Vekony, T. 1 , Marossy, H. 2 , Sefcsik, R. 2 , Nemes, L. 2 , Beres, S. 2 , Janacsek, K. 2 & Nemeth, D. 2, 3
1 Department of Neurology, University of Szeged, Szeged
2 Institute of Psychology, Eotvos Lorand University, Budapest
3 CRNL - Lyon Neuroscience Research Center, Université de Lyon, Lyon
In the cognitive neuroscience of learning, the most commonly used method to study behavior is to measure reaction times and accuracy. Studies show that emphasizing speed or accuracy during serial decision making can affect the performance of a participant. But does it lead to alterations in the degree of the acquired statistical knowledge as well? In the current study, 66 participants completed the Alternating Serial Reaction Time Task: in the learning phase, half of the participants were instructed to be as fast as possible (Speed Group), while the other half of them to be as accurate as possible (Accuracy Group). In the following test phase, all participants were asked to be as fast and as accurate as possible. In the learning phase, the two groups showed similar learning performance measured by reaction times. Contrary, only the Speed group showed learning when measured by accuracy. However, in the test phase, a similar level of statistical knowledge was found between groups. We can conclude that temporally distributed statistical learning is so robust that changes in instructions cannot affect the picking-up of regularities of the environment. Our study also shed light on the importance of differentiating between performance and competence during learning.