Hadar Fisher
@hadarfisher.bsky.social
65 followers 72 following 36 posts
Instructor in Psychiatry | Harvard Medical School | McLean Hospital
Posts Media Videos Starter Packs
hadarfisher.bsky.social
My hope is that, with the right boundaries, knowledge, and caution, and with the incredible work of so many researchers in the field, AI can advance our long-standing efforts to close the urgent gap in mental health access.
hadarfisher.bsky.social
I argue that we can’t stop this train, but as clinicians and researchers, we can and *should* shape its track. Rather than resist change, we should be more active in guiding it, wisely, ethically, and collaboratively (with the developers and the users).
hadarfisher.bsky.social
In this letter, I respond to Ziv Ben-Zion’s important World View column, which recommended ways to stop, or at least slow down, the use of AI (such as ChatGPT) for emotional support.
👉Ben Zion’s world View column: www.nature.com/articles/d41...
Why we need mandatory safeguards for emotionally responsive AI
Virtual chatbots that simulate conversations with famous actors or sci-fi characters can have real-world consequences.
www.nature.com
hadarfisher.bsky.social
My correspondence on emotionally responsive AI is now published in Nature (!)
I’m thankful for the opportunity to share this little drop of thought. 🌱
👉 Read the full correspondence here: www.nature.com/articles/d41...
Emotional AI is here — let’s shape it, not shun it
Letter to the Editor
www.nature.com
hadarfisher.bsky.social
Thanks so much for tagging me, Christian! This sounds like an absolutely amazing dataset. @eikofried.bsky.social I’m traveling now with little reception (and lots of kids’ noise 😅), but once I’m somewhere quiet I’ll send an email with my thoughts on de-identification and some collab suggestions.
hadarfisher.bsky.social
How emotions unfold over time, their flexibility and adaptability, may be just as important as what people feel for understanding depression vulnerability.
This opens new doors for early identification and prevention.
hadarfisher.bsky.social
📉 The results: Participants with more rigid emotional systems were significantly more likely to develop depressive symptoms later, even after controlling for risk factors, sex, emotional intensity, and variability.

This was specific to depression! Emotion rigidity didn't predict anxiety symptoms.
hadarfisher.bsky.social
We followed adolescents without depression and had them report emotions 4x daily for a month. Then we tracked depressive symptoms for 2 years.
We used dynamic systems methods to build individual emotion networks and calculated emotional rigidity (how densely interconnected emotional states were).
hadarfisher.bsky.social
We know that emotions help us make sense of the world, guide decisions, and navigate (social) life. But what happens when emotions become too rigid to flexibly adapting to changes?
In this study, we asked whether emotion rigidity might precede and predict the onset of depressive symptoms.
hadarfisher.bsky.social
🚨New paper on emotion rigidity as a risk factor for future depression out in BRAT!
Thank you to @christianwebb.bsky.social @diegopizzagalli.bsky.social and the awesome teddylab team @nigeljaffe.bsky.social, KristinaPidvirny, and Anna Tierney 🙌

Free 50d access: authors.elsevier.com/a/1lRqe1KMdl...
authors.elsevier.com
hadarfisher.bsky.social
Check out our thoughts on the growing use of passive sensing for digital phenotyping. In this commentary, we reflect on excellent work by @whitneyringwald.bsky.social et al., who use a large, clinically heterogeneous sample to examine assoc. bw passive sensing and domains of psychopathology.
hadarfisher.bsky.social
📣 Curious about how to measure and work with emotions in therapy? Join us for the panel Emotional Processes in Psychotherapy: Breaking New Ground, today at 11:30 in Aula A (LB). New tools, task-based methods, and fresh data on emotional change. #SPRKrakow2025
hadarfisher.bsky.social
This means we can use digital tools to provide therapists with objective, real-time measures of therapeutic progress, potentially personalizing treatments and improving outcome.
hadarfisher.bsky.social
While LLM ratings predicted immediate daily emotional changes (more positive, less negative affect), passive sensors (places visited (↑), time at home(↓))predicted weekly improvements in depression and anhedonia.
hadarfisher.bsky.social
Both LLM-derived activation (daily texts) and passive sensing features (places visited ↑, time at home ↓) correlated significantly with self-reported activation —supporting their convergent validity.
hadarfisher.bsky.social
Passive sensing features (higher activity scores (↑), more places visited (↑) less time spent at home (↓)) correlated with daily LLM-rated activation.
hadarfisher.bsky.social
In this study, we tracked 38 adolescents undergoing BA therapy over 12 weeks. We measured daily activation from free-text entries using an LLM (GPT-4o) and from movement/location patterns using smartphone sensors. 📲

Key Findings:
hadarfisher.bsky.social
Thrilled to collaborate with the dream team: @nigeljaffe.bsky.social , @christianwebb.bsky.social, Habiballah Rahimi-Eichi, Erika Forbes, @diegopizzagalli.bsky.social , Justin Baker
hadarfisher.bsky.social
Can we track therapeutic change outside the clinic?
In a new preprint, we explored whether digital tools - smartphone passive sensors📱 and LLM (GPT-4o) 🤖 - can track daily-life activation in adolescents with anhedonia receiving Behavioral Activation.

osf.io/preprints/ps...
hadarfisher.bsky.social
🚨 Now out in Emotion! 🚨
📄 psycnet.apa.org/record/2026-...
Grateful to all collaborators 🙏
Reposted by Hadar Fisher
toddrjones.bsky.social
Here are some ways the world has gotten better.
hadarfisher.bsky.social
7/7 Want to see it for yourself?
Make sure to explore our interactive companion app to try the models, view feature importance, and ask our GPT-powered chatbot about the study:
👉 emotracknlp.streamlit.app
ML Analysis Dashboard
This repository contains the source code for the Streamlit web application, which explores whethe...
emotracknlp.streamlit.app