Exploring how humans and automation establish and use common ground is the subject of a new University of Birmingham research project, which has been awarded funding by CREST today.
Human Engagement Through Artificial / Augmented Intelligence led by Professor Chris Baber (School of Computer Science, University of Birmingham) teaming with Ian Apperly (School of Psychology, University of Birmingham) is the latest project to be awarded funding from CREST’s recent commissioning call.
Baber’s project looks at ‘Augmented intelligence’ and how it can extend human cognitive ability. The capabilities of Artificial Intelligence / Machine Learning (AI / ML) for exploring vast data resources and discovering patterns, exceeds that of the human. However the human’s expertise will allow insight into unusual or unfamiliar patterns. The project therefore focuses on the need to ensure collaboration in pursuit of sense-making.
This requires that the AI is able to explain itself to the human, that the human can provide explanation to the AI, and that human-AI engagement progresses through the establishment and maintenance of common ground.
This engagement occurs in a ‘system’; that is, the cooperation between human and AI / ML is one interaction among many, e.g., humans cooperate with other humans, humans programme the AI / ML, humans could be involved in selecting and preparing the data that the algorithms use, the AI could interact with other algorithms etc.
Not only is it important that humans and automation establish and use common ground, but also that humans who communicate through the automation have this. The project asks how common ground might breakdown in order to explore consequences and mitigations.
These are the latest research projects to come out of CREST’s 2019 commissioning call, in addition to the other recently announced projects:
The adaptable law enforcement officer: Developing a measure of adaptive effectiveness
Dr Simon Oleszkiewicz at the University of Twente, Netherlands
The ability to adapt to changing situations is vital for law enforcement officers who are charged with the objective of establishing contact and building relationships with sources in criminal environments (i.e., covert law enforcement). Not only do these officers have to maintain a guise of adhering to a criminal conduct, they have to react fittingly to novel and uncertain situational demands. This project aims to develop a behavioural measure of adaptability relevant for police contexts. Link to project.
Understanding Twenty-First Century Militant Anti-Fascism: An Analytical Framework And Matrix
Professor Nigel Copsey at Teesside University
Anti-fascist militancy has existed for as long as fascism has, but as a form of contentious politics, militant anti-fascism is still largely neglected across both academic and policy-practitioner communities. Since the societal conditions behind the current right-wing populist surge are unlikely to disappear anytime soon, there is a pressing need for research that addresses the security implications of radical extra-parliamentary groups who hold that violent confrontation is essential to effective anti-fascist opposition. Link to project.
Simulated phishing and employee cybersecurity behaviour (SPEC)
Dr John Blythe at CybSafe
This project will conduct two studies with differing approaches to investigate (i) how policies on simulated phishing emails are currently implemented in organisations using a cross-sectional survey and (ii) the impact of simulated phishing emails policies on employees’ cyber security awareness and their perceptions of key factors (organisational trust, procedural fairness, stress and perceived monitoring) through an experimental study. Link to project.
Why do people spread disinformation on social media?
Professor Tom Buchanan at the University of Westminster
Individual social media users are key to the spread of disinformation online. By interacting with disinformation, they share it to their own social networks. This can greatly increase its reach, and potential impact on society. Why do people do this? Are they fooled by the disinformation, and spread it because they believe it is true? Do they know the information is fake, but spread it anyway? How does the way a disinformation message is presented influence our likelihood of sharing it? Are some people more likely to share disinformation than others? This project will address those questions. Link to project.
Collecting and Leveraging Identity Cues with Keystroke Analysis (CLICKA)
Dr Oliver Buckley at the University of East Anglia
The project is based on the idea of ‘motor learning’, which suggests that a task becomes more automatic and requires less conscious thought the more it is repeated. In the first instance the project will develop an experimental framework to capture a user’s typing behaviours. This will then be used to create a predictive model, using state-of-the-art machine learning techniques capable of inferring some or part of an anonymous individual’s name. Link to project.
‘Hot periods’ of anti-minority activism and the threat of violent domestic extremism: Towards an assessment framework.
Dr Joel Busher at Coventry University
The aim of this project is to develop a stronger understanding of the dynamics of violent escalation, non-escalation and de-escalation during periods of intense anti-minority activism, and in doing so enhance the ability of state and civil society actors to (a) assess the threat of violent escalation during and in the aftermath of such ‘hot periods’, and (b) more accurately anticipate how planned interventions are likely to play out on the ground. Link to project.
Understanding moral injury and belief change in the experiences of police online child sex crime investigators
Dr Peter Lee at University of Portsmouth and Dr Mark Doyle at Solent University
This project will start by analysing and exploiting primary data from moral injury-related findings. These will subsequently be used to inform a focused enquiry into the causes of moral injury, and consequences such as changes in attitudes, beliefs and behaviour, among police internet child abuse investigators and relevant forensic teams. Link to project.
For more information about the successful applicants please visit the CREST website at: www.crestresearch.ac.uk/projects/.
More projects to come…
Watch this space as the other successful projects (subject to contracts being finalised) will be announced very soon.