We predicted strategy choice could explain our findings, but reinforcement learning did not show a greater probability of selecting the theoretically best strategy or each skier's estimated best strategy. However, they incurred lower costs for their suboptimal strategies

What information guided the strategy decisions in the two learning groups? We found that both learning groups’ decisions could be accounted for by a ’win-stay, lose-shift’ heuristic. The difference between the learning groups was not statistically significant, however

We have also analyzed the ranking and performance of each strategy. More details on this can be found in the paper. In summary, we found significant differences in race times between the strategies.

Based on our findings, reinforcement learning appears to be an effective training strategy for skilled performers. We believe that reinforcement learning can serve as a valuable supplement to traditional instruction. More on this in the discussion.

Thanks to all skiers and coaches who participated in this project. Thanks also to Norges idrettshøgskole and Norwegian School of Sport Sciences | IFP and for their support. Thanks to all volunteers who have worked hard to make this study possible ⛷️

We would very much like to receive feedback on the manuscript before submitting it to a journal. If there is anything you believe can be improved, please write to us. All code and data for the analysis are available at osf.io/uw986/

With Simen Leithe Tajet, Thomas Losnegard, Romy Frömer, Reid, Gilgien and Haugen. #motorlearning #skilllearning #reinforcementlearning #alpineskiing #slalom #expertperformance #decisionmaking

⛷️If you're a coach aiming to help your skiers perform better in slalom, our new preprint might be of interest: biorxiv.org/content/10.110…. In this study, 98 alpine skiers (13/98 ranked in the top 100 in the world 🏅) tested four strategies (panel b) (1/4)

We've received great and valuable inputs on our new preprint about using reinforcement learning to improve sports training ⛷️🏌️‍♀️ 🎾⚽️ - Thank you! We are still seeking feedback and comments before submitting it to a journal. Please retweet! Link: biorxiv.org/content/10.110…

As my PhD comes to an end, I want to acknowledge the methods books that have been instrumental in my journey. *Missing from the list is statistical rethinking This book by Dale Barr has also been exceptionally helpful: psyteachr.github.io/stat-models-v1/

🎯🎯🎯

New book version up! solomon.quarto.pub/sr2rstan/ This is the 0.0.3 version of the #rstan sibling to my #brms translation of McElreath's 2nd edition. #Rstats 1/4

We have two new preprints up today! One is on exploratory research, written for sport scientists doi.org/10.51224/SRXIV…, the other is on measurement development, reporting the next step in developing an academic research values scale by a.kis! osf.io/preprints/psya…

Time for a thread on our new preprint on exploratory analyses! This paper was started by Max Ditroilo during his sabbatical at my lab. It started from the observation that exploratory research is often not openly reported. How can we improve this? sportrxiv.org/index.php/serv… >

🚨New pub out now! 🚨In this study we showed that performance level moderated learners’ sensitivity to reward-prediction errors. For some insights into what this might mean for instructors interested in motor skill acquisition, grab a free copy here 👇 doi.org/10.3389/fnbeh.…

🍿

Time for the World Cup slalom in Levi, which features a long flat section. Remember to 'extend with rock skis forward' in this section to achieve the best race time! ⛷️⛷️⛷️ #worldcup #alpineskiing #reinforcementlearning Full paper here: biorxiv.org/content/10.110…

I really enjoyed reading this world view in nature about the importance of exploratory research in the social sciences by Balazs Aczel and am wondering what people think. Do you generally agree with his point? nature.com/articles/d4158…

Anyone who takes Huberman too seriously might want to complete my Metascience course assignment where you get to dive into his understanding of statistics and his confirmation bias: daniellakens.blogspot.com/2024/10/andrew…