How working memory capacity modulates the time course of semantic integration at sentence and discourse level

Conventionality determines the time course of indirect replies comprehension: An ERP study

Indirect language comprehension requires decoding both the literal meaning and the intended meaning of an utterance, in which pragmatic inference is involved. This study tests the role of conventionality in the time course of indirect reply processing by comparing conventional and non-conventional indirect replies with direct reply, respectively. We constructed discourses which consist of a context and a dialogue with one question (e.g., May I buy a necklace for you) and one reply (e.g., I really have too many). The reply utterance was segmented into three phrases and presented orderly for EEG recording, e.g., with the subject as the first phrase (e.g., I), the adverbial as the second phrase (e.g., really), and the predicate as the third phrase (e.g., have too many). Our results showed that for conventional indirect replies, the second phrase elicited a larger anterior negativity, and the third phrase elicited a larger anterior N400 compared with those in direct replies. By contrast, for the non-conventional indirect reply, only the third phrase elicited a larger late negativity than the direct replies. These findings suggest that conventionality determines the time course of the pragmatic inferences for the most relevant interpretation during indirect replies comprehension.

The influences of working memory updating on word association effects and thematic role assignment during sentence processing

The current study investigated how individual variability in working memory (WM) updating affects real-time processing of thematic role assignment and word association during sentence reading comprehension when ERPs were recorded. By adopting a factorial design, four types of sentences were formed by crossing word association and role assignment as independent variables. The results indicated that associated words evoked a smaller N400 effect but a larger P600 effect than unassociated words in the high WM group, whereas no word association effect was found in the low WM group. In contrast, role reversal elicited larger N400 effects for both groups. These results suggest that individual differences in WM updating influenced whether and how readers retrieved and integrated the associated word in whole sentences but did not influence the online assignment of thematic roles during sentence reading. Individuals with high WM updating, in contrast to those with low WM updating, were good at making use of word-associated information provided by the preceding context in current processing.

The effect of aging on context use and reliance on context in speech: A behavioral experiment with Repeat–Recall Test

Purpose

To elucidate how aging would affect the extent of semantic context use and the reliance on semantic context measured with the Repeat–Recall Test (RRT).

Methods

A younger adult group (YA) aged between 18 and 25 and an older adult group (OA) aged between 50 and 65 were recruited. Participants from both the groups performed RRT: sentence repeat and delayed recall tasks, and subjective listening effort and noise tolerable time, under two noise types and seven signal-to-noise ratios (SNR). Performance–Intensity curves were fitted. The performance in SRT50 and SRT75 was predicted.

Results

For the repeat task, the OA group used more semantic context and relied more on semantic context than the YA group. For the recall task, OA used less semantic context but relied more on context than the YA group. Age did not affect the subjective listening effort but significantly affected noise tolerable time. Participants in both age groups could use more context in SRT75 than SRT50 on four tasks of RRT. Under the same SRT, however, the YA group could use more context in repeat and recall tasks than the OA group.

Conclusion

Age affected the use and reliance of semantic context. Even though the OA group used more context in speech recognition, they failed in speech information maintenance (recall) even with the help of semantic context. The OA group relied more on context while performing repeat and recall tasks. The amount of context used was also influenced by SRT.

Eye-closure effects and the influence of short-term storage and processing capacity on episodic memory

Two experiments investigated differences in short-term storage and processing capacity on the magnitude of eye-closure effects on episodic memory. Experiment 1 compared individuals with high (vs. low) forward and backward spans in the free-recall of words retrieved under both eyes closed and open conditions. Main effects of both forward and backward span capacity (greater recall for the high span group) and eye-closure (higher recall with eyes closed) were found. Eye-closure was also associated with more "remember" responses. Experiment 2 compared individuals with high (vs. low) reading spans and found both main effects for reading span and eye-closure (greater recall for the high span group and with eyes closed). Remember responses were associated with both high reading span and eye-closure. The absence of interactions is discussed in terms of explanations of eye-closure effects that differentiate between modality-general and modality-specific processes.

Neural correlates of contributions of interpreting training to second language semantic integration

How interpreting training may enhance semantic integration in general language processing (instead of language processing during interpreting) was not clear, and its neural correlates remained unexplored. Two groups of university students with different amounts of interpreting training experience (more-IE and less-IE) performed a second language (L2) reading task, in which the target sentence ended with three types of lateralized words: expected words, words of related semantic violation (related violation), and words of unrelated semantic violation (unrelated violation). N400 responses elicited by the final words were used to index semantic integration. Two major findings suggested potential contributions of interpreting training to L2 semantic integration. First, the more-IE group exhibited smaller N400 amplitudes than the less-IE group for expected words, and no group differences were obtained for the two violations. Second, N400 amplitudes for related violations were generally smaller than those for unrelated violations in the right visual field, and this difference was more fully captured (more broadly distributed over the scalp) in the more-IE group than in the less-IE group. Therefore, smaller N400 amplitudes for expected words but not for violations, and broader scalp distribution of the differences between related and unrelated violations constitute the neural correlates for the potential contributions of interpreting training to general L2 semantic integration. The results also suggest possible connections between different types of language experience.