B2B Cognition & Mental Health Lab talks, with unintention twinning titles 🎭

Me: The promise and pitfalls of 📱/ real-life data
Alex: The promise and pitfalls of computational modelling 🧮

📍Centre for Decision Science Summer School
@alexandrapike.bsky.social

People higher in compulsivity made the most habit errors - suggesting stronger habit formation 🏋️‍♀️

… but so did younger adults, men & those with more education! Because habits aren’t 'bad'. They’re efficient ✅. Until they're not 🚧.

Introducing ... 🥁🥁🥁... COGQUEST

Training habits = lots of trials + multiple in-lab visits. Burdensome & inaccessible 🥱🚫 That’s why we gamified a habit task via @clairegillan.bsky.social's Neureka app

1000 people trained habits over a week, while also reporting on their mental health. All without stepping foot inside the lab!

I promise I missed the #CPConf2025 Travel Award photo because I was boarding my ✈️ and NOT due to punting 🛶 .. Grateful for the opportunity to present my work 📱, met up with friends & eat delicious german apple cake

Come to my fan-tastic poster at #CPConf2025 🪭 Poster 35 to find out all things habits, gamification or to get a quick cool down 🥵

& that’s a wrap! 💫 4 amazing years done at this wonderful lab 🧠📱

Ending on a high having passed my PhD-viva 👩🏻‍🎓 Endless thank yous to (AWOL) @clairegillan.bsky.social & the lab for all their support (& attention to detail 🎈💞🎀🌸) and to @tobiasuhauser.bsky.social & Sonia Bishop for examining

Finally (and a shameless plug) our workshop Game On got accepted for @rldmdublin2025.bsky.social, with a poster like this - do you really want to miss out!? More info found here: sites.google.com/view/game-on... GO SIGN UP!! 📱👾🎮

Finally coming down to earth after some very big and exciting academic/personal wins recently ✨☁️ …

First, last Friday I submitted my PhD thesis 🥳 which used 📱 games to explore the cognitive mechanisms underlying repetitive behaviours in psychiatry 🧠☄️⚙️ Yes, I did wear a tiara 👸🏻

our signal will get better with more data per person, right?

In contrast to the traditional, we could detect the association between model-based planning and compulsivity in as few as 25 trials and this association did not significantly change when estimated with 300 trials per person 🤯

.. but what can we do MORE with 📱's?

In a series of A/B style manipulations, we found we could boost the signal between model-based planning and compulsivity when transition ratios were less deterministic & for certain reward probability drifts ⏫

Then we replicated several prior associations between model-based planning and age, gender, education and specific clinical associations with compulsivity - adding more evidence to the validity of this gamified task 💎💎💎

But, does this gamified task detect similar behaviours to those measured by the traditional format? ... Looks like we can ✅

Cannon Blast embeds the 2-step RL task into a diamond-shooting game 💎💥 Players aim a cannon, choose a container to load from and 🤞 they hit the diamond. To make things tricky, the probability of drawing the selected colour is not fully deterministic & not all balls 'work' - some explode!

My first post has to be dedicated to my 1st study of the PhD, now published in Communications Psychology 🎉 We leveraged gamification, via a 💎-shooting game, and citizen science to scale-up the assessment of model-based planning & tested what task-related features could be used to boost signal 📈 1/n