Sign in

Romane Cecchi

@romanececchi.bsky.social
154 followers 71 following 11 posts

Postdoc in the Human Reinforcement Learning team led by @stepalminteri.bsky.social at École Normale Supérieure (ENS) in Paris ✨

PostsRepliesMedia
Reposted by Romane Cecchi
Stefano Palminteri @stepalminteri.bsky.social · 23/06/2026
New paper in Nature Communications 🎉 Led by @romanececchi.bsky.social, with @sgluth.bsky.social, we show that attention shapes value normalization in human reinforcement learning. Eye-tracking models explain nonlinear reward scaling better than an ad hoc parameter. www.nature.com/articles/s41...
nature.com
Attention modulates value normalization in human reinforcement learning by shaping reward encoding - Nature Communications
People’s reward learning is shaped by context, but the cognitive origins of this bias remain poorly understood. Here, the authors provide evidence that attention may underlie these distortions by shap...
03112
Reposted by Romane Cecchi
Stefano Palminteri @stepalminteri.bsky.social · 27/05/2026
My lab will be present en masse at #sbdm2026 Paris. Here is a first sample of the posters, presented by @romanececchi.bsky.social "Dynamic range adaptation in vast decision spaces" and @fabiencerrotti.bsky.social "Correcting mis-conceptions and shaping preferences about energy sources with RL"
0143
Reposted by Romane Cecchi
Stefano Palminteri @stepalminteri.bsky.social · 03/02/2026
Very happy that @PNASNews agreed to publish our (w/ @romanececchi.bsky.social) response to Prakhar's thought-provoking study! You can find the final version at the link below. See the following tweet for Prakhar's response to our response. Happy to hear your thoughts! www.pnas.org/doi/10.1073/...
pnas.org
Genuine learning biases persist after accounting for temporally decreasing learning rates: Insight from fitting six datasets | PNAS
Genuine learning biases persist after accounting for temporally decreasing learning rates: Insight from fitting six datasets
1196
Romane Cecchi @romanececchi.bsky.social · 30/04/2025
Thanks for the kind mention @nclairis.bsky.social ! And congrats to the authors – really interesting to see mood being explored across species
020
Romane Cecchi @romanececchi.bsky.social · 22/04/2025
9/9 🙌 Huge thanks to @sgluth.bsky.social & @stepalminteri.bsky.social! 📄 Read the full preprint: doi.org/10.31234/osf... 💬 Feedback and discussion welcome! #reinforcementlearning #attention #computationalneuroscience #eyetracking 🧵/end
doi.org
OSF
040
Romane Cecchi @romanececchi.bsky.social · 22/04/2025
8/9 🎯 Conclusion Attention doesn’t just follow value – it shapes it. We showed that: 👁️ Gaze during learning causally biases value encoding ⏱️ Stimulus-directed attention sets the stage for processing the outcomes 🧩 Our model offers a mechanistic account of why mid-value options are undervalued in RL
110
Romane Cecchi @romanececchi.bsky.social · 22/04/2025
7/9 ⏱️ When does attention matter? 🏆 Best fit? The stimulus-only model – consistent across experiments. ➕ Complemented by fine-grained gaze analysis: 👉 Value computation relies mainly on fixations during stimulus presentation, with minimal contribution from outcome-related fixations.
110
Romane Cecchi @romanececchi.bsky.social · 22/04/2025
6/9 🧩 Modeling attention in value learning We formalized these findings in an attentional range model, where visual fixations modulate absolute value before range normalization. We tested 3 versions using: • Only stimulus fixations • Only outcome fixations • A weighted combination
130
Romane Cecchi @romanececchi.bsky.social · 22/04/2025
5/9 ⬆️ Exp 2 & 3: Bottom-up manipulation We used saliency to guide gaze to the mid-value option: • Exp 2: Salient stimulus → higher valuation in transfer • Exp 3: Salient outcome → no significant effect 👉 Only attention to stimuli during learning influenced value formation
110
Romane Cecchi @romanececchi.bsky.social · 22/04/2025
4/9 ⬇️ Exp 1: Top-down manipulation We made the high-value option unavailable on some trials – forcing attention to the mid-value one. Result: participants looked more at the mid-value option – and later valued it more. 👉 Attention during learning altered subjective valuation.
110
Romane Cecchi @romanececchi.bsky.social · 22/04/2025
3/9 Design We ran 3️⃣ eye-tracking RL experiments, combining: • A 3-option learning phase with full feedback • A transfer test probing generalization & subjective valuation Crucially, we manipulated attention via: • Top-down control (Exp 1) • Bottom-up saliency (Exp 2 & 3)
110
Romane Cecchi @romanececchi.bsky.social · 22/04/2025
2/9 💡 Hypothesis Could attention be the missing piece? Inspired by the work of @krajbichlab.bsky.social, @yaelniv.bsky.social, @thorstenpachur.bsky.social and others, we asked: 👉 Does where we look during learning causally shape how we encode value?
120
Romane Cecchi @romanececchi.bsky.social · 22/04/2025
1/9 🎨 Background When choosing, people don’t evaluate options in isolation – they normalize values to context. This holds in RL... but in three-option settings, people undervalue the mid-value option – something prior models fail to explain (see @sophiebavard.bsky.social). ❓Why the distortion?
110
Romane Cecchi @romanececchi.bsky.social · 22/04/2025
🧵 New preprint out! 📄 "Elucidating attentional mechanisms underlying value normalization in human reinforcement learning" 👁️ We show that visual attention during learning causally shapes how values are encoded w/ @sgluth.bsky.social & @stepalminteri.bsky.social 🔗 doi.org/10.31234/osf...
doi.org
OSF
1176
Reposted by Romane Cecchi
Stefano Palminteri @stepalminteri.bsky.social · 23/01/2025
Epistemic biases in human reinforcement learning: behavioral evidence, computational characterization, normative status and possible applications. A quite self-centered review, but with a broad introduction and conclusions and very cool figures. Few main takes will follow osf.io/preprints/ps...
13819
Reposted by Romane Cecchi
Maëva L'Hôtellier @maevalhotellier.bsky.social · 10/12/2024
New preprint! 🚨 Performance of standard reinforcement learning (RL) algorithms depends on the scale of the rewards they aim to maximize. Inspired by human cognitive processes, we leverage a cognitive bias to develop scale-invariant RL algorithms: reward range normalization. Curious? Have a read!👇
1147
Reposted by Romane Cecchi
Stefano Palminteri @stepalminteri.bsky.social · 05/12/2024
🚨New preprint alert!🚨 Achieving Scale-Invariant Reinforcement Learning Performance with Reward Range Normalization. Where we show that things we discover in psychology can be useful for machine learning. By the amazing @maevalhotellier.bsky.social and Jeremy Perez. doi.org/10.31234/osf...
osf.io
OSF
12310