OS_45. Working memory
Sunday, October 02nd, 2011 [17:00 - 18:00]
OS_45.1 - Declarative and procedural working memory: analogous processing principles?
Souza da Silva, A. , Oberauer, K. , Gade, M. & Druey, M.
University of Zurich
Oberauer (2009) distinguishes two working memory (WM) sub-systems: The declarative WM provides information input (memory-sets) for processing; whereas the procedural WM provides access to the processing operations themselves (task-sets). The present study tested the assumption that these sub-systems select representations in analogous ways. Participants selected a memory-list and a digit within the list (declarative representations). They selected a task-set to be applied to the digit, and a response within that task-set (procedural representations). The number of lists (one, two, or three) and the number of tasks (one, two, or three) to be switched between were manipulated independently. Switching between lists and switching between tasks produced time costs (list-switch and task-switch costs, respectively). List-switch costs increased as a function of the number of lists (two versus three); whereas task-switch costs were not affected by the number of tasks to be switched between. Increasing the number of lists also affected task-switch costs, indicating some degree of interaction between the sub-systems. Furthermore, analogous patterns of list-mixing and task-mixing costs were observed. These findings support the hypothesis of analogous, but not completely independent, WM sub-systems.
OS_45.2 - Declarative and procedural working memory - two separate systems?
Gade, M. , Druey, M. & Oberauer, K.
University of Zuerich, Institute for Psychology, General Psychology (Cognition)
Oberauer (2010) proposed a two-fold working memory system underlying the human ability to pursue goal-directed actions: a declarative working memory and a procedural one. Whereas the declarative part is thought to maintain and ensure access to representations of goal-relevant objects, the procedural part contains the representations of goal-relevant cognitive or overt operations. Capacity limits in declarative working memory have been studied in the working-memory literature. Capacity limits on procedural working memory have been revealed by research on action control, showing that people can typically hold only one task set available for direct control of action. The question of our research is whether the two systems have separate capacity limits or must share a common capacity. We combined classical tasks from both research areas in a dual-task experiment: letter recognition with varying memory set size served to manipulate declarative load, and digit classification with varying number of stimulus-response rules served to manipulate procedural load. The data suggest that declarative and procedural working memory operate independently. Two control experiments show that loading either one system with two tasks creates interference due to the irrelevant load in the second task.
OS_45.3 - How to distract the focus of attention in working memory
Hein, L. & Oberauer, K.
Department of Psychology (Cognitive Psychology Unit), University of Zurich, Switzerland
The study investigates how the single-item focus of attention (foa) in working memory can be distracted. The foa is assumed to hold on to the last item processed, such that sequential operations on the same item are faster than switches to a different item in working memory. To explore whether the foa can be detracted from an item by perceptual input, we used an interruption paradigm. We combined a sequential arithmetic memory-updating task with rare interruptions by perceptual stimuli, to which an immediate choice response was required. If the last item processed before the interruption remains in the foa after the interruption, object-repetition benefits should remain when updating is resumed after the interruption. Results indicate that memory items do not remain in the foa after interruptions by responses to perceptual stimuli. Interruptions by infrequent unfilled time intervals of the same duration as the responses to the interruption task had the same effect. However, repetition benefits remained when unfilled intervals were interspersed more frequently and hence, were not surprising. We conclude that the foa can be detracted by surprise, as indicated by the disappearance of repetition benefits after infrequent events, but not when the same events occur frequently.