Abstract
Spatial locations are perceived and represented more accurately by vision than by audition, while time intervals are perceived more accurately by audition (Welch & Warren 1980). These task-modality relationships also occur in working memory tasks (e.g. Guttman 2005, Lehnert & Zimmer 2008). Here, we test whether differences between vision and audition when remembering locations and intervals arise from corresponding differences in working memory capacity. We hypothesized that, particularly at low memory loads, capacity estimates and sensory modality would interact, with capacity increasing more quickly in the task-appropriate modality (i.e., vision for spatial tasks) as memory load increased. To test this, participants (n=10) performed eight change detection tasks with memory loads of two or four items (varying sequence length). Participants compared two short sequences presented either visually or auditorily, and looked for changes in either the order of spatial locations (left/right), or the order of inter-item intervals (short/long). Visual stimuli comprised an array of two static images (photographs of animals on a white background); visual sequence elements were an instantaneous mirror flip of one image. Auditory stimuli comprised a stream of complex tones, each of which was one auditory sequence element. Visual stimuli had spatial positions on a computer monitor; auditory stimuli were lateralized by interaural time differences of ±1 ms. Inter-item intervals were 300 and 720 ms. As hypothesized, larger memory loads led to increased capacity estimates on both spatial and temporal change detection tasks, with greater increases for the task-appropriate modality. Capacity estimates in task-inappropriate modalities remained near or below one item. This interaction between task domains and information modality supports an account of separate working memory stores for spatial locations and for time intervals, with privileged access for visual and auditory modalities, respectively.
Meeting abstract presented at VSS 2016