Visual search is an important component of many human-computer interactions and a critical task in a variety of domains. These include screening for prohibited items in x-ray imagery of luggage in airports, identifying anomalies in medical imagery, and looking for unusual activity and changes over time in satellite imagery. Unlike many other human-machine interface tasks where a small error rate is quite acceptable, a crucial element of visual search tasks in these application areas is that a single mistake can have catastrophic consequences. Thus, any improvements in technology, interfaces, or human processes that can be made to aid visual search would be invaluable in these safety critical areas. Furthermore, even when one does not consider consequential activities such as cancer screening and airport security, most human-computer interactions involve some search component, and the impact of even small improvements is magnified by a high frequency of use. This thesis explores how some of these issues affect individuals' and teams' ability to perform visual searching tasks. Considerations such as group size and display configuration are examined, as are novel interfaces that aid search on tabletop and wall displays. The overarching goal of this work is to provide system designers with immediately implementable advice and guidelines on how to improve their systems in respect to visual search and to outline further research in this critical area.

Visual search is an important component of many human-computer interactions and a critical task in a variety of domains. These include screening for prohibited items in x-ray imagery of luggage in airports, identifying anomalies in medical imagery, and looking for unusual activity and changes over time in satellite imagery. Unlike many other human-machine interface tasks where a small error rate is quite acceptable, a crucial element of visual search tasks in these application areas is that a single mistake can have catastrophic consequences. Thus, any improvements in technology, interfaces, or human processes that can be made to aid visual search would be invaluable in these safety critical areas. Furthermore, even when one does not consider consequential activities such as cancer screening and airport security, most human-computer interactions involve some search component, and the impact of even small improvements is magnified by a high frequency of use. This thesis explores how some of these issues affect individuals' and teams' ability to perform visual searching tasks. Considerations such as group size and display configuration are examined, as are novel interfaces that aid search on tabletop and wall displays. The overarching goal of this work is to provide system designers with immediately implementable advice and guidelines on how to improve their systems in respect to visual search and to outline further research in this critical area.

Aim: The aim of the present study was to examine the outcome of repetitive-part training and dual-task training in the acquisition of dribbling technique and decision making in prepubescent novice soccer players. Method: Twenty four preschool pupils, with an age of 6.2 ± 0.4 years (mean ± SD) and with no previous experience of partaking in any organized team or ball sport participated in the study. The participants were tested (pre test) for initial performance level of dribbling technique and decision making and were then divided into two balanced experimental groups. The repetitive-part group underwent four weeks of part-task training (isolated dribbling-technique training) which were followed by four weeks of dual-task training (simultaneous training of dribbling technique and decision making). The dual-task group underwent eight weeks of dual-task training. The outcome in acquisition of dribbling technique and decision making was examined after four (mid test) and eight (post test) weeks of training. Both training regimes were followed by four weeks of non-training and thereafter a test (retention test) of the retention effects was carried out. Effect sizes (d) of the acquisition within each group and of the difference in performance and acquisition between the two groups were evaluated based on Cohen's effect size criteria (|d| > 0.2, small effect; |d| > 0.5, moderate effect; |d| > 0.8, large effect). Results: Between the pre and mid test, both groups showed large improvements in dribbling technique. The corresponding improvement was small in the repetitive-part group and moderate in the dual-task group between the mid and post test. There were large differences between the groups' decision-making performances at the mid, post, and retention test where the dual-task group was superior to the repetitive-part group. The dual task group showed large improvements of decision making between the pre and mid test. Between the mid and post test, the corresponding effect was large within the repetitive-part group and moderate in the dual-task group. The improvement of the dual-task group's decision making between the pre and mid test was greater than the corresponding improvement showed by the repetitive-part group, between the mid and post test. This difference produced a moderate effect size. Conclusions: These findings challenge conventional repetitive-part training regimes and, with references to previous research, emphasize the importance of dual-task training for skill acquisition in prepubescent novice soccer players.

I investigated the effects of probability on visual search. Previous work has shown that people can utilize spatial and sequential probability information to improve their performance on visual attention tasks. My task was a simple visual search in which the target was always present among a field of distractors, and could take one of two colors. The absolute probability of the target being either color was 0.5; however, the conditional probability – the likelihood of a particular color given a particular cue combination – varied from 0.1 to 0.9. I found that participants searched more efficiently for high conditional probability targets and less efficiently for low conditional probability targets. This modulation of efficiency was reduced or abolished when participants were not explicitly informed of the cue-target relationships. After establishing this effect, I investigated its mechanism using eye tracking methods. Early in trials, participants fixated preferentially, but not exclusively, on areas of the screen which contained predominantly stimuli of the color to which they had been cued. As the trial progressed, this color bias shifted to the target color. I conclude that search efficiency is modulated by the conditional probability of target features and that this is a top-down process that benefits from explicit knowledge of the probabilistic relationship between cues and targets, and that the modulation is a result of more efficient eye movements towards stimuli with a greater probability of being the target of search.

Previous research has shown that simultaneous auditory identification of the target in a visual search task can lead to more efficient (i. e. ?flatter?) search functions (Spivey et al. , 2001). Experiment 1 replicates the paradigm of Spivey et al. , providing subjects with auditory identification of the search target either before (<em>Consecutive</em> condition) or simultaneously with (<em>Concurrent</em> condition) the onset of the search task. RT x Set Size slopes in the <em>Concurrent</em> condition are approximately 1/2 as steep as those in the <em>Consecutive</em> condition. Experiment 2 employs a distractor ratio manipulation to test the notion that subjects are using the simultaneous auditory target identification to ?parse? the search set by colour, thus reducing the search set by 1/2. The results of Experiment 2 do not support the notion that subjects are parsing the search set by colour. Experiment 3 addresses the same question as Experiment 2, but obtains the desired distractor ratios by holding the amount of relevantly-coloured items constant while letting overall set size vary. Unlike Experiment 2, Experiment 3 supports the interpretation that subjects are using the auditory target identification to parse the search set.

Three experiments examine how collaboration influences visual search performance. Working with a partner or on their own, participants reported whether a target was present or absent in briefly presented search displays. The search performance of individuals working together (collaborative pairs) was compared to the pooled responses of the individuals working alone (nominal pairs). Collaborative pairs were less likely than nominal pairs to correctly detect a target and they were less likely to make false alarms. Signal detection analyses revealed that collaborative pairs were more sensitive to the presence of the target and had a more conservative response bias than the nominal pairs. This pattern was observed when the search difficulty was increased and when the presence of another individual was matched across pairs. The results are discussed in the context of task sharing, social loafing and current theories of visual search.

The goal of this dissertation is to take initial steps towards understanding concealment behaviour and ultimately developing a theory of visual concealment. Since there are relatively few studies of concealment in the literature and given the natural relationship between search and concealment, five strategies used in the development of traditional visual search theory and scene-based search theory were applied to the study of concealment. These strategies are: 1) establish a methodology, 2) identify dimensions, 3) categorize dimensions, 4) prioritize dimensions, and 5) integrate results into a theoretical framework that may involve inferences about the mechanisms involved. In Chapter 2, participants placed target objects within luggage in locations that were easy or hard to find (i.e., the placement task). Participants’ subjective reports of their thought processes and strategies were analyzed to identify dimensions that are important during concealment in real-world settings. Once a list of dimensions was generated, the dimensions were then categorized into three categories: Stimulus Properties dimensions such as visual similarity, Embodiment dimensions such as confrontation, and Higher Order dimensions such as schema. In Chapter 3, the dimensions uncovered in Chapter 2 were used in a forced-choice task, and participants’ choices were evaluated to determine whether the dimensions affected hiding behaviour. To further develop the methodological techniques available to study concealment behaviour and to examine the generalizability of previous findings, in Chapter 4, the placement task was used in another context – an office environment – and the locations chosen by participants to make objects easy or hard to find were coded on relevant dimensions. In Chapter 5, an initial attempt was made to prioritize the dimensions. The forced-choice task was used in a new way to explore the relative importance of the dimensions by examining which dimensions participants chose to use over other dimensions. Finally in the General Discussion in Chapter 6, an attempt was made to integrate available results and previous theories and to make inferences about the mechanisms involved in visual concealment. Methodological considerations and future directions for the study of visual concealment are also discussed.

Nearly all imaginable human activities rest on a context-appropriate dynamic control of the flow of retinal data into the nervous system via eye movements. The brain’s task is to move the eyes so as to exert intelligent predictive control over the informational content of the retinal data stream. An intelligent oculomotor controller would first model future contingent upon each possible next action in the oculomotor repertoire, then rank-order the repertoire by assigning a value v(a,t) to each possible action a at each time t, and execute the oculomotor action with the highest predicted value each time. We present a striking evidence of such an intelligent neural control of human eyes in a laboratory task of visual search for a small target camouflaged by a natural-like stochastic texture, a task in which the value of fixating a given location naturally corresponds to the expected information gain about the unknown location of the target. Human searchers behave as if maintaining a map of beliefs (represented as probabilities) about the target location, updating their beliefs with visual data obtained on each fixation optimally using the Bayes Rule. On average, human eye movement patterns appear remarkably consistent with an intelligent strategy of moving eyes to maximize the expected information gain, but inconsistent with the strategy of always foveating the currently most likely location of the target (a prevalent intuition in the existing theories). We derive principled, simple, accurate, and robust mathematical formulas to compute belief and information value maps across the search area on each fixation (or time step). The formulas are exact expressions in the limiting cases of small amount of information extracted, which occurs when the number of potential target locations is infinite, or when the time step is vanishingly small (used for online control of fixation duration). Under these circumstances, the computation of information value map reduces to a linear filtering of beliefs on each time step, and beliefs can be maintained simply as running weighted averages. A model algorithm employing these simple computations captures many statistical properties of human eye movements in our search task. / text

The goal of this dissertation is to take initial steps towards understanding concealment behaviour and ultimately developing a theory of visual concealment. Since there are relatively few studies of concealment in the literature and given the natural relationship between search and concealment, five strategies used in the development of traditional visual search theory and scene-based search theory were applied to the study of concealment. These strategies are: 1) establish a methodology, 2) identify dimensions, 3) categorize dimensions, 4) prioritize dimensions, and 5) integrate results into a theoretical framework that may involve inferences about the mechanisms involved. In Chapter 2, participants placed target objects within luggage in locations that were easy or hard to find (i.e., the placement task). Participants’ subjective reports of their thought processes and strategies were analyzed to identify dimensions that are important during concealment in real-world settings. Once a list of dimensions was generated, the dimensions were then categorized into three categories: Stimulus Properties dimensions such as visual similarity, Embodiment dimensions such as confrontation, and Higher Order dimensions such as schema. In Chapter 3, the dimensions uncovered in Chapter 2 were used in a forced-choice task, and participants’ choices were evaluated to determine whether the dimensions affected hiding behaviour. To further develop the methodological techniques available to study concealment behaviour and to examine the generalizability of previous findings, in Chapter 4, the placement task was used in another context – an office environment – and the locations chosen by participants to make objects easy or hard to find were coded on relevant dimensions. In Chapter 5, an initial attempt was made to prioritize the dimensions. The forced-choice task was used in a new way to explore the relative importance of the dimensions by examining which dimensions participants chose to use over other dimensions. Finally in the General Discussion in Chapter 6, an attempt was made to integrate available results and previous theories and to make inferences about the mechanisms involved in visual concealment. Methodological considerations and future directions for the study of visual concealment are also discussed.

This work aims to integrate Byrne’s theory of visual salience computation (2006) with Salvucci’s model of eye movements (2001) by testing participants on a visual search task similar to Findlay (1997). By manipulating the number, salience, and spacing of targets, participants exhibited the global effect averaging phenomena during the first recorded saccade, whereby short‐latency saccades land in between adjacent objects. Previous work has argued that the saccadic targeting system causing the averaging is influenced both by the salience and arrangement of objects displayed (Rao, Zelinsky, Hayho, & Ballard, 2002). However, to accurately account for these results, we did not have to couple the salience system with the saccadic targeting system. Instead, the systems work sequentially and in isolation, whereby the salience system simply hands off the next object to examine to the targeting system, whose accuracy depends only on saccadic latency and the location of the targeted and non‐targeted items.