Meeting URL: https://quantling.org/EDLL2022/

Location: Online / Tübingen, Germany

Research Field(s): Cognitive Science; Language Acquisition; Computational Linguistics; Psycholinguistics; General Linguistics

Meeting Description

The Second International Conference on Error-Driven Learning in Language (EDLL 2022).

Error-driven learning models, such as Rescorla and Wagner (1972) and Widrow and Hoff (1960) have had a major influence on many areas of psychology related to human and animal learning. However, research on language learning took a separate path for a long time. Recently, insights from error-driven learning have begun to be applied to a broad range of language phenomena with very promising results. For example, error-driven learning models have addressed questions relating to reading, spoken word comprehension, colour and number acquisition, word learning, first and second language speech sound acquisition, morphological processing, sentence processing, neural correlates of prediction error and more.

The Second International Conference on Error-Driven Learning in Language (EDLL 2022) aims to bring together researchers interested in error-driven learning in speech and language. We would also be interested in submissions from neighbouring fields, such as implicit learning, statistical learning and applied linguistics, especially if a connection can be made to error-driven learning. Work in progress is also welcome.

Due to continued uncertainty regarding travel restrictions, the conference will be held online. The conference will take place from the 1^st-3^rd of August, 2022.

Important Dates

• Abstracts due: 18 April, 2022 Extended deadline! 25 April, 2022 (midnight GMT-12, i.e., as long as it is the 18th of April anywhere in the world)
• Notification of acceptance sent: 17 June, 2022
• Revised abstracts due: 1 July, 2022
• Conference: 1-3 August, 2022

Submission Guidelines

Please submit your anonymous abstracts to https://easychair.org/conferences/?conf=edll2022

Abstracts should be written in English and be no more than one page of text, single line spacing, with an optional second page for tables, figures and references. In order to produce cohesive formatting in the proceedings, we ask that authors use the following formatting. Please use 11-point Arial or Helvetica font for the whole abstract, including the title. The titles should appear in bold font, centred and in sentence case (capitalise only the first letter of the first word and lower case elsewhere). For LaTeX users, you can use the package `helvet` or any other package that produces an Arial or Helvetica-like font.

Abstracts should be anonymous, i.e. author names should not be included on the abstract file itself. Please also ensure that when entering author information in the Easychair system, you enter the authors in the order you would like them to be on the published abstract. Requests for any change in authorship (adding, deleting or changing the order) after the call is closed are possible, but will need to be confirmed by all original authors. 

List of Topics

We are inviting experimental, computation or theoretical abstracts on any topic in error-driven learning of speech or language. Suitable topics include but are not limited to:

The role of error in

• first and second language acquisition
• learning or processing phonetic, morphological, syntactical or lexical information
• sentence processing, syntax and grammar acquisition and processing

as well as

• neural processing of error feedback during speech and language comprehension, production or learning
• the relationship between error-driven learning and information theory
• comparison of error-driven learning with different learning models such as Hebbian learning, statistical learning, Bayesian learning, distributional learning.

Invited Speakers

We are pleased to announce that the following Keynote speakers have confirmed:

• Prof Peter Dayan, Max Planck Institute for Biological Cybernetics
• Dr Elizabeth Wonnacott, University of Oxford
• Prof Padraic Monaghan, University of Lancaster and University of Amsterdam
• Dr Jacolien van Rij, University of Groningen

Publication

We will publish accepted abstracts in an online proceedings. Authors will retain copyright.

Registration

There will be no registration fee. Participation is free. However, participants will be required to register. Registration information coming soon.

Venue

The conference will be held online. Details will follow.

Program Committee

• Jessie Nixon
• Fabian Tomaschek
• Harald Baayen

Contact

All questions about submissions should be e-mailed to

• jessie.nixon@uni-tuebingen.de
• fabian.tomaschek@uni-tuebingen.de