idea generation through empathy: reimagining the "cognitive walkthrough"

COLIN M. GRAY1, SEDA YILMAZ1, SHANNA R. DALY2, COLLEEN M. SEIFERT2, & RICHARD GONZALEZ2 1 Iowa State University; 2 University of Michigan

IDEA GENERATION THROUGH EMPATHYREIMAGINING THE “COGNITIVE WALKTHROUGH”

TECHNICAL COMPETENCIES

+HUMAN-CENTERED

COMPETENCIES

HUMAN-CENTERED COMPETENCIES

[Zowghi & Paryani, 2003; Mohedas et al., 2014; Lande & Leifer, 2010; Walther et al., 2012; Strobel et al., 2013]

limited examples of methods to systematically encourage the

development of empathy

EMPATHY

the cognitive knowing of what another person is feeling

the emotional feeling what another individual is feeling

the act of responding to another’s experience with compassion

[Levenson and Ruef, 1992; Walther et al. 2012]

EMPATHY

COGNITIVE WALKTHROUGHCHI 2000 • 1-6 APRIL 2000 Papers

1. Define inputs to the walkthrough a. Identification of users b. Sample tasks for evaluation c. Action sequences for completing the tasks

d. Description or implementation of interface 2. Convene the walkthrough

a. Describe the goals of the walkthrough

b. Describe what will be done during the CW c. Describe what will not be done during the

walkthrough

d. Explicitly defuse defensiveness e. Post ground rules in a visible place f. Assign roles g. Appeal for submission to leadership

3. Walkthrough the action sequences for each task a. Tell a credible story for these two questions:

- Will the user know what to do at this step? - If the user does the right thing, will they know

that they did the right thing, and are making progress towards their goal?

b. Maintain control of the CW, enforce the ground rules

4. Record critical information

a. Possible learnability problems b. Design ideas c. Design gaps d. Problems in the Task Analysis

5. Revise the interface to fix the problems

DETAILED DESCRIPTION OF THE STREAMLINED WALKTHROUGH PROCEDURE 1. Define inputs to walkthrough Before the CW session, the usability professional is responsible for defining the important user task scenario or scenarios and producing a task analysis of those scenarios by explicating the action sequences necessary for accomplishing the tasks in the scenarios. Wharton et al. should be used as a resource for determining how to decide on the scenarios and how to describe the task sequence.

2. Convene the walkthrough The first step is to describe the goals of performing the walkthrough. Namely, the walkthrough is an opportunity to evaluate the user interface in terms of learnability. This is the first opportunity to address SC 3 - Design defensiveness, by defusing defensiveness on the part of any team members. It is important that the usability professional points out that learnability is only one aspect of usability,

and that the team recognizes that learnability may have been traded off for other aspects of usability. Nonetheless, there is inherent value in knowing when users may encounter problems learning an interface as the issue could be explicitly addressed elsewhere, for example, though marketing or the help system.

A CW session is analytical in nature, and therefore lacks the definitiveness of an empirical usability tests. In light of the CW method's tentative nature, specification owners may resent absolute proclamations that "this is a problem". The usability specialist should, therefore, take care to use softer language, like "this is a potential problem, we need to think about it". Constant reference to the tentative nature of the finding should help defuse defensiveness. The usability professional then points out for the first time that the CW is an evaluation session, not a design session, and goes on to describe the process of walking through the task sequence and answering the two questions for each step (See table 4). The usability professional then gives an example of an action sequence from software not currently under consideration and that has plausible answers to the two questions, and the team is encouraged to produce those answers. Then the usability professional gives another example, one without plausible answers, and the team is prompted to try to provide answers. For each example, the usability professional should model the format that the data is captured in before proceeding with preparing the team for the CW.

Table 4 2 questions from the streamlined CW 1. Will the user know what to do at this step? 2. If the user does the right thing, will they know that they did the right thing, and are making progress towards their goal?

After describing what the team will do during the walkthrough, describe what the team will not do during the walkthrough. This is the first opportunity to directly address SC 2 - Lengthy design discussions, and indirectly address SC 3 - Design defensiveness. In particular, the usability professional explains that if the team finds a step with possible learnability issues, they will note the possible problem and move on to the next step, but they won't redesign the interface. Furthermore, the usability professional should explain that if the team encounters a gap in the design (for example when it is not clear from the specification what action sequence the user is supposed to perform), the team will note the gap and move on, but they won't stop and design the missing actions. Also, if a design idea is suggested, the team may briefly discuss the design idea, note it, and then move on, but the team will not flesh it out. Lastly, if the task analysis appears to be faulty, or only describes one of multiple possible paths towards achieving

355

[Spencer, 2000 from Wharton et al., 1994]

COGNITIVE WALKTHROUGHCHI 2000 • 1-6 APRIL 2000 Papers

1. Define inputs to the walkthrough a. Identification of users b. Sample tasks for evaluation c. Action sequences for completing the tasks

d. Description or implementation of interface 2. Convene the walkthrough

a. Describe the goals of the walkthrough

b. Describe what will be done during the CW c. Describe what will not be done during the

walkthrough

d. Explicitly defuse defensiveness e. Post ground rules in a visible place f. Assign roles g. Appeal for submission to leadership

3. Walkthrough the action sequences for each task a. Tell a credible story for these two questions:

- Will the user know what to do at this step? - If the user does the right thing, will they know

that they did the right thing, and are making progress towards their goal?

b. Maintain control of the CW, enforce the ground rules


a. Possible learnability problems b. Design ideas c. Design gaps d. Problems in the Task Analysis


DETAILED DESCRIPTION OF THE STREAMLINED WALKTHROUGH PROCEDURE 1. Define inputs to walkthrough Before the CW session, the usability professional is responsible for defining the important user task scenario or scenarios and producing a task analysis of those scenarios by explicating the action sequences necessary for accomplishing the tasks in the scenarios. Wharton et al. should be used as a resource for determining how to decide on the scenarios and how to describe the task sequence.

2. Convene the walkthrough The first step is to describe the goals of performing the walkthrough. Namely, the walkthrough is an opportunity to evaluate the user interface in terms of learnability. This is the first opportunity to address SC 3 - Design defensiveness, by defusing defensiveness on the part of any team members. It is important that the usability professional points out that learnability is only one aspect of usability,

and that the team recognizes that learnability may have been traded off for other aspects of usability. Nonetheless, there is inherent value in knowing when users may encounter problems learning an interface as the issue could be explicitly addressed elsewhere, for example, though marketing or the help system.

A CW session is analytical in nature, and therefore lacks the definitiveness of an empirical usability tests. In light of the CW method's tentative nature, specification owners may resent absolute proclamations that "this is a problem". The usability specialist should, therefore, take care to use softer language, like "this is a potential problem, we need to think about it". Constant reference to the tentative nature of the finding should help defuse defensiveness. The usability professional then points out for the first time that the CW is an evaluation session, not a design session, and goes on to describe the process of walking through the task sequence and answering the two questions for each step (See table 4). The usability professional then gives an example of an action sequence from software not currently under consideration and that has plausible answers to the two questions, and the team is encouraged to produce those answers. Then the usability professional gives another example, one without plausible answers, and the team is prompted to try to provide answers. For each example, the usability professional should model the format that the data is captured in before proceeding with preparing the team for the CW.

Table 4 2 questions from the streamlined CW 1. Will the user know what to do at this step? 2. If the user does the right thing, will they know that they did the right thing, and are making progress towards their goal?

After describing what the team will do during the walkthrough, describe what the team will not do during the walkthrough. This is the first opportunity to directly address SC 2 - Lengthy design discussions, and indirectly address SC 3 - Design defensiveness. In particular, the usability professional explains that if the team finds a step with possible learnability issues, they will note the possible problem and move on to the next step, but they won't redesign the interface. Furthermore, the usability professional should explain that if the team encounters a gap in the design (for example when it is not clear from the specification what action sequence the user is supposed to perform), the team will note the gap and move on, but they won't stop and design the missing actions. Also, if a design idea is suggested, the team may briefly discuss the design idea, note it, and then move on, but the team will not flesh it out. Lastly, if the task analysis appears to be faulty, or only describes one of multiple possible paths towards achieving

355

[Spencer, 2000 from Wharton et al., 1994]

1. Define inputs to the walkthrough

2. Convene the walkthrough

3. Walkthrough the action sequences for each task



DESIGN HEURISTICS

DESIGN HEURISTICS• Provide prompts to help designers generate alternatives that vary in

nature, discouraging fixation and encouraging divergent patterns of thinking [Yilmaz, Daly, Seifert, & Gonzalez, 2011; Yilmaz, Seifert, & Gonzalez, 2010]

• Derived from empirical evidence of industrial and engineering designs [Daly et al., 2012; Yilmaz, Christian, Daly, Seifert, & Gonzalez, 2012; Yilmaz & Seifert, 2010]

• Validated through a range of product analysis, case studies, and protocol analyses, in both educational and professional contexts [e.g., Yilmaz & Seifert, 2009; Yilmaz et al., 2011; Yilmaz et al., 2010; Yilmaz et al., 2013; Yilmaz, Daly, Christian, Seifert, & Gonzalez, 2014]

Property Definition Design Heuristics cards

Form Form of the product#32: Expand or collapse #38: Impose hierarchy on functions #55: Repurpose packaging

Function Functions embedded in the product#5: Adjust function through movement #16: Bend #50: Provide sensory feedback

Temporal Use/function of the product over time Relation to sociocultural environment

#13: Apply existing mechanism in new way #21: Change product lifetime #46: Mimic natural mechanisms

Use/User Situated use of the product User interactions with the product

#9: Allow user to customize #10: Allow user to rearrange #40: Incorporate user input

System Context in which the product is used Systems/services the product relies on

#24: Contextualize #28: Create service #29: Create system

HEURISTICS BY CATEGORY

EMPATHIC WALKTHROUGH

5-10 MINUTES

PERSONA TASK

TASK+

WALK THROUGH THE USER STORY

5 MINUTES

LIST & GROUP INFORMATION

FormFunctionUse/UserTemporalSystem

10 MINUTES

TARGETED IDEA GENERATION

Flatten 35

Compress the product to a flat surface by removing connections or deflating it, or using flexible materials or joints. This can improve portability, durability, and

© Design Heuristics, LLC 2012

Create service 28

Develop a service related to the product by defining interaction between the user and a service provider. This can improve user experience, or make the product easier to acquire, install, or use.


EMPATHIC WALKTHROUGH

1 Define inputs to the walkthrough

Identification of users, problem definition, and concept sketch

2 Convene the walkthrough Paired students

3Walkthrough the action sequences for each task

Role-play or “talk through” each other’s concept with a user story

4 Record critical information

Record parts of the design that are confusing or strange, that don’t appear to work correctly, or otherwise seem inappropriate for the user

5 Revise the interface to fix the problems

Generate alternatives or additional concepts that address user concerns

10 MINUTES

Dyad Classification

Concept Initiator

Initial Concept (Target Population)

# Generated Concepts

# Concepts using Design Heuristics

Industrial Design Junior

P1 Herb garden in fridge (millennial) 8 (P1=5; P2=3) 8

P2

Mechanical Engineering Sophomore

P3 Hand-cranked generator (rural India) 11 (P3=6; P4=5) 5

P4Latrine shovel/toilet(nomadic areas of South Africa)

10 (P3=3; P4=7) 2

Mechanical/ Aeronautical Engineering Graduate

P5 Latching galley curtain(flight attendant) 9 (P5=3; P6=6) 2

P6Rubber stopper to reduce galley curtain noise(flight attendant)

6 (P5=2; P6=4) 2

Industrial Design Graduate

P7 Silly Putty as fire starter (20s male survivalist) 9 (P7=4; P8=5) 1

P8 Tetley tea “lilypods”(college-aged girl) 14 (P7=6; P8=8) 2

PERSONA TASK

TASK+





Flatten 35



Create service 28


© Design Heuristics, LLC 2012CASE BUILDING A FLIGHT

ATTENDANT-FRIENDLY AIRLINE CURTAIN

PERSONA TASK

TASK+





Flatten 35



Create service 28



PERSONA TASK

TASK+





Flatten 35



Create service 28



“After entering the airplane, the catering is already done. So—hopefully the curtain is

stored away so I can pass through the area, you know 20 to 50 times as the passengers are starting to boarding. And I have to look

through to my other flight attendants that are working in other areas on the aircraft; we are

talking huge distances like a few yards.”

PERSONA TASK

TASK+





Flatten 35



Create service 28



“After having takeoff, I would really appreciate to have my privacy during working inside the galley, preparing food; the new galley design gives me the opportunity—there are not so

many passengers hanging around where I am working, because it is more separated than

before. Furthermore, the two—the Lycra curtain which were introduced now offers us

the opportunity to reduce the noise incredibly.”

PERSONA TASK

TASK+





Flatten 35



Create service 28



“On the other hand, there are a few like—I don’t like that I always have to pass two

curtains to get to the galley. Just imagine how often I have to pass through—50 times a day

—I guess even more per flight. And especially with the damn cart [steering with hands]—it’s always stuck as a car, and it’s not really fixed,

so even with the magnetic fixation, it’s—it doesn’t offer me the flexibility that I would

really like to have.”

PERSONA TASK

TASK+





Flatten 35



Create service 28



“Well, different part which is bad on the new curtain—it separates you for the passenger area from the area where you are working.

And the plane doesn’t have ECS outcomes…so that means as we have chiller in our area,

our area gets colder and colder, and it’s freezing cold so it’s around 40. And the work

environment is getting worse and worse, even if it’s not as loud. Maybe more secure.”

PERSONA TASK

TASK+





Flatten 35



Create service 28



Form Function Use/User System

Would a transparent curtain help w/ not

knowing what’s happening

Is the curtain staying closed easily?

Did the more open environment help keep the cold air dispersing

to the cabin?

How do they decide when the curtain is

open/closed

Is it loud when opening/ closing?

PERSONA TASK

TASK+





Flatten 35



Create service 28



Is the curtain staying closed

easily?

How do they decide when the curtain is open/

closed

Adjust function through movement 5

Allow the user to adjust stages or degrees of function by moving the product or its parts. Consider applying different types of motion (e.g., rotating, sliding, rolling) and control mechanisms. This can give flexibility to the user through the transitions created between stages.


PERSONA TASK

TASK+


Is the curtain staying closed

easily?

How do they decide when the curtain is open/

closed

Adjust function through movement 5

Allow the user to adjust stages or degrees of function by moving the product or its parts. Consider applying different types of motion (e.g., rotating, sliding, rolling) and control mechanisms. This can give flexibility to the user through the transitions created between stages.





Flatten 35



Create service 28



DISCUSSION• Even without prompting or explicit training, all participants were

able to tell a credible user story

• Students were generally not able to exhaust their ideas in the 10-15 minute period provided

• Design Heuristics were primarily used when the students got “stuck” on a piece of critical information

• Categorization of critical information was generative, promoting conversation about concerns or issues

SO WHAT ABOUT EMPATHY?• Our goal was to foster an empathic connection between the

designer, their solution, and ultimately the end user

• The empathic walkthrough pointed out areas where their knowledge of the user or use context was deficient

• Empathy displayed was most often cognitive knowledge about what the user was feeling, with occasion articulations of the emotional feeling itself

• No instances of participants responding with compassion (i.e., product-centric, not user-centric)

the empathic walkthrough stimulates empathy

strengthening existing human-centered approaches to idea generation and development

THANK YOU

COLINGRAY.MEDESIGNHEURISTICS.COM

THANK YOUThis research is funded by the National Science Foundation,

Division of Undergraduate Education, Transforming Undergraduate Education in Science, Technology, Engineering and Mathematics

(TUES Type II) Grants # 1323251 and #1322552.

idea generation through empathy: reimagining the "cognitive walkthrough"

Engineering

cognitive walkthrough

user interface

cw session

task analysis

design defensiveness

possible learnability

important user task

task sequence