Vanessa Cheung (@vanessachg_) header
Vanessa Cheung (@vanessachg_) profile picture

PhD student @EP_UCL | cognitive science, moral & legal decision-making | πŸ‡­πŸ‡°

Joined April 2018
Photo by Vanessa Cheung (@vanessachg_)Photo by Vanessa Cheung (@vanessachg_)Photo by Vanessa Cheung (@vanessachg_)Photo by Vanessa Cheung (@vanessachg_)Photo by Vanessa Cheung (@vanessachg_)
Excited to share that our paper is now published in PNAS! pnas.org/doi/10.1073/pnas.24…
We updated this preprint (osf.io/preprints/psyarxiv/aj…) with a new study investigating sources of this yes-no framing bias and amplified omission bias in LLM moral decision-making. Results suggest that they likely arise from fine-tuning for chatbot applications. @MaxMa1er @FalkLieder
4
6
959
In this article, Maier et al. show that metacognitive learning from consequences can shape moral decision-making. @MaxMa1er @vanessachg_ @FalkLieder @RtnlAltruismLab nature.com/articles/s41562-0…
5
18
1,876
Vanessa Cheung retweeted
New preprint w/ @MaxMa1er, @vanessachg_, Fiery Cushman, & @FalkLieder: Disentangling Model-Based and Model-Free Moral Learning doi.org/10.31234/osf.io/azhn… Excited to present this work at #CogSci2025!
1
6
8
609
Suppose Person A orders Person B to do something harmful, and Person B complies. Who is then seen as more of a cause of the harm - Person A or Person B? Intriguing new research from @vanessachg_ and @david_lagnado osf.io/preprints/psyarxiv/86…
6
13
1,495
We updated this preprint (osf.io/preprints/psyarxiv/aj…) with a new study investigating sources of this yes-no framing bias and amplified omission bias in LLM moral decision-making. Results suggest that they likely arise from fine-tuning for chatbot applications. @MaxMa1er @FalkLieder
New preprint with @MaxMa1er and @FalkLieder, titled "Large Language Models Amplify Human Biases in Moral Decision-Making." In this paper, LLMs and participants make decisions in realistic moral dilemmas. We found that LLMs exhibited stronger omission bias than participants.
6
1
9
2,459
We updated our preprint on moral learning, now titled "Metacognitive Learning from Consequences of Past Choices Shapes Moral Decision-Making," with three new experiments and computational modeling to capture differences in individual learning styles. @vanessachg_ @FalkLieder
1
6
12
1,183
New preprint with @MaxMa1er and @FalkLieder, titled "Large Language Models Amplify Human Biases in Moral Decision-Making." In this paper, LLMs and participants make decisions in realistic moral dilemmas. We found that LLMs exhibited stronger omission bias than participants.
2
16
1
44
8,167
Finally, we also looked at decisions in social dilemmas that pitted self-interest against the greater good. Here, the LLMs were, on average, more altruistic than participants.
1
252
New Cogsci preprint, "Introducing the Extinction Gambling Task," with Adam Harris, David Kellen, and @HenrikSingmann. We introduce a risky choice task in which participants can lose their (entire) bonus pay when choosing the risky option and drawing the extinction event.
1
7
2
16
5,289
New preprint with Mengxuan Qiao and @david_lagnado: "Attribution of Responsibility Between Agents in a Causal Chain of Events," which was accepted as a talk at CogSci 2024! osf.io/preprints/psyarxiv/86…
1
6
1
17
2,227
In this paper, we explored how people attribute causal responsibility in a causal chain of events, where an agent A instructs an intermediate agent B to harm a victim. Do laypeople consider A or B to be more responsible of this crime?
1
1
167
We find that people assigned more responsibilty to the agent who executed the act (B). Their moral judgments were somewhat consistent with what the law presumes: relatively unforeseeable, more proximal causes weaken, but do not break, the chain of causation.
2
174
πŸŽ‰ We are delighted to announce that our abstract got accepted by #SPSP2024 pre-conference on Moral Psychology. spsp.org/events/annual-conve… Malleable Morality: Investigating Moral Learning and Its Mechanisms 🧡 (@MaxMa1er, @vanessachg_ , @FalkLieder) dx.doi.org/10.13140/RG.2.2.1…
1
4
6
759
Happy to share our manuscript, "A reinforcement-learning meta-control architecture based on the dual-process theory of moral decision-making", which was accepted to the NeurIps workshop "AI meets Moral Philosophy and Moral Psychology"! (with @vanessachg_ , @FalkLieder )
1
6
17
2,587
New preprint with @vanessachg_, @BartosFra, & @FalkLieder "Learning from Consequences Shapes Reliance on Moral Rules vs. Cost-Benefit Reasoning". Using realistic moral dilemmas with outcomes, we show that moral decision-making is shaped by experience (doi.org/10.31234/osf.io/gjf3…).
1
15
2
48
9,541
Suppose you do something bad. Will people blame you more if you knew it was wrong? Or will they blame you less? The answer seems to be: They will think your act is more wrong, but your character is less bad @yorl sciencedirect.com/science/ar…
2
21
1
72
20,756
New CogSci proceedings preprint: "Defendant Character Influences Mock Juror Judgments of Blame, Guilt, and Punishment" with @david_lagnado ! psyarxiv.com/69jh7
6
1
15
1,373
New paper w/ @david_lagnado now out in Cognition! sciencedirect.com/science/ar… Here, we assess how competing explanations of the same evidence are structured, evaluated and compared in the context of legal scenarios.
3
11
2
63
17,532