pressing a key, and to withhold response to each F-stimulus (No-Go stimuli).
EEG/ERP studies using a Go/No-Go design report that a negative peak is
elicited at around 200 ms post-stimulus onset, commonly referred to as N2.
This component is larger (more negative) for No-Go stimuli (Bokura et al.,
2001; Eimer, 1993; Falkenstein et al., 1999; Thorpe et al., 1996). Such effects
are present even when participants are not asked to give a motor response,
for example when the response involves counting. Pfefferbaum et al. (1985)
asked participants to count the total number of a class of stimuli (vertical
lines, Go stimuli) and ignore another class of stimuli (horizontal lines, No-Go
stimuli). The experimental results of this counting task showed that the
N2 component elicited by No-Go stimuli was significantly larger than
the N2 component elicited by Go stimuli, just like in experiments requiring a
motor response.
The use of a Go/No-Go design has the advantage of moving the partici-
pants’ focus away from the violation, as participants are actively involved in
a decision task that has seemingly nothing to do with the grammaticality of
the sentence. This is arguably closer to real-life interactions, whereby listen-
ers do not usually think consciously about grammaticality. Although errors,
repetitions, false starts and so on form part of everyday speech, in the con-
scious mind of the listeners these are normally secondary to the communica-
tive function of the utterances they hear.
In the experiment, participants were asked to make a decision when
exposed to a stimulus that met a specific criterion (word/sentence matching
a picture that was previously presented on a given side of the screen, Go
stimuli). They were instructed to ignore stimuli that did not provide an accu-
rate description of the picture (No-Go stimuli). Our study thus resembles the
experiment by Pfefferbaum et al. (1985) in that the response required upon
seeing Go stimuli was not a motor response. Following the results reported
by Pfefferbaum et al. (1985), we expected that No-Go stimuli should elicit a
significantly larger N2 component compared to Go stimuli. However, this
might not be the case for both participant groups in the present experiment,
which involved bilingual and monolingual participants.
At this point, a detailed account of the experimental design is in order:
first, we aimed at testing sentence comprehension and opted for a picture –
sentence relatedness task. We presented participants with a picture either to
the left or to the right side of the screen (e.g. a picture of a book drawn with
red lines). A sentence describing the position of the picture on the screen then
followed, for example The red book was on the left. Participants were asked to
indicate whether the position mentioned in the sentence matched the posi-
tion of the object, by pressing a key. For example, if the sentence mentioned
left, and the object indeed appeared on the left side of the screen, participants
were asked to press a designated ‘yes’ key. If, on the other hand, the sentence
mentioned left, but the object had appeared on the right side of the screen,
they were instructed to press a designated ‘no’ keyboard key.
220 Part 5: The Bilingual Brain