Distributed neural networks of tactile working memory

Abstract

Microelectrode recordings of cortical activity in primates performing working memory tasks reveal some cortical neurons exhibiting sustained or graded persistent elevations in firing rate during the period in which sensory information is actively maintained in short-term memory. These neurons are called "memory cells". Imaging and transcranial magnetic stimulation studies indicate that memory cells may arise from distributed cortical networks. Depending on the sensory modality of the memorandum in working memory tasks, neurons exhibiting memory-correlated patterns of firing have been detected in different association cortices including prefrontal cortex, and primary sensory cortices as well. Here we elaborate on neurophysiological experiments that lead to our understanding of the neuromechanisms of working memory, and mainly discuss findings on widely distributed cortical networks involved in tactile working memory.

Keywords: Monkey; Single unit recording; Somatosensory; Tactile; Working memory.

Figure 1

Diagram of two working memory tasks. In a trial of the hapti-chaptic (HH) task (top), the monkey touches the tactile sample (without seeing it), one of two objects (rods) differing by a surface feature (orientation of parallel ridges, vertical vs. horizontal). At the end of the delay period that follows the sample touch, the animal is presented with the two rods simultaneously and must choose by touching and pulling the one that matches the sample to get a certain amount of fluid as a reward. In a trial of the visual-haptic (VH) task (bottom), the animal first views an icon of vertical or horizontal stripes, and then at the end of the delay, the animal makes a behavioral choice by touching and pulling one of the two rods with ridges oriented in the same direction as the stripes of the sample icon. (The figure is adapted from Zhou et al. 2007, Cerebral Cortex)

Figure 2

Spike rasters and histograms from a sample- and delay-differential unit in SI cortex through horizontal (hor)- and vertical (ver)-sample trials. The unit's receptive field is marked (white) on the hand diagram, the unit's position (red triangle) on the outlined parietal section on upper left. (The figure is adapted from Zhou et al. 2007, Cerebral Cortex)

Figure 3

Rasters and perievent histograms (bin size = 50 msec) of a cell recorded in the hand area of SI cortex in the HH task, showing the activity in both sample and choice periods. Left: The time-locking event for histograms is the first touch of the sample. The average firing rate (non-differential) for both rods in the sample period increases significantly from the baseline level. Right: The time-locking event for histograms is the onset of the last touch prior to the pull of the chosen rod. The red raster in each trial indicates the onset of the choice (the pull). The firing rate of the cell is significantly higher in touch of the horizontal rod (p < 0.001). This cell shows the choice-only differential activity. (The figure is adapted from Wang et al. 2012, Journal of Cognitive Neuroscience)

Figure 4

Average frequency histograms (bin size 500 ms) from an SI cell showing differential activity in two periods, cue and choice. The locking event for the left histograms is the onset of the visual cue, and for the right histograms it is the first contact of the matching object. The middle part of the delay is omitted from the histograms (dashed line). The cell favors the horizontal ridges at the choice (P < 0.01). Accordingly, it favors the horizontal visual cue (P < 0.05). (The figure is adapted from Zhou and Fuster 2000, PNAS)

Figure 5

The differential delay activity of a frontal cell in the HH task. Rasters and histograms are aligned to the beginning of the delay. Blue rasters and histograms indicate the neural activity in horizontal trials, and black ones, in vertical trials. Only correct trials are included.