Shifting Attention in Multimedia: Stochastic Roles, Design Principles and the SSF Model

Abstract

This paper examines the research on design methods for shifting attention between visuals (for example, text, graphics or moving images) and auditory prompts (such as speech, music or effects) in multimedia environments. Stochastic design methods are most prevalent in the literature. Stochastic approaches reflect outdated bottleneck theories of human attention. Results using stochastic design methods have been mostly poor. Studies using the Structured Sound Function (SSF) model, however have shown good immediate results in student retention, and even better results following a latency period. The SSF model relies on the 'explicitness' and 'gist' requirements inherent in tasks and the interrelatedness of spatial and language representations. It is recommended that designs that incorporate gist and explicit requirements using the SSF model be applied developmentally, because as we get older we get better at extracting gist from sound than details from text.

A comparison of temporal speech and text cueing in educational multimedia

“…The procedure for adding sound prompts to hypermedia learning environments was used in both programs, as reported in Mann (1996). ‘Temporal speech cueing’ was used as information about future or past events digitised in enthusiastic female speech prompts, as suggested in Mann (1997b). ‘Temporal text cueing’ was used as information about future or past events presented on the screen as text.…”