|
Learning a Foreign Language Beyond the Final Exam
Ravi Purushotma University of California, Los Angeles
Abstract Although students are required to learn a foreign language in school, many do not retain what they learn and have trouble using it once they leave classroom. Often they feel they are simply poor language learners, yet all successfully learned at least one language as a child. Some strategies adults commonly use while learning languages include relying on corrective feedback, narrowly focusing practice on specific items and analyzing abstract concepts. While these strategies help accelerate learning in the short-term, they often lead to poor long-term results (Elman, 1993; Bjork, 1999). This paper identifies how current practices in adult language learning might be hindering long-term retention and outlines ways in which the learning process could be restructured to prolong retention and increase ultimate fluency. Learning a Foreign Language Beyond the Final Exam
Perhaps one of the most remarkable accomplishments of human learning
occurs universally within the first decade of life. By the age of six,
children acquire a vocabulary exceeding 10,000 words (Anglin, 1993) and a
full command of basic grammar (Brown, 1973). By contrast, foreign language
learners who begin learning in adulthood, even with their far more refined
cognitive systems, will often never fully reach the grammatical
proficiency a child does -- even after decades of exposure to the target
language. For example, around the age of three children learn how to
correctly distinguish between the English articles ‘the’ and
‘a/an’ (Brown, 1973), whereas an adult immigrant not used to this
distinction might go their whole life distinguishing the two incorrectly.
In fact, after age six, it appears the older a learner is when they start
learning a language, the worse their long-term success in the language
will be. This decrease continues until around puberty -- after which time,
the outcome of learning a new language becomes uniformly poor (Johnson
& Newport, 1989). This discrepancy between the highly developed
cognitive capacity of adults and their poor ability to learn language has
baffled linguists, leading many to take the position that language
development in children must be assisted by an innate ability to decipher
and learn languages. This ability then begins to atrophy between age six
and puberty, at which time it is no longer useable and a foreign language
must instead be learned through the same learning processes one would use
for other non-language related skills (Chomsky, 1957; Lenneberg, 1967;
Pinker, 2002).
While this explanation provides an account for why ultimate
attainment patterns drop between six and puberty, the correlational nature
of its support makes it important that we be cautious about making
decisions based upon it for our classrooms. As with any correlation
involving age, a myriad of confounding variables exist, making it
difficult for us to see the whole picture. Two alternates deserving
consideration are either that a third variable also correlated with age is
responsible for diminishing ultimate attainment, or that age is simply one
of numerous factors contributing to the decline. For example, early last
century a series of correlational studies argued that raising a child bilingually would lower their intelligence
(Smith 1923, Saur 1923). Consequently, parents acting on these findings
unfortunately took steps to purposely limit the foreign language exposure
of their child and prevent them from being raised bilingual. Evidence
developed since then, however, has shown this not to be the case (Peal
& Lambert, 1962) and that such children were needlessly deprived of
one of the simplest ways of achieving multilingual fluency. Many
researchers now see these early findings against bilingual intelligence as
conclusions drawn by mistakenly using correlation to implicate causation,
in which extraneous socioeconomic variables such as stress and poverty
were confounded with the immigrant populations sampled for bilingual
participants.
One alternative explanation for why adult language learning
potential is so low is that the learning environments constructed for
adults may be oriented toward increasing the speed at which adults learn,
even if at the expense of the long-term potential they can attain. Bjork
(1999) provides numerous examples of teachers modifying their teaching
practices in ways that increase immediate performance on students’
exams; however, performance gains are not sustained over time and may
actually impair long-term learning. Likewise, adults who move to a foreign
country and learn a new language through immersion may be tempted to
employ similar strategies to expedite their learning. This explanation
would then be consistent with findings that although children are often
regarded as miraculous language learners, adults do, in fact, progress
through their initial stages of language learning at a faster rate (Snow
& Hoefnagel-Höhle, 1978).
To date, little work has been done looking into the long-term
effects of various language teaching practices on students’ eventual
attainment or outlining how adults could be brought closer to the levels
of ultimate fluency that children reach. Traditionally this has been
because of the extensive duration needed to conduct longitudinal studies
on language learning, and the belief that adults’ diminished potential
is due to a biological and irreversible loss of the ability to reach the
levels they were able to attain as children. Still, as can be observed by
talking to non-native speakers of English, there remains enormous
variation in how proficiently adults learn a foreign language – with
little speculation about how the cognitive processes involved in their
initial acquisition of the second language might be affecting their
current proficiency. As a starting point for understanding why this might
be, we can look at how differences between the strategies children and
adults use to learn languages affect long-term learning when used to
acquire skills besides language – where long-term effects are more
visible. Corrective
Features
Initial observations examining the differences between child and
adult language learning identified the discrepancy in the amount of
corrective feedback each receives when making a grammatical mistake.
Though adults may stop to correct children when they say something untrue,
they almost never correct children when they say something ungrammatical (Chomsky,
1975). By contrast, if an adult produces an ungrammatical sentence, the
listener is far more likely to inform the adult of the errors in their
sentence structure. In the short-term, this use of corrections helps
learners to quickly address mistakes they have made and rectify them
before the final exam. In the end, however, frequently providing immediate
corrections when a mistake is made can stifle results.
When first observed, children’s ability to learn without
corrective feedback was thought to be a phenomenon unique to child
language acquisition. However, recent findings show that the ability to
learn without explicit feedback is by no means specific to first language
acquisition, nor is it unique to children. Studies employing vernier
discrimination tasks present participants with two lines segments almost
parallel, but with one line offset minutely to the left or right. Adults
trying to report the direction of the offset line are able to improve
their accuracy simply by repeating the task – even if no feedback is
given as to the correctness of their answers (Fahle & Edelman, 1993).
While providing feedback on participant’s individual responses in these
tasks can accelerate learning, providing rare and infrequent feedback
achieves the same rate of improvement (Herzog & Fahle, 1997). If, for
long-term results, only minimal levels of feedback are necessary for
learning, it is possible children could get all the feedback they need
from implicit cues -- such as a parent going out of their way to model the
correct answer: “Mommy,
I goed to school yesterday!” Participants
in another study were taught the computer language LISP through a
software-training program (Anderson et al., 1989). In one group, the
program was set to give corrective feedback to learners on mistakes they
made, whereas in the other group this feedback function was turned off.
While the group with the explicit feedback began learning the language
significantly faster, neither group showed an advantage by the end of
training.
Besides being unnecessary, further evidence suggests that immediate
corrective feedback, such as that a teacher might give when a student says
something ungrammatical, can be damaging to long-term learning (Goldstein
& Rittenhouse, 1954). In training motor tasks, learning can be
accelerated by providing corrective feedback after each attempt at a task,
however, participants who receive feedback only after each set of 15
attempts ultimately master the task with greater proficiency (Schmidt,
1988).
With this negative relationship between corrective feedback and
ultimate potential in mind, we can speculate about how teaching strategies
may have a similar relationship in language learning. A teenage or adult
learner is likely to seek out extensive corrective feedback on their
language skills – perhaps hoping to improve their short-term performance
for the next exam or encounter with somebody who only speaks the target
language. However, similar to other learning tasks where corrective
feedback is high, the long-term results might be poor. By contrast,
children below the age of six learn through few grammatical corrections
and ultimately achieve native-level grammatical competence. While this is
again a correlational relationship, given that it is a causal factor in
other domains of learning, it deserves consideration as at least one
potential causal factor in long-term language proficiency. Textbook
Organization
Like corrective feedback, two other learning strategies, massing
practice of a specific item and reducing variations in examples, are also
uniquely used in adult language learning environments and known to create
poor long-term memory representations. When opening any standard language
textbook, content tends to be massed into distinct sections: Chapter 1: Present tense,
Chapter 2: Food, Chapter 3: Numbers. In the short term, this taxonomic
organization can allow learners to feel more in control because they can
focus their studying on the specific content for a given section. In the
long term, however, this organizational structure might be less efficient
than a more integrated design. To
better understand why massing could be ultimately harmful, we should first
look at the memory framework outlined by Bjork (1992) in his “New Theory
of Disuse.” His theory proposes that memory strength is doubly indexed
by both retrieval strength and storage strength. Retrieval
strength represents what a layperson might intuitively call memory. Simply
put, if retrieval strength for an item is above a certain threshold, it
will be easily recalled -- if it drops below that threshold, the item will
appear forgotten. Building retrieval strength alone is helpful for keeping
an item retrievable in the short-term — such as an upcoming exam. This
is built up rapidly through massed practice sessions, though it
deteriorates with time. For
those concerned with long-term retention, storage strength is the index
that determines both how well memories will be maintained over time and
how effectively prior learning can “come
back to” learners – such as once they are placed in a foreign country
and need to recall everything they learned in the classroom. For example,
think of the phone number you had when you were 11 years old. For many
people this number cannot be immediately recalled, as the retrieval
strength for that item is below threshold. Should they, however, move
somewhere new where they was assigned the exact same phone number they had
as a child, they would likely re-learn this number with ease due to the
extensive storage strength they built up for it as a child.
With this distinction between retrieval and storage strengths in
mind, the primary goal of foreign language teachers should be to maximize
their student’s storage strength of the target language. One way to
achieve this is to space practice of a given item, such that retrieval
strength will drop more in the intervening time between practice sessions.
This way, instead of quickly retrieving words by using associations
specific to events the week they first learned the word, students must
work to think through and rebuild the retrieval strength that was lost
during the intervening time; this process of rebuilding retrieval strength
simultaneously increases storage strength. While this may frustrate
learners more as they will have to think harder to be able to retrieve the
item from memory, eventually, storage strength will be boosted
considerably higher than if retrieval strength remained high enough to
retrieve the item effortlessly. For example, in a unit on the present
tense, students might first be assigned a massed set of fill-in-the-blank
exercises focusing on present tense is-are usage: (1)
Mervin ___ caffeinated. [is] (2)
Mervin and Raoul ____ caffeinated. [are] (3)
Vladimir ____ melting. [is] Students performing these
exercises in one session will likely answer question three with little
effort. This is because having just performed exercise one, the
association between a single person subject and using ‘is’ instead of
‘are’ remains highly accessible, such that when answering question
three they simply need to remember the act of answering question one,
instead of remembering why the answer is singular. On the other hand, if
question three were practiced a week later, it would be more difficult,
but the long-term benefits of a successful retrieval would be far greater
– as the extra work put into recovering the diminished retrieval
strength needed to answer question three would also increase storage
strength.
In addition to the problems created by massed practice of
grammatical skills, the thematic structuring of language textbooks creates
difficulties for memory of vocabulary items. One of the most important
mechanisms of memory involves the importance of associations and cues. As
an item enters memory, it establishes links and ties with everything that
might be associated with it: other learned items, the mood of the learner,
the presentation style of the item, the smell of the environment, words
that might rhyme with the item, etc. (see Roediger & Guyunn, 1996).
All of these associations then become the keys for being able to
successfully retrieve the item back out of memory by providing cues in
which a specific item can be differentiated from other items in memory.
One common way in which memory retrieval fails is when too many items are
all associated with the same cue – thus making it no longer useful for
distinguishing a specific item to be recalled (Watkins & Watkins,
1975). For
example, try to remember what you were doing both last New Years Eve and
the evening of January the 13th. Most likely, you had a much easier time
remembering last New Years Eve, than January 13. This is because everyone
around you was acting outside of their normal routine on New Years Eve --
leaving a unique set of associations to all memories encoded on that day.
All cues for January 13, by contrast, are probably similar to those for
January 12 and January 14, making it more difficult to retrieve only what
was happening January 13.
To illustrate this further, one can imagine driving down a country
road and seeing a series of telephone poles going by (Crowder, 1976). If
two poles were built with only a one-foot distance separating them, as the
observer travels further away, it becomes visually more difficult to
identify each individual pole – but rather only the clump of poles can
be seen. However, the greater the spacing between the poles, the greater
the distance becomes at which a clear perceptual distinction between
individual poles can be made. Now
consider what happens when a foreign language teacher begins a unit on
food items. The teacher is likely to stand in front of the class, write
‘fruits’ on the board, and present the vocabulary for a series of
fruits in the target language. By presenting items to be remembered in
this manner, the teacher leaves very little room for distinctions in cues
associated with the different fruits to develop -- as little changes in
either the learner’s mind or environment during the short intervening
time between items. In an experiment designed to measure the effectiveness
of learning items in categories, Roediger (1973) presented participants
with words to memorize, divided into different categories containing four,
five, six, or seven items. Correct recall percentage for items within each
category decreased linearly as a function of how many items the category
contained: .69, .65, .64, .59 Thus, integrating a few vocabulary items
from a variety of different categories may be more effective than
assigning a full category for a week’s vocabulary assignment. Another
dangerous long-term consequence of teaching categorized vocabulary is that
when words are then recalled together, items must compete against one
another until the correct item is distinguished – suppressing all the
other items in the category. This suppression appears to impair how easily
the suppressed item can be subsequently retrieved. To demonstrate this,
Anderson et al. (1994) introduced participants with various categories
(fruit, drink, etc), and a set of common items from that category
(fruit-orange, fruit-banana; drink-scotch, drink-gin; etc). As one item
was practiced (i.e. fruit-orange) recall ability for other items of its
category (i.e. fruit-banana) decreased – though items in other
categories remained unaffected (drink-scotch/gin). This
effect, however, might be avoidable in the foreign language classroom if
the lessons were interspersed with items from different categories, in
order to increase the temporal spacing between presentations of similar
vocabulary items. If students in a language class learn the word for
‘banana’ in a lesson on fruit vocabulary, later when they recall
‘fruit lesson-banana’ they must repress ‘fruit lesson-apple.’
However, if the students learned through the associations ‘week on
reading curious George -banana’ and ‘week on reading snow white –
apple’, even as time goes by these items are more likely to remain
distinct and not interfere with one another. Additionally, some students
may find it easier to make meaningful associations to something natural
like a children’s story than something abstract like a fruit lesson. Overusing
Analytical Abilities
In addition to providing a greater potential for meaningful and
long-term associations to form, orienting curriculums away from massed
practice of abstract grammar rules and categorized vocabulary toward
real-world contexts could improve transfer of classroom knowledge to
communicative fluency. One can easily imagine, or has perhaps experienced,
the frustration of having learned the grammar of a foreign language in a
university, yet being unable to understand and communicate in even
child-level conversations when traveling to a foreign country and being
immersed in that language.
There are likely many reasons for why this poor transfer of
classroom learning to fluency in another language occurs. One explanation
is found in the framework of exemplar theory (Logan, 1988), which proposes
that each time a task is performed, a record of it is made in memory as an
‘exemplar.’ At first, when learners have no exemplars stored, they
will tend to use some sort of algorithm or explicitly memorized
instruction to complete a task. As they perform the task, they begin to
store more exemplars of how to complete the task. Eventually, they will
possess a large enough repository of exemplars that the original algorithm
will fade away and they can base their knowledge of the task based on the
stored exemplars instead. How versatile and able to handle variations of
the task learners will be is then dependant on the diversity of the
underlying exemplars supporting their ability to perform the task.
In a traditional curriculum designed to progressively teach a
sequence of grammar rules, time is first dedicated to teaching an explicit
algorithm used to generate sentences of that rule. Next, a series of
practice exercises are assigned to help solidify memory of that rule;
generally, these exercises are stripped of all variability to try to focus
explicitly on the target rule and resemble the final exam as closely as
possible. This lack of variability allows only a very narrow and unnatural
set of exemplars to be stored. When learners are then immersed in a native
context, they rarely finds themselves in a context close enough to their
classroom-learning environment to be able to communicate fluently only
from their exemplars. Rather, they must use their meta-cognitive skills to
quickly fill in gaps unaccounted for by their underlying exemplars and
rapidly bring themselves up to the fluency needed for regular
communication. This, however, leaves them with an awkward patch-and-fix
development sequence, based more upon which modifications are needed than
accurately observing the language.
Child learners, by contrast, are never taught any explicit grammar
from which to formulate an initial algorithm. Rather, they begin by
recording a large and diverse set of exemplars based on years of hearing
correct examples, slowly putting them together as their cognitive
abilities develop. While this path leads to a slower development of
language ability, there is reason to believe it leads to better long-term
results. Recurrent neural networks programmed to simulate the
learning of grammar are able to successfully generalize the grammar rules
of a language from a corpus of examples when their meta-cognitive
abilities, represented as a variable simulating working memory capacity,
is gradually increased while the computer works through the examples. By
starting the network’s capacities at levels designed to simulate adults,
however, the network no longer synthesizes the optimal rules of the
language’s grammar (Elman, 1993). Besides
allowing them to begin using a language feature before they have stored a
full knowledge about its variation, adults use of more advanced cognitive
functions to accelerate learning of a specific language feature can run
into long-term problems when they generate the wrong algorithms to begin
with. This effect can be most clearly illustrated in the case of
pronunciation. Letters in an alphabet often represent a series of motor
movements. For example, ‘z’ in English usually represents a movement
of the tongue to the front of the mouth and vibration of the voice box --
in German the voice box is less active. As a student begins reading from a
German textbook, they will likely think about words containing a ‘z’
as being pronounced with a voice box vibration; subsequently, English
learners can end up with an accent over emphasizing voice box vibrations
on the letter ‘z’ when speaking German – which often lasts
throughout their life. If students are to learn a vocabulary word for a
weekly unit by reading a textbook, the fastest option they have is to use
knowledge of their first language to generate a guess as to how the word
would be pronounced. If, however, they had regularly heard and learned to
pronounce the word beforehand, by the time they are exposed to the written
form of the word they would not need to rely on external knowledge to
guess how it might sound. Reforming Adult
Language Learning One
belief that may be preventing the reform of classrooms to address
long-term concerns is the notion that learners should do whatever is
necessary to rapidly reach the proficiency demonstrated by a final exam
– and then from there they can work on solidifying their knowledge for
long-term use. As mentioned earlier, patching up holes in one’s language
development caused by using meta-linguistic abilities to accelerate the
learning of specific language features may not lead to the same results as
having learned them through exposure to a variety of examples in different
contexts. Jacoby (1978) further demonstrated how when an item to be
remembered is first learned poorly and then later learned well, it is
remembered worse than if it was simply learned well once. Here
participants first read a word to be remembered in a one study session and
then in a second study session were to perform a task in which they had to
fill in certain letters that were missing from the word they were to
remember – leading to a deeper level of processing for that word than
simply reading it. A second group only did the task filling in the missing
letters, without having a prior reading study session; they exhibited
better recall than the first group. Likewise, once language learners
develops a foundation for speaking a foreign language that allows them to
be understood and pass tests, they will likely only modify that foundation
to meet their immediate needs, rather than starting from scratch to build
a deeper representation. While
the ideal solution for language learners trying to form a solid foundation
for learning a language might be to go to that country and start acting
like a child again for several years until they slowly build up adult-like
communication skills, this is obviously not a practical solution for most
adults. Rather, what is needed is an investigation of how the principles
outlined in this paper that appear in child language learning and are
likely to have an effect on long-term learning in adults can be addressed
without sacrificing the faster rate at which adults proceed initially. One
step could be to re-orient the organizational structure of our textbooks
to increase the amount of time learners are exposed to a target language
feature before they are expected to begin using it and start receiving
extensive corrective feedback. Currently, a few narrowly focused new items
are introduced at the beginning of a chapter, with the goal that a learner
will have mastered them by the end of the chapter – often encouraging
corrective feedback, massed practice and reduced variations in order to be
able to produce the item by the end of the chapter test. Instead, we
should introduce as many items as possible early on, but keep our initial
expectations low as to how much the learner can do with the items. For
example, rather than first spending a week on food, then a week on numbers
and then a week on shopping, the first week all three would be introduced
(e.g. “I want three bananas”) though learners might at this point only
be expected to recognize everything they are to learn. By the second week,
students might be expected to speak and not until the third week write
about and be strictly corrected on all three units; by the end of the
third week, students would be at the same position as if they had learned
each chapter progressively. This triples the spacing each item receives,
gives a longer duration in which learners can see and record exemplars of
the correct forms without corrective feedback and allows content designers
to create more natural contexts with which learners can associate new
items -- as they have a wider range of words to incorporate into their
examples and exercises. Besides
changes within the classroom, many of the concerns raised in this paper
could be addressed by changes in the homework assignments we assign. So
far, trying to get learners to engage in extensive foreign language
practice outside of the classroom by assigning additional practice has
seen limited success -- as the traditional methods of rote memorization
and massed studying require strong motivation and perseverance to
complete. Modern technological developments, however, are opening numerous
possibilities for providing learners with convenient ways to practice a
foreign language outside the classroom in ways that are conducive to
long-term fluency (Purushotma, unpublished manuscript). For example,
rather than massing memorization of vocabulary into a dedicated study
session, a ticker symbol program could display vocabulary at expanding
intervals while waiting for web pages to download. Many junior high
students have both Spanish class, where they practice writing Spanish
sentences for homework and typing class, where they repeatedly type out
English sentences. Rather than assigning additional rote homework,
translations and pronunciation could be built into a Spanish typing
tutorial. Beyond being incorporated into everyday routines, correct
exemplars should be made both more effective and interesting by
incorporating variation and context. Popular computer adventure games have
already been translated into most popular foreign languages; coupled with
voice recognition technologies this offers enormous potential for
recreating natural foreign language contexts. Recently, Simon and Schuster
Corporation released a popular foreign language learning series dubbed
“The Pimsleur Series” based on all-audio prompts for the learner to
imagine situational contexts while driving or walking to class. With
imagination drawing on the same brain structures (Wheeler, Petersen,
Buckner, 2000) and often providing the same results (Eich, 1985) as direct
sensation, the ultimate effectiveness on pronunciation and knowledge
transfer of prescribing this series to students should be of great
interest to researchers and teachers alike. With
the upcoming potential for new innovations in language teaching, it is
especially important for evaluators to understand the long-term effects of
different techniques and to remain cautious of strategies that feature
short-term gains, but might ultimately incur harm; such as over relying on
overt feedback, massing content by categories, reducing variations by
streamlining classroom examples to match the exam, or relying on advanced
analytical skills to generate answers. Rather than dismissing the
miraculous language learning abilities of children as purely biological, I
hope we instead are able to recognize and implement any potentially
beneficial conditions in their learning environments that may be lacking
in that of adults.
References Anderson,
J.R., F.G. Conrad, & Corbett, A.T. (1989). Skill acquisition and the
LISP tutor. Cognitive Science 13, 467-506. Anderson,
M.C., Bjork, R.A., & Bjork, E.L. (1994). Remembering can cause
forgetting: Retrieval dynamics in long-term memory. Journal
of Experimental Psychology: Learning, Memory, and Cognition, 20,
1063-1087. Bjork,
R.A. & E. L. Bjork. (1992). A new theory of disuse and an old theory
of stimulus fluctuation. In A. Healy, S. Kosslyn, & R. Shiffrin
(Eds.), From learning processes to
cognitive processes: Essays in honor of William K. Estes. (Vol.2, pp.
35-67). Hillsdale, NJ: Erlbaum. Bjork,
R.A. (1999). Assessing our own competence: Heuristics and illusions. In D.
Gopher and A. Koriat (Eds.), Attention
and performance XVII. Cognitive Regulation of Performance: Interaction of
Theory and Application (pp. 435-459). Cambridge, MA: MIT Press Brown,
R. (1973). A First Language: The Early Stages. Cambridge, Mass. Harvard
University Press. Chomsky,
N. (1957). Syntactic structures.
Mouton, The Hague. Chomsky,
N. (1975). Reflections on language. Mouton,
The Hague. Crowder,
R.G. (1976). Principles of learning and memory. Hillsdale, NJ: Lawrence Erlbaum
Associates. Eich,
E. (1985). Context, memory, and integrated item/context imagery. Journal
of Experimental Psychology: Learning, Memory, and Cognition, 11,
764-770. Elman,
J.L. (1993). Learning and development in neural networks: The importance
of starting small. Cognition, 48, 71-99. Fahle,
M. & S. Edelman. (1993). Long-term learning in hyperacuity. Vision
Research, 33, 397-412. Goldstein,
M., & C.H. Rittenhouse. (1954). Knowledge of results in the
acquisition and transfer of a gunnery skill. Journal
of Experimental Psychology, 48, 187-196. Jacoby,
L. L. (1978). On interpreting the effects of repetition: Solving a problem
versus remembering a solution. Journal
of Verbal Learning and Verbal Behavior, 17, 649–667. Johnson,
J. & E. Newport. 1989). Critical period effects in second language
learning: The influence of maturational state on the acquisition of
English as a second language. Cognitive
Psychology, 21, 60-99. Lave,
J. (1988). Cognition and Practice:
Mind, Mathematics, and Culture in Everyday Life. Cambridge, England:
Cambridge University Press. Logan,
G.D. (1988). Towards an instance theory of automatization. Psychology Review, 95, 492-527. Lenneberg,
E.H. (1967). Biological Foundations
of Language. John Wiley and Sons Inc. McGeoch, J.A. (1932). Forgetting and the law of disuse. Psychology
Review, 39, 352-370. Pinker, S. (2002). The Blank
Slate: The Modern Denial of Human Nature. Viking Press. Purushotma, R. (2003). Outside the Classroom: Engaging Learners in the
L2 Around the Clock. Unpublished manuscript. Roediger,
H.L. (1973). Inhibition in recall from cueing with recall targets. Journal
of Verbal Learning and Verbal Behavior, 12, 644-657. Roediger,
H. L., III, & M. J. Guynn. (1996). Retrieval processes. In E. L. Bjork
& R. A. Bjork (eds.), Human
Memory (pp. 197-236). San Diego, CA: Academic Press. Schmidt,
R.A. (1988). Motor Control and
Learning: A Behavioral Emphasis. Champaign, Illinois: Human Kinetics
Publishers. Snow,
C. & H.M. Hoefnagel-Höhle. (1978). The critical period for language
acquisition: Evidence from second language learning. Child
Development, 49, 1114-1128. Watkins,
O.C. & J.M. Watkins. (1975). Buildup of proactive inhibition as a
cue-overload effect. Journal of Experimental Psychology: Human Learning and Memory, 104,
442-452. Wheeler M.E., Petersen S.E., & Buckner R.L. (2000). Memory's echo: vivid remembering reactivates sensory-specific cortex. Proc. Natl. Acad. Sci. USA, 97, 11125-11129. Copyright 2003 by the Undergraduate Psychology Journal. |