was more negative-going in the No-Go trials, as previously shown in the
literature (Pfefferbaum et al., 1985). These findings were obtained in both
monolingual and bilingual participants (see Figure 10.1).
In an analysis of trials in which nouns were in first position, there was a
main effect of congruence (F(1,32) = 20.606, p < 0.001) and a congruence by
group interaction (F(1,32) = 7. 2 0 7, p = 0.011). Crucially, follow up tests showed
that there was a main effect of congruence for bilinguals (F(1,15) = 21.459,
p < 0.001), such that the N2 elicited by mismatching nouns in first position
were more negative-going than that elicited by matching nouns. No such
effect of congruence was found when comparing the same conditions in
monolinguals (F(1,17 ) = 2.103, p = 0.165).
In other words, in the conditions where the noun preceded the adjective,
results of the two participant groups (monolinguals versus bilinguals) were
markedly different. While the ERPs of the bilingual group exhibited the same
modulations in the N2 component typically observed in Go/No-Go compari-
sons, the effect was not significant in the ERPs of the monolingual group (see
Figure 10.2).
Discussion
These findings directly support our predictions, since presenting the
noun in first position seems to cancel expectations in the monolingual group
but not in the bilingual group, that is a noun in first position mismatching
the picture triggered a larger N2 than a noun matching the picture in the
Juggling Two Grammars 225Figure 10.1 Averaged waveforms obtained from electrode FCZ for each of the language
groups. Adjectives. Go and No-Go conditions