Limits...
Item parameters dissociate between expectation formats: a regression analysis of time-frequency decomposed EEG data.

Monsalve IF, Pérez A, Molinaro N - Front Psychol (2014)

Bottom Line: During language comprehension, semantic contextual information is used to generate expectations about upcoming items.This has been commonly studied through the N400 event-related potential (ERP), as a measure of facilitated lexical retrieval.On the other hand, gamma-power (~50-70 Hz) was also modulated by predictability of the item in all conditions, which is interpreted as an index of a similar "matching" sub-step for both types of contexts, binding an expected representation and the external input.

View Article: PubMed Central - PubMed

Affiliation: BCBL. Basque Center on Cognition, Brain and Language Donostia, Spain.

ABSTRACT
During language comprehension, semantic contextual information is used to generate expectations about upcoming items. This has been commonly studied through the N400 event-related potential (ERP), as a measure of facilitated lexical retrieval. However, the associative relationships in multi-word expressions (MWE) may enable the generation of a categorical expectation, leading to lexical retrieval before target word onset. Processing of the target word would thus reflect a target-identification mechanism, possibly indexed by a P3 ERP component. However, given their time overlap (200-500 ms post-stimulus onset), differentiating between N400/P3 ERP responses (averaged over multiple linguistically variable trials) is problematic. In the present study, we analyzed EEG data from a previous experiment, which compared ERP responses to highly expected words that were placed either in a MWE or a regular non-fixed compositional context, and to low predictability controls. We focused on oscillatory dynamics and regression analyses, in order to dissociate between the two contexts by modeling the electrophysiological response as a function of item-level parameters. A significant interaction between word position and condition was found in the regression model for power in a theta range (~7-9 Hz), providing evidence for the presence of qualitative differences between conditions. Power levels within this band were lower for MWE than compositional contexts when the target word appeared later on in the sentence, confirming that in the former lexical retrieval would have taken place before word onset. On the other hand, gamma-power (~50-70 Hz) was also modulated by predictability of the item in all conditions, which is interpreted as an index of a similar "matching" sub-step for both types of contexts, binding an expected representation and the external input.

No MeSH data available.


Related in: MedlinePlus

Temporal evolution of power between 0 and 0.6 s post TW onset over selected channel-frequency windows: (A) alpha/theta, (B) mid gamma band. Error bars indicate standard errors around the mean, for illustrative purposes; dotted lines mark analyzed time-window. Although the control condition was not analyzed in the low frequency contrast, it is included in the figure as a reference.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4129372&req=5

Figure 2: Temporal evolution of power between 0 and 0.6 s post TW onset over selected channel-frequency windows: (A) alpha/theta, (B) mid gamma band. Error bars indicate standard errors around the mean, for illustrative purposes; dotted lines mark analyzed time-window. Although the control condition was not analyzed in the low frequency contrast, it is included in the figure as a reference.

Mentions: At the boundary between alpha and theta bands (7–9 Hz), from 260–420 ms over left frontal and temporal electrodes (F7, F3, FC5, T7). Power over the selected interval (see Figures 1, 2) was lower for MWE (mean: 0.97; SE: 0.03) than SEM (mean: 1.08; SE: 0.04).


Item parameters dissociate between expectation formats: a regression analysis of time-frequency decomposed EEG data.

Monsalve IF, Pérez A, Molinaro N - Front Psychol (2014)

Temporal evolution of power between 0 and 0.6 s post TW onset over selected channel-frequency windows: (A) alpha/theta, (B) mid gamma band. Error bars indicate standard errors around the mean, for illustrative purposes; dotted lines mark analyzed time-window. Although the control condition was not analyzed in the low frequency contrast, it is included in the figure as a reference.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4129372&req=5

Figure 2: Temporal evolution of power between 0 and 0.6 s post TW onset over selected channel-frequency windows: (A) alpha/theta, (B) mid gamma band. Error bars indicate standard errors around the mean, for illustrative purposes; dotted lines mark analyzed time-window. Although the control condition was not analyzed in the low frequency contrast, it is included in the figure as a reference.
Mentions: At the boundary between alpha and theta bands (7–9 Hz), from 260–420 ms over left frontal and temporal electrodes (F7, F3, FC5, T7). Power over the selected interval (see Figures 1, 2) was lower for MWE (mean: 0.97; SE: 0.03) than SEM (mean: 1.08; SE: 0.04).

Bottom Line: During language comprehension, semantic contextual information is used to generate expectations about upcoming items.This has been commonly studied through the N400 event-related potential (ERP), as a measure of facilitated lexical retrieval.On the other hand, gamma-power (~50-70 Hz) was also modulated by predictability of the item in all conditions, which is interpreted as an index of a similar "matching" sub-step for both types of contexts, binding an expected representation and the external input.

View Article: PubMed Central - PubMed

Affiliation: BCBL. Basque Center on Cognition, Brain and Language Donostia, Spain.

ABSTRACT
During language comprehension, semantic contextual information is used to generate expectations about upcoming items. This has been commonly studied through the N400 event-related potential (ERP), as a measure of facilitated lexical retrieval. However, the associative relationships in multi-word expressions (MWE) may enable the generation of a categorical expectation, leading to lexical retrieval before target word onset. Processing of the target word would thus reflect a target-identification mechanism, possibly indexed by a P3 ERP component. However, given their time overlap (200-500 ms post-stimulus onset), differentiating between N400/P3 ERP responses (averaged over multiple linguistically variable trials) is problematic. In the present study, we analyzed EEG data from a previous experiment, which compared ERP responses to highly expected words that were placed either in a MWE or a regular non-fixed compositional context, and to low predictability controls. We focused on oscillatory dynamics and regression analyses, in order to dissociate between the two contexts by modeling the electrophysiological response as a function of item-level parameters. A significant interaction between word position and condition was found in the regression model for power in a theta range (~7-9 Hz), providing evidence for the presence of qualitative differences between conditions. Power levels within this band were lower for MWE than compositional contexts when the target word appeared later on in the sentence, confirming that in the former lexical retrieval would have taken place before word onset. On the other hand, gamma-power (~50-70 Hz) was also modulated by predictability of the item in all conditions, which is interpreted as an index of a similar "matching" sub-step for both types of contexts, binding an expected representation and the external input.

No MeSH data available.


Related in: MedlinePlus