Postdoctoral Project: Decoding partner-specific neural preparation in task-oriented human-human and human-robot interaction
Part of research project: Multimodal Interaction and Communication
Description of the research project
To fully understand the mechanisms of social interaction and communication in humans and to replicate this complex human skill in technological artifacts, we must provide effective means of knowledge transfer between agents. This project consists of a series of fMRI experiments on human-human and human-robot task oriented verbal interactions. Participants inside the scanner interact via a live video feed with human or robot partners outside of the scanner. During one trial, participants need to instruct their task partners where to position objects on a grid. Using multivariate pattern classification we will aim to (1) identify brain structures that encode information specific to the particular task partner, (2) understand whether this partner-specific representation also remains active during speaking and informs brain areas associated with speech production, (3) clarify the temporal dynamics with which task-related information builds up over time in the task partner’s brain. Moreover, in the case of human-robot interaction we aim to (4) evaluate from a human user perspective to what degree human-robot interactions are processed in exactly the same fashion just as human-human interaction (“Brain-Based Turing Test”). Robots will differ in degrees of intelligence.
Description of the postdoctoral project
The postdoc associated with this project will be involved in all project stages, including the technical preparation of the experimental setting and design, the scanning of the participants, the analysis of the data, and the dissemination of the results. The project consists of at least two large-scale fMRI-based data collections (N=50 for each experiment) including, for human-human experiments, the coordination of several participants (i.e., task partners) within one scanning session. Data will be analyzed using (1) univariate analyses contrasting trials with respect to the type of task partner, (2) multivariate pattern classification through whole-brain decoding using a spherical searchlight, and (3) functional connectivity analysis based on comparing the correlation structure between specified regions of interest as a function of partner condition.
Project start date: October 1, 2019
Applicants must hold a Ph.D. degree in Psychology or related fields (e.g., Cognitive Science, Neuroscience, Biology) and should have a proven background in the following specifications:
- Training in experimental methods
- Familiarity with fMRI data acquisition and analyses
- Programming skills in Matlab and experiences in advanced statistical analysis, ideally with multivariate data analysis using machine learning techniques and/or functional connectivity analyses
- Scientific track record
- Interest (and ideally, research experience) in language processing, social cognition, and/or advanced analysis of neuroimaging data
- Good academic writing and presentation skills
- Willingness and ability to work in interdisciplinary teams
- Strong interpersonal and organizational skill