Modality Constraints Revealed in Tactile, Visual, and Auditory Sequential Learning Chris Conway Cornell University cmc82@cornell.edu   Our world is temporally bounded: events do not occur all at once but rather are distributed in time. As such, it is crucial for organisms to be able to encode and represent temporal order information. One potential method for encoding temporal order is to learn the statistical relationships of elements within sequential input. That is, certain elements of a sequence may be more or less likely to follow certain other elements, and being sensitive to such sequential regularities might help organisms learn temporal structure. In humans, this kind of statistical learning appears to support a wide range of cognitive activities, including auditory, visual, and visuo-motor processing. Although many researchers have focused on the similarities between statistical learning in different domains, it is likely that modality constraints exist that may differentially affect such processing. For instance, traditionally, vision and audition have been viewed as ``spatial" and ``temporal" senses, respectively (e.g., Kubovy, 1988). Empirical evidence from perceptual and temporal processing experiments supports claims that vision and audition are each biased to process specific types of information (Glenberg & Swanson, 1986; Mahar, Mackenzie, & McNicol, 1994). However, it is currently unknown whether these modality constraints may affect the learning of statistical/sequential relationships. The current work explores such potential modality constraints. First, we investigate whether statistical learning can occur in the tactile domain. Experiment 1 provides the first direct evidence that touch can mediate the acquisition of statistical information. Second, we compare learning across two additional sensory modalities, vision (Experiment 2) and audition (Experiment 3). Our data reveal commonalities between the ways in which these three modalities afford the learning of statistical/sequential information. However, the data also reveal the presence of possible modality constraints: the auditory modality was most effective at learning the sequential input. These results are discussed in relation to current theories of the mechanisms underlying statistical learning. This work was done in collaboration with Morten H. Christiansen.   Back to main page