Development of a German reading span test with dual task design for application in cognitive hearing research

Abstract

Objective: To report the development of a standardized German version of a reading span test (RST) with a dual task design. Special attention was paid to psycholinguistic control of the test items and time-sensitive scoring. We aim to establish our RST version to use for determining an individual's working memory in the framework of hearing research in German contexts. Design: RST stimuli were controlled and pretested for psycholinguistic factors. The RST task was to read sentences, quickly determine their plausibility, and later recall certain words to determine a listener's individual reading span. RST results were correlated with outcomes of additional sentence-in-noise tests measured in an aided and an unaided listening condition, each at two reception thresholds. Study sample: Item plausibility was pre-determined by 28 native German participants. An additional 62 listeners (45–86 years, M = 69.8) with mild-to-moderate hearing loss were tested for speech intelligibility and reading span in a multicenter study. Results: The reading span test significantly correlated with speech intelligibility at both speech reception thresholds in the aided listening condition. Conclusion: Our German RST is standardized with respect to psycholinguistic construction principles of the stimuli, and is a cognitive correlate of intelligibility in a German matrix speech-in-noise test.

Key Words::

• Test design
• verbal working memory
• dual task
• speech intelligibility

Declaration of interest: The authors report no conflict of interest.

This work was conducted within the BMBF-funded project Modellbasierte Hörgeräte II, (grant number 01EZ1127A), and the Lower Saxony Research Network Design of Environments for Aging (grant number ZN 2701) funded by the Lower Saxony Ministry of Science and Culture.

Notes

Notes

1. Transitive: The verb requires two arguments (a subject and an object) to constitute a grammatically correct sentence. For example, the man kisses the woman. Omitting the woman would result in an ungrammatical sentence.

2. Intransitive: The verb only requires one argument (the subject) to constitute a grammatically correct sentence. For example, the man laughs. Additional adverbials are optional.

3. The reason for the minor differences in number across these conditions was due to elimination of sentences based on their plausibility scores. We suggest that unambiguous identification of a given sentence as either absurd or sensible (and an equal number of this factor) is more important than an equal distribution of transitive and intransitive verbs.

4. The word is 2ⁿ, i.e. 2⁹–2¹⁴, times less frequent than the most frequent lexical item (der ‘the’). Frequency classes in the Leipzig corpus range from 1–24.

5. Two training lists of 20 sentences are required before testing the first test list (cf. Wagener et al, 1999); one test list is required before testing a new setting/condition.