| Chapter 2: | Background and Hypothesis |
It is probably true that identifying an individual content feature is not as obvious as listing an individual structure feature, but previous research using the LC3MP has successfully identified a number of content features affecting either automatic or controlled resource allocation. They include emotional content (A. Lang et al., 1995, 2004), risky products (A. Lang et al., 2005), and difficult content (Fox et al., 2004). For example, A. Lang et al. (2005) found that the existence of riskiness in pictures and words elicited emotional and physiological arousal, which results in automatic allocation of processing resources to the encoding and storage processes. In addition to the automatic allocation, some aspects of content can elicit controlled resource allocation. For example, Fox et al. (2004) examined how differently television viewers process easy and difficult science-related news. Their study showed that the viewers allocated more processing resources to difficult news than they did to easy news.
As described above, the LC3MP conceptualizes a mediated message from a psychological point of view; that is, the conceptualization of a media message is based on which aspect of the message affects what type of the resource allocation processes. Thus, a researcher who wants to use the LC3MP ought to be able to break a message into its components and determine the separate effects of these parts on various dependent variables of interests (A. Lang, 1990).
The same process can be used to study banner advertisements. The first step is to identify the aspects of structure and content that are likely to affect the distribution of processing resources. Recent advances in Web technology have added to banner ads the capability of featuring multimedia content (e.g., similar to a short version of a TV commercial or a simple video game). However, the majority of banner ads use a relatively small number of structural features due to limited size and downloading speed.