Search

The Mental Attributes Profiling System was developed in 2002 (Laouris and Makris, Proceedings of multilingual & cross-cultural perspectives on Dyslexia, Omni Shoreham Hotel, Washington, D.C, 2002), to provide a multimodal evaluation of the learning potential and abilities of young children's brains. The method is based on the assessment of non-verbal abilities using video-like interfaces and was compared to more established methodologies in (Papadopoulos, Laouris, Makris, Proceedings of IDA 54th annual conference, San Diego, 2003), such as the Wechsler Intelligence Scale for Children (Watkins et al., Psychol Sch 34(4):309-319, 1997). To do so, various tests have been applied to a population of 134 children aged 7-12 years old. This paper addresses the issue of identifying a minimal set of variables that are able to accurately predict the learning abilities of a given child. The use of Machine Learning technologies to do this provides the advantage of making no prior assumptions about the nature of the data and eliminating natural bias associated with data processing carried out by humans. Kohonen's Self Organising Maps (Kohonen, Biol Cybern 43:59-69, 1982) algorithm is able to split a population into groups based on large and complex sets of observations. Once the population is split, the individual groups can then be probed for their defining characteristics providing insight into the rationale of the split. The characteristics identified form the basis of classification systems that are able to accurately predict which group an individual will belong to, using only a small subset of the tests available. The specifics of this methodology are detailed herein, and the resulting classification systems provide an effective tool to prognose the learning abilities of new subjects.

To help reduce the gap between the cellular physiology of motoneurons (MNs) as studied "bottom-up" in animal preparations and the "top-down" study of the firing patterns of human motor units (MUs), this article addresses the question of whether motoneuron adaptation contributes to muscle fatigue. Findings are reviewed on the intracellularly recorded electrophysiology of spinal MNs as studied in vivo and in vitro using animal preparations, and the extracellularly recorded discharge of MUs as studied in conscious humans. The latter "top-down" approach, combined with kinetic measurements, has provided most of what is currently known about the neurobiology of muscle fatigue, including its task and context dependencies. It is argued that although the question addressed is still open, it should now be possible to design new "bottom-up" research paradigms using animal preparations that take advantage of what has been learned with the use of relatively noninvasive quantitative procedures in conscious humans.

Associations were quantified between the control force and fatigue-induced force decline in 22 single fast-twitch-fatigable motor units of 5 deeply anesthetized adult cats. The units were subjected to intermittent stimulation at 1 train/s for 360 s. Two stimulation patterns were delivered in a pseudo-random manner. The first was a 500-ms train with constant interpulse intervals. The second pattern had the same number of stimuli, mean stimulus rate, and stimulus duration, but the stimulus pulses were rearranged to increase the force produced by the units in the control (prefatigue) state. The associations among the control peak tetanic force of these units, 3 indices of fatigue, and total cumulative force during fatiguing contractions were dependent, in part, on the stimulation pattern used to produce fatigue. The associations were also dependent, albeit to a lesser extent, on the force measure (peak vs. integrated) and the fatigue index used to quantify fatigue. It is proposed that during high-force fatiguing contractions, neural mechanisms are potentially available to delay and reduce the fatigue of fast-twitch-fatigable units for brief, but functionally relevant, periods. In contrast, the fatigue of slow-twitch fatigue-resistant units seems more likely to be controlled largely, if not exclusively, by metabolic processes within their muscle cells.

Spinal recurrent inhibition via Renshaw cells and proprioceptive feedback via skeletal muscle and muscle spindle afferents have been hypothesized to constitute a compound feedback system [Windhorst (1989) Afferent Control of Posture and Locomotion; Windhorst (1993) Robots and Biological Systems--Towards a New Bionics]. To assess their detailed functions, it is necessary to know their dynamic characteristics. Previously we have extensively described the properties of signal transmission from motor axons to Renshaw cells using random motor axon stimulation and data analysis methods based thereupon. Using the same methods, we here compare these properties, in the cat, with those between motor axons and group Ia muscle spindle afferents in terms of frequency responses and nonlinear features. The frequency responses depend on the mean rate (carrier rate) of activation of motor axons and on the strength of coupling between motor units and spindles. In general, they are those of a second-order low-pass system with a cut-off at fairly low frequencies. This contrasts with the dynamics of motor axon-Renshaw cell couplings which are those of a much broader band-pass with its peak in the range of c. 2-15 Hz [Christakos (1987) Neuroscience 23, 613-623]. The second-order non-linearities in motor unit-muscle spindle signal lines are much more diverse than those in motor axon-Renshaw cell couplings. Although the average strength of response declines with mean stimulus rate in both subsystems, there is no systematic relationship between the amount of non-linearity and the average response in the former, whilst there is in the latter. The qualitative appearance of motor unit-muscle spindle non-linearities was complicated as was the average response to motor unit twitches. Thus, whilst Renshaw cells appear to dynamically reflect motor output rather faithfully, muscle spindles seem to signal local muscle fibre length changes and their dynamics. This would be consistent with the hypothesis that the two feedback pathways monitor different state variables determining the production of muscle force: neural input and length and its changes. Specifically, the dynamic properties of both subsystems may combine favourably to decrease the risk of instability (tremor) in the motoneuron-muscle spindle loop.

1. The main purpose of this study was to quantify the adaptation of spinal motoneurons to sustained and intermittent activation, using an extracellular route of stimulating current application to single test cells, in contrast to an intracellular route, as has been used previously. In addition, associations were tested between firing rate properties of the tested cells and other type (size)-related properties of these cells and their motor units. 2. Motoneurons supplying the medial gastrocnemius muscle of the deeply anaesthetized cat were stimulated for 240 s with microelectrodes which passed sustained extracellular current at 1.25 times the threshold for repetitive firing. Many cells were also tested following a rest period with intermittent 1 s current pulses (duration 600 ms) at the same relative stimulus strength. Cell discharge was assessed from the EMG of the motor unit innervated by the test neuron. The motoneurons and their motor units were assigned to four categories (i.e. types FF, FR, S and F; where F = FF + FR) based on conventional criteria. In all, twenty F (16 FF, 4 FR) and fourteen S cells were studied with sustained stimulation. Thirty of these cells (17 F, 13 S) and an additional two cells (1 F, 1 S) were studied with intermittent stimulation. 3. The mean threshold current required for sustained firing for a period of > or = 2 s was not significantly different for F and S cells. However, most of the other measured parameters of motoneuron firing differed significantly for these two cell groups. For example, at 1.25 times the threshold current for repetitive firing, the mean firing duration in response to 240 s of sustained activation was 123 +/- 88 s (+/- S.D.) for F cells vs. 233 +/- 19 s for S cells. These values were significantly longer than those from a comparable, previously reported study that employed intracellular stimulation. With intermittent stimulation, the firing durations of F and S cells were not significantly different from each other. 4. All cells exhibited a delay from the onset of current to the first spike, followed by a brief accelerating discharge that was followed by a slower drop in firing rate. Some cells (21 of 34 with sustained activation; 20 of 32 with intermittent) exhibited doublet discharges (interspike intervals < or = 10 ms) that were intermingled with the more predominant singlet discharges. Doublets were more common in the S cell type.(ABSTRACT TRUNCATED AT 400 WORDS)

The fatigue of fast-twitch, glycolytic mammalian motor units [i.e., type FF; nomenclature of (3)] is dependent, in part, on the stimulation regimen (total number of stimuli, frequency, duty cycle, temporal patterning of stimuli, etc.) used to induce fatigue. To study the effect of the temporal pattern of the stimulus train on the rate and extend of fatigue in single FF units, one theoretically acceptable approach would be to use each motor unit as its own control: i.e., a sequential testing with two fatigue tests that differ only in the temporal organization of their stimuli. The purpose of this communication is to provide evidence that such an approach is not feasible when studying FF units, due to the delayed recovery of force following their repetitive activation. It was shown that 1/s activation of single FF units for only 15 or 45 s with intermittent 40-Hz, 300-ms duration trains significantly reduced their force response to a double-pulse shock for several hours. This finding suggests that in studies designed to test for the effects of different stimulation patterns on the fatigue of single motor units, deeply anaesthetized, reduced animal preparations are not appropriate models for the sequential application of different stimulation regimens to fast-twitch, glycolytic, mammalian motor units.

1. The main purpose of this study was to examine the effects of two subtly different stimulus patterns on the force developed by fast-twitch, fatiguable motor units in a cat hindlimb muscle during control (pre-fatigue) and fatiguing contractions. 2. The peak force and the force-time integral responses of nineteen high fatigue (FF) and three intermediate fatigue (FI) motor units of the tibialis posterior muscle in five deeply anaesthetized adult cats were measured at selected times during the course of a 360-s fatigue test. 3. The fatigue test involved a pseudo-random alternation of two patterns of stimulation. One pattern (regular) was composed of a train of stimuli with constant interpulse intervals, set at 1.8 x the twitch contraction time of each unit (interval range, 27-51 ms), and delivered for 500 (or 400) ms. For the total (FF + FI) motor-unit sample, the mean (+/- S.D.) stimulation frequency was 26 +/- 4 Hz (range, 19-37 Hz). The other stimulus pattern (optimized) consisted of three initial stimuli with short (10 ms) interpulse intervals, followed by a constant interpulse-interval train that was adjusted (interval range, 29-62 ms; frequency, 23 +/- 5 Hz; frequency range, 16-36 Hz) such that the total train had the same number of pulses, and the same average frequency and duration as the regular train. 4. The stimulus trains were delivered at 1 s-1 for 360 s, using three-train sequences of each pattern, randomly alternating with one another. The response of the third train in each sequence was selected for the force measurements. The force profile obtained from the fatigue test was subsequently decomposed into two profiles: one attributable to regular and one to optimized stimulation. 5. During the initial responses to the fatigue test, the optimized stimulus pattern produced significantly more force than the regular stimulus pattern. For FF units, the mean increase in peak force (141%) was significantly greater than the increase in the force-time integral (59%). 6. All motor units exhibited an initial potentiation of peak force with the regular stimulation pattern, whereas peak force declined monotonically with the optimized pattern. In contrast, the force-time integral potentiated in the first 30 s for both regular and optimized stimulus patterns. 7. Each motor unit maintained an increased force response to optimized stimulation during the fatigue test, with the greatest relative increase occurring about 120 s into the test, well after the potentiation effect had subsided.(ABSTRACT TRUNCATED AT 400 WORDS)

The purpose of this study was to quantify the changes in motor-unit action potentials (MUAP) and force during a standard motor-unit fatigue test. MUAP waveforms were characterized by the measurement of amplitude, duration, area, and shape (as reflected in a coefficient of proportionality). Fatigue-resistant motor units exhibited small, but statistically significant, changes in MUAP amplitude and area during the fatigue test, whereas fatigable motor units displayed variable changes in MUAP amplitude, duration, and area. For all motor-unit types, the coefficient of proportionality did not change, and hence the change in MUAP area was proportional to the combined changes in amplitude and duration. The between- and within-train changes in MUAP were also distinct for the fatigue-resistant and fatigable motor units. Although several mechanisms could be responsible for the changes in the MUAP as the fatigue test proceeded, the dissociation of the time courses for MUAP and force indicated that these MUAP changes were not the principal reason for the decline in force under these conditions.

Measurement systems used in the collection and processing of laboratory data must be calibrated periodically to obtain accurate results. Because calibration factors can change over time or may be reset to optimize measurements for specific tests, care must be taken to assure that calibration factors and data are aligned correctly. Users should be able to process current data or re-process older data using appropriate calibration factors. The alignment of calibration factors and data should occur in a simple, automatic and transparent way. This document describes one approach to calibration procedures and computer programs used to collect, process, document, measure and display laboratory data. The examples are from our neurophysiology laboratory, where investigators study the mammalian spinal cord and peripheral neuromuscular system. Typical calibration problems, some workable solutions, and computer programs (described in pseudocode) are presented.

A full circuit description is provided for a triggering module used to assist a small laboratory computer in digitizing muscle force- and EMG waveforms. During the stimulation of individual motor units using a standard fatigue test, a train of 13 pulses are delivered at a rate of 40 pps either intracellularly to a motor neuron, or extracellularly to functionally isolated single motor axons from among divided ventral-root nerve filaments. Trains are delivered at a rate of 1/s for the duration of the test, which may range from 120 to 3600 s. Both the force and EMG profiles undergo changes during such tests and the quantification of parameters associated with their waveforms are of interest to neurobiologists. The triggering module allows a typical small laboratory computer to capture user-selected waveforms and thereby reduces the programming problems, timing constraints, storage requirements and analysis time associated with obtaining these parameters. The versatile circuit may be easily adapted to solve similar data-acquisition problems. The method was implemented on an Apple Macintosh II computer but can also be applied to other systems equipped with appropriate software and a data-acquisition card.

A software-based system is presented for feature extraction of compound, action-potential (EMG) recordings from single motor units. It simplifies and automates the measurement and analysis of several parameters of the action potential: peak-to-peak amplitude, total duration, peak-to-peak duration, and total area. The software is based on a simple algorithm that first finds the baseline (isoelectric line; including a noise level) of each single EMG potential (waveform) and then searches for the minimum and maximum values in the array of data points representing it. The algorithm searches in both directions starting from the minimum and maximum data points (the waveform peaks) to find the beginning and ending points of the waveform. Using the indices (i.e., array-point numbers) of the four data points provided by the algorithm, the desired features are extracted and/or calculated and saved in a standard-format spreadsheet. The algorithm has a potentially widespread usefulness in a broad array of electrophysiological studies.

The experiments were performed on 18 albino newborn rats of both sexes. The postnatal development of the air-righting-reaction (ARR) of the rats was tested for 4 different falling heights (30, 40, 50 and 60 cm) with 3 different loading conditions: artificial alteration of the head, thorax and pelvis mass, respectively. The reaction was analyzed in terms of a three-segment model of the rat (three rotatable parts: head-thorax-pelvis with only one degree of freedom between successive segments). Loading of the head resulted in a delayed development of the first phase of the ARR, whereas loading of the trunk or pelvis slowed the second phase of development. In general, it appears that loading of a body segment results in a delayed development of the rotation of that segment during free fall. Furthermore, loading of a caudal region was associated with an earlier rotation of cranial segments. The completion of the ARR maturation process showed no dependency on either the location or the size of the load, suggesting that the postnatal development of the ARR is pre-programmed.

The aim of this study was to describe the ontogenesis of the air-righting reaction (ARR) in rats. The first experiment was performed on 6 newborn albino rats of both sexes and followed the development of the ARR over postnatal days 1-21. The degree of rotation achieved after falling from different heights was quantified according to a rating scheme. It appeared that the air-righting reaction is effected by a spiral movement which spreads in a cranio-caudal direction. The reaction develops between postnatal day 8 and 18. On postnatal day 10 only a few animals are able to turn their heads, this being possible only from a falling height of 60 cm and corresponding to a falling time of 350 ms. A rapid development of the reaction was found between days 10 and 14. The second experiment on 8 rats involved the use of immobilization in order to isolate the mechanisms that trigger the ARR. The immobilization prevented neck-torso rotation, torso-pelvis rotation, and both rotations in different animals. Despite the disruption of important (afferent) feedback systems, the reaction developed within the same age period as in control rats. Thus, the Magnus "chain reflex hypothesis' as basis for the ARR is rejected in favor of a central motor program hypothesis.

Renshaw cell responses to random motor axon stimulation exhibit second-order non-linearities in that they depend on the occurrence of a preceding stimulus, although these non-linearities are not strong enough to significantly depress the coherence. However, higher-order non-linearities have not been checked for so far. This is carried out here. Lumbosacral Renshaw cells were recorded with micropipettes in anaesthetized cats. Their responses to random (pseudo-Poisson) stimulation of motor axons in peripheral nerves or ventral roots were quantified by calculating peristimulus-time histograms of various sorts, conventional and conditional. Conventional peristimulus-time histograms were computed with respect to all the stimuli in a train. Conditional peristimulus-time histograms were calculated with respect to "test" stimuli which were sorted out (by computer) from the original stimulus train when they were preceded by "conditional" stimuli at average intervals of delta 1 or delta 2 or both. These conditioned responses were compared with those to be expected from hypothetical linear superposition. Renshaw cell responses showed small third-order non-linearities to pairs of conditioning stimuli at small intervals (up to some tens of milliseconds before the test stimuli). These third-order effects were smaller than each of the second-order non-linearities elicited by any of the single-conditioning stimuli. Also, further higher-order non-linearities were apparent, but of little average significance. Hence, the non-linearities in Renshaw cell responses to random inputs are essentially of second-order and fairly small.

The system between cutaneous (suralis) afferents and dorsal horn neurons was studied for comparison with studies previously performed on the motor axon-Renshaw cell system, using the same methods. In anaesthetized or decerebrated cats, 27 dorsal horn neurons of segments L5 to S1 were recorded extracellularly in depths of 1-2.3 mm from cord dorsum. Cutaneous afferents in branches of the ipsilateral suralis nerve were stimulated with sequences of randomly occurring electrical pulses at two levels of mean rate. The responses of the dorsal horn neurons to the stimuli were evaluated in the frequency and time domain. Calculation of coherence, gain and phase functions (via spectral analysis) showed that the frequency response depended on the precise pattern on cell discharge and could vary from broad-band to low-pass or occasionally band-pass characteristics. There were minor differences in these characteristics with those of Renshaw cells. A special type of nonlinear analysis, using conditional peristimulus-time histograms, showed that the responses to test stimuli were facilitated, depressed or both by conditioning stimuli occurring some tens to a few hundred milliseconds before. Early and late response components could be conditioned individually and differently. Exponential fits to such conditioning curves yielded two time constants for depression (means of 21 and 94 ms) and one for facilitation (14 ms). Similar conditioning effects and time constants were previously found for the motor axon-Renshaw cell system although a few differences were apparent. By analogy, it is suggested that part of the long-lasting conditioning effects (with long time constants) are probably due to presynaptic mechanisms.(ABSTRACT TRUNCATED AT 250 WORDS)

1. We investigated the responses of cat lumbosacral Renshaw cells to pseudo-Poison stimulus sequences (of three different mean rates) delivered to motor axons in ventral roots or various muscle nerves. The Renshaw cell responses were evaluated by computation of peristimulus time histograms (PSTHs). 2. PSTHs computed with respect to all the stimuli showed, before the reference time, near-constant bin contents corresponding to the mean firing probability (rate), and an initial excitatory component (increase in discharge probability) after the reference time, followed by a small but longer-lasting reduction of firing rate. These two response components were strongly correlated linearly. It is suggested that the postexcitatory rate reduction is predominantly due to afterhyperpolarization. 3. In general, Renshaw cell responses to any stimulus in a stimulus train depended upon the stimulation history. In the averaged record, the response to the second of a pair of stimuli was affected by the first stimulus independently of intervening (random) stimuli. Very often, the second response showed a long-lasting depression (from 25 to greater than 250 ms). In a number of cases a briefer facilitating effect preceded the depression. 4. These conditioning effects were largely homosynaptic, i.e., confined to the particular input channel that was stimulated. This was shown by stimulating two different nerves (or nerve branches) with independent random patterns of similar mean rates and determining the cross-conditioning exerted by one input channel on the excitatory effects of the other. At small intervals between conditioning and test stimuli of some tens of milliseconds, a facilitatory effect could often be seen, which almost certainly reflected spatial summation. However, the subsequent depressant effect was largely accounted for by the postexcitatory rate reduction consequent to the conditioning stimulus in the parallel channel. Autoconditioning was still present. 5. The amount of facilitation and depression as well as their balance depended on the average Renshaw cell response. This in turn depended, at each mean stimulus rate, on the strength of synaptic coupling between an input channel and the cell, and on the mean stimulus rate, declining with an increase in mean rate. That is, the facilitation increased and the depression decreased with decreasing synaptic coupling and increasing mean stimulus rate. 6. Several factors may contribute to facilitation and depression; these are discussed with respect to their relative quantitative significance.(ABSTRACT TRUNCATED AT 400 WORDS)

In pentobarbitone-anesthetized cats, the spike sequences of dorsal horn neurons were recorded in response to random stimulation of branches of the suralis nerve. Combined frequency- and time-domain analysis was performed on the stimulus and spike trains. Coherence function estimates computed by spectral analysis were compared with peristimulus time histograms (PSTHs). The cell responses were divided into 4 main types: PSTHs with a single high and narrow peak were associated with broad-range high coherence; PSTHs with two (or sometimes 3) distinct peaks concurred with a coherence which was high a low frequencies, low at intermediate ones and higher again at high frequencies; broad unstructured PSTH peaks of varying height concurred with coherence declining from high values at low frequencies to lower values at higher ones; and small and broad PSTH peaks were associated with generally low coherence. Thus, the variation of coherence with frequency depends on the precise pattern of cell discharge.

In 9 adult anaesthetized cats, 22 lumbosacral Renshaw cells recorded with NaCl-filled micropipettes were activated by random stimulation of ventral roots or peripheral nerves. The stimulus patterns had mean rates of 9.5-13 or 20-23 or 45 pulses per second and were pseudo-Poisson; short intervals below ca. 5 ms (except in two cases) were excluded. The Renshaw cell responses were evaluated by two kinds of peristimulus-time histograms (PSTHs). "Conventional" PSTHs were calculated by averaging the Renshaw cell discharge with respect to all the stimuli in a train. These PSTHs showed an early excitatory response which was often followed by a longer-lasting slight reduction of the discharge probability. These two response components were positively correlated. "Conditional" PSTHs were determined by averaging the Renshaw cell discharge with respect to the second ("test") stimulus in pairs of stimuli which were separated by varied intervals, delta. The direct effect of the first "conditional" response was subtracted from the excitation following the second ("test") stimulus so as to isolate the effect caused by the second stimulus per se. After such a correction, the effect of the first "conditioning" stimulus showed pure depression, pure facilitation or mixed facilitation/depression. Analysis of such conditioning curves yielded two time constants of facilitation (ranges: ca. 4-35 ms and 93-102 ms) and two of depression (ranges: ca. 7-25 ms and 50-161 ms). It is concluded that these time constants are compatible with processes of short-term synaptic plasticity known from other synapses. Other processes such as afterhyperpolarization and mutual inhibition probably are of less importance.

Lumbosacral Renshaw cells were activated by random stimulation of motor axons in muscle nerves or ventral roots. The stimulus patterns had mean rates of 9.5-13 or 20-23 pulses per second. The Renshaw cell responses were evaluated by two kinds of peristimulus-time histograms. "Conventional" peristimulus-time histograms were calculated by averaging the cell discharge with respect to all the stimuli in a train. "Conditional" peristimulus-time histograms were determined by averaging the cell discharge with respect to the second ("test") stimulus in pairs of stimuli which were separated by varied intervals. The effects of the conditioning stimuli were evaluated after correcting for the effect of linear superposition of the conditioning and test stimuli. The conventional peristimulus-time histograms showed an excitatory response which often consisted of two distinct components: a narrow and high "early" peak and a broad and low "late" elevation of firing probability. The early and late excitatory components were conditioned in different ways. Whereas the late component was virtually always depressed, the early component showed three patterns: (1) uniform depression; (2) uniform facilitation; (3) a mixture of depression and facilitation. Frequency responses (coherence and gain estimates) were also calculated separately for the cell discharges underlying either the early or the late components. The estimates for the "late spikes" showed a stronger decline with increasing frequency than those for the "early spikes". The origin of the different conditioning effects probably lies in a combination of pre and postsynaptic factors. They may play a role in tremor mechanisms.

We here present a method to study the interaction of parallel neural input channels regarding their effects on a neurone. In particular, the method allows to disclose the effects of oligosynaptic pathways that may exist in parallel to direct monosynaptic connections to the cell. Two (or more) inputs (nerves) are stimulated with random patterns of stimuli. The response of the cell to these patterns is evaluated by the computation of peristimulus-time histograms (PSTHs). One of the two stimulus trains is selected as the one to yield reference events for the PSTH computation. From this stimulus train are selected those stimuli as reference events which are preceded, at defined mean intervals, by stimuli in the same or a parallel channel. These "conditioning" stimuli are determined (1) separately from each single stimulus train and (2) concomitantly from the two trains as events occurring simultaneously in both. The effects exerted by these various conditioning events on the effects of the "test" pulses on the cell response yield insights into the interactions between the two (or more) inputs. These methods are demonstrated on spinal Renshaw cells activated by independent random stimulation of two muscle nerves and on dorsal horn neurones responding to cutaneous nerve stimulation.

Cat spinal Renshaw cells were activated by stimulating muscle nerves or ventral roots with random (pseudo-Poisson) patterns of brief electrical stimuli. This input pattern is optimal for a comparative study in both the frequency- and time-domain. The frequency-dependent variable of particular interest in this study was the coherence as a measure of the degree to which signal transmission is linear and noise-free; it was estimated via spectral analysis. Time-domain analysis consisted of calculating peri-stimulus time histograms in order to estimate the amount of nonlinearity in the cell responses to pairs of stimuli. The main result was that the amount of nonlinearity measured in this way did not profoundly depress the coherence. Two types of peri-stimulus time histogram were calculated: the "conventional" peri-stimulus time histogram (as a reference) computed with respect to all the stimuli in a train, and the "conditional" peri-stimulus time histogram computed with respect to the second in pairs of stimuli which were separated from each other by varied intervals delta. The latter type of peri-stimulus time histogram showed that Renshaw cell responses to stimuli were conditioned by preceding stimuli, which could facilitate (at small delta s) and/or more often depress (up to several hundreds of milliseconds) the subsequent responses in a nonlinear manner. The objective of this study was to test the hypothesis that nonlinear characteristics contribute significantly to depress the coherence from its optimal value (1).(ABSTRACT TRUNCATED AT 250 WORDS)