1
Q
Logical reasoning
A
- induction
- deduction
2
Q
deduction
A
- ## a logical process in which the conclusion does not contain more info than the premises from which it’s based
3
Q
deduction: Modus Ponens
A
P→ Q (conditional statement) - All birds have feathers
P (propositional stated) - Robins are birds
Q (conclusion deduced) - Robins have feathers.
4
Q
deduction: Syllogism
A
P→ Q - If John is sick, then he will be absent.
Q → R - If John is absent, then he will miss his classwork.
P→ R - If John is sick, then he will miss his classwork
5
Q
Deductive arguments are evaluated in terms of ….
A
their validity and soundness
6
Q
validity
A
follows deductive logic
7
Q
soundness
A
premises are true and the argument is valid
8
Q
can you have a valid argument that isn’t sound? how about a sound argument that isn’t valid?
A
-you can have a valid argument that isn’t sound
P→ Q (conditional statement) - All birds have language
P (propositional stated) - Robins are birds
Q (conclusion deduced) - Robins have language
VALID BUT NOT SOUND
-but no way to have a sound argument that isn’t valid
-if it’s not valid and it’s not sound, it’s not a deduction-it’s a weak induction
9
Q
Induction
A
a logical process in which the conclusion contas more info than the premises from which it’s based. It’s uncertain. Ex: John is a grandfather John is bald All grandfathers are bald
10
Q
how are inductive arguments evaluated?
A
Inductive arguments are not evaluated for validity or soundness - they are strong or weak
11
Q
theory
A
- an idea about how something in the world works-don’t need to test to have a theory-it’s a guess
- we don’t really know anything-it’s all theories
- real definition: a statement about the probable relationship between things in the world
- Scientific theories are testable and make falsifiable predictions (falsifiable=something you can test and prove wrong)
- theories are inductive, but must allow for deductive reasoning
12
Q
The problem of induction
A
- can inductive arguments give us knowledge?
- there is no justification for generalizing from some number of observations, or an assumption that the world will always work the same in the future
13
Q
hypothesis
A
a testable prediction deduced from a theory
14
Q
the null hypothesis
A
- a general statement or default position that there is no relationship between 2 measured phenomena
- A null result falsifies the theory generated through inductive reasoning
15
Q
statistical inference
A
- estimate properties of a population based on a sample
- inferring properties of population based on what happened in an experiment with a subgroup of that population
16
Q
Karl Popper and falsification
A
- challenged the idea that findings confirmed a theory. Instead, Popper claimed that positive findings corroborated a theory, and tests should be designed to falsify theories
- If P, then Q; not P, therefore not Q
- Negative results require a revision of the theory
- doesn’t totally escape induction
- theories have more or less verisimilitude (i.e., degree of positive empirical support)
17
Q
Problems: one must assume a null model. How is reality distributed? is it always normal?
A
Induction again!
-we assume a normal distribution-if falls within that, accept null
18
Q
What are the alternatives to a null model?
A
- Have more sound apriori assumptions (bring in more realistic prior assumptions into the equation)
- Compare diff models
19
Q
What are the alternatives?
A
- Have more sound apriori assumptions
- Compare diff models
20
Q
What is the difference between induction and deduction?
A
Deduction: start with lots of info-funnel your way down to conclusion-only with info from top
Induction: all info follows logically from the statement above it-generalizing from a very specific point-conclusion has more info than premises
21
Q
What do validity and soundness refer to?
A
- validity: has logical flow-the deductive format flows logically
- Soundness: has to do with accuracy-the premises being true
- for deductive arguments
22
Q
How do you evaluate inductive arguments?
A
-by weak or strong
pretty subjective, so on exam will be pretty extreme ends of spectrum-very weak or very strong-so easy to tell
23
Q
John is a guy/John is a brunette/all guys are brunette. What is this?
A
-An induction
24
Q
Scientific theories must be ___ and _____
A
testable and falsifiable
25
What purpose does a null hypothesis serve?
-it’s better to prove things false time and time again to keep refining theory than to prove something true then let it go/not continue to test
26
What is verisimilitude?
How close your theory resembles reality-as you keep reworking it, your results are more and more mirroring reality
27
Theories can be ___, while hypotheses must be ____.
Inductive, deductive
28
Levels of explanation in the behavioral sciences
- For any trait, one can describe it at multiple levels
- proximate
- ultimate
- the proximate-ultimate distinction is a direct consequence of the theory of natural selection
- these interact-cultural mechanisms feed back into the genes and environment
- in order to fully understand a trait, need to understand both the ultimate and proximate mechanisms-they’re compatible-complementary-both important, though many social scientists are not interested in the ultimate explanation
29
ultimate explanation
we need to explain why these traits are present in the environment-why they have evolved
- species level events (not much of this-our genes don’t care about the species-evolution doesn’t try to save the species)
- the why
- concerned with the fitness consequences of a trait or behavior
- why is the trait favored by selection?
30
proximate explanation
- What are the mechanism by which those traits are implemented in the world?
- the how
- concerned with the mechanisms that underpin a trait or behavior
- behavior generators
31
proximate and ultimate explanations: why do we want ice cream?
- ultimate: Mechanism in our body that makes us want ice cream-because it tastes good-for centuries, sugar has been important to find in environment and difficult to find-behaviors to motivate you to find sugar and consume it
- not useful anymore, but leftover-used to be hard to find fats and sugars-not anymore-but still evolutionarily evolved
- proximate: doctors can explain how that works-taste buds-neurohormones-without even mentioning evolution
32
proximate and ultimate explanations: why do babies cry?
- proximate: includes both the external triggers of crying (e.g., physical separation from the caregiver, cold, or a lack of food) and internal mechanisms (the limbic system and the endogenous opioids involving the cessation of crying)
- has immediate causal triggers
- **as you can see, proximate explanations can sometimes seem like the why-but doesn’t actually explain why that trait exists for that species in the first place-just the apparent immediate cause (baby cries because cold)-so more the how of how that evolutionary function manifests itself, not the why
- ultimate: elicits care and defense from mothers and other caregivers
- appeals to the fitness benefits of the trait: infants that do not cry when in need of assistance are less likely to survive
33
proximate and ultimate explanations: why do male songbirds sing?
- proximate: increased daylight in the spring leads to increased testosterone production, which acts to activate a brain center than controls singing. Also, the role of experience in song-learning
- ultimate: singing functions to attract females and defend territory from other males
- enhances fitness
34
problems w proximate/ultimate distinctions reading
-designed to help clarify some errors the authors have noted in the literature in terms of how we think of proximate and ultimate-letter to social scientists
-but offers us a great learning opportunity to think about how to talk about different levels of analysis in these problems social scientists try to assess
-cooperative behavior: why are people cooperative? Why are we so nice to each other?
-some people say religion-that’s a proximate answer
-but why evolutionarily do we have that trait?
-linguistic structure and epigenetics-confusing, will not ask about on exam
-infant crying-already went over
-human reasoning-we engage in social exchange and so we’re particularly adept at IDing in a social exchange whether we’re getting a good benefit for the cost we paid-a lot of experiments revealed people good at reasoning tasks when frame them as being a cost and benefit-you need to think about why we’d be good at those reasoning problems-proximate: dedicated cognitive module for social exchange (there’s a part in our brain dedicated to reasoning), ultimate: need to pay attention to cost and benefit so not cheated in social exchange
-linguistic structure: lots of adaptive reasons why languages differ, and some cultural ones-but cultural transmission isn’t the reason for cling structure, why it varies (proximate)-we have it because there’s something about how our brains work that allow us to learn language-this has evolved evolutionarily (ultimate)
-why are misunderstandings of level of analysis common in the behavioral sciences
Proximate explanations are desired in the social sciences
The question of ultimate functionality is often not raised-don’t usually worry about evolutionary reason for why traits exist
Some terms have 2 different meanings: one at the ultimate level and another at the proximate level
E.g., cooperation, altruism, and spite are defined in terms of their net effects on inclusive fitness in biology (ultimate), but in psychology are often considered mental states (proximate)
Fitness: how many genes in your body are actually given on to next generation-how many of your genes make it to the next generation and the one after
Inclusive fitness: total count of your genes that are manifesting in the world-counting your offspring, and your family’s offspring (like siblings)
Altruism-just means nice to many, but to others about helping behavior for evolutionary purpose
Intentional language used in biology is an anthropomorphic shorthand for describing evolved strategies. But humans have conscious inventions that may or may not correspond to ultimate explanations.
Creates misunderstandings-people have conscious intentions that may or may not correspond to the ultimate explanation
We have a conscious narrative of why we’re doing something-moving towards goals we’ve set
Many things we do are unconscious
“Offspring are selected to demand more food than the parent wants to give” could be put more neutrally: “during the course of evolution selection acting on genetic differences in the begging behavior of offspring will have favored an increase in the intensity of begging, and this will have been favored to the extent that the level of begging by an individual offspring exceeds the optum level for the parent”
35
Marr’s 3 levels of analysis/Tri level of analysis
- David Marr-vision scientist-developed theoretical paradigm for levels of analysis describing psychological mechanisms
- can apply to any animal, trait, or piece of tech
- maps onto the ultimate proximate difference we’ve already described
1. computational
2. algorithmic
3. implementation
36
Marr’s 3 levels of analysis/Tri level of analysis: 1. computational
- what is the goal of the computation?
- Adaptive problem-what is this trait designed to do, through evolution? What is that thing for?
- Technological engineering
- Not saying how it works-just what is it designed to do?
37
Marr’s 3 levels of analysis/Tri level of analysis: 2. algorithmic
- what is the strategy of the computation? What is the mechanism?
- Information processing-you get some kind of info in some format-how does it actually get routed?
- Tells you exactly what is the processing strategy that tells you how something works
- The how-proximate explanation
38
Marr’s 3 levels of analysis/Tri level of analysis: 3. implementation
- How is it realized in physical material?
| - Neurons, electronic circuitry, etc.
39
Human reasoning
- domain specificity
- dedicated systems to solve 1 domain, diff systems for another-dissociated from each other-so can be good at 1 thing and bad at another-smart in some areas but not others
- human reasoning is a domain specific activity
- proximate: there are specialized computational devices that allow people to detect cheaters in a social exchange
- ultimate: in social species w interessed cooperation, those who better defend against exploitation(i.e. cheating) will have better inclusive fitness
- 1 adaptive solution: cheater detection (?)
40
domain specificity
-good at one thing, but not another
41
human social reasoning for social exchange: Wason selection task
- tool used by psychologists to measure human reasoning ability
- the task had always shown a “content effect” but it was not understood
- there’s a rule, 4 cards, which do you have to turn over to see if rule is followed?
- p and not q
- but when the problem involves cost benefit structure-performance goes from about 4% correct to 74% correct
- testing if reasoning is sensitive to the content
42
human social reasoning for social exchange
- Wason selection task - tool used by psychologists to measure human reasoning ability
- the task had always shown a “content effect” but it was not understood
- there’s a rule, 4 cards, which do you have to turn over to see if rule is followed?
- p and not q
- but when the problem involves cost benefit structure-performance goes from about 4% correct to 74% correct
- testing if reasoning is sensitive to the content
43
Information theory
-a mathematical theory of communication (1948)
-addresses the technical engineering problem of transmitting info over a noisy channel
-it’s not about info content
-”reproducing at one point either exactly or approximately a message selected at another point”-by message he means anything you can encode
-at this time, people discovering ways to take info and put them in 2 states-constitute a communication system-morse code developed from this
-he had idea that his could be applied to way circuits work
-then 11 years later came up with this mathematical theory of communication
-this addresses problem of transmitting info over noisy channel
-he was not thinking about people-trying to make a telegraph work
-not about info content-just about the plausibility of creating a communication system where you take digital info and transmit it over some channel
-the more elements of a message we sent simultaneously, the shorter it takes for the transmission
Info source--(message)--> transmitter--(signal)--> received signal→ receiver--message→ destination
**anything that’s not signal=noise
44
source coding theorem (Shannon’s model)
- on average, the number of bits (binary digits-0s and 1s) needed to represent the result of an uncertain event is given by its entropy (info uncertainty)
- info entropy is the average info of all possible outcomes (calculated through equation)
- higher entropy equals greater disorder (rolling a dice 1 time has higher entropy than flipping a coin because more possible outcomes with dice)
- info is the reduction of uncertainty (in any way-not content)
- flip coin: info in 1 flip=1 bit-encode that with a 0 or a 1
- how many possible outcomes are there, and how many outcomes did you deserve-then can get measure of how much info is in that 1 flip
- but dice-6 outcomes-have to calculate with equation in book
- so back to model: If entropy reduced, then you’ve communicated information-if entropy not reduced, you didn’t successfully communicate info
- information is reduction of uncertainty from left part of chart/model to the other
45
noisy channel coding theorem
- reliable (i.e., relatively error free) communication is possible over noisy channels provided that the rate (R) of communication is below a certain threshold, called the channel capacity (C)
- if R
46
Applying the mathematical theory of communication
- continuous signals in the world can be digitized into a binary code
- allows for compression and transmission
- ex: zipping files-reorganizing digital codes in a way that can be read on the another end-shrinks it down, sends it, opens it back up
- perceptual noise shaping-exploit properties of our hearing to throw out information we don’t perceive-rip a cd onto comp into wav file-10mb-then convert to mp3, now 1mb, sounds the same-gets rid of parts of the sound we can’t perceive-makes for faster transmission because file is smaller, and we can’t tell the difference anyways-so same as zipping files but then doesn’t open back up, just getting smaller
- Weaver really liked this idea-thought you could apply it to people
47
What is actually important if trying to apply this mathematical theory of comm to people?
- human communication as a problem of engineering
- Level A: How accurately can the symbols of communication be transmitted? (the technical problem-what Shannon was showing-if something can be coded, and capacity of channel allows it, then anything could be potentially encoded-the was thinking of tv signals-but you could potentially encode language this way-coming up with some kind of code to represent this is the technical problem)
- Level B: How precisely do the transmitted symbols convey the desired meaning? (The semantic problem-how well does the thing encoded capture the meaning of what you’re trying to convey-Language is symbolic-uses symbols that refer to things in the world-how do you talk about how well the code transmits the meaning-huge problem in language, not solved-what people actually mean when they talk-not always clear-people can understand a sentence in different ways)
- Level C: How effectively does the received meaning affect conduct in the desired way? (The effectiveness problem-if you do transmit the signal, does it actually accomplish what you wanted it to? Does it work? A language problem as well-did you figure out what you wanted to know from a person’s response to how the movie was?)
- we have to solve these problems to apply the mathematically model to communication-many think can’t be done-Weaver thinks can
48
problems with adapting to mathematical model to human comm
- issue is people trying to adapt model are not mathematicians-they just took the chart model and adapted it for language-the problem is these models don’t make any predictions-just intuitive suggestions on how language works, what’s important for communication-don’t really get us anywhere
- there's a bunch of these models-not sure what's right
- linear model of communication-but only 1 way-has to be 2-so changed to interactional model of communication-then transactional-because shared experience matters
- these are just intuitions about what matters about communication-doesn’t predict anything
- then shit starts getting crazy-like the Ross model-adding all these extra things that matter-context
49
7 Traditions in Communication Theory (reading)
- objective territory (objective analysis)-->interpretive territory (understanding human experience through subjective analysis)
- just different approaches-no better or worse here
- Objective: socio-psychological-->cybernetic-->rhetorical-->semiotic-->sociocultural--->critical-->phenomenological: Subjective/Interpretive
50
7 Traditions in Communication Theory: 1. socio-psychological
-cognitive science-measuring something specific
-Careful systematic observation of Cause and effect relationships in communicative interactions
The socio-psychological tradition operates on assumption that human reaction is rational and predictable-observe people and predict what they’re gonna do
-focuses on patterns of interactions rather than individual characteristics
-pro: if it’s true that we are predictable, then we could predict what causes people to act the way they do
51
7 Traditions in Communication Theory: 2. cybernetic
-similar-study of information processing, feedback and control on communication systems-the people trying to adapt Shannon’s model fall here
-Study of info processing, feedback and control
Different computer interactions that we have
-example: Siri
-pro: leads to more efficiency in communication
-con: lead to computational errors, nonhuman responses, system differences
52
7 Traditions in Communication Theory: 3. rhetorical
- art of using all available means of persuasion-public speaking, the socratic method, etc.-goes back to ancient Greece-how you make an argument, persuade someone
- Public speaking-socratic method-tools of persuasion-spoken-argumentation
- The study of the way we organize our words and ideas to create a persuasive article
- Limitation: just focusing on words could miss context or nonverbal cues, writing is different
- Put into the wrong hands, leads to things like dictators
53
7 Traditions in Communication Theory: 4. semiotic
old tradition, not really followed anymore-the study of verbal and nonverbal sayings that can stand for something else and how they impact society-has to do with the nature of signs and the nature of meaning
- The study of verbal and nonverbal signs that can stand for something else and how their interpretation affect society
- Nonverbal example: emoji or street sign
- Verbal: the inflection of your voice
- Drawback: among different cultures, different gestures have different meanings
54
7 Traditions in Communication Theory: 5. sociocultural
- everything is due to your own interpretative-we are subject to our interpretation of the world, that’s all we have-we can never be sure what’s really out there-incorporates idea of linguistic relativity-structure of language determines thought
- Analyzes communication as a transference of knowledge and experience which results in creation of culture and meaning
- large component: linguistic relativity: the language affects the way we perceive reality-creates own social reality
- pros: middle of spectrum, encompases lots of elements involved in communication
- difficult to test to experiment with-relies on ethnographic studies-hard to pinpoint in objective way
55
7 Traditions in Communication Theory: 6. critical
- certain way of talking about phenomena implicitly discriminate against certain groups, ways we think in science are biased against another viewpoints-creates power differential over time that really matters politically-subjective approach can determine everything about our experience-should look at ways of looking at things as just different ways-no one better-don’t pit against each other
- Talking and thinking is inherently discriminatory-this manifest itself to how power is divided
- Pro: can challenge unjust discourse and structures-
- Con: focuses just on the end result and not really on the communicative process itself-more outcome based than specifics of how the discourse does that
- Example: words themselves like policemen or firemen can have gender biases
56
7 Traditions in Communication Theory: 7. phenomenological
- what’s it like to do certain things-everything has its own associated experience-interested in trying to flesh out the specific of what constitutes an experiences in the world-and how does that play into communicative interaction-important but hard to measure, do experiment on
- tries to understand human communication through perspective of an individual and their experience and how they interpret their experiences
- pro: people perceive things through an idiosyncratic worldview, different interpretations of things-allows for that nuance and differences between people’s worldviews
- con: really subjective, rejects 1 overarching reality, hard to make sweeping claims about things, hard to generalize-not scalable
57
Shannon
- Often when communication theory is taught, these engineering/technical histories of the field are kinda ignored
- but this is where comm started-slowly started adding more factors that moved our models of communication more and more into the qualitative realm
- encoded, sent through channel, decoded, perceived by person on other side
- source coding theorem and noisy coding them-will need to know
58
Theories can be ___, while scientific hypotheses must be ____
inductive, deductive
| Can’t test an inductive theory
59
What is the ultimate level of explanation?
How it relates to inclusive fitness/fitness of a species
| concerned with the fitness of consequences of a trait or behavior - the “why”
60
Offspring are selected to demand more food than the parent wants to give” is an example of what?
Intentional language-actions that are performed unconsciously are made to seem like being performed intentionally-kind of a shorthand in writing, but can confuse people
61
Which marr’s level(s) correspond to proximate explanations? To ultimate?
Computational (the problem), algorithmic (how, not why), implementation (how manifested in physical systems)
Proximate goes with algorithmic and implementation
Ultimate goes with computational
Greg argued relationship between those 2 sets
62
People prefer symmetrical faces in pursuing romantic partners-what would ultimate and proximate explanations for this behavior be?
Proximate: chemicals in brain signal you’re attracted to someone
Ultimate: face symmetry means fit-will have fit offspring
63
A drawing of a circuit board corresponds to which level?
Algorithmic (planning out how you’re going to accomplish your goal-the physical/algorithmic would’ve been the physical part of board-wiring)
64
One of the purposes of this article is to address that social science commonly mistakes ___ for ___
Proximate for ultimate
65
Example of a null hypothesis
There’s no difference in achievement between boys and girl.
| There is no difference in care levels between a baby that cries and a baby that does not cry.
66
What does the computational level deal with?
Ultimate goal, problem to be solved-abstract level
67
```
What are these examples of:
Blueprint for a building
A recipe for how to make cookies
Directions to the mall
Programming code for an iphone
(how to accomplish something-strategy)
```
the algorithmic level
68
What is the proximate level of explanation?
The mechanism for how something happens-that underpin the trait or behavior-the how
69
What is the difference between algorithmic and implementation?
Implementation is the physical manifestation of the algorithmic
70
What is transmitted culture?
Things we learn through imitation or by being taught, or another forms of social transmission
71
Peppered moths are colored to blend into trees. Give a proximate and ultimate explanation for this trait
- ultimate: it camouflages them so they can better survive to reproductive age
- proximate: genetic makeup
72
You have an intersection with 10 accidents/ady. Propose a solution to the problem using marr’s 3 levels
Computational: you’re the city planner-your goal is to decrease accidents at this intersection
Algorithmic: strategies-draw out different road maps you could change, figuring out timetable of lights changing
Implementation: pour in concrete for new road, install stop light
73
Talk about Marr’s 3 Levels with regards to a wooden pencil
Computational: want to be able to write ideas down on paper
Algorithmic: design a pencil-figure out shape that might fit in the hand, how long to make the point, etc
Implementation: assembling the wood, graphite, metal, etc.
74
Give one ultimate and one proximate explanation for human cooperative behavior.
Ultimate: Helping behavior is reciprocated, increasing chances for survival
Proximate: empathy, compassion that causes you to want to help-also expected and encouraged in society-strong reciprocity-it makes you happy to make friends-religion
75
Cultural evolution-main principle
-culture evolves in a Darwinian manner—that
is, through a process of variation,
competition, and inheritance.
-Relative importance of different factors varies
between biological and cultural evolution, but
they can both be described as Darwinian.
76
Culture
```
-Information that is transmitted from
one individual to another via social
learning, rather than information that
is transmitted genetically, or that is
invented de novo by an individual
independently and not transmitted.
-Cultural phenomena are everywhere:
Language, art, technology
-what about nonhuman animals? Humpback whale bubble net technique-work together to get food
```
77
Models of cultural evolution, and why do we use models?
-Modeling can shed light on complex
phenomena by simplifying them, removing
inessential features, and focusing solely on
those features we suspect are important,
or necessary to test a particular
hypothesis.
-mathematical
-experimental
-Modeling—both mathematical and experimental—is an important
complement to observational research such as ethnographic field studies,
historical and archeological research, and statistical data mining.
78
Mathematical models of cultural evolution
-Mathematical models provide a way of
running formal thought-experiments, by
analyzing the consequences of a certain set
of assumptions that are thought to hold in
the real world, and provide a level of
precision that is unattainable through
purely verbal models.
79
Experimental models of cultural evolution
-Laboratory experiments can also be used to
model real-life instances of cultural evolution,
allowing hypotheses to be tested under
controlled conditions (again, removing all
inessential features of reality)-Allows detailed records of behavior that are
often not possible with real-life (e.g.,
archeological or ethnographic) data.
-Experiments add some degree of external
validity in that they measure the decision making
processes of real people, who, due to
imperfections or biases in that decision making,
may act in ways not captured or predicted by
formal models.
80
How are cultural traits changed
| during transmission?
-Traits change simply through the process of learning, or copying, on the part of
a naive individual-The ‘transmission chain’ method (Bartlett, 1932)
-Social information in stories was transmitted more accurately and lost less
frequently than nonsocial information (mutations aren’t random)-Social brain theories
-Models build in priors (Bayes)
-Open diffusion methods (e.g., danger learning)
81
The ‘transmission chain’ method (Bartlett, 1932)
• As they were transmitted, traits tended to lose detail and increasingly
resemble the individuals’ preconceived notions (e.g., War of the Ghosts
memory experiment)
82
social brain theories
posit that human cognition evolved primarily
| to deal with social information
83
War of the Ghosts
| memory experiment-Barlett, 1932
Aim:
To investigate whether people’s memory for a story is affected by previous knowledge (schemas) and the extent to which memory is reconstructive.
Procedure:
Bartlett asked British participants to hear a story and reproduce it after a short time and then repeatedly over a period of months or years (serial reproduction).
The story was an unfamiliar Native American legend called “The War of the Ghosts”.
Results
The participants remembered the main idea of the story (the gist) but they changed unfamiliar elements to make sense of the story by using terms more familiar to their own cultural expectations.
The story remained a coherent whole although it was changed.
It became noticeably shorter for each reproduction. Bartlett concluded that remembering is an active process.
Memories are not copies of experience but rather ”reconstructions”.
Evaluation
The results of the study confirm schema theory (and reconstructive memory), but it was performed in a laboratory and can be criticized for lack of ecological validity.
Participants did not receive standardized instructions and some of the memory distortions may be due to participants’ guessing (demand characteristics).
In spite of these methodological limitations, the study is one of the most important in the study of memory.
84
How and why do cultural traits accumulate over time?
- Cumulative cultural evolution
- Transmission chain method
- mechanisms of social learning
- cultural traits can be altered through transmission
- cumulative cultural transmission can create more advanced artifacts
85
Cumulative cultural evolution
- the presence of traits that have been gradually modified and built upon over successive generations
- standing on the shoulders of giants
- take existing idea-build upon it gradually
- beyond the capabilities of an individual to invent in 1 lifetime
86
Transmission chain method
- individuals solve some task with clear goals and measures of success
- onlookers then try to solve asme problem
- quality of solutions improved over time
- getting the benefit of earlier people’s trials and errors
- selection of good and bad approaches to some problem
- people aren't even aware what went into their solution once solved-how many other people have tried to solve that brought you here-just work with what have and solve it-but that’s from other people
- solutions evolve from pre-existing structure-gradual selection of variance of pre-existing structure
87
Mechanisms of social learning matter
-to have cultural evolution, going to need to have systematic social learning happen-not just copying random things people do (though kids will sometimes do this-copy even irrelevant behs in experiments-but when it comes to solving problem, realize what needs to be done, stop copying stuff that’s not helpful-certain social learning biases will guide you in different ways-we don’t just copy each other randomly)
88
Mechanisms of social learning
- imitation
- emulation
- teaching
89
Mechanisms of social learning: imitation
- copying behaviors (e.g., motor actions/procedure)
| - see a lot in technology
90
Mechanisms of social learning: emulation
- copying end products
- some sort of goal or task-don’t necessarily have to do it same way as person before, if can think of way to get there/to their end product another way -different way to get to same answer
- this happens in technology as well-people see idea, try to copy end product, may do it totally differently-apple vs microsoft
91
Mechanisms of social learning: teaching
transmission of explicit advice
- seems to be unique to humans
- actually informing through some signaling how to go about doing a thing
92
What social learning biases do individuals use?
1. content biases
2. model biases
3. frequency-dependent biases
93
Social learning biases
- reliably developing rules that people use when copying other individuals
- specifying, for instance, what or from whom to learn
94
social learning biases: 1. content biases
individuals pick a particular trait over others because of predisposed preferences for a characteristic of that trait (e.g., dangerous animals)
-example: show people pic of wombat looking scary-show to small child in Peru-they’ve never seen a wombat-then say a wombat likes to eat carrots and if it bites you, you’ll die-then show another pic of animal haven’t seen-say likes to eat meat and not going to hurt you-give whole series of animals-diet and whether or not will hurt you-go back 2 weeks later with same cards-ask what it eats-they probably won’t remember-random in remembering that-but they will definitely remember if it hurts them or not-this experiment performs pretty similarly anywhere you go -shows content bias in terms of learning danger-kids learn danger information a lot better than they learn diet information-reliably developed system looking for information about danger-important for them to know-they run around jungle all day-really good at learning about danger-just like any other kids-pick up on particular trait over others due to predisposed preferences-in part because of evolution
95
social learning biases: 2. model biases
- individuals adopt traits because of a characteristic of the person who knows the trait (e.g., prestige bias)
- if someone really knows a topic, or has prestige in some way, kids will listen to/believe them over another, even someone they know well
- also works when talking about how fashion goes-prestigious person wears something, others copy
- applies to things other than social learning-including how memes spread
- people pay attention to prestigious and sometimes dominant people
- dominant has to do with physical formidability
- prestige about social status
- these are independent of each other
- in learning, prestige dominates
96
What are the population level consequences of different social learning biases?
- the mathematical tools used to address this question are those relating to dynamical systems (using fixed rules to describe a position in space that is dependent on time)
- models and experiments can elucidate the different trait frequencies across populations
- example: prosociality-why are people nice? By default pretty much-we tend to help others in distress-people are generous-how does that trait spread-people who study cultural evolution have tried to create mathematical models for how prosociality has evolved culturally-everyone is nice, some people are more nice-why? These models can help us figure out by quantifying different aspects we think are relevant-figure out what aspects of prosociality are important for its spread throughout a population
- basically: you can use mathematical tools to look at how different traits vary in frequency across different populations
97
phenotypic value
- Your phenotype is you-your genotype are your genes underneath you that can give rise to many different phenotypes
- phenotypic value: What is the value of that particular trait in its success being expressed as a phenotype-something that causes a trait to be more frequent than another one-winning the evolutionary trait-some traits will spread through a population and take over, and others will die out-or maybe will have traits that will compete, sometimes 1 more popular, sometimes the other-keep switching-or like a 1 hit wonder song, winning for a minute, then quickly fall behind
98
What is a meme?
- a replicating cultural unit operating under Darwinian evolutionary principles analogous to a gene (Dawkins, 1976)
- in layman terms: a concept or idea, embodied by a word, a phrase, a riff, image, gesture, etc. A meme “exists” in the world of ideas and replicates by imitation
- a meme is a behavior or set of behaviors (memplex) which is learnt by observation and imitation
- internet memes have properties of real memes-but not the real definition of memes
- the word meme is a meme-all words are memes-anything can be a meme
99
memeplexes
- religions, ideologies, and wider concepts are proposed as “memeplexes” or groups of memes analogous to an organism (sets of genes replicate as single organisms, sets of members may replicate as memeplexes)
- when a bunch of genes traveling together, make organisms-they usually travel together-couples of genes that usually seem to co-occur-free rider genes that just attach to successful gene and go with it-happens with memes too-free rider memes-goes along with a more successful meme or song or joke
100
how are memes a form of evolution?
-you can take units of information - how to divide those units up can be tricky - these units of information replicate in some fashion that looks like biological evolution - because variation, survival rate, and inheritable-not genetic-but they have an environment they exist in, they propagate in some way-some will survive differentially more than others
101
What’s the difference between a trend and a meme?
- trends are memetic-a trend is a way of describing the increasing frequency of a particular meme or set of memes
- a trend is like a meme pattern
102
theoretical problems with memes?
- memes are often systematically transformed during transmission - a process quite unlike natural selection, which depends on 1 meme spreading more quickly than competing alternatives
- gate example: when put “-gate” on the end of something-like “watergate”-add “gate” for government scandal-but many people don’t know the origins now-use it in all sorts of ways without knowing original context-someone stole my toothbrush-”toothbrush-gate”-original meaning gets lose
- nonselective forces: intentional modification for improvement-people synthesize their own beliefs after being exposed to a number of people who behave differently
103
to replicate, memes must pass through 4 key stages:
1. assimilation: multimodal perception by an individual
2. retention: memory
3. expression: by some motoric act, speech, or gesture, which can be perceived by others
4. transmission: to another individual
104
functional division of a memetic system:
i-memes
e-memes
-internal thing, then expression of it
105
i-memes
-the internal transmission/storage of memes within our CNS
-neural network connectivity profile
-An i-meme (internal representation of a meme)
comprises the neural network which encodes it.
106
e-memes
-the external transmission/storage of memes in the world.
• Fidelity concerns
-How close if the meme to where it started by the time it gets to destination
107
Can we measure memes?
- so yes, we can measure memes if we can measure neural connectivity that’s associated closely with these thematic units-we can measure their implementation
- the i-meme constitutes the full set of “meme-specific” neural connections enabling e-memes to be perceived and transmitted by a communicative motor action in any one of its forms
- an i-meme (internal representation of a meme) comprises the neural network which encodes it
- lots of emotional and cognitive connections to different words- say “ammonia”, you can smell it, disgusted reaction
- there is tech of measuring neural connectivity-maybe with tight enough controls and good exemplars, could get neural connectivity signature of different memes -see how 1 person's representation of something is affecting another’s representation of it
Comm 100
flashcards
Decks in class (2)
# Cards
