Frontiers in Artificial Intelligence and the Guest Editors
Michael Zock, Simon De Deyne, Massimo Stella, and Vito Pirrelli
are curating an Article Collection entitled

"The Mental Lexicon, Blueprint of the Dictionaries of Tomorrow: Cognitive Aspects of the Lexicon".

We welcome your papers to our peer-reviewed Article Collection. Papers can be original research, reviews, or perspectives, among other article types.

Submission Deadlines:
28 February 2021 Abstract
30 June 2021 Manuscript

More information: https://www.frontiersin.org/research-topics/17890/the-mental-lexicon-blueprint-of-the-dictionaries-of-tomorrow-cognitive-aspects-of-the-lexicon

The mental lexicon, blueprint of the dictionaries of tomorrow: Cognitive Aspects of the Lexicon


Whenever we listen, speak, read, or write, we always use words, the shorthand labels of more or less well-specified thoughts. Obviously, words are important. Yet, despite their ubiquity, we still need to clarify how they are learned, stored, and organized to become accessible on the fly.

In human language production, we encounter either of the following two situations: (a) one where everything flows, words popping up one after another as water in a fountain spring, and another (b) where discourse is hampered. Being blocked somewhere along the road, the author is forced to look deliberately and often painstakingly for a specific, possibly known word (tip of the tongue problem). This is the moment to reach for a dictionary. Alas, most of them have been built with the reader in mind. Yet, alphabetically-organized dictionaries are only of limited use for the language producer. While allowing to check spelling, grammar (gender, preposition), or the definition of a word, they will not reveal the form expressing the concept an author has in mind (conceptual input). Yet, this is our concern.

Of course, there are a number of resources designed for word-finding: thesauri, wordnets, and specialized dictionaries (bilingual-, reverse-, synonym- and collocation dictionaries) being examples in case. While they are excellent with respect to coverage (number of word forms stored), outperforming even humans, they cannot rival the human brain when it comes to speed, flexibility, and accessibility, i.e., the use of partial information. Hence, the question: can we use (certain features of) the human mind as a model, or, put differently, can we use the mental lexicon (ML) as a blueprint for the dictionaries of tomorrow?

Of course, more work is needed to get a clearer picture concerning our claim, and this is the goal of this Research Topic which is devoted to two different issues in terms of scope:

Take a look at the lexicon from a cognitive point of view. This is a broad view, taken at CogALex (https://sites.google.com/view/cogalex-2020/home) a workshop devoted to the Cognitive Aspects of the Lexicon;
Consider a more specific view, namely, the development of an ecosystem to support word-finding (word access at the moment of speaking or writing)


This being so, we welcome articles addressing the problems raised in the CogALex workshop series as well as the following points:

- Given our goal, and to be fruitful for our discussion, what shall we put behind the term ‘mental lexicon’?
- How can we reconcile the fact that words are decomposed, hence subsymbolic, in the human brain, while humans can only provide and interpret symbolic information, which is why we have words in the lexicon?
- How to build a representative map of the mental lexicon for a given user group?
- Which corpora (knowledge graphs, BabelNet, ConceptNet) and which combinations have the best potential to yield a representative encyclopedic lexicon?
- How to achieve the right mix between encyclopedic and (collective) episodic knowledge?
- How to ‘dynamize’ the lexicon? The weight of the links between words is not frozen. During communication, our focus shifts all the time. Likewise, any event (news) may have an impact on the evocation potential of a given word.
-Which lexical resources are relevant to support word finding? What layers of Levelt’s lexical access model are relevant, and to what extent do they need to be worked out? While we have a good understanding of the time course of lexical access, we still do lack details concerning the components at the various levels. For example, what are the specificities at the conceptual level? Which features typically get activated to evoke concepts like ‘justice’, ‘nationality’, ‘panda’?
-How to combine existing lexical resources to allow for their joined usage?

In sum, in this Research Topic, we welcome papers devoted to the mental lexicon, its electronic emulation in the form of an electronic resource likely to support humans in wordfinding (word access), and other cognitive aspects of the lexicon. Both theoretical/conceptual and empirical contributions are welcome.

In order to be practical, we suggest to send us by the 28th of February 2021 an abstract, so that we can tell you whether it complies with the goals of this specific issue. The fully written document is expected by 30 June 2021.