Is Cross-Modal Matching Necessary? A Bayesian Analysis of Individual Reference Cues

The number of research activities on multi-modal feedback cues and their potential to enhance the performance of human operators during teleoperation tasks is growing. Yet, it is still unclear how stimulus intensities of cues in different modalities should be matched when investigating the effects of feedback modality on task performance. Previous work has shown a high within- and between-subject variability of multi-modal intensity adjustments. The source of this variability is not yet clear. In this study, we investigate the individual perception of the cues by setting individual reference intensities for cross-modal matching. In addition to traditional frequentist models, Bayesian models are used for data analysis. The results suggest that haptic as well as auditory cue intensity adjustments are not influenced by modality when combining it with a visual reference. Thus, it seems sufficient to let subjects adjust the intensity for haptic and auditory cues individually prior to investigating the effectiveness of haptic-auditory cues. In contrast, with a 99% probability, visual cues are adjusted to lower intensities by subjects if combined with haptic or auditory cues, which points to the importance of individual cross-modal matching when investigating the effects of visual-haptic or visual-auditory feedback cues.