lecture notes—first midterm Flashcards Preview

🚫 PSY100H1: Introduction to Psychology (Winter 2016) with J. Vervaeke > lecture notes—first midterm > Flashcards

Flashcards in lecture notes—first midterm Deck (34)
Loading flashcards...

The Five Insights of Psychology

  • insight 1: experience is a projection.
  • insight 2: projection is due to unavoidable limitations and problems brains face in processing information (plethora, pollution, partiality, and projection).
  • insight 3: every experience is ultimately pattern processing in your neural network because every experience is your brain solving information processing problems using the same machinery.
  • insight 4: science fine tunes our brain by both extending its natural tendency to experiment and by challenging its projection of common sense experience.
  • insight 5: science is a complex multilevel process in which framework, theory, and explanation, prediction and observation interact in a complex, self-correcting, but not perfect, manner.
    • this is especially difficult in psychology because we have to explain the projections, and our study “objects” are projecting onto us.


The Six Influential Frameworks of Psychology

  • the six frameworks: psychoanalysis, behaviourism, information processing/computational, humanist/existential, positive, and evolutionary.
    • are the most current and influential in the practice of psychology.
  • everything you’ll come across is based in at least one of these frameworks, and you must understand these to be literate in psychology.
  • main lesson: history and current practice have produced multiple and influential frameworks that generate powerful insights and ways of thinking about the mind and behaviour.
  • but be on your guard: determine which framework is at work, and know its problems.
  • there’s no perfect experiment, no perfect theory, no perfect framework.



  • was founded by Freud, and Jung.
  • Freud's insights:
    • the unconscious—it’s so pervasive that we automatically think that we have an unconscious.
    • interactionism in development—development isn’t due to just nature or nurture, but how they interact with one another.
      • the effects of these interactions could be reflected in different stages of development for different parts of the psyche.
      • these mismatches caused difficulty for the individual.
    • multiple systems in the psyche
      • each system works according to difference principles and motivations; very similar to modern day dual and triple processing models of cognition.
      • id: operates according to the pleasure principle and works largely by association.
      • ego: works according to the reality principle and makes use of inferential/logical processes.
      • the contrast between id and ego is very similar to current proposals of the difference between System 1 and System 2 styles of processing.
        • e.g. when you go grocery shopping, you expect a certain amount of groceries to have a certain sum despite not having actually added all of it up.
      • superego: represents the internalization of parental and cultural values.
  • Jung challenged some of Freud’s specific claims but more importantly made changes to the picture of the psyche that are currently influential.
    • challenged conflict “hydraulic” model (Freud was in ancient times and used a lot of language that had to do with things “under pressure” or being “repressed”) with a dialogical organic model.
    • more in line with current concepts of how parts of the brain “talk” to each other and how the mind/brain is self-organizing in nature; we shouldn’t understand the mind as a machine, but an organic being.
  • main problems with Freud:
    • reliance on case study methods, and often vague theoretical constructs makes many consider psychoanalysis not even scientific psychology.
    • his patients came to him with very specific problems, and are probably not a good representation of the entire population.
    • many of his patients would be cured because they come to believed Freud’s ideas, which doesn’t really say if they were good ideas.
    • vague words such as “experience” don’t actually explain anything.
  • clear instance of the power of framework to generate important insights without creating clearly testable theories.



  • emerged out of the work of Thorndike, Pavlov, and especially Watson.
  • Thorndike in America; interested in intelligent pets and the law of effect.
  • Pavlov in Russia; classical conditioning.
  • Watson in America; rejected introspection-based methods of psychology.
  • behaviour explained only in directly observable terms = stimuli and response.
  • developed very rigorous and reliable experimental methods and procedures.
  • a lot was learned about the basic machinery of learning.
  • main problems: too simplistic a view of experimental science.
    • what about atomic theory or evolution? we can’t see those either; can’t limit ourselves.
    • also, learning is not passive storage but can be insightful in nature.
    • organisms usually respond not to the stimulus but its “meaning” or relevance to them.
      • e.g. if you saw a video of fire, you wouldn’t run away because you know it’s not dangerous.
    • many features of the mind, such as language, cannot easily be explained by stimulus response models, Chomsky vs. Skinner.
    • the failure of perceptrons (more later on); there are many things that a computer can’t be programmed to learn that mammals do.
    • creation of computer theory and computers that gave a new rigorous way of talking about inner processes.


Information Processing/Computational Approach

  • this was the main alternative that drove what was called the cognitive revolution in psychology.
  • cognition is how information is transmitted/communicated from one processor to another, i.e. how information is stored, transformed, and applied in action.
  • computer = logic machine
    • information is organized logically; this is the software.
    • the machine is organized causally; this is the hardware.
    • one machine state causes another machine state that corresponds exactly to how one logical state implies another.
    • they perfectly track each other = a program.
  • main analogy: as software is to the hardware, so the mind is to the brain.
  • psychology studies the software, the programs, while neuroscience studies the hardware, i.e. the states of the machine.
  • the information processing approach wed mathematically rigorous theory to experimental testing so that inner processes could become a part of scientific psychology.
  • gave birth to cognitive psychology as one of the most important sub-disciplines…
  • main problems:
    • very hard to get original meaning into the system.
      • your computer doesn’t care about what it’s processing, whereas everything as meaning to you.
    • very hard to get relevance into the system.
    • hardware and software are independent and not qualitatively developmental.
      • qualitative development is when something acquires new abilities to get new kinds of information.
      • e.g. if you’re using a computer and all of a sudden it starts baking a cake.
      • your brain can do this, but computers can’t, so your brain isn’t really a machine.
    • concentration on order of functions and not timing.
  • in a lot of important ways, you both are/aren’t a machine.
    • AI often involves more than computation nowadays.
  • this method, although powerful, is too limited in its understanding of the brain.


Humanistic / Existential Framework

  • emphasizes the role of meaning making, qualitative change, and self-interpretation/self-definition missing from the information processing/computational framework.
  • existence precedes essence for human beings.
  • human beings have existential models, i.e. ways in which they co-define themselves and their world.
    • we’re constantly engaged in existential modes; we’re constantly defining ourselves and our world at the same time
  • consider the work of Fromm and the being mode vs. the having mode.
    • having mode is organized around having needs, relies on categorical cognition, I-it identification, and manipulative problem solving,
      • “I have this bottle of ginger ale, but I would be okay if you replaced it with another bottle of ginger ale”
    • being mode is organized around developmental needs of becoming, relies on expressive cognition, I-thou identification, and rational meaning making.
      • becoming mature or moral.
      • you can’t treat your significant other the same way as the bottle of ginger ale.
      • your significant other wants to matter to you; engaging in an I-thou relationship.
  • neither mode is good or bad, what is bad is modal confusion.
    • that is, we use the wrong mode for the need that we have.
    • maturity is a being mode, but many people just go out and buy a car or drink alcohol.
    • instead of being in a relationship, you have a lot of sex.
  • consider the work of Maslow and his theory of the hierarchy of needs and the roles of self-actualization and peak experiences.
  • main problems:
    • relies too much on philosophical argumentation (not a bad thing!) and not enough on experimental evidence. kind of the opposite of behaviourism. yet improved ethics in psychology.
    • difficult to know how to compare theories and get theoretical debate and empirical competition going between them.
    • influence on positive psychology framework.


Positive Psychology

  • positive psychology and the complement to the pathological framework.
  • properties of the mind only revealed in excellence.
    • e.g. Csikzentmihalyi and flow as optimal experience.
    • when everything else drops away and you just get into that “zone”.
    • this is a universal experience; this tells us something important about how the human brain works (“nirvana”?)
  • positive psychology reveals things that we wouldn’t notice by just looking at the average state.
  • or wisdom
  • main problems:
    • in its infancy and has not yet developed into a coherent framework.
    • other than the contrast with the pathological approach, it’s not clear what unifies the positive framework.
    • central constructs such as happiness and wisdom are difficult to clarify and operationalize.


Evolutionary Psychology

  • evolutionary psychology is based on understanding cognition and behaviour in terms of its original adaptive function.
  • basics of evolution by natural selection:
    • reproduction
    • scarcity and competition, and reduction in design option
    • mutation
    • sexual reproduction and variation, and increase in design option
  • evolutionary psychology pays attention to the Evolutionary Environment of Adaptation (EEA)
  • evolutionary psychology often offers plausible explanations for behaviours that initially seem bizarre or “unnatural”
  • main problems:
    • circular just so stories because little independent evidence for EEA specifics.
    • some things are just a part of your genetic code and aren’t exactly advantageous to you (e.g. spandrels; men’s nipples)
    • overly simplistic accounts for behaviour (e.g. religious)
    • selection for exaptation: a trait that has been co-opted for a use other than the one for which natural selection has built it.


Five Qualities of Scientific Research

  • based on measurements that are objective, valid, and reliable; it can be generalized; it uses techniques that reduce bias (projection); it’s made public; it can be replicated.
  • objective measurement: within a margin of error, the measurement is consistent across instruments’ and observers.
  • consistency: don’t contradict each other
  • coherence: speak the language
  • convergence: independently come to the same result
  • the more we have of these, the more we can be sure that what we’re finding is real and not a projection or individual bias.


Operational Definitions

  • in order to check for consistency and coherence, we need to be able to compare results using the same measures; this requires operational definitions.
  • e.g. you wanted to study if shyness causes loneliness. if different researchers use different definitions of shyness and loneliness they could come to exactly the opposite conclusions about the hypothesis.
    • shyness = infrequent communication or contact with others (doesn’t want to connect).
    • shyness = high anxiety in the presence of others (wants to connect but finds it difficult).
    • depending on how you define shyness, you will come to completely different conclusions about whether or not the original question is true or false.


Measurement Validity

  • measurement validity: the degree to which an instrument or procedure actually measures what it claims to measure.
  • not the same as logical validity.
  • e.g. how do I measure anxiety?
    • how about perspiration? is this valid? people perspire for a number of reasons, so it’s doubtful that you could measure anxiety validly with perspiration.
  • infinite regress of validity? i.e. how do I measure a measure?
  • you make these judgments about the validity of your measurement relative to how much objectivity it’s giving you based on good theory and reasoning.
  • you have to do a lot of work before you gather data.



  • reliable: when a measure provides consistent and stable answers across multiple observations and points in time.
  • three types: test-retest, alternate-forms, and inter-rater reliability.
  • using alternate forms of the test makes it even more reliable than retesting.
  • in some cases, you video tape people doing the experiment and have a person watch the video back.
    • e.g. do children react violently to violent movies? make them watch a video then hit the blowup clown doll.
    • you can’t just have people watch the child’s behaviour; what exactly determines violent behaviour?
    • you want a test so that different raters will come to similar conclusions about how violent the child is in the video.
    • you have to give very careful instructions to define what “violence” is.



  • generalizable: the degree to which one set of results can be applied to other individuals, or events.
  • science, competence (not just performance), explanation, and causation as opposed to just description.
  • laws as universal generalizations.
    • e.g. performance is every sentence you’ve ever said. but it doesn’t measure your competence.
    • if someone says to you, “currently, there are no hyper-intelligent squirrels on mars mining for emeralds.”
    • even if you’ve never said this sentence before in your life, you can still understand what this means. this is competence.
    • not just explaining what has been done, but what can happen.
  • generalizations point to causation and laws, and therefore supports deep explanation.
  • science affords intervention in reality, i.e. power to change things.
    • it’s not just about describing the world, but explaining the causes behind things.


The Generalizability-Discrimination Tradeoff

  • the more generalizable I make my claim, the less it applies to specific people in specific contexts.
  • sometimes, you have to give up some generalizability for discrimination; to give a more contextual explanation for what’s going on.
  • increasing the evidence pool (e.g. number of participants) is one way to increase generalizability.
  • the optimum way is to measure all people at all time; you can’t really ever measure the whole population—the population is the group the researcher wants to generalize about.
  • science almost always relies on samples.
  • the problem is properties and patterns of the sample may not be properties of the population, i.e. will not generalize.
  • most of the samples we take are from WEIRD people (White, Educated, Industrialized, Rich, Democratic). yet we generalize for everybody based off of these samples.


Sampling Bias

  • random sample: a sample technique in which every individual of a population has an equal chance of being included.
    • a true random sample would mean that every single person on the planet has an equal chance of being selected for your study.
    • however, this is highly impractical.
  • convenience sample—a sample that is convenient for people to use (e.g. the PSY100 students at U of T).
    • the more convenient it is, the less random it is, the more it’s likely to show sampling bias.
    • there’s no solution or way out of this.
  • ecological validity: the results of a laboratory study can be applied to repeated in a natural environment.
    • the more you design your experiment to be precise, the more artificial (unlike the real world) the situation becomes; the real world is messy and uncontrolled.
    • generalizability vs. experimental precision (opposite of vagueness or ambiguity) tradeoff; this is also unavoidable. 


Single and Double-Blind Studies: Reducing Bias

  • single-blind study; necessary deception; no Hawthorne effect.
  • a double-blind study is even better. no clever Hans effect.
    • Hans was a horse, whose owner claimed he could solve arithmetic problems.
    • there was no deception involved—he would stamp out the answer.
    • but it turns out that Hans would just stop when the owner tensed up.
  • a while ago, in the media, claims came out that people had psychic abilities.
    • they completely blocked off a person’s senses and brought other people into the room; the people then said that they could sense when there was actually someone starting at them.
    • someone repeated the experiment and removed a feedback element, people performed just at chance.
    • in the first test, the experimenters used a very complex pattern of bringing in another person and the participants were implicitly figuring it out.
  • memory tries to be predictive; think about those weird coincidences: “I was thinking about X and then X called.” well, think about all the times you think about A, B, C, and D, and they don’t call.


Peer Review

  • science produces results that are made public; peer review
  • peer review isn’t perfect, but it means that whenever you want to present something it has to go through theoretical filter.
  • openness to criticism, theoretical challenge and counter-argument and counter-evidence.
  • avoid errors due to lack of significant alternative explanation.
  • science uses IBE (Inference to the Best Explanation)—we put a bunch of explanations against one another, and the one that wins is accepted.
  • sometimes, there just isn’t any significant other explanation—yours just seems like the best because it’s the only one.


Replication and Convergence

  • making the results public also affords another important feature of scientific research; namely, replication: the process of repeating a study and finding the same result each time.
  • although we talk about replication a lot, we don’t actually do it nearly as much as we should because it’s hard to get replication published.
  • convergence so even if it is a competing or neutral theorist they get the same results.


Five Features of Poor Research

  • five features of poor research: not falsifiable, anecdotal evidence, data selection bias, appeal to authority, and appeal to common sense.
  • note how all of these characteristics make bias (projection) much more likely and powerful.


Descriptive Methods of Collecting Data | Empirical

  • descriptive methods: these are methods that establish the appearance of some form of behaviour, and/or its duration, and/or its frequency.
  • one of the main difficulties facing descriptive methods is coming up with an apt description, i.e. one that covers many cases but not too vaguely to include other forms of behaviour.
  • three types of descriptive methods: case study, naturalistic observation, and surveys and questionnaires.


Case Study | Empirical

  • case study: (used in psychoanalysis) an in-depth report about the details of a specific case.
  • excellent for falsifying a particular claim.
    • e.g. amygdala is needed for fear.
  • excellent for providing evidence of individual differences.
  • very poor for generalizability, again the tradeoff.


Naturalistic Observation | Empirical

  • naturalistic observation: unobtrusive observation and recording of behaviour as it occurs in the subject’s natural environment.
  • requires clear operational definition and attention to specific variables to increase objectivity.
  • excellent ecological validity at the expense of very poor precision.
  • no systematic study of the relation between variables, no good causation or explanation.
    • e.g. there are no studies of why kids are crying on the playground; you don’t know what factors are actually causing it. you don’t even know that they’re crying because they’re actually sad.
  • possibility of paying attention to irrelevant factors.


Surveys and Questionnaires | Empirical

  • surveys and questionnaires—this lets one get at participants’ attitudes, beliefs, and opinions.
  • main difficulties:
    • not asking leading questions,
    • checking for validity (against clinical diagnosis, or pretest norms),
    • only accessing largely conscious and introspectively available information (believing what they want to believe) which often is not the cause of behaviour.


Correlational Research | Empirical

  • correlational research: measuring the degree to association between two variables.
  • direction: positive correlation and negative correlation
  • zero correlation is when changes in one variable are not predictive of changes in the other and vice versa.
  • strength of association: how closely are changes in one variable linked to changes in the other variable?
  • correlation coefficient: +1 very strong positive correlation, -1 very strong negative correlation.
  • correlation is not causation! we don’t know if A causes B or B causes A or if they’re both caused by a third thing (the third variable problem).


Manipulating Variables | Empirical

  • experimental research is how we control variables in order to test for causation; causation affords explanation.
  • causation and the method of differences (lecture 1); independent variable and dependent variable.
  • main problem is how do you manipulate just one variable?
  • e.g. does sleep deprivation cause memory failure?
  • could be unfamiliar environment, so you want to make sure that an alternative explanation is ruled out.
  • this is controlling for a confounding variable.
  • placebo effect is the classic example of a confounding variable.


Between-Subject Design | Empirical

  • between-subject design; there is the experimental group and the control group.
  • a lot of what is published should be held in question because a lot of the time there is no relevant control group to compare the results to.
    • e.g. sleep deprivation experiment.
    • have two groups, one group in the lab getting sleep and then being tested for memory failure, and the other group also in the lab but also sleep deprived and then test for memory failure.
    • so the only relevant difference between them is the amount of sleep.
    • if there are differences in the memory tests it is plausible (not certain) that the difference is due to sleep differences.
  • random assignment and other sources of variability evenly spread across conditions.
  • you have to do a lot to make sure that it’s not placebo.


With-in Subjects Design | Empirical

  • within-subjects design: the same participants experience all types of stimuli or experimental conditions.
  • e.g. in the lab with and without sleep.
  • here, the order of conditions needs to be randomly assigned so that you are testing the independent variable and not the change between the experimental condition and the control condition.


Problems with the Experimental Method

  • main problem with experimental method is that as we control variables we make the laboratory situation more and more artificial and we lose ecological validity (inescapable trade-off).
  • important role of theoretical argumentation and establishing the plausibility of the interpretation of the experimental data, i.e. arguing for the relevance of the differences between the experimental situation and the real world.
  • the degree to which the experimental result affords consistent and powerful intervention in the real world provides evidence for ecological validity.



  • dealing with sampling bias is the main job of statistics.
  • random assignment is important but there’s important information in the sample that can be used to assess how accurate it represents the population.
  • information about distribution, central tendency, and variability are important for determining this.
  • distribution of the data; two pieces of information are whether some scores occurred more often than others and whether all scores were more clumped in the middle or more evenly spread across the whole range.
  • normal distribution, negatively skewed distribution, and positively skewed distribution.
    • generally, results are better when you have a normal distribution (that’s why it’s called normal)
  • central tendency, mode, mean, and median.
  • variability, standard deviation, hypothesis testing.
  • usually differences between the experimental group and the control group.
  • trying to show that the most likely explanation for differences is due to your manipulation of the variable.
  • problem arises when there is overlap between the sets of results for both groups, this is the signal to the noise problem talked about in lecture 1.
  • overlap means those results are ambiguous, i.e. either the point to the role of the independent variable or point to an alternative variable (possible confound) that the experimenter tried/failed to control.
    • i.e. ambiguity means that an alternative explanation has not been ruled out; no clear information.
  • scientists use statistical significance to address the problem of overlap/ambiguity/no clear information.
    • use a measure of central tendency (usually the mean) and a measure of variability (usually standard deviation) to calculate how much the difference could be due to chance.
    • 5% rule—if there’s only a <5% that it could’ve been caused by chance, then it’s probably not caused by chance; if it’s 10% then it may have been chance.
  • one problem with statistics: multiple comparison means that the more likely you will luck out and get a fluke result because 5% of the difference will be produced by chance.
    • yet the experimenter will conclude that it was due to the manipulation of the independent variable.
    • one response is to raise the standard as you increase the number of times you make the comparison, e.g. only accept answers where the chances of the difference being produced by luck are 1% etc.
  • second problem: as you increase the number of participants, then small differences between groups become more and more statistically significant.
    • an alternative is power analysis: you do not just answer the yes-no question of whether or not the difference is significant; you also answer the question as to how large the difference is—how much of a difference does the manipulation of the independent variable actually make.


The Biology of the Brain

  • resting potential, the electrostatic gradient, and the concentration gradient
  • ions and movement across the membrane
  • the action potential and how it travels down the axon
  • the refractory period
  • fire together wire together principle—not enough to explain how neural networks learn, but is a foundational principle
  • patterns of activity are connected to patterns of connectivity
  • how our brain is connected affects how it fires, and how our brain fires affects how it wires
  • layers of learning: brain is modeling itself as it models the world
  • the brain isn’t just interested in predicting the world, but predicting the other parts of your brain
  • bottom up processing and top down processing
    • e.g. I have to read the word to read the letters, but I have to read the letters to read the word, so reading is impossible
    • the brain is dynamically interacting with itself in a layered fashion
    • 1988; one of the first computers was made that could read text
    • we’re now getting AI that can understand spoken text
  • patterns of patterns
  • why is the brain wired in such a complex way?
  • your brain is not just an electrochemical processing, but a gland that’s responding to tons of chemicals that are diffusing through the body and the brain
  • the full array of neurotransmitters and their functions is extremely complex, with many variations of NTs, many functions, many interactions and different scales
  • Siegel’s brain in the hand
    • your fist is the brain and your wrist is the brain stem
  • cerebellum has a lot to do with fine tuning the parts of your cortex
    • looks for complex patterns and looks for ways to process those patterns
    • you have more neurons in your cerebellum than cortex
    • being exacted to help the frontal cortex with balance
    • kind of like a second brain that helps model the primary brain
  • evolutionary direction of function and morphology (back to front, inside to outside)
  • brainstem/hindbrain, sub-cortical regions, cerebellum, 4 lobes of the cortex, pre-frontal cortex

Decks in 🚫 PSY100H1: Introduction to Psychology (Winter 2016) with J. Vervaeke Class (50):