Turning a workshop into a digital feedback session: adaptive research practices in the era of social distancing and remote work

Once upon time the world was different. Everyday was spent at the university campus, where the GramAdapt team was preparing for the launch of our project workshop to be held on March 25-27. The guest list was confirmed, as well as the program, flights and accommodation for international guests were finalised. We were making the final sprint to touch up the research materials to be presented. Then on March the 18th, the state of emergency was implemented in Finland, and travel and gathering restrictions were introduced in many countries due to the spreading of the coronavirus epidemic. This meant we all had to change the way we do work. In our case, this meant that the workshop was cancelled.

Luckily for us, the nature of the GramAdapt research is not so heavily affected by voluntary social distancing, or working remotely. But we still found ourselves having to improvise. The March workshop represented a milestone for our team’s workflow. We planned it as an opportunity to showcase the state of our research and, most importantly,  to get feedback from the participants, a wide range of researchers in the language sciences partly from abroad and partly from the University of Helsinki.  Given the circumstances, this was no longer possible. But the project must go on, and we would still want to gather feedback on our project design to weedout any problems that might be identified at this early stage.

The question we confronted ourselves with during the days that followed the decision to cancel the workshop was: how to recreate the discussion venue of a large workshop remotely and with the purpose of receiving feedback from fellow researchers external to the project?

Key considerations we made in order to answer this question were:

Managing turn-taking is quite an involved task even in physically shared spaces. It’s all the more difficult if  participants of a discussion are not physically together, and their ability to rely on highly tuned cues of timing is diminished. Research in linguistics has found that turn taking gaps for humans is very short (200-600ms)(1), and it seems we plan our responses by predicting how the sentence will end (2); all before we even start moving our mouths. Part of the reason teleconferencing feels taxing is likely because the technology isn’t good enough for us to seamlessly use our highly honed linguistic response senses.

Creating an online feedback session which retains the characteristics of a group discussion, but where the number of people is reduced to the minimum and all participants previously know each other, may work better than larger sessions where participants are more than four-five and do not necessarily know each other from before.

After considering the pros and cons of the above we decided to schedule two remote feedback sessions, each of which would feature two members of the GramAdapt team as hosts and two experts from the University of Helsinki as fellow experts from whom to gather feedback.

About ten days before the meetings, we sent our fellow experts a research package which we had prepared for the workshop. The package was meant to summarize the objectives and interim achievements of the four research paths we are working on: (1) explanatory factors of contact-induced change, (2) comparisons of sociolinguistic environments, (3) sampling, and (4) the sociolinguistic questionnaire. It also included a draft of the sociolinguistic questionnaire, which was the main topic of discussion during both feedback sessions. Before the lockdown, our intention was to share this package with all workshop participants as a basis for discussion, so, in this sense, the workflow we had envisaged for the workshop did not change.

For each feedback session, one team member would field the feedback and navigate the discussion, while the other would take notes on the discussion points that came up. After each session ended, the note taker would summarise the session. The two of us would meet the next day to recap what happened. After both sessions were completed, we took it to the wider team and reported back. It feels quite piecemeal when describing the process, but we felt that being deliberate in each step was necessary and important in order for this to work as we hoped for.

And, as a matter of fact, we feel it worked out. The feedback sessions were very instructing and inspirational. They allowed us to get a sense of how the work we have been doing as a team between January and March can be received by fellow experts, and what we can improve upon. Ultimately, this has geared us up for the work we are doing this spring, and we are very grateful for that.  

All in all, this experience was a reminder of how much information is condensed and collapsed in real, face-to-face interaction. The words we utter are obviously important for communication, but the “extra-linguistic” factors that make for the situation in which communication is carried out are equally important. No doubt there are many other projects out there trying to adapt their work by trying to make up for these “extra-linguistic” factors that make face-to-face interaction feel smooth and (relatively!) seamless.

Links to references:

Levinson, Stephen C., & Torreira, Francisco. (2015). Timing in turn-taking and its implications for processing models of language. Frontiers in Psychology, 6(JUN), 1–17.


Christiansen, Morten H., & Chater, Nick. (2016). The Now-or-Never bottleneck: A fundamental constraint on language. Behavioral and Brain Sciences, 39, 1–72.