You are viewing the SNL 2017 Archive Website. For the latest information, see the Current Website.

Poster C74, Thursday, November 9, 10:00 – 11:15 am, Harborview and Loch Raven Ballrooms

Tonal triggers to word-level and sentence-level predictions

Pelle Soderstrom1, Merle Horne1, Mikael Roll1;1Lund University

The present contribution describes two studies of anticipatory processing at two different levels of language structure. The investigations take advantage of predictive prosodic cues occurring in Swedish to study the effect of prediction triggered by a tonal cue before the predicted structure has been heard. The neural underpinnings of both word-internal and sentence-level predictions were investigated using concurrent ERP/fMRI paradigms in two separate experiments. We propose that both types of prediction are indexed by an ERP component, the ‘pre-activation negativity’ (PrAN), which has previously been linked to predictive certainty (Söderström et al. 2016). In the word-level experiment, participants listened to sentences with critical words in which the tone (low or high) of a word’s stem is predictive with regard to suffixes attached to the stem. Critical words were nouns which had either singular or plural endings. Low stem tones cue singular suffixes while high stem tones cue plural suffixes in Central Swedish. The high tone is also a cue for compound word structure. Thus, high stem tones always cue a larger number of possible continuations. This is also the case for high tones at the beginning of clauses, where they can cue different kinds of structures (statements, questions) as opposed to low clause-initial tones which only cue subordinate clause structure. In the second experiment, we therefore took advantage of sentence-level tones that cue either an upcoming main or subordinate clause structure in embedded clauses. ERPs and event-related fMRI were time-locked to both predictive cue onset (i.e. the tone) and onset of the predicted structure. Results show that more predictive cues, both at word level and sentence level, gave rise to a left-lateralised and anterior pre-activation negativity in the ERPs, at around 100 ms after cue onset. At the word level, predictive cues also increased activity in the left inferior parietal lobe and temporal lobe, while the pre-activation negativity for more predictive cues at the sentence level correlated with activity in left inferior frontal gyrus (BA44) and anterior insula. Disconfirmed predictions led to P600 effects in both studies, with an additional N400 found for unexpected suffixes. We suggest that the pre-activation negativity reflects the pre-activation of strongly expected linguistic information at both the word and sentence level.

Topic Area: Perception: Speech Perception and Audiovisual Integration

Back to Poster Schedule