Scott Bolland

A Theory and Model of Conflict Detection in Air Traffic Control: Incorporating Environmental Constraints

A performance theory for conflict detection in air traffic control is presented that specifies how controllers adapt decisions to compensate for environmental constraints. This theory is then used as a framework for a model that can fit controller intervention decisions. The performance theory proposes that controllers apply safety margins to ensure separation between aircraft. These safety margins are formed through experience and reflect the biasing of decisions to favor safety over accuracy, as well as expectations regarding uncertainty in aircraft trajectory. In 2 experiments, controllers indicated whether they would intervene to ensure separation between pairs of aircraft. The model closely predicted the probability of controller intervention across the geometry of problems and as a function of controller experience. When controller safety margins were manipulated via task instructions, the parameters of the model changed in the predicted direction. The strength of the model over existing and alternative models is that it better captures the uncertainty and decision biases involved in the process of conflict detection. (PsycINFO Database Record (c) 2009 APA, all rights reserved).

Target similarity effects: Support for the parallel distributed processing assumptions.

Recent research has begun to provide support for the assumptions that memories are stored as a composite and are accessed in parallel (Tehan & Humphreys, 1998). New predictions derived from these assumptions and from the Chappell and Humphreys (1994) implementation of these assumptions were tested. In three experiments, subjects studied relatively short lists of words. Some of the lists contained two similar targets (thief andtheft) or two dissimilar targets (thief andsteal) associated with the same cue (ROBBERY). As predicted, target similarity affected performance in cued recall but not free association. Contrary to predictions, two spaced presentations of a target did not improve performance in free association. Two additional experiments confirmed and extended this finding. Several alternative explanations for the target similarity effect, which incorporate assumptions about separate representations and sequential search, are rejected. The importance of the finding that, in at least one implicit memory paradigm, repetition does not improve performance is also discussed.

What You Get Out of Memory Depends on the Question You Ask

Following study, participants received 2 tests. The 1st was a recognition test; the 2nd was designed to tap recollection. The objective was to examine performance on Test 1 conditional on Test 2 performance. In Experiment 1, contrary to process dissociation assumptions, exclusion errors better predicted subsequent recollection than did inclusion errors. In Experiments 2 and 3, with alternate questions posed on Test 2, words having high estimates of recollection with one question had high estimates of familiarity with the other question. Results supported the following: (a) the 2-test procedure has considerable potential for elucidating the relationship between recollection and familiarity; (b) there is substantial evidence for dependency between such processes when estimates are obtained using the process dissociation and remember-know procedures; and (c) order of information access appears to depend on the question posed to the memory system.

Development and Validation of a Multilevel Model for Predicting Workload Under Routine and Nonroutine Conditions in an Air Traffic Management Center

Objective: The aim of this study was to develop a model capable of predicting variability in the mental workload experienced by frontline operators under routine and nonroutine conditions. Background: Excess workload is a risk that needs to be managed in safety-critical industries. Predictive models are needed to manage this risk effectively yet are difficult to develop. Much of the difficulty stems from the fact that workload prediction is a multilevel problem. Method: A multilevel workload model was developed in Study 1 with data collected from an en route air traffic management center. Dynamic density metrics were used to predict variability in workload within and between work units while controlling for variability among raters. The model was cross-validated in Studies 2 and 3 with the use of a high-fidelity simulator. Results: Reported workload generally remained within the bounds of the 90% prediction interval in Studies 2 and 3. Workload crossed the upper bound of the prediction interval only under nonroutine conditions. Qualitative analyses suggest that nonroutine events caused workload to cross the upper bound of the prediction interval because the controllers could not manage their workload strategically. Conclusion: The model performed well under both routine and nonroutine conditions and over different patterns of workload variation. Application: Workload prediction models can be used to support both strategic and tactical workload management. Strategic uses include the analysis of historical and projected workflows and the assessment of staffing needs. Tactical uses include the dynamic reallocation of resources to meet changes in demand.

Using Maintenance Rehearsal to Explore Recognition Memory

We examined associative and item recognition using the maintenance rehearsal paradigm. Our intent was to control for mnemonic strategies; to produce a low, graded level of learning; and to provide evidence of the role of attention in long-term memory. An advantage for low-frequency words emerged in both associative and item recognition at very low levels of learning. This early emergence casts doubt on explanations based on the traditional concept of recollection. A comparison of false alarms supports a role for item information or the joint use of cues but not familiarity in producing associative false alarms. We may also have found a way to measure the amount of attention being paid to a to-be-learned item or pair, independently of memory performance on the attended item. This result may be an important step in determining whether coherent theories about the role of attention in long- and short-term memory can be created. (PsycINFO Database Record (c) 2009 APA, all rights reserved).

Evaluation of a model of expert decision making in air traffic control

The resolution of conflicts between aircraft is regarded as a very complex optimisation problem, yet air traffic controllers are able to perform the task with high rates of success under demanding circumstances. Much of the expertise of the air traffic controller appears to lie in the ability to select the appropriate strategy for the problem. In this project, we describe a model of expert decision making for the air traffic control conflict resolution task. A key assumption of the approach is that controllers adopt relatively simple heuristics to solve the complex trajectory problem. We model expert decision making as a serial search process in a hierarchical tree, in which the selection of a decision option for further evaluation is constrained by the situation. This is part of a broader project that is developing a new approach for simulating the tasks that a human operator performs, and the workload that the human experiences while carrying out those tasks. In this paper, we will present an analysis of the behaviour of an initial version of this model. The key aim of this analysis is to compare the models behaviour against the behaviour of expert controllers under varying scenario complexity. The analysis relies on both data from our model simulation runs and recordings of the activities of 14 air traffic controllers. It is based on static aircraft scenarios of varying complexity.

Exploring the Periphery of Knowledge by Intrinsically Motivated Systems

Intrinsically motivated learning is essential for the development of a wide range of competences. However, the neural substrate for the motivational signal as well as how this signal facilitates the processes of building competences are poorly understood. In this paper we exploit a biologically plausible approach, showing that an intrinsically motivated system where the motivation depends on stimulus familiarity as an inverted U-shape, exhibits well-structured exploration behaviour. Furthermore, we show that such behaviour may lead to the emergence of complex competences such as object affordances.

Dissociable Forms of Repetition Priming: A Computational Model

Nondeclarative memory and novelty processing in the brain is an actively studied field of neuroscience, and reducing neural activity with repetition of a stimulus (repetition suppression) is a commonly observed phenomenon. Recent findings of an opposite trend—specifically, rising activity for unfamiliar stimuli—question the generality of repetition suppression and stir debate over the underlying neural mechanisms. This letter introduces a theory and computational model that extend existing theories and suggests that both trends are, in principle, the rising and falling parts of an inverted U-shaped dependence of activity with respect to stimulus novelty that may naturally emerge in a neural network with Hebbian learning and lateral inhibition. We further demonstrate that the proposed model is sufficient for the simulation of dissociable forms of repetition priming using real-world stimuli. The results of our simulation also suggest that the novelty of stimuli used in neuroscientific research must be assessed in a particularly cautious way. The potential importance of the inverted-U in stimulus processing and its relationship to the acquisition of knowledge and competencies in humans is also discussed.