Theories in implicit learning indicate that implicitly learned information is processed in encapsulated modules that work independently from each other and do not interact. These modules are domain specific. Thus, visual and spatial information is for example processed in different modules (e.g., Keele et al., 2003).
For the serial reaction time task (SRTT; Nissen & Bullemer, 1987), this assumption implies that a visual and a spatial sequence can be learned implicitly in parallel (e.g., Haider et al., 2018). However, if processing of visual and spatial information occurs in different independent modules, a cross-modal sequence that contains both, visual and spatial information should not be learned.
Here, we investigated whether the learning of such a cross-modal sequence is possible. To answer this question, we created a SRTT that contained an eight element sequence consisting of visual (target colors) and spatial (motor responses) information. This cross-modal sequence was realized with four possible transitions (visual-visual; visual-spatial, spatial-spatial, spatial-visual). We expect that the transitions within a dimension (visual vs. spatial) should be learned whereas the cross-modal transitions should not.
Such a finding would support the assumption that in implicit learning, the processing of the implicitly learned information occurs in encapsulated modules that are specific for different dimensions (e.g., visual or spatial) and that do not interact.