Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • Article
  • Published:

Layer-dependent activity in human prefrontal cortex during working memory

Nature Neuroscience volume 22, pages 1687–1695 (2019)Cite this article

Subjects

Abstract

Working memory involves storing and/or manipulating previously encoded information over a short-term delay period, which is typically followed by a behavioral response based on the remembered information. Although working memory tasks often engage dorsolateral prefrontal cortex, few studies have investigated whether their subprocesses are localized to different cortical depths in this region, and none have done so in humans. Here we use high-resolution functional MRI to interrogate the layer specificity of neural activity during different periods of a delayed-response task in dorsolateral prefrontal cortex. We detect activity time courses that follow the hypothesized patterns: namely, superficial layers are preferentially active during the delay period, specifically in trials requiring manipulation (rather than mere maintenance) of information held in working memory, and deeper layers are preferentially active during the response. Results demonstrate that layer-specific functional MRI can be used in higher-order brain regions to noninvasively map cognitive processing in humans.

This is a preview of subscription content, access via your institution

Access options

Buy this article

  • Purchase on Springer Link
  • Instant access to full article PDF
Buy now

Prices may be subject to local taxes which are calculated during checkout

Fig. 1: Task, hypothesis and region of interest.
Fig. 2: Different trial types evoke distinct spatiotemporal patterns of activity.
Fig. 3: Activity contrasts across layers and conditions of interest.
Fig. 4: Single-subject, layer-dependent activity profiles.

Similar content being viewed by others

Data availability

Data are available via OpenNeuro at the following link: https://doi.org/10.18112/openneuro.ds002076.v1.0.1

Code availability

All code is available in the following GitHub repository: https://github.com/layerfMRI/repository/tree/master/DLPFC_Emily

References

  1. Courtney, S. M., Ungerleider, L. G., Keil, K. & Haxby, J. V. Transient and sustained activity in a distributed neural system for human working memory. Nature 386, 608 (1997).

    Article  CAS  Google Scholar 

  2. Courtney, S. M., Petit, L., Maisog, J. M., Ungerleider, L. G. & Haxby, J. V. An area specialized for spatial working memory in human frontal cortex. Science 279, 1347–1351 (1998).

    Article  CAS  Google Scholar 

  3. Goldman-Rakic, P. Cellular basis of working memory. Neuron 14, 477–485 (1995).

    Article  CAS  Google Scholar 

  4. D’esposito, M. et al. The neural basis of the central executive system of working memory. Nature 378, 279 (1995).

    Article  Google Scholar 

  5. Sawaguchi, T., Matsumura, M. & Kubota, K. Depth distribution of neuronal activity related to a visual reaction time task in the monkey prefrontal cortex. J. Neurophysiol. 61, 435–446 (1989).

    Article  CAS  Google Scholar 

  6. Sawaguchi, T., Matsumura, M. & Kubota, K. Catecholaminergic effects on neuronal activity related to a delayed response task in monkey prefrontal cortex. J. Neurophysiol. 63, 1385–1400 (1990).

    Article  CAS  Google Scholar 

  7. Arnsten, A. F., Wang, M. J. & Paspalas, C. D. Neuromodulation of thought: flexibilities and vulnerabilities in prefrontal cortical network synapses. Neuron 76, 223–239 (2012).

    Article  CAS  Google Scholar 

  8. Markowitz, D. A., Curtis, C. E. & Pesaran, B. Multiple component networks support working memory in prefrontal cortex. Proc. Natl Acad. Sci. USA 112, 11084–11089 (2015).

    Article  CAS  Google Scholar 

  9. Bastos, A. M., Loonis, R., Kornblith, S., Lundqvist, M. & Miller, E. K. Laminar recordings in frontal cortex suggest distinct layers for maintenance and control of working memory. Proc. Natl Acad. Sci. USA 115, 1117–1122 (2018).

    Article  CAS  Google Scholar 

  10. Barbey, A. K., Koenigs, M. & Grafman, J. Dorsolateral prefrontal contributions to human working memory. Cortex 49, 1195–1205 (2013).

    Article  Google Scholar 

  11. Mackey, W. E., Devinsky, O., Doyle, W. K., Meager, M. R. & Curtis, C. E. Human dorsolateral prefrontal cortex is not necessary for spatial working memory. J. Neurosci. 36, 2847–2856 (2016).

    Article  CAS  Google Scholar 

  12. Postle, B. R. et al. Repetitive transcranial magnetic stimulation dissociates working memory manipulation from retention functions in the prefrontal, but not posterior parietal. Cortex. J. Cogn. Neurosci. 18, 1712–1722 (2006).

    Article  Google Scholar 

  13. Garey, L. J. et al. Reduced dendritic spine density on cerebral cortical pyramidal neurons in schizophrenia. J. Neurol., Neurosurg. amp; Psychiatry 65, 446–453 (1998).

    Article  CAS  Google Scholar 

  14. Glantz, L. A. & Lewis, D. A. Decreased dendritic spine density on prefrontal cortical pyramidal neurons in schizophrenia. Arch. Gen. Psychiatry 57, 65–73 (2000).

    Article  CAS  Google Scholar 

  15. Cannon, T. D. et al. Dorsolateral prefrontal cortex activity during maintenance and manipulation of information in working memory in patients with schizophrenia. Arch. Gen. Psychiatry 62, 1071–1080 (2005).

    Article  Google Scholar 

  16. Wang, M., Vijayraghavan, S. & Goldman-Rakic, P. S. Selective D2 receptor actions on the functional circuitry of working memory. Science 303, 853–856 (2004).

    Article  CAS  Google Scholar 

  17. Opris, I., Hampson, R. E., Stanford, T. R., Gerhardt, G. A. & Deadwyler, S. A. Neural activity in frontal cortical cell layers: evidence for columnar sensorimotor processing. J. Cogn. Neurosci. 23, 1507–1521 (2011).

    Article  Google Scholar 

  18. Polimeni, J. R., Fischl, B., Greve, D. N. & Wald, L. L. Laminar analysis of 7 T bold using an imposed spatial activation pattern in human V1. Neuroimage 52, 1334–1346 (2010).

    Article  Google Scholar 

  19. Kok, P., Bains, L. J., van Mourik, T., Norris, D. G. & de Lange, F. P. Selective activation of the deep layers of the human primary visual cortex by top-down feedback. Curr. Biol. 26, 371–376 (2016).

    Article  CAS  Google Scholar 

  20. Muckli, L. et al. Contextual feedback to superficial layers of V1. Curr. Biol. 25, 2690–2695 (2015).

    Article  CAS  Google Scholar 

  21. De Martino, F. et al. Frequency preference and attention effects across cortical depths in the human primary auditory cortex. Proc. Natl Acad. Sci. USA 112, 16036–16041 (2015).

    Article  Google Scholar 

  22. Huber, L. et al. High-resolution CBV-fMRI allows mapping of laminar activity and connectivity of cortical input and output in human M1. Neuron 96, 1253–1263.e1257 (2017).

    Article  CAS  Google Scholar 

  23. Corbitt, P. T., Ulloa, A. & Horwitz, B. Simulating laminar neuroimaging data for a visual delayed match-to-sample task. NeuroImage 173, 199–222 (2018).

    Article  Google Scholar 

  24. D’Esposito, M., Postle, B. R., Ballard, D. & Lease, J. Maintenance versus manipulation of information held in working memory: an event-related fmri study. Brain Cogn. 41, 66–86 (1999).

    Article  Google Scholar 

  25. Lu, H., Golay, X., Pekar, J. J. & Van Zijl, P. C. Functional magnetic resonance imaging based on changes in vascular space occupancy. Magn. Reson. Med.: Off. J. Int. Soc. Magn. Reson. Med. 50, 263–274 (2003).

    Article  Google Scholar 

  26. Poser, B. A., Koopmans, P. J., Witzel, T., Wald, L. L. & Barth, M. Three dimensional echo-planar imaging at 7 tesla. Neuroimage 51, 261–266 (2010).

    Article  CAS  Google Scholar 

  27. Petrides, M. & Pandya, D. Dorsolateral prefrontal cortex: comparative cytoarchitectonic analysis in the human and the macaque brain and corticocortical connection patterns. Eur. J. Neurosci. 11, 1011–1036 (1999).

    Article  CAS  Google Scholar 

  28. Melchitzky, D. S., Sesack, S. R., Pucak, M. L. & Lewis, D. A. Synaptic targets of pyramidal neurons providing intrinsic horizontal connections in monkey prefrontal cortex. J. Comp. Neurol. 390, 211–224 (1998).

    Article  CAS  Google Scholar 

  29. Medalla, M. & Barbas, H. Diversity of laminar connections linking periarcuate and lateral intraparietal areas depends on cortical structure. Eur. J. Neurosci. 23, 161–179 (2006).

    Article  CAS  Google Scholar 

  30. Wang, X.-J. Synaptic basis of cortical persistent activity: the importance of NMDA receptors to working memory. J. Neurosci. 19, 9587–9603 (1999).

    Article  CAS  Google Scholar 

  31. Wang, M. et al. NMDA receptors subserve persistent neuronal firing during working memory in dorsolateral prefrontal cortex. Neuron 77, 736–749 (2013).

    Article  CAS  Google Scholar 

  32. Bunge, S. A., Hazeltine, E., Scanlon, M. D., Rosen, A. C. & Gabrieli, J. Dissociable contributions of prefrontal and parietal cortices to response selection. Neuroimage 17, 1562–1571 (2002).

    Article  Google Scholar 

  33. Jiang, Y. & Kanwisher, N. Common neural substrates for response selection across modalities and mapping paradigms. J. Cogn. Neurosci. 15, 1080–1094 (2003).

    Article  Google Scholar 

  34. Curtis, C. E. & D’Esposito, M. Persistent activity in the prefrontal cortex during working memory. Trends Cogn. Sci. 7, 415–423 (2003).

    Article  Google Scholar 

  35. Hadland, K. A., Rushworth, M. F. S., Passingham, R. E., Jahanshahi, M. & Rothwell, J. C. Interference with performance of a response selection task that has no working memory component: an RTMS comparison of the dorsolateral prefrontal and medial frontal cortex. J. Cogn. Neurosci. 13, 1097–1108 (2001).

    Article  CAS  Google Scholar 

  36. Schumacher, E. H. & D’Esposito, M. Neural implementation of response selection in humans as revealed by localized effects of stimulus–response compatibility on brain activation. Hum. Brain Mapp. 17, 193–201 (2002).

    Article  Google Scholar 

  37. Takada, M. et al. Organization of prefrontal outflow toward frontal motor‐related areas in macaque monkeys. Eur. J. Neurosci. 19, 3328–3342 (2004).

    Article  CAS  Google Scholar 

  38. Tomio, A., Kyoko, W. & Kisou, K. Connections of area 8 with area 6 in the brain of the macaque monkey. J. Comp. Neurol. 277, 21–40 (1988).

    Article  Google Scholar 

  39. Arikuni, T. & Kubota, K. The organization of prefrontocaudate projections and their laminar origin in the macaque monkey: a retrograde study using Hrp-Gel. J. Comp. Neurol. 244, 492–510 (1986).

    Article  CAS  Google Scholar 

  40. Yeterian, E. & Pandya, D. Laminar origin of striatal and thalamic projections of the prefrontal cortex in rhesus monkeys. Exp. Brain Res. 99, 383–398 (1994).

    Article  CAS  Google Scholar 

  41. Perlstein, W. M., Carter, C. S., Noll, D. C. & Cohen, J. D. Relation of prefrontal cortex dysfunction to working memory and symptoms in schizophrenia. Am. J. Psychiatry 158, 1105–1113 (2001).

    Article  CAS  Google Scholar 

  42. Kay, K. et al. A critical assessment of data quality and venous effects in ultra-high-resolution fMRI. Neuroimage 189, 847–869 (2019).

    Article  Google Scholar 

  43. Goense, J., Merkle, H. & Logothetis, N. K. High-resolution fMRI reveals laminar differences in neurovascular coupling between positive and negative bold responses. Neuron 76, 629–639 (2012).

    Article  CAS  Google Scholar 

  44. Kim, T. & Kim, S.-G. Cortical layer-dependent arterial blood volume changes: improved spatial specificity relative to bold fmri. Neuroimage 49, 1340–1349 (2010).

    Article  Google Scholar 

  45. Yacoub, E., Ugurbil, K. & Harel, N. The spatial dependence of the poststimulus undershoot as revealed by high-resolution BOLD- and CBV-weighted fMRI. J. Cereb. Blood Flow. Metab. 26, 634–644 (2006).

    Article  Google Scholar 

  46. Petridou, N. & Siero, J. C. Laminar fMRI: what can the time domain tell us? NeuroImage 197, 761–771 (2017).

    Article  Google Scholar 

  47. Stephan, K. E. et al. Laminar fMRI and computational theories of brain function. Neuroimage 197, 699–706 (2017).

    Article  Google Scholar 

  48. Lawrence, S. J. D., Formisano, E., Muckli, L. & de Lange, F. P. Laminar fMRI: applications for cognitive neuroscience. Neuroimage 197, 785–791 (2019).

    Article  Google Scholar 

  49. Peirce, J. W. Psychopy—psychophysics software in python. J. Neurosci. Methods 162, 8–13 (2007).

    Article  Google Scholar 

  50. Mildner, T. et al. Mapping of arterial transit time by intravascular signal selection. NMR Biomed. 27, 594–609 (2014).

    Article  Google Scholar 

  51. Marques, J. P. et al. MP2RAGE, a self bias-field corrected sequence for improved segmentation and T1-mapping at high field. Neuroimage 49, 1271–1281 (2010).

    Article  Google Scholar 

  52. Friston, K. J. et al. Statistical parametric maps in functional imaging: a general linear approach. Hum. Brain Mapp. 2, 189–210 (1994).

    Article  Google Scholar 

  53. Polimeni, J. R., Renvall, V., Zaretskaya, N. & Fischl, B. Analysis strategies for high-resolution UHF-fMRI data. NeuroImage 168, 296–320 (2018).

    Article  Google Scholar 

  54. Worsley, K. Statistical analysis of activation images. Funct. MRI: Introd. Methods 14, 251–270 (2001).

    Google Scholar 

  55. Greenhouse, S. W. & Geisser, S. On methods in the analysis of profile data. Psychometrika 24, 95–112 (1959).

    Article  Google Scholar 

  56. Girden, E. R. Anova: Repeated Measures (Sage, 1992).

  57. Huber, L., Uludağ, K. & Möller, H. E. Non-bold contrast for laminar fMRI in humans: CBF, CBV, and CMRO2. Neuroimage 197, 742–760 (2017).

    Article  Google Scholar 

  58. Huber, L. et al. Ultra-high resolution blood volume fMRI and bold fMRI in humans at 9.4 T: capabilities and challenges. Neuroimage 178, 769–779 (2018).

    Article  Google Scholar 

  59. Huber, L. et al. Techniques for blood volume fMRI with VASO: from low-resolution mapping towards sub-millimeter layer-dependent applications. Neuroimage 164, 131–143 (2018).

    Article  Google Scholar 

  60. Fischl, B. & Dale, A. M. Measuring the thickness of the human cerebral cortex from magnetic resonance images. Proc. Natl Acad. Sci. 97, 11050–11055 (2000).

    Article  CAS  Google Scholar 

  61. Williams, S. M. & Goldman-Rakic, P. S. Characterization of the dopaminergic innervation of the primate frontal cortex using a dopamine-specific antibody. Cereb. cortex 3, 199–222 (1993).

    Article  CAS  Google Scholar 

  62. Blazejewska, A. I., Fischl, B., Wald, L. L. & Polimeni, J. R. Intracortical smoothing of small-voxel fMRI data can provide increased detection power without spatial resolution losses compared to conventional large-voxel fMRI data. Neuroimage 189, 601–614 (2019).

    Article  Google Scholar 

  63. Gagnon, L. et al. Quantifying the microvascular origin of bold-fMRI from first principles with two-photon microscopy and an oxygen-sensitive nanoprobe. J. Neurosci. 35, 3663–3675 (2015).

    Article  CAS  Google Scholar 

  64. Fracasso, A., Luijten, P. R., Dumoulin, S. O. & Petridou, N. Laminar imaging of positive and negative BOLD in human visual cortex at 7t. Neuroimage 164, 100–111 (2018).

    Article  Google Scholar 

  65. Geyer, S., Weiss, M., Reimann, K., Lohmann, G. & Turner, R. Microstructural parcellation of the human cerebral cortex–from Brodmann’s post-mortem map to in vivo mapping with high-field magnetic resonance imaging. Front. Hum. Neurosci. 5, 19 (2011).

    Article  Google Scholar 

  66. Ding, S.-L. et al. Comprehensive cellular-resolution atlas of the adult human brain. J. Comp. Neurol. 524, 3127–3481 (2016).

    Article  Google Scholar 

  67. Stüber, C. et al. Myelin and iron concentration in the human brain: a quantitative study of MRI contrast. Neuroimage 93, 95–106 (2014).

    Article  Google Scholar 

  68. Avants, B. B., Epstein, C. L., Grossman, M. & Gee, J. C. Symmetric diffeomorphic image registration with cross-correlation: evaluating automated labeling of elderly and neurodegenerative brain. Med. Image Anal. 12, 26–41 (2008).

    Article  CAS  Google Scholar 

  69. Cox, R. W. AFNI: software for analysis and visualization of functional magnetic resonance neuroimages. Comput. Biomed. Res. 29, 162–173 (1996).

    Article  CAS  Google Scholar 

Download references

Acknowledgements

We thank A. Arnsten for guidance on experimental design and interpretation. We thank B. Poser and D. Ivanov for the 3D-EPI readout that is used in the VASO sequence here. We thank A.H. Hall and K. Chung for administrative support of human volunteer scanning. We thank S. Kashyap for helpful tips on adjusting manual initial registration used to generate Supplementary Fig. 2. We thank S. Marrett and D. Handwerker for technical advice and support. We thank A. Thomas and T. Riddle for support with data formatting, organization and sharing. Portions of this study used the high-performance computational capabilities of the Biowulf Linux cluster at the National Institutes of Health (biowulf.nih.gov). The research was funded by the National Institute of Mental Health Intramural Research Program (no. ZIAMH002783) to P.A.B., which also funded E.S.F., L.H., D.C.J. and P.J.M. During the latest periods of data analysis, L.H. was funded from the NWO VENI project 016.Veni.198.032.

Author information

Author notes

  1. Laurentius Huber

    Present address: MR-Methods Group, Maastricht Brain Imaging Center, Department of Cognitive Neuroscience, Faculty of Psychology and Neuroscience, Maastricht University, Maastricht, The Netherlands

Authors and Affiliations

  1. Section on Functional Imaging Methods, Laboratory of Brain and Cognition, National Institute of Mental Health, Bethesda, Maryland, USA

    Emily S. Finn, Laurentius Huber, David C. Jangraw, Peter J. Molfese & Peter A. Bandettini

Authors
  1. Emily S. Finn
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Laurentius Huber
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. David C. Jangraw
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Peter J. Molfese
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Peter A. Bandettini
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

E.S.F. conceptualized the study, designed the task paradigm, collected the data, analyzed the data, generated visualizations and wrote the original manuscript. L.H. conceptualized the study, contributed to the task paradigm design, designed and optimized the data acquisition and analysis methodology, collected the data, analyzed the data, generated visualizations and wrote portions of the original manuscript. D.C.J. contributed to the task paradigm design, generated visualizations and provided comments on the manuscript. P.J.M. analyzed the data, generated visualizations and provided comments on the manuscript. P.A.B. supervised study conceptualization, design and interpretation, and provided comments on the manuscript.

Corresponding author

Correspondence to Emily S. Finn.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Integrated supplementary information

Supplementary Figure 1 Graphical depiction of the analysis pipeline.

(A) Multiple runs are collected per session with two different types (‘alpha / rem’, or manipulation versus maintenance, and ‘act / non-act’, or action versus non-action). Each run consists of interleaved images with (VASO) and without (BOLD) blood nulling. (B) Nulled and not-nulled images are motion corrected separately for the whole session. It is manually checked that the motion traces of nulled and not-nulled images are matching. (C) The time series are sorted by imaging contrast. (D) Runs of the same type are averaged within contrast. (E) Time series from all trials are averaged based on their task condition. In the 2 × 2 design used here, this results in 4 average trial types. Nulled images are corrected for BOLD contamination with a time-wise division of not-nulled images to provide a clean VASO contrast. (F) Based on the approximate location of activation in the low-resolution functional localizer, two layer ROIs are manually drawn on the T1-EPI anatomical images. (G) Time series for all four task conditions are extracted from the ROI of superficial and deeper layers for BOLD and VASO contrasts, resulting in 8 time series for each contrast, or 16 total time series.

Supplementary Figure 2 Reproducibility of activation loci in two subjects across two different days.

(A) Reproducibility of activation loci in one subject scanned with the axial imaging protocol on two different days. The same parts of gray matter are engaged in the task across the same acquisition protocol on two different days (first session, top row; second session, bottom row). The consistency of the locus of activation is within the accuracy of the registration quality. (B) Reproducibility of activation loci in one subject across the axial and sagittal imaging protocols acquired on two different days. The same parts of gray matter are engaged in the task across the two different acquisition protocols (axial-like, top row; and sagittal-like, bottom row) acquired on two different days. The consistency of the locus of activation is within the accuracy of the registration quality.

Supplementary Figure 3 Layer ROIs for individual subjects and sessions.

Masks of superficial and deeper layers for each subject/session that were used to extract the timecourses in Figs. 2, 3, and Supplementary Figs. 5–7. The grayscale background contrast refers to the T1-contrast in functional VASO data. Here the T1-contrast in the functional data is used to identify superficial and deeper voxels (manually drawn green and red masks). Asterisks (* and **) are used to denote test-retest scans from the same participant (i.e., sub-06 is the same as sub-01; sub-10 is the same as sub-07).

Supplementary Figure 4 Comparison of cortical depth with cytoarchitectonic cortical layers.

(A) Grayscale Nissl-stained images of the dlPFC from the Ding Atlas (ref. 66 in Methods) were converted into single-slice NIFTI format. Borderlines of the GM-CSF surface and GM-WM surface are manually drawn in the same space. Layers are extracted with LAYNII as described in Methods subsection “Layering for sagittal protocol”. The position of the internal pyramidal layer is visible as a small bump (blue arrow). Just superior to it is the outer band of Bailager (black arrow), which is taken as a landmark to identify the approximate transition area between layer III (green arrow) and layer V (red arrow). This layer has an elevated myelin content, which comes along with reduced T1 values in MRI contrast (ref. 67 in Methods); as such, this landmark is also visible in in-vivo T1 and T2* weighted profiles (see black arrows in panel B-D). The reason to show the histological slice in panel A) is to confirm that the transition area between later III and V is approximately 40-60% of the cortical thickness away from WM as visible in myelin-sensitive MRI data. This confirmation is helpful because the relative depth of cytoarchitectonic cortical layers may differ across brain areas. Based on this depth calibration, we believe it is appropriate to pool fMRI signal (E-F) and time courses from two independent subsets of voxels, superficial voxels (superficial ≈ 30-40% of the cortex) and deeper voxels (deeper ≈ 30-40%) as shown in Supplementary Fig. 3. This dip between cytoarchitectonic layers III and IV is used here as the border between so-called “superficial layers” and “deeper layers” which show different responses to the different functional task contrasts (E-F). (Data in panels A-D come from single subjects: data in panel A are from the Ding Atlas, which is based on a single adult female brain, and data in panels B-F come from the same subject in the present study. Error bars in panel A-D refer to the standard error across columns within the area of interest [without layer smoothing]. In panel A this represents the variability across a distance of 2 mm sampled at 7270-8949 points. In panel B, this represents the variability of across a distance of 17 mm sampled at 29-45 points.) The data in panel E-F) refer to the manipulation (alphabetize) versus. remember runs. 10 trial of each condition where acquired per run. The data shown here are extracted after averaging across 4 runs with (20 trials each). This totals to 80 trials. The green contrast refers to the time average across the delay period, whereas the red contrast refers to the time average across the response period (Fig. 1b). As in Fig. 4, the green contrast refers to the signal difference of alphabetizing minus remember trials ((Salpha-Sremember)/Srest) and the red refers to the signal difference of the response period minus the rest period ((Sresponse-Srest)/Srest) taken from the inter stimulus interval. For more subjects, see Fig. 4. The uncertainty area in panel E-F refers to the standard error of the mean across the ten respective trials (upon averaging across four runs).

Supplementary Figure 5 Activity timecourses and statistical contrasts for two layers and all four trial types using the BOLD contrast.

(A) Left panel: mean BOLD percent signal change in superficial layers (top) and deeper layers (bottom) for the first contrast, which consisted of trial types manipulation (‘alpha’) and maintenance (‘rem’). Right panel: mean BOLD percent signal change in superficial layers (top) and deeper layers (bottom) for the second contrast, which consisted of trial types action (‘act’) and non-action (‘non-act’). Lines represent mean and shaded area represents 95 percent confidence intervals for the mean (determined via bootstrapping with 1,000 iterations) across n = 15 sessions (13 unique subjects). While BOLD is more sensitive than VASO, it is less spatially specific. Discrepancies between BOLD timecourses shown here and VASO timecourses shown in Fig. 2—for example, it appears from BOLD that superficial-layer activity is higher at the time of the response in action versus non-action trials (panel A, top right), but this difference is absent in VASO (see Fig. 2a, top right)—are likely due to asymmetrical layer cross-talk, in which signal from deeper layers “leaks” into superficial layers because of draining-vein artifacts (but not the other way around). Thus BOLD results must be interpreted with caution. (B) Top: Superficial-layer BOLD activity during maintenance (‘rem’) trials subtracted from activity during manipulation (‘alpha’) trials [purple line]. The largest difference can be seen during the delay period. Bottom: Deeper-layer BOLD activity during non-action (‘non-act’) trials subtracted from activity during action (‘act’) trials [teal line]. The largest difference can be seen during the response period. Lines represent mean and shaded area represents 95 percent confidence intervals for the mean (determined via bootstrapping with 1,000 iterations) across n = 15 sessions (13 unique subjects).

Supplementary Figure 6 Average VASO timecourses for each individual subject-session for manipulation and maintenance in superficial layers.

Data from n = 15 individual subject-sessions are shown. In the majority of subjects, it can be clearly seen that activity is higher for manipulation (‘alpha’, blue lines) over maintenance (‘rem’, green lines) specifically during the delay period (gray shading).

Supplementary Figure 7 Average VASO timecourses for each individual subject-session for action and non-action in deeper layers.

Data from n = 15 individual subject-sessions are shown. In the majority of subjects, an increase in activity can be clearly seen at the timepoint when response-related signal would be expected to peak (gray dashed line) in action (‘act’, red lines) but not non-action (‘non-act’, orange lines) trials. This peak occurs at the end of the response period (gray shading), or approximately 6 seconds after the appearance of the probe (accounting for behavioral and hemodynamic delay).

Supplementary Figure 8 Discussion of columnar variance.

Layer-dependent activity maps in Fig. 4 exhibit some variance across the columnar dimension. As such, there is no response activity for VASO at the location of the green arrow in panel A and there is no response activity at the location of the blue arrow in panel D). Panels B-C) and E-F) depict how the layer-dependent activity maps depend on the arbitrary threshold chosen. It can be seen in panel A) that here is indeed sub-threshold VASO response activity at the location of the green arrow. Panel G) depicts columnar coordinates that are used to unfold the fMRI signal along the cortical ribbon. The estimation of the columnar coordinates and the unfolding is done in LAYNII with the program LN_COLUMNAR_DIST. (https://layerfmri.com/quick-example-of-cortical-unfolding-in-laynii/). Note that the batch of cortex investigated here is highly folded with respect to the external magnetic field. This means that the BOLD signal change can be substantially variable dependent on the columnar position along the sulcus (refs. 63,64 in Methods). At the location of the blue arrow in panels J,L-O, it can be seen that at the location of larger draining veins the BOLD sensitivity is reduced and the functional responses exhibit larger inhomogeneities.

Supplementary information

Supplementary Figs. 1–8.

Reporting Summary

Supplementary Video 1

ROI versus FEF: sub-05. Slice-by-slice visualization of the selected ROI for sub-05. The area circled in green is the selected DLPFC ROI, while the area circled in blue refers to what we believe is the frontal eye field.

Supplementary Video 2

ROI versus FEF: sub-06. Slice-by-slice visualization of the selected ROI for sub-06. The area circled in green is the selected DLPFC ROI, while the area circled in blue refers to what we believe is the frontal eye field.

Supplementary Video 3

ROI versus FEF: sub-16. Slice-by-slice visualization of the selected ROI for sub-16. The area circled in green is the selected DLPFC ROI, while the area circled in blue refers to what we believe is the frontal eye field.

Supplementary Video 4

ROI versus FEF: sub-17. Slice-by-slice visualization of the selected ROI for sub-17. The area circled in green is the selected DLPFC ROI, while the area circled in blue refers to what we believe is the frontal eye field.

Supplementary Video 5

ROI versus FEF: sub-19. Slice-by-slice visualization of the selected ROI for sub-19. The area circled in green is the selected DLPFC ROI, while the area circled in blue refers to what we believe is the frontal eye field.

Supplementary Video 6

ROI versus FEF: sub-21. Slice-by-slice visualization of the selected ROI for sub-21. The area circled in green is the selected DLPFC ROI, while the area circled in blue refers to what we believe is the frontal eye field.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Finn, E.S., Huber, L., Jangraw, D.C. et al. Layer-dependent activity in human prefrontal cortex during working memory. Nat Neurosci 22, 1687–1695 (2019). https://doi.org/10.1038/s41593-019-0487-z

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/s41593-019-0487-z

This article is cited by

Search

Advanced search

Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing