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.

DATE

August 1-3, 2022

PLATFORM

The conference will be held online on Gather Town. See this page for more details.

CONFERENCE CHAIRS

Jessie Nixon
Fabian Tomaschek
Harald Baayen

TIMETABLE

The conference timetable is available here as PDF.

Times are in local Tuebingen (Germany) time, CEST (UTC+2).

We have done our best to accommodate the widely varying time zones. The timetable reflects the geographical diversity of our conference delegates from across the globe.

MATERIALS

Conference proceedings are available here as pdf.

REGISTRATION IS OPEN!

There is no registration fee. Participation is free. However, to access the conference, participants will be required to register. Please register here by 28 July.

SPECIAL ISSUE on ERROR-DRIVEN LEARNING IN SPEECH AND LANGUAGE

You may be interested in our upcoming Special Issue to be published in the journal Language, Cognition and Neuroscience: Emergence of speech and language from prediction error: error-driven language models, edited by Jessie Nixon and Fabian Tomaschek. We are finalising the last submissions and anticipate the special issue to be completed in late 2022.

This project has received funding from the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation programme under grant agreement No 742545.

Important Dates

Abstracts due

18 April, 2022

25 April, 2022

Deadline extended! Due to Easter, we have extended the deadline by one week.

(Midnight, GMT -12. That is, submission will stay open as long as it's still 25 April anywhere in the world.)

Notification of acceptance

17 June, 2022

Revised abstracts due

1 July, 2022

Conference

1-3 August, 2022

Call for abstracts: closed

We are inviting experimental, computational 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
learning models such as Hebbian learning, statistical learning, Bayesian learning or distributional learning, especially if a connection can be made to error-driven learning.

CALL FOR ABSTRACTS IS NOW CLOSED

Presentation guidelines

Please see the Platform page for more details about the talk and poster sessions in the gather town space.

Talks

The talks will be held in gather.town lecture hall. Each talk slot is 18+7 = 25 minutes. Presentation time is 18 minutes. After the presentation, 7 minutes are reserved for questions, discussion, and speaker transitions.

Posters

The posters will be presented and discussed in gather.town poster hall. The poster should have the aspect ratio of A0 in portrait mode. Please send the posters to us by July 27, 2022 as a PDF file.

Invited Speakers

Peter Dayan

(Max Planck Institute for Biological Cybernetics)

Generative and discriminative reinforcement learning as model-based and model-free control

Elizabeth Wonnacott

(University of Oxford)

Language input and generalization:
Evidence from language learning experiments with adults and children

Padraic Monaghan

(University of Lancaster and University of Amsterdam)

How multiple and multimodal sources of information support word learning

Jacolien van Rij

(University of Groningen)

Contact

For inquiries please contact:

Jessie Nixon

jessie.nixon@uni-tuebingen.de

Fabian Tomaschek

fabian.tomaschek@uni-tuebingen.de