TY - JOUR
T1 - Working memory storage is intrinsically domain specific
AU - Fougnie, Daryl
AU - Zughni, Samir
AU - Godwin, Douglass
AU - Marois, René
N1 - Publisher Copyright:
© 2014 American Psychological Association.
PY - 2015
Y1 - 2015
N2 - A longstanding debate in working memory (WM) is whether information is maintained in a central, capacity-limited storage system or whether there are domain-specific stores for different modalities. This question is typically addressed by determining whether concurrent storage of 2 different memory arrays produces interference. Prior studies using this approach have shown at least some cost to maintaining 2 memory arrays that differed in perceptual modalities. However, it is not clear whether these WM costs resulted from competition for a central, capacity-limited store or from other potential sources of dual-task interference, such as task preparation and coordination, overlap in representational content (e.g., object vs. space based), or cognitive strategies (e.g., verbalization, chunking of the stimulus material in a higher order structure). In the present study we assess dual-task costs during the concurrent performance of a visuospatial WM task and an auditory object WM task when such sources of interference are minimized. The results show that performance of these 2 WM tasks are independent from each another, even at high WMload. Only when we introduced a common representational format (spatial information) to bothWM tasks did dual-task performance begin to suffer. These results are inconsistent with the notion of a domain-independent storage system, and suggest instead that WM is constrained by multiple domainspecific stores and central executive processes. Evidently, there is nothing intrinsic about the functional architecture of the human mind that prevents it from storing 2 distinct representations in WM, as long as these representations do not overlap in any functional domain.
AB - A longstanding debate in working memory (WM) is whether information is maintained in a central, capacity-limited storage system or whether there are domain-specific stores for different modalities. This question is typically addressed by determining whether concurrent storage of 2 different memory arrays produces interference. Prior studies using this approach have shown at least some cost to maintaining 2 memory arrays that differed in perceptual modalities. However, it is not clear whether these WM costs resulted from competition for a central, capacity-limited store or from other potential sources of dual-task interference, such as task preparation and coordination, overlap in representational content (e.g., object vs. space based), or cognitive strategies (e.g., verbalization, chunking of the stimulus material in a higher order structure). In the present study we assess dual-task costs during the concurrent performance of a visuospatial WM task and an auditory object WM task when such sources of interference are minimized. The results show that performance of these 2 WM tasks are independent from each another, even at high WMload. Only when we introduced a common representational format (spatial information) to bothWM tasks did dual-task performance begin to suffer. These results are inconsistent with the notion of a domain-independent storage system, and suggest instead that WM is constrained by multiple domainspecific stores and central executive processes. Evidently, there is nothing intrinsic about the functional architecture of the human mind that prevents it from storing 2 distinct representations in WM, as long as these representations do not overlap in any functional domain.
KW - Working memory
UR - http://www.scopus.com/inward/record.url?scp=84925808986&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84925808986&partnerID=8YFLogxK
U2 - 10.1037/a0038211
DO - 10.1037/a0038211
M3 - Article
C2 - 25384164
AN - SCOPUS:84925808986
SN - 0096-3445
VL - 144
SP - 30
EP - 47
JO - Journal of Experimental Psychology: General
JF - Journal of Experimental Psychology: General
IS - 1
ER -