Target-distractor distinctiveness guides efficient search

Visual search is a common activity that people engage in regularly. Efficient search, where the target is distinct from distractors, is typically easy to perform. It is believed that efficient search primarily relies on a stage of parallel processing where all the locations in the scene are examined simultaneously to determine the target’s location. However, the specific type of parallel processing and its guidance on efficient search are still subjects of debate. The Target Contrast Signal Theory proposes that during parallel search, a comparison is made between the target template and display items, with distinctiveness signals being accumulated for each item. These distinctiveness signals then guide the search, determining whether an item should be rejected as a distractor, or should be attended to as a potential target. In this thesis, I use computational modeling to investigate the nature of comparisons made during parallel processing and the composition of distinctiveness signals when comparisons are made on multiple feature dimensions. Chapter 2 explores the nature of the parallel comparisons that provide distinctiveness signals in efficient search. Chapter 3 investigates how distinctiveness along shape and texture dimensions combines to guide bidimensional search. Chapter 4 delves into how distinctiveness along color, shape, and texture dimensions combine to guide tridimensional search. The studies conducted in this thesis offer a better understanding of the attentional guidance mechanisms involved in efficient visual search, and also provide a mathematical modeling framework to predict complex search performance based on simpler search performance.

Name thesis:degree_name

Ph.D.

Level thesis:degree_level

Dissertation

Discipline thesis:degree_discipline

Psychology

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2024

Author dc:creator

• Xu, Zoe Jing

Contributors dc:contributor

• Lleras, Alejandro
• Buetti, Simona
• Beck, Diane
• Christianson, Kiel
• Xia, Yan

• Visual Search
• Feature Dimension Combination

Statement dc:rights

• Copyright 2024 Zoe Jing Xu

Language dc:language

en, eng

Handle dc:identifier

https://hdl.handle.net/2142/124501

OAI record oai:identifier

https://www.ideals.illinois.edu/oai-pmh?verb=GetRecord&metadataPrefix=etdms&identifier=oai%3Awww.ideals.illinois.edu%3A2142%2F124501