Evaluating methods of crossmodal matching for multimodal displays in younger and older adults. Issue 126 (June 2019)

• Record Type: Journal Article
• Authors: Gomes, Kylie M.; Riggs, Sara L.
• Is Part Of: International journal of human-computer studies. Issue 126(2019)
• Page Start: 1
• Page End: 13
• Publication Date: 2019-06
• Subjects: Multimodal displays -- Age -- Auditory -- Visual -- Tactile -- Psychophysical methods; Human-machine systems -- Periodicals; Systems engineering -- Periodicals; Human engineering -- Periodicals; Human engineering; Human-machine systems; Systems engineering; Periodicals; Electronic journals; 004.019
• DOI: 10.1016/j.ijhcs.2019.01.001
• Languages: English
• ISSNs: 1071-5819

Abstract

Highlights: Three types of crossmodal matching methods – one based on a method of bracketing (i.e. MoB interface) and two based on a method of adjustment (i.e. MoA interfaces) – were evaluated with younger and older individuals to determine whether method used, modality order, or age contributed to subject variability in crossmodal matching. The MoA interfaces were shown to have less variability amongst match values compared to the MoB interface; however, participants indicated in the debriefing questionnaire they relied on the visual feedback of the sliding scale to make their matches rather than matching solely based on their perception of the stimuli, which may have contributed to the lower variability of the MoA interface methods. Although the MoB interface resulted in increased variability for both age groups, older participants had less variability in their match values compared to younger participants for matches including audition and touch (i.e., A-T and T-A). Across both age groups, there was a preference towards using the less complicated MoA interface; however, it is recommended that visual feedback be eliminated as it may influence crossmodal match selections. Abstract: A limited number of multimodal studies incorporate crossmodal matching – matching the intensity of stimuli across different sensory modalities – which is an important step that can help account for individual differences in the perception of cue salience. This study evaluates three crossmodal matching methods in terms of how age, matching method, and modality combination may contribute to subject match variability. One method was based on the method of bracketing (MoB; participant zeroes in on a match) and two methods were based on the method of adjustment (MoA; participant directly controls the stimulus intensity). This study evaluated three different matching methods with younger adults (ages 18–30 years) and older adults (ages 50–69 years). Matched values (stimulus intensity chosen by the participant) and subjective data were collected. The findings from both studies support the need to incorporate crossmodal matching in multimodal studies, as the results showed high match variability (between-subject variability in matched values selected) when using the MoB, where participants seemed to rely more on their perception compared to the MoA interface. Compared to younger participants, older participants exhibited less variability for matches with modality combinations that included the tactile and auditory channel – suggesting that age may contribute to subject variability when matching with these sensory channels. Furthermore, very few significant differences were found between interfaces at higher intensities suggesting that the MoB may be a suitable alternative to the MoA because while variability was highest with the bracketing method, the findings show it may be a truer reflection of participants' responses. These findings can help inform the design and development of a reliable crossmodal matching method for use in multimodal experiments.