INTRODUCTION TO EXPERIMENTAL PSYCHOLOGY.ppt

1、Introduction to Experimental Psychology,Psychology 220,Chapter 1: Basics of Scientific Psychology,Goals of Psychological Research Make you think like a scientist about behavior Better understand how to conduct research This allows use to test hypotheses This allows us to solve practical problems Bet

2、ter understand how to evaluate research This involves learning how apply critical thinking to research As a critic, you learn how to make an informed judgment about the value of something This critique can result in either support or criticism of a theory or hypothesis,Chapter 1 continued:,We make i

3、nformed decisions in science using Experiments In an experiment, we test a hypothesis (a very specific testable idea based upon a theorywhich is a collection of ideas whose purpose is to describe, predict, or explain phenomena) In an experiment we manipulate the independent variable (IV) while holdi

4、ng other potential IVs constant (control variables, or CVs), and then we examine the effect of the IV(s) on the dependent variable (DV)the performance variable If the proper experimental design is not followed, then our results can be confoundede.g., fatigue or practice effects in a within-subjects

5、design,Chapter 1 continued:,Example of a theory: Cognitive aging is primarily a function of the brain slowing down as a function of task complexity This is the “general slowing” theory of cognitive aging (e.g., Cerella, 1985; Salthouse, 1996) An example of a hypothesis is that age differences will b

6、e proportional to task complexity We can test this hypothesis using an experiment in which we vary word frequency (high vs. low), case type (lowercase vs. mixed-case), and response type (two-choice vs. go/no-go) on a lexical decision task (does a letter string form a real word, or not?) testing youn

7、ger and older adults What are the IVs, CVs, and DVs? After collecting our data, we test whether age differences in word frequency and case type are consistent with processing speed. It turns out that there are no appreciable age differences in word frequency, but there are large differences in case

8、type and response type, even though older adults are much slowerthese results are inconsistent with general slowing (see Allen et al., 1993),Chapter 1: Continued,This type of experiment is designed to isolate age effects at one, or multiple, information processing stages We cannot directly observe t

9、hinking, so we indirectly infer it based upon the relation between stimulus conditions and subject behavior This allows us to peer inside the “black box” of the human mind The mind is like trying to see what is present outside after dark when you are sitting in a lighted room This is systems enginee

10、ring of the mind,Chapter 1: Continued,In our design, we are manipulating case type (encoding), word frequency (lexical access), and response type (response selection) We assume that humans go through sequential, functionally discrete processing stages i.e., the tend to finish a stage before they beg

11、in another stage Using this approach, we can determine whether age differences are isolated at a given stage or whether they are generalized across stages,Chapter 1 continued:,These results are important because they suggest that we do not simply “go down hill” as we age Indeed, in some situations,

12、older adults actually perform better than younger adults (e.g., Allen et al., 2002; Lien et al., 2006) So it appears that aging is a combination of neural degeneration (that hurts performance) and skill acquisition (that helps performance),Chapter 1 continued:,Sources of Research IdeasWhere do resea

13、rch ideas come from? Coming up with interesting, testable research ideas is probably the most difficult part of science (including psychology). Scientists must be creative, and they need to practice this skill Observation is a good place to begin generating research ideas (this allows you to observe

14、 what the important pieces of the puzzle are, and how they fit together) New scientific ideas need to extend the existing literature Conducting a literature search allows you to confirm that your idea is novel so that you do not steal others ideas (plagiarism). Additionally, others work can give you

15、 good research ideas PsycINFO is an effective search engine (although Google works well in many circumstances, as well),Chapter 1 continued:,Typical Steps in Research: Develop an idea (this is very difficult and takes creativity as well as a comprehensive knowledge of the field) Formulate a testable

16、 hypothesis (make a statement about a presumed or theoretical relation between two or more variables) This implies that these variables are measureable It also specifies the relations among the variables Reviewing the literaturethis term we will examine Time-of-Day effects on cognition Conduct pilot

17、 research (to see if everything works) Complete your data collection (with a large enough sample size) Conduct statistical analyses (statistics is the language of science) Interpret your results Write up your results in an article,Chapter 1 Continued:,Present Project: Age Differences in Naming Theor

18、y: Cognitive aging effects are a combination of cognitive and sensory/motor loss and skill increases Hypothesis: Loss of visual acuity should make mixed-case presentation stimuli particularly difficult for older adults (compared to younger adults), but phonological regularity effects should be relat

19、ively constant across age Method: have two sets of participants (younger and older adults) tested on a naming task that varies word frequency, phonological regularity, and case type,Chapter 1 continued:,Research Pitfalls to avoid: Breaches of ethical practice (Chapter 12) (falsification, fabrication

20、, plagiarism) The “golden rule” is not to do anything to a research subject that you would not do to yourself This is a VERY complicated issue and we will discuss it in detail (e.g., non-human animal research, human research, research on impoverished individuals who participate just for the money) B

21、iased research Inadvertent research biase.g., if your political beliefs resulted in you interpreting results in a manner that went way beyond the data E.g., NPR caller that thought John McCain was too old to run for office and was showing signs of either MCI or dementia Morton, skull size, intellige

22、nce and race (and Goulds clarification) Double-blind design (both the researcher and the participant are blind to the treatment type) Avoiding anthropomorphizing with non-human animals Unreliable communication The cold war Science is inherently empirical, and it is important to have all the availabl

23、e data in order to make our best conclusion about what theories work best,Chapter 2: Explanation in Scientific Psychology,Making Sense of the Worldour attempt to discover how and why things work the way they do Social Loafingpeople working in a group do not work as hard as the same people working by

24、 themselves Why does social loafing occur?,Chapter 2: continued:,On March 13, 1964, Kitty Genovese was attacked and stabbed by Winston Moseley Mr. Moseley left for 15-60 minutes, and then returned to finish murdering Kitty Genovese There were approximately 38 witnesses to one or both of the stabbing

25、s, and no one called the police or went out to help Kitty Genovese This phenomena in which groups of people seem more unwilling to help people in need has been referred to as the “unresponsive bystander” effect,Chapter 2 continued:,The unresponsive bystander effect (or social loafing) has been expla

26、ined by the concept of “diffusion of responsibility”people working by themselves think that they are responsible for completing a task (or helping someone), whereas people working in a group diffuse the responsibility for work to the group as a whole (e.g., it is societys responsibility) Mention Bib

27、 Latanes “smoke study” in Ohio Stadium While there is some evidence that diffusion of responsibility can be attenuated (or even eliminated) by observing individual behavior in when someone is working in a group, this is still a powerful effect that can have an impact on work productivity and helping

28、 behaviors for people in need This has implications for whether “team” approaches to work really can work as well as an approach that emphasizes individual responsibility,Chapter 2 continued:,Sources of Understanding Fixation of Beliefhow we decide what we believe: The American philosopher Charles S

29、anders Peirce (1877) discussed four types of fixation of belief: Method of Authoritytake someone elses word on faith for deciding what you believe (e.g., a priests or ministers view) Method of Tenacitywhen a person steadfastly refuses to alter acquired beliefs in spite of evidence to the contrary (e

30、.g., the earth is flat) A Priori MethodWhen something is believed without prior study or examination (an extension of the Method of Authority, except now, it is not based on a particular authority, but on a general cultural outlook Scientific Methodfixes belief on the basis of experience and evidenc

31、e Science is repeatable and self-correcting (e.g., when results are not replicated consistently, we likely had a “false positive”) Empirical observation and manipulation (with experimental control) means that in science, fixation of belief is based at least in part on actual data By self-correcting,

32、 we mean that science offers methods for establishing the superiority of one belief over another,Chapter 2 continued:,Nature of Scientific Explanation Sir Francis Bacon (1561-1626) is credited with developing much of the conceptual basis of empiricismthe empirical and self-correcting aspects of scie

33、nce All accepted approaches to science share the same basic elementsdata (empirical observations collected under experimental control) and theory (organization of concepts that permit prediction and explanation of data) Induction is working from data to theory (reasoning from specific to general), a

34、nd deduction is working from theory to data (reasoning from general to specific) It is important to note that induction is probabilistic and that deduction is completely deterministic In a very real sense, there is no need for data in deduction (logic is used instead Laws in science almost involve d

35、eductive logic, but even laws do not always holdparticularly in psychology!,Chapter 2 continued:,Nature of Scientific Explanation Continued: Falsifiabilitybecause of the temporary, probabilistic nature of science, Popper (1961) has argued that negative evidence is more important than positive eviden

36、ce, and that one can disprove a theory but cannot prove it *This is the null hypothesis testing approach to science, but your professor heartily disagrees (people do not win Nobel Prizes for disproving a theoryit is usually for providing evidence in favor of a theory) Direct hypothesis testing (why

37、you did not remember some of what you learned in Statistics) This is why we need to have better quantitative skills is psychology Theory-Deduction-Data-Induction-Theory Strong Inferenceeliminating possible alternative explanations by pitting two (or more) possible explanations (theories) against eac

38、h other in a series of experiments in which each theory makes different predictions We will discuss converging operations later on,Chapter 2 continued:,Nature of Scientific Explanation Continued: From Theory to Hypothesistheories typically cannot be tested directly because they are too broad Instead

39、 theories generate very specific testable statements (hypotheses) that can be evaluated based upon observable data from an experiment Generalizationa broader statement (than a hypothesis) that cannot be tested directly but tends to come in between a theory (even more general) and hypotheses (more sp

40、ecific) E.g., Older adults slow down compared to younger adults (theory), older adults are unsafe drivers at any speed (generalization), drivers older than 65 have a higher incidence of left-turn accidents across oncoming traffic than do younger adults (hypothesis),Chapter 2 continued:,Nature of Sci

41、entific Explanation Continued: What is a theory?it is a set of related statements that explains a variety of occurrences. The more the occurrences and the fewer the statements, the better the theory (at least, in terms of parsimony) A theory in psychology performs at least three functions: organizat

42、ion (a framework for the systematic and orderly display of data), prediction (it allows the scientist to generate predictions for situations in which no data have been obtained), and explanation (it allows the scientist to interpret results) Explanation cannot occur directly from predictionit can oc

43、cur only on the basis of an experiment (sometimes several) Inferring causality requires not only relating two or more variables, but also maintaining experimental control to rule out alternative causes,Chapter 2 continued:,Nature of Scientific Explanation Continued: Intervening variablesthese link i

44、ndependent variables (IVs) to dependent variables (DVs) even though we cannot directly observe them (e.g., perception and attention occur between a stimulus and a responsewe can directly observe the latter two, but not the former two, although the intervening variables are required for a complete ex

45、planation Cognitive Psychology is based largely on unobservable intervening variables”systems engineering of the mind” (see Garner, Hake, and Erickson, 1956),Chapter 2 continued:,Evaluating theories: Parsimonythe simplest theory is the best (Occams razor) Precisiontheories based upon math equations

46、or computer programs have many advantages because they are more precise than verbal statements Precision has advantages over parsimony in mature sciences because it allows you to account for more results Testabilityeven some precise theories cannot be tested, and this means they are not scientific t

47、heories Testability is the hallmark of science,Chapter 2 continued:,The Science of Psychology: hard and soft science Applied versus basic research The importance of realism You must have internal validity to do science (ecological validity is not required, but external validity is),Chapter 3: Explor

48、ing the Literature in Psychology,How to do a literature search: Psychological Abstractscontains brief abstracts of articles (150 words or less) pertinent to psychology PsycINFOan electronic reference service that covers more than 1300 journals in psychology and related fields (e.g., neuroscience) Go

49、ogleis a useful search engine for your beginning search on a topic Social Sciences Citation Indexcan be used to search for articles or to determine how many times and an article has been cited (this is used to evaluate scholarshipe.g., for promotion processes) Academic Search Complete (EBSCO)a very

50、broad electronic reference service for all of science (4,400 peer-reviewed journals) Use Advanced Search, though,Chapter 3 continued:,Parts of an article: Title: Provides an idea of what the article is about Authors: knowing the authors past theoretical orientation can frequently provide you with an

51、 idea of the approach that will be taken in this article Abstract: “ A brief comprehensive summary of the contents of an article ” (p. 12, Publication manual of the APA). Introduction: Specifies the problem to be studied, some shortcomings in the present level of understanding, and how this study will attempt to overcome some of these shortcomings It also specifies the hypotheses to be tested and the rationale behind these predictions This is typically the most difficult portion of the paper to write,