Short answer

When designing AI coaching tools, consider using embodied agents with active listening behaviors to improve user engagement and perception of naturalness.

Field: User-Centred Design
Source: Paladyn Journal of Behavioral Robotics (2024)
Method: Online survey with video stimuli
Sample: 168 participants
Evidence: Moderate effect

When designing AI agents for coaching, a humanoid robot exhibiting active listening behaviors is perceived as more likable and natural than a voice assistant or a robot with passive listening. This user-centred design research insight is drawn from a 2024 study published in Paladyn Journal of Behavioral Robotics. Using Online survey with video stimuli with 168 participants, researchers explored how this design variable affects real-world outcomes. The key design takeaway: When designing AI coaching tools, consider using embodied agents with active listening behaviors to improve user engagement and perception of naturalness.

Study

User-Centred DesignRecentModerate effect

Humanoid robots with active listening cues enhance perceived likability and naturalness in public speaking coaching

When designing AI agents for coaching, a humanoid robot exhibiting active listening behaviors is perceived as more likable and natural than a voice assistant or a robot with passive listening.

Paladyn Journal of Behavioral Robotics · 2024

Key Findings

01The humanoid robot (both active and passive listening) was perceived as more human-like and likable than the voice assistant.
02The active listening robot was rated as more satisfying, engaging, natural, and warmer than the voice assistant.
03The active listening robot was perceived as more natural than the passive listening robot.
04No significant differences were found in perceived intelligence, competence, discomfort, or helpfulness across the agents.
05Participant gender and personality traits influenced agent evaluations.

Application

Design takeaway

When designing AI coaching tools, consider using embodied agents with active listening behaviors to improve user engagement and perception of naturalness.

How to apply

When developing AI tutors or coaches, explore the use of avatars or physical robots that can demonstrate active listening through subtle head nods, eye contact, or other nonverbal signals.

Project actions

01When evaluating AI agents, consider both their functional capabilities and their social presence.
02Think about how the physical form and interaction style of an AI can impact user trust and engagement.

Method & Evidence

AimHow do users perceive the likability, naturalness, and effectiveness of humanoid robots versus voice assistants as public speaking coaches, considering different interaction modes?

MethodOnline survey with video stimuli

ProcedureParticipants watched videos of three public speaking coaching agents: a humanoid robot in active listening mode, a humanoid robot in passive listening mode, and a voice assistant. They then rated their perceptions of the agents on various attributes.

Sample168 participants

ContextPublic speaking rehearsal and coaching

Variables

IV["Type of agent (humanoid robot active listening, humanoid robot passive listening, voice assistant)","Agent's nonverbal behavior (active vs. passive listening)"]

DV["Perceived human-likeness","Perceived likability","Perceived satisfaction","Perceived engagement","Perceived naturalness","Perceived warmth","Perceived intelligence","Perceived competence","Perceived discomfort","Perceived helpfulness"]

CV["Context of public speaking rehearsal","Content of coaching interaction (implied)","Participant demographics (potentially, though also a moderator)"]

Strengths & Limitations

Strengths

+Direct comparison of different agent types and interaction styles.
+Inclusion of both embodiment and nonverbal behavior as factors.

Limitations

The study's findings might be specific to public speaking coaching and may not generalize to all AI coaching applications. User preferences can also be highly individual.

Reliability & validity

The study's use of standardized video stimuli and quantitative ratings likely contributes to reliability. Validity is supported by measuring multiple facets of user perception related to the AI's social behavior and effectiveness.

Think critically

To what extent do the perceived 'human-like attributes' of a robot contribute to its effectiveness as a coach, versus its actual functional capabilities?

Design Principles

"Embodied AI agents with responsive nonverbal cues enhance user perception of social presence and likability."

This research highlights the critical role of embodiment and nonverbal communication in user perception of AI coaching tools. Designers can leverage these findings to create more engaging and effective AI companions by focusing on nuanced social cues.

What This Means for Your Design

People like robots that act like they're listening more than just a voice, especially for practicing speeches.

How to use in your project

1.Use this study to justify the choice of an embodied agent over a non-embodied one in your design proposal.
2.Refer to the findings on active listening cues when explaining how your design will foster user engagement.

Add to My Project

Quick Cite

Paragraph starter

This research indicates that embodied agents, particularly those exhibiting active listening behaviors, are perceived more favorably in coaching contexts than non-embodied agents. The study by Forghani et al. (2024) found that a humanoid robot with nonverbal backchannelling was rated higher in likability, naturalness, and engagement compared to a voice assistant, suggesting that social cues are crucial for user acceptance and experience in AI-driven coaching tools.

Source

Paladyn Journal of Behavioral Robotics

Evaluating people's perceptions of an agent as a public speaking coach

journal · 2024

View source

Questions About This Research

What does the research say about humanoid robots with active listening cues enhance perceived likability and naturalness in public speaking coaching?: When designing AI coaching tools, consider using embodied agents with active listening behaviors to improve user engagement and perception of naturalness. Evidence: Paladyn Journal of Behavioral Robotics (2024).
Why does "Humanoid robots with active listening cues enhance perceived likability and naturalness in public speaking coaching" matter for design?: This research highlights the critical role of embodiment and nonverbal communication in user perception of AI coaching tools. Designers can leverage these findings to create more engaging and effective AI companions by focusing on nuanced social cues.
How can designers apply this research?: When designing AI coaching tools, consider using embodied agents with active listening behaviors to improve user engagement and perception of naturalness.
What were the main findings?: The humanoid robot (both active and passive listening) was perceived as more human-like and likable than the voice assistant.. The active listening robot was rated as more satisfying, engaging, natural, and warmer than the voice assistant.. The active listening robot was perceived as more natural than the passive listening robot.. No significant differences were found in perceived intelligence, competence, discomfort, or helpfulness across the agents.
What research method was used?: Online survey with video stimuli with 168 participants.
How strong is the evidence?: Evidence strength is rated Moderate effect, based on a 2024 journal from Paladyn Journal of Behavioral Robotics.
What should I do differently in my next project?: When developing AI tutors or coaches, explore the use of avatars or physical robots that can demonstrate active listening through subtle head nods, eye contact, or other nonverbal signals.
What are the limitations?: Perceptions of intelligence, competence, and helpfulness did not differ significantly, suggesting that while likability is influenced, core functional perceptions might require different design interventions. The study relied on video stimuli, not live interaction.