Human performance in visual search for multiple targets

다국어 초록 (Multilingual Abstract)

Previous studies of visual search of an extended search field have been limited to search tasks for locating a single target. This study extended the models and data for locating a single target to search tasks for locating multiple targets. Multiple-target search tasks can be categorized into two types by the exhaustiveness of the task. In exhaustive visual search tasks, observers have to find all given targets, and in non-exhaustive visual search tasks, observers must stop at some point during the task. The first topic of this study was to derive human performance models for exhaustive multiple-target search and then to validate the models. Random and systematic models for an exhaustive search task were derived which were expected to describe upper and lower bounds of human search performance. The proposed random search model was described by a hypo-exponential distribution, while the proposed systematic search model was a piece-wise curvilinear function. These models were general models, including those of the single-target search as special cases. Human search performance data from a visual search experiment fitted between the two search performance models. On the other hand, judging from the analysis of interview data and search performance data, participants' search behavior changed during exhaustive multiple-target search tasks. The models showed that they searched the search fields (1) at faster speed, (2) with shorter dwell time in a fixation, and (3) with less revisits in a fixation in the initial period of the search task than in the late period. Such a phenomenon was more evident in a multiple-target search task than in a single-target search task, and in a multiple-target search task for different-type targets than in a multiple-target search task for targets of the same type. The second topic of this study was to derive an optimal stopping time model for a non-exhaustive search task where multiple targets can be detected. Three usage strategies of the optimal stopping time were compared: a self-stopping strategy, an externally forced stopping strategy and a hybrid stopping strategy. The self-stopping strategy was the most effective among the three strategies under almost all task conditions of different time pressure and different pre-information on the number of targets (known and unknown number of targets). Such effectiveness of the self-stopping strategy might be caused by human observers' situation awareness ability, using decision cues. Finally this study suggested a way to bridge between search theory in extended visual fields and theories of search in a single fixation. Although our study proposed computational performance models, the visual lobe (an important input parameter in both models) was measured by an experiment, instead of being computed. Comparing FVS (Function-based Visual Search) theory with AVS (Attention-Based Visual Search) theory, it was recognized that AVS theory can support the computational prediction of a visual lobe. As an extension to the current study, such a bridging between FVS theory and AVS theory should be performed to develop more useful and general human performance models.

목차 (Table of Contents)

• Abstract = iv
• Table of Contents = vi
• List of Tables = x
• List o Figures = xii
• CHAPTER 1 INTRODUCTION = 1
• 1.1 WHAT IS VISUAL SEARCH? = 1
• 1.2 TWO PARADIGMS OF VISUAL SEARCH STUDY = 2
• 1.3 MODELING OF VISUAL SEARCH ACTIVITY = 5
• 1.4 NEED OF MULTIPLE-TARGET SEARCH MODELS = 6
• 1.5 PURPOSE OF THIS DISSERTATION = 8
• 1.6 ORGANIZATION OF THIS DISSERTATION = 9
• CHAPTER 2 RESEARCH BACKGROUND = 12
• 2.1 VISUAL SEARCH THEORY AND PERFORMANCE = 12
• 2.1.1 Attention-Based Search Theories = 12
• 2.1.2 Measures of Visual Search Performance = 19
• 2.1.3 Parameters for Search Performance Models = 20
• 2.1.4 Search Performance Models for FVS = 29
• 2.1.5 Multiple-Target Search Models in AVS Study = 36
• 2.2 STOPPING RULES AND OPTIMAL STOPPING TIME IN VISUAL SEARCH = 40
• 2.2.1 Search Termination Theory in AVS Study = 41
• 2.2.2 Economic Models for Search Stopping in FVS = 42
• 2.2.3 Theories for Search Stopping in the Other Research Fields = 46
• 2.2.4 Time Awareness and Probability Perception = 49
• 2.2.5 Economic Model and Descriptive Theory = 51
• CHAPTER 3 GMST MODELS FOR MULTIPLE TARGETS OF THE SAME TYPE. = 54
• 3.1 INTRODUCTION = 54
• 3.2 SEARCH PERFORMANCE MODELS (GMST) = 55
• 3.3 SIMULATION EXPERIMENT = 58
• 3.4 EXPERIMENT = 62
• 3.5 RESULTS = 64
• 3.6 DISCUSSION AND CONCLUSIONS = 68
• CHAPTER 4 SATO MODELS FOR MULTIPLE TARGETS OF THE SAME TYPE. = 74
• 4.1 INTRODUCTION = 74
• 4.2 SEARCH PERFORMANCE MODELS FOR MULTIPLE TARGETS OF THE SAME TYPE = 75
• 4.2.1 Random Search Model = 75
• 4.2.2 Systematic Search Model = 79
• 4.3 EXPERIMENTS = 93
• 4.4 RESULTS = 97
• 4.4.1 Probability of Target Detection in a Single Fixation = 97
• 4.4.2 Comparison of Human Search Performance with Search Models = 102
• 4.5 DISCUSSION AND CONCLUSIONS = 107
• CHAPTER 5 SEARCH PERFORMANCE MODELS FOR MULTIPLE TARGETS OF DIFFERENT TYPES = 111
• 5.1 INTRODUCTION = 111
• 5.2 SEARCH PERFORMANCE MODELS FOR MULTIPLE TARGETS OF DIFFERENT TYPES = 113
• 5.2.1 A Random Search Model for Multiple Different Targets = 113
• 5.2.2 A Systematic Search Model for Multiple Different Targets = 120
• 5.3 PREDICTION MODELS FOR TARGET DETECTION ORDER IN TWO TARGET SEARCH = 122
• 5.3.1 Prediction Model Based on Random Search Strategy = 122
• 5.3.2 Prediction Model Based on Systematic Search Strategy = 125
• 5.4 EXPERIMENTS = 127
• 5.5 RESULTS = 131
• 5.5.1 Analysis of Target Conspicuity:Lobe Size Measurement = 131
• 5.5.2 Relationship between GMSTs in Single-Target Search and Typical Visual Lobes = 133
• 5.5.3 Inter-fixation Distance in Single-Target Search = 135
• 5.5.4 Search Performance in Multiple-Target Search = 139
• 5.5.5 Which Target is Located First in Two-Target Search = 151
• 5.6 DISCUSSION AND CONCLUSIONS = 152
• CHAPTER 6 STOPPING STRATEGIES AND STOPPING TIMES IN MULTIPLE-TARGET SEARCH = 158
• 6.1 INTRODUCTION = 158
• 6.1.1. Task Conditions in Multiple-Target Search Tasks = 160
• 6.1.2 Three stopping strategies = 161
• 6.2 AN OPTIMAL STOPPING TIME MODEL IN MULTIPLE-TARGET SEARCH = 163
• 6.2.1. An Optimal Stopping Time Model in Visual Search for a Known Number of Targets = 163
• 6.2.2. An Optimal Stopping Time Model in Visual Search for an Unknown Number of Targets = 169
• 6.3 EXPERIMENTS = 171
• 6.4 RESULTS = 181
• 6.4.1 Search Performance in Multiple-Target Search = 181
• 6.4.2 Optimal Stopping Times and Maximum Expected Task Values = 183
• 6.4.3 Effectiveness of Externally-Forced Stopping Strategy = 184
• 6.4.4 Effectiveness of Hybrid Stopping Strategy = 186
• 6.4.5 Effectiveness of Self-Stopping Strategy = 187
• 6.4.6 Comparison of a Self-Stopping Strategy with the Other Strategies = 188
• 6.4.7 Decision Cues Weighting Policy for Self-Stopping = 195
• 6.5 DISCUSSION AND CONCLUSIONS = 200
• CHAPTER 7 GENERAL DISCUSSION = 205
• 7.1 A TAXONOMY OF VISUAL SEARCH TASKS = 205
• 7.2 COMPARISON OF THE STUDIES IN THIS DISSERTATION WITH PREVIOUS STUDIES = 210
• 7.2.1 From Single Target to Multiple Targets (exhaustive visual search) = 212
• 7.2.2 From Same Types to Different Types (exhaustive visual search) = 213
• 7.2.3 From Exhaustive Search to Non-exhaustive Search = 215
• 7.3 PRACTICAL USES OF VISUAL SEARCH THEORY = 217
• 7.4 FROM AVS TO FVS = 220
• CHAPTER 8 GENERAL CONCLUSIONS = 224
• REFERENCE = 227
• APPENDIX 1. ANOVAS OF SEARH TIMES IN MULTIPLE-TARGET SEARCH TASKS (for Chapter 3) = 245
• APPENDIX 2. VISIBILITY AREAS AND VISUAL LOBES IN DIFFERENT FIXATION DURATIONS (for Chapter 4) = 247
• APPENDIX 3. VISIBILITY AREAS FOR DIFFERENT-TYPE TARGETS AND SEARCH PERFORMANCES (for Chapter 5) = 257
• APPENDIX 4. ANOVAS AND K'S VALUES (for Chapter 6) = 279
• APPENDIX 5. CONSENT FORM = 303