Analysis of Cognitive Attitudes to Commercial Films on Basis of Eye Tracking Data

Abstract

The present paper proposes a method for analyzing viewers' cognitive attitudes to commercial films (CF) on TV based on the eye-movement recording (EMR) technique and scenario description of a CF. In this method, a scenario is described in a time-line transition of displaying CF ingredients in a scene. CF ingredients are classified into three groups in terms of involvement of advertised claims: direct claimed items, indirect claimed items and no-claim involved items. This classification allows us not only to provide with a framework for analyzing EMR data but also to estimate claimed quantity containing a CF. A viewer's EMR data during watching a CF is transcribed according to this scenario description scheme. His/her cognitive attitudes to a particular CF are analyzed mainly based on two attitudinal measurements that can be generated by the EMR transcription combined with the scenario description. One measurement, effective gaze ratio, is relating to how intensively a viewer watches a CF. The other measurement, claim perception ratio, is an index for estimating how well a viewer acquires claims or messages regarding an advertised product, service or organization from the commercial film. A series of experiments were carried out with four subjects to obtain their eye-movement data using an ASL 4000 tracking system. Each subject performed six CF-watching sessions each of which was conducted on a different day using the same sixteen commercial films. The proposed method was applied to the experimental data to analyze the viewer's cognitive patterns to CF's. The obtained cognitive attitudes are discussed from relations between their individual attributes such as watching style of CF's and daily TV watching hours. Based on these results, we mentioned some implications on design parameters of CF's such as frequency of changing scene, and effects of celebrity.