Complementary neural mechanisms for tracking items in human working memory

Science

Volumn 287, Issue 5453, 2000, Pages 643-646

  Jiang, Yang ^a,b   Haxby, James V ^a   Martin, Alex ^a   Ungerleider, Leslie G ^a   Parasuraman, Raja ^b  

^a NATIONAL INSTITUTE OF MENTAL HEALTH   (United States)

^b CATHOLIC UNIVERSITY OF AMERICA   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

ARTICLE; FRONTAL CORTEX; HUMAN; NUCLEAR MAGNETIC RESONANCE IMAGING; PRIORITY JOURNAL; RECOGNITION; VISUAL CORTEX; WORKING MEMORY;

CEREBRAL CORTEX; FACE; FRONTAL LOBE; HUMANS; MAGNETIC RESONANCE IMAGING; MEMORY; REGRESSION ANALYSIS; VISUAL CORTEX;

EID: 0034723225     PISSN: 00368075     EISSN: None     Source Type: Journal    
DOI: 10.1126/science.287.5453.643     Document Type: Article

References (29)

1. For reviews, see D. Schacter, R. Buckner, W. Koutstaal, Philos. Trans. R. Soc. London Ser. B Biol. Sci. 353, 1861 (1998); C. Wiggs and A. Martin, Curr. Opin. Neurobiol. 8, 227 (1998).
2. For reviews, see D. Schacter, R. Buckner, W. Koutstaal, Philos. Trans. R. Soc. London Ser. B Biol. Sci. 353, 1861 (1998); C. Wiggs and A. Martin, Curr. Opin. Neurobiol. 8, 227 (1998).
3. With event-related fMRI, it is possible to distinguish neural responses to different stimuli presented in an unpredictable sequence, spaced as short as 2 s apart; e.g., A. Dale and R. Buckner, Hum. Brain Mapp. 5, 329 (1997); V. Clark, J. Maisog, J. V. Haxby, J. Neurophysiol. 79, 3257 (1998).
4. With event-related fMRI, it is possible to distinguish neural responses to different stimuli presented in an unpredictable sequence, spaced as short as 2 s apart; e.g., A. Dale and R. Buckner, Hum. Brain Mapp. 5, 329 (1997); V. Clark, J. Maisog, J. V. Haxby, J. Neurophysiol. 79, 3257 (1998).
5. Visual stimuli consisted of 33 photographs of monochromatic faces and 83 photographs of nonmeaningful visual stimuli (scrambled versions of the faces). Before scanning, each participant practiced the working memory task until no errors were made on six successive trials. On average, participants took 12 to 18 trials to reach this criterion, during which each face was seen five to eight times. Participants were also highly accurate in performing the task during MR scanning (98.4% correct identification).
6. 3) with Fourier analysis to estimate the phase for alterations between the working memory and control tasks. Regions showing significant signal enhancement or reduction were defined as voxels with Z > 3.09 (P < 0.001, one-tailed) for the overall experimental effect and |Z| > 1.96 (P < 0.05, two-tailed) for either the contrast between responses to targets and nonrepeated distracters or the contrast between repeated distracters and nonrepeated distracters. Each participant completed 72 memory trials inside the scanner. A GE 1.5 T magnet was used to obtain T2*-weighted gradient echo echo-planar images with blood oxygen level-dependent signals. Whole brain volumes, each consisting of 22 5-mm-thick axial slices, were acquired for each participant (repetition time = 3 s, echo time = 40 ms, flip angle = 90°).
7. 3) with Fourier analysis to estimate the phase for alterations between the working memory and control tasks. Regions showing significant signal enhancement or reduction were defined as voxels with Z > 3.09 (P < 0.001, one-tailed) for the overall experimental effect and |Z| > 1.96 (P < 0.05, two-tailed) for either the contrast between responses to targets and nonrepeated distracters or the contrast between repeated distracters and nonrepeated distracters. Each participant completed 72 memory trials inside the scanner. A GE 1.5 T magnet was used to obtain T2*-weighted gradient echo echo-planar images with blood oxygen level-dependent signals. Whole brain volumes, each consisting of 22 5-mm-thick axial slices, were acquired for each participant (repetition time = 3 s, echo time = 40 ms, flip angle = 90°).
8. J. Talairach and P. Tournoux, Co-planar Stereotaxic Atlas of the Human Brain (Thieme, New York, 1988). The face-sensitive fusiform regions identified here are in the same areas noted in previous studies of face perception and working memory (within 1 to 9 mm) [e.g., S. M. Courtney et al., Nature 306, 608 (1997); J. V. Haxby et al., Hum. Brain Mapp. 3, 68 (1995); J. V. Haxby et al., Neuron 22, 189 (1999)]. Average Talairach coordinates for enhanced responses to targets were localized to the inferior frontal (left: -34, 9, 10; right: 43, 36, 9; BA 44), left insular (-27, -9, 9), superior temporal (left: -36, -19, 10; right: 56, -13, 6; BA 22/42), ventral temporal cortices/fusiform gyrus (left: -30, -48, -18; right: 30, -60, -20; BA 20/37), and left primary motor (-27, -33, 51; BA 4) and supplementary motor areas (1, -11, 47; BA 6).
9. J. Talairach and P. Tournoux, Co-planar Stereotaxic Atlas of the Human Brain (Thieme, New York, 1988). The face-sensitive fusiform regions identified here are in the same areas noted in previous studies of face perception and working memory (within 1 to 9 mm) [e.g., S. M. Courtney et al., Nature 306, 608 (1997); J. V. Haxby et al., Hum. Brain Mapp. 3, 68 (1995); J. V. Haxby et al., Neuron 22, 189 (1999)]. Average Talairach coordinates for enhanced responses to targets were localized to the inferior frontal (left: -34, 9, 10; right: 43, 36, 9; BA 44), left insular (-27, -9, 9), superior temporal (left: -36, -19, 10; right: 56, -13, 6; BA 22/42), ventral temporal cortices/fusiform gyrus (left: -30, -48, -18; right: 30, -60, -20; BA 20/37), and left primary motor (-27, -33, 51; BA 4) and supplementary motor areas (1, -11, 47; BA 6).
10. J. Talairach and P. Tournoux, Co-planar Stereotaxic Atlas of the Human Brain (Thieme, New York, 1988). The face-sensitive fusiform regions identified here are in the same areas noted in previous studies of face perception and working memory (within 1 to 9 mm) [e.g., S. M. Courtney et al., Nature 306, 608 (1997); J. V. Haxby et al., Hum. Brain Mapp. 3, 68 (1995); J. V. Haxby et al., Neuron 22, 189 (1999)]. Average Talairach coordinates for enhanced responses to targets were localized to the inferior frontal (left: -34, 9, 10; right: 43, 36, 9; BA 44), left insular (-27, -9, 9), superior temporal (left: -36, -19, 10; right: 56, -13, 6; BA 22/42), ventral temporal cortices/fusiform gyrus (left: -30, -48, -18; right: 30, -60, -20; BA 20/37), and left primary motor (-27, -33, 51; BA 4) and supplementary motor areas (1, -11, 47; BA 6).
11. J. Talairach and P. Tournoux, Co-planar Stereotaxic Atlas of the Human Brain (Thieme, New York, 1988). The face-sensitive fusiform regions identified here are in the same areas noted in previous studies of face perception and working memory (within 1 to 9 mm) [e.g., S. M. Courtney et al., Nature 306, 608 (1997); J. V. Haxby et al., Hum. Brain Mapp. 3, 68 (1995); J. V. Haxby et al., Neuron 22, 189 (1999)]. Average Talairach coordinates for enhanced responses to targets were localized to the inferior frontal (left: -34, 9, 10; right: 43, 36, 9; BA 44), left insular (-27, -9, 9), superior temporal (left: -36, -19, 10; right: 56, -13, 6; BA 22/42), ventral temporal cortices/fusiform gyrus (left: -30, -48, -18; right: 30, -60, -20; BA 20/37), and left primary motor (-27, -33, 51; BA 4) and supplementary motor areas (1, -11, 47; BA 6).
12. Identification of motor-related activation associated with right-hand button responses to the target faces validated our rapid, event-related fMRI method. All six participants showed robust activation in the left primary motor cortex and the supplementary motor area (SMA). These findings provide an important internal control for our rapid, event-related fMRI design.
13. Average Talairach coordinates for reduced responses to repeated distracter faces were as follows: ventral temporal cortex/fusiform gyrus (left: -31, -54, -19; right: 30, -64, -18; BA 37), occipital (left: -2, -100, -15; BA 17; right: 27, -78, 9/6, -97, 9; BA 18), superior occipital/intraparietal sulcus cortex (left: -24, -66, 39; right: 35, -65, 39; BA 19), the precuneus (2. -66, 35; BA 7), and posterior frontal (-33, 2, 28; BA 44).
14. Increased MR responses to repeated distracters were found in only two participants. In both participants, these increased responses were in the intraparietal sulcus.
15. Analysis of the MR responses to nonrepeated distracters at each of the 13 stimulus positions within a trial did not show the same trend as the targets or distracters. Thus, we discount the possibility that the repetition reduction effect reflects a "position effect" or linear trend within a trial. Similar results were found for the intraparietal area responses.
16. MR responses to repeated distracters reset in subsequent trials to initial levels in both left and right ventral temporal areas (Fig. 4B presents the mean over two hemispheres) as well as in the left and right intraparietal areas. In all cases, the response to the first presentation in later trials was not significantly different from the response to the first presentation in first trials (P > 0.1).
17. R. Buckner et al., J. Neurosci. 15, 12 (1995); J. Demb et al., J. Neurosci. 15, 5870 (1995); D. Schacter et al., Proc. Natl. Acad. Sci. U.S.A. 93, 321 (1996); L. Squire et al., Proc. Natl. Acad. Sci. U.S.A. 89, 1837 (1992); C. Büchel, J. T. Coull, K. J. Friston, Science 283, 1538 (1999).
18. R. Buckner et al., J. Neurosci. 15, 12 (1995); J. Demb et al., J. Neurosci. 15, 5870 (1995); D. Schacter et al., Proc. Natl. Acad. Sci. U.S.A. 93, 321 (1996); L. Squire et al., Proc. Natl. Acad. Sci. U.S.A. 89, 1837 (1992); C. Büchel, J. T. Coull, K. J. Friston, Science 283, 1538 (1999).
19. R. Buckner et al., J. Neurosci. 15, 12 (1995); J. Demb et al., J. Neurosci. 15, 5870 (1995); D. Schacter et al., Proc. Natl. Acad. Sci. U.S.A. 93, 321 (1996); L. Squire et al., Proc. Natl. Acad. Sci. U.S.A. 89, 1837 (1992); C. Büchel, J. T. Coull, K. J. Friston, Science 283, 1538 (1999).
20. R. Buckner et al., J. Neurosci. 15, 12 (1995); J. Demb et al., J. Neurosci. 15, 5870 (1995); D. Schacter et al., Proc. Natl. Acad. Sci. U.S.A. 93, 321 (1996); L. Squire et al., Proc. Natl. Acad. Sci. U.S.A. 89, 1837 (1992); C. Büchel, J. T. Coull, K. J. Friston, Science 283, 1538 (1999).
21. R. Buckner et al., J. Neurosci. 15, 12 (1995); J. Demb et al., J. Neurosci. 15, 5870 (1995); D. Schacter et al., Proc. Natl. Acad. Sci. U.S.A. 93, 321 (1996); L. Squire et al., Proc. Natl. Acad. Sci. U.S.A. 89, 1837 (1992); C. Büchel, J. T. Coull, K. J. Friston, Science 283, 1538 (1999).
22. Six participants performed the same working memory task as in the fMRI study, except that they responded to both targets and distracters. The median RT for repeated distracters (429 ms) was significantly shorter than that for nonrepeated distracters (491 ms): F(1,5) = 6.7, P < 0.05. RTs for repeated distracters were computed separately for each of five repetitions. A significant main effect of repetition, F(4,20) = 7.2, P < 0.001, indicated that RT declined with repeated presentation (from 447 to 396 ms). Thus, repetition of familiar objects during the working memory task was associated with improved performance in detecting distracters.
23. E. K. Miller and R. Desimone, Science 263, 520 (1994); E. K. Miller, C. Erickson, R. Desimone, J. Neurosci. 16, 5154 (1996).
24. E. K. Miller and R. Desimone, Science 263, 520 (1994); E. K. Miller, C. Erickson, R. Desimone, J. Neurosci. 16, 5154 (1996).
25. L. Li and R. Desimone, J. Neurosci. 13, 1460 (1993); W. Suzuki, Neuron 24, 295 (1999).
26. L. Li and R. Desimone, J. Neurosci. 13, 1460 (1993); W. Suzuki, Neuron 24, 295 (1999).
27. R. Parasuraman, in The Attentive Brain, R. Parasuraman, Ed. (MIT Press, Cambridge, MA, 1998), pp. 3-15.
28. R. Desimone and J. Duncan, Annu. Rev. Neurosci. 18, 193 (1995).
29. We thank R. Desimone for insightful comments on an earlier version of the manuscript, J. Maisog for implementing data analysis software, J. Schouten and E. Hoffman for participant recruitment and training, L. Kikuchi and C. Chavez for conducting the behavioral study, J. Szczepanik for help with data analysis, S. Courtney and L Petit for valuable discussions, and the NIH in vivo Nuclear Magnetic Resonance Center for assistance with MR imaging. Y.J. and R.P. were supported by NIH grant AG07569.