Abstract
This thesis re-examines popular theories of attention grounded in cognitive psychology by watching people watching film, an example of a prolonged stimulus. Previous work used the laboratory timed trial, which is known to lack ecological validity. Three topics are examined. Firstly, models of attentional capacity. Secondly, oculomotor capture.Lastly, individual differences. The central three chapters cover the experimental work.
Chapter 2 presents studies on the dual-task, limited capacity paradigm, which is a way of measuring levels of attention capacity. Here, dual-task measurements were compared to predictions from a biologically inspired computational model. Factors such as time, narrative order, motion and flicker were found to be important predictors. Twenty-eight percent of the observed variance was successfully predicted by the model. In addition, a comparison between dual-task and presence revealed both a qualitative relationship and a quantitative, log-linear relationship, indicating that dual-task can track levels of
presence.
Chapter 3 addresses oculomotor capture from irrelevant distractors. Surprisingly, no evidence of capture was found above chance levels. However, a remote distractor type effect did occur, indicating that distractors were effective in disturbing low-level oculomotor processing, but were insufficient to cause capture. It was concluded that the oculomotor capture effect was an artefact of the laboratory timed-trial. Relevance to film restoration and compression research is discussed.
Chapter 4 explores questions of individual difference. The study found that individual gaze patterns allowed significant predictions of film ratings. However, disappointing results were obtained from comparing gaze patterns with personality trait data, and film choice. One possible explanation is that the study used very short concatenated film clips.. Further experimental work is recommended.
I
t was concluded that using film to study attention does lead to new findings. It will be argued that attention can only be understood in relationship to prolonged experiences such as watching film.
Chapter 2 presents studies on the dual-task, limited capacity paradigm, which is a way of measuring levels of attention capacity. Here, dual-task measurements were compared to predictions from a biologically inspired computational model. Factors such as time, narrative order, motion and flicker were found to be important predictors. Twenty-eight percent of the observed variance was successfully predicted by the model. In addition, a comparison between dual-task and presence revealed both a qualitative relationship and a quantitative, log-linear relationship, indicating that dual-task can track levels of
presence.
Chapter 3 addresses oculomotor capture from irrelevant distractors. Surprisingly, no evidence of capture was found above chance levels. However, a remote distractor type effect did occur, indicating that distractors were effective in disturbing low-level oculomotor processing, but were insufficient to cause capture. It was concluded that the oculomotor capture effect was an artefact of the laboratory timed-trial. Relevance to film restoration and compression research is discussed.
Chapter 4 explores questions of individual difference. The study found that individual gaze patterns allowed significant predictions of film ratings. However, disappointing results were obtained from comparing gaze patterns with personality trait data, and film choice. One possible explanation is that the study used very short concatenated film clips.. Further experimental work is recommended.
I
t was concluded that using film to study attention does lead to new findings. It will be argued that attention can only be understood in relationship to prolonged experiences such as watching film.
| Original language | English |
|---|---|
| Publication status | Published - 15 Feb 2017 |
Research Groups and Themes
- Cognitive Science
- Visual Perception
Keywords
- ATTENTION
- Movies
- Immersion