10th Speech in Noise Workshop, 11-12 January 2018, Glasgow

Laboratory paired comparisons based on the Common Sound Scenarios (CoSS) framework

Karolina Smeds(a)
ORCA Europe, Widex A/S, Stockholm, Sweden

(a) Presenting

In a previous study, we compared two hearing-aid settings in the field and in the laboratory using paired comparisons. We found that the correlation between the data collected in the laboratory and in the field was low. Specifically, the laboratory data could not predict the individual overall preference in the field (Dahlquist et al 2015). Potential explanations for the low predictive value of the laboratory test included the selection of the laboratory sound stimuli, and the artificial task used in the laboratory.

In another study, we investigated the listening situations people encounter in real life. Based on data from a literature study, a framework called the Common Sound Scenarios (CoSS) was developed. Three intention categories were formed: “Speech communication”, “Focused listening” (without own speech), and “Non-specific” (including monitoring surroundings and passive listening) (Wolters et al 2016).

When studying the CoSS framework, it becomes obvious that most laboratory tests only tap into the “Focused listening” intention category. Neither real speech communication, nor more passive listening situations are usually included in the laboratory.

In the current study, a new laboratory paired-comparisons paradigm was tested. The method focused on a test participant’s intention and the task the test participant had to solve in a scenario. Five mandatory scenarios (representing all three intention categories in CoSS) and up to six individual test scenarios were included. The individual test scenarios were selected from a list of situations experienced in an accompanying field trial. Real conversations between the test participant and one or two test leaders was central to the method, but we also included ecologically valid scenarios with focused listening and scenarios including sound monitoring and passive listening.

Test participants judged the new laboratory test to be ecologically valid. When the results from the field and the laboratory were compared, the correspondence was satisfactory, even though the background sounds experienced in the field were not strictly matched in the laboratory, and the laboratory loudspeaker setup was simple. Laboratory testing with a focus on scenarios with a variety of intentions and tasks and on commonly experienced sounds can of course be used with more acoustically accurate setups.

The pros and cons of the method will be discussed.

Dahlquist M, Larsson J, Hertzman S, Wolters F, Smeds K. (2015) Predicting individual hearing-aid preference in the field using laboratory paired comparisons. In: 5th International Symposium on Auditory and Audiological Research (ISAAR). Nyborg, Denmark.
Wolters F, Smeds K, Schmidt E, Christensen EK, Norup C. (2016) Common Sound Scenarios: A context-driven categorization of everyday sound environments for application in hearing-device research. Journal of the American Academy of Audiology 27: 527-40.

Last modified 2017-11-17 15:56:08