Nicholas A. Coles
@nicholascoles.bsky.social
Assistant Professor at the University of Florida
Big Team Science | Computational and Quantitative Methods | Emotion
Big Team Science | Computational and Quantitative Methods | Emotion
E.g.,
Paper A uses a specific dataset, architecture, and approach to training/testing/evaluation
Paper B makes totally different decision about each one of these issues
And this leaves us with puzzle pieces that can’t even really be connected 🧩
Paper A uses a specific dataset, architecture, and approach to training/testing/evaluation
Paper B makes totally different decision about each one of these issues
And this leaves us with puzzle pieces that can’t even really be connected 🧩
June 25, 2025 at 8:51 AM
E.g.,
Paper A uses a specific dataset, architecture, and approach to training/testing/evaluation
Paper B makes totally different decision about each one of these issues
And this leaves us with puzzle pieces that can’t even really be connected 🧩
Paper A uses a specific dataset, architecture, and approach to training/testing/evaluation
Paper B makes totally different decision about each one of these issues
And this leaves us with puzzle pieces that can’t even really be connected 🧩
Things are more erratic when you look at all models
Inferences about the accuracy and theoretical implications of machine learning efforts depended not only on their architecture, but also how they were trained, tested, and evaluated 😵💫
Inferences about the accuracy and theoretical implications of machine learning efforts depended not only on their architecture, but also how they were trained, tested, and evaluated 😵💫
June 25, 2025 at 8:51 AM
Things are more erratic when you look at all models
Inferences about the accuracy and theoretical implications of machine learning efforts depended not only on their architecture, but also how they were trained, tested, and evaluated 😵💫
Inferences about the accuracy and theoretical implications of machine learning efforts depended not only on their architecture, but also how they were trained, tested, and evaluated 😵💫
Some models were very inconsistent with notions of universality [e.g., see purple line]
They performed much better when tested on the same [black arrow] vs. different [grey arrow] people
But other models yielded the exact opposite conclusion [see red line] 😂
They performed much better when tested on the same [black arrow] vs. different [grey arrow] people
But other models yielded the exact opposite conclusion [see red line] 😂
June 25, 2025 at 8:51 AM
Some models were very inconsistent with notions of universality [e.g., see purple line]
They performed much better when tested on the same [black arrow] vs. different [grey arrow] people
But other models yielded the exact opposite conclusion [see red line] 😂
They performed much better when tested on the same [black arrow] vs. different [grey arrow] people
But other models yielded the exact opposite conclusion [see red line] 😂
If machine learning can uniquely capture links between physiology and emotion, perhaps it can tackle difficult theoretical questions
E.g., are these links universal?
Perhaps that could be tested by seeing how well models perform when tested on the same (vs. different) people
E.g., are these links universal?
Perhaps that could be tested by seeing how well models perform when tested on the same (vs. different) people
June 25, 2025 at 8:51 AM
If machine learning can uniquely capture links between physiology and emotion, perhaps it can tackle difficult theoretical questions
E.g., are these links universal?
Perhaps that could be tested by seeing how well models perform when tested on the same (vs. different) people
E.g., are these links universal?
Perhaps that could be tested by seeing how well models perform when tested on the same (vs. different) people
𝗢𝘂𝗿 𝗯𝗶𝗴 𝘁𝗲𝗮𝗺 𝗮𝗽𝗽𝗿𝗼𝗮𝗰𝗵
12 machine learning teams competed to predict affect using multiple measures of peripheral nervous system activity (e.g. heart rate)
We tested the models in 4 ways & made everything openly available
12 machine learning teams competed to predict affect using multiple measures of peripheral nervous system activity (e.g. heart rate)
We tested the models in 4 ways & made everything openly available
June 25, 2025 at 8:51 AM
𝗢𝘂𝗿 𝗯𝗶𝗴 𝘁𝗲𝗮𝗺 𝗮𝗽𝗽𝗿𝗼𝗮𝗰𝗵
12 machine learning teams competed to predict affect using multiple measures of peripheral nervous system activity (e.g. heart rate)
We tested the models in 4 ways & made everything openly available
12 machine learning teams competed to predict affect using multiple measures of peripheral nervous system activity (e.g. heart rate)
We tested the models in 4 ways & made everything openly available
𝗦𝘂𝗺𝗺𝗮𝗿𝘆
12 teams competed to predict affective experience using multiple measures of peripheral nervous system activity (e.g., heart rate)
Many models appeared to have captured something interesting
But attempts to link the methods to emotion theory still seem premature
12 teams competed to predict affective experience using multiple measures of peripheral nervous system activity (e.g., heart rate)
Many models appeared to have captured something interesting
But attempts to link the methods to emotion theory still seem premature
June 25, 2025 at 8:51 AM
𝗦𝘂𝗺𝗺𝗮𝗿𝘆
12 teams competed to predict affective experience using multiple measures of peripheral nervous system activity (e.g., heart rate)
Many models appeared to have captured something interesting
But attempts to link the methods to emotion theory still seem premature
12 teams competed to predict affective experience using multiple measures of peripheral nervous system activity (e.g., heart rate)
Many models appeared to have captured something interesting
But attempts to link the methods to emotion theory still seem premature
Very excited to announce that I will be recruiting 1-2 Social Psychology PhD students to join my new lab at the University of Florida!
We will be studying emotion, big team science, quantitative methods (e.g., machine learning), and research methodology.
psych.ufl.edu/graduate/pro...
We will be studying emotion, big team science, quantitative methods (e.g., machine learning), and research methodology.
psych.ufl.edu/graduate/pro...
July 22, 2024 at 1:08 PM
Very excited to announce that I will be recruiting 1-2 Social Psychology PhD students to join my new lab at the University of Florida!
We will be studying emotion, big team science, quantitative methods (e.g., machine learning), and research methodology.
psych.ufl.edu/graduate/pro...
We will be studying emotion, big team science, quantitative methods (e.g., machine learning), and research methodology.
psych.ufl.edu/graduate/pro...
The momentum the Psychological Science Accelerator already had, IMO, was more than enough.
Evidence: during my directorship, the community published 5 MASSIVE papers.
Another 3 studies are finished and accepted for publication.
2 are accepted in-principle.
Evidence: during my directorship, the community published 5 MASSIVE papers.
Another 3 studies are finished and accepted for publication.
2 are accepted in-principle.
April 1, 2024 at 3:09 PM
The momentum the Psychological Science Accelerator already had, IMO, was more than enough.
Evidence: during my directorship, the community published 5 MASSIVE papers.
Another 3 studies are finished and accepted for publication.
2 are accepted in-principle.
Evidence: during my directorship, the community published 5 MASSIVE papers.
Another 3 studies are finished and accepted for publication.
2 are accepted in-principle.
One of my favorite parts about #SPSP2024 is meeting collaborators -- both old and new. Come stop by to see what we've been up to in the world of big team science.
Apparently the session will be "highly attended" despite being at 8 AM on the last day? We'll see about that! 😅
Apparently the session will be "highly attended" despite being at 8 AM on the last day? We'll see about that! 😅
February 6, 2024 at 6:31 PM
One of my favorite parts about #SPSP2024 is meeting collaborators -- both old and new. Come stop by to see what we've been up to in the world of big team science.
Apparently the session will be "highly attended" despite being at 8 AM on the last day? We'll see about that! 😅
Apparently the session will be "highly attended" despite being at 8 AM on the last day? We'll see about that! 😅
3. Big team science is well positioned to help tackle big problems facing our world
To help address the climate crisis, Madalina Vlasceanu and colleagues leveraged big teams to identify *and* test numerous behavioral interventions around the world.
Very thought-provoking talk!
To help address the climate crisis, Madalina Vlasceanu and colleagues leveraged big teams to identify *and* test numerous behavioral interventions around the world.
Very thought-provoking talk!
October 26, 2023 at 6:20 PM
3. Big team science is well positioned to help tackle big problems facing our world
To help address the climate crisis, Madalina Vlasceanu and colleagues leveraged big teams to identify *and* test numerous behavioral interventions around the world.
Very thought-provoking talk!
To help address the climate crisis, Madalina Vlasceanu and colleagues leveraged big teams to identify *and* test numerous behavioral interventions around the world.
Very thought-provoking talk!
2. The open science and big team science movements are synergistic
Open science has been a huge enabler of big team science. Reciprocally, big team initiatives often serve as open science community of practices.
Here, I particularly enjoyed a talk on open hardware!
Open science has been a huge enabler of big team science. Reciprocally, big team initiatives often serve as open science community of practices.
Here, I particularly enjoyed a talk on open hardware!
October 26, 2023 at 6:19 PM
2. The open science and big team science movements are synergistic
Open science has been a huge enabler of big team science. Reciprocally, big team initiatives often serve as open science community of practices.
Here, I particularly enjoyed a talk on open hardware!
Open science has been a huge enabler of big team science. Reciprocally, big team initiatives often serve as open science community of practices.
Here, I particularly enjoyed a talk on open hardware!
1. Big teams can spur innovation at *multiple* parts of the research pipeline This was nicely illustrated in a talk by
the OIS center.
Big team science isn't just about data collection, but also idea generation, method development, data analysis, and more.
the OIS center.
Big team science isn't just about data collection, but also idea generation, method development, data analysis, and more.
October 26, 2023 at 6:18 PM
1. Big teams can spur innovation at *multiple* parts of the research pipeline This was nicely illustrated in a talk by
the OIS center.
Big team science isn't just about data collection, but also idea generation, method development, data analysis, and more.
the OIS center.
Big team science isn't just about data collection, but also idea generation, method development, data analysis, and more.
The 2023 Big Team Science conference is officially done!
I may be sleep deprived after 3 busy days. But Im more energized and hopeful than ever!
Big team science is an important part of scientific reform — and its clearly on the rise
3 things that stood out 🧵
I may be sleep deprived after 3 busy days. But Im more energized and hopeful than ever!
Big team science is an important part of scientific reform — and its clearly on the rise
3 things that stood out 🧵
October 26, 2023 at 6:17 PM
The 2023 Big Team Science conference is officially done!
I may be sleep deprived after 3 busy days. But Im more energized and hopeful than ever!
Big team science is an important part of scientific reform — and its clearly on the rise
3 things that stood out 🧵
I may be sleep deprived after 3 busy days. But Im more energized and hopeful than ever!
Big team science is an important part of scientific reform — and its clearly on the rise
3 things that stood out 🧵