| Popis: |
When learning a novel visuomotor mapping (e.g., mirror writing), accuracy can improve quickly through explicit learning (e.g., move left to go right) but after considerable practice, implicit learning takes over, producing fast, natural movements. This implicit learning occurs automatically, but it has been unknown whether explicit learning is similarly obligatory. Using a reaching task with a 90-degree rotation between screen position and movement direction, we found that explicit learning could be “turned off” by introducing the rotation gradually (increments of 10-degrees) and instructing participants to move quickly. These specific conditions were crucial, because both explicit and implicit learning occurred if the rotation occurred suddenly, if participants were told to emphasize accuracy, or if visual feedback during movement was removed. We reached these conclusions by examining the time course of learning (e.g., whether there was fast improvement followed by a long tail of additional improvement), by examining the aftereffects of learning when the rotation was abruptly removed, and by using formal model comparison between a dual-state (explicit and implicit) versus a single-state learning model as applied to the data.Author summaryIn some situations, the relationship between motion direction and what we see is different than normal. For instance, try using a computer mouse that is held sideways (a 90-degree rotation). When first encountering this situation, people move carefully, using explicit strategies (e.g., move right to go up). However, after many learning trials, motion becomes automatic (implicit) and natural. Prior results found that implicit visuomotor learning always occurs with enough experience. In our study, we found that this is not true of explicit visuomotor learning; in some situations, explicit learning can be turned off. More specifically, we found that this occurs when the novel visuomotor situation is: 1) introduced gradually (e.g., a gradual introduction of 90-degree rotation in steps of 10 degrees); 2) when there is pressure to move quickly; and 3) with real-time onscreen views of the motion path. If any of these three components are missing, then people use explicit learning. These conclusions were reached by examining the time course of learning (e.g., whether there was both fast and slow learning as assessed with mathematical models) and by examining the tendency to automatically move in the opposite direction from the rotation when the rotation is abruptly removed after learning. |
