FeedBook an interactive workbook for English foreign language teaching


The FeedBook is an intelligent electronic workbook is designed to accompany the English teaching at the German Gymnasium. It is developed by computational linguists at the University of Tübingen in collaboration with the publisher Diesterweg on the basis of a current workbook for 7th grade English classes, Camden Town.

The main goal of the FeedBook is to develop a system that provides immediate, individualized feedback to English language learners working on a range of tasks. To provide this kind of scaffolding feedback, state-of-the-art computational analyses of meaning and form are implemented and further developed.

FeedBook was tested for an entire school year between September 2018 and August 2019. First analyses show that students who received specific feedback benefit significantly more from using the system than students who don't receive specific feedback. A detailed analysis is published in this article.


The University of Tübingen is one of the oldest universities in Europe and one out of only 11 excellence universities in Germany. It hosts the project and provides financial resources.

Diesterweg is a publisher of educational materials. They provide the materials used as a basis for the digital form.

The development of the FeedBook is funded through the Knowledge Transfer Project (T1) as part of the of the collaborative research center 833 "The Construction of Meaning". The computational analyses implemented in the FeedBook are based on research results of the SFB Project A4 (Comparing Meaning in Context) on state-of-the-art meaning assessment of learner answers.

Project progress

Project Start

Start of the development of the system prototype.

July 2016

Prototype Rollout

Rollout of prototype to five schools, start of data collection phase.

October 2016

Update 1.1

First major system update focusing on mobile and touchscreen optimization.

November 2016

Update 1.2

Second major system update introducing among other features an intelligent feedback memory system.

December 2016

Update 1.3

Third big system update with lots of system polishing and the introduction of a statistics module.

February 2017

Update 1.4

Fourth big system update with the introduction of a mosaic design and improved automatic feedback.

February 2017

Update 1.4.1

System update with reworked navigation bar and options for automatic system suggestions.

April 2017

Update 1.5.0

System update with first automatic input analysis and many functional improvements.

June 2017

Update 1.5.1

Update with 3rd generation of student perspective and new functions for adding new tasks to the system.

July 2017

Update 2.0

First version of interactive, direct, intelligent feedback.

September 2017

Update 2.3

Form feedback for all grammar topics of the seventh grade..

Mai 2018

Update 2.4

Extension of meaning feedback for semi-open tasks, real-time feedback for spelling errors, implementation of a learner model and automatic sequencing of exercises.

August 2018

Start of Study

Begin of a 1-year study to test the effectiveness of FeedBook in an authentic context..

September 2018

Completion of Study

Completion of Study after 1 year of testing..

July 2019

First Analysis

Publication of article showing student using the FeedBook with feedback learn significantly better. [link]

August 2019

Professor of Computational Linguistics and head of the Theoretical Computational Linguistics group at the Department of Linguistics of the University of Tübingen.​

Researcher in the domain of information structure in the knowledge center 833 at the University of Tübingen.

Responsible for learner model in FeedBook and PhD student in Computational Linguistics at the University of Tübingen.

Feedback expert at FeedBook and researcher in the domain of content assessment at the University of Tübingen.

Teacher and author providing a pedagogical perspective.