Schultz, B. 1, 2 , Stevens, C. 1 , Tillmann, B. 2 & Keller, P. 3
1 MARCS Auditory Laboratories, University of Western Sydney, Australia
2 Lyon Neuroscience Research Center, CNRS-UMR 5292, INSERM U1028, Université de Lyon, France
3 Max Plank Institute for Human Cognitive and Brain Sciences, Leipzig, Germany
Implicit learning (IL) of musical rhythm and its properties (e.g. meter) has received minimal attention. According to the Dynamic Attending Theory (DAT; Jones, 2009), metrical frameworks facilitate temporal expectancies. It was hypothesized that learning occurs more readily for metrical patterns (MP) than non-metrical patterns (NMP). Experiment 1 used a serial reaction-time task (SRT) where learning is characterized by: RT decreases over blocks containing the exposure rhythm; RT increases when new rhythms are introduced; and RT recovery when the exposure rhythm is reintroduced. A generation task using the Process Dissociation Procedure (Jacoby, 1991) assessed IL. Experiment 1 demonstrated IL of MPs and NMPs but the presence of meter did not improve the rate of learning. However, there was evidence of metric binding: the presentation of novel rhythms with different metrical frameworks resulted in greater RT increases than novel rhythms with the same meter. The generation task indicated that learning of MPs and NMPs was implicit. Overall, Experiment 1 indicates that metrical and non-metrical rhythms can be learned implicitly and, as per DAT, metrical frameworks can facilitate temporal expectancies. Facilitation from meter in sequence-reproduction tasks is investigated in Experiment 2 using SRT and serial recall tasks. Results for Experiment 2 are forthcoming.