Task 7 Flashcards
Why were emotions initially ignored in AI research?
Early AI research followed a Stoic perspective, treating emotions as irrational and counterproductive to intelligence.
What changed the view on emotions in AI?
Appraisal theories showed that emotions guide adaptive responses.
Neurological studies found that emotional impairments lead to poor decision-making.
Social research revealed that emotions facilitate communication and coordination.
How are emotions useful in AI and robotics?
Prioritizing cognitive resources (focusing on urgent issues).
Enhancing human-computer interaction (trust, rapport).
Improving decision-making (emotion-based heuristics).
What are the three main theories of emotion?
Discrete Theories – Fixed basic emotions (e.g., anger, joy, fear) that are universal.
Dimensional Theories – Emotions are a continuous space (e.g., PAD model: Pleasure, Arousal, Dominance).
Appraisal Theories – Emotions arise from evaluating person-environment relationships.
Why are appraisal theories widely used in AI?
They map well onto AI decision models, such as belief-desire-intention (BDI) architectures.
What is a criticism of appraisal theories?
They assume slow, deliberate evaluation, while many emotions occur rapidly and automatically.
What is EMA?
A computational model of emotion that integrates:
Causal Interpretation – Tracks agent-environment relationships. Appraisal Component – Evaluates relevance, desirability, likelihood, and controllability of events. Emotion Derivation – Determines emotional responses. Coping Strategies – Adjusts beliefs, desires, or attention to regulate emotions.
How does EMA model coping strategies?
Attention-based coping – Redirecting focus (e.g., ignoring distractions).
Belief-based coping – Changing interpretations (e.g., blaming others).
Desire-based coping – Adjusting goals (e.g., deciding success is unimportant).
Intention-based coping – Taking action (e.g., studying harder after failing an exam).
What are the benefits of AI expressing emotions?
Enhances trust and cooperation in human-machine interaction.
Improves user engagement (e.g., virtual assistants).
Facilitates teamwork in multi-agent systems.
Aids in education, therapy, and elderly care.
Do AI emotions need to be real?
No—expressions alone can influence user perception and social interaction, regardless of internal experience.
What is reverse appraisal in AI?
Inferring an agent’s goals and future actions based on its emotional expressions.
How do people react to AI emotions?
People tend to treat machines socially (e.g., politeness, gender biases).
Machines that mirror human emotions increase rapport.
AI avatars that resemble users improve persuasion and compliance.
What is the uncanny valley effect in emotional AI?
AI that is almost human-like but slightly unnatural disturbs users rather than creating trust.
How do brain responses differ when interacting with humans vs. AI?
People show lower emotional activation and less guilt when exploiting AI compared to humans.
What is joint attention, and why is it important?
When two individuals focus on the same object or event, facilitating communication and learning.
Can robots induce joint attention?
Yes—embodied robots (not just 2D screens) can elicit joint attention responses from humans.
What is the bystander effect in human-robot interaction?
People feel less personal responsibility in a task when interacting with a robot rather than alone.
What is affective computing?
The study of how to detect, interpret, and simulate human emotions in machines.
What are the two main goals of computational emotion modeling?
Understanding human emotion through simulation.
Enhancing AI decision-making with emotional intelligence.
Why is standardizing AI emotion models difficult?
Psychology lacks a unified theory of emotion, making implementation inconsistent.
What are the three main AI architectures used for modeling emotions?
Classical Cognitive Architectures (e.g., ACT-R, SOAR).
Belief-Desire-Intention (BDI) Architectures (e.g., FAtiMA).
Affective Agent Architectures (e.g., MAMID).
How does SOAR model emotions?
By incorporating appraisal processes as production rules that shape decision-making.
How does FAtiMA Modular simulate emotions?
Uses a BDI-based reasoning process.
Generates prospect-based emotions (hope, fear) based on agent goals.
Incorporates coping strategies to manage emotions dynamically.
How does MAMID model emotions differently?
It allows trait-based personality differences in emotional reactions (e.g., anxious vs. non-anxious responses).
What is the benefit of formalizing emotions in set theory?
It forces clarity in defining assumptions and improves comparisons between models.
How does BDI logic help model emotions?
It integrates emotions into a rational agent framework, allowing AI to weigh beliefs, desires, and goals in emotional responses.
What are the limitations of BDI models for emotion?
BDI logic assumes perfect rationality, which is unrealistic for human emotions.
It lacks mechanisms for intensity and unconscious processing.
What are some key challenges in AI emotional intelligence?
Creating standardized emotion models for AI.
Ensuring authenticity of AI expressions to avoid user distrust.
Balancing realistic emotional responses with efficiency.
Understanding ethical implications (e.g., manipulating users).
What is the main goal of future emotion-based AI?
To create AI that can interact naturally with humans, adapt emotionally, and enhance decision-making.
What are some real-world applications of AI emotional modeling?
Healthcare (robotic therapy, emotional support).
Customer service (empathetic virtual assistants).
Education (emotionally responsive tutors).
Gaming & storytelling (adaptive NPC emotions).
How can AI ethically integrate emotions?
By ensuring emotional expressions enhance user experience rather than manipulating emotions for profit.
Why were emotions originally considered unimportant in AI?
Early AI research viewed emotions as irrational and assumed intelligence was purely logical and rule-based.
How has the view of emotions in AI changed?
Research shows emotions enhance decision-making, social interaction, and adaptability, making them critical for intelligent behavior.
What are three key reasons emotions are essential for AI?
Prioritization – Helps AI focus on relevant tasks.
Social Bonding – Facilitates interaction with humans.
Decision-Making – Affects risk-taking and judgment.
What are the three main emotion theories used in AI?
Discrete Theories – Emotions are fixed categories (e.g., joy, fear, anger).
Dimensional Theories – Emotions exist in continuous spaces (e.g., arousal vs. valence).
Appraisal Theories – Emotions arise from evaluating goals, control, and expectations.
Why are appraisal theories popular in AI?
They align well with goal-oriented AI models (e.g., Belief-Desire-Intention frameworks).
What is the PAD emotional model?
A dimensional model representing emotions using:
Pleasure (P) – How positive/negative an emotion is. Arousal (A) – The intensity of the emotion. Dominance (D) – The level of control over the situation.
What is EMA (Emotion and Adaptation Model)?
A computational framework that models emotions based on appraisal and coping mechanisms.
What are the main ethical concerns of AI emotion modeling?
Manipulation – AI emotions could be used to deceive or persuade.
Privacy – Emotional data could be misused.
Authenticity – Users may form unrealistic emotional attachments.
How can AI ethically integrate emotions?
Clearly indicate that emotions are simulated.
Ensure AI serves human well-being rather than exploiting users.
Avoid creating excessively deceptive emotional expressions.
