chapter 2 - methods of cognitive psychology: behavior "can we use behavioral data to test cognitive theories"

we can’t observe a mental process, so how can we observe behavior in order to explain behavior?

measuring behavior, not mental processes is the answer and we can avoid the circulatory problem if we do two things:

1. create theories that are detailed enough to make specific predictions about what the pattern of data should look like if the theory is correct and what the data will look like if it is not correct

2. we must create experiments that carefully control and manipulate behavior; we need theories that generate specific predictions and experimental conditions that will test these predictions

can we use neuroscientific data to test cognitive theories?

more complicated than we would guess

we determine temporal love is more active when people read than when they watch a movie → cannot say we are able to read bc of the temporal lobe → cannot assume temporal lobe is the “reading area” (maybe its important for attention)

3 broad classes of behavioral research that contribute to the goal of learning how abstract constructs change behavior

1. descriptive research - describes the world

2. relational research - describes how different factors change together

3. experimental research - most often used when one factor is changed and the other is observed

3 properties of scientific method

1. empiricism

2. public verifiability

3. solvable problems

empiricism

means we are dedicated to learning about the world by trying things out; we develop a hypothesis about the world, and then we test it

→ one of three principles of scientific method

→ means we learn about the world by conducting experiments

public verifiability

means that we must make our hypotheses and our experiments available to everyone to examine and critique

→ one of three principles of scientific method

solvable problems

recognizes that science is well suited to studying some problems and ill suited to studying other problems

→ one of three principles of scientific method

2 methods to improve public verifiability in science

1. there need to be professional incentives to adhere to these values

2. there need to be tools that facilitate these values

behaviorists argued that

the cognitive approach was unscientific because it was not empirical; one couldn’t conduct studies to confirm or disconfirm cognitive theories

→ believed it failed the “solvable problems” criterion; one couldn’t use the scientific method to explain thought, they argued, any more than one could use it to explain morality

behaviorists had a point

unobservable mental processes are very powerful as theoretical devices because there is no limit to what they can do; the theorist is free to invent any process or representation they should

3 theories of search in short-term memory

1. search through list, one item at a time, halting when you find the item

   → will be faster when target is present than when it is not

2. search through all items in list even if you find the item

   → will take same amount of time when target is present and when it is not

3. don’t search one item at a time rather compare the target to all items simultaneously, searching through the entire list at once

   → number of items in short term memory will not affect how long the search will take

target

the item the participant is expected to find

cognitive psychologists solved this problem by

articulating not only their theories of mental processes, but also how different unobservable mental processes might interact with the observable world through behavior

→ they didn’t create theories of behavior but they specified the behavior that could be expected if their cognitive theory were right

steps to experimentally testing cognitive theories

1. develop alternative theories

   → our understanding if much deeper if we have at least two possibilities in mind and choose among them rather than having one model which we try to confirm or disconfirm

2. derive signature predictions for each theory

   → we can’t distinguish among different theories if they all make the same predictions

3. obtain data to compare the theories

descriptive research

a type of scientific research in which one seeks only to describe the world as it is, not to describe relationships among different entities in the world; it can be contrasted with relational research and experimental research

3 commonly used methods are descriptive

naturalistic observation, case studies, and self-report

→ never provide decisive data

→ US census

naturalistic observation

a method of collecting scientific data wherein the researcher observes behavior in its natural setting, for example, an ethologist observing a world in the wild

case study

a type of scientific research in which a single individual is observed on a number of occasions; usually used only when there is a rather unusual individual to be studied

→ victor (found in forest)

relational research

a type of scientific research in which one seeks to describe the relationship of two variables (e.g. income and intelligence) without specifying whether changes in one causes changes in the other

→ more likely to be useful to psychologists

→ not decisive; we know they are associated but not why they are associated

what relational research tells us/doesn’t tell us

tells you changes in Factor A (e.g. self esteem) tend to go with changes in Factor B (e.g. GPA), but it doesn’t tell you whether changes in A causes changes in B, or whether changes in B cause changes in A, or whether Factor C (e.g. home environment) causes changes in A and in B

experimental research

a type of scientific research in which the value of one variable (e.g. self esteem) is changed to observe its affect on another variable (e.g. memory ability); allows one to draw conclusions about causality, that is, that changes in one variable cause changes in another

independent variable

something that the researcher believes will influence cognition

great advantage of an experiment

we can be much more confident that we understand the relationship between two factors when they have been evaluated in an experiment rather than in a case study or relational research

2 objections to cognitive experiments

1. majority of them are conducted on college students who are young, wealthy, and of european descent

2. experiments don’t seem like real life

   → low in ecological validity

   → they might create strange little worlds and still be informative

WEIRD

western, educated, industrialized, rich, and democratic

→ recently realized culture can shape cognitive processes

muller-lyer illusion

the perceived length of a line is affected by the context of arrows around it

→ the degree to which people are susceptible to this illusion varies across cultures

ecological validity

the extent to which an experiment represents “the real world”; high ecological validity means that the conditions if the experiment seem similar to the ones that would be encountered in the real world

summary

• cognitive psychologists can use behavioral data to test cognitive theories

• to do so, they must specify how the unobservable abstract constructs in the theory are influenced by observable manipulations and express this influence in observable behavior

• psychologists use experiments to evaluate whether the predicted relationships between these variables hold true