 Hello, my name is Bethany, I'm part of HEAR ECO, which is a Horizon 2020 project looking at testing listening effort in more ecologically valid conditions and using novel outcome measures. Today I'm going to tell you about an experiment we recently conducted using pupil and cardiovascular measures combined. Now physiological measures of listening effort often don't correlate well with one another and they also don't seem to correlate well with the subjective ratings of the participant and it's probably because they reflect different things. We decided to implement seven different features in a classifier to see what we could learn. We recruited 29 hearing impaired participants who undertook the Danish hint speech and noise test at two different difficulty levels. SNRs individually adapted to 50 or 80% correct. They also undertook the task alone and in the presence of two observers to stimulate a more ecologically valid listening condition. Whilst they were doing the task, we measured seven different features from them, four from the cardiovascular system and three from their pupil. After the experiment, we asked the participants in a semi-structured interview about their perception of the experience, including whether they were affected by the presence of the observers. We trained K nearest neighbour classifiers using our seven physiological features to determine with which accuracy level we could predict the intelligibility level, so whether it was a 50% or 80% trial, whether they were alone or observed, and finally the participants experience of being observed, whether they were affected by the presence of the observers. You'll have to visit my presentation to find out more about the results and more of the details of our analysis. With that, I'd like to say thank you very much for listening to my pitch and I hope I can share more with you soon. Thank you.