Return to search

Direct selection by colour for visual encoding

The goal of this thesis was to investigate the role of colour in visual selective attention. Previous experiments exploring this topic in tasks where location varied led to mixed results. Some studies only found evidence of colour as a guide to a specific location where selection then takes place (e.g., Nissen, 1985). Others reported an effect, but could not decide clearly if the benefit was due to direct selection of colour in perception (e.g., Humphreys, 1981). One major contributor to the inconsistencies of findings seems to be the confounding of colour and location in these tasks. For that reason the initial paradigm used here was a rapid serial visual presentation (RSVP) task. Previous studies using similar paradigms have found no evidence for direct selection by colour (Poder, 2001; Shih & Sperling, 1996), but in these studies advance colour information was of limited usefulness because it only reduced the set of candidate stimuli by half. To assess an effect of colour in selection similar to the one associated with location, in all experiments reported here valid colour information led to only one item, as is typical in location cuing tasks.
The first RSVP experiment explored whether colour certainty improved performance over a colour uncertainty condition. Colour was the defining feature of the target participants had to discriminate. In one condition the target colour was certain; in the other it could be one of two colours. Performance was improved when participants could focus on one colour. Further experiments used colour not as a defining feature of the target but as additional information presented in the form of cues, similar to the typical use of location cues. The participants� task was to discriminate whether a target letter within the RSVP sequence appeared in its upper or lower case version, and an advance cue indicated the colour in which the target letter was most likely to occur. An accuracy benefit of valid colour information was found, supporting the hypothesis that colour cuing allows the direct selection of objects for further perceptual processing. In addition, an effect of invalid colour cues was also observed. Subsequent experiments investigated possible factors influencing the colour cuing effect. Together, task requirements and properties of the stimulus set were shown to have an influence on the effect size, whereas an increase in perceptual load had no impact. Furthermore, the colour cuing effect seems to be due partially to both automatic and strategic processes. In all these experimental variations, benefits of colour cuing remained, indicating that the effect is very robust. Colour cuing effects were also found in a design where location could vary, extending the previous findings from selection in the time domain to selection in space. The two last experiments investigated whether advance colour knowledge would also lead to a performance benefit in single item tasks. No effect of colour cuing was found, indicating that colour information is only helpful in multiple item displays when a selection of one target stimulus among distractor items is necessary.

Identiferoai:union.ndltd.org:ADTP/266174
Date January 2005
CreatorsVierck, Esther, n/a
PublisherUniversity of Otago. Department of Psychology
Source SetsAustraliasian Digital Theses Program
LanguageEnglish
Detected LanguageEnglish
Rightshttp://policy01.otago.ac.nz/policies/FMPro?-db=policies.fm&-format=viewpolicy.html&-lay=viewpolicy&-sortfield=Title&Type=Academic&-recid=33025&-find), Copyright Esther Vierck

Page generated in 0.0023 seconds