Sequence-Specific and Task-General Knowledge: Parallel Learning Mechanisms Within an Implicit Perceptual-Motor Skill Learning Task
Poster Session B - Sunday, March 30, 2025, 8:00 – 10:00 am EDT, Back Bay Ballroom/Republic Ballroom
Antonio P. Santa Cruz1 (antoniosantacruz2026@u.northwestern.edu), Paul J. Reber1; 1Northwestern University
In perceptual-motor skill learning tasks, practice leads to improved performance. The Serial Interception Sequence Learning (SISL) task is a perceptual-motor memory paradigm that fosters robust implicit learning of a covertly embedded repeating sequence. In this task, participants make precisely timed interception responses to visual cues moving toward a target region. The speed of the cues is individually adjusted based on performance to maintain a consistent accuracy level. If participants make too many mistakes, the cues slow down, and if accuracy is near ceiling, the cues speed up. Sequence-specific learning is measured by a performance advantage (SSPA) reflected by increased accuracy of the trained sequence versus the non-trained sequences. Participants also exhibit learning via the adaptive cue speed (ACS) measure that gradually increases over practice. Participants practiced the SISL task over 12 240-cue blocks separated by self-terminated breaks. At block six, the covertly embedded repeating sequence switched to a novel sequence. As expected, measures of sequence-specific learning reset, reflecting a restart of the learning process. However, the task-general ACS measure continued to improve without disruption, indicating that this form of learning is general to the basic interception task and not tied to the specific sequence. This dissociation between learning measures suggests parallel mechanisms during SISL practice. In prior research, we have shown that sequence-specific measures are rigidly tied to the learning context and do not generalize well to novel conditions. In contrast, the ACS measure reflects an improving ability in the core motor task, effectively generalizing across conditions to novel sequences.
Topic Area: LONG-TERM MEMORY: Skill Learning