PSYC 339 MIDTERM STUDY GUIDE
PSYC 339 MIDTERM STUDY GUIDE PSYC 339
Popular in Origins of the Mind
verified elite notetaker
Popular in PSYC
This 10 page Study Guide was uploaded by Isaac Lemus on Sunday September 25, 2016. The Study Guide belongs to PSYC 339 at University of Southern California taught by Justin Wood in Fall 2016. Since its upload, it has received 178 views. For similar materials see Origins of the Mind in PSYC at University of Southern California.
Reviews for PSYC 339 MIDTERM STUDY GUIDE
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 09/25/16
BASIC INFO ALL animals have some innate cognitive systems, ancient building blocks in the mind, core systems but only humans have the ability for cultural evolution and knowledge (science, religion, medicine, and politics). To study in this class, we’ll look at a particular trait in a person then see how this trait is carried out by babies, other animals, and people of other cultures or circumstances to determine if that trait is an innate mechanism or something we learn. ● John Locke: The idea that we are all born with a blank slate and everything in the mind is learned through experience. Empiricism, Berkeley ● Immanuel kant: Knowledge begins with experience, but it can be innate, programmed into our DNA. Nativism, Descartes Class topic roadmap: space perception, navigation, object representation, numbers, social cognition (agents/tools, social partners, language, teaching, religion, and morality). PART 1: SPACE PERCEPTION How we see the world around us & understand concepts like depth, direction, shape, & size. it’s so hard to study perception is because it’s literally just us interpreting light hitting our eyes. 1/3 of our brain is dedicated to processing vision. Here are some ways to tell what’s where: ● Interposition: If one object blocks another, the blocking object is in front ● Linear perspective: We account linear tilts as distance. Ex: Train tracks in the distance ● Convergence: Closer an object is to eyes, larger the angle eyes have to pivot to focus From a Nativism perspective (Rene Descartes) these traits are core knowledges in our mind so our minds have natural trianglement and instinctually analyze the angle of convergence to figure out the distance. From an Empiricist perspective (George Berkeley) we learn about our environment by experience and effort so we consciously process the distance. Introspection. Let’s put it to the test: Molyneux's question: If a person who’s been blind their whole life felt a cube and sphere and suddenly could see, could they tell which is which? ● Descartes: Yes, through natural understanding of geometry ● Berkeley: Nope, because the person never gained visual association Answer? Berkeley. The results show extremely low transfer rates but it also kinda goes against Berkeley because the patient never picked up/learned overtime (unless really young) EX: Gregory (Blind 50 years) never developed skills and never fell for visual illusions after 3 months. Patient S.D. (Blind 12 years) picked up skills and fell for visual efforts after 12 years. ● The problem? Patients rely more on color and segmentation to perceive. two overlapping squares looks like three shapes and a ball looks like multiple objects because of shading. And, studying a person who has been blind for such a long time can be erroneous because their brain capabilities have been altered. The absence of one sense means that the other senses take over So, let’s look at a regular human adults: If we shift around their visual perceptions? Would they be able to adapt to their new circumstances? ● Helmholtz’s study: Walking through a forest with prism shifting glasses. If they can adapt, plasticity. If no adaptation, no plasticity. ● Results: Over time they could adjust and after they took the glasses off their brain took time go back to normal vision. The brain is actually changing to some capacity. So is space perception is flexible and able to be modified? If so, then adults can learn then babies can learn and babies wouldn’t have to have innate mechanisms. So empiricism over nativism is definitely plausible. ● BUT as it turns, it has to do, not with vision, but with where you think your body is. In other words, your vision isn’t the sense that is adjusting but where you sense your arm is in comparison to everything else. When there is conflicting information being sent from these two sense, touch (proprioception) is likely to be affected rather than vision. ○ Conclusion: Space perception isn’t modifiable and actually supports Nativism Let’s study other animals: If same abilities appear in many animals, humans likely have it to. ● Gibson’s visual cliff experiment. Baby goats automatically know not to step off a stool. Goats, chickens, and lambs all have this ability at birth. Rats and cats would avoid it after a week and babies after 6 months (once they learn to walk). ● But a week for animals and 6 months for humans is a long time to wait to do this study. Babies and animals could learn about visual perception during the waiting period. ● Modified experiment: raised 2 sets of animals in 2 different settings while they learned to walk: one in a light based group (normal) the other without any light or visual cues. ○ Results: Regardless of light or the dark, once walking, they avoid cliffs. ○ But cats, at first would go off the cliff but after three days of light they would learn not to do this. We theorized two possible options. ■ Specific experience: cat learns by experience that the cliff is safe ■ Non specific experience: Being able to see activates innate capabilities ■ New test: Cats automatically raised on deepside of cliff then would be shown the other side of the cliff. If it was specific effect, then the cat would stay on the deepside, because it is learned that it’s safe. If it is nonspecific, then the cat would see the deep end, and its innate process would determine that the shallow end is what is safer. ● Conclusion: Cliff avoidance is present and therefore nonspecific. ● How does movement affect visual perception? ○ Experiment: Held and heins experiment: cat locomotive. 2 cats were attached to each other, one can move freely and associates their movement with their visual intake. Other cat couldn’t move and had to follow what the other cat did. ■ Results: The cat that could move didn’t go off the cliff. The cat that couldn’t move didn’t associate their movement with their visual input, went off the cliff 50% of the time. Strong evidence for nativism ● Homology: same mechanism across species. ○ Hox genes; genes that have stayed constant regardless of time. ■ Example: The Body structure of flies and mice are similar ● Convergence: opposite of homology: Animals use different mechanisms to solve tasks Studying on babies ● Problems: takes 4.5 months to crawl, 7 months to walk, 3 years to start talking ● Solution: we examine where babies look at and for how long (to show preference) ● Robert Frantz: Do babies prefer looking at bullseyes, vertical stripes, or others? ○ Babies like looking at faces, text, bulls eyes but not homogenous colors. This tells us they can see! There is a dramatic increase in visual acuity during the first half year and It takes 3 years for babies to obtain 20/20 vision. ● Can they perceive depth? To test this: Do babies prefer to look at disks over spheres? ○ Anywhere 1 to 6 months they would rather look at something that is 3D. But this supports both Descartes and Berkeley. Descartes would say babies are perceiving 3D (Wow 3D!), Berkeley would say 3D requires more effort, confusing, therefore attention. Three ways to solve this problem: ○ Looking for signatures of depth perception ■ Stereopsis: putting together two slightly different views of each eye to create 3D perception AKA binocular disparity. used for tasks like threading a needle. To test this trait on babies, we put stereovision glasses on them (3D glasses) and showed them two images, one without disparity (would show up flat) The other with disparity (would show up in double imagines=3D). Results? At 3 months there is a dramatic increase in preference for the disparity imagines. But again, does this mean they can perceive depth or that the just like seeing double images? Well to fix this we tested for mature stereopsis signatures ● Hyperacuity= babies prefer stereopsis version rather than just stripes ● Limited range= if the stripes get too far apart, the perception of depth is lost of the baby ● Horizontal disparity= babies only prefer disparities that are horizontal ● So in summary, babies have stereopsis like adults, it appears suddenly at 3 months and the trait comes before they are able to move or reach and develop spatial association….nativism ○ Putting together different cues to depth ■ We showed babies two different cues for depth (motionan object moving in front of another, and stereoa triangle showing a shadow on the pattern below). And if babies perceived them as different it would support empiricist because it’s a new experience of depth, whereas if the baby thought they were the same, it would support nativism because it’s using innate cues to see the same thing, depth. Results? Babies don’t look longer at either only look longer if the depth arrangement has been changed but not between the two cues. ■ So in summary, this supports nativism because empiricism would say that each cue would give off a new sensation in perception. ○ Size Constancy (Computation of depth) ■ After showing newborn infants a small block close up, would they perceive a big block far away as new or the same thing (its retinal size would be the same). Results? Babies are able to perceive distance because they perceive each block as a novel experience. Supports nativism ● Studies of once blind patients ○ What we found: When they could see for the first time they couldn’t match the object with what they felt (Empiricism) ○ Vision seems to get better over time (empiricism) ○ Tracking objects. We all have the ability to discern objects once motion is added to them. Seems to be built into us (Nativist) ○ Stereovision can never be developed. Can’t learn through experience. (Nativist) ● Studies in normal human adults with altered perception ○ Vision seems to be rigid and unchanging (whereas our sensation of where our bodies are can be more easily changed) supports Nativism ● Studies on newborn animals ○ Seem to have depth perception since day one (Nativism) ● Studies on human infants ○ Infants too, know not to crawl of the edge once they can start walking ○ They also have the same signatures of stereopsis as adults: Hyperacuity, limited range, and disparities need to be horizontal rather than vertical ○ Different cues can trigger depth ○ Can perceive the true size of object, therefore understand depth ● All four areas point to core knowledge of space, which favors descartes and nativism. But experience is still important, especially to maintain these abilities. Just because it’s innate doesn’t mean it’s going to say rigid throughout a lifespan. PART 2: NAVIGATION How do we navigate the three dimensional world? We are horrible at navigation on our own without landmarks or references. 1920’s study: Blindfolded people when asked to walk/swim forward would start out straight then would veer off course dramatically. Socrates asked geometric questions 2000 years ago to figure out if trait is innate. Today we focus on these main questions: What are the origins of this geometric knowledge, What aspects can change? Studying on animals and how they compare to humans ● Three systems that humans share with other animals. First is path integration! ○ Ants follow smell trails for navigation. But In the Tunisian desert, the wind and dirt wipe away the smell and there are no landmarks yet they are somehow able to B line there way back home. Is it using something we can’t see or is it doing some natural trigonometry? ■ Experiment: Once ant finds food, we move it ten meters without it knowing. Results? it walked home but was 10 meters shifted, so they have fixed location in their head. How? ● The sun seems to be used as a compass for direction. They update their position depending on the time. When the sun is covered the can internally track where it would be. ■ Three different hypothesis for the distance part: ● Maybe they have an energy cue like humans; the more tired they are the farther they must have walked. ○ Not the answer because when we added weights to the ant it still got home perfectly ● Maybe they use optic flow. If they’ve seen a lot of stuff means they’ve traveled farther ○ Not the answer, even if the ants eyes are covered, they can still navigate the right distance ● Maybe they count their steps, like an internal pedometer. A study added stilts to some ants and cut some ants’ legs in half. ○ Ants with stilts over estimated the distance ○ Ants with stumps underestimated the distance ○ Ants with no change made it back home perfectly ● In summary, for direction the desert ants use the sun, and they count steps for distance. ○ Bee’s. Use optic flow rather than counting wing flaps because wind could affect distance. Do a special dance to tell other bees the exact location of food ■ Point: animals evolve to use strategies suited for the restraints given. ○ Gerbils: regardless of where the noise of her babies are coming from, the gerbils went the wrong way because they were following their path direction ○ Geese on their first flight always fly towards home if taken away ○ Newborn chicks will imprint to whatever it sees and naturally stay close to it. Even if barriers are in the way, the chick will know how to stay the closest to it. ○ Studying on young children sighted and blind children were asked to put a toy in the basket. They all took a novel path. Navigation doesn’t seems to be vision/experience based. ○ Studying on adults. We have path integration. 1st world cultures use it less. Adults need a compass to determine surrounding. Like heat of a light in a room. ● Conclude: Path integration is used by every animal studied and seems to be innate. Likely a homology trait because researches think this trait was passed down by a common ancestor. ● There is 2 degrees of direction error and 10% of distance error ● Second system: snapshots representation: Viewdependent scene recognition. We create a vivid image in our head as a reference. This is in pretty much all animals. Instead of a map in our head, we store millions of snapshots in our head to create a bunch of places that are linked together. ○ Bees: will go around an area and take snapshots before going to the food. Then you could release them from anywhere and they can B line straight to the food. Don’t seem to have a map in their head. ○ Rats. Water mazes: once a rat knows where’s the platform, it can B line straight to it by using the shape and color references on the side of the pools. To test if this is snapshots, we tested on new rats but blocked off a section. When we put them on the new section they had no idea where to go until they made it to a familiar zone then went straight to the platform ○ Adult Humans? Comparative experiments: if a fundamental trait is present in a basic animal, the mechanism is likely in more advanced animals, like humans. To test humans; objects are put on a table. Bring in participants, they look at table. Close eyes, one object is moved. Participants open eyes and can tell what objects are moved. Now scientist move the table when the participants eyes are closed, humans are terrible at telling what changed, why? Because they never developed a snapshot from this position. Put adults in a virtual maze with wormholes and they don’t notice, which means they don’t rely on a map ○ Young kids can’t see other people’s viewpoint (3 mountain doll experiment) ○ We know that the hippocampus helps with memory and therefore navigation. Three different types of cells used in the hippocampus: ■ Place cell Snapshot part, fires with familiar perceptual image. We link the place cells together to navigate (like google maps street view) ■ Direction cell like a compass. fire’s when body faces the right direction ■ Grid cells Spatial organization seems to be a collection of triangles, will fire when the body meets a vertex of triangles. ■ Are these neurons innate? Study on baby rats says yes ● Third core system: reorientation ○ Rats only use geometry (ignore patterns, colors, smells) and only look at the surface layout when disorientated. Signatures of what makes them do this ■ Its domain specific it only operates over some areas and not others ■ Its task specific it only happens when you disorientate a rat ■ Encapsulation the information is still used in the brain but it's not used in the process of reorientation ○ Young children, do they have these same signatures? ■ Hide sticker in rectangular room. Disorient them. when all the walls are white they go look at the long corners. When one wall is red they still go to the long corners. Completely ignore red wall ● Domain specific because they go to long corners geometry of extended surfaces. ● Task specific If the child isn’t spun around, the kid will go off of color. But if disoriented, will base finding sticker off of geometry ● Encapsulation kid remembers red wall, doesn’t use the info. ● So these knowledges appear to be innate but this can’t be the whole story, adult humans must have more than these core knowledges! Adults can develop theory of mind (three hill doll experiment) and can reference the red wall as a reference. Humans can adjust and mold their snapshots to the current perspective (Kids can’t do this until they are 6). At 6 we start using spatial language, could that be what causes further development. Two attempts to figure this out ○ Use human adults with both language and representation and take away their language abilities. Had to recite a textbook or following a beat with their hands while trying to remember where something was put. ● With drumming most can still go to the red wall ● With reciating, we mess up. We can’t put together “left of the red wall” in our head because we are using our language process to recite the textbook. Just our core systems must be used. ○ Teach little children left and right and see if they can navigate more effectively ● Learners: they navigate more like an adult. ● Non learners: search equally ● something about language formulates how we think, acts like a bridge that can link info from various core knowledges and form thought. Connect geometric representation with object representation. Language unencapsulated these core values. Animals can't think “Water is west of the rock”. It's all about spatial language. ● Humans also have other tools to create spatial representation like pictures. All humans have pictures, but no other animals do. Do humans understand pictures at birth? ○ Hochberg & Brooks experimented on their own kids. 19 months without pictures. Kids could recognize pictures at first sight but they think it’s the object itself, especially if it’s more realistic looking until 19 months when those traits disappear. So they recognize pictures but they don't understand them until 1.5 ○ Do kids know that a picture represents something else? Find snoopy experiment in a little room to a big room. At 2 they can't do this, but at 2.5 most can now do it. Start to understand symbolic capacity. So what’s hard for younger kids is dual representation. On one hand it's an object but on the hand it symbolizes something else. What happens when we trick them to get rid of the symbolic relationship, by a ‘shrinking and growing machine?’ Children have no problem finding the object with the help of the shrinking machine, because they don’t need to form a symbolic relationship ○ When do kids understand that toys are representation of real things? They don’t act appropriately because of scale errors. They are perceiving the size correctly, and they understand it’s too small, but they still try to use the objects in the same way. Conceptual not perceptual. Overview: We have innate mechanisms that are shared with other animals (3 core systems) ● Path integration (desert ants): the sun is used as a compass, steps are counted and remembered, natural geometry. ● Snapshot/ view dependent system (Rats in swimming maze, Bees): We match what we’re currently navigating with memories in our head. ● Reorientation system: When the two other systems can’t be used, this comes into play. Only cares about the geometry layout of an environment. Properties of the system: Domain specific, Task specific, Encapsulated From these three core processes, humans develop incredible representation: Understand other perspectives, reorientate using non geometrical cues, pictures, models, and maps. PART 3: OBJECT REPRESENTATION How do we associate shapes, colors, lines, texture, and so on to discern objects? We figure out what goes with what, what’s the complete shape of an object, we continue to represent objects even after they disappear, we have to separate different objects. What aspects of adult human knowledge allows us to recognize objects? Are these traits innate or learned? To answer this, first, we’ll find the object representation systems, then find their signatures, before finally testing these signatures on babies and animals ● Note: attention has severe limits. We’re never processing entire scene: Jabbawockee ● System one: Object tracking: Doesn’t care if objects change, it track motions. Doesn’t care about what, but where AKA spatiotemporal information. Signatures of system: ○ We can track 3 to 4 objects at once. ○ Can track across (in)visible barriers (inclusion) Can’t track an object not disappear along an edge ○ It’s easier to track an entire rather than just a part of it. ○ Brain operates over objects not fluid objects. ● Stem two: Object recognition. Complete opposite of tracking, Looks at color and shape of object. Signatures of this system: ○ Color and shapes. We need to have hues and outlines to discern objects ○ We are more inclined to group perceived shape together rather than physical shape. EX: Smoothies, cubies and spikes ○ We are sensitive to non accidental cues. Certain lines like edges and vertices provide more info rather than random outlines ○ We pick up categories and kinds of things. let's look at tracking systems and its signatures to see if they support nativist or empiricist ● Lets first look at how do babies deal with objects from birth onwards? ○ From 01 months there is no responsiveness to objects ○ From 14 months focus on objects but aren’t able to grab or motion for them yet ○ From 48 months reach for objects only when in sight, no object permanence ○ From 812 months search but only in fixed location. ‘A not B area’ phenomenon ○ From 1218 months they make errors when the change is invisible. They think that the object must be in the last place where they saw it ○ Piaget's proposed an empiricist perspective saying that babies go from only being able to intake visual information to developing the skills listed above. Stems from the coordination of action. Not an innate mechanism. Called a Copernican revolution. ○ As of right now it’s 50% a 50% split in support. But maybe the problem is their motor system not mental capacity or maybe they just don’t have good memory. Solution: need to figure out what babies think without relying on their actions. ○ For example: Gestalt psychology, we create the easiest way of interpretation This trait is apparent in adults, what about babies? To test, we create a habituation to a triangle with a block in the middle then show two new triangles, one fully complete, the other missing the edges where the block use to be. If they already assumed the triangle was complete under the block (like adults) they’ll stare at the broken triangle longer. They end up looking equally long at both. There is no establishment of an expectation ○ When we add movement there is a massive responsive to broken object, they see that as new. Meaning they establish a representation to a single object ○ Take away, babies don't have the gestalt principle and instead focus on spatial temporal information. Another example: Duck on a truck. The babies learn to associate the duck and truck as one object because they move together. What babies do is what blind people do who regain their vision, they rely on tracking and movement information. ● Studying adults ○ Change in rings. Even though the shapes and colors of the objects change, when there is motion added, our minds want to create as sense of stability. Things that move overpower our tracking of color and shapes. ● Now, let's look at monkeys ○ Tunnel effect illusion. Monkeys search for only one of the fruits because they think it’s just one fruit. Even though the fruit changes color and shape, the continuous motion overpowered their other system. ● Chick study with imprinted objects also shows that these traits are innate! ○ Object permanence task, mom is covered, yet they still found her. ○ Chicks only develop object permanence when the object runs smoothly. So they have it early on but it's still learned. ● So in summary, do animals and infants have the signature of tracking? ○ 3 to 4 tracking system; Yes. babies can tell 1 vs 2, 2 vs 3, but not 3 vs 4. (grahm crackers). Monkeys get 1vs2, 2vs3, and 3 vs 4 but not 4 vs 5 (fruit) ○ We can still represent things that disappear from view. Babies will look at the other edge of a blockage waiting for the object to reappear only when the objects disappears along an edge. ○ We track a whole object rather than its parts. Breaking an object into two, babies perform at chance level and their ability to keep track of an object ○ Cohesin. Babies can’t track things that pour, like the sand experiment. They can only track things that are rigid. ● In conclusion, infants have the same signature as adults, and so do other animals suggesting that this is a core feature that seems to require visual experience with smooth and slow natural movement Now, Let’s focus at the four systems of object recognition ● When it comes to color to shape, we can turn to illusions.The box/stick illusion and the illusion where drawn objects look 3D show how we build information of colors and shape. We perceive connecting lines as edges and use color to form sides. ● Mature species use non accidental properties. We focus more when a shape edges change from straight to curved rather than from curved to more curved. By looking at examples of Expansion, Curvature in and out, Curvature of shape direction we see that even though two blocks are equally different from the middle block, we sense that the straight block is the outlier. We are more honed in on shifts from straight to curved ● Studies Himba tribe in Northwestern Namibia show they are also sensitive to non accidental error rather than metric change. Even though they have different visual inputs they have the same abilities ● Pigeons and monkeys have these characteristics, widely shared in the animal kingdom ● Newborns can use visual information to perceive objects. At 4 months they can distinguish shape. 7 months they can distinguish patterns. 12 months distinguish color ○ Helps both sides. Empiricist: they learned it and gets better or time. Nativist: Maintained in the brain until hooks up get developed. ● It’s hard to test babies right at birth so next best thing: New born chicks on imprinting ○ Color; they can recognize change in shape and color above chance level ○ Perception: They go by perceived shape not by physical shape. Even though the object imprinted might change positions they still know to follow that shape rather than a different shape that is in a similar position as the original shape. ■ So they can recognize even in Abstract, view invariant, and across different backgrounds. Nativism! ■ But experience with the natural world is needed. Our world is based off of slow and smooth aspects. As the virtual world become less smooth it breaks chicks capacity to process and recognize. The quicker world moves, the less chicks can recognize and recognize. Therefore needs to be in the sweet spot of a slow and smooth world ● Now do animals and robots recognize objects in kinds of things ○ Monkeys using tools and throwing rocks (weapons and tools) ○ Birds use tools and have fun (fun and tools) ○ Simple set of computations that filters, thresholds, pools ,and normalizes through repeated layers. We create the same set of systems in them to look for ● All in all, it is clear that we have these four signatures for recognition. Non human animals have the same signature of both systems of object recognition In humans these traits emerge universally, across different cultures and groups These traits develop quickly within the first couple of days Object recognition is also learned from experience, but we have to build from some powerful inherent mechanisms Both nativist and empiricist are right for this one!
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'