Here is a plain-text order form to order reports by mail. Some of the following reports can be downloaded as PostScript, or compressed PostScript files.
Joe Chase, Deborah Hix, David Tate, and James Templeman
"Use of the User Action Notation at the Naval Research Laboratory
Human-Computer Interaction Laboratory," Naval Research Laboratory Memorandum
Report, NRL/MR/5530--94-7488, June 30, 1994, (NCARAI Report: AIC-94-015).
Not available on-line at this time. Please see order form.
Abstract:
This report describes introduction of researchers in the
Information Technology Division of the Naval Research Laboratory (NRL) to
the User Action Notation (UAN), a notation developed at Virginia Tech for
representing the design of the interaction component of interactive
systems. This report also presents use of the UAN in describing a variety
of unique, innovative interaction techniques to evaluate the notation for
its ability to represent such techniques, and explores the possibilities
for future research and technology transition with the UAN collaboratively
between NRL and Virginia Tech.
Abstract:
Following a brief introduction to the Naval Research Laboratory
(NRL), specific ongoing research in speech and human language technology
will be presented. Areas of research fall across two laboratories in the
Information Technology Division, the Human Computer Interaction Laboratory
(HCI) and the Navy Center for Applied Research in Artificial Intelligence
(NCARAI). Speech research focuses on narrowband algorithm development and
uses human based intelligibility to evaluate success. Demonstrations of
results will be played. Navy has been a prominent participant in
evaluation of narrowband speech. Current efforts on human-human
communication evaluation will provide a basis for new approaches to
multi-mode interactions which include speech interfaces. Finally, a video
of ongoing efforts at NCARAI on the EUCALYPTUS system which include
graphics and spoken language interaction will be shown.
Abstract:
How can we use multiple channels of communication to enhance the
bi-directional communication between human and machine? This question
forms the premise of a new research effort at the Naval Research
Laboratory called the Human Machine Dialogue Project. The goal is to
use human-human interactions during decision making problem solving to
develop human-machine interactions which can maximize the interchange of
information in a beneficial way between the computational counterpart of
the team and the user. Aspects being studied include defining new modes
of interaction using multi-odal pathways, assessing how and when these are
appropriate to use and under what contexts they are best used alone.
Evaluation forms a pivotal aspect of the initial study, design, and
implementation and is used to determine what is a "better solution" for a
given task. The effort assumes that user models provide crucial
information about organizational principles that can be instantiated in a
module which is independent from an application, a UIMS. How to represent
user information, communication information, and application information,
what to represent in each, and how to integrate their use is a focal
research consideration in the work.
Abstract:
The problem of human-computer interaction can be viewed as two powerful
information processors (human and computer) attempting to communicate with each
other via a narrow-bandwidth, highly constrained interface (Tufte, 1989). To
address it, we seek faster, more natural, and more convenient means for users
and computers to exchange information. The user's side is constrained by the
nature of human communication organs and abilities; the computer's is
constrained only by input/output devices and interaction techniques that we can
invent. Current technology has been stronger in the computer-to-user direction
than user-to-computer, hence today's user-computer dialogues are rather
one-sided, and the bandwidth from the computer communication medium can help
redress this imbalance. This chapter describes the relevant characteristics of
the human eye, eye tracking technology, how to design interaction techniques
that incorporate eye movements into the use-computer dialogue in a convenient
and natural way, and the relationship between eye movement interfaces and
virtual environments.
Abstract:
Current input device taxonomies and other frameworks typically
emphasize the mechanical structure of input devices. We suggest that
selecting an appropriate input device for an interactive task requires
looking beyond the physical structure of devices to the deeper perceptual
structure of the task, the device, and the interrelationship between the
perceptual structure of the task and the control properties of the device.
We affirm that perception is key to understanding performance of
multidimensional input devices on multidimensional tasks. We have
therefore extended the theory of processing of perceptual structure to
graphical interactive tasks and to the control structure of input devices.
This allows us to predict task and device combinations that lead to better
performance and hypothesize that performance is improved when the
perceptual structure of the task matches the control structure of the
device. We conducted an experiment in which subjects performed two tasks
with different perceptual structures, using two input devices with
correspondingly different control structures, a three-dimensional tracker
and a mouse. We analyzed both speed and accuracy, as well as the
trajectories generated by subjects as they used the unconstrained
three-dimensional tracker to perform each task. The results support our
hypothesis and confirm the importance of matching the perceptual
structure of the task and the control structure of the input device.
Abstract:
Communications are a crucial aspect of military decision making.
NATO restricted brevity codes are often used in verbal communications of
AEGIS team members during training sessions and in the fleet. An
understanding of the relationship between brevity code use and
communication performance during AEGIS team training is important for
brevity code training issues. We recorded the internal Combat Information
Center (CIC) communications during AEGIS team training exercises and
performed an analysis on speech turns that contained brevity codes. (U)
Brevity codes were identified, categorized, and analyzed and on average,
58% of the brevity codes used were used as defined. The remaining were
used consistently as ordinary English or in a manner we designated as
"Professional" Brevity codes were categorized as Professional if they were
used in a consistent manner by team members in what appeared to be a
mutually accepted and consistent meaning different from either the NATO or
English meanings.
The title is classified. Under limitations on distribution: DoD & DoD
contractors only; test and evaluation; Feb 1994. Other requests shall be
referred to : Commanding Officer, NRL
Abstract:
The potential to deliver information through sound is rapidly
expanding with new technology, new techniques, and significant advances in
our understanding of hearing. Although these changes raise important new
issues about the design of sound delivery systems, there is already a wide
range of knowledge scattered through different disciplines about
communicating information through nonspeech sound such as sonification. An
overview of how sound can deliver information is presented using a
framework of linguistic analogies. Areas that will be discussed in some
detail include contextual and expectance effects, which operate when tonal
sounds as well as realistic sounds are interpreted.
Abstract:
The Human-Computer (HCI) Laboratory at the Naval Research
Laboratory (NRL) is developing high-performance interactive computer
systems for use in Naval command and control applications. New technology
allows computerized systems to make greater use of a person's natural
physical, perceptual, and cognitive skills. These systems make it faster
and more straightforward for users to access and enter information within
the context of time critical, real world situations.
Our work focuses on developing novel interaction techniques-ways
of using physical input and output devices to perform user tasks. Previous
work has demonstrated the effectiveness of individual interaction
techniques. This study is now integrating these techniques into a
command and control testbed, that will be tested through empirical
user-based evaluation.
Such new capabilities increase the complexity of software
development and maintenance requirements of already complex systems. New
software architecture's must co-evolve with the technology to structure
and manage these additional requirements. This report describes the new
demand s placed on interactive systems and explains how we are addressing
this challenge, through development of a flexible software architecture.
Abstract:
Data are presented from two experiments related to accurate
identification of everyday sounds: what aspects of a complex sound with
several transients are important for accurate identification, and how much
of a sound must be heard to identify it. The first experiment was
conducted to examine further what parts of a sound contribute to accurate
identification, or conversely, confusion and ambiguity. The sounds used
were a light switch and stapler and were chosen because they include a
pattern of impacts that is somewhat similar. Multiple exemplars of each
sound were produced by changing the instruments and circumstances of the
event. Results showed that there may be two levels of analysis necessary
in selecting and designing a sound for accurate identification. At one
level one is concerned with the general type of sound and what other
events it can signify. At another level, one is concerned with the details
of the sound. Good design of the details can reduce the ambiguity, but in
some cases ambiguity cannot be eliminated simply because two different
events can legitimately produce the same acoustic pattern. The second
experiment was conducted to get some preliminary data on how much of a
sound is needed for its identification. Although in the usual case, one
would generate the complete sound, it would be useful to know when
sufficient information has been presented for its identification. Although
in the usual case, one would generate the complete sound, it would be
useful to know when sufficient information has been presented for
identification. Forty-one sounds were presented for identification with
the duration of the sound successively incremented in 50 ms steps. The
results showed that listeners generally needed to hear very little of the
sound to make accurate decisions. Accuracy was above 70% for all but three
of the 41 test sounds. In general less than half of the sound needed to be
heard, and in many cases less than 20% of the sound was needed for an
accurate response.
Abstract:
We recently completed an initial study to develop a paradigm for
software prototyping tool evaluation. Prototyping tools are used to
produce "an executable unit that demonstrates particular aspects of the
behavior or functionality of the desired software product" (Lee, Blum,
Kanellakis, Crisp, and Caruso, 1994). Our study was conducted with a
particular perspective of assessing how the tools support an aspect of
software design that has received little attention: exploration of the
design space. In order to pursue this perspective, an initial definition
of design space exploration was developed and hypotheses were proposed on
what outcomes would be observed if a tool supported design space
exploration. Finally, techniques were designed to obtain data to test the
hypotheses in a general manner. Data were collected with these techniques
during a session in which a particular tool was employed to design
software.
Abstract:
Today's graphical user interfaces remember little from one
transaction to the next; each command exists nearly independently. Humans,
however, typically draw on previous elements of a dialogue in their
communications. We are seeking to add some of the characteristics of human
dialogues to graphical interfaces. This paper describes our research into
this problem and our initial results in answering three questions: What
are the appropriate analogues of conversational focus in a graphical
human-computer conversation? Where does this type of processing best fit
within a user interface management system paradigm? What mechanisms can be
used to realize it?
Abstract: This document describes how to use the Modechart Toolset (MT). MT is a set of tools designed to facilitate the specification, modeling, and analysis of real-time embedded systems using the Modechart language. MT supports the creation, modification, and storage of Modechart specifications. It also supports the analysis of Modechart specifications via a consistency and completeness checker, a simulator and a verifier.
Computational Reasoning
Intelligent M4 Systems
Machine Learning
Sensor-Based Systems
