Advanced

Anticipating morphological and syntactic structures : investigating the pre-activation negativity

Söderström, Pelle LU ; Horne, Merle LU ; Mannfolk, Peter LU ; van Westen, Danielle and Roll, Mikael LU (2017) Cognitive Neuroscience Society Annual Meeting 2017
Abstract
It is known that listeners can predict upcoming words based on constraining contexts (e.g. DeLong et al., 2005). In a recent study, we proposed a left frontal brain potential, the pre-activation negativity, PrAN (Söderström et al., 2016), thought to reflect pre-activation of expected word continuations. Time-locked to word-initial fragments, PrAN’s amplitude was found to increase in a 136-280 ms time window as the number of possible continuations decreased, suggesting that PrAN increased with increased predictive certainty about a word’s ending. In the present study, we tested whether a similar effect could be found for pre-activation of expected syntactic structures. In Swedish, intonation is used to signal whether an unfolding embedded... (More)
It is known that listeners can predict upcoming words based on constraining contexts (e.g. DeLong et al., 2005). In a recent study, we proposed a left frontal brain potential, the pre-activation negativity, PrAN (Söderström et al., 2016), thought to reflect pre-activation of expected word continuations. Time-locked to word-initial fragments, PrAN’s amplitude was found to increase in a 136-280 ms time window as the number of possible continuations decreased, suggesting that PrAN increased with increased predictive certainty about a word’s ending. In the present study, we tested whether a similar effect could be found for pre-activation of expected syntactic structures. In Swedish, intonation is used to signal whether an unfolding embedded clause is a main or subordinate clause. Specifically, a clause-initial word with a low boundary tone cues only subordinate clause structure. Conversely, a corresponding high tone signals that any kind of embedded main clause structure may follow, i.e. it cues a more open set of structures. Test participants listened to complex sentences and judged the word order of the verb (V) and negation (NEG) after the boundary tone as quickly as possible (NEG–V word order occurs in subordinate clauses and V–NEG in main clauses). ERPs were time-locked to the tone-bearing syllable. A repeated-measures ANOVA showed a negativity in left anterior electrodes at 136-280 ms for low initial boundary tones, which cue only subordinate clauses. We propose that this effect is a PrAN, but that it here reflects pre-activation of syntactic structures rather than possible word endings. (Less)
Please use this url to cite or link to this publication:
author
organization
publishing date
type
Contribution to conference
publication status
published
subject
keywords
ERP, fMRI, Morphology, Syntax, Prediction, Pre-activation
conference name
Cognitive Neuroscience Society Annual Meeting 2017
language
English
LU publication?
yes
id
42606551-62b7-4b4d-9788-1a940cdd8ba3
date added to LUP
2017-04-04 12:26:49
date last changed
2017-04-04 12:52:06
@misc{42606551-62b7-4b4d-9788-1a940cdd8ba3,
  abstract     = {It is known that listeners can predict upcoming words based on constraining contexts (e.g. DeLong et al., 2005). In a recent study, we proposed a left frontal brain potential, the pre-activation negativity, PrAN (Söderström et al., 2016), thought to reflect pre-activation of expected word continuations. Time-locked to word-initial fragments, PrAN’s amplitude was found to increase in a 136-280 ms time window as the number of possible continuations decreased, suggesting that PrAN increased with increased predictive certainty about a word’s ending. In the present study, we tested whether a similar effect could be found for pre-activation of expected syntactic structures. In Swedish, intonation is used to signal whether an unfolding embedded clause is a main or subordinate clause. Specifically, a clause-initial word with a low boundary tone cues only subordinate clause structure. Conversely, a corresponding high tone signals that any kind of embedded main clause structure may follow, i.e. it cues a more open set of structures. Test participants listened to complex sentences and judged the word order of the verb (V) and negation (NEG) after the boundary tone as quickly as possible (NEG–V word order occurs in subordinate clauses and V–NEG in main clauses). ERPs were time-locked to the tone-bearing syllable. A repeated-measures ANOVA showed a negativity in left anterior electrodes at 136-280 ms for low initial boundary tones, which cue only subordinate clauses. We propose that this effect is a PrAN, but that it here reflects pre-activation of syntactic structures rather than possible word endings.},
  author       = {Söderström, Pelle and Horne, Merle and Mannfolk, Peter and van Westen, Danielle and Roll, Mikael},
  keyword      = {ERP,fMRI,Morphology,Syntax,Prediction,Pre-activation},
  language     = {eng},
  month        = {03},
  title        = {Anticipating morphological and syntactic structures : investigating the pre-activation negativity},
  year         = {2017},
}