gms | German Medical Science

The relation between task demand and listening effort (LE) is often assumed to form an inverse U-shape: While effort generally increases with increasing task demand (e.g., lower SNR) [1]), people tend to give up (i.e., invest less effort) when a task becomes overly difficult. Another effect which increases task demand is listening through hearing aids. Strikingly, even listening through a hearing aid in a transparent setting (i.e., gain set to 0 dB insertion gain) has been shown to have a detrimental effect on speech intelligibility for normal hearing listeners [2]. We do not know, however, whether such detrimental effects due to a transparent hearing aid can also be observed on objective and subjective indicators of LE.

With the present study, we aim to gain new insights in the relationship between task demand (i.e., speech intelligibility) and listening effort with and without a transparent hearing aid. The amount of LE was derived from four objective measures used in the past to reflect LE (electroencephalography, pupil size, skin resistance, and heartrate variability) and a subjective 13-level scale for LE (ACALES). We hypothesise that the amount of LE decreases when the task becomes overly difficult, and that the usage of a transparent hearing aid shifts this LE function further towards more favourable SNRs.

Fifteen young normal-hearing volunteers underwent a speech-in-noise task at five levels of task demand reflecting the individual speech reception thresholds from almost unsolvable SNR levels to practically fully solvable SNR levels, with and without a transparent hearing aid (HA) of 60 trials each (5 SNR x 2 HA design). Measurements were split into two fully balanced appointments. Individual SRTs were re-determined in each appointment to ensure the intended performance levels. In addition, we measured LE using the ACALES with and without a transparent hearing aid within each appointment.

Preliminary results show performance differences between each of the five SNRs but no differences between appointments. Listening through a transparent hearing aid shifts the SNRs required for a certain performance to positive values, i.e., decreases speech intelligibility. A comparable effect can be observed for the subjective LE measured by ACALES. Listening through a transparent hearing aid shifts the SNRs for a certain amount of effort to higher values, i.e., increases subjective amount of LE. Following our hypothesis, the pupil data show the inverted U-shape, i. e. the pupil decreases when the task becomes overly difficult.

The final results will provide valuable insight into the dependence of objective and subjective LE from long-recognised factors such as task demand, but also the perceptual burden resulting from even transparent hearing aid processing.