A comparison of cognitive performance and listening effort test procedures

Workers in open-plan offices are objected to irrelevant speech, which leads to decreased cognitive performance. With acoustical treatments, such as sound masking, the cognitive impact can be reduced, but there is no objective procedure to assess this improvement in real office environments. For in-situ evaluation, a robust test procedure is required. The serial recall test used in laboratories requires masking levels to be higher than the speech concealed, which seldom is the case in real offices. Therefore, a dual-task paradigm was investigated, which measures latencies for auditory stimuli presented in noise and silence. The method, originating from audiology for listening effort measurement, showed significant differences even for little masked, clearly intelligible speech. The underlying cognitive processes leading to those differences in response times need to be understood. Experiments with 21 normal hearing participants were conducted to investigate the memory component and the auditory identification process in the listening effort experiment. A comparison to the serial recall test was carried out. The findings suggest an arousal effect at work, with faster replies for higher masking in serial recall and listening effort test. The two tests seem to analyse different cognitive aspects.