Basic Neuroscience
High-density multielectrode array with independently maneuverable electrodes and silicone oil fluid isolation system for chronic recording from macaque monkey

https://doi.org/10.1016/j.jneumeth.2012.08.019Get rights and content

Abstract

Chronic multielectrode recording has become a widely used technique in the past twenty years, and there are multiple standardized methods. As for recording with high-density array, the most common method in macaque monkeys is to use a subdural array with fixed electrodes. In this study, we utilized the electrode array with independently maneuverable electrodes arranged in high-density, which was originally designed for use on small animals, and redesigned it for use on macaque monkeys while maintaining the virtues of maneuverability and high-density. We successfully recorded single and multiunit activities from up to 49 channels in the V1 and inferior temporal (IT) cortex of macaque monkeys. The main change in the surgical procedure was to remove a 5 mm diameter area of dura mater. The main changes in the design were (1) to have a constricted layer of heavy silicone oil at the interface with the animal to isolate the electrical circuit from the cerebrospinal fluid, and (2) to have a fluid draining system that can shunt any potential postsurgical subcranial exudate to the extracranial space.

Highlights

► Fluid drain system for reducing the postsurgical intracranial pressure. ► Heavy silicone oil system for guarding electrical circuit from CSF backflow. ► Maneuverable electrode array implant to macaque IT cortex from temporal surface.

Introduction

Recent studies have suggested that sensory information is represented in distributed manners in cortical areas using various levels of functional structures, such as neurons, columns, and other structures larger than columns (Haxby et al., 2001, Howard et al., 2009, Hung et al., 2005, Tsunoda et al., 2001, Tsao et al., 2003, Tsao et al., 2006). One approach to investigate distributed codes of sensory information is to record activities from many neurons by penetrating an electrode repeatedly and to combine these activities together for analysis of population responses (Hung et al., 2005, Kiani et al., 2007, Zhang et al., 2011). However, to relate population activities to functional structures, we need to map population responses in cortical space. Although fMRI and optical imaging techniques are used to map cortical activities (Haxby et al., 2001, Howard et al., 2009, Tsunoda et al., 2001), measured signals do not necessarily correspond neural activities because they are secondary hemodynamic responses induced by neural activities. Especially, we lose the information buried in temporal structure of neural activities, such as correlation and synchrony among the cells that are involved in sensory information representation (Eckhorn et al., 1988, Gray et al., 1989). Multiple electrode arrays are only the available technique so far to investigate population neural activities with high temporal and spatial resolution (Blake and Merzenich, 2002, deCharms et al., 1999, Nicolelis et al., 2003, Hochberg et al., 2006). deCharms and colleagues specially developed a densely arranged multi-electrode array to address response patterns in spatial scales equivalent to columnar representation (1999). This array led them to map spectrotemporal representation of sound input in primary auditory cortex (Blake and Merzenich, 2002), and to analyze plastic changes of the functional map of auditory cortex in rats (Blake and Merzenich, 2002, Blake et al., 2005). Our goal of the present study is to make this type of array feasible to map sensory information representation in macaque monkeys.

Because the fine electrodes (75 μm in diameter) adequate for above high-density recording cannot penetrate the thick dura mater of macaques, we took the approach of making an opening in the dura mater over the implant sites, as was done on some other macaque choric recordings (Jackson and Fetz, 2007, Nicolelis et al., 2003). Leaving a dura window under the recording device will increase the risk of causing (1) biological reaction at the dural scar, (2) cerebrospinal fluid invasion into the electrical circuit, and will often limit the number of feasible electrodes and the duration of successful recording. To reduce these risks, we developed a new device which (1) has a fluid drain system at the interface between the chamber and the subcranial space over the implant site and (2) has a layer of highly viscous silicone oil to maintain isolation of the electrical circuit, in the present study.

To map spatial patterns of activity across the cortical surface, we have to take into account recent studies showing that nearby neurons generally behave very differently (DeAngelis et al., 1999, Reich et al., 2001, Sato et al., 2009, Vinje and Gallant, 2000, Yen et al., 2007). For example, Yen and colleagues revealed that even in cat V1, where columnar functional structure has been known for decades, responses of nearby cells are uncorrelated while the animal is looking at natural scenes, thus, a single cell activity cannot be treated as the representative activity in the vicinity of the recording electrode. In the present study, to make the array feasible to map spatial patterns of activity, we examined whether electrodes with large exposed tips can detect neural activity to reflect common properties across the cells within the local region. In short, the newly designed electrode array achieved higher neural activity yields in macaque monkeys. In practice, we were able to show representative spatial patterns of activity obtained from inferior temporal (IT) cortex of macaque monkeys which gave qualitatively consistent results with spatial map obtained by optical intrinsic signal imaging.

Section snippets

General experimental conditions

Six rhesus monkeys (Macaca mulatta) were used in this study. One monkey was used for semi-acute electrophysiology experiments to search for the optimal electrode tip configuration. Five other monkeys were tested with the actual chronic multielectrode array. Chronic array was implanted on V1 cortex of two monkeys, posterior IT of another monkey and anterior IT of the other two. Electrophysiological recording experiments were conducted while the monkeys were under neuroleptoanalgesia (NLA). The

Results

The multielectrode array developed by deCharms et al. is unique in realizing high-density spatial configuration of electrodes (spacing, 350 μm) and post-implant depth adjustment feature simultaneously. In their array, electrodes were penetrated through densely arranged metal guide tubes. Electrical signals picked up by an electrode were delivered to the head-amplifier through an electrical contact between uninsulated part of the electrode and inner surface of the metal guide tube. Since the

Summary

In the present study, we introduced an implant recording system with densely arranged and independently maneuverable electrodes applicable to macaque cortices and achieved high yield of active electrodes. High yield of active electrodes enable us to map cortical activities across cortical surface.

To make implantation feasible for exposed cortices in macaques, we made multiple critical modifications. The oil system achieved (1) great reduction of the electrical shunting problem, (2) maintaining

Acknowledgements

We thank Ms. Kei Hagiya for technical assistance, Dr. Hideyuki Watanabe for making graphical software, and Ms. Toshiko Ikari for providing comments on the manuscript. N.M. was supported by Grant-in-AID for Young Scientists (B) 21700442 from the Ministry of Education, Culture, Sports, Science, and Technology (MEXT). M.T. was supported by Grant-in-AID for Scientific Research 22300137 and Grant-in-AID for Innovative Areas, “Face Perception and Recognition”.

References (31)

  • A. Arieli et al.

    Dural substitute for long-term imaging of cortical activity in behaving monkeys and its clinical implications

    J Neurosci Methods

    (2002)
  • R.C. deCharms et al.

    A multielectrode implant device for the cerebral cortex

    J Neurosci Methods

    (1999)
  • A.H. Bell et al.

    Relationship between functional magnetic resonance imaging-identified regions and neuronal category selectivity

    J Neurosci

    (2011)
  • D.T. Blake et al.

    Changes of AI receptive fields with sound density

    J Neurophysiol

    (2002)
  • D.T. Blake et al.

    Experience-dependent plasticity in S1 caused by noncoincident inputs

    J Neurophysiol

    (2005)
  • G.C. DeAngelis et al.

    Functional micro-organization of primary visual cortex: receptive field analysis of nearby neurons

    J Neurosci

    (1999)
  • R. Eckhorn et al.

    Coherent oscillations: a mechanism of feature linking in the visual cortex? Multiple electrode and correlation analyses in the cat

    Biol Cybern

    (1988)
  • I. Fujita et al.

    Columns for visual features of objects in monkey inferotemporal cortex

    Nature

    (1992)
  • C.M. Gray et al.

    Oscillatory responses in cat visual cortex exhibit inter-columnar synchronization which reflects global stimulus properties

    Nature

    (1989)
  • J.V. Haxby et al.

    Distributed and overlapping representations of faces and objects in ventral temporal cortex

    Science

    (2001)
  • L.R. Hochberg et al.

    Neuronal ensemble control of prosthetic devices by a human with tetraplegia

    Nature

    (2006)
  • J.D. Howard et al.

    Odor quality coding and categorization in human posterior piriform cortex

    Nat Neurosci

    (2009)
  • D.H. Hubel et al.

    Receptive fields and functional architecture of monkey striate cortex

    J Physiol

    (1968)
  • C.P. Hung et al.

    Fast readout of object identity from macaque inferior temporal cortex

    Science

    (2005)
  • A. Jackson et al.

    Compact movable microwire array for long-term chronic unit recording in cerebral cortex of primates

    J Neurophysiol

    (2007)
  • View full text