Shifts in Audiovisual Processing in Healthy Aging

Abstract

Purpose of review: The integration of information across sensory modalities into unified percepts is a fundamental sensory process upon which a multitude of cognitive processes are based. We review the body of literature exploring aging-related changes in audiovisual integration published over the last five years. Specifically, we review the impact of changes in temporal processing, the influence of the effectiveness of sensory inputs, the role of working memory, and the newer studies of intra-individual variability during these processes.

Recent findings: Work in the last five years on bottom-up influences of sensory perception has garnered significant attention. Temporal processing, a driving factors of multisensory integration, has now been shown to decouple with multisensory integration in aging, despite their co-decline with aging. The impact of stimulus effectiveness also changes with age, where older adults show maximal benefit from multisensory gain at high signal-to-noise ratios. Following sensory decline, high working memory capacities have now been shown to be somewhat of a protective factor against age-related declines in audiovisual speech perception, particularly in noise. Finally, newer research is emerging focusing on the general intra-individual variability observed with aging.

Summary: Overall, the studies of the past five years have replicated and expanded on previous work that highlights the role of bottom-up sensory changes with aging and their influence on audiovisual integration, as well as the top-down influence of working memory.

Keywords: Aging; Inverse effectiveness; Multisensory integration; Sensory processing; Speech perception; Temporal processing; Variability; Working memory.

Figure 1. Methods of measuring audiovisual perception

Panel A depicts the experimental design for both the simultaneity judgment task and the temporal order judgment task. Visual and auditory stimuli are presented with parametrically varies stimulus onset asynchronies, and participants are asked to report either if the stimuli were synchronous, or which modality came first, with each respective task. Panel B depicts the McGurk task, where participants are presented with a visual token of a speaker articulating the syllable “ga” and a simultaneous auditory “ba”. Participants often report perceiving the syllable “da” or “tha”, which was not presented in either modality, and is evidence of multisensory integration. Panel C depicts the sound-induced flash illusion, where an individual is presented with a single visual flash paired with varying numbers of auditory beeps, and instructed to count the number of flashes while ignoring the beeps. When the auditory and visual stimulus inputs are integrated, the presence of multiple beeps induces the perception of a second (or third) illusory flash. Panel D depicts the stream-bounce task. Two visual circles are presented at opposite sides of a computer screen, and then move towards each other, overlap and cross, and proceed to the opposite side of the screen from which they started. This motion can be perceived in one of two ways, with the circles “streaming” passed each other, or “bouncing” off each other and changing directions. Importantly, when an auditory click or beep is played concurrently with the two circles overlapping, the “bounce” perception increases in prevalence, indicative of audiovisual integration.