Phase 1: Speech production study
83 adults, all speakers of Standard Southern British English, came to our lab for the recording phase of the project, over three separate sessions. These were 57 older adults aged between 65 to 84 years; 30 had normal hearing (OANH) and 27 had mild age-related hearing loss. There were also 26 younger adults aged 18-26 who had normal hearing.
In Session 1, participants had their hearing tested and completed a number of short tests of their cognitive and hearing abilities. In Sessions 2 & 3, they completed two types of tasks with a conversational partner (‘Talker B’) who was a young adult and who took a more passive role in the task. These tasks were carried out in different conditions where communication was easy or difficult for one or both partners. We were primarily interested in analysing the speech behaviour of Talker A who was the person having to adapt their speech to make sure the information was transmitted during the task.
Task 1: ‘Spot the difference’ diapix task
Conversational partners collaborate to find 12 differences between their two pictures.
Task 2: Sentence repetition task
Talker A reads BKB sentences to Talker B who had to repeat them back.
Conditions
Both tasks are done in four conditions where communication is either easy (NORM) or made difficult for one (HLS, BAB-1) or both (BAB-2) conversational partners.
Modes
All tasks and conditions are done in 2 modes:
Audio alone: Conversational partners communicate via headphones and can’t see each other.
Audiovisual: Conversational partners communicate via headphones and see each other via a two-way window.
Phase 2: Perception study
A number of perception studies have been run with young adults, older adults with normal hearing and older adults with age-related hearing loss. These use as test materials either the BKB sentences recorded in Phase 1 or short snippets extracted from the diapix task (e.g. ‘there’s a zebra crossing and a large green car’). Sentences or snippets from the different conditions (NORM, HLS, BAB2) are presented in babble noise. In some studies, participants respond by saying what they have heard, while in others they rate how intelligible the speech was.