1. Spatiotemporal dynamics of complex word processing in the brain
Alina Leminen, PhD, Cognitive Brain Research Unit, Cognitive Science, University of Helsinki

The spatiotemporal dynamics of the neural processing of complex words, consisting of several meaningful elements (e.g., homeless, houses, darkness) are still poorly understood. I will present two EEG/MEG studies which investigated word complexity in the brain. In the first EEG/MEG study we investigated cortical correlates of the processing of spoken inflected (house+s) and derived (dark+ness) forms. The participants were instructed to judge the acceptability of each stimulus. Derived words elicited larger sensor-level and neural source responses than inflected words ~100 ms after the suffix (e.g., --s or -ness) onset. At ~200 ms after the suffix onset, inflected words elicited larger responses than other words. The early processing stage was suggested to reflect automatic mapping of incoming acoustic information onto stored representations, whereas later compositional processes at the morphosyntactic-semantic level. In the second study (EEG), the participants were presented with derived and inflected words in a passive oddball paradigm. Derived words elicited larger mismatch negativity (MMN) responses and neural source activity than inflected words at ~100 msec after the suffix onset. These results suggest stronger unified cortical memory circuits for derived than inflected words, which are activated automatically, without participants’ focused attention on the stimuli. Overall, the results show that during listening, inflected words are decomposed into their lexical elements (e.g., house and –s), whereas derived words activate full-form representations (darkness). Analysis of cortical sources underlying these processes suggest the predominant role of left perysilvian cortices, which exhibit activation dynamics 100-150 ms after the information is available at the sensory input.


2. Cortical mechanisms of rapid novel word learning and memory trace formation: EEG evidence
Lilli Kimppa, Cognitive Brain Research Unit, Inst. for Behavioural Sciences, University of Helsinki

How does the human brain acquire new words from spoken input? It has been shown that real words elicit stronger ERPs than meaningless word-like phonological stimuli (“pseudowords”), which reflects the neural activation of lexical memory traces for words. Crucially, some studies suggest that only within ~15 minutes of repetition, the ERP response to pseudowords increases to the level of real words, likely reflecting automatic rapid formation of new memory representations for novel speech content. In the current study, we addressed the reliability of these results and investigated if the rapid perceptual learning of novel wordforms depends on their phonology and on the level of attention paid to the input.

We recorded ERPs from 22 adults for spoken CVCV word stimuli in three categories: real words, pseudowords with native phonology and non-native pseudowords, matched for their acoustic properties. Subjects were listening to multiple trials (N=150) of the stimuli both in a passive condition and an attentive condition with a combined target detection and memory task.

We found that, in the passive condition a negative-going ERP peak at ~50 ms after the disambiguation point (second syllable onset) was strongest for the real words and weakest for the non-native pseudowords. Further, while there was a habituation in this early word response, ERPs to both pseudoword types did not show such habituation effects. Crucially, the ERP response to pseudowords with native phonology increased during the short recording session to resemble the initial response magnitude of words, while non-native stimuli did not reach the same level. A similar pattern of activation dynamics was seen in a later time window, at ~150 ms, in both attend and non-attend passive conditions.

The results suggest that rapid formation of novel memory traces for phonologically plausible stimuli does take place automatically in both passive and attentionally demanding conditions, whereas the role of attention appears to be in successful categorization of the spoken word-like input. Memory trace formation for words with non-native phonology, however, does not seem to occur in the passive exposure with the same efficiency as that for phonologically native spoken input, at least with the amount of repetition used in this experiment. The results suggest phonetically restricted, attention-modulated rapid perceptual learning and plastic changes in the brain reflected by activation pattern changes for novel words. They also imply that rapid mapping phenomena are most expressed for spoken stimuli with native phonology, benefiting from pre-existing perception-articulation links in the brain, and suggest different neural strategies for the learning of new words in the native and non-native language.