Characterization of hippocampal and cerebellar activity

J. Karhu and C. Tesche

A widely held view of cortical somatosensory organization is that the sensory scene is completed by serial, hierarchical processing of gradually more complex stimulus features. This view is supported by the presence of large, somatotopically organized primary receiving cortices with converging projections to smaller association areas. Supporting neurophysiological evidence of this view is provided by responses to median nerve stimulation recorded by MEG arrays: Initial cortical activation in primary somatosensory cortex (SI) about 20 ms after stimulation is followed by processing in second somatosensory cortex (SII) at ~100 ms. However, the anatomical connectivity of the somatosensory system also suggests simultaneous participation of widely separated cortical areas in the early processing of sensory input. We have reported previously the detection of evoked neuromagnetic responses from human primary (SI) and secondary (SII) somatosensory cortices 20-30 ms after intermittent median nerve stimulation. These results indicate that SII and / or associated cortices in parietal operculum, often viewed as higher order processing areas for somatosensory perception, are coactivated with SI during the early processing of intermittent somatosensory input. 

Activation of human hippocampus is usually taken to follow the final stages of sensory processing relevant to recognition and memory encoding at latencies of ~ 300 ms. We observed magnetoencephalographic signals consistent with activation of hippocampus and surrounding cortices as early as 120-150 ms after intermittent median nerve stimulation. These responses followed the observed activation of both SI and SII cortices, suggesting a complex flow of somatosensory information to hippocampal formation. A gradual build-up of activity was also observed at 100500 ms following omissions. We hypothesize that this anticipatory "readiness field" reflects the identification of an omission by networks which characterize the timing of somatosensory input. The increased salience of the subsequent stimulus as a potentially novel event is reflected in the enhanced responses observed in hippocampal formation. 

Single cell and intracranial field potential recordings in hippocampus reveal 4-12 Hz oscillatory (theta) activity which is remarkably prominent in the rodent brain. There is evidence in rodents for a 4-7 Hz "cognitive" component associated with attentive behavior, and an 8-12 Hz "motor" component associated with movement. Reports of hippocampal theta in man are very rare. The first observations in normal human hippocampal formation of both 4-7 Hz theta and activity at 8-12 Hz have been reported recently by one of us utilizing MEG data recorded during a mental arithmetic task. We have now continued the characterization of human theta with recordings of MEG responses during a working memory task. Spectral components at 4-7 and 8-12 Hz were observed in hippocampal formation both in stimulus- and movement-triggered averaged evoked responses. These observations support the initial report of oscillatory phenomena in normal human hippocampal formation in these frequency bands.

The traditional view of cerebellum is a structure that modifies and synchronizes elements of motor performance. Recent evidence indicates that human cerebellum is involved in a wide range of non-motor sensory and cognitive functions. A common feature in these diverse motor and non-motor tasks may be the capacity of cerebellar neuronal circuits to anticipate and process sensory input with high temporal acuity. We have presented evidence for this hypothesis from observations of scalp magnetic fields evoked by neuronal population activity in human cerebellum. Electrical stimulation of the median nerve at the wrist elicited stimulus-locked cerebellar responses with oscillatory components at 6-12 Hz and 25-35 Hz. Activity commenced prior to the initiation of stimulation, with sustained responses continuing even during the epochs which contained random omissions of stimuli. Moreover a slow shift of cerebellar evoked responses was initiated in the omission epoch and preceded an overt, expected stimulus.

We have also continued the characterization of cerebellar evoked responses to somatosensory stimulation with recordings for repetitive median nerve and finger stimulation, for predictable and random omissions and for stimulation at various interstimulus intervals and attentional levels. We observed responses to the first stimuli after omissions which were clearly larger than the responses to subsequent stimuli at all interstimulus intervals for random omissions, but were reduced when the train of stimuli were interrupted by a regular, predictable pattern of omissions. Alpha-range activity during the omission epochs was also suppressed for regular omissions compared to values for random omissions. Moreover anticipatory alpha-range cerebellar responses just prior to the first stimulus after an omission and those elicited by the first stimuli were clearly diminished when attention was consciously directed outside the somatosensory system.

The anticipatory slow shift of averaged cerebellar evoked responses was similarly suppressed. Slow shifts of activity in cortical and subcortical neural networks have been attributed both to the control of involuntary vs. focused attention and to the expectancy of a cue stimulus. Presuming that cerebellum participates to the control of attentional networks as suggested by recent functional imaging studies, we hypothesize that cerebellar stimulus-locked alpha-range activity may index the level of expectancy and attention in cerebellar neuronal populations. These observations support engagement cerebellar networks in the processing of temporal features of somatosensory input independent of motor performance or response. The moment-to-moment cerebellar neuronal representations of real and anticipated stimuli may be utilized to optimize the performance of large-scale sensory and integrative systems. The observed short-term maintenance of a cerebellar template for predictable somatosensory input may thus reflect a physiological substrate for fine-grained temporal tuning of large-scale neuronal networks.


RH 4/10/99