Collaborative Action (CoAct) Theory: Socially Constructing Shared Knowledge through Mutual Attunement of Shared Affordances

John Teofil Paul Nosek

CONTENTS

Introduction............................................................................................................134

Background............................................................................................................134

Available Affordances versus Perceived and Conceived Affordances..............135

Available Affordances.......................................................................................135

Extending Gibson’s Affordance-Based Theory beyond Physical

Affordances..................................................................................................136

Example of Available versus Perceived Affordances........................................138

Collaborative Action (CoAct) Theory...............................................................138

Example 1: Command and Control..............................................................142

Example 2: Medical Emergency Surgery.....................................................143

Example Applications of CoAct........................................................................143

User Experience Design...............................................................................143

Reducing Data Overload/More Effective Use of Bandwidth.......................143

Using Coact to Precisely Design User Experience—Exemplar.............................143

Tools and Methods............................................................................................144

CoAct Analysis.............................................................................................144

Claims Analysis.................................................................................................144

Analysis Results of Specific Feature.................................................................147

The Feature...................................................................................................147

Original Design.................................................................................................147

CoAct Analysis—Original Design....................................................................147

Affordances—Original Design.....................................................................148

Semantic and Syntactic Knowledge—Original Design...............................148

Goodness of Design—Original Design........................................................149

Claims Analysis—Original Design...................................................................150

Claims—Original Design.............................................................................150

Implementation Requirements...............................................................................151

Recommendations.............................................................................................151

Redesign............................................................................................................151

CoAct Analysis of Redesign.............................................................................151

Affordances—Redesign...............................................................................152

Semantic and Syntactic Knowledge—Redesign..........................................152

Goodness of Redesign..................................................................................153

Claims Analysis—Redesign..............................................................................153

Claims—Redesign........................................................................................154

Discussion of Original Design and Redesign Using CoAct..............................154

Have the Affordances Changed?.......................................................................155

Has Semantic and Syntactic Knowledge Been Reduced?.................................155

Has Just-in-Time Training Been Improved?.....................................................155

Summary................................................................................................................156

Acknowledgments..................................................................................................156

References..............................................................................................................157

INTRODUCTION

Collaborative Action Theory (CoAct), an affordance-based theory based on ecological psychology, explicates how constructing shared affordances provides an alternative model to constructing shared mental models to achieve effective team action (Nosek, 2011). CoAct provides preciseness in what is needed to develop shared affordances to achieve appropriate action. Computer-based applications are integral to achieving effective team action and user-experience design of these computer-based applications is critical to their success. User-experience design is an example of asynchronous work between designer and user to effect appropriate action at the appropriate time. This chapter reviews CoAct and provides a detailed exemplar of how CoAct can be used to achieve preciseness in user experience design.

BACKGROUND

Nosek (2011) explains CoAct in detail and describes initial efforts to test CoAct. This section reviews key elements of CoAct. As actors move within an environment, they discern available informational structures that afford action possibilities. These action possibilities, affordances, are available for the class of actors who have the same potential to discern informational structures that provide these affordances from the same observation point (Gibson, 1979).

Cognition is the embodied, embedded and always situated process whereby life forms bound to their respective environments in an essential dialectical relationship thrive “to persist and prevail”1 within the existential spatio/temporal framework defined by their own corporeal dynamics.... A life form and its environment constitute a “closed purposive organization” bound by a relationship of mutual influence.

(Ferreira, 2019, pp. 1,2)

Available Affordances versus Perceived and Conceived Affordances

While an affordance is potentially available for the class of actors with certain capacities, a perceived or conceived affordance is what emerges or surfaces for a given member of the class at a specific moment in time as the member moves through the environment to achieve some goal, i.e., while affordances are available to all actors with similar capacities and can be defined statically, perceived and conceived affordances are a subset of available affordances that emerge dynamically based on what the actor is engaged in at the time and what the goal of the actor is. Affordances can be perceived or conceived. When actors perceive and act, they are not self-aware of the action opportunity of the affordance, whereas, when actors conceive and act, they are self-aware of the action opportunity of the affordance. Most actors perceive and then act. To simplify reading, perceived is used in the text, but conceived is usually possible in those circumstances.

Available Affordances

The critical difference between available affordances and perceived affordances is often overlooked and may be the cause of some confusion. This may stem from researchers who have been introduced to Gibson’s work on affordances through Norman and Draper’s (1986) work in design of human-computer interaction. Affordances, as used in human interface design, are usually inferred to be available to all humans and not dependent on individual differences. Norman is associated with the idea that interfaces should be “user friendly,” i.e., so clear that they intuitively afford any human user what to do with the interface at any time.

However, affordances only make sense when actor capacities are coupled with available informational structures within an environment, i.e., actor and environment are inexorably linked, and affordances exist at the intersection of actor capacities and the environment (Gibson, 1979) (see Figure 6.1).

For a given environment, a change in an actor’s capacities can change available action possibilities within the environment (affordances); and for a given actor’s capacities, a change in the informational structures in the environment changes affordances (see Figure 6.2a and b).

Affordances = intersection of actor capacities and the environment

FIGURE 6.1 Affordances = intersection of actor capacities and the environment.

(a) Changed capacities (increased); (b) changed environment (increased)

FIGURE 6.2 (a) Changed capacities (increased); (b) changed environment (increased).

Extending Gibson's Affordance-Based Theory

beyond Physical Affordances

While Gibson (1979) focused on actor effectiveness based on affordances that are discernable to a class of actors with certain physical capacities to act, CoAct extends the notion of capacities to include more than just physical attributes. This is consistent with Gibson’s view of the potentially broad applicability of his theory and the principles of ecological psychology (Heft, 2001). “Gibson’s idea of the environment as shared by all perceivers is understood not just for the physical but also for the social environment” (Szokolszky & Read, 2018, p. 20). “An ecological approach presumes a symbiotic relationship between the environment and an agent who is actively seeking out and realizing novel functional relations and therefore is constantly changing the functional environment” (Szokolszky & Read, 2018, p. 9). Szokolsky and Read further argue that Gibson’s work should be at the heart of a new, broad science of developmental ecological psychology with the goal to explain “how the organism develops by keeping continuous, active, and reliable contact with its richly structured environment via direct perception and action, over time” (2018, p. 27).

Capacities to act can broadly include capacities to do, think, feel, etc. (Szokolszky & Read, 2018). For example, informational structures in the environment may afford sadness or happiness for one class of actors with certain capacities, but not another. In addition, the idea of an actor moving through an environment is not restricted to the physical movement of the actor within a physical environment but includes the notion that actors are active with respect to what informational structures they pick up, i.e., the notion of static is the exception. Moving through the environment can mean the working through a problem, the reading of a book, the engagement in a discussion. The importance of the concept of moving through the environment means that one is continually experiencing one’s environment and is forever changed by this experience, i.e., “cognitively rewired” (Pizlo, 2007). Even visual perception to recognize objects in the environment appears to be dependent on prior experience and/or genetic encoding (Pizlo, 2001, 2007). “Cognitive systems do not passively react to events; they rather actively look for information and their actions are determined by purposes and intentions as well as externally available information and events” (Hollnagel & Woods, 2005, p. 16).

Figure 6.3 depicts how perceived affordances are dynamic and change over time given the goal and current activity. For simplicity, in Figure 6.3 the informational structures within the environment and actor capacities remain the same through time, which means that the set of affordances remain the same for actors in the class with these capacities, while what is perceived within this set changes over time.

Perceived affordances = subset of affordances based on goal and current activity over time

FIGURE 6.3 Perceived affordances = subset of affordances based on goal and current activity over time.

Example of Available versus Perceived Affordances

As noted previously, affordance are action opportunities in the world identified at the intersection of the informational structures in the environment and the capacities of actors. They are static for a given environment and actor capacities and always exist whether perceived or not. Perceived and conceived affordances are those subsets of affordances that emerge when an actor desires to achieve some goal and is engaged in some current activity.

Assume you are sitting in a room. There is a door with a typical doorknob that you used to enter the room. For the typical person of average human height and two arms and hands, in normal conditions, the door affords safe egressibility; the doorknob affords graspability (it would not afford graspability for a one-foot tall person with no arms) and turnability (clockwise or counter-clockwise), and the door affords pushability and/or pullability depending on which way it opens. In normal conditions these affordances always exist for you whether you perceive or conceive them.

When you want to exit the room (goal) and move in the direction of the door (current activity), you grasp the doorknob, turn it, push or pull, and exit. For most of these actions you will not be self-aware of the affordances. For an unfamiliar door you may consciously decide if the door should be pushed or pulled, i.e., conceive of the affordance of pushablity or pullability. Otherwise, these affordances are perceived (you are not self-aware) and you act out these action opportunities.

Assume the door no longer affords safe egressiblity because there is a fire or some danger on the other side of the door. The affordances of doorknob graspability and turnability and the affordance of door pushability or pullability continue to exist for you. However, now you seek safe egressibility and move towards the window. You will not perceive or conceive of doorknob graspability and turnability nor door pushability or pullability. The window affords safe egressibility if you can safely egress through the window, if you can’t, then it does not afford safe egressability. As you move to the window, you are more likely conceiving (consciously evaluating) the openability or breakability of the window because of the unfamiliar situation.

You may overestimate your capacities and injure or die egressing the window. In that case the window did not afford you safe egressibility and does not afford safe egressibility for someone with similar capacities.

Collaborative Action (CoAct) Theory

CoAct can be used at multiple levels of granularity, from fine granularity of understanding a single interaction, to tracking intermediate progress and final results of interactions (see Figure 6.4).

For effective collaborative action, actors, whether human or non-human, must perceive shared, relevant affordances at the appropriate time. A shared affordance is an affordance that is shared by more than one actor. To achieve a shared, perceived affordance: (1) the actors must share sufficient capacities so that a given environment affords the same action opportunity for the actors; and (2) this affordance or action opportunity is available (or perceived) at the appropriate time, i.e., an affordance that is available to an actor, but is not perceived when it is supposed to be, is

Attunement

Capacity Sufficiency

Relevant Informational Structures

Selectivity Calibration (Aid Perceiving)

Affordances (Perceived & Conceived)

Target Actor’s

Target Actor’s

Model of Collaborative Action (CoAct)

FIGURE 6.4 Model of Collaborative Action (CoAct).

not considered a shared, perceived affordance. A shared, perceived affordance may occur at the same time among actors, but it may also occur asynchronously, as long as this is considered an appropriate time to support collaborative action.

The process to achieve and the end-state of achieving shared, perceived affordances use the same label, “attunement.” Attunement (n.d.) means “being or bringing into harmony; a feeling of being ‘at one’ with another.” For example, one can say that collaborators attune each other, i.e., undergo mutually attunement (bringing into harmony) to achieve attunement (being in harmony) of perceiving shared relevant affordances at the appropriate time. Context determines which form of the word is meant.

Actors engaged in collaborative action must take responsibility for mutual attunement by (1) sharing relevant informational structures, (2) bringing each other up to the sufficient capacities of the class of actors for whom the environment affords the desired, relevant action opportunities (affordances), and (3) assisting each other in perceiving the subset of relevant affordances at the appropriate time (selectivity calibration).

In Figure 6.5, we identify the actor who is actively attuning as the Source Actor, and the actor who is the target of this attunement effort as the Target Actor.

However, these roles alternate as collaborators engage in mutual attunement. Initially, at Time 1, the actors do not share the relevant informational structures

Capacities

Informational Structures

Capacity Sufficiency

Capacities

Time 1

Environment

Affordances

Enviro

Afford. mees

Attunement

Selectivity Calibration (Aid in Perceiving - Goal + Activity)

Environment

Time 2

FIGURE 6.5 Attunement to perceive shared, relevant affordances at the right time.

within the environment and sufficient capacities. Therefore, they cannot perceive relevant affordances. This is depicted by the actor at the bottom (the Target Actor) only sharing half of what the actor on the top (Source Actor) has available.

In Time 2, the Source Actor attunes the Target Actor by sharing relevant informational structures in the environment, assisting the Target Actor to achieve sufficient capacities (this is shown by the capacities of the Target Actor are now the same as the Source Actor), and assisting the Target Actor in perceiving relevant informational structures at the appropriate time (selectivity calibration). At the end of Time 2, both actors are perceiving similar, relevant affordances. For simplicity’s sake, although we only depict one Target Actor in Figure 6.5, but there could be more than one Target Actor in a given collaborative act.

In Gibsonian terms, actors attune each other to perceive relevant affordances by building sufficient capacities in each other, exchanging informational structures so they are available to both, and bringing each other to a common observation point within the environment of available informational structures at the appropriate time (see Figure 6.6).

Attunement can be intended and unintended, explicit and implied, verbal and nonverbal. For example, an actor may explicitly expose his/her identity intending to imply positive informational structures as to status, however, the receiving actor may not treat this datum as positive. As noted earlier, actors attune each other by sharing relevant informational structures in the environment, assisting each other to achieve sufficient capacities, and assisting each other in perceiving relevant informational structures at the appropriate time. Table 6.1 summarizes components of attunement.

Note, not all components exist in every act of attunement. Attunement related to selectivity calibration aligns the goals and activities of actors. This relates to Gibson’s idea that an actor moves within the environment picking up informational structures to achieve some goal and affects what affordances are perceived at a given moment in time.

Gibsonian mutual attunement

FIGURE 6.6 Gibsonian mutual attunement.

TABLE 6.1

Summary of Attunement Components

Informational Structures: Relevant aspects of the environment/situation.

Capacity Sufficiency Building: Building sufficient, similar capacity needed so that affordances needed for collaboration are available to the actors.

Perceived Affordances: Perceived action opportunities in the environment (inferred through observation).

Conceived Affordances: Conceived action opportunities in the environment (explicit through self-aware reflection).

Selectivity Calibration: Assistance in perceiving the relevant affordances at the appropriate time.

Achieve goal or fulfil) need—align goals; Current activity—align current activity.

One aspect of the model may not be adequately depicted but deserves special mention. Target Actors may accept the affordances of the Source Actor without being brought up to sufficient capacities and having shared informational structures so that these affordances are available for Target Actors themselves. For example, Source Actors, who are trusted, may be able to transfer their affordance, but the Target Actors do not have sufficient capacities so that they can perceive the affordances for themselves.

If Source Actors expose their affordances (perceived and conceived) as part of their attunement for a given shared environment, is this a form of learning? That is, does this teach the Target Actor, either explicitly through conceived affordances transmitted or implicitly through perceived affordances ascertained through observation, that these and future, similar informational structures afford these action possibilities? Does repetition reinforce building sufficient capacities? Is learning by example a form of this process?

Example of stratifying the shared social world of an individual actor

FIGURE 6.7 Example of stratifying the shared social world of an individual actor.

Note, attunement is neither inherently good nor correct and can result in building sufficient capacities so that Target Actors perceive good, correct affordances or warped, incorrect affordances (Boyle, Kacmar, & George, 2008). For example, Source Actors who enjoy very high trust, such as some fanatical religious and political leaders, may be able to transfer their affordances of world events in a continuously, w'arped manner such that Target Actors will perceive affordances within current and future environments that will result in unwarranted and apparently irrational actions, but consistent with CoAct.

Habermas (1984) offers the concept of lifeworlds to explain the difficulty in achieving shared affordances among actors from different lifeworlds. Borrowing loosely from Habermas, actors of dissimilar backgrounds may be able to find some common elements of lifeworlds that can be shared. Figure 6.7 provides an example of stratifying the shared social world of an individual actor.

The unshared subjective world can only be contained within the shared social world. An actor living in the world is part of and helps to create a shared social world.

For example, the Target Actor may be an integral part of the decision making process but is not experiencing the situation for himself or herself and may not have the ability to understand the relevancy of facts of the situation at hand. In this case, actors must find a common aspect of a lifeworld to share, for example, the capacities of actors from the two different classes may intersect on similar cultural or human capacities to pick up informational structures to share that the situation affords danger. Two examples are offered to illustrate this.

Example 1: Command and Control

In a military situation, a commander on the ground feels threatened, perceives the attackability of the enemy, but must obtain permission to engage from a distant political leader with no military background. In this case, the Source Actor from the class of experienced warfighters must obtain permission from this politician, the Target Actor, who is not within the class of experienced warfighters, and may not be attunable within limited time requirements, to perceive the affordance of attackability of the enemy. However, the Source Actor (warfighter) can then expose behavior that will permit the political decision maker to perceive that the situation affords danger. The decision maker at headquarters would then trust the local onscene commander to make the best decision (including attacking) and implicitly trust the advancement system (shared culture) that puts commanders in this position of responsibility.

Example 2: Medical Emergency Surgery

In emergency exploratory surgery, family members and patients may share decisionmaking responsibility with surgeons but do not have the capacities of the surgeon to pick up informational structures that afford necessary surgery. It may be even more difficult for family members than for patients, who may be experiencing pain, to perceive affordances available to the class of surgeons. Surgeons should, and the best most likely do, focus on selectivity calibration and informational structure sharing to achieve the shared affordance of imminent danger among surgeon, patient, and family members. Patients and family members must then trust the surgeon to make the best decision and implicitly trust the medical system (shared culture) that puts surgeons in this position of responsibility.

Example Applications of CoAct

In addition to the examples in group decision making provided above, two examples of other applications of CoAct include user experience design and information over-load/reduced bandwidth.

User Experience Design

Humans, using computers to complete their work, progress through a sequence of cycles of interaction steps and information displays, where the last information display provides affordances for the next interaction step as the user acts to achieve some goal. For each step in the cycle, the designer assumes an envisioned user, i.e., the design will provide appropriate affordances at the appropriate time for a user with certain capacities engaged in an activity to achieve some goal. Differences between the envisioned user and the actual user determines design quality and attunement needed, such as training for the actual user.

Reducing Data Overload/More Effective Use of Bandwidth

Because some data may provide more informational value that other data, in times of reduced bandwidth, CoAct may provide guidance on prioritizing what data should be transmitted or received.

USING COACT TO PRECISELY DESIGN USER

EXPERIENCE—EXEMPLAR

In this exemplar, CoAct and Claims Analyses (Rosson & Carroll, 2002) are integrated to assess usability of a computer-based interface, to make informed recommendations for improving usability, and to evaluate the usability of a redesigned interface. This process allows developers and usability experts to uncover key usability issues in their design and to determine if the interface satisfies usability requirements for their audience. The evaluation also helps to locate opportunities for just-in-time training to mitigate the need for the user to hold specific semantic and syntactic knowledge prior to using the system.

This exemplar is divided into the following sections: (1) an introduction to the utilized tools and methods, (2) an initial evaluation of a simple interface feature and redesign recommendations, (3) an evaluation of the redesigned feature, and (4) a usability comparison between the original and redesigned features.

Tools and Methods

CoAct Analysis

CoAct Analysis provides a technique to assess the usability of interface features involved in completing a given activity. This involves identifying an activity to analyze and then proceeding through the activity until completion while determining the affordances provided by the interface at each step. An expanded version of the CoAct Analysis which also evaluates effectiveness of the design in providing certain affordances was used. Figure 6.8 presents the template that was used in this analysis.

The following list briefly describes the fields within the template:

  • Activity: A specific activity supported by the interface. Usability of the interface in achieving completion of this activity will be assessed.
  • Emerging Goals/Subgoals: The smaller tasks involved in completing the larger activity. Consideration of the following questions for each subgoal:
  • Is the first step to take to achieve the subgoal obvious?
  • After achieving the subgoal, is the next step after that obvious?
  • Affordances: The possible actions that are available to the user based on the capacities of the user and the design of the interface, both of which we consider in further detail:
  • Capacities of the User
  • Semantic Knowledge: The conceptual knowledge needed by the user to make an affordance available.
  • Syntactic Knowledge: The knowledge of specific steps needed to be completed by the user to make an affordance available.
  • Design of the Interface
  • Design to Achieve: The interface design and features that allow an available affordance to be recognized by the user.
  • Goodness of Design: The effectiveness of the design to achieve and any possible improvements that may allow an affordance to be more readily recognized by the user.

Claims Analysis

The Claims Analysis provides a methodology for assessing specific design features of an interface and determining if their design effectively supports their intended function. This analysis deals specifically with the organization and aesthetics of the features and relies on Gestalt principles to determine which interface features effectively support usability of the interface. This analysis can provide insight for both the current design and suggested redesigns, and considers the implementation involved in a redesign. Figure 6.9 presents the template used in this analysis.

CoAct Analysis

Activity title

Page Title

QI: Is thefrst step to taketo achieve thesubqoalobvious?

02: After cchievnq the subqoal, is the next step after that obvious?

Screen#

Informal Cognitive Walkthrough

Emerging

Goal/Subgoal

Affordance

Essential Capabilities

Knowledge) Required

QI

Q2

Semantic

Design to Achieve

Goodness of Design

Syntactic

Design to Achieve

Goodness of Design

1

2

3

FIGURE 6.8 CoAct Analysis Template.

Collaborative Action (CoAct) Theory 145

4—

Claims Analysis put/V/ty title

I Page Title

Feature If

Design Feature

Claims

Positive

Negative

How to Impement

Difficulty to Impement

FIGURE 6.9 Claims Analysis Template.

Foundations and Theoretical Perspectives of Distributed Team Cognition

The following list briefly describes the fields within the template:

  • Design Feature: A specific feature included in the interface.
  • Options: A specific design option for the feature, either the current design or a recommended redesign.
  • Claims: Statements about the design of an interface feature, including:
  • Positive Claims: Those that demonstrate a design supports usability.
  • Negative Claims: Those that demonstrate a design does not support or detracts from usability.
  • For possible redesign options, the analysis considers the following from the development perspective:
  • How to Implement: A brief description of the development needed to implement the redesign.
  • Difficulty to Implement: A rough estimate of the difficulty of the development needed to implement the redesign (e.g., low difficulty, moderate difficulty).

Analysis Results of Specific Feature

The Feature

For this exemplar, a simple interface feature of the Virtual Teaching Assistant (VTA) web application was analyzed. VTA provides an interface for autism therapy providers to perform therapy management and progress tracking of their students. On the Therapy Assignment Page, VTA users can perform the activity of assigning and/or updating therapy programs and behaviors for a student. Selecting a student emerges as a subgoal involved in completing this activity. The analysis that follows will focus specifically on the interface feature that allows a user to select a student during the therapy assignment activity.

Original Design

The original design feature for student selection combines an input box with a dropdown list (see Figure 6.10). The input box allows a user to search for a specific student, while the dropdown list allows the user to view all available students.

CoAct Analysis—Original Design

The following sections highlight key information gathered from the CoAct Analysis, including affordances provided by the interface, semantic and syntactic knowledge needed by the user to utilize these affordances, and the effectiveness of the interface design in supporting user recognition of these affordances.

! ones. Saadi (=323) » □ cwy m, smaenu’

Jones, Sandi (#323)

Smyth, Bob (#324)

Student, One (#244)

Student, Another (#277)

• Clicking the dropdown arrow opens a dropdown list of all available students.

b «il On(, My SluOenK?

Smyth, Bob (#324)

p~i

h ’ O Only My SlwtonU?

Jones, Sandi (#323)

Smyth, Bob (#324)

Student, One (#244) |

Student, Another (#277)

Student. One (»24[1] ’ J Only My Sluaents? &i

Indent. One (»244) — Ci Only My Students’ &

Jone1 ^andi

Smy student Selection Select from list, Stuc or enter Student's last name. Press Stuc [ENTER]

Coins

  • Typing in the input box opens a dropdown list of filtered results.
  • Hovering over a student name with the cursor highlights the active student name and makes it available for selection.
  • Clicking the name selects that student.
  • The tooltip overlaps the dropdown list of students.

FIGURE 6.10 Original design.

Affordances—Original Design

The following list highlights the affordances provided by the interface for the user with the necessary semantic and syntactic knowledge:

  • • Revealability of the tooltip for selecting and searching students
  • • Revealability of available students
  • • Searchability of student’s name
  • • Selectability of student
  • • Clearability of previously selected student’s name

Semantic and Syntactic Knowledge—Original Design

In general, most of the semantic and syntactic knowledge needed to make these affordances available is possessed by the average computer user, including understanding the conceptual processes and executing the steps involved in moving the cursor over a feature and interacting with the feature through clicking. However, analysis reviews a few key areas in which semantic and syntactic knowledge needed by the user could be reduced or simplified:

  • • Upon continued use of the therapy assignment page, the user must know that the student’s name in the input box can be removed by backspacing the text and that another student’s name can be entered. This knowledge cannot be readily acquired or understood through the interface design.
  • • The user must know' that a tooltip with further instructions will open upon hovering over the dropdown arrow. The tooltip’s connection to the arrowmay be beneficial when a user decides to use the dropdown arrow, and then receives further information. However, if a user does not attempt to use the dropdown arrow, the knowledge of the tooltip or the knowledge provided by the tooltip will not be readily available.
  • • Since the tooltip contains information for both methods (input box search or dropdown list selection) of selecting a student, revealing the tooltip should require the same semantic and syntactic knowledge when using either method to simplify the knowledge needed to utilize the tooltip.
  • • Since using either method (input box search or dropdowm list selection) leads to completion of the same goal (selecting a student), revealing the list of available students should require the same semantic and syntactic knowledge when using either method. In general, the methods for selecting a student should be more tightly integrated to simplify the knowledge needed to select students.

Goodness of Design—Original Design

The above issues related to semantic and syntactic knowledge needed by the user can be remedied thorough improving the design of the interface, especially in providing the user more thorough just-in-time training. The following list highlights issues related to goodness of design:

  • • Upon first use of the therapy assignment page, the input box should not be blank. Rather, the blank space should be used to provide brief just-in-time training for the user.
  • • Upon continued use of the therapy assignment page, the interface should provide clear visual indication that the name of a previously searched student can be removed to perform a new' search. The removal process should require few'er steps than current design of manual backspacing.
  • • An expanded area should be able to trigger the tooltip. This allows the just-in-time training that it provides to be more easily accessible, no longer requiring the user to navigate to a highly specific part of the screen (i.e., the dropdown arrow).
  • • The tooltip should be positioned such that it no longer overlaps the dropdown list of students. Since it currently interferes with functionality, a user may find it disruptive and be less inclined to benefit from the just-in-time training that it provides.
  • • The tooltip offers the most thorough just-in-time training for the user. Therefore, its instructions should be more comprehensive to benefit all levels of users.

• The combination box should be more tightly integrated so the process of selecting a student becomes more specific and streamlined. Currently, the combination box seems to offer two functions: student searching using the input box and student selection using the dropdown list. Rather, the interface should approach this feature to perform one coherent action: selecting a student.

Claims Analysis—Original Design

The following sections highlight important positive and negative claims made for both the original interface design, as well as for possible redesigns which aim to solve specific issues identified in the CoAct Analysis. The analysis also considers implementation factors for each redesign option.

Claims—Original Design

Performing a Claims Analysis on both the original design and possible redesigns allows for the most effective interface features to be chosen based on the ratio between positive and negative claims. For most features, the analysis allowed for identification of designs that eliminated negative claims. The following list describes the claims made for each feature in the interface for both the original design and any effective redesigns:

  • Tooltip for student selection/tooltip for student searching
  • Current Design: Tooltip triggered when user hovers on down arrow
  • Positive Claims: Does not clutter screen.
  • Negative Claims: Just-in-time training not initially visible, user must hover over specific area to trigger tooltip, tooltip may be overlooked by a user that does not use the down arrow, tooltip overlaps dropdown list, and limited instructions provided by the tooltip.
  • Effective Redesign: Tooltip triggered when user hovers on entire combination box, with a placeholder in the blank input box, repositioning of the open tooltip, and revised tooltip content.
  • Positive Claims: Does not clutter screen, placeholder provides brief just-in-time training, tooltip will be triggered when the user interacts with any area of the combination box, tooltip does not interfere with the dropdown list, and comprehensive instructions provided by the tooltip.
  • Negative Claims: None.
  • Student selection dropdown/ student search box
  • Current Design: Dropdown list opens when user clicks down arrow or user begins entering text to search for student
  • Positive Claims: Down arrow often associated with dropdown lists and input boxes often associated with text input.
  • Negative Claims: Must either begin searching or click a specific area (i.e., down arrow) to see all available students, the dropdown list and search functions not seamlessly integrated, and no visual indication that a previous student name can be removed to begin a new search.
  • Effective Redesign: Dropdown list opens when user clicks input box, with down arrow removed from the design and an (x) button in the input box to easily clear the previous search.
  • Positive Claims: Provides only one simple course of action for interacting with the input box (i.e., clicking into it) which automatically opens list of available students, tightly integrates dropdown list and searching into one function, and provides a visual indication that a previous student name can be removed to begin a new search.
  • Negative Claims: None.

IMPLEMENTATION REQUIREMENTS

All redesigns can be implemented through simple changes in HTML, CSS, and JavaScript. Implementing the (x) button for clearing a search term from the input box may be slightly more involved. However, existing implementation solutions are readily available from open source websites. All suggested redesigns could be classified as low difficulty implementation. Thus, implementation factors did not affect the redesign recommendations.

Recommendations

Based on findings from the CoAct and Claims Analyses, the following redesigns are recommended to increase usability of the interface:

  • • Add a placeholder in the empty input box that reads “Select or search a student.”
  • • Remove down arrow, leaving only the input box.
  • • Trigger tooltip when user hovers over input box.
  • • Add more comprehensive instructions to the tooltip: “Click to see list of available students. Enter student’s name to search. Click student name to select. Click (x) to clear student name and begin new search.”
  • • Reposition tooltip such that it does not overlap the dropdown list.
  • • Add (x) button to input box to clear previously selected student.

Redesign

The redesign incorporates recommendations derived from the CoAct and Claims Analyses performed on the original design (see Figure 6.11). The redesigned feature provides an input box that allows the user to view all available students from a dropdown list as well as search for a specific student.

CoAct Analysis of Redesign

The following sections highlight key information gathered from the CoAct Analysis of the redesign, including affordances provided by the interface, semantic and

  • • Add a placeholder in the empty input box that reads: "Select a student."
  • • Remove the down arrow, leaving only the input box.
  • • Trigger the tooltip when a user hovers over input box.
  • • Add more comprehensive instructions to the tooltip.
  • • Reposition the tooltip such that it does not overlap the dropdown list.
  • • Open the dropdown list when user clicks input box.
  • • Add an (x) button to the input box to clear previously selected student.

FIGURE 6.11 Recommended redesign.

Select a student

Select a student

Click to see list of available students. Enter student's name to search. Click student name to select. Click (x) to clear student name and begin new search.

■m:

X Only My

Jones, Sandy (#323)

Smith, Bob (#324)

Student, One (#244)

Student, Another (#277)

syntactic knowledge needed by the user to utilize these affordances, and the effectiveness of the interface design in supporting user recognition of these affordances.

Affordances—Redesign

The following list highlights the affordances provided by the redesigned interface:

  • • Revealability of the tooltip for selecting and searching students
  • • Revealability of available students
  • • Searchability of student’s name
  • • Selectability of student
  • • Clearability of previously selected student’s name

Semantic and Syntactic Knowledge—Redesign

When you compare the semantic and syntactic knowledge required in the original design to the redesign, to make similar affordances available the knowledge requirement has been reduced. The following list describes the semantic and syntactic knowledge needed by the user to make affordances available in the redesigned interface:

  • • Interacting with the input box will allow for student selection.
  • • Clicking the input box will open a list of available students.
  • • Typing in the input box will filter the list of available students.
  • • Moving the cursor over a student’s name and clicking will select that student.

• Clicking the (x) button will clear the previously selected student in order to search again.

Goodness of Redesign

The semantic and syntactic knowledge requirements of the user have been optimized through the good design of the interface. The design reduces prior semantic and syntactic knowledge required of a user and therefore expands the potential number of users with sufficient existing semantic and syntactic knowledge to complete their work. The design reduces the need for a priori training. Incorporating just-in-time training decreases the knowledge needed in advance by a user to utilize the interface, facilitates user perception of the affordances available in the interface, and ultimately guides the user through an activity. The following list highlights the design features that contribute to goodness of design:

  • • Upon first use of the therapy assignment page, the input box contains a placeholder that reads “Select or search a student.” The placeholder prompts the user to interact with the input box in order to search or select a student. The user must only have semantic and syntactic knowledge related to moving the cursor.
  • • Upon interacting with the input box after the initial prompt, a tooltip with comprehensive instructions immediately opens and provides the following training for the user: “Click to see list of available students. Enter student’s name to search. Click student name to select. Click (x) to clear student name and begin new search.” Thus, when a user chooses to interact with the tool for student selection, the interface offers the semantic and syntactic knowledge needed to perform any action related to student selection: viewing the available students, searching a student, selecting a student, and beginning a new search. Again, the user must only hold knowledge related to moving the cursor, clicking, and typing with the keyboard.
  • • In addition to the verbal instruction in the tooltip, the (x) button provides clear visual indication that the name of a previously searched user can be removed. This feature simplifies the knowledge needed by the user to perform subsequent student searches by reducing the action of clearing a previously selected student into one step (i.e., clicking the (x) button).

Claims Analysis—Redesign

The following section highlights important claims made for the redesigned interface. The Claims Analysis also considers Gestalt principles that “describe the configurai properties of visual information” (Rosson & Carroll, 113). When used successfully in an interface, “the principles of Gestalt perception direct attention to the many relationships among information elements in a complex display” (Rosson & Carroll, 114). Therefore, effective use of Gestalt principles in interface organization contributes to positive claims made about an interface.

Claims—Redesign

The redesign resolves issues present in the original design, thus reducing negative claims for the student selection feature. The Claims Analysis for the redesign focuses on positive claims and demonstrates the redesign’s consistency with Gestalt principles. The following list describes the positive claims made and Gestalt principles followed for each feature in the interface:

  • Tooltip for student selection/tooltip for student searching
  • Current Design: Tooltip triggered when user hovers input box, with a placeholder in the blank input box, repositioning of the open tooltip, and revised tooltip content.
  • Positive Claims: Does not clutter screen, placeholder provides brief just-in-time training, tooltip will be triggered when the user interacts with any area of the input box, tooltip does not interfere with the dropdown list, and just-in-time training provided by comprehensive instructions in the tooltip.
  • Gestalt Principles: Proximity (tooltip opens near the input box), Closure (placeholder located within the input box, encapsulating the just-in-time training within the figure), and Area (placeholder located within the input box, reducing the size of the overall figure). These principles indicate the relationship between various parts of the figure.
  • Student selection dropdown/ student search box
  • Current Design: Dropdown list opens when user clicks input box and (x) button in the input box clears the previous search.
  • Positive Claims: Provides only one simple course of action for interacting with the input box (i.e., clicking into it) which automatically opens a list of available students, tightly integrates using dropdown selection and searching into one function, and provides a visual indication that a previous student name can be removed to begin a new search.
  • Gestalt Principles: Closure (the input box, the (x) button, and the dropdown list all form one closed figure), Area (the (x) button resides within the input box, and the dropdown list opens immediately beneath the input box, reducing the space used by the overall figure), and Similarity (student names in the dropdown list are formatted consistently). These principles indicate the relationship between various parts of the figure and reveal when certain elements (i.e., student names) represent the same information type.

Discussion of Original Design and Redesign Using CoAct

The following discussion compares findings from the analyses performed for both the original design and the redesign and explores the effect of the analyses on improving key elements of the interface. The section addresses affordances, semantic and syntactic knowledge needed by the user to utilize each interface, and just-in-time training offered by each interface to increase usability.

Have the Affordances Changed?

Based on the CoAct Analyses, the affordances remain the same for both the original design and the redesign:

  • • Revealability of the tooltip for selecting and searching students
  • • Revealability of available students
  • • Searchability of student’s name
  • • Selectability of student
  • • Clearability of previously selected student’s name

Has Semantic and Syntactic Knowledge Been Reduced?

Yes. In the original design, the user needed more extensive semantic and syntactic knowledge for the affordances to be available. For example, the user needed to initially understand without any prompt that:

  • • The input box provided for student searching.
  • • The dropdown arrow triggered a tooltip with further instructions.
  • • Backspacing the name of a previously selected student would allow for another search.

The user should not be required to hold this knowledge before using the interface. Instead, the interface should reduce the need for previous knowledge to a minimum through just-in-time training. The redesign effectively uses just-in-time training to decrease the knowledge required by the user to do the following:

  • • Move the cursor
  • • Click a feature
  • • Use the keyboard to type

Thus, the redesign allows for users with a minimum knowledge to utilize the interface efficiently.

Has Just-in-Time Training Been Improved?

Yes. In the original design, the interface did provide just-in-time training within the tooltip. However, revealing the tooltip required the prior knowledge that hovering over the dropdown arrow would trigger the tooltip. If the user did not interact with the dropdown arrow, the just-in-time training would not be readily accessible. In addition, the tooltip did not contain entirely clear instructions.

To improve the effectiveness of just-in-time training, a minimum amount of knowledge should allow the user to access it for all levels of users to benefit from the knowledge increase that it provides. Therefore, the user should only be required to hold basic semantic and syntactic knowledge (e.g., moving the cursor, clicking) with just-in-time training and visual cues bridging the gap between the user’s knowledge and the required knowledge to utilize the interface. The following interface features of the redesign support just-in-time training:

  • • The placeholder in the input box instructs the user that the student selection feature can be used to select or search students and acts as a visual cue to interact with this feature if the user aims to perform the action of student selection. Without this brief training, the user might not hold the knowledge that this feature performs the specified activity.
  • • The placeholder acts as a prompt to interact with the student selection feature. When the user interacts with any area of the input box, the interaction triggers a tooltip with further instructions to open. Thus, when the user demonstrates an interest in using this feature, the interface immediately offers the user the knowledge required to perform any action afforded by the feature: viewing the available students, searching a student, selecting a student, and beginning a new search.

The just-in-time training provided by the redesigned interface reduces the a priori semantic and syntactic knowledge needed by the user to a minimum and delivers further instructions when the user needs to perform tasks requiring greater than minimum knowledge. Thus, the training equips all users with the knowledge required to use this interface feature, regardless of prior experience using the application.

SUMMARY

CoAct extends the notions of affordance and moves away from idiosyncratic, subjective mental models of the world to the notion that actors with similar capacities to act can potentially discern similar action possibilities in the world. CoAct has the potential to impact user experience design, provide guidance on reducing data overload among classes of users with different capacities to act, and improve collaborative decision making. In this chapter, an exemplar was provided where CoAct Analysis was integrated with Claims Analysis to improve user experience design. CoAct and Claim Analyses of the original design identified precision in what semantic and syntactic knowledge was required to make affordances available and identified areas for design improvement. CoAct and Claims Analyses of the redesign were used to precisely identify the reduced semantic and syntactic knowledge required to make the same affordances available and how redesign will aid in helping users to perceive available affordances.

ACKNOWLEDGMENTS

It was a pleasure to work with Sharon Tartarone in completion of the exemplar. She quickly grasped CoAct and provided a clear analysis of how CoAct can provide precision in user experience design.

REFERENCES

attunement. (n.d.). Dictionary.com’s 21st Century Lexicon. Retrieved September 25, 2009, from Dictionary.com website http://dictionary.reference.com/browse/attunement,

Boyle, R. J., Kacmar, C. J., & George, J. F. (2008). Distributed deception: An investigation of the effectiveness of deceptive communication in a computer-mediated environment. International Journal of e-Collaboration, 4(3), 14-39.

Ferreira, M. I. A. (2019). Cognitive architectures: The dialectics of agent/environment. In M. I. A. Ferreira. J. S. Sequeira, & R. V. Ventura (Eds.), Intelligent systems, control and automation: Cognitive architectures (Vol. 94, pp. 1-12). Switzerland: Springer Nature.

Gibson, J. J. (1979). The ecological approach to visual perception. Boston, MA: Houghton-Mifflin.

Habermas, J. (1984). The theory of communicative action: Reason and the rationalization of society. Boston, MA: Beacon Press.

Heft, H. (2001). Ecological psychology in context. Mahwah, NJ: Laurence Erlbaum.

Hollnagel, E., & Woods, D. D. (2005). Joint cognitive systems. New York: CRC Press.

Norman, D. A., & Draper, S. W. (1986). User centered system design. Mahwah, NJ: Laurence Erlbaum.

Nosek, J. T. (2011). Towards an Affordance-based Theory of Collaborative Action (CoAct). International Journal of e-Collaboration, 7(4), 37-60.

Pizlo, Z. (2001). Perception viewed as an inverse problem. Vision Research, 41(24), 3145-3161.

Pizlo, Z. (2007). Human perception of 3D shapes. In W. G. Kropatsch, M. Kampel. & A. Hanbury (Eds.), Computer analysis of images and patterns (Vol. 4673, pp. 1-12). Berlin: Springer.

Rosson, M. B., & Carroll, J. M. (2002). Usability engineering: Scenario-based development of human computer interaction. New York: Morgan Kaufmann.

Szokolszky, A., & Read, C. (2018). Developmental ecological psychology and a coalition of ecological-relational developmental approaches. Ecological Psychology, 30(V), 6-38. https://doi.org/10.1080/10407413.2018.1410409.

~7 Mismatches between

Perceiving and Actually Sharing Temporal Mental Models

  • [1] Upon first use of the Therapy Assignment Page, the user must know that the input box can be used to search for students. This knowledge can be acquired through the tooltip, yet opening the tooltip requires even further semantic and syntactic knowledge.
 
Source
< Prev   CONTENTS   Source   Next >