Concomitant classical conditioning of the rabbit nictitating membrane and eyelid responses: correlations and implications

Simultaneous recordings of muscle unit activity from the muscles of the left and right eyelids (M. obicularis oculi) and recordings of the movement of the left nictitating membrane (NM) were taken during classical conditioning in the rabbit using a tone CS paired with an airpuff UCS to the left cornea. The unconditioned eyelid responses were found to be bilateral. The conditioned eyelid responses were also bilateral in most animals. Both the conditioned and unconditioned eyelid responses were larger on the left side. The conditioned responses of the left and right eyelids and the left NM were found to increase in magnitude and decrease in latency from the onset of the CS over training trials in almost the exactly same manner (correlations as high as .99). Behaviorally, the three responses could occur independently, suggesting that the cranial nuclei which control them (left abducens/accessory abducens, left facial nucleus, right facial nucleus) are not strongly coupled. Thus, for the learned response, it is suggested that the three nuclei are controlled by a common central system. This finding has implications for the nature of the engram--the essential neuronal circuitry encoding the learned response.

Selective, naloxone-reversible morphine depression of learned behavioral and hippocampal responses

Morphine administered intravenously causes immediate and complete abolition of a simple learned response (classically conditioned nictitating membrane extension in rabbit) and of the associated learning-induced increase in hippocampal neuron activity. Both effects are completely reversed by low doses of naloxone. Morphine has no effect at all on behavioral performance of the unconditioned reflex response.

Auditory signal detection and decision processes in the nervous system

Neuronal unit activity was recorded from auditory nuclei, and dorsal hippocampus, and the cerebellum in rabbits behaviorally detecting a threshold-level constant intensity white noise stimulus. Stimulus-evoked neuronal unit activity was present and identical on both detection and nondetection trials in auditory nuclei but was dichotomous in the hippocampus and in the cerebellum, the latter two systems predicting the occurrence of behavioral detection. It is concluded that the behavioral absolute auditory threshold is not determined by differential activation of neurons in the primary auditory relay nuclei.

Initial localization of the memory trace for a basic form of learning

Electrophysiological recording of neuronal unit activity during paired training trials from various regions of the ipsilateral cerebellum in rabbits well trained in the classically conditioned eyelid/nictitating membrane response have revealed both stimulus-evoked responses and responses that form an amplitude/temporal model of the learned behavioral response. Ablation of the ipsilateral, lateral cerebellum completely and permanently abolished the behavioral conditioned response in well-trained animals but had no effect at all on the unconditioned reflex response. In marked contrast, conditioned responses were easily trained in the eye contralateral to the cerebellar lesion. We suggest that at least part of the essential neuronal plasticity that codes the learned response may be localized to the cerebellum.

Hippocampal cellular plasticity during extinction of classically conditioned nictitating membrane behavior

Hippocampal unit responses were recorded during extinction of classically conditioned nictitating membrane (NM) behavior in the rabbit. Prior studies have shown that during the acquisition phase of nictitating membrane conditioning, the frequency of hippocampal cell firing increases at a faster rate (across trials) than learned behavior. Results reported here show that, during the early phases of extinction, conditioned hippocampal unit responses decrement at a faster rate than learned NM behavior. Furthermore, only certain components of the conditioned hippocampal unit response display robust spontaneous recovery across successive days of extinction training. In all, results show that changes in the activity of hippocampal neurons predict changes in learned behavior over trials of conditioning and extinction, and demonstrate that hippocampal cellular activity is particularly sensitive to stimulus configurations or environmental contingencies that produce changes in behavior.

Reciprocal anatomical connections between anterior thalamus and cingulate--retrosplenial cortex in the rabbit

Reciprocal anatomical connections between posterior limbic (i.e. cingulate-retrosplenial) cortex and anterior thalamus in the rabbit were investigated using horseradish peroxidase histochemistry. Results showed that the cells of origin for corticofugal fibers are contained only within deep cortical layers, while thalamofugal projections terminate only within superficial laminae. That is, only layer VI neurons of cingulate-retrosplenial cortex project to thalamic regions, while cortical layers I and IV are the primary targets of thalamic afferents.

Hippocampal unit-behavior correlations during classical conditioning

The correspondence that develops over the course of classical conditioning between the temporal distribution of increased unit activity in the rabbit hippocampus and the amplitude--time distribution of the behavioral nictitating membrane response is analyzed. Results reveal a high degree of correspondence between neural and behavioral measures. The real time correlation between the within-trial probability and increased hippocampal unit discharge and amplitude--time course of the nictitating membrane response grows substantially with learning. Further analyses reveal that this apparent increase in correlation results from a growth in amount of hippocampal unit activity per se (i.e., a differentiation of the hippocampal unit response from background firing rates), rather than an increase in the correspondence between cellular and behavioral measures (i.e. a repatterning of hippocampal discharges to more accurately code spatio-temporal aspects of the behavioral response). These and other results indicate that the neuronal 'temporal model' of the behavioral response either develops within the hippocampus from the first few conditioning trials or develops first in entorhinal cortex to subsequently influence hippocampal discharge patterns. On the other hand, the increase in amount of hippocampal unit activity developing with conditioning appears to occur within the hippocampus.

Alterations in spontaneous miniature potential activity during habituation of a vertebrate monosynaptic pathway

Intracellular recordings of spontaneous miniature synaptic potentials were made from motoneurons of the isolated spinal cord of the frog, as a function of habituation of the monosynaptic pathway originating with lateral column fibers. The frequency and amplitude of spontaneous miniature potentials were used to assess the possibility of several proposed mechanisms underlying habituation in this system. These studies provide clear evidence in eliminating the role of transmitter depletion, incomplete vesicle filling and receptor desensitization in the habituation process occurring within this vertebrate central nervous system.

Effect of the interstimulus (CS-UCS) interval on hippocampal unit activity during classical conditioning of the nictitating membrane response of the rabbit (Oryctolagus cuniculus)

Neuronal activity (multiple unit) was recorded from the dorsal hippocampus during classical conditioning of the nictitating membrane response in the rabbit. Groups of subjects were trained with conditioned stimulus/unconditioned stimulus (CS-UCS) interstimulus intervals of 250 msec, 150 msec, or 50 msec, and an additional group received unpaired presentations of the CS and UCS, with a tone CS and a corneal air puff UCS. Increased hippocampal unit activity and nictitating membrane conditioned responses (CRS) occurred only in the 250 msec and 150 msec groups. Both the temporal distribution of hippocampal unit activity and the averaged CR topographies were strongly and similarly influenced by the CS-UCS interval, with both neuronal and behavioral response peaks occurring near the time of UCS onset. The peak of hippocampal unit activity preceded the peak of the nictitating membrane CR by 30-60 msec. When the 250 msec group was shifted to a CS-UCS interval of 500 msec, the peaks of both the hippocampal unit response and the behavioral CR topography shifted in the same direction. The shift appeared to occur sooner in the hippocampal unit response. It is suggested that the hippocampus may function to provide a "temporal map" for learned behaviors.

Classical conditioning of the rabbit (Oryctolagus cuniculus) nictitating membrane response, with electrical brain stimulation as the unconditioned stimulus

Rabbits were given classical conditioning training by using paired tone CS (conditioned stimulus) and brain-shock UCS (unconditioned stimulus), with the stimulating electrode localized in the vicinity of the abducens (6th nerve) nucleus such that the electrical stimulus elicited a low-threshold nictitating membrane (NM) extension response as the unconditioned response. Eight of the 27 animals developed clear conditioned NM responses to the tone CS. Control procedures, e.g., subsequent explicity unpaired training, argue against sensitization, pseudoconditioning, and "kindling" as possible explanations. There was no clear-cut anatomical differentiation of electrode-tip locations between learners and nonlearners. Both learners and nonlearners were subsequently trained with paired tone-corneal air puff, and neuronal unit activity was recorded from the stimulating electrode. Animals that had learned with the prior brain-shock UCS exhibited substantially greater increases in neuronal unit activity during air-puff training, which suggests the existence of a differentially effective anatomical-physiological substrate for conditionability.

Reciprocal anatomical connections between hippocampus and subiculum in the rabbit evidence for subicular innervation of regio superior

Anatomical connections between the dorsal hippocampus and subiculum were examined in the rabbit, using horseradish peroxidase (HRP) and autoradiographic methods. A previously undescribed pathway was found to project from the dorsal prosubicular-subicular region to dorsal hippocampal cell fields CA1 and CA2. Autoradiographic findings showed that subicular afferents travel via two routes. One pathway projected through the alveus and stratum oriens, with results suggesting collateral input to the basal dendritic pyramidal cell region. The other projection coursed through the stratum lacunosum-moleculare with apparent termination onto CA1 and CA2 apical dendrites. Regions of subiculum providing afferents to hippocampus were compared with subicular areas receiving efferent terminations from hippocampal CA1 and CA3 cell zones. Distribution of hippocampal-subicular terminations were regionally distinct from subicular retrograde cell fields in rostral areas of the subicular complex, extended over a much wider area of subiculum than was seen for retrograde-labeled cells, and was cytoarchitectonically organized. In total, findings indicated that a reciprocal anatomical relationship exists between dorsal hippocampus and subiculum in the rabbit.

Neuronal unit activity in the abducens nucleus during classical conditioning of the nictitating membrane response in the rabbit (Oryctolagus cuniculus)

Neuronal unit activity was recorded from the abducens (6th nerve) nucleus, the "final common path," during classical conditioning of the nictitating membrane (NM) response in the rabbit, with the use of a tone conditioned stimulus, an air puff unconditioned stimulus (UCS), 250-msec interstimulus interval, and 60-sec intertrial interval. Animals were given 2 days of conditioning training (104 trials in eight blocks per day) and 1 day of extinction. Control animals were given comparable periods of stimulus presentations, explicitly unpaired. Activity of small clusters of units--"multiple unit" recording--was compared with the amplitude-time course of the NM response. Between-blocks comparisons of neural and behavioral responses indicated an essentially perfect correlation during acquisition of the conditioned response (Day 1, r = .99; Day 2, r = .98) and a slightly lower correlation during extinction (r = .93) for the conditioning animals. Within-blocks comparisons indicated a close correspondence between the histograms of unit activity and the amplitude-time course of the NM response for the conditioning animals in all phase of training and for the control animals in the UCS trial blocks.

Medial septal lesions retard classical conditioning of the nicitating membrane response in rabbits

Lesions of the medial septum were produced in 7 of 14 rabbits prior to classical conditioning of the nictitating membrane response. Lesions significantly altered the hippocampal electroencephalogram, attenuated conditioned hippocampal unit responses, and slowed the behavioral rate of acquisition. The contrast of the behavioral results with those of studies of massive septal or hippocampal ablation suggests a functional subdivision of the septo-hippocampal system in learning.

Neuronal plasticity in the limbic system during classical conditioning of the rabbit nictitating membrane response. II: Septum and mammillary bodies

Neuronal unit activity was recorded from several limbic system structures during classical conditioning of the rabbit nictiating membrane response to a tone CS. Air puff to cornea was used as a UCS. The present and past investigations of hippocampal activity using this paradigm show a neuronal plasticity which develops early in training, increases rapidly and shifts forward in time as behavioral conditioning develops. Recordings from the lateral septal region demonstrate the same within-trial pattern of unit discharge seen in hippocampus, indicating a projection of hippocampal plasticity over precommissural fornix pathways. Medial septal neurons, on the other hand, respond in an excitatory manner to the onsets of tone and air puff stimulation. While unit discharges seen in hippocampus and lateral septum occur only during the paired (learning) paradigm, medial septal activity is identical under both paired and unpaired (control) conditions. The latter fact lends support for a sensory interpretation of medial septal responses, and is consistent with anatomical evidence of a major septohippocampal projection originating from this region. In contrast to results for lateral septum, recordings from medial and lateral mammillary nuclei indicate only small, diffuse excitation that exhibits no consistent changes over training, and is not related to activity seen in hippocampal or septal regions. The apparent lack of correspondence between learning dependent unit measures obtained from pre- and postcommissural fornix structures is entirely consistent with current modified descriptions of limbic system anatomy.

Analysis of temporal relations among units and slow waves in rabbit hippocampus

Arrays of triple microelectrodes were stereotaxically lowered into CA1, CA3 and dentate areas of the dorsal hippocampal formation in anesthetized rabbits. Recordings of action potentials and waves were analyzed on a PDP-11 computer using auto-, cross-, and multiple-correlation programs to determine temporal relations during 90 sec samples of spontaneous activity. It was found that temporal periods of neuronal firing and inhibition were strongly related to the pattern of waves. During periods of high amplitude synchronous waves (theta), the correlation between the activities of different groups of neurons was directly related to the periodicity of the wave. During instances of lower amplitude, desynchronous wave activity, the correlations between spikes recorded from those same cells were less periodic, varying according to the amount of wave synchrony. Variations in wave synchrony due to anatomical location, eserine effects, or spontaneous fluctuations under anesthesia produced corresponding variations in the relations between the activities of different groups of neurons. It is suggested that these relations between neuronal activity and gross waves may be implicated in processes which are at the basis of learning and memory.

Prediction of learning rate from the hippocampal electroencephalogram

Samples of spontaneous electroencephalographic (EEG) activity from the dorsal hippocampus of rabbits were recorded immediately before classical conditioning of the nictitating membrane response. Computer analysis revealed a significant predictive relationship between EEG frequency characteristics and the subsequent rate of learning.

Neuronal plasticity in the limbic system during classical conditioning of the rabbit nictitating membrane response. I. The hippocampus

Hippocampal unit responses were recorded throughout classical conditioning of the rabbit nictitating membrane response to a tone conditioned stimulus (CS) using a corneal air-puff unconditioned stimulus (UCS). Multiple unit analysis revealed that a rapidly developing increase in cell discharges (relative to spontaneous activity) occurs within the first block of paired trials and continues to increment with subsequent training, initially in the UCS period and then in the CS period. The pattern of hippocampal activity within paired trials closely parallels the amplitude-time course of the behavioral response and precedes it temporally. Identical recordsings from animals given unpaired CS-alone and UCS-alone presentations showed no such changes. These control results and additional lines of evidence point to the critical necessity of the learning paradigm for the development of the hippocampal response seen in conditioning animals. A single unit analysis indicates that not all hippocampal neurons exhibit the described conditioned discharge pattern. Hippocampal long-term potentiation is considered as a possible mechanism for mediating this early and rapid neuronal plasticity dependent on specific 'contingent' patterns of stimulation.

Identification of pyramidal cells as the critical elements in hippocampal neuronal plasticity during learning

The activity of single neurons recorded from rabbit hippocampus during classical conditioning of the nictitating membrane reflex was studied. All cells were first categorized according to their responses after fornix stimulation--i.i., antidromic activation, orthodromic activation, or no activation. The majority of cells that were antidromically activated--pyramidal cells--showed a highly positive correlation between the pattern of unit discharge and the topography of the nicititating membrane response within trial periods. Units that were orthodromically driven by fornix stimulation tended to inhibit during the presentation of trial stimuli, whereas most non-activated cells maintained low spontaneous levels of activity at all times. Thus, the major output neurons of the hippocampus appear to be the neuroanatomical substrate for the large and rapidly developing neuronal plasticity induced by this classical conditioning paradigm.

Descending fibres of the lateral funiculus of the amphibian spinal cord: their course and terminal distribution

Descending fibres of the lateral funiculus (LF) of the spinal cord have been mapped in the adult bullfrog, Rana catesbeiana. Following large and small lesions of the cervical LF, Nauta preterminal and terminal degeneration staining techniques revealed a large bilateral projection to intermediate regions of the lumbar central grey. Dorsal LF fibres were seen only contralaterally, and ventral fibres mainly ipsilaterally, Medical LF fibre and terminal degeneration occurred bilaterally. Terminal fields of ipsilateral fibres often included the large motoneuron somata situated dorsolaterally in the ventral grey. The smaller contralateral contingent of fibres terminated most often in areas of the central grey dorsal or medial to the motoneuron pool.

Limbic system interrelations: functional division among hippocampal-septal connections

Neuronal activity was recorded simultaneously from hippocampus and medical or lateral septum during classical conditioning of the rabbit nictitating membrane response. Although similarities exist between hippocampal and lateral septal patterns of activity, medial septal unit discharges indicate a different role during learning.

Habituation of a monosynaptic response in frog spinal cord: evidence for a presynaptic mechanism

1. Using the isolated spinal cord of bullfrogs (Rana catesbeiana), intracellular correlates of habituation-like depression of the monosynaptic response elicited in motoneurons by lateral column (LC) stimulation were investigated. The following properties of the motoneuron were compared before and after response depression produced by stimulation of the LC at 0.5/s: resting membrane potential, membrane conductance, critical firing level, and rheobasic current. No alteration was found in any of these parameters. 2. To determine whether transmitter release mechanisms were changing over trials, the LC was stimulated with pairs of stimuli separated by 6 ms presented at 0.5/s. While the amplitude of the first EPSP declined (74% of initial value), the amplitude of the second EPSP increased (111% of initial value). Facilitation ratios thus increased. 3. The following conclusions can thus be drawn: 1) habituation involves a process intrinsic to the LC-motoneuron synapse; 2) habituation is not totally mediated by receptor desensitization; 3) habituation is not mediated by a mechanism extrinsic to the LC-motoneuron synapse that depolarizes terminal endings, e.g., presynaptic inhibition or accumulation of extracellular potassium; 4) habituation is not produced by transmitter depletion. Any of these possibilities has as a necessary consequence that facilitiation ratios remain unchanged. 4. Possible mechanisms that could mediate habituation are: 1) alterations in mobilization and/or release of transmitter; 2) decreased probability of invasion of terminal branches of the presynaptic fiber by the action potential.

Tone-induced changes in excitability of abducens motoneurons and of the reflex path of nictitating membrane response in rabbit (Oryctolagus cuniculus)

Excitability of the reflex nicitating membrane (NM) response to air puff and of the abducens motoneurons (final common path) to direct electrical stimulation (measured as amplitude of the evoked NM response) in rabbit was tested at various times after onset and offset of a 350-msec tone. Excitability to air puff showed a substantial increase during tone on a gradual decrease following tone off, in agreement with Ison and Leonard. Excitability of motoneurons showed a similar marked increase during tone on a transient decrease immediately following tone off, and then a gradual decrease. It is suggested that these excitability changes, which parallel closely the interstimulus-interval conditionability function reported by Gormezano for the same preparation, may provide an independent measure of the "molar stimulus trace". Excitability of abducens motoneurons tested in the intertrial intervals during subsequent tone-air puff conditioning showed no consistent changes. Surprisingly, animals given tone-abducens nucleus shock testing developed conditioned responses. In subsequent conditioning, using standard tone-air puff training, they showed 85% savings in acquisition relative to nonstimulated controls.

Mechanisms of efferent neuronal control of the reflex nicitating membrane response in rabbit (Oryctolagus cuniculus)

Efferent mechanisms controlling the nicitating membrane (NM) reflex response to air puff in the albino rabbit were analyzed using stimulation, lesions, and recording techniques. In brief, stimulation of the sixth nerve (abducens) yields short-latency NM extension. Stimulation of the fourth and seventh nerves and the superior cervical ganglion has essentially no effect on the NM. Stimulation of the third nerve causes short-latency retraction of the NM. Lesions and recording data are consistent with this result - the sole efferent neuronal control of NM extension is the sixth cranial nerve and of NM retraction is the third cranial nerve. The NM extension response appears to be mediated by mechanical actions via retraction of the eyeball by the retractor bulbi muscle, and NM retraction appears to result from direct activation of muscle fibers in the NM by the third nerve. The superior cervical ganglion appears to play no role in reflex NM retraction in the rabbit, in contrast to its action in the cat.

Neuronal substrate of classical conditioning in the hippocampus

Neuronal activity in dorsal hippocampus was recorded in rabbits-during classical conditioning of nictitating membrane response, with tone as conditioned stimulus and corneal air puff as unconditioned stimulus. Unit activity in hippocampus rapidly forms a temporal neuronal "model" of the behavioral response early in training. This hippocampal response does not develop in control animals given unpaired stimuli.

Unit activity in posterior association cortex of cat

1. "Association" neurons in the posterior middle suprasylvian gyrus of cat were found to be predominantly polysensory, with 82% of the units in the chloralose-anesthetized preparation responding to auditory, visual, and somatic stimuli. There was no evidence of response differentiation associated with cortical depth distribution. Most units responded with a short-latency response (median 35-60 ms) to all stimulus modalities, with the response to visual stimulation occurring at the shortest latency. Among polysensory cells, almost half responded with equal probability to auditory, visual, and somatic stimulation. The visual stimulus was the most potent for those cells responding with a higher probability to a single modality. Varying degrees of response complexity were noted in some cells in terms of changes in responsivity over time, discharge to stimulus offset, and inhibition of spontaneous activity. The unitary discharge was seen to occur on the negative peak and slope of the locally recorded evoked potential. When only the larger amplitude spikes were analyzed, most of the unitary activity occurred on the negative peak of the evoked potential. Almost half of trimodally responsive cells displayed similar phase relationships between unitary activity and evoked potentials for all three modalities. The absolute refractory period for most cells was from 200 to 300 ms, with relative refractory periods extending up to 30 s...

Unit responses to moving visual stimuli in the motor cortex of the cat

Neurons in the pericruciate cortex of the cat were tested with moving visual stimuli for responses to specific properties of the visual receptive field. Specific response patterns were shown by cells of origin of the pyramidal tract as well as by other cells.

Development of polysensory responses in association cortex of kitten

Sensory responsiveness of single neurons in posterior association cortex of kittens that were 7 to 50 days old was investigated. The percentage of trimodal cells (that is, cells that respond to visual, auditory, and somesthetic stimulation) increased gradually until day 50, when percentages of trimodally responsive cells approached the adult level. In the youngest kittens, cells were predominantly responsive to only visual stimulation. With maturation, responsiveness to auditory and then to somesthetic stimulation was observed in increasing percentages of cells.