Coordinating Human-Robot Teams with Dynamic and Stochastic Task Proficiencies

As robots become ubiquitous in the workforce, it is essential that human-robot collaboration be both intuitive and adaptive. A robot’s ability to coordinate team activities improves based on its ability to infer and reason about the dynamic (i.e., the “learning curve”) and stochastic task performance of its human counterparts. We introduce a novel resource coordination algorithm that enables robots to schedule team activities by (1) actively characterizing the task performance of their human teammates and (2) ensuring the schedule is robust to temporal constraints given this characterization. We first validate our modeling assumptions via user study. From this user study, we create a data-driven prior distribution over human task performance for our virtual and physical evaluations of human-robot teaming. Second, we show that our methods are scalable and produce high-quality schedules. Third, we conduct a between-subjects experiment (n = 90) to assess the effects on a human-robot team of a robot scheduler actively exploring the humans’ task proficiency. Our results indicate that human-robot working alliance ( p\lt 0.001 ) and human performance ( p=0.00359 ) are maximized when the robot dedicates more time to exploring the capabilities of human teammates.

Through (Tracking) Their Eyes: Abstraction and Complexity in Program Comprehension

Previous studies on writing and understanding programs presented evidence that programmers beyond a novice stage utilize plans or plan-like structures. Other studies on code composition showed that learners have difficulties with writing, reading, and debugging code where interacting plans are merged into a short piece of code. In this article, we focus on the question of how different code-composition strategies and the familiarity with code affect program comprehension on a more abstract, i.e., algorithmic level. Using an eye-tracking setup, we explored how advanced students comprehend programs and their underlying algorithms written in either a merged or abutted (sequenced) composition of code blocks of varying familiarity. The effects of familiarity and code composition were studied both isolated and in combination. Our analysis of the quantitative data adds to our understanding of the behavior reported in previous studies and the effects of plans and their composition on the programs’ difficulty. Using this data along with retrospective interviews, we analyze students’ reading patterns and provide support that subjects were able to form mental models of program execution during task performance. Furthermore, our results suggest that subjects are able to retrieve and create schemata when the program is composed of familiar templates, which may improve their performance; we found indicators for a higher element-interactivity for programs with a merged code composition compared to abutted code composition.

“Late-onset” ADHD symptoms in young adulthood: Is this ADHD?

Objective: We investigated whether “late-onset” ADHD that emerges in adolescence/adulthood is similar in risk factor profile to: (1) child-onset ADHD, but emerges later because of scaffolding/compensation from childhood resources; and (2) depression, because it typically onsets in adolescence/adulthood and shows symptom and genetic overlaps with ADHD. Methods: We examined associations between late-onset ADHD and ADHD risk factors, cognitive tasks, childhood resources and depression risk factors in a population-based cohort followed-up to age 25 years ( N=4224–9764). Results: Parent-rated late-onset ADHD was like child-onset persistent ADHD in associations with ADHD polygenic risk scores and cognitive task performance, although self-rated late-onset ADHD was not. Late-onset ADHD was associated with higher levels of childhood resources than child-onset ADHD and did not show strong evidence of association with depression risk factors. Conclusions: Late-onset ADHD shares characteristics with child-onset ADHD when parent-rated, but differences for self-reports require investigation. Childhood resources may delay the onset of ADHD.

Relevant Physiological Indicators for Assessing Workload in Conditionally Automated Driving, Through Three-Class Classification and Regression

In future conditionally automated driving, drivers may be asked to take over control of the car while it is driving autonomously. Performing a non-driving-related task could degrade their takeover performance, which could be detected by continuous assessment of drivers' mental load. In this regard, three physiological signals from 80 subjects were collected during 1 h of conditionally automated driving in a simulator. Participants were asked to perform a non-driving cognitive task (N-back) for 90 s, 15 times during driving. The modality and difficulty of the task were experimentally manipulated. The experiment yielded a dataset of drivers' physiological indicators during the task sequences, which was used to predict drivers' workload. This was done by classifying task difficulty (three classes) and regressing participants' reported level of subjective workload after each task (on a 0–20 scale). Classification of task modality was also studied. For each task, the effect of sensor fusion and task performance were studied. The implemented pipeline consisted of a repeated cross validation approach with grid search applied to three machine learning algorithms. The results showed that three different levels of mental load could be classified with a f1-score of 0.713 using the skin conductance and respiration signals as inputs of a random forest classifier. The best regression model predicted the subjective level of workload with a mean absolute error of 3.195 using the three signals. The accuracy of the model increased with participants' task performance. However, classification of task modality (visual or auditory) was not successful. Some physiological indicators such as estimates of respiratory sinus arrhythmia, respiratory amplitude, and temporal indices of heart rate variability were found to be relevant measures of mental workload. Their use should be preferred for ongoing assessment of driver workload in automated driving.

Continuous learning and employee performance: a moderated examination of managers' coaching behavior in India

PurposeBy integrating organizational support theory (OST) and social cognitive theory, this study investigates types of managers' coaching behavior as experienced by the employees. Furthermore, the study examines whether employees would exhibit greater task and contextual performance when organizational learning is blended with a specific coaching behavior of their manager.Design/methodology/approachUsing primary data from 298 software engineers working in select information technology companies across India, the current study attempts to assess moderating effect of managers' coaching behavior in two relationships, including continuous learning and employees' task performance (CL-TP) and continuous learning and employees' contextual performance (CL-CP).FindingsResult of exploratory factor analysis suggests that managers of select organizations exhibit two major types of coaching behavior: inspiration-based coaching behavior and facilitation-based coaching behavior. On the moderating role of coaching behavior, it is documented that facilitation-based coaching behavior significantly positively moderates both stated (CL-TP and CL-CP) relationships, whereas inspiration-based coaching behavior of supervisors has positive significant effect on CL-TP relationship but negatively moderates the CL-CP relationship.Research limitations/implicationsThe extent to which the findings of this study can be generalized is constrained by the limited sample and organizational context.Practical implicationsThe most important managerial implication for all learning organizations is that both kinds of coaching behaviors help improving the task performance of the employees, but managers should prefer facilitation-based coaching style in order to generate higher contextual performance of employees.Originality/valueThis study contributes to practitioners and existing literature by explaining how individual performance of employees is affected by the investment made by organizations in facilitating continuous learning.

Optimism and pessimism in optimised replay

The replay of task-relevant trajectories is known to contribute to memory consolidation and improved task performance. A wide variety of experimental data show that the content of replayed sequences is highly specific and can be modulated by reward as well as other prominent task variables. However, the rules governing the choice of sequences to be replayed still remain poorly understood. One recent theoretical suggestion is that the prioritization of replay experiences in decision-making problems is based on their effect on the choice of action. We show that this implies that subjects should replay sub-optimal actions that they dysfunctionally choose rather than optimal ones, when, by being forgetful, they experience large amounts of uncertainty in their internal models of the world. We use this to account for recent experimental data demonstrating exactly pessimal replay, fitting model parameters to the individual subjects’ choices.