We examined sequential learning of actions in an experiment in which four different actions (push, twist, pinch, switch) were placed at four horizontal locations. At transfer, participants responded to a sequence that required performing the same sequence of actions at different locations and to a different sequence of actions at the same sequence of locations. Participants with explicit knowledge demonstrated only learning the sequence of response locations. However, participants with implicit knowledge learned the sequence of actions just as well as the sequence of locations, and performance on individual sequences was just as good as performance when both sequences were presented. These results demonstrated that two types of sequences, one of actions and another of response locations, can be learned simultaneously, suggesting that parallel representations are involved in implicit motor skill acquisition.
Skilled actions in everyday life require sequential knowledge in order to prepare responses in advance. Motor skill learning has often been explored in the lab using the serial response time (SRT) paradigm (Nissen & Bullemer, 1987). In this task, participants respond to stimuli that, unbeknownst to the participants, sometimes appear in a repeating sequence and other times appear randomly. Learning is assessed in two ways. One measure is the time it takes to respond to the sequenced stimuli relative to random stimuli; this measure is considered implicit because it is based on performance and is not associated with awareness of the sequence. The second measure is an interview, or recognition or recall measure. These measures are considered explicit because participants know that their memory is being tapped, and performance is necessarily associated with awareness of die material remembered. Interestingly, response times (RTs) often decrease even though the participant is unaware of the sequence (Willingham, Nissen, & Bullemer, 1989). Some researchers claim that the two types of learning represent the workings of different memory systems (e.g., Squire & ZolaMorgan, 1991), which raises the possibility that the two systems rely on different types of representations.
Most investigations on the nature of the representation for motor skill acquisition have used buttonpressing as the response, and it is well established that people are capable of learning sequences of response locations implicitly (e.g., Nattkemper & Prinz, 1997; Willingham, Wells, Farrell, & Stemwedel, 2000; but see Remillard, 2003). This task provides a useful way to study skills, such as piano playing and typing, for which the actions (keypresses) vary little and responses are differentiated by spatial location. However, other types of motor skills require learning a sequence of different actions. For example, dancers and martial artists learn sequences of different kinds of movements; however, this learning is usually explicit. To our knowledge, no research has investigated whether action sequences can be learned implicitly. By action sequence, we mean a sequence in which the responses can be differentiated by the action required to make the response. In the typical implicit SRT tasks, participants respond with the same action throughout the sequence (keypresses), so the sequence cannot be learned as a sequence of distinct actions and, rather, is typically learned as a sequence of response locations.
To address whether action sequences can be learned implicitly, we used a modified SRT task. Participants responded to spatial stimuli on a response board with four manipulanda, each requiring a unique action. At transfer, the participants responded on a different board in which the manipulanda were placed at different spatial locations. They responded to a sequence that required performing the same sequence of actions at different locations and a sequence that required performing the same sequence of response locations but with a different sequence of actions. …