, which most likely greatly affected my results.  I would like to repeat this on a larger scale and with a similar, yet lesser known picture.

After reading the article in from the nature journal about magic tricks, I was reminded of the experiment (because it was mentioned) on unintentional blindness where people are supposed to count the number of passes a certain team makes and will not notice that a big gorilla walks through the scene because they are focusing so much on counting. I decided to send that video to people (http://viscog.beckman.uiuc.edu/flashmovie/15.php) along with my survey about it and told them the following:

"Please watch the following video clip.

this is a video clip of two different teams passing a basketball around.

focus VERY CAREFULLY on how many passes the team in the black shirts makes and make sure to count the passes.
the following survey will ask you how many passes they make along with a few other questions."

The survey asked 5 questions. question 1 asked the participant to state how many passes the team he or she was watching made. The second asked if they noticed the gorilla that walked through the middle of the players (yes, no). The third asked "If yes, have you seen this video before?" (Yes, No, I did not see the gorilla). Question four asked "If yes but you have not seen this video before, did you accurately count the number of passes made?" (Yes, No, I have already seen this video, I did not see the gorilla). Question five asked the participant, if they did not see the gorilla, to watch it again and rate how surprised they were to have not seen it at first (Not surprised, somewhat surprised, very surprised, utterly speechless, I have already seen this video, I did not see the gorilla). 

After 10 people had taken it, and every single person had seen the gorilla (when we did it in my psych class last year, only another girl and me saw the gorilla because we were not paying attention to counting...everybody else missed it!), I realized that something was wrong with the way I was conducting my experiment. I decided to send out my survey again, only this time asking the new participants to count the passes made by the team in the white shirts, not the black shirts as before. I was wondering if the black gorilla was getting caught by people's eyes who were watching the black shirts. However, everyone there too saw the gorilla. I have concluded that the way which this video was presented, online in an email, greatly inhibited my results. If I were telling the participants in person to really focus on the passes made, I could make my voice sound really important and probably make them focus more, as my psych teacher did to my class. However, the fact that the video was presented so impersonally made people not focus that much on the counting. This is shown in my graphs:

The first graph just shows that almost everyone saw the gorilla, except for one person counting the black shirts. The second graph just shows what percent of those who saw the gorilla had not seen the video before, because once you see the video once you are sure to see the gorilla the second time you watch it because you will not be concentrating on the passes as hard or you will know the gorilla is there. The third graph, however, is the most interesting to me. This shows that the majority of the people who had not seen the video already did not count the passes accurately. This shows that they were not focusing extremely hard and therefore probably noticed the gorilla (the person who did not see the gorilla left question 3 and 4 blank for some reason, so they are not represented on those graphs). My last graph just shows that the person who did not see the gorilla was "very surprised" that he or she did not see it. 

I asked people 3 questions about this image:

/exchange/ambigfig/cheetahs

The graph about the cheetah on the left had 17 people say that the cheetah is facing backwards and 9 people say that the cheetah is facing forwards.

The graph about the cheetah on the right had 18 people say that the cheetah is facing backwards and 8 people say that the cheetah is facing forwards.

The graph about if the people thought the cheetahs could go both ways had 23 people say yes and 6 people say no.

What I was trying to prove was that the picture is not only ambigious, but it all depends on a personal view. I only had the people look at the graph for 10 seconds and answer each question. The interesting part was that even though people said they could go both ways, only one person answered that both cheetahs were facing fowards. The point of my survey was to make people decide which direction the cheetahs were facing, and to see if there was more of a trend. I wanted to see what way the majority of people though the cheetah was facing, I thought it was going to be 50/50 but it wasn't so maybe next time instead of asking on surveymonkey.com I will ask in person which they think to make sure that they only look at the survey for 10 seconds at a time.

Leigh 

For the optical illusion survey, I used the very popular picture of the two faces and the vase to collect results on what was seen, and a possible explanation for what was seen. I showed it to several people on my floor, and in my Spanish class, and the results varied rather greatly from person to person. I decided to pick a rather popular image to use in this survey because prior to our class discussion I had not really considered what was happening behind the image to create an optical allusion. However, after learning more about optical illusions I thought it would be interesting to collect the opinions of other people with respect to the image.

The results that I collected showed that the image that was seen first varied greatly amongst the people that I asked. For example, while most of the people said that they saw the two people looking “face-to face at one another”, some first saw the vase, and one of the participants first saw a goblet.

I asked the participants if the original image that they saw was vague or definite and I found it interesting that all of the participants responded that the image that they saw was definite. I also asked if they saw more than one image in the photo, however, and all of the participants answered yes.

Lastly, I asked the participants if they could elaborate on their reasoning, if any behind what they saw, or if they could give a description of what they saw and how they went about seeing it. Some people responded that they felt like their eyes were sliding out of focus when they tried to see both of the images.I researched this at this is called multistability or literally multiple perception. Other people said that what they saw was just immediate and there was no reasoning, at least not apparent to them, for why they saw what they saw. Finally, some people responded that they saw the people or the vase first, then, after focusing, were able to see the faces of the two people.

Though the degree of ambiguity differs , every input to the brain can , and is , interpreted in more than one way by different individuals . These are the informed best guesses of each brain of what is out there that is meaningful to itself .
 
The fifteen randomly selected subjects under study were shown Figure 1 ( /exchange/ambigfig/gilbert ) . The overwhelming majority ( 13 out of 15 respondents or 87% ) identified the figure as that of a skull . When probed about what led to their answer , they typically identified the white expanse of the forehead , cheekbones and jaw . Only two respondents ( 13% ) said that the picture depicted two women , framed by a porthole / round window , sitting across each other at a formal dinner .
When exposed to Figure 2 ( /exchange/ambidfig/cheetahs )  , every respondent saw two leopards , one standing , and the other seated .
Working on the hypothesis that the sharp contrast in Figure 1 and the largely uniform tone of Figure 2 influenced the answers of the majority ,  respondents were shown Figure 3 ( /exchange/ambigfig/etching ) . The majority ( 87% )  responded with answers ranging from a grandfather to William Shakespeare though all indisputably related to a man . When asked what decided their answer , they again replied that it was the white expanse of the forehead and , in some cases , the upper lip . Only two ( seasoned test - takers obviously ! ) said that it  showed a meadow with  grand buildings in the background and a robed woman and the bole of a tree in the foreground .
Though each figure could have elicited more than one equally good spontaneous answer , there was one popular informed guess every time driven by the same reason . The brain evidently pays more attention to comparisons and contrasts than to absolutes . In a world driven by information overload and constant flux , the brain sifts through signals to decide which are worth its while to interpret . It allows large rivers of data to pass through almost entirely unassimilated , peeling off selected data for a close , careful  view ( Jeremy Wolfe , Harvard Medical School ) . Picking the " out of the ordinary " or the different , the high contrast , is perhaps one way in which the brain decides what to focus on .
 
According to Natalie Angier ( The New York Times , April 2008 ) , the mechanisms that succeed in seizing our attention fall into two basic classes : bottom up and bottom down . Bottom-up attentiveness originates with the stimulus , with something in our sensory field that is the equivalent of a sensory shout - the expanses of white standing out of the dark background in the examples discussed earlier . In comparison , top-down attentiveness is volitional - the decision by the subject that an item , even in the absence of sensory shouts , nonetheless merits attention .
Respondents were shown two photographs of themselves as part of a group , taken in quick succession and with barely perceptible differences . With the photographs removed from view , they were asked if the two photographs were different in any way . Only three ( 20% ) said they were ; interestingly  the only three who had  changed position in some way ( sitting cross-legged / moving hair out of the eyes / smiling in the second photograph ) between the two photographs . The brain concentrates only on what it wants to . Here , one's own image leapt out to the exclusion of all else in the limited time available for viewing the photographs .
Similar behavior was exhibited when participants heard the pun , " the land of the rising sun/son " . Asked what kind of society it might describe , every participant but one came up with " a Japanese society " - a response clearly influenced by geograpy textbooks though the question was a sociological one . The solitary exception ( probably a student of sociology ) averred that it was certainly not a meritocracy . The brain tries to ease its workload by using readily available decoding devices ( in this case knowledge of geography or sociology directing perception ) whenever it can do so .  

Respondents heard the word " teapot " repeatedly and then the word " potty " several times , without any break in the repetition from beginning to end . When asked to identify the word that had just been repeated , more than half answered " teapot " . Only a little over a third caught the switch to a new word .
Respondents were also shown ten cue cards with the phrase , " the bear eats shoots and leaves " on the first nine and " the bear eats , shoots and leaves " on the last card . When asked to explain the sentence , every respondent but one said it described the bear's diet . The sole dissident volunteered that the last card seemed to speak about some kind of a cowboy bear . The change - i.e. the addition of a comma after the word " eats " - registered with just one respondent .
Taking a leaf out of the magician's manual , repetition was used to hide the trick - when observers are faced with repetition , they naturally assume that each repetition is identical . In this , they are encouraged to do so since the brain tends to try and anticipate future perceptions in trying to make sense of ambiguous input in a data-overloaded rapidly changing world .
Covert misdirection ( by a sigh to indicate fatigue with the repetition at the point when the change took place ; not memorable so as not to arouse suspicion ) , was used with half the sample . However , both halves largely overlooked the change in stimuli . People fail to notice that something is different from the way it was before even if it is staring them in the face - the brain , in cutting corners and making do , makes us change blind .
 
The mind moves in mysterious ways its wonders to perform . The more one studies it , the more one realises how much remains to be understood .
" What is mind ? No matter . What is matter ? Never mind . " ( Ascribed to Thomas H. Key ; 1799-1875 ) .      

To me, it seems like these illusions are all successfull because of weaknesses in our minds.  Well, maybe not "weaknesses", but visual flaws.  Because I had seen some of the illusions before and learned about color theory in psych, the illusions themselves did not have the best effect on me.  However, I enjoyed our discussion.  But if the mind is so easily tricked, then how can it have the capacity to make sense out of all of these constantly moving particles that make up objects?  That's not necessarily my opinion, but I think the notion that our perception is flawed and that everything just exist because our brains are putting order to chaos was unsettling to the majority of the class.  Much like when you find out how an optical illusion works, some theories can really suck the fun out of things...NOT THAT THE CONVERSATION WASN'T FUN!  In one of Plato's dialogues, Socrates asks Euthyphro, "And if a thing is being led, is it being led because of the leading?  And if being seen, being seen because of the seeing?"  Euthyphro replies with, "Certainly."  The opinion is like what Paul suggested in our discussion: that the table is only in existence, or being seen, because our seeing it and interpreting it exists.

 Just a thought, I could be totally wrong:

What Paul Grobstein said really made me think about how we perceive the world (or how our brain perceives the world). But I still am confused about the idea that our brain assigns meaning to randomness. If we imagine that our eyes are capable of seeing molecules ( like we did in class), Paul Grobstein says that everything will be one big blu of molecules. I agree with this understanding, but it made me think: aren't molecules held together by attractive forces? (this is why a table is a solid object, right?) If so, even if our eyes were capable of seeing molecules, doesn't it make sense that the bonds or forces between the molecules would still exist? This would mean that even though everything would be a blur of molecules, solid objects would exist (until the bonds were broken). So assuming objects (held together by forces) still exist we (by process of thought) give an object a name, "table". But is the idea that the brain assigns a name (classification) the samee as assigning it a meaning? I think that although the brain interprets things in a way I don't understand, and classifies them, it is people's (physical) interacions with randomness that gives it meaning.

I'm still trying to wrap my head around our discussion yesterday. Someone in class referred to our shared perception as a collective construct and it started me on the following train of thought:

Even as we are composed of clustered molecules, our joining together each week for class creates another cluster (that disperses in a relatively short period of time compared to the molecules of a table).  What are the defining characteristics of this cluster? What traits could we ascribe to our group (if our eyes were too big to see people as individuals) and how could our cluster be manipulated (as we can manipulate the molecules of a tree by cutting it down to make a table). I also wonder how the molecules that create the tree choose to group/adhere? Do they self select intuitively (like moths to light), intentionally (like us?), or through external design? Are the ties that bind molecules together, which gradually weaken over time, like the ties that bind people to each other?

I noticed that there's a string on Paul's serendip blog re: the benefits of dialogue/collaberation between scientists and humanists - I was struck by the same thing in class. The way our conversation grew by flowing from literary ideas to scientific explanations and back again was illuminating.

I'm more in the humanist camp myself and I'm interested in getting together with some scientists to further explore the ways we interpret our world. If anyone else feels the same way, let me know!

Found this on the internet:

Proof

hope it helps haha