bennett, k. b. and flach, j. m. (2012). visual momentum

Bennett, K. B. and Flach, J. M. (2012). Visual momentum redux. International Journal of Human Computer Studies, 70, 399-414.

Visual Momentum Redux

KEVIN B. BENNETT1 and JOHN M. Flach, Wright State University, Dayton, Ohio

Over 25 years ago Woods (1984) introduced the concept of visual momentum: the extent to

which an interface supports a practitioner in transitioning between various information-seeking

activities that are required for understanding and exploring work domains. Increasing visual

momentum requires the consideration of a range of “cognitive couplings” that span all levels of

the interface: between multiple screens, within individual screens, and within a display on a

screen. Although the concept has been well-received, we believe that its potential to improve the

quality of human computer interaction may be under-appreciated. Our purpose in this review is

to provide a better understanding of visual momentum: to provide concrete and diverse examples

of its successful application, to review empirical findings, to refine and expand the original

design techniques that were proposed, and to integrate diverse terms that appear across different

research communities.

1 Correspondence and request for reprints should be addressed to Kevin B. Bennett,

Psychology Department, 335 Fawcett Hall, Wright State University, Dayton, OH 45435 or e-

mailed to [email protected].

Running title: Visual Momentum

Key words: visual momentum, cognitive systems engineering, long shot, focus + context,

overview + detail, fish eye views, center surround, landmarks, ecological interface design

BENNETT & FLACH Visual Momentum Redux 1

mailto:[email protected]

mailto:[email protected]

1. Introduction

An area that has received little attention in the design of computer-based display

systems is how the user integrates data across successive displays (cf. Badre,

1982) ... Failure to consider the requirements for effective across-display user

information processing can produce devastating effects on user performance ...

The "getting lost" and "keyhole" phenomena are not inevitable consequences of

using computer-based displays; neither do they represent human limitations (for

example, short-term memory) that must be compensated for through memory aids

or walls of VDUs. Across-display processing difficulties are the result of a failure

to consider man and computer together as a cognitive system ... (Woods, 1984, pp.

229-230).

Today’s work domains are complicated, often involving large, complex workspaces and

databases. This is true for traditional work domains (e.g., process control, military command and

control) and work domains as more broadly defined (e.g., consumer electronics). Interaction in

these work settings has become increasingly “computer mediated” over time: the advent of

computational technologies has resulted in the computer interface becoming the practitioner’s

primary window on the world of work. Since the physical size of viewing surfaces in the

interface is very limited, this window is often an extremely small one.

This combination (i.e., complex work domains and inherently limited “display real estate”)

has had a fairly direct impact on the ability of practitioners to conduct work effectively. All of the

information required for effective control cannot possibly be displayed in parallel; the

practitioner must selectively view glimpses serially, over time, through the limited “keyhole”


provided by the small display surface (i.e., the “keyhole” effect, Woods, 1984). Thus,

information must often be remembered and mentally integrated across successive glimpses, an

activity that practitioners are not particularly well-equipped to handle.

A direct consequence is that practitioners must navigate these complicated workspaces and

databases to find information that is relevant for the task at hand. To navigate successfully, the

practitioner must know their current location within the workspace, the space of possible

locations that might be navigated to, and a navigational route that will get them there.

Unfortunately, information systems rarely provide interface resources that support these needs

and the end result is often what Woods (1984) referred to as the “getting lost” phenomenon: “...

users often do not know where they are, where they came from, or where they have to go

next” (Ziefle and Bay, 2006, p. 395).

We recently had the opportunity to observe an example at first hand. In 2007 we

participated in exercises for modern air combat operations. The work setting was large

(auditorium-sized), collaborative (nearly 100 people) and it featured cutting-edge information

technology (e.g., display walls, chat rooms, etc.) Our observations include the following. Even

simple operations often required practitioners to navigate across multiple display windows and to

integrate information (sometimes presented in different formats). Multiple overlapping windows

occluded, and sometimes even locked out, critical information. Practitioners were unable to

perform reasonably simple (and critical) tasks in some instances because of the lack interface

support. “Work-arounds” (e.g., jotting down critical information using pen and paper as a

memory aid) were common. Ironically, despite cutting edge technology (e.g., display walls), the

most common complaint was the need for more screen real estate.


Other researchers have observed similar problems across a variety of work settings and

information technologies. Examples include studies of cell phones (Ziefle and Bay, 2006),

database queries (Vicente, Hayes, and Williges, 1987), medical systems (Cook and Woods, 1996;

Obradovich and Woods, 1996), process control (Easter, 1991; Elm and Woods, 1985), aviation

(e.g., Aretz, 1991), spread sheets (Watts-Perotti and Woods, 1999) and hypertext documents

(McDonald and Stevenson, 1996).

Over a quarter of a century ago Woods (1984) explicitly defined the fundamental problems

and issues (i.e., complex worlds, limited display real estate, the keyhole effect and the getting

lost phenomenon) and the associated challenges for system design. Despite significant advances

in interface technologies (or perhaps due to it!), problems like the keyhole effect and the getting

lost phenomenon have become even more commonplace in the interim.

A primary factor that appears to have contributed to the lack of progress is what we will

refer to as the “myth of information technology.” Briefly, it is a belief that the application of

computational technologies alone will be sufficient to address these challenges. For example, a

belief that simply increasing the size of the window on the world (e.g., data walls) will alleviate

problems like the keyhole effect. Or, a belief that artificial intelligence can be applied to

automate data management and presentation (e.g., intelligent interfaces) or even task

performance (e.g., expert systems) to alleviate problems like the getting lost phenomenon.

1.1. Visual Momentum (VM)

There is an alternative to devising and throwing new forms of technology at these problems.

Woods (1984, p. 231) proposed the concept of “visual momentum,” defining it as “... a measure

of the user’s ability to extract and integrate information across displays ...” In this approach


existing interface technologies, both powerful and mature, are used wisely with the explicit goal

of increasing the degree of “cognitive coupling” between practitioners and the work domain.

Note that the term was originally coined by Hochberg (Hochberg, 1986; Hochberg and Brooks,

1978) who studied perceptual processes in the context of cinematic film editing.

The need to support visual momentum is most obvious at the highest levels of structure in

the interface (i.e., when an interface has multiple screens). Interaction at this level, the

“workspace” level, involves transitions between successive computer screens or pages: one

collection of displays on a screen is replaced with another. The degree of visual momentum in an

interface is determined by the “smoothness” of these segues. Does the practitioner know where

they are, where they have been, and where they might want to go? Does the interface prepare the

practitioner (i.e., set up an appropriate set of expectations) for this transition by perceptually and/

or cognitively orienting them to the particulars of their new destination?

There is also a need to support visual momentum at lower levels of structure in the

interface, including transitions within a screen or page (i.e., between collections of displays or

windows-- the “view” level) or within a display (i.e., between the visual elements of a display

itself -- the “form” level). This need is perhaps less obvious due to our phenomenological

experience of a stable and unbounded visual world (e.g., Gibson, 1966), which here refers to the

subjective impression of parallel data presentation that one experiences when looking at a

computer screen or display. However, this impression does not match the anatomical and

physiological facts (e.g., blind spot, eye movements, field of view). In the end it is clear that the

visual system can only obtain information about the world by scanning it in a serial fashion (and

therefore, that visual momentum techniques are needed to support transitions at this level).


1.1.1. Techniques To Increase Visual Momentum

Woods (1984) originally proposed and then refined (Watts-Perotti and Woods, 1999; Woods

and Watts, 1997) a set of design strategies or techniques to increase visual momentum. Similar

techniques have been proposed and evaluated by researchers in the human computer interaction

(HCI) and computer science communities, although these researchers have chosen a different set

of descriptive labels. Figure 1 (adapted from Woods, 1984) organizes these techniques in terms

of their potential to increase visual momentum and integrates interdisciplinary terminology.

Figure 1. Design techniques for increasing visual momentum. (Adapted with permission

from Woods, D. D., “Visual momentum: A concept to improve the cognitive coupling of

person and computer,” International Journal of Man-Machine Studies, 21, 229-244.

Copyright 1984. All rights reserved)

The purpose of this article is to reexamine the concept of visual momentum and the

progress that has been made in its application. One goal is pedagogical in nature: to provide

additional descriptions and concrete examples of these techniques and how they have been

applied in a variety of different work domains. A second goal is practical in nature: to review


instances where these techniques have been implemented successfully and to document the

benefits to overall human-machine performance that have been obtained. The article will be

structured around the techniques in Figure 1, starting at the left and working towards the right.

2. Fixed Format Data Replacement

A simple (yet powerful) example of the “fixed format, data replacement” technique is

spatial dedication. A natural consequence of interfaces designed with older-generation, hardwired

technology was that controls and displays were spatially dedicated. In contrast, with

computerized interfaces this becomes a design goal: care must be taken to place interface objects

(e.g., menus, commands, categories of information, display frames, labels, etc.) in specific and

consistent spatial locations (both within and across views). Spatial dedication increases visual

momentum via a form of cognitive orientation. Practitioners will learn where specific objects and

information are located as they gain experience with an interface; required transitions are

facilitated because accessing information and executing control input are more efficient.

Spatial dedication provides a form of “consistent mapping” across views, which has been

shown to contribute to the development of automatic processing that can improve the efficiency

of human performance (e.g., Schneider and Shiffrin, 1977; Shiffrin and Schneider, 1977).

Specific results supporting spatial dedication include Watts (Watts, 1994; Watts-Perotti and

Woods, 1999) who compared performance between versions of a spreadsheet with and without

spatial dedication. They found that participants “... were able to accurately predict where they

needed to travel to accomplish their tasks...” (p. 312) with spatial dedication. Spatial dedication

appears to be a compelling and general requirement for effective interface design. For example,

Woods and Watts (1997, p. 642) have observed practitioners “... throwing away flexibility ... to


create their own spatially dedicated set of views of the underlying process or device.”

2.1. Example: RAPTOR Interface

The “signature” version of the fixed format data replacement technique allows practitioners

to selectively change the informational content of a display without changing the viewing

context that it resides within (e.g., surrounding information; general format of the display

including axes, units, labels). An example will be drawn from our work (e.g., Bennett, Posey, and

Shattuck, 2008; Hall, Shattuck, and Bennett, 2011; Talcott et al., 2007) on an ecological interface

(RAPTOR, Representation Aiding Portrayal of Tactical Operations Resources) for military

command and control. Figure 2 illustrates a coordinated set of displays used represent the

combat resources (e.g., ammunition, fuel, personnel, tanks, and armored personnel carriers) of an

Army battalion. Note the hierarchically-nested organizational structure (i.e., one battalion, four

companies, 12 platoons, and 48 vehicles) of the battalion that is represented in the tree structure

of the aggregation control / display at the top of Figure 2a.

The fixed format data replacement technique is used in the primary and secondary display

slots (see the labels in Figure 2a). The number and content of the displays that appear in these

two slots are variable. The primary display slot represents the combat resources of a

superordinate unit (e.g., the battalion in Figure 2a); the secondary display slot represents the

combat resources of a subordinate unit (e.g., the companies in Figure 2a). Rolling over (or

pointing at and clicking on) a node in the tree structure replaces current displays in both slots

with those associated with that particular node. Figures 2b (company level) and 2c (platoon

level) illustrate changes in the content of the slots for views at lower levels in the organizational

structure (note that one more level is possible, but not illustrated -- vehicles and the soldiers they


contain).

Figure 2. The fixed format data replacement and the long shot design techniques as

implemented in the RAPTOR interface. (Adapted with permission from Bennett, K. B.,


Posey, S. M., and Shattuck, L. G., “Ecological interface design for military command and

control,” Journal of Cognitive Engineering and Decision Making, 2(4), pp. 349-385.

Copyright 2008 by the Human Factors and Ergonomics Society. All rights reserved.)

This is a prototypical example of the fixed format data replacement technique: the entire

database is too complex to be viewed in parallel; a limited amount of display real estate is used

to provide detailed, selective, and serial glances into the database. The primary and secondary

slots are located in fixed areas of the interface (i.e., spatial dedication). All representations

appearing in these slots utilize variations of the same general display format (e.g., horizontal bar

graphs, color-coding conventions, boundaries), modified as needed to meet data requirements at

the different levels (fixed format data replacement). Thus, with a single control input the

practitioner can choose from 65 different perspectives on the database, allowing access to

information that ranges from summarized combat resources for the battalion all the way down to

the status of hundreds of individual soldiers.

2.2. Summary

In general, the fixed format data replacement technique increases visual momentum by

supporting alternative, manageable glances into a larger workspace. This allows the practitioner

to selectively focus their attention on relevant subsets of information (e.g., combat resources at

various echelon levels). Visual momentum is increased in several ways. First, the fixed formats

(e.g., primary and secondary display slots) provide spatially-dedicated locations that frame the

presentation of information. This limits the cognitive effort required for search (i.e., consistent

mapping; automatic processing). Second, the new informational content is made available in

parallel within the context of the original view. The practitioner does not need to navigate to an


alternative view, remember the new data, and then return to consider what those data mean

within the broader context (e.g., combat resources in light of extensive tactical information

presented in other areas of the RAPTOR interface). Third, the transition between old and new

information is facilitated by a common viewing context between views (e.g., the same general

display format that is customized for different organizational levels in RAPTOR). This reduces

the amount of cognitive effort that is required for reorientation; the focus can be on changes in

the data instead.

We are not aware of any empirical research that has addressed the fixed format data

replacement technique directly. The technique is likely to have contributed heavily to significant

performance advantages found for the RAPTOR interface (Hall, Shattuck, and Bennett, 2011),

but it was not an isolated manipulation. There is some empirical evidence supporting a very

closely-related presentation technique. Tufte’s (1990) “small multiples” information visualization

technique embodies exactly the same principle (i.e., changes in information content displayed in

the same viewing context). Several empirical evaluations (Archambault, Purchase, and Pinaud,

2011; Javed, McDonnel, and Elmqvist, 2010; Robertson et al., 2008) have found significant

performance advantages for this technique. The only difference between small multiples and data

replacements is that the former are presented in parallel, while the latter are presented serially.

3. Long Shot (Overview + Detail); Zoom + Pan

In this section we describe two complementary techniques that increase visual momentum

by changing the resolution or focus of the information that is being considered. The long shot

design technique (Watts-Perotti and Woods, 1999; Woods, 1984; Woods and Watts, 1997) is

comprised of two distinct components. One component provides an abstract overview of the


global “space of possibilities” of a workspace. The second component provides, in parallel, a

more detailed view of a specific portion of the overall workspace. The HCI / computer science

communities have used the term “overview + detail” (Beard and Walker, 1990; Cockburn,

Karlson, and Bederson, 2008; Plaisant, Carr, and Shneiderman, 1995) to describe the technique,

explicitly referring to both essential components.

The zoom + pan technique also has two distinct components. One component allows the

practitioner to change the resolution of the current glance into a workspace by zooming in (e.g.,

magnification) or zooming out (e.g., demagnification). The second component allows the

practitioner to shift the focus of the current glance into a workspace (i.e., pan). This is a powerful

technique that is a routine feature of many applications.

3.1. Examples: RAPTOR And Adobe® Photoshop®

Figure 2a illustrates the long shot technique as implemented in RAPTOR. The overview

component is embodied in the aggregation control/display. This overview captures the

hierarchically-nested, whole-parts organization of the battalion, as described previously. Each

node is color-coded (using the Army convention of green, amber, red, and black to represent

successively fewer resources) to provide a categorical summary of the combat-readiness of each

unit or vehicle. The node of the currently-selected view (e.g., the battalion level in Figure 2a) is

differentiated from all other nodes by distinct perceptual encodings (a brighter color and a bolder

outline). The detail component (i.e., graphical displays in the primary and secondary slots) and

control mechanisms for changing focus were described in the previous section.

Figure 3a illustrates the long shot technique as implemented in Adobe Photoshop. The

overview component is located in the upper right portion of the screen. This overview (shown in


enlarged proportions on the right side of the figure) presents a low-resolution view of the entire

graphics file that is being edited. The detail component (the majority of the application screen)

contains the portion of the overall graphics file that is currently being worked on. The red

rectangle in the long shot display specifies the current boundaries of this detailed view.


Figure 3. The long shot and zoom + pan techniques as implemented in Adobe Photoshop.

A. Overview and detail components. B. Zoom in. C. Pan right. Adobe product screenshot(s)

reprinted with permission from Adobe Systems Incorporated.

Figures 3b and 3c illustrate how the zoom + pan technique has been integrated with the

long shot technique. In Figure 3b the practitioner has changed the resolution of information in

the detail view by zooming in. Note that the size and location of the red rectangle in the long shot

display has changed to define the currently viewed portion of the graphics file. In Figure 3c the

practitioner changed the focus in the detail view by panning to the right in the long shot. In this

case the pan has been accomplished via direct manipulation (i.e., the practitioner points, clicks,

and drags the rectangle to another physical location within the long shot display).

Woods and Watts (1997) describe three primary functions that a well-designed long shot (or

overview + detail) will provide: orienting, movement, and status summary. The orienting

function refers to the provision of interface resources that illustrate the space of possibilities and

a practitioner’s current location within that space (e.g., the long shot displays in Figures 2 and 3).

The movement function refers to the capability to switch the focus and/or the resolution of a

particular glance into the overall workspace (e.g., the manipulations of the long shot displays).

The status summary function refers to the provision of information that “... allows users, in

a mentally economical way, to step back and assess their overall situation with respect to the

underlying process, device, or activity they are engaged in.” (Woods and Watts, 1997, p. 637). To

be successful, status summary information must meet several criteria: it must be distilled,

abstracted, dynamic (with regard to change), relevant, and economical. This function is fulfilled

in RAPTOR through the visual appearance of the tree nodes in the long shot display. These


nodes specify the organizational structure of the battalion (i.e., relevant). The color-coding of

each node provides a categorical summary (i.e., distilled, abstracted) of associated combat

resources at each level. The color-coding is dynamically updated (i.e., a salient cue illustrating

change in status). Thus, at a glance, the practitioner can see interrelated summaries of status

throughout the entire organizational structure (i.e., economical).

3.2. Summary

The long shot display technique “... provides an overview of the display structure as well as

summary status data. It is a map of the relationships among data that can be seen in more

detailed displays and acts to funnel the viewer’s attention to the ‘important’ details.” (Woods,

1984, p. 236). The long shot technique increases visual momentum by supporting transitions

between alternative glances into the overall workspace. The “getting lost” phenomenon is

avoided by providing an overview of the overall layout of the workspace and by highlighting the

particular “glance” that is currently being displayed in detail (e.g., “you are here”). The

“keyhole” effect is avoided by providing integrated sets of selective and detailed glances that

correspond to the various perspectives that might need to be considered (thereby meeting the

dual challenge of complexity and limited display real estate). Visual momentum is increased

because practitioners know where they are currently located, where they might go to in the

future, and how intuitive control mechanisms can be used to get them there easily.

The long shot (overview + detail) technique has been analyzed, implemented, and evaluated

extensively. Hornbaek, Bederson, and Plaisant (2002) and Cockburn, Karlson, and Bederson

(2008) provide summaries of the empirical evaluations in the HCI literature. In the human

factors literature, the long shot technique is often embedded within large-scale interface design


and evaluation efforts (e.g., the RAPTOR example and its evaluation, Hall, Shattuck, and

Bennett, 2011). However, several empirical studies have evaluated the long shot technique in a

reasonably direct fashion (Billingsley, 1982; Errington et al., 2005; Roth et al., 1998;

Tharanathan et al., 2010). In general, well-designed long shot displays have been found to

enhance performance in a variety of settings.

Empirical evaluations of the zoom + pan technique and its variations (zoom + tilt +pan;

integrated zoom + pan) have produced results that are somewhat mixed (e.g., Hornbaek,

Bederson, and Plaisant, 2002). Literature reviews (e.g., Bederson, 2011; Cockburn, Karlson, and

Bederson, 2008) have found practitioner acceptance (practitioners generally like the technique),

some performance benefits (particularly when well- implemented) and some performance trade-

offs (e.g., mis-orientation after zooming). Recent work (Morison et al., 2009) has applied VM to

go beyond single video camera zoom + tilt + pan techniques to aid in coordinating multiple

perspectives, particularly of feeds from multiple cameras (surveillance or robots) when the

practitioner is removed from the scene of interest.

4. Perceptual Landmarks

A third design strategy for increasing visual momentum is to provide perceptual landmarks

in the interface. Siegel and White (1975) describe “landmark” knowledge in the following

manner (p. 23): “Landmarks are unique configurations of perceptual events (patterns). They

identify a specific geographical location. The intersection of Broadway and 42nd Street is as

much a landmark as the Prudential Center in Boston ... These landmarks are the strategic foci to

and from which the person moves or travels ...We are going to the park. We are coming from

home.” This concept is directly applicable to display and interface design. Perceptual landmarks


in the interface are representations that 1) occupy dedicated spatial locations and 2) provide cues

with regard to meaningful aspects of the underlying work domain that can be navigated to.

4.1. Example: The BookHouse Interface

The Bookhouse system (1980, 1988, 1992) equates the process of finding a book of fiction

in a public library to the process of navigating through a virtual library. Figure 4a provides a

representative navigational sequence. The practitioner initiates a search by entering the virtual

library (pointing at and clicking on the entrance). The entry hallway has three adjoining rooms

(Figure 4b) representing alternative sections of library holdings (e.g., children’s section) which

can be searched. The practitioner narrows the search by navigating to one of these rooms (Figure

4c). This room contains patrons, involved in activities, which represent alternative search

strategies (e.g., search by analogy) that can be executed. The practitioner chooses a search

strategy (clicks on a patron) and enters another room (Figure 4d) containing objects (e.g., the

globe, the clock, the eyeglass icons) that represent the various dimensions of search (e.g.,

geographical setting, time period, font size) that can be specified.

Thus, many of the spatial metaphors in the BookHouse interface (e.g., virtual library,

rooms, patrons, objects) are landmarks (i.e., “configurations of perceptual events”) that point to

locations in a virtual workspace and associated functionality. The landmark technique is perhaps

most clearly evident in the “navigation bar” located at the top portion of the screen. A metaphor

is placed in the navigation bar each time the practitioner leaves an area of the virtual library

(compare the contents of the navigational bar in Figures 4a-4d). The physical appearance of these

metaphors mimics the physical appearance of the area being left (i.e., it provides a small-scale

replica of the corresponding room, hallway, or entrance). These “replica” metaphors also serve as


controls that can be pointed at and clicked on to produce navigation back to that location. Thus,

the replica metaphors in the navigation bar provide perceptual landmarks that specify locations

in the virtual workspace that can be navigated to so that an ongoing search can be modified.

Figure 4. The landmark technique as implemented in the BookHouse fiction retrieval

system. (Screenshots from the Bookhouse system. Used with permission from system

designer, Pejtersen, A. M. Copyright 1992. All rights reserved)

The BookHouse interface uses a global metaphor that is explicitly tied to a real world

setting. It should be emphasized that this is not necessary for implementation of the landmark


technique. Apple’s iPhone® and iPad® devices employ a general interface design strategy that

provides an excellent example. These interfaces are organized around multiple “home pages”

that can be configured (and navigated to via a “swipe” gesture). Each home page organizes

multiple application metaphors in a matrix. Tapping a metaphor produces navigation to a view in

the application. Each view, in turn, has metaphors for objects (and sometimes modes) that can be

tapped. Thus, each home page provides a “navigational matrix” that is functionally similar in

purpose to the BookHouse’s navigational bar; each manipulable metaphor in the interface

(application, mode, or object) is a landmark to associated functionality in the workspace.

4.2.Summary

Perceptual landmarks at the workspace level will increase visual momentum by facilitating

navigation between views. At a general level they are spatially-dedicated (promoting consistency

and automatic processing) and therefore facilitate the location of navigation-related information.

They are powerful mnemonic devices, providing external landmarks that remind practitioners of

destinations that can be navigated to and the associated functionality that can be accessed. They

provide a visual “preview” of potential destinations (either new destinations or destinations

previously visited) in the workspace. This preview orients the practitioner, both visually and

cognitively, thereby smoothing the ensuing transitions between views. Of course, perceptual

landmarks in the interface must be both perceptually salient and distinct (relative to other

landmarks) if they are to be effective.

The role of landmark knowledge (Lynch, 1960; Siegel and White, 1975) in navigation has

been studied extensively. For example, May and Ross (2006) investigated landmark-based

navigation aids in a driving field study; they found significantly better performance when high-


quality landmark information was used. Empirical studies have indicated that the extensive

research on navigation in real worlds generalizes to computer settings (Darken and Sibert, 1996;

Ruddle, Payne, and Jones, 1997). Vinson (1999) reviews the literature on the use of landmarks in

these settings and provides guidelines for their design. Aretz (1991) found that a navigational

display based on visual momentum principles provided a global set of landmarks that facilitated

pilots’ abilities to orient to and locate terrain-based targets. The utility of landmarks has also been

demonstrated in more traditional HCI applications. For example, Watts-Perotti and Woods

(Watts, 1994; Watts-Perotti and Woods, 1999) found landmarks (e.g., table headings and other

forms of formatting) to be an essential design feature in supporting effective spreadsheet

navigation (in fact, they observed that practitioners manually added landmarks when they were

missing).

5. Overlap: Rooms, Center-Surround, Side Effects, Integrated Representations

Woods (1984) originally described the overlap technique as a very general strategy to

increase visual momentum by providing supplemental viewing contexts to frame the presentation

of information. These contexts are presented in parallel and are used to testify with regard to

physical contiguity, functional similarity, or other meaningful relations. For example, in

cartography an area of interest (e.g., the state of Ohio) is presented in a detailed map situated

within a larger, abbreviated context (i.e., partial maps of the surrounding states). Subsequent

publications (Watts-Perotti and Woods, 1999; Woods and Watts, 1997) describe variations of

overlap referred to as “rooms,” “center surround,” “side effects,” and “integrated

representations.” Brief examples of each technique will be provided, followed by a summary.


5.1. Example: Rooms

The rooms technique is designed to increase visual momentum by addressing a fundamental

challenge often facing practitioners: configuring and accessing sets of functionally-related

information that are required for the completion of multiple, asynchronous activities. Clumsy

information systems, in combination with limited display real estate, often make this a daunting

task (e.g., the command center described earlier). The everyday challenge of windows

management (e.g., organizing, layering, resizing, opening, closing, moving) that often results in

“window thrashing” (Henderson and Card, 1986) is a common example. Practitioners often adapt

existing interface resources in an attempt to meet these needs (e.g., Cook and Woods, 1996;

Watts-Perotti and Woods, 1999). Visual momentum can be increased by providing interface

resources that allow the configuration of alternative visualization contexts or environments and

facilitate effective navigation between them.

Some of the key features of the rooms technique will be illustrated using Apple’s Spaces®

operating system feature. Figure 5a represents a current view (i.e., a desktop containing sets of

windows, applications, files, etc.); assume that the practitioner has previously defined two

additional views. When activated, Spaces provides a parallel overview of the workspace via a

matrix (in this instance a 2 by 2 matrix, see Figure 5b) that organizes all existing views. One way

the practitioner can navigate between views (i.e., re-establish a previously-defined computing

context) is by pointing at and clicking on a cell in the matrix. For example, a click on the upper

right cell in Figure 5b causes View 2 to become the current view (Figure 5c).


Figure 5. The rooms technique as implemented in Apple’s Spaces® operating system

feature.

Alternatively, practitioners can also navigate by activating a long shot display. This display

provides an abstract, simplified version of the matrix with highlighting of the current view (see


Figure 5c). Navigation between alternative views is accomplished via arrow keys. For example,

navigation from View 2 (Figure 5c) to View 4 (Figure 5d) to View 3 (Figure 5e) is accomplished

by pressing the down and left arrow key in succession. In addition, all overlapping windows

within a view can be spatially separated instantaneously (via Apple Exposé®), making them

available for individual inspection (see Figure 5f). This feature also works for the matrix (e.g., all

three of the views illustrated in Figure 5b would have spatially separated windows).

Woods and Watts (1997) provide a summary of design goals for the rooms technique. The

first goal is to provide alternative views (or contexts) with resources that can be tailored to meet

task-specific needs (e.g., up to 16 desktops in Spaces). A second goal is to provide a coordinated

workspace (i.e., explicit representation of the relationships between views and navigational

resources to change views, as exemplified by the Spaces matrix). A third goal is to allow views

to be seen in parallel (e.g., the matrix and the long shot displays of Spaces). A fourth goal is to

allow practitioners to “... compose, save, and manipulate sets of views as a coherent

unit ...” (Woods and Watts, 1997, p. 644). Spaces supports this goal in a variety of ways.

Composing a view can be achieved via simple (i.e., starting an application in one of the

desktops) or more innovative means (i.e., dragging an application or window from one view in

the matrix to another). Entire views can be manipulated (i.e., point, click, and drag an entire view

to a different cell in the matrix) as a way to organize the matrix along functional lines (e.g.,

views with similar purposes can be placed close in space).

Early research projects addressing these needs include Smalltalk Projects (Goldberg, 1984),

Rooms (Henderson and Card, 1986, an early and comprehensive effort after which the technique

was named), 3D Rooms (Robertson, Card, and Mackinlay, 1993), and Elastic Windows


(Kandogan and Shneiderman, 1998). More recent examples include the Task Gallery (Robertson

et al., 2000), the TimeSpace system (Krishnan and Jones, 2005), and Piles Across Space (Wang,

Hsieh, and Paepckea, 2009, for hand-held devices). Empirical evidence has been obtained in

support of overlap techniques. Information visualization and windows management systems

(e.g., Rooms) have been evaluated both informally (e.g., Krishnan and Jones, 2005; Robertson et

al., 2000) and formally (e.g., Kandogan and Shneiderman, 1998; Wang, Hsieh, and Paepckea,

2009) with generally positive results.

5.2. Example: Center-Surround (Focus + Context)

The center-surround design technique is used to increase visual momentum through parallel

presentation: information in the practitioner’s current focus of attention is presented in high

resolution (i.e., “center”); related information is presented in lower resolution (i.e., “surround”).

The center surround technique is conceptually similar to the long shot technique; one often-cited

difference is that the center and surround are spatially integrated (i.e., they form a single

integrated display) while the overview and detail displays are spatially separated. The HCI /

Computer Science community has used the label “focus + context” to describe this technique.

The integration of center and surround is often achieved through a variety of systematic

visual transformation and magnification functions (e.g., Leung and Apperley, 1994). Several

representative examples will be illustrated via comparisons to a uniformly-spaced rectangular

grid (Figure 6a).


Figure 6. Abstract representations of various forms of information compression used in the

implementation of the center surround technique. See text for details.

The “bifocal display” (Spence and Apperley, 1982) provides a center region where

information is magnified and surround regions with uniform compression (Figure 6b). The “table


lens system” (Rao and Card, 1994) provides uniform compression in both the horizontal and

vertical dimensions (and support for multiple centers and surrounds, see Figure 6c). The

“perspective wall” (Mackinlay, Robertson, and Card, 1991) applies continuous, rather than

discrete, compression (based on perspective geometry) for two surround regions (Figure 6d). The

“fisheye view” (e.g., Furnas, 1986) applies circular compression with continuous distortion

towards the periphery. Figure 6e illustrates a generic fisheye transformation using polar

coordinates (Sarkar and Brown, 1994). The “document lens” (Robertson and Mackinlay, 1993)

applies a modified fisheye transformation using cartesian coordinates (Figure 6f). Lamping and

Rao (1996) also describe a modified fisheye transformation using compression based on

hyperbolic geometry (not illustrated).

5.2.1. Side Effect Views

Note that while the information presented in a center surround display can be spatial in

nature (and often is), other types of relationships can also be portrayed. Woods (Watts-Perotti and

Woods, 1999; Woods and Hollnagel, 1987; Woods and Watts, 1997) propose a variation of the

center surround technique that portrays functional relationships. Because of the many-to-many

functional mappings that exist in certain work domains, the actions of practitioners will

propagate through the system and produce side effects. These side effects can be either intended

or unintended (and either for good or bad). The “side effect views” technique increases visual

momentum by ensuring that practitioners are aware of these consequences (Woods and Watts,

1997, p. 646): “If an operator is interested in one function [i.e., the center] ... the system also

provides a summary of the status of other functions which could be influenced by changes in the

primary area of interest in a side effects window [i.e., the surround] ...”


The need for side effect views is readily-apparent for work domains which involve

semantics that are driven by the laws of nature and therefore tightly-coupled (e.g., process

control). Woods and Hollnagel (1987) provide an extensive discussion of side effects in this

category of domains and describe a control room display system containing a side effects view.

This need is not restricted to process control; it is relevant for any work domain in which actions

have propagating effects. For example, Watts (1994) proposed the use of side effect views to

illustrate computational interdependencies within spreadsheets and between sets of spreadsheets.

An extensive review of the center surround (focus + context) literature is provided by

Cockburn et al. (2008). They conclude “It therefore seems clear that supporting both focused and

contextual views can improve interaction over constrained single-view software” (p. 2:26).

5.3. Example: Integrated Representations

Woods (1984) originally described functional overlap as a way to increase visual

momentum “By identifying [functional] relationships between data and user tasks and by

paralleling those relationships in the structure of the display system ...” (Woods, 1984, p. 237).

Functional overlap is implied, if not explicit, in all of the overlap techniques described thus far.

Subsequently, Woods and Watts (1997) identified “integrated representations” as a research trend

related to visual momentum, but outside the scope of their review. Based on our experience in

developing these representations (e.g., Bennett and Flach, 1992, 2011; Bennett, Nagy, and Flach,

1997), we believe that they provide a particularly useful form of functional overlap and that they

qualify as a technique to increase visual momentum.

Cognitive systems engineering has developed analytical tools that have provided a more

precise definition of functional overlap. The abstraction hierarchy (Rasmussen, 1983, 1986)


provides five levels (categories or perspectives) that can be used to describe a work domain.

Each level provides an independent category of information, yet there are inherent and

systematic relations between these categories. Using this tool to model a work domain (e.g.,

Vicente, 1999) produces a description of the informational content (and inherent relations) that

need to be made apparent for a practitioner to think productively about complex problems.

Figure 7 illustrates a representation that integrates information across levels of the

abstraction hierarchy. The process represented by the display involves a liquid coolant entering

(mass in), being stored in (reservoir level), and leaving (mass out) the system. The level of

“functional purpose” contains descriptions of the design goals for the system. These are to

maintain both a level of mass in the reservoir (G1) and a flow rate for mass out (G2). The level of

“abstract functions” contain descriptions of the physical laws that govern system behavior. For

example, the change in reservoir level (ΔR) should be determined by the difference between the

rate of mass flowing into (I1 + I2) and out of (O) the reservoir. The “generalized functions” level

refers to a source, a store, and a sink for fluid in the system. The level of “physical functions”

contains descriptions of physical limits (e.g., the moment-to-moment values of each variable)

and control (e.g., changing a valve setting). Finally, the level of “physical form” contains

descriptions of the physical configuration of the system. All of the levels of abstraction (except

for physical configuration , not critical for control) are represented in this display (see labels in

Figure 8 and Bennett, Nagy, and Flach, 1997, for a more complete description of the design

rationale and benefits).


Figure 7. Example of integrated display illustrating functional overlap across levels of the

abstraction hierarchy. (Adapted with permission from Bennett, K. B., A. L. Nagy, and J. M.

Flach, “Visual displays,” in Handbook of Human Factors and Ergonomics, G. Salvendy, ed.

Copyright 1997 New York, NY: Wiley. All rights reserved.)

The implications for increasing visual momentum are clear: if these relationships can be

integrated into a single display, then the requirement to navigate between windows and


remember information across these contexts is eliminated. Thus, the visual integration of

information across the levels of abstraction increases visual momentum because it provides a

“continuous graphical explanation” of events in the domain that is available in parallel (i.e.,

powerful perceptual skills are leveraged).

These integrated representations have been referred to as both “configural” (e.g., Bennett

and Flach, 1992, 2011; Bennett, Nagy, and Flach, 1997) and “ecological” (Rasmussen and

Vicente, 1990; Vicente and Rasmussen, 1992) displays in the literature and they have been

evaluated extensively. Vicente’s DURESS system (Vicente, 1991) in particular has been

systematically evaluated (e.g., Pawlak and Vicente, 1996). In addition, several studies have

empirically evaluated the impact of integrating information from various levels of the abstraction

hierarchy into representations (Burns, 2000; Ham and Yoon, 2001a, 2001b). In general,

representations that integrated information across more levels (as opposed to missing levels or

levels that were displayed separately) were found to be more effective.

In the introduction we outlined the need to support visual momentum at lower levels of

structure in the interface (i.e., the view or form level). To reiterate, obtaining information from a

display (or collections of displays on a single screen) involves visual transitions that take place

over time (serially). Visual momentum can be increased at levels lower than the workspace by

providing effective visual structures that guide successive fixations, thereby increasing the

“smoothness” of these segues. For example, does the screen provide optical structure that links

information that is meaningfully-related, but spatially separated? Does the display provide levels

of optical structure that highlight the inherent relationships between the different types of

information that it portrays? Tufte (1990) discusses the general principles of “layering and


separation” that can increase visual momentum at the view and form level. We (e.g., Bennett and

Flach, 1992, 2011; Bennett, Nagy, and Flach, 1997) discuss more specific principles to integrate

and balance the hierarchically-nested visual information (i.e., global / local “emergent features”

and low-level “graphical elements”) in analog geometrical displays.

5.4. Summary

In general, overlap techniques increase visual momentum by explicitly illustrating

meaningful relations (e.g., tasks, contexts, physical and functional perspectives). These relations

are presented in an overall context that organizes alternative views (e.g., the matrix provided in

Spaces). The currently active view is is situated, in parallel, within this viewing context.

Together, this provides a form of external survey knowledge (“you are here” and potential

destinations) that serves to increase visual momentum by eliminating the getting lost

phenomenon. Additional detail on related information can be obtained by changing to an

alternative view. These transitions are facilitated by the overall context (i.e., it provides

perceptual and cognitive orientation that allows the practitioner to anticipate changes in view), as

well as natural and effective control mechanisms. The end result is a “widening of the key hole”

through which the practitioner can view the work domain.

6. Spatial Representations: Spatial Structure

The final design technique (spatial representation, see Figure 1) has the greatest potential to

increase visual momentum. It is a general technique used to incorporate (or impose, in the case

of non-spatial data) a degree of spatial structure and organization into the controls and displays

of an interface (e.g., Bennett and Flach, 2011). This technique is described by Woods (1984, pp.

238-240) in the following passage:


Spatial organization translates the normative user internal model into a

perceptual map. The user sees, rather than remembers, the organization of data in

the system and can move within the system just as he moves in an actual spatial

layout ... One spatial access technique is to organize the data base as a topology

and then provide the viewer with a mechanism to move through the space ...inter-

display movements can be conceptualized as itineraries or paths through the

space ... the user’s perceptual and attentional skills ... can be supported ... by

constructing a conceptual or virtual space ...

This technique increases visual momentum by leveraging powerful natural skills, using

them to facilitate interactions with information technologies. Specifically, virtual spaces allow

practitioners to navigate between the views of a workspace much like they navigate through

man-made (e.g., a familiar building) or natural ecologies (e.g., from home to work). The

associated human capabilities are both general and powerful and they have been studied

extensively in a variety of ways: cognitive maps (e.g., Kitchin, 1994), spatial knowledge (e.g.,

Siegel and White, 1975), spatial thinking (e.g., Gauvain, 1993), environmental cognition (e.g.,

Evans, 1980), and way-finding (e.g., Hutchins, 1995). The spatial representation technique has,

to a large degree, been implicitly recognized as an integral part of HCI (e.g., witness the

ubiquitous use of the term “navigation”). The technique is well-represented in several of the

interfaces discussed throughout the paper (e.g. BookHouse, RAPTOR, iPhone, iPad) and will

therefore not be reviewed in any greater detail.

7. Summary

The topic of visual momentum covers an extremely broad spectrum of issues in design, as


the variety of techniques and examples discussed in this review indicate. Even so, our coverage

of visual momentum was necessarily selective in nature. We consciously chose the pedagogical

approach of covering a smaller number of particularly representative examples in detail, rather

than providing a comprehensive treatment. We did not cover some techniques that could very

well be considered as examples of visual momentum (e.g., animated transitions between views)

and did not discuss some excellent examples (Guerlain, 2007; McKenna et al., 2008; Patterson,

Watts-Perotti, and Woods, 1999).

As the review indicates clearly, the visual momentum techniques are often complementary

in nature and can be applied in combination (e.g., long shot and fixed format data replacement in

Figure 1; long shot, and zoom + pan in Figure 2). This occurs to an extent that sometimes made

it difficult to assign an example or an empirical finding to a particular category.

On the other hand, these techniques can also be competing alternatives for increasing visual

momentum. For example, the long shot and the center surround techniques are reasonably

similar in purpose and the choice between them may depend upon some inherent tradeoffs. The

center surround technique involves an integrated representation whereas the two components of

the long shot technique (e.g., overview and detail) are spatially separated (imposing a potential

cost for information access). Conversely, the compression of information in the center surround

technique saves space, but these distortions can also make information difficult to access (unlike

the long shot technique).

We had several goals in writing this review: to once again reiterate the need to consider

visual momentum as an important consideration in the design of information systems, to refine

and expand Woods’ original ideas, and to integrate related concepts and findings from


interdisciplinary literatures. In the process it has become very clear that there is no single,

preferred technique to increase visual momentum; choosing one particular technique over

another will be dependent upon very situationally-specific factors. Cockburn, Karlson and

Bederson (2008, p. 2:26) make essentially the same point:

The current state of research fails to provide clear guidelines, despite a recent

surge in empirical analysis of the techniques’ effectiveness. Results to date have

revealed that the efficiency of different techniques is dependent on many factors,

particularly the nature of the users’ task.

There is an important implication that can be drawn from these observations. Specifically,

the successful application of visual momentum techniques will depend upon a deep

understanding of the design context: “Developers should study/analyze what views need to be

seen in parallel ... Methods for cognitive task analysis that analytically map meaningful domain

contexts identify views that are inter-related and need to be seen in parallel ...” (Woods and

Watts, 1997, p. 643, emphasis original). We and others have made the same points with regard to

interface design in general (e.g., Bennett and Flach, 2011; Rasmussen, Pejtersen, and Goodstein,

1994; Vicente, 1999). Although they are sometimes viewed as unnecessary steps in the process,

these work domain analyses are pre-requisites to effective design.

In closing, experience has shown that the “myth of information technology” referred to in

the introduction has been “busted.” Larger screens and increased automation (e.g., “intelligent”

interfaces, expert systems) often wind up increasing the complexity of modern work

environments, rather than decreasing it. The keyhole effect and the getting lost phenomenon have

not disappeared with the introduction of these and similar technologies; if anything, their rate of


occurrence has increased. The techniques to increase visual momentum that have been described

in this review are successful because they leverage powerful technologies to increase the

cognitive coupling between practitioners and their work domains. They are available now (i.e.,

they do not depend upon future advances in technology to be implemented) and they offer the

very real potential to increase overall system performance.

8. Acknowledgments

We would like to express our sincere thanks to the editor and anonymous reviewers for

valuable suggestions that contributed immensely to the overall quality of the final paper.

9. References

Archambault, D., Purchase, H. C., and Pinaud, B., 2011. Animation, Small Multiples, and the Effect of Mental Map Preservation in Dynamic Graphs. IEEE Transactions on Visualization and Computer Graphics 17 (4), 539-552.

Aretz, A. J., 1991. The Design of Electronic Map Displays. Human Factors 33, 85-101.Beard, D. B., and Walker, J. Q., 1990. Navigational Techniques to Improve the Display of Large

Two-Dimensional Spaces. Behavior & Information Technology 9 (6), 451–466.Bederson, B. B., 2011. The Promise of Zoomable User Interfaces. Behaviour & Information

Technology 30 (6), 853-866.Bennett, K. B., and Flach, J. M., 1992. Graphical Displays: Implications for Divided Attention,

Focused Attention, and Problem Solving. Human Factors 34, 513-533.Bennett, K. B., and Flach, J. M., 2011. Display and Interface Design: Subtle Science, Exact Art.

Boca Raton, FL: CRC Press.Bennett, K. B., Nagy, A. L., and Flach, J. M., 1997. Visual Displays. In: Salvendy, G., (Ed.),

Handbook of Human Factors and Ergonomics. New York, NY: Wiley, 659-696.Bennett, K. B., Posey, S. M., and Shattuck, L. G., 2008. Ecological Interface Design for Military

Command and Control. Journal of Cognitive Engineering and Decision Making 2 (4), 349-385.

Billingsley, P. A., 1982. Navigation through HierarchicalMenu Structures: Does It Help to Have a Map? In: the Human Factors Society 26th Annual

Meeting, Santa Monica, CA, 103-107.Burns, C. M., 2000. Putting It All Together: Improving Display Integration in Ecological

Displays. Human Factors 42, 226–241.Cockburn, A., Karlson, A., and Bederson, B. B., 2008. A Review of Overview+Detail, Zooming,

and Focus+Context Interfaces. ACM Computing Surveys 41 (1), 1-31.


Cook, R. I., and Woods, D. D., 1996. Adapting to New Technology in the Operating Room. Human Factors 38 (4), 593-613.

Darken, R. P., and Sibert, J. L., 1996. Navigating Large Virtual Spaces. International Journal of Human-Computer Interaction 8 (1), 49-71.

Easter, J. R., 1991. The Role of the Operator and Control Room Design. In: White, J. and Lanning, D., (Eds.), European Nuclear Instrumentation and Controls. World Technology Evaluation Center, Loyola college, Baltimore, MD (available from national information technology service, Springfield, VA, Report # PB92100197): National Technical Information Service.

Elm, W. C., and Woods, D. D., 1985. Getting Lost: A Case Study in Interface Design. In: the Human Factors and Ergonomics Society 29th Annual Meeting, 927-931.

Errington, J., Reising, D. V. C., Bullemer, P., DeMaere, T., Coppard, D., Doe, K., and Bloom, C., 2005. Establishing Human Performance Improvements and Economic Benefit for a Human-Centered Operator Interface: An Industrial Evaluation. In: the Human Factors Society 49th Annual Meeting, Santa Monica, CA, 2036-2040.

Evans, G. W., 1980. Environmental Cognition. Psychological Bulletin 88 (2), 259-287.Furnas, G. W., 1986. Generalized Fisheye Views. In: the CHI'86 Human Factors in Computing

Systems Conference, New York, NY, 16-23.Gauvain, M., 1993. The Development of Spatial Thinking in Everyday Activity. Developmental

Review 13, 92-121.Gibson, J. J., 1966. The Senses Considered as Perceptual Systems. Boston, MA: Houghton

Mifflin.Goldberg, A., 1984. Smalltalk-80: The Interactive Programming Environment. Reading, MA:

Addison-Wesley.Guerlain, S., 2007. Software Navigation Design. In: Allen, G., (Ed.), Applied Spatial Cognition.

From Research to Cognitive Technology. Mahwah, NJ: Lawrence Erlbaum Associates, 317-337.

Hall, D. S., Shattuck, L. G., and Bennett, K. B., 2011. Evaluation of an Ecological Interface Design for Military Command and Control. Manuscript submitted for publication.

Ham, D. H., and Yoon, W. C., 2001a. Design of Information Content and Layout for Process Control Based on Goal-Means Analysis. Cognition, Technology & Work 3 (4), 205-223.

Ham, D. H., and Yoon, W. C., 2001b. The Effects of Presenting Functionally Abstract Information in Fault Diagnosis Tasks. Reliability Engineering & System Safety 73 (2), 103-119.

Henderson, D. A., and Card, S. K., 1986. Rooms: The Use of Multiple Virtual Workspaces to Reduce Space Contention in a Window-Based Graphical User Interface. ACM Transactions on Graphics 5 (3), 211-243.

Hochberg, J., 1986. Representation of Motion and Space in Video and Cinematic Displays. In: Boff, K., Kaufman, L. and Thomas, J., (Eds.), Handbook of Perception and Human Performance. New York, NY: John Wiley & Sons, 22-1 to 22-64.

Hochberg, J., and Brooks, V., 1978. Film Cutting and Visual Momentum. In: Senders, J. W., Fisher, D. F. and Monty, R. A., (Eds.), Eye Movements and the Higher Psychological Functions. Hillsdale, NJ: Lawrence Erlbaum Associates, 293-313.


Hornbaek, K., Bederson, B., and Plaisant, C., 2002. Navigation Patterns and Usability of Zoomable User Interfaces with and without an Overview. ACM Transactions on Computer-Human Interaction 9 (4), 362–389.

Hutchins, E. L., 1995. Cognition in the Wild. Cambridge, MA: MIT Press.Javed, W., McDonnel, B., and Elmqvist, N., 2010. Graphical Perception of Multiple Time Series.

IEEE Transactions on Visualization and Computer Graphics 16 (6), 927-934.Kandogan, E., and Shneiderman, B., 1998. Elastic Windows: Design, Implementation, and

Evaluation of Multi-Window Operations. Software - Practice and Experience 28 (3), 225–248.

Kitchin, R. M., 1994. Cognitive Maps: What Are They and Why Study Them? Journal of Environmental Psychology 14, 1-19.

Krishnan, A., and Jones, S., 2005. Timespace: Activity-Based Temporal Visualisation of Personal Information Spaces. Personal and Ubiquitous Computing 9, 46-65.

Lamping, J., and Rao, R., 1996. The Hyperbolic Browser: A Focus+Context Technique for Visualizing Large Hierarchies. Journal of Visual Languages and Computing 7 (1), 33-55.

Leung, Y., and Apperley, M., 1994. A Review and Taxonomy of Distortion-Oriented Presentation Techniques. ACM Transactions on Computer-Human Interaction 1 (2), 126–160.

Lynch, K., 1960. The Image of the City. Cambridge, MA: MIT Press.Mackinlay, J., Robertson, G., and Card, S., 1991. The Perspective Wall: Detail and Context

Smoothly Integrated. In: CHI'91 ACM Conference on Human Factors in Computing Systems, New York, NY, 173-179.

May, A. J., and Ross, T., 2006. Presence and Quality of Navigational Landmarks: Effect on Driver Performance and Implications for Design. Human Factors 48 (2), 346–361.

McDonald, S., and Stevenson, R. J., 1996. Disorientation in Hypertext: The Effects of Three Text Structures on Navigation Performance. Applied Ergonomics 27 (1), 61-68.

McKenna, B., Tittle, J., Gualtieri, J., Elm, W., Grossman, J., Jennnings, D., Voshell, M., and Woods, D. D., 2008. How Do You Know What Is Happening -- Just Beyond Your View? The Next Wave: National Security Agency's Review of Emerging Technologies 17 (2), 20-25.

Morison, A., Voshell, M., Roesler, A., Feil, M., Tittle, J., Tinapple, D., and Woods, D. D., 2009. Integrating Diverse Feeds to Extend Human Perception into Distant Scenes. In: McDermott, P. and Allender, L., (Eds.), Advanced Decision Architectures for the Warfigher: Foundation and Technology. Boulder, CO: Partners of the Army Research Laboratories Advanced Decision Architectures Collaborative Technology Alliance, 177-200.

Obradovich, J., and Woods, D. D., 1996. Users as Designers: How People Cope with Poor Hci Design in Computer-Based Medical Devices. Human Factors 38 (1), 574-592.

Patterson, E. S., Watts-Perotti, J. C., and Woods, D. D., 1999. Voice Loops as Coordination Aids in Space Shuttle Mission Control. Computer Supported Cooperative Work 8, 353-371.

Pawlak, W. S., and Vicente, K. J., 1996. Inducing Effective Operator Control through Ecological Interface Design. International Journal of Human-Computer Studies 44 (5), 653-688.

Pejtersen, A. M., 1980. Design of a Classification Scheme for Fiction Based on an Analysis of Actual User-Librarian Communication, and Use of the Scheme for Control of Librarians'


Search Strategies. In: Harbo, O. and Kajberg, L., (Eds.), Theory and Application of Information Research London, UK: Mansell, 167-183.

Pejtersen, A. M., 1988. Search Strategies and Database Design for Information Retrieval in Libraries. In: Goodstein, L. P., Andersen, H. B. and Olsen, S. E., (Eds.), Tasks, Errors and Mental Models. Bristol, PA: Taylor & Francis, 171-190.

Pejtersen, A. M., 1992. The Bookhouse: An Icon Based Database System for Fiction Retrieval in Public Libraries. In: Cronin, B., (Ed.), The Marketing of Library and Information Services 2. London, UK: ASLIB, 572-591.

Plaisant, C., Carr, D., and Shneiderman, B., 1995. Image-Browser Taxonomy and Guidelines for Designers. Software, IEEE 12 (2), 21-32.

Rao, R., and Card, S. K., 1994. The Table Lens: Merging Graphical and Symbolic Representations in an Interactive Focus + Context Visualization for Tabular Information. In: CHI'94 Human Factors in Computing Systems, New York, NY, 318-322.

Rasmussen, J., 1983. Skills, Rules, and Knowledge; Signals, Signs, and Symbols, and Other Distinctions in Human Performance Models. IEEE Transactions on Systems, Man, and Cybernetics SMC-13, 257-266.

Rasmussen, J., 1986. Information Processing and Human-Machine Interaction: An Approach to Cognitive Engineering. New York, NY: Elsevier.

Rasmussen, J., Pejtersen, A. M., and Goodstein, L. P., 1994. Cognitive Systems Engineering. New York, NY: Wiley.

Rasmussen, J., and Vicente, K. J., 1990. Ecological Interfaces: A Technological Imperative in High Tech Systems? International Journal of Human-Computer Interaction 2 (2), 93-111.

Robertson, G., Card, S. K., and Mackinlay, J. D., 1993. Information Visualization Using 3d Interactive Animation. Communications of the ACM 36 (4), 57-71.

Robertson, G., Fernandez, R., Fisher, D., Lee, B., and Stasko, J., 2008. Effectiveness of Animation in Trend Visualization. IEEE Transactions on Visualization and Computer Graphics 14 (6), 1325-1332.

Robertson, G., van Dantzich, M., Robbins, D., Czerwinski, M., Hinckley, K., Risden, K., Thiel, D., and Gorokhovsky, V., 2000. The Task Gallery: A 3d Window Manager. In: CHI' 2000 Human Factors in Computing Systems, New York, NY, 494-501.

Robertson, G. G., and Mackinlay, J. D., 1993. The Document Lens. In: ACM Symposium on User Interface Software and Technology (UIST), New York, NY, 101–108.

Roth, E. M., Lin, L., Thomas, V. M., Kerch, S., Kenney, S. J., and Sugibayashi, N., 1998. Supporting Situation Awareness of Individuals and Teams Using Group View Displays. In: the Human Factors Society 49th Annual Meeting, Santa Monica, CA, 244-248.

Ruddle, R. A., Payne, S. J., and Jones, D. M., 1997. Navigating Buildings in "Desk-Top" Virtual Environments: Experimental Investigations Using Extended Navigational Experience. Journal of Experimental Psychology: Applied 3 (2), 143-159.

Sarkar, M., and Brown, M. H., 1994. Graphical Fisheye Views. Communications of the ACM 37 (12), 73-84.

Schneider, W., and Shiffrin, R. M., 1977. Controlled and Automatic Human Information Processing: I Detection, Search, and Attention. Psychological Review 84, 1-66.


Shiffrin, R. M., and Schneider, W., 1977. Controlled and Automatic Human Information Processing: Ii Perceptual Learning, Automatic Attending, and a General Theory. Psychological Review 84, 127-190.

Siegel, A. W., and White, S. H., 1975. The Development of Spatial Representations of Large-Scale Environments. In: Reese, H. W., (Ed.), Advances in Child Development and Behavior. New York: Academic Press, 10-55.

Spence, R., and Apperley, M., 1982. Database Navigation: An Office Enironment for the Professional. Behavior & Information Technology 1 (1), 43–54.

Talcott, C. P., Bennett, K. B., Martinez, S. G., Shattuck, L., and Stansifer, C., 2007. Perception-Action Icons: An Interface Design Strategy for Intermediate Domains. Human Factors 49 (1), 120-135.

Tharanathan, A., Bullemer, P., Laberge, J., Reising, D. V. C., and Mclain, R., 2010. Functional Versus Schematic Overview Displays: Impact on Operator Situation Awareness in Process Monitoring. In: the Human Factors Society 49th Annual Meeting, Santa Monica, CA, 319-323.

Tufte, E. R., 1990. Envisioning Information. Chesire, CT: Graphics Press.Vicente, K. J. 1991. Supporting Knowledge-Based Behavior through Ecological Interface

Design. Urbana-Champaign, IL: Engineering Psychology Research Laboratory, Department of Mechanical Engineering, University of Illinois.

Vicente, K. J., 1999. Cognitive Work Analysis: Toward Safe, Productive, and Healthy Computer-Based Work. Mahwah, NJ: Lawrence Erlbaum Associates.

Vicente, K. J., Hayes, B. C., and Williges, R. C., 1987. Assaying and Isolating Individual-Differences in Searching a Hierarchical File System. Human Factors 29 (3), 349-359.

Vicente, K. J., and Rasmussen, J., 1992. Ecological Interface Design - Theoretical Foundations. Ieee Transactions on Systems Man and Cybernetics 22 (4), 589-606.

Vinson, N. G., 1999. Design Guidelines for Landmarks to Support Navigation in Virtual Environments. In: SIGCHI conference on Human factors in computing systems, Pittsburg, PA, 278–285.

Wang, Q. Y., Hsieh, T., and Paepckea, A., 2009. Piles across Space: Breaking the Real-Estate Barrier on Small-Display Devices. International Journal of Human-Computer Studies 67, 349-365.

Watts, J. C., 1994. Navigation in the Computer Medium: A Cognitive Analysis. . In: Human Factors and Ergonomics Society, 38th annual Meeting, Santa Monica, CA, 310-314.

Watts-Perotti, J., and Woods, D. D., 1999. How Experienced Users Avoid Getting Lost in Large Display Networks. International Journal of Human-Computer Interaction 11 (4), 269-299.

Woods, D. D., 1984. Visual Momentum: A Concept to Improve the Cognitive Coupling of Person and Computer. International Journal of Man-Machine Studies 21, 229-244.

Woods, D. D., and Hollnagel, E., 1987. Mapping Cognitive Demands in Complex Problem Solving Worlds. International Journal of Man-Machine Studies 26, 257-275.

Woods, D. D., and Watts, J., 1997. How Not to Have to Navigate through Too Many Displays. In: Helander, M., Landauer, T. and Prabhu, P., (Eds.), Handbook of Human-Computer Interaction. Amsterdam, The Netherlands: Elsevier Science, 617-650.


Ziefle, M., and Bay, S., 2006. How to Overcome Disorientation in Mobile Phone Menus: A Comparison of Two Different Types of Navigation Aids. Human-Computer Interaction 21, 393-433.

9.1. Author Biographies

Kevin B. Bennett is a Professor of psychology at Wright State University, Dayton, Ohio.

He received a PhD in Applied Experimental Psychology from The Catholic University of

America in 1984.

John M. Flach is a Professor and Chair of psychology at Wright State University, Dayton,

Ohio. He received a PhD in Human Experimental Psychology from The Ohio State University in

1984.

Dr.’s Flach and Bennett are co-directors of the Joint Cognitive Systems Laboratory at

Wright State University. They share interests in human performance theory and cognitive

systems engineering, particularly in relation to ecological display and interface design. Their

independent and collaborative efforts in these areas have spanned over three decades.


bennett, k. b. and flach, j. m. (2012). visual momentum

Documents