Talks

Coming up

  • 18 December: (chair) AFIRE Funder’s Forum - promoting experiments by research funders : slides

Past talks

2024

  • 16 December: “History of partial randomisation in research”. At “Using partial randomisation to award research funding – a day of discussion”, hosted by the British Academy, London. Slides

  • 20 November: (chair) AFIRE Funder’s Forum - promoting experiments by research funders : slides

  • 23 October: (chair) AFIRE Funder’s Forum - promoting experiments by research funders: slides

  • 8th September: “Research Culture in the UK”. Keynote at Pujiang Innovation Forum 2024, Shanghai, China.

    • Britain has a rich industrial and scientific heritage, with an established and diverse research culture anchored within a world-leading University sector. Successfully sustaining and building on the reputation of British research means addressing the challenges and criticisms of how research culture has developed in the 21st century. I will review some of these challenges - hypercompetition and excessive workload, employment precarity, perverse metrics, unfortunate publishing practices, lack of transparency, irreproducibility of findings and threats to international collaboration - and suggest ways forward.
    • Slides: PPT, PDF

  • 9th September: “Experiments with the funding of research”. Shanghai Institute for Science of Science.

    • Research funders share some common concerns: how do they know if they are allocating funding in the way that is most effective, fair and best balances risks versus possible rewards? Although a large variety of different funding systems exist, rigorous evidence on how funding can be designed to promote the best outcomes is not common. I will speak about the Research on Research Institute, and the programme on formal experiments that I lead there: AFIRE https://researchonresearch.org/project/a-f-i-r-e I will showcase good examples of strong evidence on aspects of funding systems, and talk about how funders might better use and generate evidence in the future
    • Slides: PPT, PDF, transcript

  • 22nd May 2024 AFIRE Project, Research on Research Institute at Seminar: Interventions for Funding Application Assessments for EDI Caucus

  • 17th April 2024, Improving The Research Funding Lottery at School of Architecture, University of Sheffield.

    • Researchers spend hours and hours writing funding proposals and in the end bad ideas get funded and good ideas get missed. Everyone has the suspicion that the funding process is biased against people in the wrong fields, from the wrong institutions or with the wrong backgrounds. Can research funding be improved, and what does the evidence say? Tom will talk about work from the Research on Research Institute, and about projects on innovations in how research funding is awarded.https://researchonresearch.org/

  • 10th April 2024, Research on Research at School of Psychology Research Day, University of Sheffield.

  • 9th February 2024, Adventures in the Multiverse at TARG, University of Bristol

    • Complex data and complex methods mean multiple statistical analyses are potentially legitimate. Multiverse analysis offers a tool to interrogate the stability of results under different analysis choices. After introducing the general history and rationale of multiverse analysis, I will talk about two multiverse projects I have been involved in, both in my research area of bias and decision making, and what we have learned from them

    • Slides: “Adventures in the Multiverse” (PDF)

  • 19th January 2024 6pm UK time, Google IQ Research Talk: When people are wrong on the internet - reasons to be optimistic (host Mevan Babakar)

    • Human reasoning in the digital age has a terrible reputation. People are often portrayed as hopelessly biased, polarised and resistant to good sense, while - strangely - also being easily manipulated by malevolent forces and bizarre fads. From the trenches of psychological science, I bring you good news! I will discuss two strands of work. In the first, we show that even minimal deliberation allows online groups to select away from intuitive but wrong answers (and that there are prospects for language-model based dialogue agents to support good group reasoning). More at delibot.xyz. In the second strand of research we show that, across polarising topics, UK partisans are surprisingly good at representing the views of the ideological opponents, and that the better people are at this the more likely they are to view people they disagree with as moral, informed and rational. More in Brand, Brady & Stafford (2023). 

    • Slides, PDF

2023

2022

These slides form the basis for events run in the:

      • Department of Urban Studies & Planning (2022-04-26)

      • Department of Economics (2022-05-04)

      • The Social Research Institutes (2022-05-13)

      • School of Architecture (2022-05-19)

      • School of Education (2022-05-25)

      • School of Clinical Dentistry (2022-05-26), Centre for Care (2022-06-07)

      • Department of Sociology and Social Care (2022-06-07)

      • Department of Politics (2022-06-07)

      • Department of Architecture (2022-06-29)

      • Department of Landscape Architecture (2022-07-06) and others.

  • 23 March and 14 April 2022 - Understanding the brain through the eyes
    What we see and how we see can be used to help us understand how the brain works. Illusions are not just visual tricks, but illustrate the deep principles which our brains use to construct reality. In this talk, with the use of many examples, I will show how the study of illusions provides important lessons for psychologists trying to understand all aspects of mind, brain and behaviour

  • 23 Feb 2022 - Open Research at the DDRI Network meeting

2021

  • 7 December 2021 - Faculty doctoral training on research professionalism and integrity (DDP Module code: FCP6101)

  • 25 November 2021 - SuperVisionaries: Open Research for Supervisors

  • 17 November 2021 - Open Research Conversations: “Open Source Licensing”, with Bob Turner

  • 27 October 2021 - Into the Multiverse, slides to support a Sheffield MetaNet discussion

  • 30 June 2021 - The rise of on-line political advertising: A ‘moral panic’?, with Dr Kate Dommett, a Top Talk from the APG.

  • 16 June 2021 - Why you should use decision models with your response time and accuracy data at Centre for Cognition and Neuroscience, University of Huddersfield (and online).

    Decision modelling is increasingly accessible and yields highly practical benefits to experimental psychologists. This talk will explain why and how you should apply decision models to your data.

    Response time and accuracy are fundamental measures of behavioural science, but discerning participants’ underlying abilities can be masked by speed-accuracy trade-offs (SATOs). SATOs are often inadequately addressed in experiment analyses which focus on a single variable or which involve a suboptimal analytic correction. Models of decision making, such as the drift diffusion model (DDM), provide a principled account of the decision making process, allowing the recovery of SATO-unconfounded decision parameters from observed behavioural variables.

    I will show, using simulations and worked examples from experimental data, how decision models allow potentially massive increases in statistical power at the same time as helping you avoid mistaken inferences due to variations in participants’ SATOs strategies. Because the modelling community has focussed on differences between different decision models, this has disguised the fact that there is now a consensus around the core mechanisms which define the family of plausible decision models, as well as increasing agreement on tools for fitting behavioural data with these decision models.

    Stafford, T., Pirrone, A., Croucher, M., and Krystalli, A. (2020). Quantifying the benefits of using decision models with response time and accuracy data. Behavioral Research Methods, 52, 2142-2155 doi.org/10.3758/s13428-020-01372-w
    See also:

  • 14 June 2021 - A researcher’s perspective on reproducibility, University of Sheffield Medical School Research Meeting

  • 3 March 2021 - Revolutions in Reproducibility, Sheffield Metascience Network Launch

  • 18 Jan 2021 - Reproducibility in Practice - Open Research Conversations at University of Sheffield. With Dr Anna Krystalli

  • Google slides

  • Presentation (PDF, 673KB)

  • Talk recording

2020

2019

2018

2017

2016

2012

  • July 2012 - A Berlin Cognitive Science Safari

    • Report

    • And I made the cover of the Wall Street Journal:

2011 and earlier

  • Beyond reinforcement learning in action acquisition, (OpenOffice, 3.7MB) 9 November 2011, Department of Psychological Sciences, Birkbeck University of London.

  • Inferring cognitive architectures from high-resolution behavioural data, 13 May 2011, York Centre for Complex Systems Analysis, University of York.

  • A novel task for the investigation of action learning, 7 July 2010, Experimental Psychology Society, Manchester, 7-9 July 2010

  • An empirical test of some philosophies of science, 21 May 2010, Heng Seng Centre for Cognitive Studies, University of Sheffield

  • How do we use computational models of cognitive processes?, Neural Computation and Psychology Workshop, 8-10 April 2010, Birkbeck, London

  • What use are computational models of cognitive processes?, (PowerPoint, 3.6MB) 19th of March 2010, Redwood Centre for Theoretical Neuroscience, UC Berkeley

  • Using the psychophysics of choice behaviours to infer mental structure from reaction times,(PowerPoint 2MB) 15th of January 2010, Department of Psychology, National University of Ireland, Galway

  • Pieron’s Law holds in conditions of response conflict, 1st August, presented at the 31st Annual Conference of the Cognitive Science Society. Amsterdam, The Netherlands.
    Presentation (PowerPoint, 1.6MB)

  • The nonconscious mere exposure effect with brand logos: real but elusive, Department of Psychology, City College, Thessaloniki, 1/6/09

  • Email: the technology and psychology of continuous partial attention, UFI, Sheffield, 12 November 2008

  • Things can be known : Teaching psychology through demonstrations (PowerPoint, 7.4MB) Keynote at 26th Annual Conference of the Association for the Teaching of Psychology, University of Lincoln, 11th of July, 2008

  • How to make students talk in seminars, HEA Psychology ‘Postgraduates who Teach’ Network, University of Birmingham, 27 May 2008

  • The psychological foundations of privacy, Privacy in Law, Ethics and Genetic Data, 1st international workshop PRIVILEGED Project (EC FP6), 10 Jan 2008

  • Debates in Cognitive Neuroscience, Centre for Inquiry Based Learning in the Arts and Social Sciences, University of Sheffield, 30 May 2007

  • Is there a science of advertising? (PowerPoint, 3.3MB) University of Hull, Management School, 27th of April 2007

  • Residents’ perception of risk on contaminated sites, Environment Agency training day, Leeds, 20th of June, 2007