Year: 2024
Type: conference-paper
Abstract: The Concurrent OLSA (CCOLSA) is a multi-talker speech test that uses a temporal overlap between sentences from alternating talkers to achieve sensitivity at high signal-tonoise ratios (SNR).The paradigm is a dual-task design that combines call-sign detection and speech recognition for switching target talkers.This study proposes an adaptive CCOLSA approach, where the overlap parameter is adjusted adaptively to measure individual thresholds with 50 percent speech recognition at fixed SNR.For this, the adaptive procedure used in the corresponding standard matrix test is applied with a transformation of SNR steps to overlap time steps.As outcome measure, a consolidated result value is determined, which is called CCOLSA costs and includes the difference between the presented SNR and the individual Speech-Reception Threshold (SRT) as well as the difference of the individual overlap time compared to the median overlap time of normal-hearing subjects.The approach was tested for 31 elderly participants with various hearing profiles.Additionally, the Comprehensive Trail Making Test 2 (CTMT-2) was conducted.The measured individual CCOLSA costs at threshold amount 4.0-7.1 dB with a median test-retest deviation of 0.4 dB.Interactions between CCOLSA costs, SRTs and CTMT-2 results are discussed.
Authors: J. Heeren, V. Hohmann, M. Schulte, K.C. Wagener
Institution: Hearing4all
Language: English
Subfield: Cognitive Neuroscience
Field: Neuroscience
Domain: Life Sciences
Open Access status: bronze
Funder: Deutsche Forschungsgemeinschaft
OpenAlex ID: W4391028702