Textbook Notes (368,125)
Canada (161,663)
Psychology (3,337)
PSYC 2650 (228)

PSYC*2650 Ch 13.doc

13 Pages
Unlock Document

PSYC 2650
Anneke Olthof

Chapter 13: Reasoning: Thinking Through the Implications of What You Know - often we draw conclusions from claims that seem solidly established - this relies on Deduction - a process through which we start with claims or assertions that we could as “given” and ask what follows from these premises - deduction allows us to make predictions about upcoming events - the same predictions can keep our beliefs in touch with reality - if deduction leads to a prediction based on our beliefs, and the prediction turns out to be wrong, this tells us that something is off track in our beliefs, and that things we thought to be solidly established aren’t so solid after all Confirmation and Disconfirmation - each of us regards that many of our beliefs are solidly established e.g. water is wet - we are far less certain about some other beliefs e.g. would he make a good secretary? - how do we proceed in cases where we are uncertain? - there are concerns with the ambiguity of confirming evidence and the great value of disconfirming evidence Confirmation Bias - in order to evaluate a belief, it’s often useful to seek out both evidence that supports that belief and also evidence that might challenge the belief - despite the usefulness of disinformation, people generally don’t do it Confirmation Bias - a strong tendency to seek out confirming evidence and to rely on that evidence in drawing their conclusions - first, when people are assessing a belief or hypothesis, they are far more likely to seek evidence that confirms the belief than evidence that might disconfirm it - second, when disconfirming evidence is made available to them, people often fail to use it in adjusting their beliefs - third, when people encounter confirming evidence, they take it at face value but when they encounter disconfirming evidence, they reinterpret the evidence to diminish its im- pact - people regularly fail to consider alternative hypotheses that might explain the available data just as well as their current hypothesis does Memory for Disconfirming Evidence - since people scrutinize disconfirming evidence, this mental activity sometimes makes people more likely to remember disconfirming evidence - e.g. in a study people who had places bets on a football game believed they had good strategies for picking winning teams - their faith was undiminished by a series of losses - gamblers don’t remember their losses as “losses”, they remember them as mere flukes or oddball coincidences - winning bets were remembered as wins, but losing bets were remembered as “near wins” Belief Perseverance - a phenomenon where even when disconfirming evidence is undeniable and out in plain view but some people do not use it - in a study, participants were asked to read a series of suicide notes and asked to find out which ones were authentic - as they offered their judgments, they were provided feedback - the feedback was predetermined and had nothing to do with their judgments - later participants were debriefed and told that the feedback they received was bogus - they were then asked additional questions where they had to assess their own “social sensitivity” and their actual ability on the previous task - participants were influenced by the feedback - those who received the above average feedback continued to think of their social sensitivity as being above average - their beliefs persevered even when the basis for the belief had been completely dis- credited - this has an important implication - maybe debriefing is not effective Logic Reasoning About Syllogisms - a number of theorists have proposed that thought does follow the rules of logic - if we make reasoning errors, it is not because of some flaw in our thinking - instead, the errors must come from other sources e.g. carelessness, misreading - errors in logical reasoning are ubiquitous - if we are careless or misread problems, we do so with great frequency - this is shown in studies of Categorical Syllogisms - a type of logical argument that begins with 2 assertions (the problem’s premises) each containing a statement about a category - the syllogism can be completed with conclusions that follow these premises - e.g. all plumbers are mortal, all sadists are mortal, but all sadists are not plumbers - participants who are asked to reason about syllogisms do remarkably poorly Sources of Logical Errors - errors in logical reasoning are extremely common - errors don’t look at all like the product of carelessness, they tend to fall into a few sim- ple categories Belief Bias - if a syllogism’s conclusion happens to be something people believe to be true anyway, they are more likely to judge the conclusion as following logically from the premises - if the conclusion happens to be something they believe to be false, they are likely to re- ject the conclusion as invalid - logic is concerned with moral “local” issues of reasoning, specifically whether a partic- ular conclusion is warranted by a particular set of premises - when people show belief-bias, they are failing to distinguish between good arguments and bad ones - other logical errors seem to be the result of people relying on a low-level “matching strategy” - endorsing conclusions if the words “match” those in the premise - e.g. if a participants sees a premise such as “all A are B” and then another premise such as “all D are B” the participant is likely to accept a conclusion like “all A are D be- cause this conclusion “matches” the wording and structure of the premises - this “matching strategy” is illogical and will lead to errors Reasoning About Conditional Statements Conditional Statements - statements of the familiar “if X then Y” format, with the first statement providing a condition under which the second statement it guaranteed to be true - the rules of logic specify how one should reason about conditional statements - one logical rule, called Modus Ponens justifies conclusions in this case: If P is true, then Q is true. P is true. Therefore, Q must be true. - another more difficult rule of Modus Tollens which justified the conclusion in this case: If P is true, then Q is true. Q is false. Therefore, P must be false. - other logical errors are Affirming the Consequent (modus ponens) and Denying the Antecedent (modus tollens) - because of these, performance tends to be rather poor when people are asked to rea- son about conditional statements - errors are more common if people are asked to reason about abstract problems in comparison to performance with concrete problems - errors are also more common if the logic problems involve negatives - belief bias can be demonstrated - people will endorse a conclusion is they believe it to be true even if the conclusion doesn’t follow from the stated premises The Four-Card Task - sometimes called the selection task - research participants are shown four playing cards and told that each card has a num- ber on one side and a latter on the other - their task is to evaluate this rule: “if a card has a vowel on one side, it must have an even number on the other side” Which cards must be turned over to put this rule to the test? - many people assert that an A card must be turned over, others assert that a 6 be, oth- ers say both - the correct answer was turning over the A and the 7 (only 4% got it right) - if we turn over the A card and find an even number, that consistent with the rule - if we turn over an odd number that’s inconsistent - if we turn over the 6 and find a vowel it fits with the rule, if you find a consonant this also fits since the rule makes no claims about what is on the reverse of a consonant - if we turn over the 7 and a consonant is on the other side, this fits with the rule, if there’s a vowel, it doesn’t fit The Effects of Problem Content - performance is much better with some variations of the four-card task - e.g. using the rule “if a participant is drinking beer, then the person must be over 19 years of age” - participants did quite well in this with 73% picking the correct cards of drinking a beer and 16 years of age - this data provides an argument against the proposal that logic describes the rules of thought - logic depends only on the form or syntax of the assertions being considered, and not at all on their content - they way we think and how well we think depends on what we’re thinking about - it de- pends on the meaning and pragmatics of the material we’re contemplating Detecting Cheaters - there are a variety of proposals for why some versions of the four-card problem are dif- ficult and others are easy, there is no clear choice among them - one proposal comes from an evolutionary perspective - our ancient ancestor’s didn’t have to reason about whether or not all A are B nor did they have to reason about vow- els and even numbers, they had to worry about issues like social interactions - an individual skills in this reasoning might have a survival advantages - the over 19 and drinking beer problem is consistent with this, it involves the detection of “cheaters” and yields good performance Pragmatic Reasoning Schemata - another hypothesis about the four-card problem is that thinking generally depends on pragmatic rules that people have learned from their day-to-day experience - e.g. people learn from experience that certain rules apply to any situations that in- volves permission - e.g. if one wishes to take a certain action, then one must have permission, if one has permission to take that action, the one may take the action - these pragmatic reasoning rules are more concrete than the rules of logic - people have pragmatic reasoning schemata for other sorts of situations, situations in- volving obligations, and cause-and-effect relationships - the original version of the four-card problem contained no practical or meaningful rela- tions, it is unlikely to evoke a reasoning schema - we should be able to improve performance on the four-card task by altering the prob- lem so that it will trigger a pragmatic reasoning schema - participants who are given a rational perform much better Necessity and Sufficiency - another approach to the four-card data begins with the fact that “if-then” sentences are actually ambiguous - e.g. if Jacob passed his driver’s test, then it’s legal for him to drive. If Solomon is eligi- ble for jury duty, then he is over 21. - these 2 sentences seem to have the same structure but they’re actually quite different - in the first sentence the “if” part identifies a Necessary Condition - something that must be true for the rest of the sentence to be true - in the second sentence, the “if” part identifies a Sufficient Condition - something that if true, guarantees the conclusion is true - Solomon might not be a US citizen so he couldn’t serve on a jury but that would tell us nothing about his age - a purely formal account of our reasoning - one that emphasizes only the logical form of the problems we face, and not the content of those problems - does not mesh with the way humans reason Deductive Logic - An Interim Summary - some people, in some circumstances, can use the rules of logic - more commonly people rely on less-formal reasoning strategies - there is debate what the other strategies are - these other reasoning strategies are sophisticated enough so that they generally yield conclusions identical to those endorsed by logical rules - our account of reasoning also needs a third layer, one describing the way people rea- son when these relatively sophisticated reasoning strategies are not triggered - the suggestion is that the rules of logic, some sort of content-specific reasoning strate- gies, and various shortcuts may all coexist in our mental repertoire and each will surface in appropriate circumstances - our theory requires multiple parts similar to judgment and reasoning - sophisticated thinking is possible even for people who have not been formally trained - this sophisticated thinking isn’t always used, and it seems to play a role only in the “right” circumstances - we find that better-quality reasoning is associated with higher levels of intellectual abili- ty, although even the smartest people are vulnerable to reasoning errors - different neural systems are involved in different specific tasks, with some neural sys- tems especially active when people are relying on heuristics, different systems active when people are relying on the detection and resolution of conflict etc - people reason in different ways on different occasions - both reasoning and judgment often rely on relatively primitive strategies when they don’t see how else to tackle the problem Mental Models - if we sometimes reason poorly, this is largely because we are using inappropriate strategies - one further reasoning strategy is creating and then reasoning about Mental Models - people may try to reason by thinking about concrete and specific cases - let’s return to categor
More Less

Related notes for PSYC 2650

Log In


Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Sign up

Join to view


By registering, I agree to the Terms and Privacy Policies
Already have an account?
Just a few more details

So we can recommend you notes for your school.

Reset Password

Please enter below the email address you registered with and we will send you a link to reset your password.

Add your courses

Get notes from the top students in your class.