Experimental psychology
Experimental psychology refers to work done by those who apply experimental methods to psychological study and the processes that underlie it. Experimental psychologists employ human participants and animal subjects to study a great many topics, including sensation & perception, memory, cognition, learning, motivation, emotion; developmental processes, social psychology, and the neural substrates of all of these.
History
Early experimental psychology
Wilhelm Wundt
Experimental psychology emerged as a modern academic discipline in the 19th century when Wilhelm Wundt introduced a mathematical and experimental approach to the field. Wundt founded the first psychology laboratory in Leipzig, Germany. Other experimental psychologists, including Hermann Ebbinghaus and Edward Titchener, included introspection among their experimental methods.Charles Bell
Charles Bell was a British physiologist, whose main contribution was research involving the nervous system. He wrote a pamphlet summarizing his research on rabbits. His research concluded that sensory nerves enter at the posterior roots of the spinal cord and motor nerves emerge from the anterior roots of the spinal cord. Eleven years later, a French physiologist Francois Magendie published the same findings without being aware of Bell's research. Due to Bell not publishing his research, this discovery was called the Bell-Magendie law. Bell's discovery disproved the belief that nerves transmitted either vibrations or spirits.Ernst Heinrich Weber
Weber was a German physician who is credited with being one of the founders of experimental psychology. Weber's main interests were the sense of touch and kinesthesis. His most memorable contribution to the field of experimental psychology is the suggestion that judgments of sensory differences are relative and not absolute. This relativity is expressed in "Weber's Law," which suggests that the just-noticeable difference, or jnd is a constant proportion of the ongoing stimulus level. Weber's Law is stated as an equation:where is the original intensity of stimulation, is the addition to it required for the difference to be perceived, and k is a constant. Thus, for k to remain constant, must rise as I increases. Weber's law is considered to be the first quantitative law in the history of psychology.
Gustav Fechner
Fechner published in 1860 what is considered to be the first work of experimental psychology, "Elemente der Psychophysik." Some historians date the beginning of experimental psychology from the publication of "Elemente." Weber was not a psychologist, and it was Fechner who realized the importance of Weber's research to psychology. Fechner was profoundly interested in establishing a scientific study of the mind-body relationship, which became known as psychophysics. Much of Fechner's research focused on the measurement of psychophysical thresholds and just-noticeable differences, and he invented the psychophysical method of limits, the method of constant stimuli, and the method of adjustment, which are still in use.Oswald Külpe
Oswald Külpe is the main founder of the Würzburg School in Germany. He was a pupil of Wilhelm Wundt for about twelve years. Unlike Wundt, Külpe believed experiments were possible to test higher mental processes. In 1883 he wrote Grundriss der Psychologie, which had strictly scientific facts and no mention of thought. The lack of thought in his book is odd because the Würzburg School put a lot of emphasis on mental set and imageless thought.Würzburg School
The work of the Würzburg School was a milestone in the development of experimental psychology. The School was founded by a group of psychologists led by Oswald Külpe, and it provided an alternative to the structuralism of Edward Titchener and Wilhelm Wundt. Those in the School focused mainly on mental operations such as mental set and imageless thought. Mental set affects perception and problem solving without the awareness of the individual; it can be triggered by instructions or by experience. Similarly, according to Külpe, imageless thought consists of pure mental acts that do not involve mental images. An example of mental set was provided by William Bryan, an American student working in Külpe's laboratory. Bryan presented subjects with cards that had nonsense syllables written on them in various colors. The subjects were told to attend to the syllables, and in consequence they did not remember the colors of the nonsense syllables. Such results made people question the validity of introspection as a research tool, and led to a decline of voluntarism and structuralism. The work of the Würzburg School later influenced many Gestalt psychologists, including Max Wertheimer.George Trumbull Ladd
Experimental psychology was introduced into the United States by George Trumbull Ladd, who founded Yale University's psychological laboratory in 1879. In 1887, Ladd published Elements of Physiological Psychology, the first American textbook that extensively discussed experimental psychology. Between Ladd's founding of the Yale Laboratory and his textbook, the center of experimental psychology in the US shifted to Johns Hopkins University, where George Hall and Charles Sanders Peirce were extending and qualifying Wundt's work.Charles Sanders Peirce
With his student Joseph Jastrow, Charles S. Peirce randomly assigned volunteers to a blinded, repeated-measures design to evaluate their ability to discriminate weights. Peirce's experiment inspired other researchers in psychology and education, which developed a research tradition of randomized experiments in laboratories and specialized textbooks in the 1800s. The Peirce–Jastrow experiments were conducted as part of Peirce's pragmatic program to understand human perception; other studies considered perception of light, etc. While Peirce was making advances in experimental psychology and psychophysics, he was also developing a theory of statistical inference, which was published in "Illustrations of the Logic of Science" and "A Theory of Probable Inference" ; both publications that emphasized the importance of randomization-based inference in statistics. To Peirce and to experimental psychology belongs the honor of having invented randomized experiments, decades before the innovations of Jerzy Neyman and Ronald Fisher in agriculture.Peirce's pragmaticist philosophy also included an extensive theory of mental representations and cognition, which he studied under the name of semiotics. Peirce's student Joseph Jastrow continued to conduct randomized experiments throughout his distinguished career in experimental psychology, much of which would later be recognized as cognitive psychology. There has been a resurgence of interest in Peirce's work in cognitive psychology. Another student of Peirce, John Dewey, conducted experiments on human cognition, particularly in schools, as part of his "experimental logic" and "public philosophy."
20th century
In the middle of the 20th century, behaviorism became a dominant paradigm within psychology, especially in the United States. This led to some neglect of mental phenomena within experimental psychology. In Europe this was less the case, as European psychology was influenced by psychologists such as Sir Frederic Bartlett, Kenneth Craik, W.E. Hick and Donald Broadbent, who focused on topics such as thinking, memory and attention. This laid the foundations for the subsequent development of cognitive psychology.In the latter half of the 20th century, the phrase "experimental psychology" had shifted in meaning due to the expansion of psychology as a discipline and the growth in the size and number of its sub-disciplines. Experimental psychologists use a range of methods and do not confine themselves to a strictly experimental approach, partly because developments in the philosophy of science have affected the exclusive prestige of experimentation. In contrast, an experimental method is now widely used in fields such as developmental and social psychology, which were not previously part of experimental psychology. The phrase continues in use, however, in the titles of a number of well-established, high prestige learned societies and scientific journals, as well as some university courses of study in psychology.
Methodology
Sound methodology is essential to the study of complex behavioral and mental processes, and this implies, especially, the careful definition and control of experimental variables.Assumptions
Empiricism
Perhaps the most basic assumption of science is that factual statements about the world must ultimately be based on observations of the world. This notion of empiricism requires that hypotheses and theories be tested against observations of the natural world rather than on a priori reasoning, intuition, or revelation.Testability
Closely related to empiricism is the idea that, to be useful, a scientific law or theory must be testable with available research methods. If a theory cannot be tested in any conceivable way then many scientists consider the theory to be meaningless. Testability implies falsifiability, which is the idea that some set of observations could prove the theory to be incorrect. Testability has been emphasized in psychology because influential or well-known theories like those of Freud have been difficult to test.Determinism
Experimental psychologists, like most scientists, accept the notion of determinism. This is the assumption that any state of an object or event is determined by prior states. In other words, behavioral or mental phenomena are typically stated in terms of cause and effect. If a phenomenon is sufficiently general and widely confirmed, it may be called a "law"; psychological theories serve to organize and integrate laws.Parsimony
Another guiding idea of science is parsimony, the search for simplicity. For example, most scientists agree that if two theories handle a set of empirical observations equally well, we should prefer the simpler or more parsimonious of the two. A notable early argument for parsimony was stated by the medieval English philosopher William of Occam, and for this reason the principle of parsimony is often referred to as Occam's razor.Operational definition
Some well-known behaviorists such as Edward C. Tolman and Clark Hull popularized the idea of operationism, or operational definition. Operational definition implies that a concept be defined in terms of concrete, observable procedures. Experimental psychologists attempt to define currently unobservable phenomena, such as mental events, by connecting them to observations by chains of reasoning.Experiments
In experiments, human participants often respond to visual, auditory or other stimuli, following instructions given by an experimenter; animals may be similarly "instructed" by rewarding appropriate responses. Since the 1990s, computers have commonly been used to automate stimulus presentation and behavioral measurement in the laboratory. Behavioral experiments with both humans and animals typically measure reaction time, choices among two or more alternatives, and/or response rate or strength; they may also record movements, facial expressions, or other behaviors. Experiments with humans may also obtain written responses before, during, and after experimental procedures. Psychophysiological experiments, on the other hand, measure brain or single-cell activation during the presentation of a stimulus using methods such as fMRI, EEG, PET or similar.Control of extraneous variables, minimizing the potential for experimenter bias, counterbalancing the order of experimental tasks, adequate sample size, the use of operational definitions, emphasis on both the reliability and validity of results, and proper statistical analysis are central to experimental methods in psychology. Because an understanding of these matters is important to the interpretation of data in almost all fields of psychology, undergraduate programs in psychology usually include mandatory courses in research methods and statistics.
A crucial experiment is an experiment that is intended to test several hypotheses at the same time. Ideally, one hypothesis may be confirmed and all the others rejected. However, the data may also be consistent with several hypotheses, a result that calls for further research to narrow down the possibilities.
A pilot study may be run before a major experiment, in order to try out different procedures, determine optimal values of the experimental variables, or uncover weaknesses in experimental design. The pilot study may not be an experiment as usually defined; it might, for example, consist simply of self-reports.
In a field experiment, participants are observed in a naturalistic setting outside the laboratory. Field experiments differ from field studies in that some part of the environment is manipulated in a controlled way. Control is typically more lax than it would be in a laboratory setting.
Other methods of research such as case study, interview, opinion polls and naturalistic observation, are often used by psychologists. These are not experimental methods, as they lack such aspects as well-defined, controlled variables, randomization, and isolation from unwanted variables.
Reliability and Validity
Reliability
Reliability measures the consistency or repeatability of an observation. For example, one way to assess reliability is the "test-retest" method, done by measuring a group of participants at one time and then testing them a second time to see if the results are consistent. Because the first test itself may alter the results of a second test, other methods are often used. For example, in the "split-half" measure, a group of participants is divided at random into two comparable sub-groups, and reliability is measured by comparing the test results from these groups, It is important to note that a reliable measure need not yield a valid conclusion.Validity
Validity measures the relative accuracy or correctness of conclusions drawn from a study. To determine the validity of a measurement quantitatively, it must be compared with a criterion. For example, to determine the validity of a test of academic ability, that test might be given to a group of students and the results correlated with the grade-point averages of the individuals in that group. As this example suggests, there is often controversy in the selection of appropriate criteria for a given measure. In addition, a conclusion can only be valid to the extent that the observations upon which it is based are reliable.Several types of validity have been distinguished, as follows:
Internal validity
refers to the extent to which a set of research findings provides compelling information about causality. High internal validity implies that the experimental design of a study excludes extraneous influences, such that one can confidently conclude that variations in the independent variable caused any observed changes in the dependent variable.External validity
refers to the extent to which the outcome of an experiment can be generalized to apply to other situations than those of the experiment - for example, to other people, other physical or social environments, or even other cultures.Construct validity
Construct validity refers to the extent to which the independent and dependent variables in a study represent the abstract hypothetical variables of interest. In other words, it has to do with whether the manipulated and/or measured variables in a study accurately reflect the variables the researcher hoped to manipulate. Construct validity also reflects the quality of one's operational definitions. If a researcher has done a good job of converting the abstract to the observable, construct validity is high.Conceptual validity
Conceptual validity refers to how well specific research maps onto the broader theory that it was designed to test. Conceptual and construct validity have a lot in common, but conceptual validity relates a study to broad theoretical issues whereas construct validity has more to do with specific manipulations and measures.Scales of measurement
Measurement can be defined as "the assignment of numerals to objects or events according to rules."Almost all psychological experiments involve some sort of measurement, if only to determine the reliability and validity of results, and of course measurement is essential if results are to be relevant to quantitative theories.
The rule for assigning numbers to a property of an object or event is called a "scale". Following are the basic scales used in psychological measurement.
Nominal measurement
In a nominal scale, numbers are used simply as labels – a letter or name would do as well. Examples are the numbers on the shirts of football or baseball players. The labels are more useful if the same label can be given to more than one thing, meaning that the things are equal in some way, and can be classified together.Ordinal measurement
An ordinal scale arises from the ordering or ranking objects, so that A is greater than B, B is greater than C, and so on. Many psychological experiments yield numbers of this sort; for example, a participant might be able to rank odors such that A is more pleasant than B, and B is more pleasant than C, but these rankings would not tell by how much each odor differed from another. Some statistics can be computed from ordinal measures – for example, median, percentile, and order correlation – but others, such as standard deviation, cannot properly be used.Interval measurement
An interval scale is constructed by determining the equality of differences between the things measured. That is, numbers form an interval scale when the differences between the numbers correspond to differences between the properties measured. For instance, one can say that the difference between 5 and 10 degrees on a Fahrenheit thermometer equals the difference between 25 and 30, but it is meaningless to say that something with a temperature of 20 degrees Fahrenheit is "twice as hot" as something with a temperature of 10 degrees. "Standard scores" on an achievement test are said to be measurements on an interval scale, but this is difficult to prove.Ratio measurement
A ratio scale is constructed by determining the equality of ratios. For example, if, on a balance instrument, object A balances two identical objects B, then one can say that A is twice as heavy as B and can give them appropriate numbers, for example "A weighs 2 grams" and "B weighs 1 gram". A key idea is that such ratios remain the same regardless of the scale units used; for example, the ratio of A to B remains the same whether grams or ounces are used. Length, resistance, and Kelvin temperature are other things that can be measured on ratio scales. Some psychological properties such as the loudness of a sound can be measured on a ratio scale.Research design
One-way designs
The simplest experimental design is a one-way design, in which there is only one independent variable. The simplest kind of one-way design involves just two-groups, each of which receives one value of the independent variable. A two-group design typically consists of an experimental group and a control group.The one-way design may be expanded to a one-way, multiple groups design. Here a single independent variable takes on three or more levels. This type of design is particularly useful because it can help to outline a functional relationship between the independent and dependent variables.