Multi-scale brain function of tactile prediction processing: from cortical layers to whole brain

2435 

Abstract Submission 

Yinghua Yu¹, Masaki Fukunaga², Laurentius (Renzo) Huber³, Peter Bandettini^4,3, Norihiro Sadato⁵, Jiajia Yang¹

¹Faculty of Interdisciplinary Science and Engineering in Health Systems, Okayama University, Okayama, Okayama, Japan, ²Section of Brain Function Information, National Institute for Physiological Sciences, Okazaki, Aichi, Japan, ³Functional MRI Core Facility, National Institute of Mental Health, Bethesda, MD, US, ⁴Section on Functional Imaging Methods, National Institute of Mental Health, Bethesda, MD, US, ⁵Research Organization of Science and Technology, Ritsumeikan University, Kyoto, Kyoto, Japan

Yinghua Yu  
Faculty of Interdisciplinary Science and Engineering in Health Systems, Okayama University
Okayama, Okayama, Japan

Masaki Fukunaga  
Section of Brain Function Information, National Institute for Physiological Sciences
Okazaki, Aichi, Japan

Laurentius (Renzo) Huber, PhD  
Functional MRI Core Facility, National Institute of Mental Health
Bethesda, MD, US

Peter Bandettini, Ph.D.  
Section on Functional Imaging Methods, National Institute of Mental Health|Functional MRI Core Facility, National Institute of Mental Health
Bethesda, MD, US|Bethesda, MD, US

Norihiro Sadato  
Research Organization of Science and Technology, Ritsumeikan University
Kyoto, Kyoto, Japan

Jiajia Yang, PhD  
Faculty of Interdisciplinary Science and Engineering in Health Systems, Okayama University
Okayama, Okayama, Japan

Human sensory processing is considered heirachally organized, having a series of discrete stages from a single cortical layer to the whole brain. In our previous studies [1,2], we demonstrated this bidirectional hierarchy processing at layer levels in the primary somatosensory cortex (S1) using laminar fMRI at 7T. We found that sensory input to S1 evoked activity in the middle layers, while prediction input yielded activity in the superficial and deep layers. Beyond the sensory cortex, our follow-up study [3] found the double-peak activity feature across midcingulate cortex (MCC) layers for tactile temporal prediction, indicating feedforward and feedback activity. Because of limited brain coverage of the VASO laminar fMRI approach, in order to explore connectivity between these regions subjects performed the prediction task both when using limited coverage laminar fMRI and whole brain lower resolution BOLD contrast.

Ten participants were asked to participate in the fMRI experiment, which consisted of one layer-specific medial surface of the cerebral hemisphere slab run and one whole brain run. We used three tactile temporal prediction (TP) tasks and one control task with random sensory input (RS) (Figure 1A). The participant was asked to predict when the left index finger will be poked in three TP tasks. The layer-specific fMRI data acquisition procedures at 7T were used as described in our previous study [1,2]. The interleaved BOLD and VASO contrasts – obtained as separate yet concomitant time series [4]. The effective TR is 4400 ms. The nominal resolution was 0.76 mm across cortical depths with 1.4-mm thick slices. For the whole brain data acquisition, multiband T2 ∗ -weighted echo planar imaging (EPI) sequence parameters were used: TR = 1785 ms, TE = 22 ms, axial slices = 90, iPAT = 2, multiband factor = 3 and spatial resolution = 1.6 × 1.6 × 1.6 mm3. The fMRI data preprocessing and general linear model analysis were conducted using AFNI and FSL. Laminar analyses were conducted with the open software suite LayNii [5]. The study protocol was approved by the local medical ethics committee at the Okayama University Hospital and National Institute for Physiological Sciences.

We found that all four tasks activated widespread brain regions, including contralateral S1, bilateral S2, MCC, and other high-level regions. Task-state functional connectivity analysis and showed different connectivity patterns between prediction and prediction error tasks (Figure 1B). Furthermore, we confirmed that the layer-specific activity of MCC and SMA modulations across tasks could be detected (Figure 1C).

In the present study, we asked the participants to perform the same tactile temporal prediction tasks in a standard whole brain and a laminar BOLD/VASO fMRI session at 7T. We found that the prediction error modulated the functional connectivity between the somatosensory cortex and medial prefrontal cortex (i.e., MCC and SMA) from the whole brain session. We also found that the activity in MCC and SMA upper layers are selectively modulated by the error processing which occurred during the delayed period of the TPoff_short, but the increased upper layer activity was temporally turned by extending the off-beat interval in the TPoff_long task. One possible interpretation is that somewhat inhibitory processing may predominate during the longer delay period. These findings reflected the tactile temporal prediction processing at different spatial scales, which would be expected to provide insights into understanding our brain prediction system.

Non-BOLD fMRI ¹

Perception: Tactile/Somatosensory ²

Cortical Layers

FUNCTIONAL MRI

HIGH FIELD MR

NORMAL HUMAN

Other - laminar fMRI