experimental psychologyOverview

Psychology is beginning to understand the mechanisms of human thought and behaviour. The more literary language — that which employs metaphor, mental imagery, synaesthesia, etc. — seems not to be simply more picturesque, but to reflect actual modes of brain behaviour.


Psychology is the science of thought and behaviour. Experiments are set up so that the clearest and most significant generalizations are possible from the results. The work aims to be replicable, so that other researchers with different expectations and cultural backgrounds get the same results. Validity is equally important: the work must ensure that it is indeed measuring what it claims to measure. Control is therefore vital for experiments — either in a single-blind manner (subjects do not know the object of the work and so cannot selectively cooperate) or double-blind manner (object is not known to the actual experimenter, so that unconscious clues cannot be passed on). Very elaborate precautions are commonly taken to ensure that the setting is as naturalistic as possible, and that other factors do not unduly influence the result (differences of age, cultural background and family history). {1}

Fields of Psychology

Psychology has very diverse aims, and is commonly divided into overlapping but fairly distinct fields. These include the areas of genetic inheritance, child development, maturation, socialization, intelligence, language development, perception, learning, emotion, concepts of self, psychology in the home and workplace, sexual differentiation, life changes, ageing and bereavement. The list is endless, but our concern here is with language and cognition (perceiving, knowing and conceiving) as they are relevant to literature and literary theory.

First a warning. This page generally adopts the scientific approach: it treats mind and body as different categories of being, not in any way interconnected. It is also mechanical — the brain not only drives the body, but is ultimately reducible to chemical and physical processes: these are the bedrock of reality. For analysis on other planes of understanding see the psychoanalysis and schemas section immediately below.


Psychoanalysis is not a science, and not a branch of psychology, though often regarded so in the popular mind. Though now fragmented into many competing schools, psychoanalysis was founded on the attempt by Sigmund Freud to treat behaviours that were thought to arise from illnesses or malfunctionings of the unconscious. Freud developed a talking cure that supposedly allowed him to enter into that part of the patient's mind that is normally hidden, and effect a cure. Freud also believed that the first five year's a child's life were crucial for later development, and identified three stages. In the oral stage the libido (the free-floating sexual energy, which was the essential motivating force behind thought and behaviour) was focused on the mouth, and a person who did not develop properly beyond this stage remained somewhat gullible in later life. In the second, anal stage the child takes a keen interest in defecation, and failures to progress from this stage may also mean that the person doesn't strike the right balance between generosity and self-interest in later life. In the phallic stage the child becomes aware of sexual differences between its parents. A girl realizes she hasn't a penis, feels that she has been castrated and so identifies with the mother who has been similarly mutilated. The boy however sees himself in competition with his father for his mother's love, and may develop feelings of hostility to authority figures. Conflicts are repressed into the unconscious, to emerge in social or sexual problems in adulthood.

Little of Freud's work survives scientific investigation. The unconscious is an elusive concept, of doubtful existence. The first five years are not as crucial as Freud believed. {2} Freud's investigations were not properly controlled, and for all his explanations there exist much more plausible and testable alternatives. There is no clinical evidence for repression. {3} Psychoanalysis is lengthy, expensive, and works no better than other therapies (which also converse with the patient but assume very different theories).

Freud's approaches were developed further by other psychoanalysts, notably Jung and Lacan. Jung's archetypes operate as schemas, and usefully account for religious and cultural symbolism. Lacan's psychoanalysis has become the mainstay of some contemporary literary theory, but is at best a myth without supporting experimental evidence.

Mental Representation

How do we represent things in our minds: is it with propositions, or with images? The two are very different. Propositions are language-orientated: they employ symbols which are somewhat abstract, explicit, combine by rules and stand for things (make reference). Images, on the other hand, are analogical: they are more concrete, implicit, without clear rules of combination and can stand alone. If this sharp distinction is wanted, then the answer is that we use both. By many techniques — laboratory experiments, introspection, examination of brain-damaged patients, study of brain physiology — psychologists attempted through the 1970s and 1980s to argue that images were only vacuous representations of propositions. This view has been abandoned. Both are needed for cognitive richness, and images are now seen to be mental constructs in their own right. The most widely-accepted theory, that of Kosslyn and his coworkers, envisages images being represented in their own spatial medium, which holds images in the greatest detail near the centre of view, is dependent on graininess for resolution, and which cannot prevent images fading in time. Long-term memory holds two types of files, image and propositional, which are nonetheless linked together. Both are processed by the brain to generate, interpret and transform images. {4}

Another approach altogether, connectionism, employs the concept of information-processing networks that partially resemble the brain's own neural networks. These computer models are far simpler, of course, and use a weighting mechanism rather than the firing or non-firing at synapses. But they do give results in line with empirical evidence, and have two strong advantages. They model complex behaviour without recourse to explicit propositional rules (they program themselves from the inputs supplied), and they represent memory as predispositions distributed throughout the network (the predispositions also programme themselves, like an artificial intelligence programme deriving and then applying rules once it is fed the data.) The networks can be extended, when memory and rules are indeed widely distributed. Or they can be more modular, with local areas operating somewhat independently of others, though carrying their results through to the larger network. {5}

Concepts, Categories and Schemas

How do we group observations and thoughts to give something a name and category? Psychology has been much influenced by Anglo-American philosophy, and its first investigations accepted the approach of Frege, the founder of its twentieth-century development. Intension was the set of attributes that define a concept, and extension was the set of examples. A name or concept was therefore the conjunction of defining attributes, all of them equally representative and providing clear-cut boundaries. Is this realistic? People tested remarked that it was sometimes difficult to be sure: are portable oil-heaters to be regarded as furniture, for example? And then there was Wittgenstein's concept of games. An alternative approach was sought.

One which became popular was that of prototypes: find a characteristic example, make this the core concept, allow the concept to have typical but not necessarily delimiting attributes, and link other instances to it by degrees of typicality. But there are still problems. Some concepts are not amenable to the approach: religious beliefs, for example. Some attributes are held to be more important than others. And concepts have to be natural and coherent, to serve some larger end. {6}

Consider a motorcar. We can study it as an example of the internal combustion machine, or as a system of inter-linked systems, electrical and mechanical. But for most of us the car serves as a means of transport — safe, speedy and convenient — and we tend to judge it by these criteria. Transportation is the overarching concept, called in psychology frames or scripts or schemas. Such schemas are naturally rather fluid and ad hoc. Many more schemas have also been proposed than have been adequately tested. {7} But even at a more primitive level, that of simple concept combination, psychology adds useful empirical ballast to arcane theorizing. One repeatedly tested model is that of Tversky which states that the similarity of two concepts A and B is quantitatively given by the number of concepts shared less the sum of the attributes distinctive to A and distinctive to B. If this is so — and it does seem to be — then the theories of post-Saussurean Structuralists (and Post-structuralists like Derrida) are seriously open to question. {8}

Speech and Reading

Experimentation becomes even more useful when applied to speech and reading. It shows that syllables (and to some extent phonemes, which tend to overlap and become blurred in rapid speech) are the basic elements of comprehension. Nonetheless, word recognition (bottom up processes) and context (top down processes) are both necessary, and there are indeed two theories to model this. The cohort model argues that the initial sounds or syllables throw up various word possibilities, which are then whittled down to the correct candidate as the context is grasped and more of the word is read or heard. The process is quite complicated, with various knowledge sources — lexical, syntactic and semantic — being accessed by the brain. The TRACE model assumes that processing units at different levels — manner of production, phonemes and words — operate in proportion to their activation and strength of interconnections. Both give reasonable matches to experimental evidence, though refinements are necessary. Words read are recognized partly by sound and partly by appearance. {9}

Schemas are obviously important when it comes to making sense of text, but the models proposed so far tend to be over-simple and not easily tested. Comprehension does seem to involve parsing, analysis of literal meaning and then an interpretation of its intended meaning. Inner speech can be important, both subvocal articulation and phonological coding, probably because it facilitates transient knowledge storage. One influential theory by Kintsch and van Dijk distinguished between a microstructure of a text (set of propositions representing meaning) and macrostructure (gist of the story) and appears to be generally correct. {10}

Though good speakers are usually good writers, the difficulties experienced by brain-damaged patients show that very different processes are involved. Grice's cooperative principle appears to be broadly true, and has been extended by the spreading activation theory of Dell and others. Though writers often say their sentences to themselves before writing them down, perfectly adequate sentences are written by patients who lack this facility. Expert writers differ from non-expert markedly in two respects: they spot more errors and know how to put them right, and they organize their scripts much more effectively. {11} Certain languages facilitate thought in certain directions, but the Sapir-Whorf hypothesis has been verified only in its weak form — i.e. that languages influence but do not control perception.{12}


For some thirty years, psychology has intensively studied the ways problems are approached and solved. Many of the results confirm intuitive expectations. Experts are better than novices because they have more knowledge and experience, can chunk steps, and have devised forward-looking strategies. As in anything else, practice makes perfect, but the expertise diminishes rapidly with increasing effort (is commonly a power law: logarithm of time spent learning is inversely proportional to the time needed to solve a problem.) Creativity in the more demanding of scientific and artistic activities is indeed analogical and often ad hoc, but few problems are actually solved by the flash of inspiration. It is much more usual to seek general strategies, break a problem into sub-problems, recall past successes and modify what worked then. Interestingly, most scientists do not try to disconfirm hypotheses in the Popper manner, but rather the opposite.{13}


Whatever the philosophic difficulties, people must surely use deduction in their everyday lives: to plan, make sense of surroundings, to interpret their experiences. But how exactly? Psychologists have investigated four possible approaches — by employing abstract logic, context-specific schemas, models that represent possible states of affairs, and reasoning swayed by emotional bias. Others have some share of the truth, but it is the model theory which seems most fully to represent how people really do perform. They seem to first extract the premises involved, often by analogy, taking into account what they know of comparable situations. They then combine these premises to form an integrated model or sets of models. Finally they validate their model by looking for alternatives. If no satisfactory alternatives exist they conclude that they have properly represented the situation. Though they make their model the simplest possible — describe it in the most parsimonious fashion — they generally find the reasoning easier if the situation is one that appears sensible or familiar to them. {14}

Literature and Emotion

Given that we are affected emotionally by what we read or hear, what has cognitive psychology to tell us about the link between cognition and feelings? Not a great deal at present. Various models have been proposed, the most successful of which — Bower's semantic network theory — does correctly predict that material is learnt better when it is congruent with the subject's mood. Other predictions of Brower's model are less supported, however. Undoubtedly anxiety and depression affect performance, and a famous law of Yerkes and Dodson (1908) that performance is best at intermediate levels of arousal or anxiety has been substantiated by recent work, but not fully explained. {15}


Some people hear in colour. Others find that some words bring up specific tastes or smells. Though developed only weakly in most people, the correspondence of the sensory modalities is recognized in literary and "colourful" writing, and was exploited by the Symbolists. Though little employed by contemporary writing, there is nonetheless a great deal of scientific evidence for this phenomenon. When the interrelationships of size, space, intensity and duration are investigated for the specific senses, they not only show common patterns but a good deal of equivalence. Visual estimates of size correspond with tactile skills. The perceived duration of a sound and image correspond in the same way to duration measured by the stopwatch. Brightness applies equally to light, touch, sound and odour. Sensory inhibitions caused by abrupt intensity changes is seen in visual and tactile experience. Moreover, there is every reason to expect these results. In the first place, those parts of the cerebral cortex specifically associated with each of the senses are somewhat adjoining: their neural systems inevitably interpenetrate. Secondly, the brain as a whole operates in a diffuse cooperative way, which further serves to link the sensory functions. Metaphor, which portrays these correspondences, is a feature of brain functioning. {16}

Concluding Remarks

Cognitive science is a fast-growing area of research, and promises to shed much useful light on mental processes. Two words of caution, however. Experimental results and their interpretations are not as clear cut as this survey suggests. {17} Secondly, there have been many theories of brain functioning, and are likely to be still more in future.{18}

This and other pages in the theory section have been collected into a free pdf ebook entitled 'A Background to Literary Theory'. Click here for the download page.


1. See any introductory psychology textbook, e.g. Nicky Hayes's A First Course in Psychology (1984), and John Neale and Robert Liebert's Science and Behaviour: An Introduction to Methods of Research (1980).
2. Chapters 6 and 7 of Hayes 1984.
3. Ibid. Also pp. 439-40 in Michael Eysenck and Mark Keane's Cognitive Psychology (1995) and D. Holmes's The evidence for repression: An examination of sixty years of research in J. Singer's (Ed.) Repression and Dissociation: Implications for Personality Theory, Psychopathology, and Health (1990).
4. pp. 220-227 in Eysenck and Keane 1995, Howard Gardner's The Mind's New Science: A History of the Cognitive Revolution (1985), and S.M. Kosslyn's Image and Brain: The Resolution of the Imagery Debate (1994).
5. pp. 227-231 in Eysenck and Keane 1995.
6. pp. 247-256, ibid.
7. Chapter 11, ibid.
8. pp. 247-249, ibid. Also A. Tversky and I. Gati's Studies in Similarity (1978).
9. Chapter 10 of Margaret Matlin and Hugh Foley's Sensation and Perception (1983) and Chapter 12 of Eysenck and Keane 1995.
10. Chapter 13, ibid. Also K Rayner and T. Pollatsek's The Psychology of Reading (1989).
11. Chapter 14, ibid.
12. E. Hunt and F. Agnoli's The Whorfian Hypothesis: A Cognitive Psychology Perspective (1991).
13. Chapter 16 of Eysenck and Keane 1995.
14. Chapter 17, ibid. Also J. Evans, S. Newstead and R. Byrne's Human Reasoning: The Psychology of Deduction (1993).
15. Chapter 18 of Eysenck and Keane 1995. Also G.H. Bower's How might emotions affect learning? in S.-A. Christianson's (Ed.) The Handbook of Emotion and Memory: Research and Theory (1992).
16. Lawrence Mark's The Unity of the Senses: Interrelations Among the Modalities (1978).
17. Churchland 1986, Jordan Scher's (Ed.) Theories of Mind (1962).
18. Stephen Priest's The Theories of the Mind (1991).

Internet Resources

1. Experimental Psychology. 2001. http://www.findarticles.com/cf_dls/g2699/0001/
. Nature of experimental psychology: short article from the Gale Encyclopedia on Psychology.
2. The Philosophy of Neuroscience. John Bickle and Peter Mandik May 2001. http://www.seop.leeds.ac.uk/entries/neuroscience/. Survey of a rapidly expanding field.
3. Psychoanalysis and Its Critics. Christian Perring. http://www.uky.edu/~cperring/PPB3.HTM. Very extensive listings of psychoanalysis, psychoanalysts and criticisms of psychoanalysis.
4. Famous Figures in Psychology. J.W. Nichols. Jun. 2001. http://www.tulsa.oklahoma.net/
. Not limited to psychoanalysts.
5. Jung, Freud, Rank, Adler: Narrators of the psyche in meta-perspective. Herman Pietersen. 2003. http://www.examinedlifejournal.com/articles/
. Differing intellectual predispositions in the founders of psychoanalysis.
6. Carl Gustav Jung (1875-1961). Paul Bishop. 2003. http://www.kirjasto.sci.fi/cjung.htm. Books and Writers article. NNA
7. Jacques Lacan. Michael Cark. 1997. http://www.press.jhu.edu/books/
. JHG entry: detailed, with good bibliography but few links.
8. Intelligence Without Representation. Hubert L. Dreyfus. 1998. http://www.hfac.uh.edu/cogsci/dreyfus.html. Discusses Merleau-Ponty's Critique of Mental Representation.
9. Mental Representation. David Pitt . Dec. 2002. http://plato.stanford.edu/entries/mental-representation/. Detailed but readable entry in the Stanford Encyclopedia of Philosophy.
10. Connectionist approaches to psycholinguistics, language acquisition, and lingusitic theory. Jeff Elman. http://crl.ucsd.edu/~elman/Bulgaria/. Course description, but with useful online material.
11. On the psychology of prediction. Kahneman & Tversky. 1973. http://faculty.babson.edu/krollag/org_site/soc_psych/
. Abstract of their paper in the Psychological Review.
12. The Kintsch and van Dijk Model of Discourse Comprehension and Production Applied to the Interpretation Process. Jennifer Mactintosh. http://www.erudit.org/revue/meta/1985/v30/n1/
Paper reviewing the theory and subsequent work.
13. Thinking Model and Tools for Understanding User Experience Related to Information Appliance Product Concepts. Anu KanKainen. http://lib.hut.fi/Diss/2002/isbn9512263076/
. PhD thesis taking matters beyond Bower's semantic network theory.
14. Synaesthesia & the Arts. "Doctor Hugo". http://www.doctorhugo.org/synaesthesia/index.htm. Attractive site explores the multisensory interactions of synesthesia: good links.
15. Dictionary of Philosophy of Mind. http://www.artsci.wustl.edu/~philos/MindDict/. Excellent: search with "mental representation" etc.
16. Psicoligica: Journal of Methodology and Experimental Psychology. http://www.uv.es/psicologica/. Free online papers in English and Spanish.
17. Behavioral and Brain Science. http://www.bbsonline.org/view-ROOT.html NNA. Excellent set of papers: free when you register.
18. AmoebaWeb: Psychology on the Web. Douglas Degelman. Feb. 2004. http://www.vanguard.edu/faculty/ddegelman/amoebaweb/. Extensive listings of free psychology resources.
19. Experimental Psychology. http://www.hhpub.com/journals/exppsy/journals.html. 20. Experimental Psychology Society. http://www.eps.ac.uk/. Journals free online to members only.
21. Society of Experimental Social Psychology. http://www.sesp.org/. Membership ($45/year) and journal available to professionals.
26. Harnessed: How Language and Music Mimicked Nature and Transformed Ape to Man by Mark Changizi. BenBella Books, 2011.
22. Incomplete Nature: How Mind Emerged from Matter by Terrence W. Deacon. W. W. Norton & Co., 2011.
28. Music, Language, and the Brain by Aniruddh D. Patel. O.U.P. 2010.