Our research

The Psy ID lab currently conducts research in the following areas:


Detecting social identity salience in text data

Photo by Julia M Cameron on Pexels

Credit: Julia M Cameron

In order to better understand how social identities play out under real-world conditions, and to make ‘big data’ from online forums and social media usable for researchers interested in social identity processes, we have developed ASIA – Automated Social Identity Assessment. We currently have validated ASIA tools to assess relative salience of a parent & feminist identity and a libertarian & entrepreneur identity, with an addiction & depression identity ASIA in development. Find out more here…


Social identity switches: Effectiveness, control, and consequences

Photo by Ketut Subiyanto from Pexels

Credit: Ketut Subiyanto

Self-categorization theory suggests frequent and effortless switches between the different social identities that we hold. But are these switches really highly effective, or do they incur costs analogous to “task switch costs”? Are these switches entirely determined by the social context, or can we “control” such identity switches ourselves? Does our ability to switch affect our well-being? Find out more here…


Social identity development

Photo by Thiago Borges from Pexels

Credit: Thiago Borges

How do we develop new social identities? Which factors help us, and which ones hinder the development? How does the development of new identities relate to mental health and well-being? In this part of the project, we are currently looking at the development of a parent identity in first-time mothers – and the relationship of different trajectories to postnatal depression; the effects of different online community memberships on changes to an addiction identity during recovery, and the role that depression and recovery identities play in sustaining recovery from addiction. Find out more here…


Group prototypicality and social influence

Photo by Oladimeji Ajegbile from Pexels

Credits: Oladimeji Ajegbile

Leadership research suggests that typical group members are more likely to become leaders in a range of circumstances and to effect change in fellow group members. Using computational social science tools, such as natural language processing, machine learning, and social network analysis, we seek to understand whether group members that speak like “us” are also the ones that garner the most attention online. Find out more here…


Collective self-understanding

Photo by ? from Pexels

Credits: ?

The Social Identity Model of Deindividuation Effects (SIDE) suggests that group members follow the values and norms of the group. This implies that the actions of group members reflect the norms and values of the group – providing us with a potential route to understanding what a group stands for. Using computational social science tools, such as natural language processing, cluster analysis and multidimensional scaling, this project seeks to examine whether a group’s linguistic style reflects the group’s collective self-understanding. Find out more here…


Technology acceptance and the digital divide

Photo by thisisengineering from Pexels

Credits: ThisIsEngineering

This project looks at the way the digital divide and different social identities affect the acceptance of service AI tools, such as a “Divorce AI”. Which opportunities but also barriers do people from different walks of life associate with such AI tools? Can psychological theories such as the Theory of Planned Behaviour (TPB) and the Social Identity Approach (SIA) help to create “Responsible Innovation”? Find out more here…


Social identity salience and disclosure of private information

In a series of experimental studies and a study based on naturally occurring online forum data, we examined whether the salience of different identities (e.g., student or customer, parent or feminist) affects a person’s disclosure of personal information. This research challenges the privacy paradox by showing that information is disclosed in line with the norms of the salient identity. Find out more here…


Skip to toolbar