The Audible Contrast Threshold (ACT™) test: a clinical spectro-temporal modulation detection test

Abstract

Over the last decade, multiple studies have shown that hearing-impaired listeners’ speech-in-noise reception ability, measured with audibility compensation, is closely associated with performance in spectro-temporal modulation (STM) detection tests. STM tests thus have the potential to provide highly relevant beyond-the-audiogram information in the clinic, but the available STM tests have not been optimized for clinical use in terms of test duration, required equipment, and procedural standardization. The present study introduces a quick-and-simple clinically viable STM test, named the Audible Contrast Threshold (ACT) test. First, an experimenter-controlled STM measurement paradigm was developed, in which the patient is presented binaurally with a continuous audibility-corrected noise via headphones and asked to press a pushbutton whenever they hear an STM target sound in the noise. The patient’s threshold is established using a Hughson-Westlake tracking procedure with a three-out-of-five criterion and then refined by post-processing the collected data using a logistic function. Different stimulation paradigms were tested in 28 hearing-impaired participants and compared to data previously measured in the same participants with an established STM test paradigm. The best stimulation paradigm showed excellent test-retest reliability and good agreement with the established laboratory version. Second, the best stimulation paradigm with 1-second noise “waves” (windowed noise) was chosen, further optimized with respect to step size and logistic-function fitting, and tested in a population of 25 young normal-hearing participants using various types of transducers to obtain normative data. Based on these normative data, the “normalized Contrast Level” (in dB nCL) scale was defined, where 0±4 dB nCL corresponds to normal performance and the greater the positive value of dB nCL, the greater the audible contrast loss. Overall, the results of the present study indicate that the ACT test may be considered a reliable, quick-and-simple (and thus clinically viable) test of STM sensitivity. The ACT can be measured directly after the audiogram using the same set up, adding only a few minutes to the process.CITE AS1) In text: Zaar/Simonsen et al. (2023)2) In reference list: Zaar, J./Simonsen L. B., Sanchez-Lopez, R., and Laugesen, S. (2023): “The Audible Contrast Threshold (ACT™) test: a clinical spectro-temporal modulation detection test,” medRxiv.

Predicting speech-in-noise reception in hearing-impaired listeners with hearing aids using the Audible Contrast Threshold (ACT™) test

“…A histogram of the 108 ACT values is shown in the left panel of Figure 5, with the vertical gray lines indicating the mean (solid line; 7.0 dB nCL) and median (dotted line; 6.4 dB nCL). A broad distribution can be observed, which covers the entire range between "normal" performance (estimated to be in the range 0±4 dB nCL, see Zaar/Simonsen et al, 2023) and the maximum measurable level of contrast loss (16 dB nCL). It should be noted that 7 out of the 8 participants with an ACT of 16 dB nCL were not able to obtain a threshold as they were not able to detect the difference between noise and fully modulated noise such that the maximum value of 16 dB nCL was assigned to them.…”

“…For participants who were unable to do the task, the maximum ACT value of 16 dB nCL was stored. For more details on the ACT test procedure and data post-processing, the reader is referred to Zaar/Simonsen et al (2023). Two runs were conducted in order to assess the test-retest reliability of the ACT test.…”

“…However, the test may not have been immediately applicable to clinical practice as it was administered using a high-precision laboratory procedure that consisted of multiple adaptive runs with a 3-alternative forced-choice procedure (3-AFC; duration about 15 minutes) and required a computer screen and keyboard for the test person to interact with. The STM test used in Zaar et al (2023b) was thus transformed into a clinically viable version named the "Audible Contrast Threshold" (ACT) test (Zaar/Simonsen et al, 2023). The ACT test was designed to (i) be quick and simple, (ii) require only the standard audiometry interface (headphones and pushbutton), and (iii) be as similar as possible to the measurement procedure applied in pure-tone audiometry (Carhart and Jerger, 1959).…”

“…The ACT test was designed to (i) be quick and simple, (ii) require only the standard audiometry interface (headphones and pushbutton), and (iii) be as similar as possible to the measurement procedure applied in pure-tone audiometry (Carhart and Jerger, 1959). Based on normative data obtained in young audiometrically NH listeners, the physical modulation level was transformed into the "normalized Contrast Level" (nCL), such that normal contrast sensitivity yields around 0 dB nCL and positive values up to maximally 16 dB nCL indicate various degrees of "contrast loss" (Zaar/Simonsen et al, 2023). The present study evaluates the novel clinical ACT test with respect to test-retest reliability, test duration in clinical settings, and crucially investigates its connection to speech-reception performance in noise with hearing aids, including effects of DIR+NR processing, in a large clinical population of HI participants.…”

“…Furthermore, a recent study showed a connection between STM sensitivity and the speech-intelligibility benefit induced by different levels of directionality and noise reduction (DIR+NR) hearing-aid processing. A further significant development has been the introduction of the "Audible Contrast Threshold" (ACT) test (Zaar/Simonsen et al, 2023), a novel STM detection test optimized for clinical use in terms of measurement time, required test equipment, and measurement procedure. The present study, conducted at two sites in Germany and Japan, evaluated the ACT test in a clinical setting using 108 HI participants (81 German; 27 Japanese) with respect to (i) test-retest reliability, (ii) measurement duration, (iii) association with aided speech-in-noise performance, and (iv) connection to DIR+NR speech-intelligibility benefits.…”