Research Projects
The advances in Human-AI Collaboration demonstrate the exciting potential to expand human capabilities and improve the quality of work, particularly when working with large amounts of data or complex and ambiguous scenarios. However, the success of this collaboration is predicated upon the ability of humans and AI to effectively communicate, which involves the ability to understand each other and to efficiently present the information to each other.
The Cycle of Human-AI Collaboration
The Cycle of Human-AI Collaboration can be roughly represented as follows:
(1) a human user presents some communicative input, (2) the system recognizes the intent of this input, (3) and produces information according to the intent, (4) then formats this (internal 4a) and other (external 4b) required information for human understanding, and (5) communicates it back to the user. (6) What is being communicated and how it is communicated affects the user's perceptions of the information and of the system, (7) further influencing the user's mental model and beliefs about the system. (8) Based on a combination of these in-situ perceptions and overall beliefs, the user formulates their communicative input with a corresponding intent and (1) presents it to the system.
The mechanisms defining how each step proceeds are inevitably affected by contextual dynamics (e.g. systematic differences between users) and might differ for specific processes (e.g. financial decision-making, creativity support, education, etc.), but the common goal of this collaboration is to bring better task outcomes while ensuring positive social impact.
The COoKIE Group Research on Human-AI Collaboration
Research projects at The COoKIE Group explore different stages of the Human-AI Collaboration cycle to gain a holistic understanding of the underlying mechanisms and to lead the development of effective, efficient, and ethical Human-AI collaborating teams.
Below are our current or recently finished projects, published projects are in bold. Click to expand.
Something Borrowed: Exploring the Influence of AI-Generated Explanation Text on the Composition o...
Mobilizing Crowdwork: A Systematic Assessment of the Mobile Usability of HITs
Drone in Love: Emotional Perception of Facial Expressions on Flying Robots
Personal Space in Play: Physical and Digital Boundaries in Large-Display Cooperative and ...
A Human-Centered Review of Algorithms for Decision-Making in Higher Education
Drone in Love: Emotional Perception of Facial Expressions on Flying Robots
Naturally Together: A Systematic Approach for Multi-User Interaction With Natural Interfaces
[Preview] Mobilizing Crowdwork: A Systematic Assessment of the Mobile Usability of HITs
Drone in Love: Emotional Perception of Facial Expressions on Flying Robots
Human-Computer Interaction and User Interface Design