Format

Send to

Choose Destination
Neuroimage. 2018 Feb 1;166:325-334. doi: 10.1016/j.neuroimage.2017.10.059. Epub 2017 Oct 28.

Overlapping frontoparietal networks for tactile and visual parametric working memory representations.

Author information

1
Neurocomputation and Neuroimaging Unit (NNU), Department of Education and Psychology, Freie Universität Berlin, Habelschwerdter Allee 45, 14195 Berlin, Germany; Berlin School of Mind and Brain, Humboldt-Universität zu Berlin, Unter den Linden 6, 10099 Berlin, Germany. Electronic address: yuan-hao.wu@fu-berlin.de.
2
Neurocomputation and Neuroimaging Unit (NNU), Department of Education and Psychology, Freie Universität Berlin, Habelschwerdter Allee 45, 14195 Berlin, Germany; Berlin School of Mind and Brain, Humboldt-Universität zu Berlin, Unter den Linden 6, 10099 Berlin, Germany.
3
Neurocomputation and Neuroimaging Unit (NNU), Department of Education and Psychology, Freie Universität Berlin, Habelschwerdter Allee 45, 14195 Berlin, Germany; Bernstein Center for Computational Neuroscience Berlin, Humboldt-Universität zu Berlin, Unter den Linden 6, 10099 Berlin, Germany; Institute of Cognitive Science, University of Osnabrück, Wachsbleiche 27, 49090 Osnabrück, Germany.
4
Neurocomputation and Neuroimaging Unit (NNU), Department of Education and Psychology, Freie Universität Berlin, Habelschwerdter Allee 45, 14195 Berlin, Germany; Department of Neurophysics, Max Planck Institute for Human Cognitive and Brain Sciences, Postfach 500355, 04103 Leipzig, Germany.
5
Neurocomputation and Neuroimaging Unit (NNU), Department of Education and Psychology, Freie Universität Berlin, Habelschwerdter Allee 45, 14195 Berlin, Germany; Berlin School of Mind and Brain, Humboldt-Universität zu Berlin, Unter den Linden 6, 10099 Berlin, Germany; Bernstein Center for Computational Neuroscience Berlin, Humboldt-Universität zu Berlin, Unter den Linden 6, 10099 Berlin, Germany.

Abstract

Previous working memory (WM) research based on non-human primate electrophysiology and human EEG has shown that frontal brain regions maintain frequencies of flutter stimulation across different sensory modalities by means of a supramodal parametric WM code. These findings imply that frontal regions encode the memorized frequencies in a sensory-unspecific, quantitative format. Here, we explored which brain regions maintain information about frequencies provided by different sensory modalities at the level of activity pattern across fMRI voxel populations. Moreover, we sought evidence for a supramodal multivariate WM representation. Participants maintained the same set of frequencies of tactile vibration and visual flicker for a 6 s WM delay in a frequency discrimination task. A support vector regression model for multivariate pattern analysis was applied. We observed that sensory cortices were only selective for memoranda of their corresponding modalities, while frontoparietal regions exhibited distinguishable activity patterns to memorized frequencies regardless of sensory modality. A common multivariate code was not evident in our data. Collectively, we show that mnemonic representations for stimulus frequencies are maintained throughout the cortical hierarchy, in line with the suggested transformation of information across different representational formats. Although evidence for a supramodal multivariate code is absent, our findings underpin the generalized role of the frontoparietal cortex for maintaining quantitative information across sensory modalities.

KEYWORDS:

Abstract quantity; Frequency discrimination task; Parametric; Vibrotactile; Working memory; fMRI MVPA

[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Elsevier Science
Loading ...
Support Center