370: Optic Flow Pt2

THE OCCLUSION HEURISTIC

Researchers wanted to know how does the visual system interpret ambiguous apparent motion situations?  We assume ecological plausibility, or that things don't disappear or fuse but that two things turning into one thing probably means that one of them is just covering the other.

IMPLIED MOTION

Still pictures that are of an object that would be moving if it were real stimulate the MT more than just still images of an object at rest.  

375: Visual Imagery Pt2

IMAGERY IS LIKE PERCEPTION

Imagery is much like perception.  The idea is that in a bottom-up process you have memory representations.  You don't usually have a memory representation unless you first have a visual experience (e.g., you can recall what your parents house looks like).  He made a quick explanation that we do actually remember auditory experience as well, without a visual representation.

Confusability
Perky took people and had them sit in front of a screen and imagine a banana.  A banana that is as lifelike as possible.  Unknown to the subjects he very faintly flashed a banana on the screen and then asked them what type of banana they imagined.  And in almost every case they described the banana that was flashed on the screen.  This shows evidence that vision and memory are very similar processes.

Other evidence for the similarity between vision and memory would be giving someone a high imagery phrase and a low imagery phrase.  Visual interference task shows subjects a 1 or 2 and have to remember the other number.  Auditory interference does the same but with sound.  Visual interfering task causes a big effect on high imagery pairs and auditory interfering task causes a big effect on low imagery pairs.  (Sorry if this is a bad description, I didn't understand his explanation very well).

Damage
Damage to ventral impairs visual imagery and damage to dorsal impairs spatial imagery.

IMAGERY IS NOT LIKE PERCEPTION

Heuristics
The rotation heuristic is where we imagine things that are tilted as much more straight then they really are.  The alignment heuristic is where we believe that Philadelphia is further north than Rome when it is not.

Some of the things that imagery allows you to do is mental inspection.  When we do mental transformations we are able to get them correct about 60% of the time.  The point is that images can be inspected to some extent but it is not as clear as perception.

Distortions, inspection and it is difficult are reasons it is not like perception.

WHY DO WE HAVE IMAGERY?

Memory
One idea is that it helps with memory.  We remember words that are concrete better than abstract words because we can make a mental image of them.  Mentioned the dual coding process as a process where you only have meaning with abstract words but there is an image with concrete words so with an image you can do deeper semantic processing.

Another idea is that if you have concrete words you can imagine them interacting bizarrely to help with memory.  Of course this is non conclusive but some of the evidence shows that bizarre interaction helps memory more than non-bizarre interaction.

Make Implicit Knowledge Conscious
Imagery helps us take implicit knowledge and make it conscious knowledge.  For example, is the writing on the coca-colo logo cursive?  To answer this question most people will visualize the coke can to remember the writing.  So the answer to a question like this is in an image that we remember but we have yet to encode and extract this specific information, per se.

Prepare for Future Actions
One reason to have visual imagery is so that we can prepare for future situations.  For example trying to decide if a bed will fit into a room.  How do we do that?  We imagine it.

382: Stress Management Through Relaxation

Relaxation

TECHNIQUES

Breathing

An effective and easy way to train your breath is to use an analog clock on the wall.  Inhale for 5 seconds and then exhale for 5 seconds and do this for about a minute a few times each day.  You will be surprised how quickly you can train your breath.

Progressive Muscle Relaxation

Slowly work your way through every muscle group systematically first tensing the muscle group for 10 seconds and then relaxing for 30 seconds.  The first time you do this start with small muscle groups, about 16, and then as you get better at this technique you will eventually be able to separate your muscles into only  4 groups.

Body Scan

A body scan involves close to the same process as progressive muscle relaxation but only cognitively.

Visualization and Imagery

Autogenics

Imagine that your hands are warm and heavy.

Self-hypnosis

This involves deepening processes, such as counting down as you imagine yourself descending in a visual scene.

304: General Mental Intelligence

General Mental Intelligence

THE CONCEPT OF g

Today in class we talked about the concept of g, or a general mental intelligence.  Binet took the concept of g and adapted it to include two kinds of intelligence; fluid and crystalized.  People can have a fluid intelligence, these are the abilities that allow us to reason, thin, and acquire new knowledge, or crystallized intelligence, which is the knowledge and understanding that we have acquired.  So what is your understanding of intelligence?  Is there really some sort of general mental intelligence that could encompass all types of intelligence or does it need to be more specific?

THE FIRST IQ TEST

The Binet-Simon Scale (1905) was designed to identify children with mental limitations.  It was revised in 1916 and the concepts of intelligence quotient (IQ), mental age, and chronological age were introduced into the results (the Binet today has a mean of 100 and a SD of 15).   

SUBTEST SCORING

There are 10 subtests and they all have scaled scores (mean of 10, SD of 3).  To be a scaled score is like being a z-score except they have a different mean and SD.  

370: Optic Flow

OPTIC FLOW

As you move through your environment you see everything moving by you 

but your perception is that they are stationary.  

APPROACHES TO OPTIC FLOW

Behavioral Approach

There is a local disturbance in the optic array (e.g., you fixate on the environment and something moves across your vision)  There is a local disturbance in the optic array when you move your eyes to follow the stimuli that moves across the scene.  A global disturbance is when everything seems to be moving because in fact you are the one that is moving.

Physiological Approach

Vestibular information is your sense of motion so when spin for a while even when you stop spinning you will feel as though you are still spinning.  This is an example of when vestibular information is contrasting with your visual information.  Why would you feel nauseous in these types of situation?  The reason for throwing up is to get rid of toxins in your body so if you activate this area of postrema (the area that controls vomiting) you will throw up.  If you feel dizzy then that is a good indication that you have toxins so your body will interpret that something is wrong and probably a toxin so it is adaptive to throw up.

COROLLARY DISCHARGE THEORY

The idea is that if you move your eyes there is a signal sent to you eyes to move eye muscles that we call a motor signal (MS).  There is a copy of the motor signal called the corollary discharge signal (CDS) that goes to the visual system and converges with the image movement signal (IMS), that comes from the eye, and if there is a difference between the IMS and the CDS then we perceive movement.

Some of the prediction of this theory include...

If you were to immobilize your eye muscles you could still get a CDS but not an IMS which means that the world will appear to be spinning around you.  This is because the eye will remain stationary when you try to move it.  

If I shine a light at your eye and then turn off all the lights and then you move your eye what would happen with the after image?  You will get a CDS but no change in the IMS so the after image will appear to be moving.

If an object moves and your eyes move then the image of the object will not move.  You will get an CDS but not an IMS change since you are following the image so you will perceive movement.

REAL-MOTION NEURONS

How do we go from V1 neurons only being able to see one portion of the scene to seeing everything?  It needs lots of inputs.  We have a global motion detector that takes these inputs from a layer right above V1 and combines all the information together to determine the global bit of motion that is going on.  Basically you are just taking a lot of bits of information and combining it to get the whole picture.  

MIDDLE TEMPORAL AREA (MT)

How sensitive are MT neurons to motion.  The idea is that you show these monkeys randomly moving dots and the question is  how coherent does the motion have to be to perceive movement.  You can vary experimentally the motion and monkeys are taught to judge direction of dot motion.  As the coherence of dot motion increased so does the firing rate of the MT neurons.  The actual percentage of coherence that you need to perceive motion is as low as 1-2%.  If there is an MT lesion then monkeys don't respond until about 20%.  

What if you stimulate MT?  Can you simulate motion?  The short answer is yes, you can. 

BIOLOGICAL MOTION

Biological motion is processed in the superior temporal sulcus (STS).  We are very good at identify biological motion even if we aren't quite sure what the actual object is.  An application of this type of biological motion is the recording of joints on the actor who played Gollum.  The reason that movies use people instead of just relying on animators is because even when there is a slight mistake in motion we are able to detect that something is off.  

375: Visual Imagery

Visual Imagery

MENTAL MODELS

We construct mental models of situations.  When we read a sentence, for example, we create a mental picture of what we are reading.  Dr. Kirwan used a specific example of a fantasy novel describing a group of people walking in single file then suddenly talking to each other side by side.  These were two contradicting mental models.

Early on it was thought that these types of mental models were remembered in semantic categories.  Usually these semantic mental models were be explained in sentences.  A type of abstract representation of this is shown in an analog vs proposition of "a ball is on a box."  In proposition you have a relation, syntax, a truth value, it is abstract and it is not spatial.  Another possibility is an analog presentation (e.g., a picture of a ball on a box) and this analog presentation has no distinct relation, no syntax, no truth value until it is described, it is concrete and it is represented in a spatial medium as a picture.  We took a got in the class as to how we experience this sentence (a ball is on a box) and we all agreed that we imagine a picture instead of a proposition.  This was controversial in the 70's but it is becoming clear that this type of analog imagery is more correct.

MEMORIES AND MENTAL MODELS

If you have hippocampal damage how do you do at these types of imagining tasks?  It turns out that people with this type of damage cannot do this very well because we draw on memories to create these imagined scenarios.  So damage to the hippocampus means you will have damage to your memory system which makes it difficult to draw on memories in order to create a mental image/model.  This could be part of the reason why our memories are so susceptible to change based on new information.  We are usually pretty good at keeping imagination and memory separate but it can create some confusion.

PROPERTIES OF MENTAL MODELS

Property 1: Rotation

This is a type of spatial rotation.  There are also sex effects for this where men are slightly quicker and more effective with this than women.  It turns out that the further degree of rotation the two objects are different from each other the longer it takes to do this mental rotation.  This seems to indicate that people are actually visually imagining this mental rotation.  If people were doing this propositionally then it wouldn't take extra time.

Property 2: Size Zooming

A task that exemplifies this property is asking people to imagine a tiger in a small pink square and then tell me what color the tigers nose is.  The smaller the box is the longer it takes to answer this question.  This indicates that we are doing a sort of zooming in to determine what the color of the nose is.  Again this is in opposition to proposition.

Property 3: Scanning

Your task is to view an island with some fictitious landmarks such as a well, a tree, etc.  Then you take the map away and are asked to imagine going from one landmark to another.  If those landmarks are close together your response is faster than if you are asked to go from landmarks that are somewhat further away.  This indicates that people are scanning across the map and is again in opposition to proposition.

Property 4: Brain Locus

If these mental models are represented by language then as you do these tasks your language centers should be activated or if it is visual your visual centers should be activated.  It turns out that the dorsal stream, a visual center, is activated when a person is asked to do these tasks.  Language centers remain inactive.  

THE PIAZZA EXPERIMENT

Imagine that you are in the city square sitting on the cathedral steps looking south, tell me what you see.  Now imagine you are on the other side, now what do you see.  Undamaged people can do this task with little difficulty.  With unilaterally left neglect patients they could only focus their attention on half of the visual scene even for internally generated scenes.  The point is that mental images require you to focus attention on them and that mental images are intact in neglect but there is an attentional problem.

304: Theories of Intelligence and the Binet Scale

 WHAT IS INTELLIGENCE?

“general potential, independent of previous learning” ~ Saccuzzo

INTELLIGENCE

In class there was a general consensus that intelligence has a lot to do with cognitive functioning.  But where intelligence begins or ends is much more difficult to decide.  For example, at what point can we say that a person’s response is an intelligent response?  Is it just when the answer is correct or is there more to it?  Another example would be from a gymnast (Jennifer in class) who is flying through the air in a flip that almost any other person could not do.  Is this an example of intelligence?  How much of a physical flip is cognitive and how much is physiological?

Saccuzzo, the author of the text we are studying, said that intelligence is a person’s  “general potential, independent of previous learning.”  If intelligence is independent of previous learning, then why is there an information section on current intelligence test that look at your acquired knowledge?  Alfred Binet has given an alternate definition of intelligence when he said it is “the tendency to take and maintain a definitive direction; the capacity to make adaptations for the purpose of attaining a desired end, and the power of autocriticism.”  This definition is one that is much more recognized in the psychological community.

TYPES OF INTELLIGENCE

There are three main ways to define intelligence; the psychometric approach, the information-processing approach, and the cognitive approach.  The first is the psychometric approach.  This is the oldest approach and the focus of chapter 8. 

The second is the information-processing approach.  This approach is the newest approach that is used most often today.  They believe that the scores a person receives are interesting but even more interesting is the way that the participant approaches the test. 

The last is the cognitive approach.  This focusing on how people adapt to real world demands.

GENERAL ABILITIES INDEX (GAI) – on the test!!!

The general abilities index (GAI) is what you calculate when the four scores on the full scale intelligence quotient (FSIQ) are not lining up.  The FSIQ is scored on working memory, processing speed, verbal intelligence quotient (VIQ), and the (PIQ).  You use the GAI usually because working memory and processing scores are not reliable so the GAI ignores those two scores.  This seems to provide good evidence that one need not concern themselves with a FSIQ and just concern themselves with individual scores and what those specifically mean.  However, this would eliminate the concept of spearman’s g.

THE HISTORY OF INTELLIGENCE QUESTIONING

Alfred Binet defined intelligence as the capacity to 1. Find and maintain a definite direction or purpose 2. Make necessary adaptation to achieve that purpose and 3. Engage in self-criticism so necessary adjustments in strategy can be made.  He believed that intelligence expressed itself through judgment, attention, and reasoning.  It is guided by 2 principles underlying all modern intelligence theories; age differentiation and general mental ability.

Age differentiation’s goal was to find tasks in which the age group completion would be 66.67 to 75% (2/3 to ¾)  General mental ability is the total product of the various separate and distinct elements of intelligence.  Binet essentially wanted talks with a strong correlation with each other.

The next big contribution to intelligence testing after Binet came from Spearman.  As we all know Charles Darwin theorized about natural selection.  His brother, Sir Francis Galton, took this idea and wanted to come up with a way to decide which humans are better than others.  Spearman is a student of Galton.  Spearman came up with the positive manifold where he took two tests and if you did well on one test then you could do well on the other.  From this relationship between these two tests he he started to come up with the concept of a general intelligence which he called g.