Study Guide for COMM402 Spring 2012 Exam 3 (FINAL EXAM)Chapter 6: 1. What is the general picture of adaptation and its relationship to communication?a. A person acts, the world reacts, and then the person needs or chooses to adapt to the world’s responseb. “Adaptively rational”2. What is the basic model of the chapter?a. P ( t + 1 ) = P ( t ) + some increment3. Understand the Alfred examplea. Reinforcement learningb. Initially random behavior, finds reinforcement, adapts to reinforcementc. Through trial and error, behavior becomes less random4. How could we model the increment? What are the possible models and which one works?a. Constant increment modeli. Poor modelii. Eventually get a probability larger than 1b. Constant proportion modeli. Represents how much the subject has yet to learnii. At each trial, subject learns a constant proportion of what he has left to learniii. Produces an asymptotic curve where the behavior approaches a probabilityof 15. What are the assumptions of the various adaptation equations?a. Alternate behaviors are availableb. State of the individual is described by probabilities that add up to 1c. World responds differently to various behaviorsd. State of the world is described by various probabilities of various responses to each behaviore. Set of possible events is a combination of possible behaviors crossed by possible world responsesf. Specify adaptation equations for each possible event6. Make sure you understand the four adaptation equationsa. 4 eventsb. L and rewardi. PL ( t + 1 ) = PL ( t ) + a [ 1 – PL ( t ) ]c. L and no rewardi. PL ( t + 1 ) = PL ( t ) – b PL ( t )d. R and rewardi. PR ( t + 1 ) = PR ( t ) + a [ 1 – PR ( t ) ]e. R and no rewardi. PR ( t + 1 ) = PR ( t ) – b PR ( t )7. What is “a” in the equations? What is “b” in the equations? What’s the difference between them? Are they related in some way?a. “a” is the rate of learning with rewardb. “b” is non-reward rate of learningc. not related in any specific way8. Understand what expected probability is and know how to calculate the expected probabilities (e.g., what is the expected probability of going left after trial 2 in the Alfred example?). Also understand the implications for expected probability. a. Have a collection of subjects we want to modelb. Randomness in their behaviors initiallyc. EP = multiply the probabilities for each option across tree, then multiply by the matching value of probability, then added to get TOTAL EXPECTED VALUE9. What factors may affect “a” and “b” in the adaptation equations?a. Ability of the subject to draw inferencesi. Individual differencesb. Motivationsi. Rewards that are highly valuableii. Extreme depravation of any kindc. Noveltyi. Familiarity or lack thereof with the subject10. Understand different reward schedules: consistent, single probability, double probability, and patterned.a. Consistent – reward is always in the same placeb. Single Probability – reward only depends on one partyc. Double probability – reward depends on two partiesd. Patterned 0 reward schedule follows a pattern; crack the pattern and you can get rewarded 100% of the time11. Understand the graph on page 291 and related creativity example.12. What are the three possible kinds of theories?a. Magic Theories – cannot be proved wrong because any possible observation can be made to fit the theoryb. Perfect theories – can be proved wrong, but is so good that it is almost always correctc. Imperfect theories – can be proved wrong, and is proved wrong; makes erroneous predictions13. What’s superstitious learning? How is superstition different from adaption? What is a benign world? What is a malevolent world? How do we model superstitious learning?a. Environment is indifferent to a person’s behaviorb. Dispenses reward and non-reward on some basis other than a person’s behaviorc. Individual continues to behave as if the environment is responsived. Benign world – rewarded either way, regardless of chosen behaviori. Falsely believe you are controlling the rewardii. Suffer from insufficiently explored optionse. Malevolent world – unresponsively non-rewarding, regardless of a person’s behaviori. Probability will oscillate around 0.5; keep trying14. If Alfred starts at PL(0)=.50, where does he end up after a lot of trials in a benign world orin a malevolent world?a. Benign – close to 1.0b. Malevolent – 0.515. Understand textbook examples of superstitious learning (e.g., hair dye, adopting personality traits, and successful executives).16. What is mutual learning? Understand joint event and how to calculate probabilities of joint events. a. When other people are learning about us as well – reward schedule is not passivei. Also called mutual accommodationb. For each person write out possible eventsc. Create a dual equation table to find nidividual probabilitiesd. To get probability of joint event, multiply the individual probabilities17. Understand the mating example: Henry & Mildred and the table on page 310. 18. How do you calculate the expected probabilities of a person’s choice in mutual adaptation? Hint: Know where the numbers on page 311 and page 312 come from. 19. Study the hunting model. What are the differences and similarities between the mating model and the hunting model? a. Hunting – mutual adaptation, but what is reward for one person is non-reward for the otherb. Mating – mutual adaptation, but reward is same; either mutual attraction or mutual avoidance20. Know where the numbers on page 319 come from.Chapter 7:1. What classic book began the study of diffusion in our field and why it that book important?a.2. Be familiar with the five assumptions of diffusion models discussed in the chapter and summarized on page 351 of your text. a. CONTACT - There must be some sort of contact, connection, or relative exposurebetween individualsb. TRANSMISSION - Object of diffusion must be transmitted by someone who already has the informationc. ACCEPTANCE - Transmission of object of diffusion must be acceptedd. YOU HAVE IT OR NOT – either have information or you do note. NEVER LOSE IT – once have information, you continue to have it indefinitely3. Be able to explain why some assumptions may not apply to diffusion in real world situations (e.g., simplifications made for mathematical reasons).a. Sometimes will get a fraction of a personb. Don’t necessarily accept the infoc. Might lose the info over time4. What is