Efficacy of Auditory Training and Combined Auditory-working-memory Training in Improving Communication in Older Adults

The Cantonese version of the Hearing in Noise Test (CHINT) (Wong & Soli, 2005) will be used to measure sentence perception in noise. The tool contains 12 lists of 20 sentences, and each sentence comprises 10 words. A high inter-list reliability has been previously demonstrated, which suggests that the lists are equivalent and, therefore, consistent results can be obtained using any of the lists (Wong et al., 2005).

The Cantonese version of the reading span test (CRST) with dual task design will be used to measure WM. It was developed based on the German reading span test (Carroll et al., 2015). The reliability and validity of the CRST have been examined to confirm that the CRST is a suitable test for investigating verbal WM (Cheung, 2021).

The CRST comprises 54 six-word sentences. The sentences are presented in different set sizes (i.e., 3, 4, 5, and 6 sentences) in random order. The participants are expected to read aloud the sentences presented on a computer screen and judge whether the sentence is semantically correct. After reading aloud all sentences in a given sentence set, the participants will be asked to recall the first or last two words of each sentence in the set. The reading span score is based on the second task: the total number of words recalled. Approximately 20 minutes is needed to complete the test.

Other cognitive functions (executive function, response inhibition, processing speed, and attention) will be examined using the CogState Battery, which is a sensitive computer-based assessment instrument. The test battery has been adapted for Chinese and its reliability and validity have been previously reported (Zhong et al., 2013). The test battery includes the following subtests and uses playing cards, digits, and pictures as stimuli: 1) One Back Test (short-term memory), 2) Groton Maze Learning Test (executive function), 3) Identification Test (Attention), 4) Go/No-Go Test (response inhibition), 5) International Digit Symbol Substitution Test-Symbols (processing speed). Each CogState subtask takes approximately 3-8 minutes. The entire CogState test battery is expected to be completed in 30-45 minutes.

The Speech, Spatial, and Qualities of Hearing scale (SSQHS) (Gatehouse & Noble, 2004) will be used to assess abilities and experience of hearing in different listening environments. The SSQHS is assessed using a self-report questionnaire that comprises 49 questions across the following three scales: 1) speech hearing (n = 14) for measuring the ability to recognize speech in a variety of competing contexts; 2) spatial hearing (n = 17) for assessing the ability of the directional, distance, and movement components of spatial hearing; 3) quality of hearing (n = 18) for evaluating the quality of hearing experience, including the clarity and naturalness of different speakers and everyday sounds, and ease of listening. The use of the SSQHS has been validated for Cantonese speakers (Lee, 2015).

The stimulus, a 175 milliseconds /ji/ with Cantonese Tone 4 (low falling), will be used for recoding of subcortical neural responses to lexical tone. Stimuli with an interstimulus interval of 80 milliseconds will be presented in alternating polarity at 80 dB SPL and at 10 dB SNR with a six-speaker multi-talker noise through an insert earphone (ER-3, Etymotic Research). This type of noise and SNR were used in previous studies where significant relationships between SIN perception and cABRs were reported (Anderson & Kraus, 2010b; Anderson et al., 2011). Responses will be collected using a vertical electrode montage (ipsilateral earlobe reference, Cz active, and forehead ground) with the SmartEP (Intelligence Hearing Systems). An average of 6,000 sweeps will be made for responses to /ji4/ in quiet and multi-talker babble noise conditions, respectively.