Methods Events
Informal events highlighting a range of methods used in affective science, hosted by expert users. Come along to ask all your burning questions!
Clicking on each speaker’s name will take you to their personal websites.
2022 Themes
- Linguistic Analysis
- Machine Learning Primer
- Virtual Reality (multi-person virtual reality)
- Health Behavior Assessment *Sponsored by ilumivu, Inc.
- Computational Modeling
- Conducting Research with Community Partners
- Multi-Person Physiology *Presented and sponsored by BIOPAC Systems, Inc.
- MEG Methods
- Longitudinal Data Analysis
- Digital Agents
Theme #1: Linguistic Analysis
Joshua Jackson
Title: From Text to Thought: How Analyzing Language Can Advance Affective Science
Date: Friday, April 1st, 2022
Time: 8:00pm – 9:00pm EDT
Description: Humans have been using language for millennia, but we have only just begun to scratch the surface of what natural language can tell us about the mind. In this methods event, I will highlight how two forms of language analysis—natural language processing and comparative linguistics—are contributing to how we understand psychological processes, with a focus on emotion. I will summarize resources for learning both methods and highlight the best way to combine language analysis with more traditional psychological paradigms
Theme #2: Machine Learning Primer
Shirley Wang
Title: Introduction to Machine Learning for Affective Scientists
Date: Saturday, April 2nd, 2022
Time: 3:15pm – 4:15pm EDT
Description: Whereas statistical methods traditionally used in the social and behavioral sciences emphasize interpretability and inference, machine learning methods focus on predictive accuracy. This Methods Event will provide a brief primer and conceptual introduction to machine learning, with a focus on supervised learning methods and how they may be applied in affective science.
Theme #3: Virtual Reality (multi-person virtual reality)
Alexia Galati
Title: Virtual Reality as a Novel Method for Affective Science
Date: Saturday, April 2nd, 2022
Time: 3:15pm – 4:15pm EDT
Description: I will introduce virtual reality (VR) as a method for affective science that can accommodate both ecological validity and experimental control. I will discuss some benefits of using VR, which include: immersing participants in dynamic task-relevant environments, designing experiments that don’t have a repetitive trial structure, using virtual agents that display relevant behaviors (e.g., postures, facial expressions) that are convincing and consistent, and recording rich, continuous, multimodal data from participants. I will also address technical and practical challenges associated with using VR as a research method.
Theme #4: Health Behavior Assessment
Sidney D’Mello
Sponsored by:
Title: Assessing Health and Wellbeing from Wearables in the Real World: How, How Well, and What Next?
Date: Thursday, March 31st, 2022
Time: 3:30pm – 4:45pm EDT
Description: Human behavior assessment of health and well-being in the “real” world – where people go about their daily routines in their homes and workplaces – is a different problem than assessment in the lab or in the “quasi” wild (e.g., YouTube videos). Building off the recently completed IARPA-sponsored MOSAIC program (Multimodal Objective Sensing to Assess Individuals with Context), the talk will emphasize the major approaches, performance achieved, challenges faced, and lessons learned. Sponsorship by ilumivu, Inc. will include a brief 30s promotional video and will not have an impact on the content of this method event. The presenter is not affiliated with the sponsor.
Theme #5: Computational Modeling
Oriel FeldmanHall
Joseph Heffner
Title: Modeling Emotions Using Computational Tools and Machine Learning
Date: Friday, April 1st, 2022
Time: 8:00pm – 9:15pm EDT
Description: Computational modeling and machine learning methods bring a level of precision and description to fuzzy theories in psychology and neuroscience. By formalizing mathematically tractable models of emotional experiences, we can improve our understanding of the mechanisms behind emotion and make progress on advancing emotion theory. We will provide an introduction for how standard computational models and data driven methods can be fruitfully adapted to the study of emotion
Theme #6: Conducting Research with Community Partners
Laura Armstrong
Title: Opportunities for Extending Affective Science into the Community
Date: Thursday, March 31st, 2022
Time: 3:30pm – 4:45pm EDT
Description: This methods event will provide a framework for envisioning affective research as a community-engaged endeavor. The focus will be on best practices for engaging diverse community stakeholders and the use of partnership-based approaches to enhance science and practice.
Theme #7: Multi-Person Physiology
Alex Dimov
Title: Research in the Metaverse: Multi-person Physiological Assessment from Dyads and Triads to Virtual Reality
Date: Thursday, March 31st, 2022
Time: 10:45am – 12:00pm EDT
Description: We will present a complete solution for experiments in virtual reality involving two or more participants. Easily immerse multiple users in a shared environment even if they are not physically at the same location. Customize avatars, 3D environments, add interactivity. Measure physiological responses related to emotion, stress, and add eye tracking analytics to provide deeper insights. Learn about deploying a fully integrated and scalable system that is designed with researchers in mind. Find out how you can get started even with minimum programming skills.
Theme #8: MEG Methods
Joachim Gross
Title: MEG in Affective Science
Date: Thursday, March 31st, 2022
Time: 10:45am – 12:00pm EDT
Description: I will lay out the benefits of using MEG in affective science, also in comparison to other neuroimaging methods. I will discuss novel ways of doing body-brain studies and introduce analysis methods that can be applied to affective science.
Theme #9: Longitudinal Data Analysis
Janne Adolf
Title: Missing Data in Ecological Momentary Assessment
Date: Thursday, March 31st, 2022
Time: 10:45am – 12:00pm EDT
Description: The topic of missing data is a complex one – maybe especially so in ecological momentary assessment. The aim of this Methods Event is to discuss different reasons for and perspectives on the emergence of missing data, such as non-compliance, night breaks, or discrete sampling in continuous time, as well as different approaches to handling missing data, such as imputation methods, full information maximum likelihood estimation, or continuous-time modeling.
Theme #10: Digital Agents
Stacy Marsella
Title: Virtual Humans in Affective Science
Date: Saturday, April 2nd, 2022
Time: 11:00am – 12:00pm EDT
Description: We will discuss the synergy between research methods in Psychology and in the design of Virtual Humans, graphical facsimiles of humans capable of interacting with people in face-to-face interaction. The design of virtual humans requires the computational specification of psychological theories and data, while the resulting computational models can be simulated to generate predictions and hypotheses that can subsequently be evaluated in human studies. Virtual humans also serve as highly controllable confederates in laboratory experiments and can be used in applications that support large scale data collection.