back to list of Articles About Arrowsmith School

STUDYING HOW THE BRAIN LEARNS
ARE THERE ANY USEFUL IMPLICATIONS FOR INSTRUCTION?

By Barbara Arrowsmith Young & Marcel Danesi

A perusal of the main journals of education, teaching, and instruction published during the last three decades reveals that a growing number of practitioners in the field have been looking to the neurosciences for insights and guidance. Is it simply a sign of the times, when anything that is clothed with "scientific method" is instantly accorded "authoritative" status? Or is it just another way of replacing good teaching with a gimmick method? Between the lines of the published reports there seems to be an implicit belief that knowledge about the brain will provide an empirical basis upon which to construct a truly coherent theory of teaching or, at the very least, a framework for assessing and interpreting theories or models of education. The fuss over the brain sciences seems to have started when Eric Lenneberg’s widely influential 1967 study put forward convincing evidence to support a "critical period" for the acquisition of language, i.e. a biologically determined timetable for language that starts at birth and is completed at adolescence. Debate on the implications that this finding had for general education was ignited almost immediately, and it continues uninterrupted to this day. The fundamental feature that differentiates teaching approaches that are based on brain research from others is an explicit sequencing and formatting of the material to be learned and practiced in ways that are purported to simulate how the brain handles incoming information.

The authors of the present study, too, have been involved in researching and using brain-based models of learning (e.g. Danesi 1986, 1988, 1991, 1994, Danesi and Mollica 1988). The first author also runs a school, called the Arrowsmith School, for thirty students in full-time attendance, of whom three quarters are of elementary school age. Each child is diagnosed as having a specific learning disability and is assessed according to dysfunction. On that basis a speciallytailored approach and curriculum are designed for the child. Such "brain-based" teaching has produced so many promising results over the years that it was introduced experimentally into the Catholic School Board of Toronto into two classrooms with highly significant results. Indeed, the Board has recently decided to expand the experiment with a concurrent research component. The Arrowsmith program is designed to strengthen specific components of various learning capacities that are necessary for learning specific skills. For example, if the student has a weak visual symbol memory, then the student’s capacity to effectively learn and remember visual symbol patterns will be impaired and the student will have significant difficulty remembering how to spell words and how to read. The strategy of the Arrowsmith program in this case is not to get the student to overlearn specific sight words through repetitive exposure to the words, but to work on increasing the student’s capacity for memorizing symbol patterns through a program of having the student memorize increasingly complex visually presented symbol patterns in 27 foreign languages. The general capacity for learning visual symbol patterns improves which leads to improvements in skill learning such as reading.

Interest on the part of the second author in this line of thinking was triggered in 1986 when he became involved with neuropsychologists and psychiatrists working with language-handicapped children in Italy (e.g. D’Alfonso, Danesi, De Lellis, and Mastracci 1986, Danesi and De Lellis 1994). Collaborative projects on how to design effective teaching materials for such children led to the framing of bimodality theory, or the view that the two modes of learning - experiential and analytical - are systematically cooperative in the processing of verbal input (language which a learner receives and from which he/she can learn) and in influencing verbal intake (input which the learner can actually utilize cognitively). Incidentally, when the term bimodality was proposed in 1986, the author was not aware of the fact that it had already been in use among neuroscientists as a synonym for complementary hemisphericity theory (e.g. Bogen, DeZure, Tenhouten, and Marsh 1972, Dunn 1985). It continues to be used in this way in the relevant literature (e.g. Ressler 1991). He was also not cognizant of the fact that the term was employed by Laurence Ridge, a professor of mathematical education at the University of Toronto, five years earlier in 1981. Ridge’s use of the term in that year was, to the best of our knowledge, the first time it was so employed in the educational literature.

Three decades after Lenneberg’s watershed study, the time has come to ask ourselves if the fuss over the neurosciences has been worthwhile. Can knowledge about the brain truly inform not only the way we teach children with disabilities, but also anyone else? And what does it mean to say that a teaching approach is "brain-based?" We doubt if these questions can be answered conclusively, simply because there is no empirical way to demonstrate that a specific teaching procedure, for instance, is capable of activating a certain part of the brain - unless we put our students through a PET scan as we teach them something! And even if it could be shown that certain parts are activated at certain stages or in response to certain instructional stimuli, what does that truly mean? We know so little about the connection between brain activities and learning processes that all it would really show is a "cooccurrence" between an input and a brain activity, not a "correlation" between the two. Nevertheless, it is our cautious opinion that the foray into the neuroscientific domain on the part of practitioners has been anything but fanciful. If nothing else, it has forced us to look more closely at the conditions we create in a classroom and at the theoretical suppositions underlying instructional practices and teaching curricula.

From a biological perspective, learning something new implies a reorganization of the structure of some, if not most, parts of the brain. However, we alert the reader to the fact that in their enthusiasm, neuroscientifically- inclined educators have perhaps not always been judicious and cautious in applying the brain research to instruction. So, the present synopsis will highlight only the main ramifications that have ensued from the use of brain-based instruction which, needless to say, confirm our own experiences in this area. We believe, in effect, that brain-based pedagogy has truly enriched the research agendas, discourses, and practices of our profession.
Some Background Historical Matters

It is now common knowledge that the left hemisphere (LH) is the primary biological locus for language and analytical thought. The apparent superiority of the LH for language was established more than a century ago in 1861 by the French anthropologist and surgeon Pierre Paul Broca, when he published his classic study of a patient who had lost the ability to articulate words during his lifetime, even though he had not suffered any paralysis of his speech organs. Noticing a destructive lesion in the left frontal lobe of the LH at the autopsy of this patient, Broca was thus able to present concrete evidence to link the articulation of speech to a specific cerebral site. Fifteen years later, in 1874, the German neurologist Carl Wernicke brought forward further evidence linking the LH with language. Wernicke documented cases in which damage to another area of the LH consistently produced a recognizable pattern of impairment to the faculty of speech comprehension. Then, in 1892 Jules Déjerine found that reading and writing deficits resulted primarily from damage to the LH alone. So, by the end of the nineteenth century the research evidence was pointing convincingly to the LH as the biological locus for language. This led to "localization theory" - the view that specific mental functions had precise locations in the brain. A corollary to this theory was the notion of "cerebral dominance" - the view that the verbal LH was the dominant one for generating the higher forms of cognition.

With a few notable exceptions (e.g. Lashley 1929, Vygotsky 1931, Jakobson 1942, Luria 1947), localization theory dictated the research agenda of the neurosciences during the first half of the present century. The dissenters argued that language in a restricted sense - i.e. as sounds, words, and meanings - could indeed have a primary locus in the LH; but as a more encompassing expressive phenomenon it was more likely to involve neural processes that were distributed throughout the brain. Vygotsky (1931) also suggested that the whole brain was endowed at birth with a unique kind of "plasticity" that rendered it highly sensitive and adaptive to environmental stimuli during childhood. Therefore, he put forward the intriguing proposal that the neurological structures associated with the mental functions were constantly subject to modifications from sociocultural influences.

It was, however, during the fifties and sixties that the first serious doubts were cast on the theory of dominance by the widely-publicized studies conducted by the American psychologist Roger Sperry and his associates on epilepsy patients who had had their two hemispheres separated by surgical section (see Springer and Deutsch 1993 for a detailed account of the relevant experiments). These studies made three crucial accomplishments possible: (1) they showed that both hemispheres, not just a dominant one, were needed in a neurologically- cooperative way to produce complex thinking; (2) they provided a detailed breakdown of the main psychological functions according to hemisphere; (3) they confirmed that the LH was the primary site for language and analytical thought. As mentioned, the latter finding was further entrenched in 1967 when Eric Lenneberg published his famous book. On the basis of a large body of clinical studies, Lenneberg noticed that most aphasias - the partial or total loss of speech due to a disorder in any one of the brain’s language centers - became permanent after the age of puberty. This suggested to Lenneberg that the brain lost its capacity to transfer the language functions from the LH to the nonverbal right hemisphere (RH) after puberty, which it was able to do, to varying degrees, during childhood. Lenneberg concluded that there must be a biologically-fixed timetable for the lateralization of the language functions to the verbal LH and, consequently, that the critical period for the acquisition of language was before adolescence. Although his time frame has been disputed (e.g. Scovel 1988), Lenneberg’s basic hypothesis that there is a fixed period of time during which the brain organizes its division of labor remains, to this day, a plausible theory and a target for much debate.

By the early seventies the neurosciences had charted out a flourishing field of inquiry. The brain research suggested, above all else, that for any new input to be comprehensible, it must occur in contexts that allow the synthetic functions of the RH to do their interpretive work first. In the case of tutored, or classroom, adult learning this has rather far-reaching implications. Above all else, it suggests that the brain is prepared to interpret new information primarily in terms of its contextual characteristics.

Today, neuroscientists have at their disposal a host of truly remarkable technologies for mapping and collecting data on brain functioning. The use of positron emission tomography (PET brain scanning), for instance, has become a particularly powerful investigative tool for neuroscientists, since it provides images or "maps" of mental activities such as language (Calvin and Ojemann 1994). Such maps have given us a fairly good idea of how the neocortex is involved in producing various psychological functions, psychomotor movements, etc. However, there are other areas of the brain of which very little is known - such as the areas below the cortex, which are involved in the emotions. In evolutionary terms, these areas are older, tying us to our primate heritage. So, although much has been learned about the neocortex since 1861, the brain in its totality still remains a largely mysterious organ.

Neuroscientifically-Raised Issues for Education

The foray into the neurosciences has made it possible to raise several issues that have far-reaching implications for both educational research and instructional practices. First, there is the question of a "critical period" for learning. If we take Lenneberg’s work to its literal conclusion then we would be faced with the proposition that learning at later stages of life is ineffectual at best. But, as all SCS instructors and students know, this is far from being the case. Learning occurs at all ages. And, moreover, the "ways" in which adults learn is not much different from how children learn. This being the case, some other explanatory framework, other than a critical period one, would have to be contemplated to account for the capacity of adults to acquire competence in any subject area after puberty.

Perhaps the most exhaustive critique of this hypothesis has come from the pen of Thomas Scovel (1988) who, in reviewing the extensive body of research evidence assessing the critical period, reached the conclusion that there are no clear-cut findings to suggest biological constraints on learning, but rather psychological ones such as motivation, cognitive style, and affective variables. Lenneberg, as Scovel points out, simply assumed that learning was easier for children.

The recent work on brain mapping suggests that the two hemispheres differ not so much in the type of stimuli they are designed to process but, rather, in the manner in which they process stimuli. In previous work (e.g. Danesi and Mollica 1988) we have preferred to adopt the terminology L-Mode and R-Mode, to refer to LH and RH functions respectively (in imitation of Edwards 1979), so as to allow for the fact that the RH may be involved in some contralateral L-Mode functions and the LH in R-Mode ones. Moreover, the research now indicates that while each hemisphere is specialized to handle a certain specific function, it does so in tandem with complementary or parallel processing patterns taking place in the other hemisphere - pure analytical thinking simply does not exist in the human brain, nor does pure intuitive thinking!

Research has shown that the R-Mode dominates the learning process during its initial stages, with the L-Mode taking on more of the burden in later stages. Thus, the learning process will enlist the RMode and/or the L-Mode according to the specific nature of the learning task at hand.

Brain-Based Methods

The foray into the neurosciences has been an especially productive one for language teaching practices. The research on the role of the RH in language, for instance, has led to the design of three major methods in the last three decades - Lozanov’s (1979) "Suggestopedia," Asher’s (1977, 1981) "Total Physical Response," and Krashen and Terrell’s (1983) "Natural Approach." These can be characterized schematically as follows:

• Lozanov stresses the importance of creating a learning environment that is capable of activating subliminal R-Mode processes. This is why he suggests the technique known as the séance - a period during which students relax and sit comfortably in reclining chairs listening to background music (usually the slow movements of Baroque composers such as Bach, Handel, Vivaldi, Corelli and Telemann) while new language input is being read in the new language and in translation.

• Asher’s Total Physical Response method is designed to impart the new language mainly through physical activities. Moreover, Asher suggests that the criterion for including an item of vocabulary, grammar, or communication at a particular point in the learning sequence should be the ease of assimilation shown by the students. If the item is not learned rapidly, then they are obviously not ready for that item. Hence, it should be withdrawn and presented again at some future time. The "flow" of learning which Asher intends to set in motion with Total Physical Response goes from concrete actions to linguistic abstractions; i.e. from the R-Mode to the LMode. Asher claims that when a sufficient amount of R-Mode learning has taken place, the L-Mode will be triggered naturally to produce the more abstract linguistic notions. So, he views grammatical training as virtually unnecessary.

• Krashen and Terrell’s Natural Approach became one of the most discussed teaching proposals in the eighties, probably because of its intuitive appeal to teachers and learners alike. It too ascribed great salience to the RMode during all stages of learning, but especially during the initial ones. Krashen and Terrell viewed the R-Mode as the natural "acquisitional" mode of the student. They deemed grammar training to be virtually useless, since they claimed that knowledge of structure would emerge inductively through the L-Mode’s inbuilt "monitoring" system. However, before his untimely death in the early nineties, Terrell (1991) modified this radical view somewhat.

Suggestopedia, Total Physical Response, and the Natural Approach have constituted the first serious attempts to organize classroom language teaching around the brain’s acquisition mode - the R-Mode. In so doing, however, they have downplayed the role of the L-Mode perhaps too drastically. They seem to generate much interest and enthusiasm in teacher and learner alike during the initial stages - the stages during which the R-Mode probably dominates the intake of novel information. But their overemphasis on this mode throughout the course of learning also probably explains why they have not caught on across the entire profession. They simply do not place enough importance on the L-Mode and on the analytical learning subsystems that it encompasses. There really can be no method or approach that is designed in a purely R-Mode or L-Mode fashion.

General Issues and Implications

In addition to spawning the methods just discussed, the foray into the neurosciences has also raised some important general issues for the entire teaching profession. Brain-based teaching models suggest at least two "instructional- design principles" for pedagogy generally: the modal directionality principle and the modal focusing principle.
Modal Directionality

It would appear, above all else, that the teaching of new notions and structures should follow an R-Mode (experiential) to L-Mode (analytical) "flow." This means that during the initial learning stages students need to assimilate new input through observation, induction, role-playing, simulation, oral tasks, and various kinds of interactive activities. But we would quickly add that formal explanations, drills, and other LMode procedures must follow these stages, since we have found that control of structure will not emerge spontaneously, as Asher and others claim. Incidentally, identifying a learning task or unit as having an L-Mode or an RMode focus implies only indicating which mode is to be emphasized in the overall design of the task, and does not necessarily indicate which specific hemispheric functions will be activated. The modal directionality principle thus claims: (1) that experiential forms of tutoring belong to the initial learning stages, and (2) that teaching should move progressively towards a more formal, analytical style in the later stages.

An analogy to music teaching can perhaps be made to illustrate the practical implications of this principle. Learning how to play a new piece on the piano, say, entails the ability to mold the component mechanical skills needed to play the notes, phrases, etc. of the piece successfully into the global skill of "playing the music." So, in order to give the learner’s L-Mode a better opportunity to analyze and organize the component skills into automatic psychomotor routines, the teacher normally starts out by playing the piece for the student, making appropriate aesthetic comments here and there. In this way, the student’s R-Mode has an opportunity to decipher the new musical input in a global aesthetic way. The component mechanical skills can now be understood separately and practiced apart from their expressive modalities. Needless to say, an advanced music student who is already in firm control of the required LMode skills through previous training will not have to spend as much time on this component as would a beginner. When the student has mastered the L-Mode aspects of the piece, then he/she will be in a position to integrate them with the R-Mode ones as he/she performs the piece. A consummate performance of the piece is, from a neurological perspective, a bimodal feat, requiring the integrated contribution of both the RMode and the L-Mode to the performative task at hand.

The modal directionality principle implies, above all else, that the teacher should leave ample room for student improvisation during the early learning stages. Instructional techniques which focus on explanations will be of little value, since the students generally have no preexisting L-Mode schemata for accommodating the new input directly. In order to make the new material accessible to the L-Mode (intake), therefore, the early stages should involve teacher and learner alike in activities enlisting exploration, imagination, spontaneity, and induction. Once the initial learning stages have been completed, the teacher can "shift modes" and begin to focus more on formal, mechanical, rule-based instruction.

Modal directionality can be seen to be a different version of the oldest principle in teaching - the inductive principle. But unlike its use in strictly inductivist methods, it does not require the deployment of induction for all learning tasks, only those that involve new input. Thus, if a learning task contains knowledge or input that the learner can already accommodate cognitively, directionality can be efficiently avoided. So, modal directionality is really a common-sensical brain-based principle that good teachers have always embodied into their modus operandi. It is virtually a "law of learning" which claims that teaching should ensure a constant movement from experiential to expository learning conditions, from practical to theoretical content, and from concrete to analytical presentation styles.

Modal Focusing

The principle of modal focusing claims that at certain points in the tutored learning process the students will need to focus on one mode or the other for various reasons. After the learners have grasped the new concepts in an R- Mode way, for example, their mental systems can be said to be prepared to assign them to appropriate L-Mode categories. At this point, the teacher can step in with suitable LMode techniques which focus on pattern practice, explanations, etc.

Modal focusing might also be required at points in the learning process when, for instance, a learner appears to need help in overcoming some error pattern that has become an obstacle to learning - L-Mode focusing allows the students an opportunity to focus on formal matters for accuracy and control; R-mode focusing on matters of understanding and conceptualization. It is important to point out that the modal focusing principle in no way implies that mechanical practice be conducted in an uncontextualized way. On the contrary, meaningful contexts should always be provided not only for new input, but also for focusing routines. This allows the R-Mode to complement and strengthen the intake operations of the L-Mode, especially during more mechanically- oriented focusing tasks. Contextualized instruction enables the learners to relate L-Mode form to R-Mode content.

In sum, the general teaching implications that modal directionality and modal focusing call forth can be summarized in point form as follows:

During an R-Mode Stage:

• Classroom activities should be student-centered and involve students and teacher in a complementary fashion.
• Novel input should be structured in ways that activate sensory, experiential, inductive forms of learning.
• The students’ inductive and exploratory tendencies should be encouraged to operate freely when introducing new information.

During an L-Mode Stage:

• The focus now shifts to the teacher.
• Explanations, drills, etc. should follow the experiential learning phases.
• Focusing on some problematic aspect of the subject being taught is to be encouraged if a student appears to have difficulty grasping it or using it.

Concluding Reflections

The reader is by now aware that we posed the question in the subtitle of this essay "Are There Any Useful Implications for Adult Instruction? " only rhetorically. As we have attempted to argue, there are fundamental implications. But we also wish to emphasize that the research applying neuroscientific findings to instruction has produced very little in the way of empirical research findings. Most of current brain-based views of instruction have been based primarily on extrapolations from the neuroscientific literature or from the observations of teachers. So, we cannot help but agree with Obler (1983) when she observes that, unless we are very careful, many unnecessary problems are bound to crystallize when extracting too many implications from the work on the brain.

Interpreting the research on the role of the RH for adult education, and then translating it into pedagogical principles, has been particularly instructive (Satz, Strauss and Whitaker 1990). It has now become apparent that the two hemispheres do share some features. The LH has been shown to have the capacity to engage in some holistic and parallel processing, and the RH in some analytic and serial processing. All attempts to construct models based on the participation of the RH at various stages, and to translate such models into instructional practices, therefore, must tread very cautiously and judiciously. Brain-based views of classroom learning are, more accurately, useful for providing additional understanding about learning, rather than constituting explanations of it.

In addition to the issues raised above, it should be pointed out that the foray into the neuroscientific domain raises another interesting question, and one that is rarely addressed. Is it possible or desirable to take account of the likelihood that learners will have different hemispheric learning styles? There exists some evidence in the neuroscientific literature that hemispheric style (a preference for one or the other learning mode) correlates with handedness, gender, and various environmental factors (Geschwind and Galaburda 1987). From an educational perspective, it is obvious that a student with a dominant LMode learning style will gain very little from an abundant use of R-Mode techniques. Similarly, analytical instruction for students with an RMode learning style would probably prove equally futile. However, much more empirical work would need to be done in this area. Nevertheless, the fact that the above question can be asked in the first place is an outcome of the foray into neuroscientific turf.

As a final word, we would like to remark that the foray should continue in the future, producing interesting hypotheses, constructs, and suggestions for conducting research on classroom learning and for modeling instruction. If teachers are truly interested in understanding how their students learn and in responding pedagogically in an appropriate way then, as Spolsky (1989: 279) put it a decade and a half ago, it is "certainly not unreasonable to seek insights from the brain sciences."

References

Asher, J. J. (1977). Learning Another Language Through Actions: The Complete Teacher’s Guidebook. Los Gatos: Sky Oaks.

Asher, J. J. (1981). The Total Physical Response: Theory and Practice. In: H. Wintz (ed.), Native Language and Foreign Language Acquisition, pp. 324-331. New York: New York Academy of Sciences.

Bogen, J. E., DeZure, R., Tenhouten, W. D., and Marsh, J. F. (1972). The Other Side of the Brain: The A/P Ratio. Bulletin of the Los Angeles Neurological Societies 37: 49-61.

Broca, P. (1861). Remarques sur le siège de la faculté du langage articulé suivies d’une observation d’aphémie. Bulletin de la Société d’Anatomie 36: 320- 357.

Calvin, W. H. and Ojemann, G. A. (1994). Conversations with Neil’s Brain: The Neural Nature of Thought and Language. New York: Addison- Wesley.

D’Alfonso, A., Danesi, M., De Lellis, M., and Mastracci, M. (1986). Problemi di neuropedagogia. Quaderni L’Ipetro 29. L’Aquila: Penne.

Danesi, M. (1986). Research on the Brain’s Hemispheric Functions: Implications for Second Language Pedagogy. Lenguas Modernas 13: 99-113.

Danesi, M. (1988). Neurological Bimodality and Theories of Language Teaching. Studies in Second Language Acquisition 10: 13-35.

Danesi, M. (1991). Neurological Learning Flow and Second Language Teaching: Some Evidence on the Bimodality Construct. Rassegna Italiana di Linguistica Applicata 23: 19-29.

Danesi, M. (1994). The Neuroscientific Perspective in Second Language Acquisition Research. International Review of Applied Linguistics 22: 201-228.

Danesi, M. and De Lellis, M. (1994). Apprendimento linguistico bimodale ed educazione emozionale: l’approccio neuropedagogico. In: M. De Lellis (ed.), Apnostressterapia: un trattamento preventivo curativo della patologia psicosomatica dell’infanzia e dell’adolescenza, pp. 103- 113. Roma: Verduci.

Danesi, M. and Mollica, A. (1988). From Right to Left: A "Bimodal" Perspective of Language Teaching. Canadian Modern Language Review 45: 76-90. Déjerine, J. (1892). Contribution à l’étude anatomopathologique et clinique des différents variétés de cécité verbale. Comptes Rendus des Sciences de la Société de Biologie 9: 61-90.

Dunn, B. R. (1985). Bimodal Processing and Memory from Text. In: V. M. Rentel,S. A. Corson, and B. R. Dunn (eds.), Psychophysiological Aspects of Reading and Learning, pp. 12-29. New York: Gordon and Breach.

Edwards, B. (1979). Drawing on the Right Side of the Brain. Los Angeles: J. P. Tarcher.

Ellis, R. (1986). Understanding Second Language Acquisition. Oxford: Oxford University Press.

Geschwind, N. and Galaburda, A. (1987). Cerebral Lateralization: Biological Mechanisms, Associations, and Pathology. Cambridge, Mass.: MIT Press.

Jakobson, R. (1942). Kindersprache, Aphasie und alge-meine Lautgesetze. Uppsala: Almqvist and Wiksell. Krashen, S. D. and Terrell, T. (1983). The Natural Approach: Language Acquisition in the Classroom. Oxford: Pergamon.

Lashley, K. S. (1929). Brain Mechanisms and Intelligence. Chicago: University of Chicago Press.

Lenneberg, E. (1967). The Biological Foundations of Language. New York: Wiley.

Lozanov, G. (1979). Suggestology and Outline of Suggestopedy. New York: Gordon & Breach.

Luria, A. (1947). Traumatic Aphasia. The Hague: Mouton.

Obler, L. K. (1983). Knowledge in Neuroscience: The Case of Bilingualism. Language Learning 33: 159-191.

Ressler, L. E. (1991). Improving Elderly Recall with Bimodal Presentation: A Natural Experiment of Discharge Planning. The Gerontologist 31: 364-370.

Ridge, H. L. (1981). A Two- Way Street: Multiculturalism in Mathematics and Mathematics in Multiculturalism. In: K. A. McLeod (ed.), Intercultural Education and Community Development, pp. 54-62. Toronto: Guidance Centre, Faculty of Education.

Rumelhart, D. E. and McClelland, J. L. (eds.) (1986) Parallel distributed processing. Cambridge, Mass.: MIT Press.

Satz, P., Strauss, E. and Whitaker, H. (1990). The Ontogeny of Hemispheric Specialization: Some Old Hypotheses Revisited. Brain and Language 38: 596-614.

Scovel, T. (1988). A Time to Speak: A Psycholinguistic Inquiry into the Critical Period for Human Speech. Rowley, Mass.: Newbury House.

Spolsky, B. (1989). Conditions for Second Language Learning. Oxford: Oxford University Press.

Springer, S. P. and Deutsch, G. (1993). Left Brain, Right Brain, 4th ed. New York: W. H. Freeman.

Terrell, T. D. (1991). The Role of Grammar Instruction in a Communicative Approach. Modern Language Journal 75: 52-63.

Vygotsky, L. S. (1931). Storia dello sviluppo delle funzioni psichiche superiori. Firenze: Giunti-Barbèra.

Wernicke, C. (1874). Der aphasische Symptomkomplex. Breslau: Cohn & Weigart.

Barbara Arrowsmith Young is Director of the Arrowsmith School, designed to provide solutions for students with learning disabilities.

Marcel Danesi is Professor of Semiotics and Communication Theory at the University of Toronto.