Educational and instructional research has provided contrasting results regarding the best representation of numerical information, with the two most common being tabular and graphical representations. This motivated us to examine the issue using a novel approach. We employed electroencephalography (EEG) with event-related synchronisation and desynchronisation (ERS/ERD) to model cognitive load, and alpha and beta band powers to model engagement. We conducted an experiment to measure the cognitive load and engagement of 48 Oregon State University students and compared their performances with respect to these two representations. Structural equation models (SEMs) were constructed to investigate the potential mediation of learning performance by engagement and cognitive load. The results indicate that learning performance was fully mediated by cognitive load and engagement. Relative to graphs, tables produced a higher cognitive load and engagement, and subsequently, greater overall learning performance. Current results showed that different representations could yield significant differences in learning performance, and that an understanding of representation-elicited affective behaviours has considerable potential for future online learning instructional design.

Signaling can be used to guide learners’ attention to relevant details of instructional materials, and consequently can foster multimedia learning. We investigate how signaling, and highlighting specifically, are predictive of individual learning performance in short-term knowledge acquisition, mediated by cognitive load and visual attention employing biometrically using electroencephalog- raphy (EEG) and eye-tracking measures. We perform a two-level experiment on short-term knowledge acquisition within the context of an online course, with content-highlighting employed as signaling cues. We use a structural equation model (SEM) to examine the potential mediation that cognitive load and visual attention may have on knowledge acquisition. The results suggest that signaling can direct attention to the crucial areas and potentially reduce the extraneous cognitive load, thereby promoting short-term knowledge acquisition, as a key stage of learning.

Drones have been used increasingly to aid in Industry 4.0 activities including inspection of the nation’s infrastructure. We investigate several potential underlying affective behaviors related to drone pilot skill acquisition, with the eventual goal of developing methods to enhance human performance. We employ Electroencephalography (EEG) and Eye tracking instrumentation to measure human affect in a series of simulated drone piloting experiments to examine performance using behavioral variables, controller input variables, as well as measures of individual cognitive ability. Current results show that task difficulty impacts the performance/learning process and varies by the nature of the task. The behavioral and biometric measures associated with performance/learning varied significantly among activities. We conclude that drone specifications and training requirements can and should be calibrated to the drone mission. In addition to developing specifications and training requirements, psychological and behavioral measures can also serve as theoretical foundations for modeling complex tasks.

Online learning has become increasingly commonplace, including the replacement and augmentation of traditional in-residence education, as well as ad-hoc training systems and just-in-time knowledge dissemination. However, design for instructional media to facilitate performance has relied on an inadequate understanding of the behavioral affect elicited from these designs. To investigate the degree to which excitement and engagement are predictive of individual learning outcomes in an online learning setting, we evaluate an experiment with two instructional representations (static and dynamic) using learning materials for a semaphore signaling system. We examine the excitement and engagement levels of the participants using electroencephalography (EEG) as potential mediators. Learning outcomes are measured by the evaluation scores of knowledge retention. We consider several structural equation models (SEMs) to see the underlying relationship between instructional representation, excitement, engagement, and learning outcomes. Notably, the models indicate the full mediation of behavioral affect on learning outcomes, in which both the range of excitement and the maximum level of engagement are mediating effects. This article illustrates the potential for biometrically measured affect aid in the modeling and understanding of how instructional design features ultimately impact performance.

Time series data are often found in diverse fields, such as science, business, medicine and engineering. In this paper, we focus on sequential pattern mining for categorical time series datasets that contain multiple independent timeseries. Frequent patterns are considered important in many applications. However, collected data are generally afflicted with noise. Conventional sequential pattern mining methods that use exact matching may meet difficulties in mining databases with long sequences and noise. We propose a framework that uncovers frequent approximate sequential patterns with multiple widths. A mined pattern in this framework is a representative of a group of sequences (with various widths) that follow the pattern's event flow order. The presentation of the patterns also gives insight into the occurrence of the pattern longitudinally and across the population. The pattern can be recognized as a common pattern across the multiple time series, time, or both. We name this novel framework MWASP: Multiple-Width Approximate Sequential Patterns.

We consider the problem of routing a vehicle making multiple intermediate stops, assuming a non-order-preserving, multiattribute reward structure. Sub-paths of optimal paths may not be optimal for such a reward structure, which may result from routing a pick-up and delivery vehicle carrying hazardous materials that is routed on the basis of minimizing cost and risk. We assume that a priori bounds exist on the rewards from the vehicle's current position to each of the intermediate destinations and to the depot through all the intermediate destinations that have yet to be visited. Precise calculations of these rewards would require additional computational effort. Two heuristic search algorithms, BU* and DU*, are developed and analyzed. Both algorithms satisfy termination, completeness, and admissibility properties. Results indicate that BU* is guaranteed to perform no worse given better heuristic information, a guarantee that cannot be made for DU*. Computational requirements are illustrated through examples based on a real network in northeast Ohio.