iiinova — where your digital self keeps evolving

Melissa Franch, PhD @mfranch.bsky.social · 20/03/2026

I have a personal connection, as my brother Stefan has profound autism and is nonverbal. LLMs generate fluent language, but many people like him still lack reliable ways to communicate. I hope these findings one day improve neural interventions for communication in ASD.

1102

Melissa Franch, PhD @mfranch.bsky.social · 20/03/2026

Why care about dimensionality? The dimensionality of the neural subspace determines the brain's capacity to represent semantic information. We think autism may be associated with a bias towards low dimensional neural subspaces for semantic coding.

161

Melissa Franch, PhD @mfranch.bsky.social · 20/03/2026

Said another way, in autistic patients, neural population activity is dominated by a smaller number of shared latent signals, causing neurons to vary more together rather than independently. Reduced dimensionality was also associated with reduced semantic decoding in 2/3 ASD patients.

151

Melissa Franch, PhD @mfranch.bsky.social · 20/03/2026

Consistent with this, all autistic patients exhibited lower effective dimensionality in the neural subspace predictive of semantics, indicating fewer independent coding axes.

181

Melissa Franch, PhD @mfranch.bsky.social · 20/03/2026

While single neurons in all patients showed prediction error to the most surprising words, autistic patients showed reduced population-level decoding of word surprisal, which may arise from their altered weighting of contextual information.

181

Melissa Franch, PhD @mfranch.bsky.social · 20/03/2026

Neural weighting of prior context was systematically altered in autistic patients, reflected in a broader distribution of weights. This broader distribution of prior word weighting in autism resembles the proposed weaker, broad distribution of priors in Bayesian theories for ASD.

1101

Melissa Franch, PhD @mfranch.bsky.social · 20/03/2026

Both control and ASD patients had comparable single neuron encoding of semantics, representations of semantic relationships between words, and neural flexibility in semantic representations as word meanings change with context.

161

Melissa Franch, PhD @mfranch.bsky.social · 20/03/2026

I’m excited to share my newest work with @benhayden.bsky.social, and the work I’m most proud of to date, on characterizing semantic coding in single-neuron hippocampal activity in patients with autism during natural language comprehension! www.biorxiv.org/content/10.6...

29531

Melissa Franch, PhD @mfranch.bsky.social · 21/11/2025

Besides the awesome neuroscience I learned at HSN and @sfn.org, I was reminded of my incredible support system and the many inspiring women in neuroscience who currently guide and influence me! @nicolecrust.bsky.social @libertysays.bsky.social @storiesofwin.bsky.social to name a few.

2286

Melissa Franch, PhD @mfranch.bsky.social · 12/11/2025

If you're attending HSN or #SfN2025 #SfN25 be sure to come by my poster to learn about my new work, "Neural signatures of impaired semantic contextualization in autism." SFN info: Board KK2, Sun Nov 16. 13-1700PM

0152

Melissa Franch, PhD @mfranch.bsky.social · 11/09/2025

Really looking forward to presenting my new research on semantic coding in single neurons in autism this Sunday at the Society for Neurobiology of Language conference! Be sure to check out all the presentations from the Hayden lab! @benhayden.bsky.social @snlmtg.bsky.social

071

Melissa Franch, PhD @mfranch.bsky.social · 27/02/2025

Similarly, a neuron may be selective to a specific meaning of a word, and therefore only fire when the word is used in that context, resulting in sparse coding.

161

Melissa Franch, PhD @mfranch.bsky.social · 27/02/2025

Contextual LLMs and the brain take this into account. Since neurons and LLMs are contextual coders, their distances correlate with word polysemy, causing the greatest changes in neural activity between high frequency and thus more polysemous words (such as function words, actions, and pronouns).

142

Melissa Franch, PhD @mfranch.bsky.social · 27/02/2025

How does the brain represent different occurrences of the same word? Neural activity varies greatly for different instances of the same word. Words that are more frequently used in the English language are more polysemous, meaning they can mean many different things, depending on their context.

141

Melissa Franch, PhD @mfranch.bsky.social · 27/02/2025

reflecting a negative correlation to LLMs only for very similar word pairs. This contrastive coding likely helps prevent confusion for different words with very similar meanings.

151

Melissa Franch, PhD @mfranch.bsky.social · 27/02/2025

After comparing the angles between neural vectors and model vectors for all word pairs, we found that the neural encoding of semantics is highly contextual and positively correlates only with LLMs that have contextual embeddings (GPT, BERT).

1114

Melissa Franch, PhD @mfranch.bsky.social · 27/02/2025

We wanted to know if the brain also employed these coding principles. For each word, its neural embedding vector comprised the firing rate response of each neuron for that word.

162

Melissa Franch, PhD @mfranch.bsky.social · 27/02/2025

When considering words related to the concept of family, of the 28 neurons encoding "father" and 19 encoding "dad," only 7 neurons encode both words. Each neuron encoded a median of 6 words and their words spanned a median of 4 semantic categories (out of 11).

1111

Melissa Franch, PhD @mfranch.bsky.social · 27/02/2025

‘Hit’ was the most densely coded word at 67 neurons and ‘was’ and ‘and’ were the most sparse at 3 neurons. Even words coding specific concepts like ‘God’ were densely coded (60 neurons).

1141

Melissa Franch, PhD @mfranch.bsky.social · 27/02/2025

We recorded the activity of hundreds of neurons in the human hippocampus while people listened to stories. We found a dense and distributed code for semantics based on comparing each neuron's firing rate across words. A single word was encoded by many neurons and each neuron encoded many words.

1151

Melissa Franch, PhD @mfranch.bsky.social · 27/02/2025

I am incredibly proud to share my first, first-author paper as a postdoc with @benhayden.bsky.social . How does the human hippocampus, known for encoding concepts, represent the meanings of words while listening to narrative speech? www.biorxiv.org/content/10.1...

1231280

Melissa Franch, PhD @mfranch.bsky.social · 03/02/2025

Highly recommend this book for learning how to be a more empathetic, compassionate, and meaningful person in your community and workplace, especially the lab! Ruchika provides practical, actionable strategies to enhance inclusivity and belonging in science.

041

Melissa Franch, PhD @mfranch.bsky.social · 21/01/2025

Finally cold enough to wear a cowl in Houston! ❄️☃️

040