An Introduction to Applied Linguistics

participants’ use of the passive in English. Kootstra et al. show that this approach 
can be used to prime specific forms of CS without interrupting the natural flow of 
the interaction.
Gestures in a second language
One of the recent trends in psycholinguistics is an enhanced interest in the link 
between verbal and non-verbal aspects of communication. In particular speech-
related gestures (hand movements) have been studied extensively. We know very 
little about the genesis of gestures as part of the communicative system and as part 
of the language production system. As McGafferty and Stam (2008) argue in their 
introduction to their edited volume on gesture, second language acquisition and 
classroom research, the Chomskian focus on language as a separate encapsulated 

---

136 An Introduction to Applied Linguistics
module has not stimulated the integration of the study of gesture as an integrated 
aspect of communication. De Ruijter (2000) presents a version of the Levelt model 
discussed earlier, to which he adds a gesture component by suggesting that just as 
verbal production is done on the basis of a limited set of syllables (‘the syllabary’), 
a similar set can be postulated for gestures (‘the gestuary’). How exactly gestures 
are produced in sync with speech, and how the distribution of labour between 
verbal and non-verbal information is given shape, is as yet unclear.
Several studies have looked at the acquisition and use of gestures in second 
language development (see Gullberg, de Bot and Volterra, 2008 for an overview). 
One of the intriguing questions is whether there is something like a ‘gestural 
foreign accent’, that is, whether language learners may be fluent on the verbal 
level but still show signs of their other languages through gesture and other non-
verbal behaviour. The study of gesture in language development is clearly a field 
in which we have only scratched the surface and one that will become more 
prominent in the years to come.
An interesting development related to the work on gesture addresses bimodal 
bilingualism among hearing individuals who speak one language and sign the 
other. In a series of studies, Emmorey and colleagues (Emmorey, Borinstein, 
Thompson and Gollan, 2008; Pyers and Emmorey, 2008) have shown that the 
language non-selectivity that appears to characterize bilinguals who speak two 
languages also appears to hold for bimodal bilinguals, suggesting that the control 
of the two languages occurs at a relatively abstract level of representation. Thus, 
using two languages whose form is completely distinct does not apparently permit 
bilinguals to keep the two languages more functionally separate.
The cognitive consequences of bilingualism
One of the most exciting developments in recent research on bilingualism 
comes from studies that show that a life as a bilingual confers a set of benefits to 
cognition within the realm of executive function. Although some studies suggest 
that bilinguals suffer relative to monolinguals in the realm of verbal fluency and 
in the speed of lexical access (for example, Gollan, Montoya, Fennema-Notestine 
and Morris, 2005; Gollan, Montoya and Werner, 2002), a now compelling body of 
literature shows that there are benefits of bilingualism on attentional control that 
extend from young bilingual children to young adult bilinguals and to elderly 
bilinguals (for example, Bialystok, 2005; Costa, Hernandez and Sebastián-Gallés, 
2008). Most notably, these benefits are observed in simple cognitive tasks that 
do not explicitly involve language. The data on older bilinguals are particularly 
striking because bilingualism appears to provide a measure of protection against 
the normal effects of cognitive aging (for example, Bialystok, Craik, Klein and 
Viswanathan, 2004). Elderly bilinguals outperform their monolingual counterparts 
on tasks that require them to ignore irrelevant information or to resolve conflict 
in the face of stimulus-response incompatibility. The hypothesis is that a life spent 
negotiating cross-language competition fine tunes a set of cognitive skills that 
benefit the ability to select targeted information, regardless of whether the context 
is linguistic or not. Thus far, the available data are correlational. It will remain to 
be seen in the next period how studies of language processing in bilinguals might 
be related to the observed cognitive consequences to provide a causal account of 
the way in which the resolution of cross-language competition might create these 
changes in cognitive performance.

---

137
Psycholinguistics
Forgetting and Relearning
In this chapter, we have looked at the storage and retrieval of L2 knowledge. A 
growing field of research now deals with the opposite of language acquisition: 
language attrition and language loss (see de Bot (1996) and Hansen (2001) for 
overviews). As discussed above, the level of acquisition of linguistic knowledge is 
crucial in production and perception. Through non-use of a language, the level 
of activation of knowledge in that language decreases, even to the point that that 
knowledge is considered lost. An important point for foreign language teaching 
is how such knowledge can be reactivated again using our knowledge of the 
mechanics of language production and perception. Unfortunately, very little has 
been done on this so far.
Many people assume that words can be lost completely, but is this true? de 
Bot and Stoessel (2000) report on a number of experiments on reactivation of 
language skills. In those studies, they made use of the ‘Savings’ method for 
establishing low levels of activation of items in memory. This method is based 
on the assumption that words, once learned, are never really lost, and that even 
for words that cannot be recognized using traditional test procedures there are 
residues of knowledge that possibly can be used in reactivating these words. The 
procedure is fairly simple: subjects are presented with a list of words; some of 
these have been learned at some point in the past but cannot be remembered 
(‘old’), whereas others have never been seen before (‘new’). The task is to 
translate the words from the second language into the first language. Then, 
the subjects are presented with the words and their translations and are asked 
to learn the translations. Finally, they are tested on those same words again. In 
a number of experiments with Dutch as a second language, and German and 
French as a foreign language, de Bot and Stoessel (2000) showed significant 
savings effects for the old words. Relearning the old words was easier than 
learning completely new words, which indicates that there was indeed residual 
knowledge of the old words, which helped to activate them. These findings can 
be used to help language learners who learned a second language at some point 
in the past reactivate the language they feel they have forgotten. The data show 
that very short relearning activities (presenting words in L1 and their translation 
in L2 for six seconds per pair) lead to high retention scores for such once-learned 
words.
In terms of language maintenance, Harley (1994) discusses a number of case 
studies of Anglophone Canadians who successfully retained their French skills 
learned at school. Among the people interviewed there was unanimity on the 
role of high initial proficiency and continuing contact with the foreign language. 
Motivation is also important, as it leads to learners actively seeking opportunities 
to use the foreign language in different settings.
Implications
Probably the one main implication of this, largely theoretical, chapter is that 
for both language production and language perception two factors determine 
accessibility of linguistic elements, in particular in non-balanced bilinguals and 
language learners: the information must have been acquired and stored, and it 
must be accessible in time. Both production and perception are incredibly fast 
processes, and information that is not readily available will hamper processing of 

---

138 An Introduction to Applied Linguistics
input and output. So far, little attention has been paid to the crucial role of speed 
of processing. An exception to this is the work by Jan Hulstijn and his colleagues 
who are now actually training early learners of a second language to access 
linguistic elements as quickly as possible; and their preliminary results show that 
there is a direct and probably causal relation between speed of processing and 
reading skills. No research has been done so far on productive skills, but it is 
quite likely that specific training of speeded processes will have a positive effect 
on those skills as well. The use of computers in language teaching will allow for 
the use of programs that can train and evaluate such processing. In addition, 
insights from psycholinguistic research should be applied more frequently in the 
area of language testing. Although the language production system in particular 
has been treated as a ‘black box’ whose hidden mechanisms are difficult to 
discern, psycholinguists have been successful in explaining many of its workings. 
This information should inspire language testers to design more sophisticated 
and valid measures of language proficiency in which the input and output sub-
processes are measured along with the global outcome of the system as a whole 
(cf. de Bot, 2000).
Another, maybe less welcome implication is that much of our linguistic 
knowledge is by definition unstable: words and rules are not always equally 
available, availability depends on similarity to the L1 or other languages 
acquired earlier in life, recency and frequency of use and many other factors. 
Having learned a word through translation lists does not mean that such a word 
is then available with all its nuances; rather, only a first connection between a 
form and a meaning are established. It is only through extensive contact with 
that word in a variety of contexts that it will gradually develop a full, close to 
native, set of links.
Psycholinguistic insights also can inform some of the discussions and 
controversies on bilingual education and bilingual upbringing. The most 
important one is that there is no support for the hypothesis that bilingualism 
or learning an additional language at any age will have negative consequences 
on cognitive processing. It is remarkable that there is a long history of negative 
attitudes towards bilingualism but basically no empirical evidence to support the 
assumptions that being or becoming bilingual has negative effects. As Hakuta 
(1986) has argued convincingly, the debate on bilingual education in the USA 
(and similar discussions on bilingual upbringing in educational circles) is based 
much more on attitudes and beliefs than on facts. Apparently, the evidence for 
positive effects of bilingualism at an early age have little effect on such beliefs. It is 
quite likely that the growing internationalization in many parts of the world will 
show the need for multilingualism on all levels. In Canada, the requirement of 
bilingualism for a good work position led to changes in attitudes towards learning 
French. Similarly, the need for proficiency in more languages in the global market 
may change attitudes in other places as well.
New trends in psycholinguistics
In this contribution we have given a description of some of the current trends in 
psycholinguistics, in particular with respect to second language development and 
multilingual processing. We want to conclude with a short summary of what we 
think will be the main trends in the years to come.

---

139
Psycholinguistics
• From monologue to dialogue: Though dialogue is more difficult to study than 
monologue, there will be a move away from taking monologue as the typical 
language production setting. The same may be true for language comprehension, 
though there are more natural settings in which listening happens without 
interaction, such as listening to lectures or sermons, watching television, and 
listening to spoken books. This shift to dialogue also implies a shift in research 
methodology: what are needed are more paradigms in which we can manipulate 
speech in dialogue as in the confederate technique described earlier. Possibly 
different computerized techniques will provide options for this.
• From words to larger units: It follows from the previous point that we will 
have to move from single words to larger units of production and interaction. 
The interactive alignment model assumes an interaction between the use of 
language and the setting in which it is used. In that interaction single words will 
be rare as units, larger units and chunks are more likely to be the building blocks 
in production and perception.
• Integration of verbal and non-verbal aspects of language use: The increase in 
interest in the use of gestures reflects a tendency to a more holistic approach 
in which speech is no longer studied in isolation but as part of communication 
that also involves non-verbal behaviour. This will lead to questions like: what 
is the added value of non-verbal behaviour, and how are different types of 
information combined in production and perception?
• The added value of neuro-imaging: Research in bilingual processing and neuro-
imaging is booming. The main aim is to find out to what extent functional 
differences in language processing between monolinguals and bilinguals is 
linked to specific neural substrates that play a role in multilingual processing. 
So far, the contribution of neuro-imaging data has been limited, but there is 
no doubt that there is great potential in the use of more advanced techniques 
like Transcortical Magnetic Stimulation (TMS) that will allow us not only to 
monitor brain activity, but also manipulate it to find out what roles different 
brain regions play in language processing.
Further Reading
Altarriba, J. and Heredia, R. (eds.). (2008) An Introduction to Bilingualism. Principles