Search

In computational neurolinguistics, it has been demonstrated that hierarchical models such as recurrent neural network grammars (RNNGs), which jointly generate word sequences and their syntactic structures via the syntactic composition, better explained human brain activity than sequential models such as long short-term memory networks (LSTMs). However, the vanilla RNNG has employed the top-down parsing strategy, which has been pointed out in the psycholinguistics literature as suboptimal especially for head-final/left-branching languages, and alternatively the left-corner parsing strategy has been proposed as the psychologically plausible parsing strategy. In this article, building on this line of inquiry, we investigate not only whether hierarchical models like RNNGs better explain human brain activity than sequential models like LSTMs, but also which parsing strategy is more neurobiologically plausible, by developing a novel fMRI corpus where participants read newspaper articles in a head-final/left-branching language, namely Japanese, through the naturalistic fMRI experiment. The results revealed that left-corner RNNGs outperformed both LSTMs and top-down RNNGs in the left inferior frontal and temporal-parietal regions, suggesting that there are certain brain regions that localize the syntactic composition with the left-corner parsing strategy., Competing Interests: Competing Interests: The authors have declared that no competing interests exist., (© 2023 Massachusetts Institute of Technology.)

One of the central debates in the cognitive science of language has revolved around the nature of human linguistic competence. Whether syntactic competence should be characterized by abstract hierarchical structures or reduced to surface linear strings has been actively debated, but the nature of morphological competence has been insufficiently appreciated despite the parallel question in the cognitive science literature. In this paper, in order to investigate whether morphological competence should be characterized by abstract hierarchical structures, we conducted a crowdsourced acceptability judgment experiment on morphologically complex words and evaluated five computational models of morphological competence against human acceptability judgments: Character Markov Models (Character), Syllable Markov Models (Syllable), Morpheme Markov Models (Morpheme), Hidden Markov Models (HMM), and Probabilistic Context-Free Grammars (PCFG). Our psycholinguistic experimentation and computational modeling demonstrated that "morphous" computational models with morpheme units outperformed "amorphous" computational models without morpheme units and, importantly, PCFG with hierarchical structures most accurately explained human acceptability judgments on several evaluation metrics, especially for morphologically complex words with nested morphological structures. Those results strongly suggest that human morphological competence should be characterized by abstract hierarchical structures internally generated by the grammar, not reduced to surface linear strings externally attested in large corpora., (Copyright © 2020 Oseki and Marantz.)

A central part of knowing a language is the ability to combine basic linguistic units to form complex representations. While our neurobiological understanding of how words combine into larger structures has significantly advanced in recent years, the combinatory operations that build words themselves remain unknown. Are complex words such as tombstone and starlet built with the same mechanisms that construct phrases from words, such as grey stone or bright star? Here we addressed this with two magnetoencephalography (MEG) experiments, which simultaneously varied demands associated with phrasal composition, and the processing of morphological complexity in compound and suffixed nouns. Replicating previous findings, we show that portions of the left anterior temporal lobe (LATL) are engaged in the combination of modifiers and monomorphemic nouns in phrases (e.g., brown rabbit). As regards compounding, we show that semantically transparent compounds (e.g., tombstone) also engage left anterior temporal cortex, though the spatiotemporal details of this effect differed from phrasal composition. Further, when a phrase was constructed from a modifier and a transparent compound (e.g., granite tombstone), the typical LATL phrasal composition response appeared at a delayed latency, which follows if an initial within-word operation (tomb + stone) must take place before the combination of the compound with the preceding modifier (granite + tombstone). In contrast to compounding, suffixation (i.e., star + let) did not engage the LATL in any consistent way, suggesting a distinct processing route. Finally, our results suggest an intriguing generalization that morpho-orthographic complexity that does not recruit the LATL may block the engagement of the LATL in subsequent phrase building. In sum, our findings offer a detailed spatiotemporal characterization of the lowest level combinatory operations that ultimately feed the composition of full sentences., (Copyright © 2018 The Authors. Published by Elsevier Ltd.. All rights reserved.)

We studied the energy transfer processes in the molecular array consisting of pyrene (Py), biphenyl (Ph2), and bisphthalimidethiophene (ImT), (Py-Ph2)2-ImT, during two-color two-laser flash photolysis (2-LFP). The first laser irradiation predominantly generates ImT in the lowest triplet excited state (ImT(T1)) because of the efficient singlet energy transfer from Py in the lowest singlet excited state to ImT and, then, intersystem crossing of ImT. ImT(T1) was excited to the higher triplet excited state (Tn) with the second laser irradiation. Then, the triplet energy was rapidly transferred to Py via a two-step triplet energy transfer (TET) process through Ph2. The efficient generation of Py(T1) was suggested from the nanosecond-picosecond 2-LFP. The back-TET from Py(T1) to ImT was observed for several tens of microseconds after the second laser irradiation. The estimated intramolecular TET rate from Py(T1) to ImT was as slow as 3.1 x 104 s-1. Hence, long-lived Py(T1) was selectively and efficiently produced during the 2-LFP.

Photoinduced intramolecular charge separation (CS) and recombination (CR) processes of the tetrathiophene-substituted benzene dyads with an amide spacer (4T-PhR, R = 4-H (1), 4-CN (2), 3,4-(CN)2 (3), 4-NO2 (4), 3,5-(NO2)2 (5)) in solvents of different polarities were investigated using various fast spectroscopies. It was revealed that the CS rates depend on the ability of the acceptor and solvent polarity. Ultrafast CS with the rate of 5 x 10(12) s(-1) was revealed for 5 in PhCN and MeCN. The ultrafast CS can be attributed to the large electronic coupling matrix element between the donor and the acceptor despite the relative long donor-acceptor distance. The existence of the state with large electron density on the spacer between 14T*-PhR and LUMO should facilitate the CS process in the present dyad system. It was also revealed that the CR rates in these dyads were rather fast because of the enhanced superexchange interaction through the amide spacer.

Photophysical properties of oligo(2,3-thienyleneethynylene)s (nTE, n denotes the number of thiophene rings, n = 2, 3) in benzene were investigated using steady-state, time-resolved fluorescence, and transient absorption spectroscopies. For 2TE, generation of the radiative S2 and nonradiative S1 states was confirmed. Upon excitation, the S2 state was initially generated and deactivated to the S1 state within 10 ps. The S1 state exhibited the transient absorption band at 470 nm, of which the lifetime was estimated to be 5.3 ns. In the case of 3TE, on the other hand, it was revealed that the radiative S1 state with a transient absorption peak at 650 nm was generated upon excitation. The T1 states of nTE were generated from the S1 states. The quantum yields were estimated to be 0.52 and 0.54 for 2TE and 3TE, respectively. Extremely fast reactions in the higher triplet excited state were indicated for both 2TE and 3TE.