Claire Gillan
@clairegillan.bsky.social
Professor in Psychology at Trinity College Dublin. Here for less gross twitter, less boring mastodon. www.gillanlab.com
Finally, Paveen Phon-Amnuaisuk will present unpublished data examining within-person fluctuations of metacognition over 8 weeks via... you guessed it... the neureka app again 📱
July 14, 2025 at 9:19 AM
Finally, Paveen Phon-Amnuaisuk will present unpublished data examining within-person fluctuations of metacognition over 8 weeks via... you guessed it... the neureka app again 📱
@glassybrain.bsky.social (Vanessa Teckentrup) will present "Understanding habitual properties of compulsivity using moment-to-moment experience sampling in daily life" with 6 weeks of EMA data tracking OCD, Binge-Eating and Phone-checking compulsions!📱https://osf.io/preprints/psyarxiv/yvhua_v1
July 14, 2025 at 9:19 AM
@glassybrain.bsky.social (Vanessa Teckentrup) will present "Understanding habitual properties of compulsivity using moment-to-moment experience sampling in daily life" with 6 weeks of EMA data tracking OCD, Binge-Eating and Phone-checking compulsions!📱https://osf.io/preprints/psyarxiv/yvhua_v1
@kellydonegan.bsky.social will present "Compulsivity is associated with an increase in stimulus- response habit learning"📱,which we are 🤏 close to pre-printing. So 🙏 give us your thoughts and ideas!
July 14, 2025 at 9:19 AM
@kellydonegan.bsky.social will present "Compulsivity is associated with an increase in stimulus- response habit learning"📱,which we are 🤏 close to pre-printing. So 🙏 give us your thoughts and ideas!
In our reply, we show that the FA solution is replicated in symptomatic patients with a different pattern of skew to m-turkers. We also show a near identical FA solution for two samples that differ in attentiveness... mturk vs. unpaid citizen scientists from the Neureka app (below).
July 1, 2025 at 8:52 AM
In our reply, we show that the FA solution is replicated in symptomatic patients with a different pattern of skew to m-turkers. We also show a near identical FA solution for two samples that differ in attentiveness... mturk vs. unpaid citizen scientists from the Neureka app (below).
This may seem at odds with the new data Sarna et al., gathered, which appeared to show that controlling for inattention and acquiescence eliminates the +association btwn confidence and OCD (p>.3). We found some issues with this analysis; when addressed the result is quite different (p=.009).
July 1, 2025 at 8:52 AM
This may seem at odds with the new data Sarna et al., gathered, which appeared to show that controlling for inattention and acquiescence eliminates the +association btwn confidence and OCD (p>.3). We found some issues with this analysis; when addressed the result is quite different (p=.009).
We also replicate the effect in unpaid citizen scientists who donate their time to research... with <1% inattention and who lack the signature of an acquiescence-confidence bias (below).
July 1, 2025 at 8:52 AM
We also replicate the effect in unpaid citizen scientists who donate their time to research... with <1% inattention and who lack the signature of an acquiescence-confidence bias (below).
We provide new analyses directly showing that in certain datasets (e.g. patients with different patterns of clinical skew) the +association with compulsivity and confidence is replicated and not affected by inattention.
July 1, 2025 at 8:52 AM
We provide new analyses directly showing that in certain datasets (e.g. patients with different patterns of clinical skew) the +association with compulsivity and confidence is replicated and not affected by inattention.
What drives this model? Self-report stuff does most of the lifting with good social support 👥 and positive expectations 🔮 about the treatment standing out. Cognitive tasks took a long time but v. little bang for buck (time) unfortunately
June 24, 2025 at 11:55 AM
What drives this model? Self-report stuff does most of the lifting with good social support 👥 and positive expectations 🔮 about the treatment standing out. Cognitive tasks took a long time but v. little bang for buck (time) unfortunately
When you do that - hey presto you have a treatment-specific model (see poor performance of 'final' model in antidepressant group on right - no better than a simple age, gender and baseline symptoms model).
June 24, 2025 at 11:55 AM
When you do that - hey presto you have a treatment-specific model (see poor performance of 'final' model in antidepressant group on right - no better than a simple age, gender and baseline symptoms model).
Our model performed OK .... r2=18.8% in held out iCBT and better than a 'baseline' model with age gender and depression. But what's cool (or not cool, depending) is the model generalised to N=110 who started antidepressants (r2 17.9%) and were recruited in a different way (compare dark blue bars)
June 24, 2025 at 11:55 AM
Our model performed OK .... r2=18.8% in held out iCBT and better than a 'baseline' model with age gender and depression. But what's cool (or not cool, depending) is the model generalised to N=110 who started antidepressants (r2 17.9%) and were recruited in a different way (compare dark blue bars)
We gathered a baseline assessment of cognitive and self-report variables over about 2 hours via the internet... from N=776 people within 2 days of starting iCBT.
June 24, 2025 at 11:55 AM
We gathered a baseline assessment of cognitive and self-report variables over about 2 hours via the internet... from N=776 people within 2 days of starting iCBT.
The story is of course complicated and nulls are nulls... so people might view the strength of the evidence differently. e.g. we found no effect in our patient groups, but a marginal effect was observed in our control group (though not when controlled for inattentive responding... p=.211)
June 18, 2025 at 1:16 PM
The story is of course complicated and nulls are nulls... so people might view the strength of the evidence differently. e.g. we found no effect in our patient groups, but a marginal effect was observed in our control group (though not when controlled for inattentive responding... p=.211)
In our preprint "Limited evidence for reduced learning rate adaptation in anxious-depression, before or after treatment", led by @stephsuddell.bsky.social and Lili Zhang, we fail to find a robust association between anxious-depression and learning rate adaptation
osf.io/preprints/ps...
osf.io/preprints/ps...
June 18, 2025 at 1:16 PM
In our preprint "Limited evidence for reduced learning rate adaptation in anxious-depression, before or after treatment", led by @stephsuddell.bsky.social and Lili Zhang, we fail to find a robust association between anxious-depression and learning rate adaptation
osf.io/preprints/ps...
osf.io/preprints/ps...
@kellydonegan.bsky.social showing our latest smartphone habit research this morning at @rldmdublin2025.bsky.social 📲
June 14, 2025 at 9:02 AM
@kellydonegan.bsky.social showing our latest smartphone habit research this morning at @rldmdublin2025.bsky.social 📲
In my room in college accom in cambridge 🆗😳… Surely some sort of test??
March 5, 2025 at 10:00 PM
In my room in college accom in cambridge 🆗😳… Surely some sort of test??
Congratulations Dr Anna Rosicka @annarosicka.bsky.social on passing your Viva today! Another amazing student flying the nest 😥🦉. Thanks to examiners Clare MacKay and Claire Howlin #AlltheClaires. Celebrating this afternoon with a very special hat from @glassybrain.bsky.social
January 28, 2025 at 3:19 PM
Congratulations Dr Anna Rosicka @annarosicka.bsky.social on passing your Viva today! Another amazing student flying the nest 😥🦉. Thanks to examiners Clare MacKay and Claire Howlin #AlltheClaires. Celebrating this afternoon with a very special hat from @glassybrain.bsky.social
Abstract submission is closing soon for the Interdisciplinary conference on Reinforcement Learning and Decision-Making #RLDM2025, taking place this June in Dublin - the first time the meeting will be in Europe! Have you submitted your abstract yet?? @rldmdublin2025.bsky.social
January 9, 2025 at 9:13 AM
Abstract submission is closing soon for the Interdisciplinary conference on Reinforcement Learning and Decision-Making #RLDM2025, taking place this June in Dublin - the first time the meeting will be in Europe! Have you submitted your abstract yet?? @rldmdublin2025.bsky.social
Hey 👋 the deadline for submitting workshop proposals to #RLDM is fast approaching! We would especially love creative structures for workshops, like hackathons, debates or other traditional formats. Follow @rldmdublin2025.bsky.social for more updates!
December 3, 2024 at 12:39 PM
Hey 👋 the deadline for submitting workshop proposals to #RLDM is fast approaching! We would especially love creative structures for workshops, like hackathons, debates or other traditional formats. Follow @rldmdublin2025.bsky.social for more updates!
Just back from a great visit to the Laureate Institute for Brain Research #LIBR in Tulsa. Thanks so much for all the great chats and esp. to mega-host (!) and recent Bluesky recruit @mariaironside.bsky.social
November 13, 2024 at 8:25 PM
Just back from a great visit to the Laureate Institute for Brain Research #LIBR in Tulsa. Thanks so much for all the great chats and esp. to mega-host (!) and recent Bluesky recruit @mariaironside.bsky.social
Out now in Alzheimer's & Dementia, @annarosicka.bsky.social uses the
@neurekaApp and data from thousands of citizen scientists to examine how well-established ricks factors relate to subjective versus objective cognitive ability across the lifespan. doi.org/10.1002/alz....
@neurekaApp and data from thousands of citizen scientists to examine how well-established ricks factors relate to subjective versus objective cognitive ability across the lifespan. doi.org/10.1002/alz....
October 16, 2024 at 1:46 PM
Out now in Alzheimer's & Dementia, @annarosicka.bsky.social uses the
@neurekaApp and data from thousands of citizen scientists to examine how well-established ricks factors relate to subjective versus objective cognitive ability across the lifespan. doi.org/10.1002/alz....
@neurekaApp and data from thousands of citizen scientists to examine how well-established ricks factors relate to subjective versus objective cognitive ability across the lifespan. doi.org/10.1002/alz....
I have been so blind
October 1, 2024 at 8:01 PM
I have been so blind
Congratulations to newly minted Dr Fox! @celinef.bsky.social absolutely knocked it out of the park yesterday in her viva - no corrections on her work on metacognition in psychiatry! Thanks to examiners @redmondoconnell.bsky.social and @micahgallen.com 🍾🦊
September 7, 2024 at 10:11 AM
Congratulations to newly minted Dr Fox! @celinef.bsky.social absolutely knocked it out of the park yesterday in her viva - no corrections on her work on metacognition in psychiatry! Thanks to examiners @redmondoconnell.bsky.social and @micahgallen.com 🍾🦊
Last thing we did was compare this unsupervised method to a supervised one - partial least squares regression, which defines the factor in part *based* on the cognitive test scores. The results were no better than for factors derived from a bog-standard EFA on items alone… 🤯
August 1, 2024 at 1:33 PM
Last thing we did was compare this unsupervised method to a supervised one - partial least squares regression, which defines the factor in part *based* on the cognitive test scores. The results were no better than for factors derived from a bog-standard EFA on items alone… 🤯
For those interested in hierarchical approaches - we also compared bifactor models (with a p-factor) to first order rotations. An orthogonal rotation did best and the p-factor was not linked to cognitive performance at all...
August 1, 2024 at 1:31 PM
For those interested in hierarchical approaches - we also compared bifactor models (with a p-factor) to first order rotations. An orthogonal rotation did best and the p-factor was not linked to cognitive performance at all...
The result - the 3 factor solution performs - what can only be described as *freakishly* well. For example, compulsivity and Intrusive thought circa 2016 was in 9th place of all 7665 in terms of its association to model-based planning.
August 1, 2024 at 1:31 PM
The result - the 3 factor solution performs - what can only be described as *freakishly* well. For example, compulsivity and Intrusive thought circa 2016 was in 9th place of all 7665 in terms of its association to model-based planning.