skip to content
Home

The Psychometrics Centre

Cambridge Judge Business School
 

Feed aggregator Subscribe to The Psychometrics Centre aggregator

Thu 20 Nov 14:00: Building efficient & useful knowledge systems

Other Psychology Seminars - Fri, 19/09/2025 - 08:05
Building efficient & useful knowledge systems

Abstract not available

Add to your calendar or Include in your list

Tue 07 Oct 12:00: Feedback Forensics: Measuring AI Personality By Comparing Observed Behaviour If you are interested in attending the talk online, please email the organiser and ask for a Teams invite.

Psychometrics Seminar Series - Thu, 18/09/2025 - 15:37
Feedback Forensics: Measuring AI Personality By Comparing Observed Behaviour

Many personality tests ask participants hypothetical questions predicting their own behaviours. Yet, as with humans, self-predicted AI behaviour does not always match observed behaviour. In this talk, I will introduce Feedback Forensics: a toolkit to measure AI traits related to personality directly based on observed behaviour data. Comparing model behaviours to the same input relative to each other, our toolkit can measure a diverse set of traits related to the underlying personality, manner, and style of AI responses. I will share results describing traits exhibited by popular AI models as well as detecting the traits encouraged by human feedback. The talk will feature a live demo of our personality visualisation tool and attendees are invited to follow along via our online platform https://feedbackforensics.com/ (laptops are encouraged).

Bio: Arduin is currently a PhD student in the Department of Computer Science in Cambridge working on AI model evaluation. His work focuses on understanding what desirable and undesirable model behaviours are reinforced by human and AI feedback. Prior to joining his current PhD programme, Arduin completed an MPhil in Machine Learning and Machine Intelligence in Cambridge’s Engineering Department. Recently, Arduin also worked on model evaluation within Apple’s Foundation Models team as an intern.

If you are interested in attending the talk online, please email the organiser and ask for a Teams invite.

Add to your calendar or Include in your list

Tue 07 Oct 12:00: Feedback Forensics: Measuring AI Personality By Comparing Observed Behaviour If you are interested in attending the talk online, please email the organiser and ask for a Teams invite.

Psychometrics Centre Seminars - Thu, 18/09/2025 - 15:37
Feedback Forensics: Measuring AI Personality By Comparing Observed Behaviour

Many personality tests ask participants hypothetical questions predicting their own behaviours. Yet, as with humans, self-predicted AI behaviour does not always match observed behaviour. In this talk, I will introduce Feedback Forensics: a toolkit to measure AI traits related to personality directly based on observed behaviour data. Comparing model behaviours to the same input relative to each other, our toolkit can measure a diverse set of traits related to the underlying personality, manner, and style of AI responses. I will share results describing traits exhibited by popular AI models as well as detecting the traits encouraged by human feedback. The talk will feature a live demo of our personality visualisation tool and attendees are invited to follow along via our online platform https://feedbackforensics.com/ (laptops are encouraged).

Bio: Arduin is currently a PhD student in the Department of Computer Science in Cambridge working on AI model evaluation. His work focuses on understanding what desirable and undesirable model behaviours are reinforced by human and AI feedback. Prior to joining his current PhD programme, Arduin completed an MPhil in Machine Learning and Machine Intelligence in Cambridge’s Engineering Department. Recently, Arduin also worked on model evaluation within Apple’s Foundation Models team as an intern.

If you are interested in attending the talk online, please email the organiser and ask for a Teams invite.

Add to your calendar or Include in your list

Tue 07 Oct 12:00: Feedback Forensics: Measuring AI Personality By Comparing Observed Behaviour If you are interested in attending the talk online, please email the organiser and ask for a Teams invite.

Psychometrics Seminars - Thu, 18/09/2025 - 15:37
Feedback Forensics: Measuring AI Personality By Comparing Observed Behaviour

Many personality tests ask participants hypothetical questions predicting their own behaviours. Yet, as with humans, self-predicted AI behaviour does not always match observed behaviour. In this talk, I will introduce Feedback Forensics: a toolkit to measure AI traits related to personality directly based on observed behaviour data. Comparing model behaviours to the same input relative to each other, our toolkit can measure a diverse set of traits related to the underlying personality, manner, and style of AI responses. I will share results describing traits exhibited by popular AI models as well as detecting the traits encouraged by human feedback. The talk will feature a live demo of our personality visualisation tool and attendees are invited to follow along via our online platform https://feedbackforensics.com/ (laptops are encouraged).

Bio: Arduin is currently a PhD student in the Department of Computer Science in Cambridge working on AI model evaluation. His work focuses on understanding what desirable and undesirable model behaviours are reinforced by human and AI feedback. Prior to joining his current PhD programme, Arduin completed an MPhil in Machine Learning and Machine Intelligence in Cambridge’s Engineering Department. Recently, Arduin also worked on model evaluation within Apple’s Foundation Models team as an intern.

If you are interested in attending the talk online, please email the organiser and ask for a Teams invite.

Add to your calendar or Include in your list

Thu 06 Nov 14:00: A meeting of minds: Modulating mentalizing in autism

Other Psychology Seminars - Thu, 18/09/2025 - 09:37
A meeting of minds: Modulating mentalizing in autism

Abstract not available

Add to your calendar or Include in your list

Thu 04 Dec 14:00: Reconnecting with the body: Repairing interoception in physical illness and mental health

Other Psychology Seminars - Thu, 18/09/2025 - 09:36
Reconnecting with the body: Repairing interoception in physical illness and mental health

Abstract not available

Add to your calendar or Include in your list

Thu 27 Nov 14:00: The neurocognition of dance

Other Psychology Seminars - Thu, 18/09/2025 - 09:35
The neurocognition of dance

Abstract not available

Add to your calendar or Include in your list

Thu 20 Nov 14:00: Talk title tbc

Other Psychology Seminars - Thu, 18/09/2025 - 09:34
Talk title tbc

Abstract not available

Add to your calendar or Include in your list

Thu 13 Nov 14:00: How would I feel tomorrow: Towards a computational understanding of subjective pain experiences

Other Psychology Seminars - Thu, 18/09/2025 - 09:24
How would I feel tomorrow: Towards a computational understanding of subjective pain experiences

Abstract not available

Add to your calendar or Include in your list

Thu 02 Oct 14:00: Attention under challenge: Addressing the consequences of modern environments

Other Psychology Seminars - Thu, 18/09/2025 - 09:23
Attention under challenge: Addressing the consequences of modern environments

Abstract not available

Add to your calendar or Include in your list

Thu 23 Oct 14:00: Grasping the invisible: Multidimensional meanings for abstract concepts

Other Psychology Seminars - Thu, 18/09/2025 - 09:16
Grasping the invisible: Multidimensional meanings for abstract concepts

Abstract not available

Add to your calendar or Include in your list

Thu 16 Oct 14:00: Understandable language models

Other Psychology Seminars - Thu, 18/09/2025 - 09:00
Understandable language models

Abstract not available

Add to your calendar or Include in your list

Thu 09 Oct 14:00: Neural and computational mechanisms of conscious visual perception in humans

Other Psychology Seminars - Thu, 18/09/2025 - 09:00
Neural and computational mechanisms of conscious visual perception in humans

Abstract not available

Add to your calendar or Include in your list

Wed 17 Sep 12:00: LLM Social Simulation is a Promising Research Method The talk is available online. Please get in touch to ask for a Teams invite.

Psychometrics Seminar Series - Tue, 16/09/2025 - 21:16
LLM Social Simulation is a Promising Research Method

The emergence of large language models as in-silico subjects for social science poses a central question: can they genuinely simulate diverse human behavior, or do they merely produce plausible, homogenized artifacts? This talk demonstrates that LLMs are powerful but imperfect simulators by presenting three core contributions. First, we establish the “Persona Effect,” showing that persona-prompting a 70B model captures 81% of explainable variance in subjective tasks, creating a strong baseline for individual-level simulation. Second, to address data scarcity, we introduce iNews, a large-scale dataset of personalized affective responses to news, enriched with persona information. Finally, we introduce SimBench, the first large-scale benchmark for group-level simulation, which reveals the strengths and critical weaknesses of current models. I conclude by arguing for the specialized datasets and training required to advance the frontier of high-fidelity human simulation.

The talk is available online. Please get in touch to ask for a Teams invite.

Add to your calendar or Include in your list

Wed 17 Sep 12:00: LLM Social Simulation is a Promising Research Method The talk is available online. Please get in touch to ask for a Teams invite.

Psychometrics Centre Seminars - Tue, 16/09/2025 - 21:16
LLM Social Simulation is a Promising Research Method

The emergence of large language models as in-silico subjects for social science poses a central question: can they genuinely simulate diverse human behavior, or do they merely produce plausible, homogenized artifacts? This talk demonstrates that LLMs are powerful but imperfect simulators by presenting three core contributions. First, we establish the “Persona Effect,” showing that persona-prompting a 70B model captures 81% of explainable variance in subjective tasks, creating a strong baseline for individual-level simulation. Second, to address data scarcity, we introduce iNews, a large-scale dataset of personalized affective responses to news, enriched with persona information. Finally, we introduce SimBench, the first large-scale benchmark for group-level simulation, which reveals the strengths and critical weaknesses of current models. I conclude by arguing for the specialized datasets and training required to advance the frontier of high-fidelity human simulation.

The talk is available online. Please get in touch to ask for a Teams invite.

Add to your calendar or Include in your list

Wed 17 Sep 12:00: LLM Social Simulation is a Promising Research Method The talk is available online. Please get in touch to ask for a Teams invite.

Psychometrics Seminars - Tue, 16/09/2025 - 21:16
LLM Social Simulation is a Promising Research Method

The emergence of large language models as in-silico subjects for social science poses a central question: can they genuinely simulate diverse human behavior, or do they merely produce plausible, homogenized artifacts? This talk demonstrates that LLMs are powerful but imperfect simulators by presenting three core contributions. First, we establish the “Persona Effect,” showing that persona-prompting a 70B model captures 81% of explainable variance in subjective tasks, creating a strong baseline for individual-level simulation. Second, to address data scarcity, we introduce iNews, a large-scale dataset of personalized affective responses to news, enriched with persona information. Finally, we introduce SimBench, the first large-scale benchmark for group-level simulation, which reveals the strengths and critical weaknesses of current models. I conclude by arguing for the specialized datasets and training required to advance the frontier of high-fidelity human simulation.

The talk is available online. Please get in touch to ask for a Teams invite.

Add to your calendar or Include in your list

Wed 17 Sep 12:00: LLM Social Simulation is a Promising Research Method The talk is available online. Please get in touch to ask for a Teams invite.

Psychometrics Seminar Series - Thu, 11/09/2025 - 10:44
LLM Social Simulation is a Promising Research Method

The emergence of large language models as in-silico subjects for social science poses a central question: can they genuinely simulate diverse human behavior, or do they merely produce plausible, homogenized artifacts? This talk demonstrates that LLMs are powerful but imperfect simulators by presenting three core contributions. First, we establish the “Persona Effect,” showing that persona-prompting a 70B model captures 81% of explainable variance in subjective tasks, creating a strong baseline for individual-level simulation. Second, to address data scarcity, we introduce iNews, a large-scale dataset of personalized affective responses to news, enriched with persona information. Finally, we introduce SimBench, the first large-scale benchmark for group-level simulation, which reveals the strengths and critical weaknesses of current models. I conclude by arguing for the specialized datasets and training required to advance the frontier of high-fidelity human simulation.

The talk is available online. Please get in touch to ask for a Teams invite.

Add to your calendar or Include in your list

Wed 17 Sep 12:00: LLM Social Simulation is a Promising Research Method The talk is available online. Please get in touch to ask for a Teams invite.

Psychometrics Centre Seminars - Thu, 11/09/2025 - 10:44
LLM Social Simulation is a Promising Research Method

The emergence of large language models as in-silico subjects for social science poses a central question: can they genuinely simulate diverse human behavior, or do they merely produce plausible, homogenized artifacts? This talk demonstrates that LLMs are powerful but imperfect simulators by presenting three core contributions. First, we establish the “Persona Effect,” showing that persona-prompting a 70B model captures 81% of explainable variance in subjective tasks, creating a strong baseline for individual-level simulation. Second, to address data scarcity, we introduce iNews, a large-scale dataset of personalized affective responses to news, enriched with persona information. Finally, we introduce SimBench, the first large-scale benchmark for group-level simulation, which reveals the strengths and critical weaknesses of current models. I conclude by arguing for the specialized datasets and training required to advance the frontier of high-fidelity human simulation.

The talk is available online. Please get in touch to ask for a Teams invite.

Add to your calendar or Include in your list

Wed 17 Sep 12:00: LLM Social Simulation is a Promising Research Method The talk is available online. Please get in touch to ask for a Teams invite.

Psychometrics Seminars - Thu, 11/09/2025 - 10:44
LLM Social Simulation is a Promising Research Method

The emergence of large language models as in-silico subjects for social science poses a central question: can they genuinely simulate diverse human behavior, or do they merely produce plausible, homogenized artifacts? This talk demonstrates that LLMs are powerful but imperfect simulators by presenting three core contributions. First, we establish the “Persona Effect,” showing that persona-prompting a 70B model captures 81% of explainable variance in subjective tasks, creating a strong baseline for individual-level simulation. Second, to address data scarcity, we introduce iNews, a large-scale dataset of personalized affective responses to news, enriched with persona information. Finally, we introduce SimBench, the first large-scale benchmark for group-level simulation, which reveals the strengths and critical weaknesses of current models. I conclude by arguing for the specialized datasets and training required to advance the frontier of high-fidelity human simulation.

The talk is available online. Please get in touch to ask for a Teams invite.

Add to your calendar or Include in your list

Tue 26 Aug 16:00: Direct brain stimulation modulate physiology and behaviour in humans.

Other Psychology Seminars - Thu, 21/08/2025 - 11:48
Direct brain stimulation modulate physiology and behaviour in humans.

Direct brain electrical brain stimulation offers a causal window on human brain function and a route to therapy. This talk contrasts open-loop perturbation, single pulses and short trains used to map effective connectivity and state dependence across wake, sleep and anaesthesia, with closed-loop paradigms that detect neural features in intracranial EEG in real time to trigger stimulation. I’ll show how closed-loop experiments can adjust stimulation based on estimated cognitive state, producing rapid changes in cortical dynamics and measurable effects on behaviour during ongoing tasks. I’ll also discuss high-frequency oscillations and ripples as candidate biomarkers for targeting, and outline opportunities and limits for translating these methods to precision neuromodulation in humans.

The Host for this talk is Tristan Bekinschtein, tb419@cam.ac.uk.

Add to your calendar or Include in your list

Upcoming events

View all events

Study at Cambridge

About the University

Research at Cambridge