UX Research
·
Uploaded
November 2024
Researching Interface-Driven Sociocultural Empathy for Conversational AI
Understanding Cultural Empathy
User Evaluation via 108 Datapoints
Statistical Data Analysis (SPSS)
UX Research
·
Uploaded
November 2024
Researching Interface-Driven Sociocultural Empathy for Conversational AI
Understanding Cultural Empathy
User Evaluation via 108 Datapoints
Statistical Data Analysis (SPSS)
I designed A/B Interfaces, and used them for UX Research.
I designed A/B Interfaces, and used them for UX Research.
Design Approach A
Design Approach A
Active Listening × Cultural Dimensions
Active Listening × Cultural Dimensions
Design Approach B
Design Approach B
Semiotic Indicator × Emotion Recognition
Semiotic Indicator × Emotion Recognition
The Aim
How can interfaces for conversational AI be more empathetic to users' sociocultural diversity?
about
Using a Research through Design (RtD) approach, I researched how interfaces can solve the problem of sociocultural insensitivity in social robots.
I A/B tested approaches to interface design, and conducted a statistical analysis to generate design guidelines.
Timeline
5 Weeks · Aug to Sep 2024
Client
MSc Human-Computer Interaction
Team
Solo Project
with Credit to
Supervisor · Gisela Reyes Cruz PhD
About
Using a Research through Design (RtD) approach, I researched how interfaces can solve the problem of sociocultural insensitivity in social robots.
I A/B tested approaches to interface design, and conducted a statistical analysis to generate design guidelines.
Timeline
5 Weeks · Aug to Sep 2024
Client
MSc Human-Computer Interaction
Team
Solo Project
with Credit to
Supervisor · Gisela Reyes Cruz PhD
about
Using a Research through Design (RtD) approach, I researched how interfaces can solve the problem of sociocultural insensitivity in social robots.
I A/B tested approaches to interface design, and conducted a statistical analysis to generate design guidelines.
Timeline
5 Weeks · Aug to Sep 2024
Client
MSc Human-Computer Interaction
Team
Solo Project
with Credit to
Supervisor · Gisela Reyes Cruz PhD
About
Using a Research through Design (RtD) approach, I researched how interfaces can solve the problem of sociocultural insensitivity in social robots.
I A/B tested approaches to interface design, and conducted a statistical analysis to generate design guidelines.
Timeline
5 Weeks · Aug to Sep 2024
Client
MSc Human-Computer Interaction
Team
Solo Project
with Credit to
Supervisor · Gisela Reyes Cruz PhD
Researching Solutions
First, I explored existing frameworks of therapy & computational empathy.
Methodology Table
I set up a 3-staged research process. Interfaces were designed in accordance with key theoretical frameworks in socioculturalism, computational empathy and applicational methods. Then, the interfaces were tested using scenarios, before an analysis was ran.
Approach A - Semiotic Interpretive Flow (SIF)
Design Approach A tests Picard’s Information Flow Diagram (1997) through the application of Barthes’ Semiotic Significance Diagram (1981), all while conforming to Hofstede’s Model of Cultural Dimensions (2010).
Approach B - Empathy Classification Monitor (ECM)
Design Approach B tests Bordoli & Biswas’ Sentiment Classification Techniques (2023) on Murray & Arnott’s Effect of Emotion on Speech Patterns (1993), giving way for the contextual implementation of Greimas’ Semiotic Square (1966).
Methodology Table
I set up a 3-staged research process. Interfaces were designed in accordance with key theoretical frameworks in socioculturalism, computational empathy and applicational methods. Then, the interfaces were tested using scenarios, before an analysis was ran.
Approach A - Semiotic Interpretive Flow (SIF)
Design Approach A tests Picard’s Information Flow Diagram (1997) through the application of Barthes’ Semiotic Significance Diagram (1981), all while conforming to Hofstede’s Model of Cultural Dimensions (2010).
Approach B - Empathy Classification Monitor (ECM)
Design Approach B tests Bordoli & Biswas’ Sentiment Classification Techniques (2023) on Murray & Arnott’s Effect of Emotion on Speech Patterns (1993), giving way for the contextual implementation of Greimas’ Semiotic Square (1966).
Methodology Table
I set up a 3-staged research process. Interfaces were designed in accordance with key theoretical frameworks in socioculturalism, computational empathy and applicational methods. Then, the interfaces were tested using scenarios, before an analysis was ran.
Approach A - Semiotic Interpretive Flow (SIF)
Design Approach A tests Picard’s Information Flow Diagram (1997) through the application of Barthes’ Semiotic Significance Diagram (1981), all while conforming to Hofstede’s Model of Cultural Dimensions (2010).
Approach B - Empathy Classification Monitor (ECM)
Design Approach B tests Bordoli & Biswas’ Sentiment Classification Techniques (2023) on Murray & Arnott’s Effect of Emotion on Speech Patterns (1993), giving way for the contextual implementation of Greimas’ Semiotic Square (1966).
Methodology Table
I set up a 3-staged research process. Interfaces were designed in accordance with key theoretical frameworks in socioculturalism, computational empathy and applicational methods. Then, the interfaces were tested using scenarios, before an analysis was ran.
Approach A - Semiotic Interpretive Flow (SIF)
Design Approach A tests Picard’s Information Flow Diagram (1997) through the application of Barthes’ Semiotic Significance Diagram (1981), all while conforming to Hofstede’s Model of Cultural Dimensions (2010).
Approach B - Empathy Classification Monitor (ECM)
Design Approach B tests Bordoli & Biswas’ Sentiment Classification Techniques (2023) on Murray & Arnott’s Effect of Emotion on Speech Patterns (1993), giving way for the contextual implementation of Greimas’ Semiotic Square (1966).
Methodology Table
I set up a 3-staged research process. Interfaces were designed in accordance with key theoretical frameworks in socioculturalism, computational empathy and applicational methods. Then, the interfaces were tested using scenarios, before an analysis was ran.
Approach A - Semiotic Interpretive Flow (SIF)
Design Approach A tests Picard’s Information Flow Diagram (1997) through the application of Barthes’ Semiotic Significance Diagram (1981), all while conforming to Hofstede’s Model of Cultural Dimensions (2010).
Approach B - Empathy Classification Monitor (ECM)
Design Approach B tests Bordoli & Biswas’ Sentiment Classification Techniques (2023) on Murray & Arnott’s Effect of Emotion on Speech Patterns (1993), giving way for the contextual implementation of Greimas’ Semiotic Square (1966).
Methodology Table
I set up a 3-staged research process. Interfaces were designed in accordance with key theoretical frameworks in socioculturalism, computational empathy and applicational methods. Then, the interfaces were tested using scenarios, before an analysis was ran.
Approach A - Semiotic Interpretive Flow (SIF)
Design Approach A tests Picard’s Information Flow Diagram (1997) through the application of Barthes’ Semiotic Significance Diagram (1981), all while conforming to Hofstede’s Model of Cultural Dimensions (2010).
Approach B - Empathy Classification Monitor (ECM)
Design Approach B tests Bordoli & Biswas’ Sentiment Classification Techniques (2023) on Murray & Arnott’s Effect of Emotion on Speech Patterns (1993), giving way for the contextual implementation of Greimas’ Semiotic Square (1966).
Methodology Table
I set up a 3-staged research process. Interfaces were designed in accordance with key theoretical frameworks in socioculturalism, computational empathy and applicational methods. Then, the interfaces were tested using scenarios, before an analysis was ran.
Approach A - Semiotic Interpretive Flow (SIF)
Design Approach A tests Picard’s Information Flow Diagram (1997) through the application of Barthes’ Semiotic Significance Diagram (1981), all while conforming to Hofstede’s Model of Cultural Dimensions (2010).
Approach B - Empathy Classification Monitor (ECM)
Design Approach B tests Bordoli & Biswas’ Sentiment Classification Techniques (2023) on Murray & Arnott’s Effect of Emotion on Speech Patterns (1993), giving way for the contextual implementation of Greimas’ Semiotic Square (1966).
Methodology Table
I set up a 3-staged research process. Interfaces were designed in accordance with key theoretical frameworks in socioculturalism, computational empathy and applicational methods. Then, the interfaces were tested using scenarios, before an analysis was ran.
Approach A - Semiotic Interpretive Flow (SIF)
Design Approach A tests Picard’s Information Flow Diagram (1997) through the application of Barthes’ Semiotic Significance Diagram (1981), all while conforming to Hofstede’s Model of Cultural Dimensions (2010).
Approach B - Empathy Classification Monitor (ECM)
Design Approach B tests Bordoli & Biswas’ Sentiment Classification Techniques (2023) on Murray & Arnott’s Effect of Emotion on Speech Patterns (1993), giving way for the contextual implementation of Greimas’ Semiotic Square (1966).
Methodology Table
I set up a 3-staged research process. Interfaces were designed in accordance with key theoretical frameworks in socioculturalism, computational empathy and applicational methods. Then, the interfaces were tested using scenarios, before an analysis was ran.
Approach A - Semiotic Interpretive Flow (SIF)
Design Approach A tests Picard’s Information Flow Diagram (1997) through the application of Barthes’ Semiotic Significance Diagram (1981), all while conforming to Hofstede’s Model of Cultural Dimensions (2010).
Approach B - Empathy Classification Monitor (ECM)
Design Approach B tests Bordoli & Biswas’ Sentiment Classification Techniques (2023) on Murray & Arnott’s Effect of Emotion on Speech Patterns (1993), giving way for the contextual implementation of Greimas’ Semiotic Square (1966).
Methodology Table
I set up a 3-staged research process. Interfaces were designed in accordance with key theoretical frameworks in socioculturalism, computational empathy and applicational methods. Then, the interfaces were tested using scenarios, before an analysis was ran.
Approach A - Semiotic Interpretive Flow (SIF)
Design Approach A tests Picard’s Information Flow Diagram (1997) through the application of Barthes’ Semiotic Significance Diagram (1981), all while conforming to Hofstede’s Model of Cultural Dimensions (2010).
Approach B - Empathy Classification Monitor (ECM)
Design Approach B tests Bordoli & Biswas’ Sentiment Classification Techniques (2023) on Murray & Arnott’s Effect of Emotion on Speech Patterns (1993), giving way for the contextual implementation of Greimas’ Semiotic Square (1966).
Methodology Table
I set up a 3-staged research process. Interfaces were designed in accordance with key theoretical frameworks in socioculturalism, computational empathy and applicational methods. Then, the interfaces were tested using scenarios, before an analysis was ran.
Approach A - Semiotic Interpretive Flow (SIF)
Design Approach A tests Picard’s Information Flow Diagram (1997) through the application of Barthes’ Semiotic Significance Diagram (1981), all while conforming to Hofstede’s Model of Cultural Dimensions (2010).
Approach B - Empathy Classification Monitor (ECM)
Design Approach B tests Bordoli & Biswas’ Sentiment Classification Techniques (2023) on Murray & Arnott’s Effect of Emotion on Speech Patterns (1993), giving way for the contextual implementation of Greimas’ Semiotic Square (1966).
Methodology Table
I set up a 3-staged research process. Interfaces were designed in accordance with key theoretical frameworks in socioculturalism, computational empathy and applicational methods. Then, the interfaces were tested using scenarios, before an analysis was ran.
Approach A - Semiotic Interpretive Flow (SIF)
Design Approach A tests Picard’s Information Flow Diagram (1997) through the application of Barthes’ Semiotic Significance Diagram (1981), all while conforming to Hofstede’s Model of Cultural Dimensions (2010).
Approach B - Empathy Classification Monitor (ECM)
Design Approach B tests Bordoli & Biswas’ Sentiment Classification Techniques (2023) on Murray & Arnott’s Effect of Emotion on Speech Patterns (1993), giving way for the contextual implementation of Greimas’ Semiotic Square (1966).
Methodology Table
I set up a 3-staged research process. Interfaces were designed in accordance with key theoretical frameworks in socioculturalism, computational empathy and applicational methods. Then, the interfaces were tested using scenarios, before an analysis was ran.
Approach A - Semiotic Interpretive Flow (SIF)
Design Approach A tests Picard’s Information Flow Diagram (1997) through the application of Barthes’ Semiotic Significance Diagram (1981), all while conforming to Hofstede’s Model of Cultural Dimensions (2010).
Approach B - Empathy Classification Monitor (ECM)
Design Approach B tests Bordoli & Biswas’ Sentiment Classification Techniques (2023) on Murray & Arnott’s Effect of Emotion on Speech Patterns (1993), giving way for the contextual implementation of Greimas’ Semiotic Square (1966).
Methodology Table
I set up a 3-staged research process. Interfaces were designed in accordance with key theoretical frameworks in socioculturalism, computational empathy and applicational methods. Then, the interfaces were tested using scenarios, before an analysis was ran.
Approach A - Semiotic Interpretive Flow (SIF)
Design Approach A tests Picard’s Information Flow Diagram (1997) through the application of Barthes’ Semiotic Significance Diagram (1981), all while conforming to Hofstede’s Model of Cultural Dimensions (2010).
Approach B - Empathy Classification Monitor (ECM)
Design Approach B tests Bordoli & Biswas’ Sentiment Classification Techniques (2023) on Murray & Arnott’s Effect of Emotion on Speech Patterns (1993), giving way for the contextual implementation of Greimas’ Semiotic Square (1966).
Methodology Table
I set up a 3-staged research process. Interfaces were designed in accordance with key theoretical frameworks in socioculturalism, computational empathy and applicational methods. Then, the interfaces were tested using scenarios, before an analysis was ran.
Approach A - Semiotic Interpretive Flow (SIF)
Design Approach A tests Picard’s Information Flow Diagram (1997) through the application of Barthes’ Semiotic Significance Diagram (1981), all while conforming to Hofstede’s Model of Cultural Dimensions (2010).
Approach B - Empathy Classification Monitor (ECM)
Design Approach B tests Bordoli & Biswas’ Sentiment Classification Techniques (2023) on Murray & Arnott’s Effect of Emotion on Speech Patterns (1993), giving way for the contextual implementation of Greimas’ Semiotic Square (1966).
Methodology Table
I set up a 3-staged research process. Interfaces were designed in accordance with key theoretical frameworks in socioculturalism, computational empathy and applicational methods. Then, the interfaces were tested using scenarios, before an analysis was ran.
Approach A - Semiotic Interpretive Flow (SIF)
Design Approach A tests Picard’s Information Flow Diagram (1997) through the application of Barthes’ Semiotic Significance Diagram (1981), all while conforming to Hofstede’s Model of Cultural Dimensions (2010).
Approach B - Empathy Classification Monitor (ECM)
Design Approach B tests Bordoli & Biswas’ Sentiment Classification Techniques (2023) on Murray & Arnott’s Effect of Emotion on Speech Patterns (1993), giving way for the contextual implementation of Greimas’ Semiotic Square (1966).
Constructing A/B Designs
I merged a product development cycle with Research through Design (RtD) to synthesise the designs.
Approach A - Wireframe
Using the combined theoretical approaches, I created lo-fi wireframes for each approach. The wireframe for Approach A (SIF) emulates a phenomenon in which high and low levels of consciousness interacted with one another, aswell as with the conversational partner.
Approach B - Wireframe
The wireframe for Approach B (ECM) classifies the system's response based on four schemes of meaning, based on the Semiotic Square theory, but responds in accordance with the effect of emotion on speech patterns, offering a more linguistic form of communication.
Scenario 1 - Individual Level
As the interfaces are built purely as testing materials, they must also be flexibile solely to the structure of the user examination, in which scenarios will play out in order to isolate each approach's success criteria as prescribed in the methodology table. Firstly, the individual level.
Scenario 2 - Relational Level
Secondly, the relational level. Labelled is how each interaction point along the scenario is designed to provide a place for the assessment criteria to be naturally achieved.
Scenario 3 - Universal Level
Thirdly, the universal level - the final and widest level of the Tripartite of Identity framework.
Wireframe Validation
With the wireframes and their scenarios drafted, I played out the user stories in order to identify discreprencies, which were addressed in the most suitable way before the next stage began.
Approach A - Wireframe
Using the combined theoretical approaches, I created lo-fi wireframes for each approach. The wireframe for Approach A (SIF) emulates a phenomenon in which high and low levels of consciousness interacted with one another, aswell as with the conversational partner.
Approach B - Wireframe
The wireframe for Approach B (ECM) classifies the system's response based on four schemes of meaning, based on the Semiotic Square theory, but responds in accordance with the effect of emotion on speech patterns, offering a more linguistic form of communication.
Scenario 1 - Individual Level
As the interfaces are built purely as testing materials, they must also be flexibile solely to the structure of the user examination, in which scenarios will play out in order to isolate each approach's success criteria as prescribed in the methodology table. Firstly, the individual level.
Scenario 2 - Relational Level
Secondly, the relational level. Labelled is how each interaction point along the scenario is designed to provide a place for the assessment criteria to be naturally achieved.
Scenario 3 - Universal Level
Thirdly, the universal level - the final and widest level of the Tripartite of Identity framework.
Wireframe Validation
With the wireframes and their scenarios drafted, I played out the user stories in order to identify discreprencies, which were addressed in the most suitable way before the next stage began.
Approach A - Wireframe
Using the combined theoretical approaches, I created lo-fi wireframes for each approach. The wireframe for Approach A (SIF) emulates a phenomenon in which high and low levels of consciousness interacted with one another, aswell as with the conversational partner.
Approach B - Wireframe
The wireframe for Approach B (ECM) classifies the system's response based on four schemes of meaning, based on the Semiotic Square theory, but responds in accordance with the effect of emotion on speech patterns, offering a more linguistic form of communication.
Scenario 1 - Individual Level
As the interfaces are built purely as testing materials, they must also be flexibile solely to the structure of the user examination, in which scenarios will play out in order to isolate each approach's success criteria as prescribed in the methodology table. Firstly, the individual level.
Scenario 2 - Relational Level
Secondly, the relational level. Labelled is how each interaction point along the scenario is designed to provide a place for the assessment criteria to be naturally achieved.
Scenario 3 - Universal Level
Thirdly, the universal level - the final and widest level of the Tripartite of Identity framework.
Wireframe Validation
With the wireframes and their scenarios drafted, I played out the user stories in order to identify discreprencies, which were addressed in the most suitable way before the next stage began.
Approach A - Wireframe
Using the combined theoretical approaches, I created lo-fi wireframes for each approach. The wireframe for Approach A (SIF) emulates a phenomenon in which high and low levels of consciousness interacted with one another, aswell as with the conversational partner.
Approach B - Wireframe
The wireframe for Approach B (ECM) classifies the system's response based on four schemes of meaning, based on the Semiotic Square theory, but responds in accordance with the effect of emotion on speech patterns, offering a more linguistic form of communication.
Scenario 1 - Individual Level
As the interfaces are built purely as testing materials, they must also be flexibile solely to the structure of the user examination, in which scenarios will play out in order to isolate each approach's success criteria as prescribed in the methodology table. Firstly, the individual level.
Scenario 2 - Relational Level
Secondly, the relational level. Labelled is how each interaction point along the scenario is designed to provide a place for the assessment criteria to be naturally achieved.
Scenario 3 - Universal Level
Thirdly, the universal level - the final and widest level of the Tripartite of Identity framework.
Wireframe Validation
With the wireframes and their scenarios drafted, I played out the user stories in order to identify discreprencies, which were addressed in the most suitable way before the next stage began.
Approach A - Wireframe
Using the combined theoretical approaches, I created lo-fi wireframes for each approach. The wireframe for Approach A (SIF) emulates a phenomenon in which high and low levels of consciousness interacted with one another, aswell as with the conversational partner.
Approach B - Wireframe
The wireframe for Approach B (ECM) classifies the system's response based on four schemes of meaning, based on the Semiotic Square theory, but responds in accordance with the effect of emotion on speech patterns, offering a more linguistic form of communication.
Scenario 1 - Individual Level
As the interfaces are built purely as testing materials, they must also be flexibile solely to the structure of the user examination, in which scenarios will play out in order to isolate each approach's success criteria as prescribed in the methodology table. Firstly, the individual level.
Scenario 2 - Relational Level
Secondly, the relational level. Labelled is how each interaction point along the scenario is designed to provide a place for the assessment criteria to be naturally achieved.
Scenario 3 - Universal Level
Thirdly, the universal level - the final and widest level of the Tripartite of Identity framework.
Wireframe Validation
With the wireframes and their scenarios drafted, I played out the user stories in order to identify discreprencies, which were addressed in the most suitable way before the next stage began.
Approach A - Wireframe
Using the combined theoretical approaches, I created lo-fi wireframes for each approach. The wireframe for Approach A (SIF) emulates a phenomenon in which high and low levels of consciousness interacted with one another, aswell as with the conversational partner.
Approach B - Wireframe
The wireframe for Approach B (ECM) classifies the system's response based on four schemes of meaning, based on the Semiotic Square theory, but responds in accordance with the effect of emotion on speech patterns, offering a more linguistic form of communication.
Scenario 1 - Individual Level
As the interfaces are built purely as testing materials, they must also be flexibile solely to the structure of the user examination, in which scenarios will play out in order to isolate each approach's success criteria as prescribed in the methodology table. Firstly, the individual level.
Scenario 2 - Relational Level
Secondly, the relational level. Labelled is how each interaction point along the scenario is designed to provide a place for the assessment criteria to be naturally achieved.
Scenario 3 - Universal Level
Thirdly, the universal level - the final and widest level of the Tripartite of Identity framework.
Wireframe Validation
With the wireframes and their scenarios drafted, I played out the user stories in order to identify discreprencies, which were addressed in the most suitable way before the next stage began.
Approach A - Wireframe
Using the combined theoretical approaches, I created lo-fi wireframes for each approach. The wireframe for Approach A (SIF) emulates a phenomenon in which high and low levels of consciousness interacted with one another, aswell as with the conversational partner.
Approach B - Wireframe
The wireframe for Approach B (ECM) classifies the system's response based on four schemes of meaning, based on the Semiotic Square theory, but responds in accordance with the effect of emotion on speech patterns, offering a more linguistic form of communication.
Scenario 1 - Individual Level
As the interfaces are built purely as testing materials, they must also be flexibile solely to the structure of the user examination, in which scenarios will play out in order to isolate each approach's success criteria as prescribed in the methodology table. Firstly, the individual level.
Scenario 2 - Relational Level
Secondly, the relational level. Labelled is how each interaction point along the scenario is designed to provide a place for the assessment criteria to be naturally achieved.
Scenario 3 - Universal Level
Thirdly, the universal level - the final and widest level of the Tripartite of Identity framework.
Wireframe Validation
With the wireframes and their scenarios drafted, I played out the user stories in order to identify discreprencies, which were addressed in the most suitable way before the next stage began.
Approach A - Wireframe
Using the combined theoretical approaches, I created lo-fi wireframes for each approach. The wireframe for Approach A (SIF) emulates a phenomenon in which high and low levels of consciousness interacted with one another, aswell as with the conversational partner.
Approach B - Wireframe
The wireframe for Approach B (ECM) classifies the system's response based on four schemes of meaning, based on the Semiotic Square theory, but responds in accordance with the effect of emotion on speech patterns, offering a more linguistic form of communication.
Scenario 1 - Individual Level
As the interfaces are built purely as testing materials, they must also be flexibile solely to the structure of the user examination, in which scenarios will play out in order to isolate each approach's success criteria as prescribed in the methodology table. Firstly, the individual level.
Scenario 2 - Relational Level
Secondly, the relational level. Labelled is how each interaction point along the scenario is designed to provide a place for the assessment criteria to be naturally achieved.
Scenario 3 - Universal Level
Thirdly, the universal level - the final and widest level of the Tripartite of Identity framework.
Wireframe Validation
With the wireframes and their scenarios drafted, I played out the user stories in order to identify discreprencies, which were addressed in the most suitable way before the next stage began.
Approach A - Wireframe
Using the combined theoretical approaches, I created lo-fi wireframes for each approach. The wireframe for Approach A (SIF) emulates a phenomenon in which high and low levels of consciousness interacted with one another, aswell as with the conversational partner.
Approach B - Wireframe
The wireframe for Approach B (ECM) classifies the system's response based on four schemes of meaning, based on the Semiotic Square theory, but responds in accordance with the effect of emotion on speech patterns, offering a more linguistic form of communication.
Scenario 1 - Individual Level
As the interfaces are built purely as testing materials, they must also be flexibile solely to the structure of the user examination, in which scenarios will play out in order to isolate each approach's success criteria as prescribed in the methodology table. Firstly, the individual level.
Scenario 2 - Relational Level
Secondly, the relational level. Labelled is how each interaction point along the scenario is designed to provide a place for the assessment criteria to be naturally achieved.
Scenario 3 - Universal Level
Thirdly, the universal level - the final and widest level of the Tripartite of Identity framework.
Wireframe Validation
With the wireframes and their scenarios drafted, I played out the user stories in order to identify discreprencies, which were addressed in the most suitable way before the next stage began.
Approach A - Wireframe
Using the combined theoretical approaches, I created lo-fi wireframes for each approach. The wireframe for Approach A (SIF) emulates a phenomenon in which high and low levels of consciousness interacted with one another, aswell as with the conversational partner.
Approach B - Wireframe
The wireframe for Approach B (ECM) classifies the system's response based on four schemes of meaning, based on the Semiotic Square theory, but responds in accordance with the effect of emotion on speech patterns, offering a more linguistic form of communication.
Scenario 1 - Individual Level
As the interfaces are built purely as testing materials, they must also be flexibile solely to the structure of the user examination, in which scenarios will play out in order to isolate each approach's success criteria as prescribed in the methodology table. Firstly, the individual level.
Scenario 2 - Relational Level
Secondly, the relational level. Labelled is how each interaction point along the scenario is designed to provide a place for the assessment criteria to be naturally achieved.
Scenario 3 - Universal Level
Thirdly, the universal level - the final and widest level of the Tripartite of Identity framework.
Wireframe Validation
With the wireframes and their scenarios drafted, I played out the user stories in order to identify discreprencies, which were addressed in the most suitable way before the next stage began.
Approach A - Wireframe
Using the combined theoretical approaches, I created lo-fi wireframes for each approach. The wireframe for Approach A (SIF) emulates a phenomenon in which high and low levels of consciousness interacted with one another, aswell as with the conversational partner.
Approach B - Wireframe
The wireframe for Approach B (ECM) classifies the system's response based on four schemes of meaning, based on the Semiotic Square theory, but responds in accordance with the effect of emotion on speech patterns, offering a more linguistic form of communication.
Scenario 1 - Individual Level
As the interfaces are built purely as testing materials, they must also be flexibile solely to the structure of the user examination, in which scenarios will play out in order to isolate each approach's success criteria as prescribed in the methodology table. Firstly, the individual level.
Scenario 2 - Relational Level
Secondly, the relational level. Labelled is how each interaction point along the scenario is designed to provide a place for the assessment criteria to be naturally achieved.
Scenario 3 - Universal Level
Thirdly, the universal level - the final and widest level of the Tripartite of Identity framework.
Wireframe Validation
With the wireframes and their scenarios drafted, I played out the user stories in order to identify discreprencies, which were addressed in the most suitable way before the next stage began.
Approach A - Wireframe
Using the combined theoretical approaches, I created lo-fi wireframes for each approach. The wireframe for Approach A (SIF) emulates a phenomenon in which high and low levels of consciousness interacted with one another, aswell as with the conversational partner.
Approach B - Wireframe
The wireframe for Approach B (ECM) classifies the system's response based on four schemes of meaning, based on the Semiotic Square theory, but responds in accordance with the effect of emotion on speech patterns, offering a more linguistic form of communication.
Scenario 1 - Individual Level
As the interfaces are built purely as testing materials, they must also be flexibile solely to the structure of the user examination, in which scenarios will play out in order to isolate each approach's success criteria as prescribed in the methodology table. Firstly, the individual level.
Scenario 2 - Relational Level
Secondly, the relational level. Labelled is how each interaction point along the scenario is designed to provide a place for the assessment criteria to be naturally achieved.
Scenario 3 - Universal Level
Thirdly, the universal level - the final and widest level of the Tripartite of Identity framework.
Wireframe Validation
With the wireframes and their scenarios drafted, I played out the user stories in order to identify discreprencies, which were addressed in the most suitable way before the next stage began.
Approach A - Wireframe
Using the combined theoretical approaches, I created lo-fi wireframes for each approach. The wireframe for Approach A (SIF) emulates a phenomenon in which high and low levels of consciousness interacted with one another, aswell as with the conversational partner.
Approach B - Wireframe
The wireframe for Approach B (ECM) classifies the system's response based on four schemes of meaning, based on the Semiotic Square theory, but responds in accordance with the effect of emotion on speech patterns, offering a more linguistic form of communication.
Scenario 1 - Individual Level
As the interfaces are built purely as testing materials, they must also be flexibile solely to the structure of the user examination, in which scenarios will play out in order to isolate each approach's success criteria as prescribed in the methodology table. Firstly, the individual level.
Scenario 2 - Relational Level
Secondly, the relational level. Labelled is how each interaction point along the scenario is designed to provide a place for the assessment criteria to be naturally achieved.
Scenario 3 - Universal Level
Thirdly, the universal level - the final and widest level of the Tripartite of Identity framework.
Wireframe Validation
With the wireframes and their scenarios drafted, I played out the user stories in order to identify discreprencies, which were addressed in the most suitable way before the next stage began.
Approach A - Wireframe
Using the combined theoretical approaches, I created lo-fi wireframes for each approach. The wireframe for Approach A (SIF) emulates a phenomenon in which high and low levels of consciousness interacted with one another, aswell as with the conversational partner.
Approach B - Wireframe
The wireframe for Approach B (ECM) classifies the system's response based on four schemes of meaning, based on the Semiotic Square theory, but responds in accordance with the effect of emotion on speech patterns, offering a more linguistic form of communication.
Scenario 1 - Individual Level
As the interfaces are built purely as testing materials, they must also be flexibile solely to the structure of the user examination, in which scenarios will play out in order to isolate each approach's success criteria as prescribed in the methodology table. Firstly, the individual level.
Scenario 2 - Relational Level
Secondly, the relational level. Labelled is how each interaction point along the scenario is designed to provide a place for the assessment criteria to be naturally achieved.
Scenario 3 - Universal Level
Thirdly, the universal level - the final and widest level of the Tripartite of Identity framework.
Wireframe Validation
With the wireframes and their scenarios drafted, I played out the user stories in order to identify discreprencies, which were addressed in the most suitable way before the next stage began.
Approach A - Wireframe
Using the combined theoretical approaches, I created lo-fi wireframes for each approach. The wireframe for Approach A (SIF) emulates a phenomenon in which high and low levels of consciousness interacted with one another, aswell as with the conversational partner.
Approach B - Wireframe
The wireframe for Approach B (ECM) classifies the system's response based on four schemes of meaning, based on the Semiotic Square theory, but responds in accordance with the effect of emotion on speech patterns, offering a more linguistic form of communication.
Scenario 1 - Individual Level
As the interfaces are built purely as testing materials, they must also be flexibile solely to the structure of the user examination, in which scenarios will play out in order to isolate each approach's success criteria as prescribed in the methodology table. Firstly, the individual level.
Scenario 2 - Relational Level
Secondly, the relational level. Labelled is how each interaction point along the scenario is designed to provide a place for the assessment criteria to be naturally achieved.
Scenario 3 - Universal Level
Thirdly, the universal level - the final and widest level of the Tripartite of Identity framework.
Wireframe Validation
With the wireframes and their scenarios drafted, I played out the user stories in order to identify discreprencies, which were addressed in the most suitable way before the next stage began.
Approach A - Wireframe
Using the combined theoretical approaches, I created lo-fi wireframes for each approach. The wireframe for Approach A (SIF) emulates a phenomenon in which high and low levels of consciousness interacted with one another, aswell as with the conversational partner.
Approach B - Wireframe
The wireframe for Approach B (ECM) classifies the system's response based on four schemes of meaning, based on the Semiotic Square theory, but responds in accordance with the effect of emotion on speech patterns, offering a more linguistic form of communication.
Scenario 1 - Individual Level
As the interfaces are built purely as testing materials, they must also be flexibile solely to the structure of the user examination, in which scenarios will play out in order to isolate each approach's success criteria as prescribed in the methodology table. Firstly, the individual level.
Scenario 2 - Relational Level
Secondly, the relational level. Labelled is how each interaction point along the scenario is designed to provide a place for the assessment criteria to be naturally achieved.
Scenario 3 - Universal Level
Thirdly, the universal level - the final and widest level of the Tripartite of Identity framework.
Wireframe Validation
With the wireframes and their scenarios drafted, I played out the user stories in order to identify discreprencies, which were addressed in the most suitable way before the next stage began.
User Testing & Data Analysis
I conducted the user research with a diverse participant population, then cleaned & analysed the data on SPSS.
The Setting
[faces anonymised due to University data protection guidlines] The therapy-esque scenarios were carried out at a place of my participants' choosing, based on logistics and availability. I recorded the likert ratings, aswell as any verbal comments during the testing.
Participants & Data
There were 6 participants in total, all of which I was able to record the 18x receptivity ratings. To allow sociocultural insights to be made from the analysis stage, I ensured to record the participant's age, gender and nationality.
Data Preparation & Process
At first, parametric data was assumed, and I prepared the data in accordance with this, ensuring homogeneity and a normal distribution. When the data proved resistant to the subsequent transformations, I made the decision to switch to non-parametric tests.
The Setting
[faces anonymised due to University data protection guidlines] The therapy-esque scenarios were carried out at a place of my participants' choosing, based on logistics and availability. I recorded the likert ratings, aswell as any verbal comments during the testing.
Participants & Data
There were 6 participants in total, all of which I was able to record the 18x receptivity ratings. To allow sociocultural insights to be made from the analysis stage, I ensured to record the participant's age, gender and nationality.
Data Preparation & Process
At first, parametric data was assumed, and I prepared the data in accordance with this, ensuring homogeneity and a normal distribution. When the data proved resistant to the subsequent transformations, I made the decision to switch to non-parametric tests.
The Setting
[faces anonymised due to University data protection guidlines] The therapy-esque scenarios were carried out at a place of my participants' choosing, based on logistics and availability. I recorded the likert ratings, aswell as any verbal comments during the testing.
Participants & Data
There were 6 participants in total, all of which I was able to record the 18x receptivity ratings. To allow sociocultural insights to be made from the analysis stage, I ensured to record the participant's age, gender and nationality.
Data Preparation & Process
At first, parametric data was assumed, and I prepared the data in accordance with this, ensuring homogeneity and a normal distribution. When the data proved resistant to the subsequent transformations, I made the decision to switch to non-parametric tests.
The Setting
[faces anonymised due to University data protection guidlines] The therapy-esque scenarios were carried out at a place of my participants' choosing, based on logistics and availability. I recorded the likert ratings, aswell as any verbal comments during the testing.
Participants & Data
There were 6 participants in total, all of which I was able to record the 18x receptivity ratings. To allow sociocultural insights to be made from the analysis stage, I ensured to record the participant's age, gender and nationality.
Data Preparation & Process
At first, parametric data was assumed, and I prepared the data in accordance with this, ensuring homogeneity and a normal distribution. When the data proved resistant to the subsequent transformations, I made the decision to switch to non-parametric tests.
The Setting
[faces anonymised due to University data protection guidlines] The therapy-esque scenarios were carried out at a place of my participants' choosing, based on logistics and availability. I recorded the likert ratings, aswell as any verbal comments during the testing.
Participants & Data
There were 6 participants in total, all of which I was able to record the 18x receptivity ratings. To allow sociocultural insights to be made from the analysis stage, I ensured to record the participant's age, gender and nationality.
Data Preparation & Process
At first, parametric data was assumed, and I prepared the data in accordance with this, ensuring homogeneity and a normal distribution. When the data proved resistant to the subsequent transformations, I made the decision to switch to non-parametric tests.
The Setting
[faces anonymised due to University data protection guidlines] The therapy-esque scenarios were carried out at a place of my participants' choosing, based on logistics and availability. I recorded the likert ratings, aswell as any verbal comments during the testing.
Participants & Data
There were 6 participants in total, all of which I was able to record the 18x receptivity ratings. To allow sociocultural insights to be made from the analysis stage, I ensured to record the participant's age, gender and nationality.
Data Preparation & Process
At first, parametric data was assumed, and I prepared the data in accordance with this, ensuring homogeneity and a normal distribution. When the data proved resistant to the subsequent transformations, I made the decision to switch to non-parametric tests.
The Setting
[faces anonymised due to University data protection guidlines] The therapy-esque scenarios were carried out at a place of my participants' choosing, based on logistics and availability. I recorded the likert ratings, aswell as any verbal comments during the testing.
Participants & Data
There were 6 participants in total, all of which I was able to record the 18x receptivity ratings. To allow sociocultural insights to be made from the analysis stage, I ensured to record the participant's age, gender and nationality.
Data Preparation & Process
At first, parametric data was assumed, and I prepared the data in accordance with this, ensuring homogeneity and a normal distribution. When the data proved resistant to the subsequent transformations, I made the decision to switch to non-parametric tests.
The Setting
[faces anonymised due to University data protection guidlines] The therapy-esque scenarios were carried out at a place of my participants' choosing, based on logistics and availability. I recorded the likert ratings, aswell as any verbal comments during the testing.
Participants & Data
There were 6 participants in total, all of which I was able to record the 18x receptivity ratings. To allow sociocultural insights to be made from the analysis stage, I ensured to record the participant's age, gender and nationality.
Data Preparation & Process
At first, parametric data was assumed, and I prepared the data in accordance with this, ensuring homogeneity and a normal distribution. When the data proved resistant to the subsequent transformations, I made the decision to switch to non-parametric tests.
The Setting
[faces anonymised due to University data protection guidlines] The therapy-esque scenarios were carried out at a place of my participants' choosing, based on logistics and availability. I recorded the likert ratings, aswell as any verbal comments during the testing.
Participants & Data
There were 6 participants in total, all of which I was able to record the 18x receptivity ratings. To allow sociocultural insights to be made from the analysis stage, I ensured to record the participant's age, gender and nationality.
Data Preparation & Process
At first, parametric data was assumed, and I prepared the data in accordance with this, ensuring homogeneity and a normal distribution. When the data proved resistant to the subsequent transformations, I made the decision to switch to non-parametric tests.
The Setting
[faces anonymised due to University data protection guidlines] The therapy-esque scenarios were carried out at a place of my participants' choosing, based on logistics and availability. I recorded the likert ratings, aswell as any verbal comments during the testing.
Participants & Data
There were 6 participants in total, all of which I was able to record the 18x receptivity ratings. To allow sociocultural insights to be made from the analysis stage, I ensured to record the participant's age, gender and nationality.
Data Preparation & Process
At first, parametric data was assumed, and I prepared the data in accordance with this, ensuring homogeneity and a normal distribution. When the data proved resistant to the subsequent transformations, I made the decision to switch to non-parametric tests.
The Setting
[faces anonymised due to University data protection guidlines] The therapy-esque scenarios were carried out at a place of my participants' choosing, based on logistics and availability. I recorded the likert ratings, aswell as any verbal comments during the testing.
Participants & Data
There were 6 participants in total, all of which I was able to record the 18x receptivity ratings. To allow sociocultural insights to be made from the analysis stage, I ensured to record the participant's age, gender and nationality.
Data Preparation & Process
At first, parametric data was assumed, and I prepared the data in accordance with this, ensuring homogeneity and a normal distribution. When the data proved resistant to the subsequent transformations, I made the decision to switch to non-parametric tests.
The Setting
[faces anonymised due to University data protection guidlines] The therapy-esque scenarios were carried out at a place of my participants' choosing, based on logistics and availability. I recorded the likert ratings, aswell as any verbal comments during the testing.
Participants & Data
There were 6 participants in total, all of which I was able to record the 18x receptivity ratings. To allow sociocultural insights to be made from the analysis stage, I ensured to record the participant's age, gender and nationality.
Data Preparation & Process
At first, parametric data was assumed, and I prepared the data in accordance with this, ensuring homogeneity and a normal distribution. When the data proved resistant to the subsequent transformations, I made the decision to switch to non-parametric tests.
The Setting
[faces anonymised due to University data protection guidlines] The therapy-esque scenarios were carried out at a place of my participants' choosing, based on logistics and availability. I recorded the likert ratings, aswell as any verbal comments during the testing.
Participants & Data
There were 6 participants in total, all of which I was able to record the 18x receptivity ratings. To allow sociocultural insights to be made from the analysis stage, I ensured to record the participant's age, gender and nationality.
Data Preparation & Process
At first, parametric data was assumed, and I prepared the data in accordance with this, ensuring homogeneity and a normal distribution. When the data proved resistant to the subsequent transformations, I made the decision to switch to non-parametric tests.
The Setting
[faces anonymised due to University data protection guidlines] The therapy-esque scenarios were carried out at a place of my participants' choosing, based on logistics and availability. I recorded the likert ratings, aswell as any verbal comments during the testing.
Participants & Data
There were 6 participants in total, all of which I was able to record the 18x receptivity ratings. To allow sociocultural insights to be made from the analysis stage, I ensured to record the participant's age, gender and nationality.
Data Preparation & Process
At first, parametric data was assumed, and I prepared the data in accordance with this, ensuring homogeneity and a normal distribution. When the data proved resistant to the subsequent transformations, I made the decision to switch to non-parametric tests.
The Setting
[faces anonymised due to University data protection guidlines] The therapy-esque scenarios were carried out at a place of my participants' choosing, based on logistics and availability. I recorded the likert ratings, aswell as any verbal comments during the testing.
Participants & Data
There were 6 participants in total, all of which I was able to record the 18x receptivity ratings. To allow sociocultural insights to be made from the analysis stage, I ensured to record the participant's age, gender and nationality.
Data Preparation & Process
At first, parametric data was assumed, and I prepared the data in accordance with this, ensuring homogeneity and a normal distribution. When the data proved resistant to the subsequent transformations, I made the decision to switch to non-parametric tests.
The Setting
[faces anonymised due to University data protection guidlines] The therapy-esque scenarios were carried out at a place of my participants' choosing, based on logistics and availability. I recorded the likert ratings, aswell as any verbal comments during the testing.
Participants & Data
There were 6 participants in total, all of which I was able to record the 18x receptivity ratings. To allow sociocultural insights to be made from the analysis stage, I ensured to record the participant's age, gender and nationality.
Data Preparation & Process
At first, parametric data was assumed, and I prepared the data in accordance with this, ensuring homogeneity and a normal distribution. When the data proved resistant to the subsequent transformations, I made the decision to switch to non-parametric tests.
Design of Research Interfaces
Interfaces for Research through Design (RtD)
The two designs both adapt to the users' responses in their own separate way, but with the same premeditated scenario paths - each isolating their assessment criteria with a different approach.
Statistical Observations
🔬 Analysis Parameters
Findings presented are based on results with a significance value of ≤0.05. H₀ = no observed significant difference in A/B user-perceived sociocultural empathy. H₁ = a significant observed difference betwen in A/B user-perceived sociocultural empathy.
🔬 Analysis Parameters
Findings presented are based on results with a significance value of ≤0.05. H₀ = no observed significant difference in A/B user-perceived sociocultural empathy. H₁ = a significant observed difference betwen in A/B user-perceived sociocultural empathy.
🔬 Analysis Parameters
Findings presented are based on results with a significance value of ≤0.05. H₀ = no observed significant difference in A/B user-perceived sociocultural empathy. H₁ = a significant observed difference betwen in A/B user-perceived sociocultural empathy.
🔬 Analysis Parameters
Findings presented are based on results with a significance value of ≤0.05. H₀ = no observed significant difference in A/B user-perceived sociocultural empathy. H₁ = a significant observed difference betwen in A/B user-perceived sociocultural empathy.
1
Gender × Design Approach B
Female participants responded more positively to Approach B/ECM than male participants.
p = 0.034
see implications +
see implications +
2
Nationality × Universal Identity
Participants with a nationality of Indian responded to A/B more consistently than participants of other nationalities on the Universal level.
p = 0.049
see implications +
see implications +
3
Gender × Relational Identity
Female participants also responded more positively in the Relational Level of Identity to A/B.
p = 0.046
see implications +
see implications +
Research Outcome
My research generated guidelines for sociocultural design of AI interfaces, that enhance empathy, user satisfaction & speed of adoption.
Generation 1 - Empathising on Gender
The recognition of emotion in speech, and the application of a Semiotic Square, were less able to meet the needs of males as they were for females. Female users must be provided with self-reflective visual motifs, while male users require visual evidence of reflective listening.
Generation 2 - Empathising on Nationality
Cross-cultural standards have an impact on the effectiveness of visual-supported active listening, so it is necessary to supplement user interactions with semiotic materials that counterweight the speech content in the direction of individual user expectations.
Generation 3 - Empathising on Identity
Satisfaction increases when placing emphasis on visually recognising user demographics, such as age, race, and socioeconomic status, as it has proven easier to empathise on the Relational level of the Tripartite of Identity (Sue, 2001) than the Individual or Universal levels.
