A Quantitative Model that Incorporates Reactive Inhibition but No Facilitating Consolidation Explains Single-Session Motor Sequence Practice Effects

Several recent papers have advanced the hypothesis that motor sequence learning occurs during breaks between performance trials rather than during performance itself. Here, we advance an alternative hypothesis in which motor learning occurs concurrently with performance and that reactive inhibition – which has a progressive slowing effect on performance during each trial but dissipates during breaks – operates with equivalent magnitude across trials, yielding the illusion that learning occurs during breaks across the range of early training trials on which analyses were conducted in the recent studies. We advance a quantitative model of motor sequence practice effects that embodies our assumptions and provides a close fit to data at the grain-size of sequences within-trials. Our modeling approach supports a level of predictive precision and theoretical transparency that in our view should be the standard in future work on this topic.