NPCEditor is a system for building a natural language processing component for virtual humans capable of engaging a user in spoken dialog on a limited domain. It uses a statistical language classification technology for mapping from user's text input to system responses. NPCEditor provides a user-friendly editor for creating effective virtual humans quickly. It has been deployed as a part of various virtual human systems in several applications.

Modeling human behavior requires vast quantities of accurately labeled training data, but for ubiquitous people-aware applications such data is rarely attainable. Even researchers make mistakes when labeling data, and consistent, reliable labels from low-commitment users are rare. In particular, users may give identical labels to activities with characteristically different signatures (e.g., labeling eating at home or at a restaurant as "dinner") or may give different labels to the same context (e.g., "work" vs. "office"). In this scenario, labels are unreliable but nonetheless contain valuable information for classification. To facilitate learning in such unconstrained labeling scenarios, we propose Community-Guided Learning (CGL), a framework that allows existing classifiers to learn robustly from unreliably-labeled user-submitted data. CGL exploits the underlying structure in the data and the unconstrained labels to intelligently group crowd-sourced data. We demonstrate how to use similarity measures to determine when and how to split and merge contributions from different labeled categories and present experimental results that demonstrate the effectiveness of our framework.

Over the past decade, we have been engaged in an extensive research effort to build virtual humans and applications that use them. Building a virtual human might be considered the quintessential AI problem, because it brings together many of the key features, such as autonomy, natural communication, sophisticated reasoning and behavior, that distinguish AI systems. This paper describes major virtual human systems we have built and important lessons we have learned along the way.

Over the past decade, we have been engaged in an extensive research effort to build virtual humans and applications that use them.  Building a virtual human might be considered the quintessential AI problem, because it brings together many of the key features, such as autonomy, natural communication, sophisticated reasoning and behavior, that distinguish AI systems.  This paper describes major virtual human systems we have built and important lessons we have learned along the way.

How is shape represented during spatial tasks such as mental rotation? This research investigated the format of mental representations of 3-D shapes during mental rotation. Specifically, we tested the extent to which visual information, such as color, is represented during mental rotation using methods ranging from reaction time studies, verbal protocol analysis, and eyetracking. Another set of studies examined whether people use piecemeal or holistic strategies to rotate complex objects. Results show that individuals with good rotation ability do not represent color during mental rotation and rotate whole shapes; whereas poor rotators do represent color and rotate individual pieces of the shape using piecemeal strategies. This work contributes to theories about cognitive shape processing by showing that different information processing strategies may be one cause of individual differences in mentally rotation performance.

Representing and comparing two-dimensional shapes is an important problem. Our hypothesis about human representations is that that people utilize two representations of shape: an abstract, qualitative representation of the spatial relations between the shape’s parts, and a detailed, quantitative representation. The advantage of relational, qualitative representations is that they facilitate shape comparison: two shapes can be compared via structural alignment processes which have been used to model similarity and analogy more broadly. This comparison process plays an important role in determining when two objects share the same shape, or in identifying transformations (rotations and reflections) between two shapes. Based on our hypothesis, we have built a computational model which automatically constructs both qualitative and quantitative representations and uses them to compare two-dimensional shapes in visual scenes. We demonstrate the effectiveness of our model by summarizing a series of studies which have simulated human spatial reasoning.

This paper describes some functionalities and features of the Maryland Virtual Patient (MVP) environment. MVP models the process of disease progression, diagnosis and treatment in virtual patients who are endowed with a “body,” a simulation of their physiological and pathological processes, and a “mind,” a set of capabilities of perception, reasoning and action that allow the virtual patient to exhibit independent behavior, participate in a natural language dialog, remember events, hold beliefs about other agents and about specific object and event instances, make decisions and learn.

Time perception and inferences there from are of critical importance to many autonomous agents. But time is not perceived directly by any sensory organ. We argue that time is constructed by cognitive processes. Here we present a model for time perception that concentrates on succession and duration, and that generates these concepts and others, such as continuity, immediate present duration, and lengths of time. These concepts are grounded through the perceptual process itself. The LIDA cognitive model is used to illustrate these ideas.

In this paper, we present a cognitive model of frequency-based decision-making applied to the task of landmine detection. The model is implemented in the ACT-R cognitive architecture and is strongly constrained by the cognitive primitives of the architecture. We then generalize the model to another task in the domain of macroeconomic decision-making using the same architecture, pursuing theoretical parsimony. We describe each model's representation requirements, assess their fits to the data, and analyze their performance scaling as a function of task and architectural parameters. Efforts to generalize the landmine detection model to macroeconomic decision making showed that reasonable fits to the macro-economic performance data could be achieved by models based either on procedural knowledge or declarative knowledge. This finding underscores the importance of distinguishing between processing strategies employed to execute tasks. Such detail appears needed to understand the neural foundations of frequency-based decision-making.

A human subject experiment was conducted to investigate caffeine’s effect on appraisal and performance of a mental serial subtraction task. Serial subtraction performance data was collected from three treatment groups: placebo, 200, and 400 mg caffeine. The data were analyzed by caffeine treat ment group and how subjects appraised the task (as challenging or threatening). A cognitive model of the serial subtraction task was developed. The model was fit to the human performance data using a parallel genetic algorithm. How the model’s parameters change to fit the data suggest how cognition changes due to caffeine and appraisal. Over all, the cognitive modeling and optimization results suggest that the speed of vocalization varies the most along with changes to declarative memory. This approach provides a way to compute how cognitive mechanisms change due to moderators.