g blue car
h car blue
i book
j car
This resulted in a total of 480 trials overall. Finally, each stimulus was incor-
porated into a sentence indicating the position of the object on the screen
(left or right), as in ‘the blue book was on the right’.
Experimental task
By exposing our groups to an ungrammatical sentence and to its gram-
matical counterpart we follow the tradition of studies that have examined
the processing of syntactic violations in bilinguals (Hahne, 2001; Hahne &
Friederici, 2001; Weber-Fox & Neville, 1996). Similarly to the above studies,
we aimed to determine whether the violations (ungrammatical sentences)
trigger different patterns of brain activity in bilinguals than they do in
monolinguals. This difference would then be attributed to the knowledge of
an extra language by the bilingual participants.
There are, however, a number of characteristics that differentiate our
study from previous ones. First, in our study, the sentences that contained a
violation, if translated word-for-word while maintaining word order, would
be grammatical in the bilinguals’ other language. This, in conjunction with
participant selection that ensured high proficiency and early age of acquisi-
tion, means that any differences in brain activity patterns between the two
groups could be attributed to the knowledge of two conflicting grammars by
the participants in the bilingual group. By testing highly proficient bilinguals
that were exposed to both languages during childhood, we made sure that
low proficiency or late exposure to the second language could not be the
reasons for any deviation from the monolingual pattern. If the experimental
results were to reveal a difference between the two groups, this would indi-
cate parallel activation of the Welsh grammar in the bilingual group, when
processing English sentences.
Second, our study presents a departure from previous research in that
syntactic violation was indexed by a different marker, the N2. Instead of
relying on the traditionally used LAN and P600 components, we combined
the syntactic violation paradigm with a Go/No-Go design, as exemplified
below. The N2 component, a marker of response inhibition, was used to
measure the inhibition associated with No-Go trials, indexing violation pro-
cessing only indirectly. In a Go/No-Go design, there are two classes of stimu-
lus, one class which requires a response (Go stimuli) and a second class which
does not (No-Go stimuli). For example, in a simple form of the task,
Falkenstein et al. (1999) used two letters (F and J) as stimuli presented on a
screen, and asked participants to respond to each J-stimulus (Go stimuli) by
Juggling Two Grammars 219