Current and Past Projects

Can a chat change the way our brains respond to the world?

Watching a film with someone can feel different from watching it alone. But what actually happens in our brains when we co-watch something — and does social interaction beforehand make a difference? In this study, I explored how face-to-face conversation can tune people’s brains to respond more similarly, even during a quiet, shared activity like watching a cartoon.

The question

Can social interaction — even one unrelated to the experience — shape brain-to-brain synchrony in the moments that follow?

What we did

We invited familiar pairs (friends, flatmates, partners) into the lab. Using fNIRS hyperscanning, we recorded brain activity from both participants simultaneously as they:

• Co-watched a cartoon episode (BBC’s Dipdap) side by side

• Engaged in a 20-minute face-to-face conversation on unrelated topics

• Watched a new cartoon episode together

By comparing the synchrony in brain signals across these phases, and contrasting real pairs with pseudo pairs (constructed from recordings of people who did not interact), we could isolate the effect of shared presence and conversation.

What we found

• Brain synchrony was stronger in real pairs (people who watched the cartoon together) than pseudo pairs (randomly shuffled pairs who never actually met), especially over the right dorsolateral prefrontal cortex (dlPFC) and superior parietal lobe (SPL) — areas involved in attention and perspective-taking.

• After the conversation, synchrony further increased — particularly over the right temporo-parietal junction (TPJ), a region known for social understanding and mentalising.

Notably, the cartoon episodes were non-verbal and unrelated to the chat. This shows that it’s not what you talk about, but the act of connecting itself that prepares our brains to see the world in more aligned ways.

Why it matters

Our results suggest that sharing an experience together - even when there is little interaction as when watching a movie - synchronises our brains. In addition, this is one of the first studies to show that brief social interaction can leave a trace in how our brains process later, unrelated experiences. It underscores the power of conversation in shaping not just our thoughts — but also in aligning our underlying neural dynamics and the way we process the world.

Rethinking assumptions about autism and learning
It’s often assumed that autistic people find social interaction effortful and therefore might not learn effectively in social contexts. In this project, I asked a different question: not how autistic people learn about the social world, but whether they can learn through social interaction just as well as neurotypical people.

The question
Can autistic adults benefit from interactive learning to the same extent as neurotypical adults? And how do social factors like enjoyment or anxiety shape this experience?

What we did
We ran two large online experiments during the post-pandemic shift to remote learning. Participants — half autistic, half neurotypical — learned about obscure real-world items (like musical instruments or ancient tools) in three different formats:

• A live Zoom session with a teacher

• A recorded video showing the teacher interacting with someone else

• A recorded video of the teacher alone

Everyone took a quiz right after learning, and again a week later. We also measured their enjoyment, anxiety, and cognitive traits like verbal fluency and mentalising.

What we found
Live interaction led to the best learning — for both autistic and neurotypical adults. Surprisingly, autistic participants reported just as much enjoyment, and no more anxiety, as the neurotypical group. Even more importantly, the advantage of live learning was equally strong in both groups, and did not fade over time.

Why it matters
Our results challenge common narratives about autism and social learning. They suggest that while autistic people may process social signals differently, they are no less capable of benefiting from social interaction when learning factual content. This study argues for a critical distinction: learning about social signals versus learning via social interaction. While autistic adults may find it difficult to decode subtle social cues, they can still thrive in interactive learning settings when the social context is structured and meaningful. Our results highlight the need to move beyond deficit-focused models of autism and reimagine inclusive pedagogical approaches that embrace diverse learners.

What happens when we learn together — even through a screen? 

It was 2019 and the Covid-19 pandemic had drastically transformed our everyday life. While much of schooling, training and work meetings had moved online, little was known about whether live teaching—where the student can interact with the teacher—actually supports better learning than simply watching a recording.This project explored how the social nature of a learning interaction can transform the way we acquire new knowledge. 

The question

Can social interaction enhance learning in online contexts—and does it matter whether we see the teacher’s face?

What we did

During the height of the COVID-19 pandemic, when most teaching was taking place remotely, we ran a series of tightly controlled Zoom experiments. Young adults were taught obscure facts about unfamiliar objects under different conditions:

• Some learned via a live video call, where they could see and speak with the teacher.

• Others watched a recording of the same session, with no interaction.

• We also varied what participants could see: in some cases the teacher’s face and hands, in others just slides or hands only. 

• Learning was tested twice—immediately, and again one week later.

What we found

Live, interactive teaching consistently led to better memory for the material—especially when the teacher’s face was visible. But in pre-recorded sessions, the opposite was true: seeing a face became distracting, and slides actually helped people learn better.

These results suggest that interaction changes how we process social cues. Rich visual input (like a human face) helps only when the learner is engaged in a real-time exchange.

Why it matters

These findings are among the first to demonstrate that true social interaction—contingent, bi-directional, and real-time—can directly enhance adult learning. They challenge assumptions in educational technology and suggest that building more interactive tools might be more effective than simply adding video content.