Citation
Effects of penicillin-induced epileptic foci in the hippocampus

Material Information

Title:
Effects of penicillin-induced epileptic foci in the hippocampus
Creator:
Hamilton, Gillian, 1946-
Publication Date:
Language:
English
Physical Description:
ix, 120 leaves : illus. ; 28 cm.

Subjects

Subjects / Keywords:
Catheterization ( jstor )
Dendrites ( jstor )
Dentate gyrus ( jstor )
Electrodes ( jstor )
Hippocampus ( jstor )
Lesions ( jstor )
Memory ( jstor )
Penicillin ( jstor )
Pyramidal cells ( jstor )
Rats ( jstor )
Dissertations, Academic -- Psychology -- UF ( lcsh )
Epilepsy ( lcsh )
Hippocampus (Brain) ( lcsh )
Penicillin ( lcsh )
Psychology thesis Ph. D ( lcsh )
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis - University of Florida.
Bibliography:
Bibliography: leaves 115-119.
General Note:
Manuscript copy.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
This item is presumed in the public domain according to the terms of the Retrospective Dissertation Scanning (RDS) policy, which may be viewed at http://ufdc.ufl.edu/AA00007596/00001. The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator(ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
022274744 ( ALEPH )
13586218 ( OCLC )

Downloads

This item has the following downloads:


Full Text










EFFECTS OF PENICILLIN-INDUCED EPILEPTIC FOCI

IN THE HIPPOCAMPUS















By
GILLIAN HAMILTON















A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY











UNIVERSITY OF FLORIDA


1970





























DEDICATION Dan Hamilton















ACKNOIMDE GEXENTS


I thank Dr. Robert L. Isaacson for everything.



I also thank my Cormmittee members: Dr. Paul Satz,

Dr. Robert King, Dr. Charles Weiss, and Dr. Fred King. And finally, I thank Dr. Carol Van Hartesveldt, Dr. and Mrs. Charles Vierck, Dr. Leonard Schmaltz, Dr. David Olton, Art Nonneman, Evan Suits, Dale Lewellyn, Mike Woodruff, Dr. Larry Means, Lewis Pinson, Pauletta Sanders the Terrific, and Mrs. Virginia Walker for material and/or moral support at various periods over the past three years.


iii














TABLE OF CONTENTS


Pa e
Acknowledgements ............. 111

List of Tables ...... ...................... v

List of Figures ....... ..................... vi

Abstract ......... ...................... viii

Chapter I: Introduction ...... ..............

Chapter II: Experimental Procedures ......... . 14 Chapter III: Histological Evaluation ............ 22

Chapter IV: EEG pesults .... .............. .. 36

Chapter V: Behavioral Results .. ........... . 71

Chapter VI: Discussion and Summary ............. 79

Appendix A ........ ..................... .. 97

Appendix B ........ ..................... .. 103

References ........ ..................... 115

Biographical Sketch ......................... 120















LIST OF TABLES

Table Page

1. Summary of avoidance response data ............... 72

2. Avoidance response data for Group 5 ............... 75

3. Locus of penicillin injections ................... 78

4. Layers within the hippocampus, beginning with
ventricular surface .............................. 81














LIST OF FIGURES


Figure Page

1. Loci of penicillin implants, S #19, 39, 40, 41 .. 25 2. Loci of penicillin implants, S #44, 2, 32, 34 ... 27 3. Loci of penicillin implants, S #35, 38, 59, 60 .. 29 4. Loci of penicillin implants, S #61, 63, 64, 62 .. 31 5. Loci of penicillin implants, S #65, 66, 67, 69 .. 33 6. Loci of penicillin implants, S #70, 74 .......... 35

7. Development of ipsilateral spiking after penicillin injection into the left hemisphere ....... 38
8. Development of ipsilateral spiking after penicillin injection into the left hemisphere
(continued) ....................... .... ....... . 39

9. Record of S #31 which showed no ipsilateral spiking .............o...................... 42

10. An example of a record with ipsilateral spiking but no contralateral spiking ................ 44

11. Examples of change in form of ipsilateral spike coinciding with spread to contralateral side .... 46 12. An example of the "break through phenomenon": ipsilateral seizures during or immediately preceding spread of spike to contralateral side .... 48 13. Examples of ipsilateral, contralateral, and bilateral seizures ................................ 51

14. Examples of independent contralateral spiking ... 53 15. Averages of 20 spikes: trigger on left (injected) side ...................... .... .. 56

16. Averages of 20 spikes: trigger on right (mirror) side .................. ... ... ..... . 58

17. Example of recover from spiking ................. 60
vi















LIST OF FIGURES
(continued)

Figure Page

18. Example of recovery from spiking,
(continued) ....... ............................. 61

19. Comparison of the records of a control S and a
penicillin-injected S .................... ......... 63

20. EEG records one hour post-injection for all
Group 5A Ss (ipsilateral penicillin; contralateral lesion) ................................... 66

21. EEG records one hour post-injection for all
Group 5A Ss, (continued) ........................ 67

22. EEG records 14 days post-injection for all
Group 5A Ss ........ ..... ..................... 69

23. EEG records 14 days post-injection for all
Group 5A Ss, (continued) ....... ................ 70

24. Diagram of the hippocampus in the rat, redrawn
from Douglas (1967) ............ ................. 83

25. Regional differences in termination of
afferents, adapted from Raisman (1965, p. 990). 88

26. The two neuronal systems plus the principal
links between them, redrawn and relabelled from
Raisman (1966, p. 105). . .......................... 92


vii














Abstract of Dissertation Presented to the Graduate Council of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

EFFECTS OF PENICILLIN-INDUCED EPILEPTIC FOCI IN THE HIPPOCAMPUS
By

Gillian Hamilton

August, 1970

Chairman: Rodbert L. Isaacson
Major Departmient: Psychology

The effects of penicillin-induced epileptogenic foci in the hippocampus on electrical activity and on acquisition of a two-way active avoidance task were studied in the rat. The following experimental groups; ,cre compared to unoperated and operated controls and unilateral and bilateral hippocamZally ablated controls: unilateral penicillin; unilateral penicillin with ipsilateral hippocamrpal ablation; unilateral penicillin with contralateral hippocampal ablation; unilateral penicillin with ipsilateral hippocampal ablation and subsequent contralateral hippocampal ablation; and unilateral penicillin with contralateral hippocamupal ablation and subsequent ipsilateral hippocampal ablation. In none of the above groups was there a significant avoidance deficit when the results xere analyzed without regard to locus of implant within the hippocampus. However, when the determination of locus of imrlant was made in those groups in which the locus


viii








was not obtained, it was found that an implant in fields CA3 and CA4, but not in CA1, CA2, or the dentate gyrus, produced severe avoidance deficits. These results indicate that there is a functional differentiation of the hippocampus, at least with regard to acquisition of this avoidance task. The penicillin-induced spiking recorded immediately after implantation did not appear to be correlated with avoidance performance, but there were indications that greater spike amplitude recorded several weeks later was positively correlated with performance deficits.















CHAPTER I: INTRODUCTION

Hippocampal lesions in humans sometimes result in

severe loss of recent memory, yet these effects are not found after similar lesions in animals. It has been suggested that epileptogenic foci (in combination with ablation) may be responsible for the memory deficit found in humans. In order to investigate this hypothesis, experimental epileptogenic foci were induced in rats, and their behavior was studied. It was expected that the abnormal electrical activity would produce animals with deficits more closely resembling the human results than animals in which the hippocampus was simply ablated.

Temporal lobe resection performed on humans with an

epileptogenic focus localized in one temporal lobe results in partial or complete destruction of that focus and a concomitant reduction in the frequency and severity of seizures. However, when the resection extends far enough posteriorly to include portions of the hippocampus and hippocampal gyrus, many patients also suffer a severe loss of memory, with no concurrent decrease of general intelligence. This memory impairment can be more accurately described as an anterograde amnesia; the patient is unable to retain the memory for any event occurring postoperatively. he can remember an events for only as long as his attention








remains fixed upon it; as soon as his attention is distracted the memory is lost. In addition, there is frequently a retrograde amnesia for events occurring in a specific preoperative time interval (several months to several years, depending on the patient). This deficit, which dissipates with time, does not affect remote memory prior to the time interval. Scoville, Penfield, and Milner (Scoville and Milner, 1957; Penfield and Milner, 1958) suggest that bilateral hippocampal destruction always results in recent memory loss. Unilateral hippocampal destruction might or might not result in memory impairment depending on the state of the remaining hippocampus. They suggest that if the remaining hippocampus is functioning improperly (due to lesions, neoplastic damage, or epileptic foci), the patient will have, in effect, bilateral hippocampal destruction and the recent memory impairment. If the remaining hippocampus is functioning normally, recent memory will be normal. Damage to the nondominant temporal lobe alone will produce deficits in pattern memory. Damage to the dominant temporal lobe alone will produce deficits in verbal memory. However, these deficits are relatively minor, and do not involve memory loss for events.

After similar experimental lesions in animals, behavioral testing reveals no evidence of memory loss. For an excellent review of this literature, see Douglas (1967). An increase of intertrial interval has no effect on acquisition, and hippocampally ablated animals seldom exhibit deficits in








retention. They show superior performance on many learning tasks (e. g., two-way active avoidance, runway or bar-press for food reward, go-nogo), no change on other tasks (e.g., discrimination tasks); and inferior performance on tasks which are generally categorized as requiring response inhibition (e. g., extinction, reversal, alternation, passive avoidance). The lesions also cause some changes in mating and maternal behavior, rage thresholds, and adrenal cortex activation, which are probably the result of hippocampohypothalamic influences, but these effects are not relevant here.

The results of these experiments with animals have generated several theories, such as implication of the hippocampus in general activation of inhibition of sensory systems (Herrick, 1933); regulating or modulating input (Klaver, 1965); decrease in attention to stimuli followed by nonreward via efferent control of sensory input (Douglas and Pribram, 1966); and production of the internal inhibition responsible for habituation (Kimble, 1968). These theories,however, contain no suggestion of hippocampal implication in memory function.

In contrast, some animal studies using chemical treatments rather than lesions do suggest that the hippocampus is involved with memory. Deutsch and his co-workers have done a series of studies in the past ten years involving injection of cholinergic drugs into the hippocampal area. They have found that anti-cholinesterase drugs such as physostigmine or









diisopropylfluorophosphate (DFP) have no effect if injected immediately after learning; cause severe temporary retention deficits if injected a week or two after learning; and cause recovery of a forgotten habit if injected one month after training (Deutsch and Hamburg, 1966; Deutsch and Leibowitz, 1966). Scopolamine, an anticholinergic drug, produced opposite effects (Deutsch and Rocklin, 1967). These authors hypothesized that learning causes an increase in the capacity of a synapse to eject transmitter substances. An accumulation of too much transmitter will therefore produce a temporary blockage at the synapse and a concomitant loss of memory.

Presumably puromycin also produces amnesia, because of inhibition of protein synthesis. Studies by Flexner and his co-workers have shown that retrograde amnesia for a two-day old habit is produced when puromycin is injected into the temporal areas of rats, mice, or fish in dosages sufficient to produce at least 80% inhibition of protein synthesis for 8-10 hours in the temporal area (Flexner, Flexner, and Stellar, 1965; Flexner, Flexner, and Roberts, 1967). However, if the habit is more than 5 days old, this hippocampal injection will not be sufficient to produce amnesia. It then becomes necessary to inhibit at least60% of protein synthesis in other brain areas for 8-10 hours by injection into the frontal poles and veniricles, in addition to the 80% inhibition of temporal protein synthesis. Flexner concluded that memory is stored for on- or two days in the








hippocampus, after which the engram spreads to other areas. He suggests a complex mechanism for memory storage, involving synthesis of messenger RNA in learning. The RNA, in turn, synthesizes a protein responsible for memory expression acting as an inducer of the mRNA gene (Flexner, et al., 1967).

ECS is also hypothesized to produce its amnesic effects by acting in the hippocampal area, since it appears to have less effect in hippocampally ablated Ss (Hostetter, 1968), and the hippocampus is more susceptible to seizures than other brain areas (Green, 1964). The ECS, however, is effective for only a few seconds or hours after training, and there is much dispute over whether it actually affects memory. Several investigators postulate that it may produce its amnesic effects through conditioned fear, disinhibition, or some other mechanism.

Nonetheless, it is clear from Deutsch's and Flexner's research that puromycin and cholinergic drugs injected into the hippocampal area exert an effect on the consolidation of long-term memory. Similarly, in human patients, Scoville and Milner hypothesize that bilateral hippocampal lesions cause inability to consolidate memory. Yet the animal lesion studies implicate the hippocampus in regulation of attention through inhibition of incoming sensory stimuli. Somehow these discrepancies must be resolved. It seems highly unlikely that a phylogenetically old structure such as the hippocampus should have an entirely different basic









function in man than in animals, although the function may be altered as the phylogenetic scale is ascended. Therefore, a more tenable hypothesis is that the recent memory loss results from a different primary disorder, such as attentional mechanisms.

Schmaltz (1968) presented an interesting hypothesis. He suggested that in humans, hippocampal destruction itself does not produce a memory loss. Rather, the presence of an epileptic abnormally functioning hippocampus is the major determinant. The most consistent difference between human and animal Ss is that the overwhelming majority of humans had preoperative psychomotor epilepsy before undergoing hippocampal removal. Schmaltz supported his hypothesis with the findings that a group of rats with bilateral hippocampal lesions learned an avoidance task at least as quickly as normal rats, while another group with an epileptogenic drug (penicillin) injected into one hippocampus was significantly impaired on the task. A third group, with one hippocampus removed and penicillin injected contralaterally, appeared slightly (although insignificantly) more impaired than the group given penicillin alone.

Nakajima (1969) found comparable results on a different avoidance task: hippocampally ablated Ss learned almost as quickly as normals, but rats with bilateral hippocampal injection of another epileptogenic drug, actinomycin D, were severely impaired.

Why should no hippocampus at all produce better perform-








ance on the avoidance task than a malfunctioning hippocampus? Schmaltz suggests that the abnormal discharge induced by the penicillin spreads through the many excitatory and inhibitory connections of the hippocampus with other portions of the brain, thus interfering with the functions of many different brain structures. It may be that removal of the hippocampi allows the release of inhibition of some auxiliary, phylogenetically older pathway which assumes the functions of the hippocampal pathway. This mechanism would presumably not be as sophisticated as the hippocampal mechanism, but would allow adequate and in some cases superior performance of the tasks. However, if the hippocampi are merely functioning abnormally, the auxiliary mechanism is not released from inhibition: instead, the hippocampus itself must perform its own functions -- but the abnormal electrical discharges cause severe deficits. Both hippocampi must be removed in order for the auxiliary mechanism to be released from inhibition. Hence if only one hippocampus is removed and the other is injected with penicillin, the one abnormally functioning hippocampus alone must perform all of the work of the hippocampal pathway by itself. Therefore, performance is even more severely retarded than if two malfunctioning hippocampi had been present.

Schmaltz did not run a group with bilaterally implanted penicillin, nor did he determine whether the unilateral penicillin-induced focus had spread to the contralateral side. However, Olton (1969) studied performance of rats with bi-









lateral penicillin injections on the same task, and found learning rates similar to Schmaltz' unilateral penicillin group.

The similarities between human patients and epileptic

rats support the hypothesis that abnormal electrical activity rather than hippocampal ablation might be involved in the memory loss found in humans. A review of the literature on human operations reveals that in most cases of memory loss after a unilateral hippocampal lesion, an abnormal focus was recorded in the contralateral side (Penfield and Milner, 1958; Terzian, 1958; Serafetinides and Falconer, 1962). When operations were performed on non-epileptics (generally during treatment of chronic schizophrenia) no EEG recordings were made to allow rejection of the possibility that some aspect of the surgery produced abnormally firing cells - a tenable view, in light of the many chemical and traumatic conditions which can precede formation of an epileptogenic focus.

Support of this hypothesis is offered by the ECS work, and also by recent unpublished work with puromycin, suggesting that this drug exerts its effects through the production of hippocampal seizure-activity rather than protein synthesis inhibition. It is possible that Deutsch's cholinesterase work will prove to have a similar basis.

A case presented by Stepien and Sierpinski (1960) provides confirmation for the hypothesis in the human data. A 15-year old girl had been exhibiting recurrent seizures since









she was 6 months old, and they were increasing in frequency and severity. An EEG revealed continuous multiple spikes, multiple spike-and-wave patterns, and slow wave (2-3/sec) complexes from the right frontal and temporal region. Sporadic sharp waves and wave-and-spike complexes occurred in the left temporal region, 0.5 seconds later than spikes in the right teniporal lobe.

The girl showed neither impairment of memory for the

distant past nor loss of attention, concentration, reasoning ability, or verbal recall. She was able to repeat nonsense syllables or series of numbers after as long as five minutes. However, when tested with task interruption for auditory and visual recent memory, she showed impairments. The task involved presentation of an auditory or visual signal, and after an interval of 0-120 seconds, the presentation of a second signal in the same modality. The S had to determine whether the second signal was identical to or different from the first signal. She was able to perform the task when the interval between the two signals was 60 seconds and when she was permitted to keep her attention on the signal. However, when a distracting stimulus of another modality was interposed between the two signals, she was unable to keep in mind the first signal, and so failed at the task.

The girl was subjected to right frontotemporal ablation. Marked abnormality was found in the region of the amygdaloid nucleus and hippocampus. This abnormal area was typical of epileptogenic tissue, being tough and yellow. Tissue was removed until all high-voltage activity had disappeared.








Postoperatively, electrical recording revealed no

epileptic focus on the right side. And, in agreement with the hypothesis, all recent memory disturbance disappeared: the S could discriminate the stimuli when they were separated by a 120-second interval, with a distracting stimulus interposed. In this case, the lesion produced an improvement in a previously impaired memory function.

In the human with an epileptogenic focus, memory deficits may or may not be present. Hippocampal ablation, however, produces severe memory deficits in these patients. A focus with a lesion is thus more debilitating than a focus alone. It is possible that additional tissue removal would alleviate rather than exacerbate the memory deterioration by removing residual epileptic tissue, and hence removing the cause of the memory loss.

To test this hypothesis, a group of experimental rats analagous to psychomotor epileptics must be established. An important experimental assumption will be that the focus resulting from penicillin implantation is closely analagous to an epileptic focus in a human S. In this study an experimental analog of human focal or partial seizure epilepsy with the focus located in the temporal lobe (temporal lobe or psychomotor epilepsy) is being investigated. In this type of epilepsy, the epileptic discharge originates in a restricted population of cortical or subcortical neurones from which it may or may not spread to other regions of the brain. Schmtidt and Wilder (1968) list five unique qualities









which characterize the abnormal cells making up the epileptogenic focus. First, the cells generate an autonomous paroxysmal discharge which can be influenced by synaptic activity. Second, they possess increased electrical excitability. Third, the cortical surface over the focus is electrically negative. Fourth, a sudden depolarization of the resting membrane potential will initiate volleys of very high-frequency impulses (700-1000/sec). Fifth, the cells have the ability to induce secondary epileptogenic foci in synaptically related areas.

Experimentally induced electrical foci appear to possess all of these five qualities. Particularly important is the autonomy of the foci (not true of other forms of epilepsy). Isolation of cortical from subcortical structures, a procedure which produces electrical silence in normal neurones, does not stop the paroxysmal activity of epileptic cells (Schmidt and Wilder, 1968). This autonomy of focal epilepsy supports the proposition that local application of penicillin can cause a group of cells to become abnormal in much the same manner as human foci.

In studying the performance of penicillin-induced

epileptic animals, the choice of task is important. Some damage is done to the hippocampus when the cannula used for penicillin injection is lowered. Also, an irritative lesion develops over the course of several days in the brain tissue surrounding the penicillin implant. Yet bilateral hippocampal destruction has been found to produce impairments on









many behavioral tasks (Douglas, 1967). Therefore, it is necessary to choose a task in which rats with hippocampal lesions are unimpaired, so that any deficits found after penicillin injection cannot be attributed to lesion effects of the drug or cannula. For this reason, the two-way active avoidance task was chosen. Several studies have shown that rats with hippocampal lesions are superior or equal to unoperated rats in acquisition and performance of this task (Isaacson, Douglas, and Moore, 1961; Olton and Isaacson, 1968; Schmaltz, 1968).

In order to study animals analogous to human temporal lobe epileptics with an abnormally discharging focus on one side, rats in this study were implanted with penicillin in one hippocampus. Through hourly and daily EEG recording, the spread of the focus to the contralateral side was carefully observed. Behavioral testing of these rats was expected to reveal performance deficits on the avoidance task.

An operation was then performed which was analogous to

the temporal lobe resection performed on humans in an attempt to reduce the severity and frequency of psychomotor seizures. In human patients, a severe recent memory loss was found after the operation. The rats were given hippocampal lesions on the same side as the focus, hence removing the focus. Behavioral testing of this group was expected to reveal a severe deficit in avoidance performance. EEG recordings revealed whether the unilateral focus had spread to the contralateral side, forming a mirror focus, and whether the degree








of spread which occurred either before or after the operation could be correlated with performance. This design allows a careful study of the behavioral effects of mirror foci.

Another group was hippocampally ablated on the side contralateral to the focus. These rats were expected to reveal behavioral deficits at least as severe as the unilateral penicillin group. EEG recordings were correlated with behavioral data.

In the last stage of the study, the remaining hippocampus was removed from some of the rats which had previously received unilateral hippocampal ablation. Any residual abnormally functioning hippocampus should thus have been removed, and performance was expected to return to initial levels. The S was expected to be "cured" of his severe performance deficits.














CHAPTER II: EXPERIMENTAL PROCEDURES


Subj ects

The Ss were 68 male Long-Evans hooded rats from Blue Spruce Farms, weighing 275-350 grams at the start of the experiment.

Surgery

Group 1 was unoperated. Ss in Groups 2 through 9 were anesthetized with pentobarbital and placed in the stereotaxic instrument for surgery. Surgical procedure for each group is outlined below. Recovery period after initial surgery for all groups was approximately 3 1/2 weeks, unless otherwise indicated.

Group 2: Untreated Operates

A midline incision was made in the scalp, and the

skull scraped free of fascia. Two small circular holes were drilled in the skull 3.8 mm posterior to bregma and 5.2 mm lateral to the midline. (These stereotaxic coordinates were derived from Pellegrino and Cushman, 1969.) Two holes were drilled anterior and posterior to each hole through which small screws were fastened to the skull. A 22 gauge blunt hypodermic needle was used as a cannula. This was placed in an electrode holder and lowered 5.3 mm below the skull into the left hippocampus. Next, a stainless steel twisted bipolar electrode insulated except at the tip and with a tip 14







separation of less than 1 mm was lowered into the homologous position on the contralateral side, and fastened to the skull with dental cement. Then, using the same procedure, a second electrode was implanted into the insilateral hippocampus using the same coordinates as were used for the cannula. The scalp was then closed with silk sutures and EEG recording was begun. In a few sample rats, recording was made on the cannulated side through the cannula itself, by using a larger 20 gauge needle and lowering an insulated wire with the tip exposed down into the cannula.

After the operation, Ss were given .25 cc bicillin injected intramuscularly into the hind leg. Group 3: Unilateral Hippocampal Penicillin

The animals in this group were subjected to the same

procedures as the animals in Group 2, except that the cannula was loaded with S-R penicillin in dry powder form (Parke, Davis, and Company, Detroit, Michigan). The penicillin was spread evenly on the bottom of a glass dish. The cannula was tapped lightly on the drug 25 times to force a plug of the substance approximately 1 mm long into the tip of the cannula. The end was then sealed with bone wax to prevent diffusion as the cannula was lowered into the brain. The plug was forced from the cannula with a stainless steel wire cut to extend just beyond the tip of the cannula. Electrodes were implanted bilaterally using the procedures outlined from Group 2.

Group 4: Unilateral Hippocampal Lesion









After incision and retraction of the scalp, a hole was drilled over the left hippocampus using a dental burr. The hole was enlarged with rongeurs. The dura was punctured with a needle, and then reflected with scissors. The exposed cortical tissue was then aspirated, using a 21 gauge aspirator tip. Finally the hippocampus was removed as completely as possible without damage to the underlying thalamus. Cotton balls soaked in saline were applied to arrest bleeding. After the bleeding had ceased, these were replaced with gelfoam plugs. The temporal muscles were pulled back in place, and the scalp was closed with silk sutures. Bicillin (.25 cc) was injected intramuscularly into the hind leg. Group 5A: Unilateral Hippocampal Penicillin; Contralateral
-cp6amoaI Lesion

Ss in this group were implanted with S-R penicillin in the left hippocampus, according to the procedures outlined for Group 3. At the same time, chronic bilateral electrodes were implanted in both hippocampi. Approximately two weeks after the drug implantation, both electrodes were pulled out, and the contralateral hippocampus was removed according to the procedures outlined for Group 4. Groups 5B, 5C, 5D, and 5E:

Ss in these additional subgroups were treated 3-4 months after Ss in the main groups of the study, and the data are analyzed separately. In these subgroups, different forms of penicillin were implanted in the left hippocampus according to the procedures outlined for Croup 3. All Ss were also given right hippocamnal lesions according to the procedures








outlined for Group 4. The subgroups differed in details of treatment.

Group 5B.--Ss in this group were implanted with S-R

penicillin. Six days later, the contralateral hippocampus was removed. The recovery period (after initial drug injection) was 13 days.

Group 5C.--Ss in this group were implanted with S-R

penicillin. The contralateral lesion was made either immediately before or immediately after the drug injection. The recovery period was 14 days.

Group 5D.--Ss in this group were implanted with a mixture of buffered potassium penicillin G and cobalt. For these rats, the cortex overlying the hippocampus was aspirated prior to lowering of the cannula. This method allowed E to observe the cannula entering the hippocampus, thus insuring correct locus of implant. Lesions were made imediately before or after the drug implant. The recovery period was 14 days for one S, and 5 days for eight Ss.

Group 5E.--Ss in this group were implanted with S-R

penicillin from a new shipment, with an expiration date of January, 1975. Again, the hippocampus was exposed prior to implantation. Lesions were made immediately before or after the drug implant. The recovery period was 5 days. Group 6: Unilateral Hippocamnpal Penicillin; Ipsilateral Hippoarpal Lesion

Ss in this group were implanted with penicillin into

the left hippocampus according to the procedure outlined for Group 3, and bilateral electrodes were implanted. Approxi-








mately two weeks later, both electrodes were removed. The left hippocampus, which had been injected with penicillin, was now ablated.

Group 7: Bilateral Hiopocamoal Lesion

Ss in this group were given bilateral hippocampal

lesions according to the procedure outlined for Group 4. Group 8: Unilateral Hippocampal Penicillin; Contralateral
Hippocapal Lesion; Ipsilateral )ippocampal Lesion

Ss in this group were first given left hippocampal

penicillin implants and bilateral electrode implants. Approximately 2 weeks later, the electrodes were removed, and the contralateral hippocampus was ablated. (Up to this point, treatment was identical with Group 5A). Then, 2 1/2 weeks after the drug injection, the implanted side was ablated. Group 9: Unilateral Hippocampal Penicillin; Ipsilateral
Hippocampal Lesion; Contralateral Hippocampal Lesion

Ss in this group were first given left hippocampal penicillin implants and bilateral electrode implants. Approximately 2 weeks later the electrodes were removed, and the same (left) hippocampus was ablated. (Up to this point, treatment was identical with Group 6.) Then, 2 1/2 weeks after the drug injection, the contralateral hippocampus was ablated.

EEG Recording

EEG recordings were obtained from all Ss with permanent electrodes. Recording from both hippocampi began immediately after the scalp was sutured (10-20 minutes post-injection) and continued for approximately two hours. Whenever it was








necessary to avoid confusion between electrocardiogram and EEG, a third channel recorded heart rate. In a few Ss from each group, the cannula itself was used as an electrode, and the recordings begun within 15 seconds of the injection.

In a few representative cases, the record was fed into a tape recorder so that it could later be analyzed on a Digital Electronics PDP-8 computer. In other cases, a Grass oscilloscope was used to measure spike latencies, and polaroid pictures were taken to record results.

EEG recordings were taken at intervals from each S

during the time that the electrodes were in place. Records were usually taken hourly during the first twelve hours postinjection, and daily thereafter. During these recording sessions, the S was awake, but generally remained motionless in the bottom of a high-walled box; care was taken to keep the room quiet. His movements were carefully observed and reported by observers, and movement artifacts noted on the record.

Approximately 2 weeks post-injection, Ss in all

groups were anesthetized with pentobarbital, and EEG recordings were made from those Ss with permanently implanted electrodes (Groups 2, 3, 5A, 6, 8, and 9).

Recording in unanesthetized Ss continued in those rats without lesions (Groups 2 and 3) until the end of behavioral testing, or about 40 days in all.








Behavioral Testing

The avoidance chamber consisted of two unpainted plywood compartments, each 11.5 in. long, 7.5 in. wide, and 8 in. high with a plexiglas top and a grid floor. The compartments were separated by metal guillotine doors which closed on a 1.75 in. high barrier. Midway between the ends of each compartment a 6 w. light bulb was located on the side and a 4 in. speaker (Quam #25A07) in the top. The grid floor was electrified by means of a Grason-Stadler shock generator set to deliver 0.6 mA shock. White noise was provided by a Grason-Stadler noise generator set at 12 dB. Time intervals were controlled by Grason-Stadler electronic timers and latencies were measured using a Grason-Stadler clock timer.

At the beginning of training, the animal was placed in the left hand compartment for 30 sec. The CS (onset of light and noise in both compartments) was presented for 10 sec. followed by the US (shock to the grid floor in the compartment occupied by the animal). The guillotine doors were then opened, allowing the animal to escape shock by moving into the other compartment. On subsequent trials, the guillotine doors were opened at CS onset, so that on all but the first trial of the first day, avoidance responses as well as escape responses could be made. Both CS and US were terminated and the guillotine doors lowered when the animal crossed the barrier into the previously unoccupied compartment. The intertrial interval was 30 sec.

On each trial E recorded whether the animal moved to








the other side before US onset (an avoidance response) or required the initiation of the US (an escape response). The avoidance or escape latency was also recorded. Twenty-five trials were given daily for a maximum of 15 days. If at any time a criterion of 23 avoidance responses in 25 consecutive trials was attained, testing was terminated.

Histology

After behavioral testing, the animals in Groups 2-9

were sacrificed and intracardially perfused with 0.9% saline followed by 10% formalin. The brains were removed and embedded in celloidin. Sections were cut at 10 microns and every tenth section was stained with thionin.















CHAPTER III: HISTOLOGICAL EVALUATION

All brains were carefully examined microscopically to determine the extent of aspiration lesion, locus of drug implant, and loci of electrodes. Reconstructions of these parameters were then made on representative diagrams of the brain (see Appendix A). Among groups with hippocampal ablations, Ss did not differ significantly in extent and locus of lesions. Brain damage was generally limited to the posterior neocortex, the corpus callosum, the fimbria, and the hippocampus.

It was sometimes difficult to differentiate between electrode tract and cannula tract, particularly since one electrode was lowered through the same skull hole as the cannula, using the same coordinates. However, the electrode tracts were generally more symmetrical, and the contralateral tract could of course be identified. Also, the cannula tract, but not the electrode tract, often contained bright orange-yellow non-nuclear material, distinguishable from the paler yellow blood cells. This deposit probably represented residual bone wax used to seal the cannula. In Group 2, electrodes were implanted without any penicillin, allowing verification of characteristics of cannula tracts.

Because the locus of drug injection is critical in this study, photographs of loci for all animals in groups with the





23


injected hippocampus intact at the time of histology (Groups 3, 5A, 5B, and 5C) are shown in Figures 1 - 6. In all other Ss, the injected side was subsequently removed, so that locus of implant into the hippocampus cannot be confirmed in these groups.
































Figure 1. Loci of penicillin implants.
Top left: S #19. Top right: S #39. Bottom
left: S #4U. Bottom right: S-#41.

















..


-V.




\N.


* K. -


'~*



* .4 . V
f* ~,
k
4~~f j


- 4/

.~ '1


r - -'- *-~


I I


* ,..* .-4-.- --4. -


a


NA;9 ft
-'-.4



A--


*~ I-.


* r


1 4


L































Figure 2. Loci of penicillin implants.
Top left: S #44. Top right: S #2.
Bottom left: S #32. Bottom right: S #34.


















~r ,


V


I -


i


*




I































Figure 3. Loci of penicillin implants.
Top left: S #35. Top right: S #38.
Bottom left: S #59. Bottom right: S #60.















.4 r , - -~


-.


/


w.


S.


j1


4,































Figure 4. Loci of penicillin implants.
Top left: S #61. Top right: S #63.
Bottom left: S #64. Bottom right: S #62.



















* *'~~ '~v.' ~ -~

I




, Il


I


.5-i


-,~



I


SKA' ~


-S.


j


. ; " .- I I .- _-,






























Figure 5. Loci of penicillin implants.
Top left: S #65. Top right: S #66.
Bottom left: S #67. Bottom right: S #69.
















* ( I ~ ~ ~ !j


K



P/v.. .


'C. . ~ - -a-- ..e-~


* 3"~.V~ a -~- ~nr rat r~ ~ ..~,a~'VW 'fl - -.


F -, ,r'~ ~rerra'.
V
I


-a


a

9/-, W~KY


\N'


-i


L.































Figure 6. Loci of penicillin implants.
Left: S #70. Right: S #74.




















-________/



'I ?. ~ ~

























N.








10

Cr)




-" - -~-.* --~-,.


















~
* S














CHAPTER IV: EEG RESULTS

Immediately after lowering of the cannula in Groups 2, 3, 5, 6, 8, and 9 the cannula was removed from the brain, an electrode was permanently implanted, the scalp was sutured, and connectors were fitted to the electrodes. Electrical recordings of brain activity in most Ss began about twenty minutes post-injection, the time required for the above procedures. At this time, spiking was invariably recorded in the left (implanted) hippocampus. In order to observe the development of this spiking, recordings were made in several rats using the cannula as an electrode. In these Ss, recording began a few seconds after injection of penicillin, and development of ipsilateral spiking could be traced. Generally, small monophasic spikes occurred at irregular intervals for the first few minutes, and then became more regular and grew in amplitude and complexity showing biphasic and multiphasic forms. An example of development of the ipsilateral spike is shown in Figures 7 and 8.

Only four Ss did not show regular ipsilateral spiking after the penicillin injections. In one of these Ss, histological examination revealed that the electrode had passed through the hippocampus and entered the ventrobasal thalamic complex. Two other Ss (No. 6, No. 8) were in Group 9, which had received bilateral hippocampal lesions




























Figures 7 and 8. Development of ipsilateral spiking after penicillin injection into the left hippocampus (S #28). Top trace: left (injected) hippocampus. Bottom trace: right hippocampus. First record: 30 sec. after injection of penicillin, bilateral seizurelike spiking. Second record: 2 min. post-injection, no spiking either side. Third record: 3 min. postinjection, first ipsilateral spikes. Fourth record:
5 min. post-injection, ipsilateral spikes increase in magnitude and become more regular. Fifth record: 10 min. post-injection, ipsilateral spikes continue to increase in magnitude and become more regular. Sixth record: 12 min. post-injection, first contralateral spikes appear. The pulses on time trace occur every second. The calibration mark at the beginning of the time trace represents 0.5 my. Gains are constant for all traces.


























[2 - -.,
, .77Z1


Y v V ''" '





































.'/ I








. - - -
I-








prior to histological analysis, so electrode loci could not be determined. It seems probable that the electrodes were not positioned correctly in these Ss. The record of the fourth S (No. 31) was particularly interesting, as it showed contralateral spikes, but no ipsilateral spikes. However, histology revealed that the electrode was correctly positioned (Figure 9).

In two Ss, contralateral spikes never appeared (Figure 10). In all other Ss, between 20 and 45 min. after the penicillin injection, the first spiking appeared in the contralateral hippocampus. In every case, the spread of spiking to the contralateral side was accompanied by some change in the form of the ipsilateral spike. In some cases, the form change was from monophasic to biphasic; in other cases the change was from single to double or from one multiple to another multiple. Many variations in the changes were found. In all cases, however, whenever the ipsilateral spike showed the new form, a contralateral spike appeared; whenever the ipsilateral spike showed its old form, no spike appeared on the contralateral side. Examples of this phenomenon are shown in Figure 11.

In some cases, but by no means always, the appearance of the first spike on the contralateral side occurred during or immediately after a short burst of ipsilateral seizures. An example of this "break-through phenomenon" is shown in Figure 12.

In addition to ipsilateral and contralateral spikes,































Figure 9. Record of S #31 which showed no ipsilateral spikes. Top trace: left (injected) hippocampus. Bottom trace: right hippocampus. First record: 20 min. post-injection. Second record: 30 min. postinjection. Third record: 40 min. post- injection. Pulses on tine trace occur every second. The calibration mark at the beginning of the time trace represents 0.5 mV. Gains are constant for all traces.





























I


~W
U V~ ~ 'v~


[117 -


-------- ------ -































Figure 10. An example of a record with ipsilateral spiking but no contralateral spiking (S #4). Top trace: left (injected) hippocampus. Bottom trace: right hippocampus. First record: 30 min. postinjection. Second record: 60 min. post-injection. Third record: 90 min. post-injection. Pulses on time trace occur every second. Calibration mark at beginning of time trace represents 0.5 mV. Gains are constant for all traces.















(


[uZffi III 7< 77.Xr:mzzz
I
, / / / 1' / I ' I
I ! J ~



V ----I..--,


- ~ui]


I, i I




** ~ ~
































Figure 11. Examples of change in form of ipsilateral spike coinciding with spread to contralateral side. First trace: left (injected) hippocampus. Second trace: right hippocampus. Third trace on bottom record: heart rate. First record: S #1. Second record: S #28. Third record: S #14. Pulses on time trace occur every second. Calibration mark at beginning of time trace represents 0.5 M.


































r7~ - 71 ~77 ____________~ZZ�2iL~____ ____4vK~

~









[7~ -


iU~j,~'~' ~ ~ ~' ~ ii urn rirrrr~ urn ~fu ~ up U
































Figure 12. An example of the "breakthrough phenomenon": ipsilateral seizures during or immediately preceding spread of spikes to the contralateral side (S #13). First trace: left (injected) hippocampus. Second trace: right hippocampus. Third trace: heart rate. First record: 15 min. post-injection. Second record: 60 sec. later. Third record: 60 sec. later. Pulses on time trace occur every second. Calibration mark at beginning of time trace represents 0.5 mV. Gains are constant for all traces.





























17 7








*'Al











~~~~~ A --- ~ 4, , ~ V..........F .








records from about one quarter of the animals also showed seizures (very short latency repetitive sDiking) from one or both hippocampi. Sometimes the seizures appeared only on the ipsilateral side; sometimes they appeared only on the contralateral side; and sometimes they appeared simultaneously in both records (Figure 13). No movements could be detected during the electrical seizures.

The spikes on the contralateral side did not always occur only in response to an ipsilateral spike. In some records, one hour or more after the injection, occasional contralateral spikes appeared which were apparently independent of any recorded ipsilateral activity (Figure 14). Independent contralateral seizures have already been mentioned. In order to confirm these findings, the data from several Ss were recorded on tape and fed into the Lab-8 averaging system (Digital Equipment Co.). The Advanced Averager Program was used to obtain the average waveform of 20 consecutive spikes for the post-implantation experimental period. The Advanced Averager offers advantages over other averaging programs in that it allows triggering of the average on actual spikes, and the waveform can be averaged for time before and after the spike.

From the tape record, a series of average waveforms

were computed using spikes on the ipsilateral side to trigger the averager. This procedure showed the initial absence of a mirror focus and its later development. The same record was then averaged again using spikes on the contralater-































Figure 13. Examples of ipsilateral, contralateral, and bilateral seizures. Top trace: left (injected) hippocampus. Bottom trace: right hippocampus. First record: ipsilateral seizures (S #27). Second record: contralateral seizures (S #7). Pulses on time trace occur every second. Calibration mark at beginning of time trace represents 0.5mV.












A - I



AXuififi7




E7.7.7-1V































Figure 14. Examples of independent contralateral spiking. Top trace: left (injected) hippocampus. Bottom trace: right hippocampus. First record: S #28, one hour post-injection. Second record: continuation of first record. Third record: S #17, 4 hours post-injection. Pulses on time trace occur every second. Calibration mark at beginning of time trace represents 0.5 mV.
























LIZILI� 77uXL�7~7Ii>7L


I I





~w~4 ~ I F7~


A
V
II ~IW I ill
v









al side to trigger the averager. This allowed observation of activity on the implanted side whenever mirror spikes occurred. An average was taken of 50 msec. before and 100 msec. after each 20 ipsilateral spikes (Figure 15), and then each 20 contralateral spikes (Figure 16). The averaged spike magnitude was found to be higher on the triggering side, indicating that although some correlation existed between spikes on both sides, the cross-correlation was less than unity: some spikes occurred on each side which were not matched on the opposite side.

Recovery followed a similar course in all Ss (see Figures 17 and 18). Approximately two hours after the implantation, spiking began to decrease in frequency and magnitude on both sides. By about 3 hours postoperative, ipsilateral spiking was frequent, but only a few independent contralateral spikes remained. By about 5 hours postoperative, no contralateral spikes occurred at all, and only about one ipsilateral spike appeared per minute. By about 7 hours postoperative, no spikes remained on either side, and the records were essentially normal, although ipsilateral or contralateral depression remained in some Ss for as long as two or three days before the records appeared completely normal. After 2 or 3 days, however, the records of Ss which had had cannulation without penicillin were indistinguishable from most of the records of Ss which had received the penicillin injection (Figure 19).

None of the electrical phenomena that occurred ii.mmedi-
































Figure 15. Averages of 20 spikes: trigger on left (injected) side. Top row, left to right: then bottom row, left to right: Beginning 150 sec. after injection, consecutive averages of 20 spikes show development of the mirror focus.
Traces 1 and 2 show left and right hippocampal averages for 50 msec. before and 50 msec. after the spike. Traces 3 and 4 show left and right averages for 100 msec. after the spike.
























L


3ii
































Figure 16. Averages of 20 spikes: trigger on right (mirror) side. Top row, left to right; then bottom row, left to right: Beginning 10 min. after injection, consecutive averages of 20 spikes. Traces 1 and 2 show left and right hippocampal averages for 50 msec. before and 50 msec. after the spike. Traces 3 and 4 show left and right averages for 100 msec. after the spike.


































Figures 17 and 18. Example of recovery from spiking (S #13). Top trace: left (injected) hippocampus. Bottom trace: right hippocampus. First record: 1 1/2 hours post-injection. Second record: 3 hours postinjection. Third record: 4 hours post-injection. Fourth record: 6 1/2 hours post-injection, ipsilateral spikes approximately once per record. Fifth record: 7 hours post-injection, no spiking, some depression. Sixth record: 36 hours post-injection, complete recovery. Pulses on time trace occur every second. Calibration- mark at beginning of time trace represents
0.5 mV. Gains are constant for all traces.
















V
I I
LIII XX IIII Ill ~ ,~~XhJ







~77i~X1i















































I~~ AN~


9-: II


-L -*.----*---**~~~~I *- -~































Figure 19. Comparison of the records of a control S and a penicillin-injected S. Top trace: left (injected) hippocampus. Bottom trace: right hippocampus. First record: control S, Day 0 (S #49). No spiking. Second record: unilateral penicillin S, Day 0 (S #41). Bilateral spiking. Third record: control S, Day 37. Fourth record: unilateral penicillin S, Day 42. Pulses on time trace occur every second. Calibration mark at beginning of time trace represents 0.5 mV.


































































IP it


IAll~
*~~ ti *ii I



* ~ ~ ~ ~ ~ ~ i A*** * *;, I A ' '*'* ~ ~ *





,,le A-









ately after the penicillin injection appeared to have any correlation with performance of the animal on the behavioral task. Among the Ss of Group 5, which had penicillin implanted in the left hippocampus and the right hippocampus ablated, two of the six Ss never learned the avoidance task; the remaining four Ss learned the task in 3-5 days. Yet all Ss showed spiking from one or both hippocampi; and the S which showed seizures learned faster than any other Ss. Only one S in the group did not show regular spiking, but instead showed infrequent ipsilateral and contralateral spikes; this S was one of the two which never learned the task (Figures 20 and 21).

However, when records taken two weeks after the injection are examined, the two Ss which showed the greatest deficits also appeared to have the greatest spike height relative to background activity (Figures 22 and 23). Since lesions removed the penicillin focus, behavioral correlations with other groups cannot be made.































Figures 20 and 21. EEG records one hour post-injection for all Group 5A Ss (ipsilateral penicillin; contralateral lesion). Top trace: left (injected) hippocampus. Bottom trace: right hippocampus. First record: S #2, never reached criterion. Second record: S #13,
never reached criterion. Third record: S #14, 4 days to criterion. Fourth record: S #32, 5 days to criterion. Fifth record: S #35, 4 days to criterion. Sixth record: S #37, 3 days to criterion. No apparent correlation between electrical activity and behavior. Pulses on time trace occur every second. Calibration mark at beginning of time trace represents 0.5 mV.





























j I IVK-

















A.
ft







.1~. -., ~


U7~7Z7YI7CZ. 77L111.ZYZILYZZIZZZI


I I

k


46 1































Figures 22 and 23. EEG records 14 days post-injection for all Group 5A Ss (ipsilateral penicillin; contralateral lesion). Nembutal anesthesia. Top trace: left (injected) hippocampus. Bottom trace: right hippocampus. First record: S #2, never reached criterion. Second record: S #13, never reached criterion. Third record: S #14, 4 days to criterion. Fourth record: S #32, 5 days to criterion. Fifth record: S #35, 4 days to criterion. Sixth record: S #37, 3 days to criterion.












p








Vt'

L11711Z1 i1�1L


I (~i ( I I








* ~X)AJ)~J..~ *J< 'AA'..~k~> *XA. AtA~., A CLXA& AA.A~AA.?.~&%AtAAA~A AAA )4A~~.AA. &~ ttAJS....
K ~.*. __





































[0 1















CHAPTER V: BEHAVIORAL RESULTS

Table 1 presents the mean number of trials which the

nine main groups of subjects took to reach the behavioral criterion of 23 out of 25 consecutive avoidance responses. Appendix B gives the number of avoidance responses made by the individual Ss for each block of 25 trials.

Table 1 also shows the standard deviation for each group, and variances were also computed. Because the within-group variances differed so greatly among groups, parametric statistical tests were consideredinappropriate to analyze the data.

The greatest variance was found in Groups 1, 5, and 6.

Groups 5 and 6, which had penicillin implanted unilaterally and a subsequent ipsilateral or contralateral lesion, were the two groups in which it was expected that behavioral deficits might be found. However, although the mean scores were highest for these two groups, and particularly for Group 5, indicating that they took more trials to reach criterion, the overlap in scores was so great that the groups do not differ significantly from control groups using the nonparametric Mann-Whitney U Test. The source of this variance will be discussed later.

Despite the lack of significant differences between groups, certain trends are evident in Table 1. Group 2








Table 1

Summary of Avoidance Response Data
Standard Deviations Given in Parentheses


Group 1 2 3 4 5A 6 7 8 9

Mean Trials 101 87 99 85 198 123 47 75 100
to
Criterion (115.9) (43.24) (32.75) (21.21) (162.3) (113.8) (13.43) (40.98) (47.85)









scores were very similar to Group 1 scores, indicating that the operative procedure and the permanent bilateral electrode implants alone had no effect on performance of the task.

The scores of animals in Group 4, which received unilateral hippocampal lesions alone, were very similar to control group scores. Group 7, which received bilateral lesions alone, learned the task more quickly than control groups. Only two subjects were run in each of these groups because the results have been confirmed in several other studies (e.g., Olton and Isaacson, 1968; Schmaltz, 1968; Olton, 1969).

Animals in Group 3, which received unilateral penicillin injections, were also unimpaired on the task. This result is in contrast to the findings of Schmaltz (1968); however, it does agree with the human data on temporal lobe epileptics prior to surgical intervention.

Animals in Group 5, which received unilateral hippocampal penicillin and a subsequent contralateral lesion, were deficient on the task, confirming the findings of Schmaltz (1968). In contrast, Ss in Group 8, which also received unilateral hippocampal penicillin and a subsequent contralateral lesion, but also received a lesion of the injected hippocampus, showed no deficits on the task.

Animals in Group 6, which received unilateral hippocampal penicillin and a subsequent lesion of this injected side to isolate a possible mirror focus, performed more









poorly than the control groups, although the difference did not reach significance. It seems likely that, at least for some Ss, the mirror focus had a behavioral effect.

Animals in Group 9, which also received unilateral

hippocampal penicillin and a subsequent lesion of this injected side but also received a lesion of the mirror focus side, showed no deficits on the task.

The large variance found in Group 5 was unexpected: in previous studies, this treatment resulted in more consistently poor performance (Schmaltz, 1968; Isaacson, Schmaltz, and Hamilton, in preparation). Four additional groups were run in an effort to determine some variable which would increase performance deficits over those found in this study. The treatment changes investigated were: decreased recovery period; decreased time elapsing between implantation and contralateral lesion; and changes in the type of drug used. Table 2 presents the mean number of trials to criterion for each subgroup of Group 5. An analysis of variance of the means yielded a significant difference between subgroups (F = 4.31, df 4/23, p,<.01).

Clearly, reducing the amount of time elapsing between drug implantation and behavioral testing did not increase performance deficits, whereas in Group 5A, 26-28 days elapsed; in Group 5B, 13 days elapsed; in Group 5C, 14 days elapsed; and in Groups 5D and 5E only 5 days elapsed.

Variation in length of time elapsing between injection and contralateral lesion also appears to have no significant effect on trials to criterion. In Groups 5C, 5D, and 5E















Table 2

Avoidance Response Data for Group 5 Standard Deviations Given in Parentheses


Group


Mean Trials 198 91 120 91 375
to
Criterion (162.3) (39.19) (125.5) (41.63) (0.000)








the lesion was made either immediately before or immediately after the drug injection; in Group 5B the lesion was made 7 days later; and in Group 5A the lesion was made 14 days later.

Penicillin G-K was used to determine whether the form of penicillin used might affect results, possibly because of active ingredients other than the penicillin itself. However, penicillin G-K appeared to have effects similar to those of S-R penicillin. In Groups 5A, 5B, 5C, and 5E S-R penicillin was injected. In Group 5D, penicillin G-K, mixed with cobalt powder, was used.

The same drug, S-R penicillin, was used in the studies of Schmaltz (1968); Isaacson, Schmaltz, and Hamilton (in preparation), and this study. However, in the present study the results were less consistent for Group 5A, and the drug appeared to have no effect at all in Groups 5B and 5C. Although spiking for the two-hour recording period immediately after injection was of the same frequency and amplitude for all groups it is quite possible that the effects of the drug used in this study were more quickly dissipated when used for Group 5A; and that when the drug was injected 2-3 months later in Groups 5B and 5C, its potency had entirely dissipated. This hypothesis is supported by the results from Group 5E, which was treated exactly like Group 5D, except that a fresh shipment of S-R penicillin was used. The Ss of this group showed large and consistent deficits.

If this is the case, then an interesting interpretation of the data can be made. The decreased potency of the drug









quite probably causes it to activate fewer cells within the area of diffusion, so that locus of drug implant becomes critical. The locus of penicillin injection for all Ss in which the site of penicillin injection was intact at the time of histology was therefore carefully analyzed by three different observers. In all cases, close agreement was found between observers: in a few cases, two observers designated the immediately neighboring field. Photographs of each of these loci have been presented in Figures 1 - 6; Appendix A gives reconstruction on coronal sections. Table

3 presents the histological analysis.

Injection of penicillin into CA3 or CA4 produced an inability to learn the avoidance task within 375 trials. Injection into CAl or CA2 did not produce this deficit, nor did dentate injections.

Two components are necessary for successful performance of the avoidance task. The first component is learning the escape response: once the shock has begun, it can be terminated by crossing to the other side. For days 1 and 2, mean escape latencies were computed for escape trials for Ss listed in Table 3. No differences were found between groups: the impaired Ss were able to learn and retain the escape response as well as other animals. The second component is, of course, learning to make the avoidance response (crossing to the other side as a response to the CS). It is this response that the Ss were unable to learn.















Table 3

Locus of Penicillin Injection


Locus of Injection


Trials to Criterion


CAl CA1
CAl-2 CA1 CAl CAl


CA2 CA2


CA3-4 CA3-4 CA3


dentate dentate dentate dentate dentate dentate dentate dentate dentate


129 70 76
152
63 77 Mean =95

95 85
Mean =90

375 375 375 idean=375

137 66 60 50
101 66
87 55 64
Mean= 76


Subject


Group















CHAPTER VI: DISCUSSION AND SUI -,ARY

The basic purpose of this study was to investigate the hypothesis that hippocampal mirror foci as well as original foci have behaviorally debilitating effects and that these effects can be alleviated by removing the epilep-togenic focus. When results were analyzed without regard to locus of implant within the hippocampus, no significant deficits were found in any of the groups receiving penicillin. However, when the determination of locus of implant was made in those groups in which the focus was not ablatedi, it was found that an implant 4 , fields CA3 and CA4, but not in CAl, CA2, or the dentate gyrus, produced severe avoidance deficits. These

results indicate a functional diffeimEntation of the hipp:ocampus. The finding that in all rats, regardless of locus, mean trials to criterion for Group 5 (unilateral penicillin; contralateral lesion) and Group 6 (unilateral penicillin; ipsilateral lesion) were larger than for the control groups, whereas Group 8 (unilateral penicillin; contialateral lesion; ipsilateral lesion) and Group 9 (unilateral penicillin: ipsilateral lesion; contralateral lesion) learned at the same rate as controls, suggests that the hypothesis merits further investigation, with careful attention to locus of implant. In view of the important implications of the hypothesis, such research soems warranted.








In order to assess the significance of the functional differentiation of the hippocampus found in this study, it is necessary to review the nature of the cytoarchitectonic fields of the hippocampus and to determine to what extent and in what manner they may represent functionally distinct entities.

An important cell type of the hippocampus is the pyramidal cell. Axons of the pyramidal cells form the hippocampal efferent and commissural systems and their collaterals and various association pathways. Lorente de No identified four different cytoarchitectonic fields of the hippocampus on the basis of the morphology and arrangement of these pyramidal cells. These layers are called CA1, CA2, CA3, and CA4.

CAl is composed of small pyramidal cells arranged in

two layers. The apical dendrites of these pyramids typically have many fine branches given off at right angles into the stratum radiatum. A review of the layers of the hippocampus is given in Table 4. The pyramidal cells of CA2 are considerably larger than those of CAl and are not arranged in two layers, nor do they have branches in the stratum radiatum. The pyramids of CA3 are the largest in the hippocampus. Their apical dendrites usually bifurcate before crossing the stratum radiatum, but they have no side branches. The proximal part of the apical dendrite shaft has thick thorns, which are in contact with mossy fibre endings. The axons give off Schaffer collaterals which leave the stratum





. 81



TABLE 4 Layers within the hippocampus, beginning with ventricular surface alveus - axons, afferent and efferent; rostrally continuous with fornix, laterally with fimbria


stratum oriens stratum pyramidale


stratum lucidum stratum radiatum


stratum lacunosum


- pyramidal axons heading for alveus
- basal dendrites of pyramidal cells
- afferents to hippocampus from fimbria
and entorhinal area
- commissural fibre inputs

- pyramidal cell bodies
- basked cell terminations

- mossy fibres (axons of granule cells
of dentate gyrus) end on thorny projections on initial part of apical
dendrites in the stratum lucidum
of CA3 and CA4

- unbranched large apical dendrites,
closely packed in radial rows

- unbranched apical dendrites
- large Schaffer collaterals from axons
of CA3 and CA4 pyramids form a
dense cuff of synapses on the pyramidal apical dendrites in CAl and
CA2


stratum moleculare - fine terminal branches of apical dendrites
- input from perforant path from entorhinal cortex

Dentate gyrus:


stratum moleculare stratum granulosura stratum polyo-ohe -


peripheral dendrites of granule cells;
continuous with stratum moleculare
of hippocampus

tightly packed granule cells axons from mossy fibres to end on thick
thorns of proximal apical dendrite
trunks of pyramids in s. lucidum

in hilum of dentate gyrus; not clearly
separable from CA 4 of hippocampus









lacunosum of CA1. CA4 cells, which lie in the hilum of the dentate gyrus, do not have a typical pyramidal shape. They do, however, send axons to the alveus, and have both thorns in contact with mossy fibres and Schaffer axon collaterals.

The spatial arrangement of the cytoarchitectonic fields of the hippocampus is sho :n in Figure 24. Figure 24-A shows the position of the hippocampus within the whole brain: the fornix extends anterior and ventral. Figure 24-B shows a section cut through the middle portion of the hippocampus. Two interlocking U-shaped structures can be seen: the hippocampus proper, and the dentate gyrus. A close-up of this section is drawn in Figure 24-C. If another section showing these structures cut on the same plane were made in the dorsal hippocampus, it would then be a coronal rather than a horizontal section, and CA2 would be located ventrally rather than medially.

With this background, the evidence concerning the

functional significance of the cytoarchitectonic fields can be reviewed. The work of Blackstad (1956, 1958, 1970), and Raisman (1965, 1966), based on earlier anatomical analyses of Cajal, Lorente de ITo (1933, 1934), Nauta (1956), and numerous other investigators, suggests that both afferent and efferent projections of the hippocampus in the rat show considerable localization with respect to field of origin and termination. The analysis below draws on the work of all these investigators.

Two major direct sources of afferent fibres to the














A. Side view of rat brain , fe


SHIPPOCAMPU

REC-009DENTATE NTERI OLES




ZLVEUS 4PV~n~





B. Schematic diagram of side view of the hippocampus in
the rat, with section through the middle portion.



















C. Close-up H IPPOCAMPAL
of section S C O
in B RVU
Llll l: VENTP ICLE
Fig. 24. Diagram of the hippocampus in the rat, redrawn and
relabelled from Douglas (1967) ,pp. 417-418.









hippocampus exist. The first source is the entorhinal area. The lateral entorhinal area sends fibres into the hippocampus via the perforant tempooammonic tract to the stratum lacunosum-moleculare of the hippocampus, contacting distal apical dendrites of pyramidal cells in CAl, and to a lesser extent in CA2. A further group of fibres reaches the stratum moleculare of granule cells. The medial entorhinal area sends fibres through the alvear tract to the basal dendrites of stratum oriens in CAl.

The second source of afferent fibres to the hippocampus is the septum. Fibres originating in the medial septal nucleus pass through the fimbria to terminate in the basal dendrites of stratum oriens and the apical dendrites of stratum radiatum of fields CA3 and CA4 of the hippocampus (not CAl), and to a lesser extent in the dentate gyrus. Relatively few axodendritic synapses are formed on the main dendritic trunk; most are on the small fine side branches.

Other proposed afferents influence the hippocampus only indirectly: the cingulum projects to the presubiculum and the entorhinal area; the pyriform cortex projects to the lateral entorhinal area; and the dorsal fornix projects primarily to the presubiculum, although it may relay a few afferents from the hypothalamus and rostral midbrain to CAl. McLardy (1963) presents evidence for one additional projection: he observed in both guinea pig and monkey brains a small fascicle cf myelinated fibres projecting between the amygdaloid nucleus and the ventral alveus of the hippocampus.









The comisural system of the hippocampus has been

studied by Blackstad (19560. At the rostral pole of the hippocampus, fibres of the fimbria separate into two distinct groups: the extrinsic hippocampal efferents continuing forward to form the anterior pillars of the fornix; and the commissural fibres. The latter cross the midline as the ventral hippocampal commissure, and turn back in the fimbria of the opposite side to terminate in the stratum oriens and radiatum in all fields of the hippocampus. However, the percentage of these fibres terminating in CAl and CA2 is much greater than in CA3 and CA4. Some fibres also project through the ventral coinissure to the stratum moleculare of the dentate gyrus. Also, a small proportion of fibres pass through the dorsal hippocampal comnmissure, projecting to the stratum moleculare of anterior CAl and to the subiculum; the precise origin of these fibres is unknown.

It is probable (and verified for CA1, Blackstad, 1956) that each cytoarchitectonic field is reciprocally connected through the ventral hippocampal commissure with its contralateral counterpart. Sharp transitions in fibre distribution occur exactly at the architectonic limits. Whenever heterotopic rather than homotopic projections are found, the fibres pass only to precise loci on the contralateral side. Thus the idea that commissural fibres from one point spread diffusely over the contralateral hippocampus, with at most a greater percentage of fibres projecting to the homologous









point, appears to be as inaccurate in the hippocampus as it is in the cortex. The function of such precise organization in unknown. However, the precision of the organization suggests that a specific function is involved.

Association fibres in the hippocampus also have precise loci. The largest group of association fibres consists of very short axonal relays within the complex fibre plexus of the various hippocampal strata. A second group of collaterals consists of Schaffer's axon collaterals of fields CA3 and CA4, which turn back through the stratum ratiatum to the stratum lacunosum-moleculare of CA3, where they pass to the same layer of CA2 and CA1. CA2 and CA1, however, do not have reciprocal interhippocampal connections with CA3 and CA4; their axon collaterals pass to the stratum lacunosum-moleculare of their own field, and also to the subiculum.

The last group of collaterals consists of those involving the dentate gyrus. Posterior CA1, but not anterior CA1, CA3, or CA4, projects to the stratum moleculare of the dentate gyrus. Reciprocally, axons of granule cells of the dentate gyrus gather into supra-and infra-pyramidal bundles and pass to the stratum lucidum of CA3 and CA4, where they make contact with the thorns at the bases of the apical dendrites of the pyramids. Blackstad (1970) found that each level of the dentate gyrus along a septo-temporal axis projects to an equally restricted transverse level of CA3 and CA4 of the hippocampus. Blackstad suggested that this precise level-tolevel localization implied that other associated systems









(such as the perforant path, Schaffer collaterals, and basket cell system) might be organized with comparable topographic precision. Concerning the functional significance of this level-to-level localization, "all that can be said at present is that the preciseness of the observed localization clearly suggests some precise function in need of a high degree of structural order " (Blackstad, 1970, p. 448).

In summary, the regional differences in termination of afferents are diagrammed in Figure 25, adapted from Raisman (1965, p. 990). The functional significance of this patterning of cellular connections can at present only be speculated upon.

The efferent projections from the hippocampus are gathered together in the alveus and distributed through two pathways, the fimbria and the dorsal fernix. These efferents are well described by Nauta (19%) (The dentate gyrus appears to have no extrahippocampal projections.)

The anterior part of CAl sends its efferents through the dorsal and nostcomrrissural fornix to terminate in the anterior thalamic nuclei and the medial and lateral mammi.llary nuclei, with the exception of the ventral lamina of the posterior part of the medial mammillary nucleus. The posterior part of CAl has fibres distributed both through the dorsal fornix and fimbria. These have a precommissural termination in the septofimbrial, medial septal, and diagonal band nuclei, the ventromedial quadrant of the lateral septal nucleus and the nucleus accumbens of the








lateral entorhinal area via perforant TAT


commissural fibers


medial entorhinal __area commu fibers


CO


CA 1 CA 3


Fig.25. Regional differences in termination of afferents,
adapted from Raisman, 1965, p. 990.









same side. They have a postcoimitssural termination in the anterior thalamic nuclei and in the medial and lateral mammillary nuclei, restricted to the mid-dorso-ventral lamina of the posterior part of the medial mammillary nucleus. CA2 distributes its fibres through the fimbria and the postcommissural fornix to the anterior thalamic nuclei and to the medial and lateral mammillary nuclei, being confined to the ventral lamina of the posterior part of the medial mammillary nucleus. It is not certain whether this field projects into the precommissural fornix. CA3 and CA4 distribute their fibres through the fimbria and precommissural fornix to terminate in the septofimbrial, medial septal and accumbens nuclei, and bilaterally in both the dorsolateral quadrant of the lateral septal nuclei and in the diagonal band nuclei. To summarize and simplify, anterior CA1 projects to anterior thalamic nuclei and mam-millary bodies; CA3 and CA4 project to septal region nuclei; and posterior CA1 projects to all of these structures. CA2 fibre projections do not fall into one clear category.

Taking together all of the findings on afferent and efferent connections of the hippocampus, the evidence is persuasive that the cytoarchitectonic fields are functionally distinct, and "each area will be shown to have a different pattern of activity, determined by its afferent connections and reflected in its efferent projection " (Raisman, 1966, p. 104). The hippocampus appears to divide functionally into two parts with reference to afferent and efferent









systems. The first part, adjacent to the subiculum, includes CAl and corresponds to Cajal's r superior. The second part, adjoining the dentate area, includes fields CA3 and CA4, and corresponds to Cajal's regio inferior. [CA2 connections seem to be intermediate between the two divisions and exact boundaries are difficult to determine. It is interesting, however, that in many epileptic patients with hippocampal sclerosis, the pyramidal cells in CA2 are the only remaining intact cells (McLardy, 1969). McLardy suggests that these cells, as a direct result of their isolation, may be responsible for a positive feedback seizure-progatating system acting through the amygdaloid complex.]

The first part, field CA1, projects to the anterior

thalamic nuclei, both directly and indirectly, through the mammillary nuclei and the marmillothalamic tract. This projection is relayed on to the cingulate gyrus, and hence through the cingulum, to the entorhinal area. From here fibres arise which are distributed principally to the cells of CA1. The alvear tract terminates solely in CAl of the hippocampus, and the perforant tract terminates mainly in CAl and the dentate gyrus, although with some overlap to CA2 and CA3. This system corresponds roughly to the "limbic circuit" of classical anatomy.

The second part, fields CA3 and CA4, projects directly to the medial septal nucleus and the nucleus of the diagonal band and also indirectly to these nuclei via the lateral









septal nucleus through an interseptal relay to the diagonal band. The medial septal and diagonal band nuclei in turn project back to CA3 and CAA.

The dentate gyrus acts as a link between these two parts. The efferents from the dentate gyrus, the mossy fibres, project only to CA3 and CA4. So the first division acts upon the second through a relay in the dentate gyrus, while a reciprocal pathway is provided by the Schaffer collaterals which allow CA3 and CA4 to act upon CAl (see Figure 26).

The significance of these two systems is not known at present. Raisman (1966) suggests that such circuits, if excitatory, could provide the mechanism for reverberation with prolonged activity; or if inhibitory, could explain a damping down of an isolated burst of activity. The hippocampus is capable of both of these types of activity.

The low seizure threshold of the hippocampus may be a

direct result of these systems. The other striking electrical characteristic of the hippocampus, the 4-7 cps theta rhythm, can also be explained by this system. Von Euler and Green (1960) describe the inactivation process: a brief initial burst of excitation, followed by inhibition. They suggest that this pattern, which occurs synchronously in many neurons, is the mechanism responsible for theta rhythm. They present evidence to support the hypothesis that the pathway from septum to dorsal fornix to granule cells of the dentate gyrus, and thence to CA3 and CA4 pyramids, is responsible for this excitatory and inhibitory process.




Full Text

PAGE 1

EFFECTS OF PENICILLIN-INDUCED EPILEPTIC FOCI IN THE HIPPOCAMPUS By GILLIAN HAMILTON A DISSERTATION PRESENTED TO THE GRADUATE COUNQL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DECREE OF DOCTOR OF PHILOSOPHY \ UNIVERSITY OF FLORIDA 1970

PAGE 2

DEDICATION Dan Hamilton

PAGE 3

ACKN0V7LDEGEI-IEMTS . ,/ M ' » I thank Dr. Robert L. Isaacson for everything. I I also thank my Coinmittee members: Dr. Paul Satz, Dr. Robert King, Dr. Charles Weiss, and Dr. Fred King. And finally, I thank Dr. Carol Van Hartesveldt, Dr. and Ilrs. Charles Vierck, Dr. Leonard Scb-maltz, Dr. David Olton, Art Nonneman, Evan Suits, Dale Lewellyn, Mike Woodruff, Dr. Larry Means, Lewis Pinson, Pauletta Sanders the Terrific, and Mrs, Virginia Walker for material and/or moral support at various periods over the past three years. iii

PAGE 4

TABLE OF CONTENTS Page Acknowledgements List of Tables v List of Figures vi Abstract viii Chapter I : Introduction 1 Chapter II: Experimental Procedures ... 14 Chapter III: Histological Evaluation 22 Chapter IV: EEG Results . 36 Chapter V: Behavioral Results 71 Chapter VI: Discussion and Surnmary 79 Appendix A 97 Appendix B 103 References 115 Biographical Sketch '-20 iv

PAGE 5

LIST OF TABLES Table Page 1. Summary of avoidance response data 72 2. Avoidance response data for Group 5 75 3. Locus of penicillin injections 78 4. Layers vjithin the hippocampus, beginning with ventricular surface 81

PAGE 6

LIST OF FIGURES Figure Page 1. Loci of penicillin implants, £ #19, 39, 40, 41 . . 25 2. Loci of penicillin implants, £ #44, 2, 32, 34 ... 27 3. Loci of penicillin implants, S_ #35, 38, 59, 60 .. 29 4. Loci of penicillin implants, S #61, 63, 64, 62 .. 31 5. Loci of penicillin implants, S #65, 66, 67, 69 .. 33 6. Loci of penicillin implants, S #70, 74 35 7. Development of ipsilateral spiking after penicillin injection into the left hemisphere" 38 8. Development of ipsilateral spiking after penicillin injection into the left hemisphere (continued) 39 9. Record of S #31 which showed no iosilateral spiking 42 10. An example of a record with ipsilateral spiking but no contralateral spiking 44 11. Examples of change in form of ipsilateral spike coinciding with spread to contralateral side .... 46 12., An example of the "break through phenomenon": ipsilateral seizures during or immediately preceding spread of spike to contralateral side .... 48 13. Examples of ipsilateral, contralateral, and bilateral seizures 51 14. Examples of independent contralateral spiking ... 53 15. Averages of 20 spikes: trigger on left (injected) side 56 16. Averages of 20 spikes: trigger on right (mirror) side 58 17. Example of recover from spiking 60 vi

PAGE 7

LIST OF FIGURES (continued) Figure Page 18. Example of recovery from spiking, (continued) 61 19. Comparison of the records of a control S_ and a penicillin-injected 63 20. EEG records one hour post-injection for all Group 5A Ss (ipsilateral penicillin; contralateral lesion) 66 21. EEG records one hour post-injection for all Group 5A Ss, (continued) 67 22. EEG records 14 days post-injection for all Group 5A Ss 69 23. EEG records 14 days post-injection for all Group 5A Ss, (continued) 70 24. Diagram of the hippocampus in the rat, redrawn from Douglas (1967) 83 25. Regional differences in termination of afferents, adapted from Raisman (1965, p. 990). 88 26. The two neuronal systems plus the principal links between them, redrawn and relabelled from Raisman (1966, p. 105). 92 vii

PAGE 8

Abstract of Dissertation Presented to the Graduate Council of the University of Florida in Partial Fulfillraent of the Requirements for the Degree of Doctor of Philosophy EFFECTS OF PENICILLIN-INDUCED EPILEPTIC FOCI IN THE HIPPOCAI-IPUS By Gillian Hamilton August, 1970 Chairman: PvObert L. Isaacson Major Department: Psychology The effects of penicillin-induced epileptogenic foci in the hippocampus on electrical activity and on acquisition of a two-v/ay active avoidance task were studied in the rat. The following experimental groups were compared to unoperated and operated controls and unilateral and bilateral hippocampally ablated controls: unilateral penicillin; unilateral penicillin with ipsilateral hippocampal ablation; unilateral penicillin with contralateral hippocampal ablation; unilateral penicillin v/ith ipsilateral hippocampal ablation and subsequent contralateral hippocampal ablation; and unilateral penicillin with contralateral hippocampal ablation and subsequent ipsilateral hippocampal ablation. In none of the above groups was there a significant avoidance deficit when the results were analyzed without regard to locus of implant within the hippocampus. However, when the determination of locus of implant was made in those groups in v;hich the locus viii

PAGE 9

was not obtained, it vas found that an implant in fields CA3 and CA4 , but not in CAl, CA2 , or the dentate gyrus, produced severe avoidance deficits. These results indicate that there is a functional differentiation of the hippocampus, at least with regard to acquisition of this avoidance task. The penicillin-induced spiking recorded immediately after implantation did not appear to be correlated with avoidance performance, but there were indications that greater spike amplitude recorded several weeks later V7as positivelv correlated with performance deficits. ix

PAGE 10

CHAPTER I: INTRODUCTION Hippocampal lesions in humans sometimes result in severe loss of recent memory, yet these effects are not found after similar lesions in animals. It has been suggested that epileptogenic foci (in combination with ablation) may be responsible for the memory deficit found in humans. In order to investigate this hypothesis, experimental epileptogenic foci were induced in rats, and their behavior v;as studied. It was expected that the abnormal electrical activity would produce animals with deficits more closely resembling the human results than animals in which the hippocampus was simply ablated. Temporal lobe resection performed on humans with an epileptogenic focus localized in one temporal lobe results "in partial or complete destruction of that focus and a concomitant reduction in the frequency and severity of seizures. However, when the resection extends far enough posteriorly to include portions of the hippocampus and hippocampal gyrus, many patients also suffer a severe loss of memory, with no concurrent decrease of general intelligence. This memory impairment can be more accurately described as an anterograde amnesia; the patient is unable to retain the memory for any event occurring postoperatively. He can remember an event for only as long as his attention 1

PAGE 11

2 remains fixed upon it; as soon as his attention is distracted the memory is lost. In addition, there is frequently a retrograde amnesia for events occurring in a specific preoperative time interval (several months to several years, depending on the patient) . This deficit, which dissipates with time, does not affect remote memory prior to the time interval. Scoville, Penfield, and Milner (Scoville and Milner, 1957; Penfield and Milner, 1958) suggest that bilateral hippocampal destruction always results in recent memory loss. Unilateral hippocampal destruction might or might not result in memory impairment depending on the state of the remaining hippocampus. They suggest that if the remaining hippocampus is functioning improperly (due to lesions, neoplastic damage, or epileptic foci), the patient will have, in effect, bilateral hippocampal destruction and the recent memory impairment. If the remaining hippocampus is functioning normally, recent memory will be normal. Damage to the nondominant temporal lobe alone will produce deficits in pattern memory. Damage to the dominant temporal lobe alone will produce deficits in verbal memory. However, these deficits are relatively minor, and do not involve memory loss for events. After similar experimental lesions in animals, behavioral testing reveals no evidence of memory loss. For an excellent review of this literature, see Douglas (1967) . An increase of intertrial interval has no effect on acquisition, and hippocampally ablated animals seldom exhibit deficits in

PAGE 12

3 retention. They show superior performance on many learning tasks (e. g. , tv/o-way active avoidance, runway or bar-press for food reward, go-nogo) , no change on other tasks (e.g., discrimination tasks) ; and inferior performance on tasks which are generally categorized as requiring response inhibition (e. g., extinction, reversal, alternation, passive avoidance) . The lesions also cause some changes in mating and maternal behavior, rage thresholds, and adrenal cortex activation, v/hich are probably the result of hippocampohypothalamic influences, but these effects are not relevant here. The results of these experiments with animals have generated several theories, such as implication of the hippocampus in general activation of inhibition of sensory systems (Herrick, 1933); regulating or modulating input (Klflver, 1965) ; decrease in attention to stimuli follov/ed by nonreward via efferent control of sensory input (Douglas and Pribram, 1966) ; and production of the internal inhibition responsible for habituation (Kimble, 1968). These theories, however , contain no suggestion of hippocarapal implication in memory function. In contrast, some animal studies using chemical treatments rather than lesions do suggest that the hippocampus is involved with memory. Deutsch and his co-workers have done a series of studies in the past ten years involving injection of cholinergic drugs into the hippocampal area. They have found that anti-cholines terase drugs such as physostigmine or

PAGE 13

4 diisopropylf luorophosphate (DFP) have no effect if injected immediately after learning; cause severe temporary retention deficits if injected a week or two after learning; and cause recovery of a forgotten habit if injected one month after training (Deutsch and Hamburg, 1966; Deutsch and Leibowitz, 1966) . Scopolamine, an anticholinergic drug, produced opposite effects (Deutsch and Rocklin, 1967). These authors hypothesized that learning causes an increase in the capacity of a synapse to eject transmitter substances. An accumulation of too much transmitter will therefore produce a temporary blockage at the synapse and a concomitant loss of memory . Presumably puromycin also produces amnesia, because of inhibition of protein synthesis. Studies by Flexner and his co-workers have shown that retrograde amnesia for a two-day old habit is produced v/hen puromycin is injected into the temporal areas of rats, mice, or fish in dosages sufficient to produce at least 80% inhibition of protein synthesis for 8-10 hours in the temporal area (Flexner, Flexner, and Stellar, 1965; Flexner, Flexner, and Roberts, 1967). However, if the habit is more than 5 days old, this hippocampal injection will not be sufficient to produce amnesia. It then becomes necessary to inhibit at least 60% of protein synthesis in other brain areas for 8-10 hours by injection into the frontal poles and ventricles, in addition to the 80% inhibition of temporal protein synthesis. Flexner concluded that memory is stored for one or two days in the

PAGE 14

3 hippocampus, after which the engram spreads to other areas. He suggests a complex mechanism for memory storage, involving synthesis of messenger RNA in learning. The RNA, in tiirn, synthesizes a protein responsible for memory expression acting as an inducer of the raRNA gene (Flexner, et al., 1967) . ECS is also hypothesized to produce its amnesic effects by acting in the hippocampal area, since it appears to have less effect in hippocampally ablated Ss (Hostetter, 1968), and the hippocampus is more susceptible to seizures than other brain areas (Green, 1964). The ECS, however, is effective for only a few seconds or hours after training, and there is much dispute over whether it actually affects memory. Several investigators postulate that it may produce its amnesic effects through conditioned fear, disinhibition, or some other mechanism. Nonetheless, it is clear from Deutsch's and Flexner 's research that puromycin and cholinergic drugs injected into the hippocampal area exert an effect on the consolidation of long-term memory. Similarly, in human patients, Scoville and Milner hypothesize that bilateral hippocampal lesions cause inability to consolidate memory. Yet the animal lesion studies implicate the hippocampus in regulation of attention through inhibition of incoming sensory stimuli. Somehow these discrepancies must be resolved. It seems highly unlikely that a phylogenetically old structure such as the hippocampus should have an entirely different basic

PAGE 15

function in man than in animals, although the function may be altered as the phylogenetic scale is ascended. Therefore, a more tenable hypothesis is that the recent memory loss results from a different primary disorder, such as attentional mechanisms. Schmaltz (1968) presented an interesting hypothesis. He suggested that in humans, hippocampal destruction itself does not produce a memory loss. Rather, the presence of an epileptic abnormally functioning hippocampus is the major determinant. The most consistent difference between human and animal Ss is that the overwhelming majority of humans had preoperative psychomotor epilepsy before undergoing hippocampal removal. Schmaltz supported his hypothesis with the findings that a group of rats with bilateral hippocampal lesions learned an avoidance task at least as quickly as normal rats, while another group with an epileptogenic drug (penicillin) injected into one hippocampus was significantly impaired on the task. A third group, with one hippocampus removed and penicillin injected contralaterally , appeared slightly (although insignificantly) more impaired than the group given penicillin alone. Nakajima (1969) found comparable results on a different avoidance task: hippocampally ablated Ss learned almost as quickly as normals, but rats with bilateral hippocampal injection of another epileptogenic drug, actinomycin D, were severely impaired. Why should no hippocampus at all produce better perform-

PAGE 16

7 ance on the avoidance task than a malfunctioning hippocampus? Schmaltz suggests that the abnormal discharge induced by the penicillin spreads through the many excitatory and inhibitory connections of the hippocampus with other portions of the brain, thus interfering with the functions of many different brain structures. It may be that removal of the hippocampi allows the release of inhibition of som.e auxiliary, phylogenetically older pathway which assumes the functions of the hippocampal pathway. This mechanism would presumably not be as sophisticated as the hippocampal mechanism, but would allow adequate and in some cases superior performance of the tasks. However, if the hippocampi are merely functioning abnormally, the auxiliary mechanism is not released from inhibition: instead, the hippocampus itself must perform its own functions — but the abnormal electrical discharges cause severe deficits. Both hippocampi must be removed in order for the auxiliary mechanism to be released from inhibition. Hence if only one hippocampus is removed and the other is injected with penicillin, the one abnormally functioning hippocampus alone must perform all of the work of the hippocampal pathway by itself. Therefore, performance is even more severely retarded than if two malfunctioning hippocampi had been present. Schmaltz did not run a group with bilaterally implanted penicillin, nor did he determine whether the unilateral penicillin-induced focus had spread to the contralateral side. However, Olton (1969) studied performance of rats with bi-

PAGE 17

8 lateral penicillin injections on the same task, and found learning rates similar to Schmaltz' unilateral penicillin group. The similarities between human patients and epileptic rats support the hypothesis that abnormal electrical activity rather than hippocampal ablation might be involved in the memory loss found in humans. A review of the literature on human operations reveals that in most cases of memory loss after a unilateral hippocampal lesion, an abnormal focus was recorded in the contralateral side (Penfield and Milner, 1958; Terzian, 1958; Seraf etinides and Falconer, 1962). When operations were performed on non-epileptics (generally during treatment of chronic schizophrenia) no EEG recordings were made to allow rejection of the possibility that some aspect of the surgery produced abnormally firing cells a tenable view, in light of the many chemical and traumatic conditions which can precede formation of an epileptogenic focus. Support of this hypothesis is offered by the ECS work, and also by recent unpublished work with puromycin, suggesting that this drug exerts its effects through the production of hippocampal seizure-activity rather than protein synthesis inhibition. It is possible that Deutsch's cholinesterase work will prove to have a similar basis. A case presented by Stepien and Sierpinski (1960) provides confirmation for the hypothesis in the human data. A 15-year old girl had been exhibiting recurrent seizures since

PAGE 18

she was 6 months old, and they v/ere increasing in frequency and severity. An EEG revealed continuous multiple spikes, multiple spike-and-wave patterns, and slow wave (2-3/sec) complexes from the right frontal and temporal region. Sporadic sharp waves and wave-and-spike complexes occurred in the left temporal region, 0.5 seconds later than spikes in the right temporal lobe. The girl showed neither impairment of memory for the distant past nor loss of attention, concentration, reasoning ability, or verbal recall. She was able to repeat nonsense syllables or series of numbers after as long as five minutes. However, when tested with task interruption for auditory and visual recent memory, she showed impairments. The task involved presentation of an auditory or visual signal, and after an interval of 0-120 seconds, the presentation of a second signal in the same modality. The S had to determine whether the second signal was identical to or different from the first signal. She was able to perform the task when the interval between the two signals was 60 seconds and v^hen she was permitted to keep her attention on the signal. However, when a distracting stimulus of another modality was interposed between the two signals, she was unable to keep in mind the first signal, and so failed at the task. The girl was subjected to right f rontotemporal ablation. Marked abnormality was found in the region of the amygdaloid nucleus and hippocampus. This abnormal area was typical of epileptogenic tissue, being tough and yellow. Tissue was removed until all high-voltage activity had disappeared.

PAGE 19

10 Postoperatively, electrical recording revealed no epileptic focus on the right side. And, in agreement with the hypothesis, all recent memory disturbance disappeared: the S could discriminate the stimuli when they were separated by a 120-second interval, with a distracting stimulus interposed. In this case, the lesion produced an improvement in a previously impaired memory function. In the human with an epileptogenic focus, memory deficits may or may not be present. Hippocampal ablation, however, produces severe memory deficits in these patients. A focus with a lesion is thus more debilitating than a focus alone. It is possible that additional tissue removal would alleviate rather than exacerbate the memory deterioration by removing residual epileptic tissue, and hence removing the cause of the memory loss. To test this hypothesis, a group of experimental rats analagous to psychomotor epileptics must be established. An important experimental assumption will be that the focus resulting from penicillin implantation is closely analagous to an epileptic focus in a human S. In this study an experimental analog of human focal or partial seizure epilepsy with the focus located in the temporal lobe (temporal lobe or psychomotor epilepsy) is being investigated. In this type of epilepsy, the epileptic discharge originates in a restricted population of cortical or subcortical neurones from which it may or may not spread to other regions of the brain. Schmidt and Wilder (1968) list five unique qualities

PAGE 20

11 which characterize the abnormal cells making up the epileptogenic focus. First, the cells generate an autonomous paroxysmal discharge \tfhich can be influenced by synaptic activity. Second, they possess increased electrical excitability. Third, the cortical surface over the focus is electrically negative. Fourth, a sudden depolarization of the resting membrane potential will initiate volleys of very high-frequency impulses (7 00-1000/sec) . Fifth, the cells have the ability to induce secondary epileptogenic foci in synaptically related areas. Experimentally induced electrical foci appear to possess all of these five qualities. Particularly important is the autonomy of the foci (not true of other forms of epilepsy) . Isolation of cortical from subcortical structures, a procedure which produces electrical silence in normal neurones, does not stop the paroxysmal activity of epileptic cells (Schmidt and Wilder, 1968). This autonomy of focal epilepsy supports the proposition that local application of penicillin can cause a group of cells to become abnormal in much the same manner as human foci. In studying the performance of penicillin-induced epileptic animals, the choice of task is important. Some damage is done to the hippocampus when the cannula used for penicillin injection is lowered. Also, an irritative lesion develops over the course of several days in the brain tissue surrounding the penicillin implant. Yet bilateral hippocampal destruction has been found to produce impairments on

PAGE 21

12 many behavioral tasks (Douglas, 1967). Therefore, it is . necessary to choose a task in which rats with hippocampal lesions are unimpaired, so that any deficits found after penicillin injection cannot be attributed to lesion effects of the drug or cannula. For this reason, the tv/o-way active avoidance task was chosen. Several studies have shown that rats with hippocampal lesions are superior or equal to unoperated rats in acquisition and performance of this task (Isaacson, Douglas, and Moore, 1961; Olton and Isaacson, 1968; Sclimaltz, 1968). In order to study animals analogous to human temporal lobe epileptics with an abnormally discharging focus on one side, rats in this study were implanted with penicillin in one hippocampus. Through hourly and daily EEG recording, the spread of the focus to the contralateral side was carefully observed. Behavioral testing of these rats was expected to reveal performance deficits on the avoidance task. An operation was then performed which was analogous to the temporal lobe resection performed on humans in an attempt to reduce the severity and frequency of psychomotor seizures. In human patients, a severe recent memory loss was found after the operation. The rats were given hippocampal lesions on the same side as the focus, hence removing the focus. Behavioral testing of this group was expected to reveal a severe deficit in avoidance performance. EEG recordings revealed whether the unilateral focus had spread to the contralateral side, forming a mirror focus, and whether the degree

PAGE 22

13 of spread which occurred either before or after the operation could be correlated with performance. This design allows a careful study of the behavioral effects of mirror foci. Another group was hippocampally ablated on the side contralateral to the focus. These rats were expected to reveal behavioral deficits at least as severe as the unilateral penicillin group. EEG recordings were correlated with behavioral data. In the last stage of the study, the remaining hippocampus was removed from some of the rats which had previously received unilateral hippocampal ablation. Any residual abnormally functioning hippocampus should thus have been removed, and performance was expected to return to initial levels. The S_ was expected to be "cured" of his severe performance deficits.

PAGE 23

CHAPTER II: EXPERIMENTAL PROCEDURES Subjects The Ss were 68 male Long-Evans hooded rats from Blue Spruce Farms, weighing 275-350 grams at the start of the experiment. Surgery Group 1 was unoperated. Ss in Groups 2 through 9 were anesthetized vrith pentobarbital and placed in the stereotaxic instrument for surgery. Surgical procedure for each group is outlined below. Recovery period after initial surgery for all groups was approximately 3 1/2 weeks, unless otherv/ise indicated. Group 2; Untreated Operates A midline incision was made in the scalp, and the skull scraped free of fascia. Two small circular holes were drilled in the skull 3.8 mm posterior to bregma and 5.2 mm lateral to the midline. (These stereotaxic coordinates were derived from Pellegrino and Cushrnan, 1969.) Two holes were drilled anterior and posterior to each hole through which small screws were fastened to the skull. a 22 gauge blunt hypodermic needle was used as a cannula. This was placed in an electrode holder and lowered 5.3 mm below the skull into the left hippocampus. Next, a stainless steel twisted bipolar electrode insulated except at the tip and with a tip 14

PAGE 24

15 separation of less than 1 mm v/as lowered into the homologous position on the contralateral side, and fastened to the skull v/ith dental cement. Then, using the same procedure, a second electrode v/as implanted into the ipsilateral hippocampus using the same coordinates as were used for the cannula. The scalp was then closed with silk sutures and EEG recording V7as begun. In a few sample rats, recording was made on the cannulated side through the cannula itself, by using a larger 2 0 gauge needle and lowering an insulated wire with the tip exposed dov;n into the cannula. After the operation, S_s were given .25 cc bicillin injected intramuscularly into the hind leg. Group 3; Unilateral Hippocampal Penicillin The animals in this group were subjected to the same procedures as the animals in Group 2, except that the cannula was loaded with S-R penicillin in dry powder form (Parke, Davis, and Company, Detroit, Michigan). The penicillin v^as spread evenly on the bottom of a glass dish. The cannula was tapped lightly on the drug 25 times to force a plug of the substance approximately 1 mm long into the tip of thei cannula. The end was then' sealed with bone wax to prevent diffusion as the cannula was lowered into the brain. The plug was forced from the cannula with a stainless steel wire cut to extend just beyond the tip of the cannula. Electrodes v;ere implanted bilaterally using the procedures outlined from Group 2. Group 4 ; Unilateral Hippocampal Lesion

PAGE 25

16 After incision and retraction of the scalp, a hole was drilled over the left hippocampus using a dental burr. The hole v/as enlarged with rongeurs. The dura was punctured v/ith a needle, and then reflected with scissors. The exposed cortical tissue was then aspirated, using a 21 gauge aspirator tip. Finally the hippocampus was removed as completely as possible without damage to the underlying thalamus. Cotton balls soaked in saline were applied to arrest bleeding. After the bleeding had ceased, these were replaced with gelfoam plugs. The temporal muscles were pulled back in place, and the scalp v/as closed with silk sutures. Bicillin (.25 cc) was injected intramuscularly into the hind leg. Group 5A; Unilateral Hippo campal P enicillin; Contralateral HippocampalTesion Ss in this group were implanted with S-R penicillin in the left hippocampus, according to the procedures outlined for Group 3. At the same time, chronic bilateral electrodes were implanted in both hippocampi. Approximately two weeks after the drug implantation, both electrodes were pulled out, and the contralateral hippocampus was removed according to the procedures outlined for Group 4. Groups 5B, 5C, 5D, and 5E : Ss in these additional subgroups were treated 3-4 months after Ss in the main groups of the study, and the data are analyzed separately. In these subgroups, different forms of penicillin were implanted in the left hippocampus according to the procedures outlined for Group 3. All Ss v/ere also given right hippocampal lesions according to the procedures

PAGE 26

17 outlined for Group 4. The subgroups differed in details of treatment. Group 53 . — Ss in this group were implanted with S-R penicillin. Six days later, the contralateral hippocampus was removed. The recovery period (after initial drug injection) was 13 daysGroup 5C . — Ss in this group were implanted with S-R penicillin. The contralateral lesion was made either immediately before or immediately after the drug injection. The recovery period was 14 days. Group 5D . — Ss in this group were implanted with a mixture of buffered potassium penicillin G and cobalt. For these rats, the cortex overlying the hippocampus was aspirated prior to lov;ering of the cannula. This method allowed E to observe the cannula entering the hippocampus, thus insuring correct locus of implant. Lesions V7ere made immediately before or after the drug implant. The recovery period v/as 14 days for one S, and 5 days for eight Ss. Group 5E . — Ss in this group were implanted with S-R penicillin from a new shipment, with an expiration date of January, 1975. Again, the hippocampus was exposed prior to implantation. Lesions v;ere made immediately before or after the drug implant. The recovery period was 5 days. Group 6: Unilateral Hippocampal Penicillin; Ipsilateral Hippocampal Lesion £s in this group v/ere implanted with penicillin into the left hippocampus according to the procedure outlined for Group 3, and bilateral electrodes were implanted. Approxi-

PAGE 27

18 raately two v/eeks later, both electrodes were reraoved. The left hippocampus, which had been injected with penicillin, was now ablated. Group 7: Bilateral Hippocam p al Lesion Ss in this group were given bilateral hippocampal lesions according to the procedure outlined for Group 4. Group 8: Unilateral Hippocampal Penicillin; Contralateral Hippocampal Lesion; Ipsilateral Hippocampal Lesion Ss in this group were first given left hippocam.pal penicillin implants and bilateral electrode implants. Approxi mately 2 weeks later, the electrodes were removed, and the contralateral hippocampus v;as ablated. (Up to this point, treatment v/as identical with Group 5A). Then, 2 1/2 v/eeks after the drug injection, the implanted side was ablated. Group 9: Unilateral Hippocampal Penicillin; Ipsilateral Hippocampal Lesion; Contralateral Hippocampal Lesion Ss in this group were first given left hippocampal penicillin implants and bilateral electrode implants. Approximately 2 weeks later the electrodes were removed, and the same (left) hippocampus was ablated. (Up to this point, treatment was identical with Group 6.) Then, 2 1/2 weeks after the drug injection, the contralateral hippocampus was ablated. EEG Recording EEG recordings were obtained from all Ss with permanent electrodes. Recording from both hippocampi began immediately after the scalp was sutured (10-20 minutes post-injection) and continued for approximately two hours. Whenever it was

PAGE 28

19 necessary to avoid confusion between electrocardiogram and EEG, a third channel recorded heart rate. In a few Ss from each group, the cannula itself was used as an electrode, and the recordings begun v/ithin 15 seconds of the injection. In a few representative cases, the record was fed into a tape recorder so that it could later be analyzed on a Digital Electronics PDP-8 computer. In other cases, a Grass oscilloscope was used to measure spike latencies, and polaroid pictures were taken to record results. EEG recordings were taken at intervals from each S during the time that the electrodes were in place. Records were usually taken hourly during the first twelve hours postinjection, and daily thereafter. During these recording sessions, the S was av/ake, but generally remained motionless in the bottom of a high-walled box; care was taken to keep the room quiet. His movements were carefully observed and reported by observers, and movement artifacts noted on the record. Approximately 2 weeks post-injection, Ss in all groups were anesthetized with pentobarbital, and EEG recordings were made from those Ss with permanently implanted electrodes (Groups 2, 3, 5A, 6, 8, and 9). Recording in unanesthetized Ss continued in those rats without lesions (Groups 2 and 3) until the end of behavioral testing, or about 40 days in all.

PAGE 29

20 Behavioral Testing The avoidance chamber consisted of two unpainted plywood compartments, each 11.5 in. long, 7.5 in. wide, and 8 in. high with a plexiglas top and a grid floor. The compartments were separated by metal guillotine doors which closed on a 1.75 in. high barrier. Midway between the ends of each compartment a 6 w. light bulb was located on the side and a 4 in. speaker (Quam #25A07) in the top. The grid floor was electrified by means of a Grason-Stadler shock generator set to deliver 0.6 mA shock. White noise was provided by a Grason-Stadler noise generator set at 12 dB. Time intervals were controlled by Grason-Stadler electronic timers and latencies were measured using a Grason-Stadler clock timer. At the beginning of training, the animal was placed in the left hand compartment for 3 0 sec. The CS (onset of light and noise in both compartments) was presented for 10 sec. followed by the US (shock to the grid floor in the compartment occupied by the animal) . The guillotine doors were then opened, allowing the animal to escape shock by moving into the other compartment. On subsequent trials, the guillotine doors were opened at CS onset, so that on all but the first trial of the first day, avoidance responses as well as escape responses could be made. Both CS and US were terminated and the guillotine doors lowered when the animal crossed the barrier into the previously unoccupied compartment. The intertrial interval v;as 30 sec. On each trial E recorded whether the animal moved to

PAGE 30

21 the other side before US onset (an avoidance response) or required the initiation of the US (an escape response) . The avoidance or escape latency was also recorded. Twenty-five trials were given daily for a maximum of 15 days. If at any time a criterion of 23 avoidance responses in 25 consecutive trials was attained, testing was terminated. Histology After behavioral testing, the animals in Groups 2-9 were sacrificed and intracardially perfused with 0.9% saline followed by 10% formalin. The brains were removed and embedded in celloidin. Sections were cut at 10 microns and every tenth section was stained with thionin.

PAGE 31

CHAPTER III: HISTOLOGICAL EVALUATION All brains were carefully examined microscopically to determine the extent of aspiration lesion, locus of drug implant, and loci of electrodes. Reconstructions of these parameters were then made on representative diagrams of the brain (see Appendix A) . Among groups with hippocampal ablations, Ss did not differ significantly in extent and locus of lesions. Brain damage v;as generally limited to the posterior neocortex, the corpus callosum, the fimbria, and the hippocampus. It was sometimes difficult to differentiate between electrode tract and cannula tract, particularly since one electrode was lowered through the same skull hole as the cannula, using the same coordinates. However, the electrode tracts were generally more symmetrical, and the contralateral tract coold of course be identified. Also, the cannula tract, but not the electrode tract, often contained bright orange-yellow non-nuclear material, distinguishable from the paler yellow blood cells. This deposit probably represented residual bone wax used to seal the cannula. In Group 2, electrodes were implanted without any penicillin, allowing verification of characteristics of cannula tracts. Because the locus of drug injection is critical in this study, photographs of loci for all animals in groups with the 22

PAGE 32

23 injected hippocampus intact at the time of histology (Groups 3, 5A, 5B, and 5C) are shown in Figures 1-6. In all other Ss, the injected side was subsequently removed, so that locus of implant into the hippocampus cannot be confirmed in these groups.

PAGE 33

Figure 1. Loci of penicxllm implants. Top left: S ?fl9. Top right: S #39. Bottom left: S #40. Bottom right: S #41.

PAGE 35

5 Figure 2. Loci of penicillin implants. Top left: S #44. Top right: S #2. Bottora left: S #32. Bottom right: S #34

PAGE 36

27

PAGE 37

Figure 3. Loci of penicillin implants. Top left: S #35. Top right: S #38. Bottom left: S #59. Bo'ttom right: S #60 V

PAGE 38

29

PAGE 39

Figure 4. Loci of penicillin implants. Top left: S #61. Top right: S #63. Bottom left: S #64. Bottom right: S #62.

PAGE 40

31

PAGE 41

\ Figure 5. Loci of penicillin implants. Top left: S #65. Top right: S #66. Bottom left: S #67. Bottom right: S #69.

PAGE 42

33

PAGE 43

Figure 6. Loci of penicillin implants. Left: S #70. Right: S #74

PAGE 45

CHAPTER IV: EEG RESULTS Immediately after lov/ering of the cannula in Groups 2, 3 , 5 , 6 , 8 , and 9 the cannula was removed from the brain , an electrode was permanently implanted, the scalp Was sutured, and connectors were fitted to the electrodes. Electrical recordings of brain activity in most Ss began about tv/enty minutes post-injection, the time required for the above procedures. At this time, spiking was invariably recorded in the left (implanted) hippocampus. In order to observe the development of this spiking, recordings x/ere made in several rats using the cannula as an electrode. . In these Ss, recording began a few seconds after injection of penicillin, and development of ipsilateral spiking could be traced. Generally, small monophasic spikes occurred at irregular intervals for the first few minutes, and then became more regular and grew in amplitude and complexity showing biphasic and multiphasic forms. An example of development of the ipsilateral spike is shown in Figures 7 and 8. Only four Ss did not show regular ipsilateral spiking after the penicillin injections. In one of these Ss, histological examination revealed that the electrode had passed through the hippocampus and entered the ventrobasal thalamic complex. Two other Ss (No. 6, No. 8) were in Group 9, which had received bilateral hippocampal lesions 36

PAGE 46

Figures 7 and 8. Development of ipsilateral spiking after penicillin injection into the left hippocampus (£ #28). Top trace: left (injected) hippocampus. Bottom trace: right hippocampus. First record: 30 sec. after injection of penicillin, bilateral seizure like spiking. Second record: 2 min. post-injection, no spiking either side. Third record: 3 min. postinjection, first ipsilateral spikes. Fourth record: 5 min. post-injection, ipsilateral spikes increase in magnitude and become more regular. Fifth record: 10 min. post-injection, ipsilateral spikes continue to increase in magnitude and become more regular. Sixth record: 12 min. post-injection, first contralateral spikes appear. The pulses on time trace occur every second. The calibration mark at the beginning of the time trace represents 0.5 mV. Gains are constant for all traces.

PAGE 47

M. I'Mi uvilHi'Uliln Tgyr.^enBS'i'.'-v'J'.Trr'' "• 1 ' — • — ' — — ^ — ' — — — • — — — ' — ^ — — — Vv— ^ E

PAGE 48

JjW— 39

PAGE 49

40 prior to histological analysis, so electrode loci could not be determined. It seems probable that the electrodes V7ere not positioned correctly in these Ss. The record of the fourth S_ (No. 31) was particularly interesting, as it showed contralateral spikes, but no ipsilateral spikes. However, histology revealed that the electrode was correctly positioned (Figure 9) . In two S_s, contralateral spikes never appeared (Figure 10). In all other Ss, betv/een 20 and 45 min. after the penicillin injection, the first spiking appeared in the contralateral hippocampus. In every case, the spread of spiking to the contralateral side was accompanied by some change in the form of the ipsilateral spike. In som.e cases, the form change was from monophasic to biphasic; in other cases the change was from single to double or from one multiple to another multiple. Many variations in the changes were found. In all cases, however, whenever the ipsilateral spike shotted the new form, a contralateral spike appeared; whenever the ipsilateral spike showed its old form, no spike appeared on the contralateral side. Examples of this phenomenon are shown in Figure 11. In some cases, but by no means always, the appearance of the first spike on the contralateral side occurred during or immediately after a short burst of ipsilateral seizures. An example of this "break-through phenomenon" is shown in Figure 12. In addition to ipsilateral and contralateral spikes.

PAGE 50

Figure 9. Record of S_ #31 which shov/ed no ipsilateral spikes. Top trace: left (injected) hippocampus. Bottom trace: right hippocampus. First record: 20 min. post-injection. Second record; 30 min. postinjection. Third record: 40 min. postinjection. Pulses on tine trace occur every second. The calibration mark at the beginning of the time trace represents 0.5 mV. Gains are constant for all traces.

PAGE 51

"I " i \ ' • ' 1' 1 1 4 2

PAGE 52

Figure 10. An example of a record with ipsilateral spiking but no contralateral spiking (S #4) . Top trace: left (injected) hippocampus. Bottom trace: right hippocampus. First record: 30 min. postinjection. Second record: 60 min. post-injection. Third record: 90 min. post-injection. Pulses on time trace occur every second. Calibration mark at beginning of time trace represents 0.5 mV. Gains are constant for all traces.

PAGE 53

'I'll I I ' ' I ' I 44

PAGE 54

Figure 11. Exaraples of change in form of ipsilateral spike coinciding with spread to contralateral side. First trace: left (injected) hippocampus. Second trace: right hippocampus. Third trace on bottom record: heart rate. First record: __S_ #1. Second record: S #28. Third record: S #14.' Pulses on time trace occur every second. Calibration mark at beginning of time trace represents 0.5 laV.

PAGE 55

/ / ^^^UJUUUUjUUIUjUJUJUUjUJ^^ 46

PAGE 56

Figure 12. An example of the "breakthrough phenomenon": ipsilateral seizures during or iramediately preceding spread of spikes to the contralateral side (S #13) First trace: left (injected) hippocampus. Second trace: right hippocampus. Third trace: heart rate. First record: 15 min. post-injection. Second record: 60 sec. later. Third record: 60 sec. later. Pulses on time trace occur every second. Calibration mark at beginning of time trace represents 0.5 mV. Gains are constant for all traces.

PAGE 57

48

PAGE 58

49 records from about one quarter of the animals also showed seizures (very short latency repetitive spiking) from one or both hippocampi. Sometim.es the seizures appeared only on the ipsilateral side; sometimes they appeared only on the contralateral side; and sometimes they appeared simultaneously in both records (Figure 13) . No movements could be detected during the electrical seizures. The spikes on the contralateral side did not always occur only in response to an ipsilateral spike. In some records, one hour or more after the injection, occasional contralateral spikes appeared which were apparently independent of any recorded ipsilateral activity (Figure 14) . Independent contralateral seizures have already been mentioned. In order to confirm these findings, the data from several Ss were recorded on tape and fed into the Lab-8 averaging system (Digital Equipment Co.). The Advanced Averager Program was used to obtain the average waveform of 20 consecutive spikes for the post-implantation experimental period. The Advanced Averager offers advantages over other averaging programs in that it allows triggering of the average on actual spikes, and the v/aveform can be averaged for time before and after the spike. From the tape record, a series of average waveforms were computed using spikes on the ipsilateral side to trigger the averager. This procedure showed the initial absence of a mirror focus and its later developm.ent . The same record was then averaged again using spikes on the contralater-

PAGE 59

Figure 13. Examples of ipsilateral, contralateral, and bilateral seizures. Top trace: left (injected) hippocampus. Bottom trace: right hippocarapus . First record: ipsilateral seizures {S_ #27). Second record: contralateral seizures (S #7). Pulses on time trace occur every second. Calibration mark at beginning of time trace represents 0.5mV.

PAGE 60

JL 51

PAGE 61

Figure 14. Examples of independent contralateral spiking. Top trace: left (injected) hippocampus. Bottom trace: right hippocampus. First record: S #28, one hour post-injection. Second record: continuation of first record. Third record: S #17, 4 hours post-injection. Pulses on time trace occur every second. Calibration mark at beginning of time trace represents 0.5 mV.

PAGE 62

53

PAGE 63

54 al side to trigger the averager. This allov/ed observation of activity on the implanted side whenever mirror spikes occurred. An average v/as taken of 5 0 msec, before and 100 msec, after each 20 ipsilateral spikes (Figure 15), and then each 20 contralateral spikes (Figure 16) . The averaged spike magnitude was found to be higher on the triggering side, indicating that although some correlation existed between spikes on both sides, the cross-correlation was less than unity: some spikes occurred on each side which were not matched on the opposite side. Recovery followed a similar course in all Ss (see Figures 17 and 18) . Approximately tv/o hours after the implantation, spiking began to decrease in frequency and magnitude on both sides. By about 3 hours postoperative, ipsilateral spiking was frequent, but only a few independent contralateral spikes remained. By about 5 hours postoperative, no contralateral spikes occurred at all, and only about one ipsilateral spike appeared per minute. By about 7 hours postoperative, no spikes remained on either side, and the records were essentially normal, although ipsilateral or contralateral depression remained in some Ss for as long as two or three days before the records appeared completely normal. After 2 or 3 days, however, the records of Ss which had had cannulation without penicillin were indistinguishable from most of the records of Ss which had received the penicillin injection (Figure 19) . None of the electrical phenomena that occurred immedi-

PAGE 64

Figure 15. Averages of 20 spikes: trigger on left (injected) side. Top row, left to right: then bottom row, left to right: Beginning 150 sec. after injection, consecutive averages of 20 spikes show development of the mirror focus. Traces 1 and 2 show left and right hippocampal averages for 50 msec, before and 50 msec, after the spike. Traces 3 and 4 show left and right averages for 100 msec, after the spike.

PAGE 65

56

PAGE 66

Figure 16. Averages of 20 spikes: trigger on right (mirror) side. Top row, left to right; then bottom row, left to right: Beginning 10 min. after injection, consecutive averages of 20 spikes. Traces 1 and 2 show left and right hippocampal averages for 50 msec, before and 50 msec, after the spike. Traces 3 and 4 shov; left and right averages for 100 msec, after the spike.

PAGE 67

i f: fk f 1 ! 1 i. i y • t: J 4 / h 4 I: 1 58

PAGE 68

Figures 17 and 18. Example of recovery from spiking (S #13). Top trace: left (injected) hippocampus. Bottom trace: right hippocampus. First record: 1 1/2 hours post-injection. Second record: 3 hours postinjection. Third record: 4 hours post-injection. Fourth record: 6 1/2 hours post-injection, ipsilateral spikes approxiiiiately once per record. Fifth record: 7 hours post-injection, no spiking, some depression. Sixth record: 36 hoars post-injection, complete recovery. Pulses on time trace occur every second. Calibration mark at beginning of time trace represents 0.5 mV. Gains are constant for all traces.

PAGE 69

/ V J JJ^^OJ J A.I jjU-.'JJ.I. \ \ \ \ \ r i { .1 /I ^ '1 '1

PAGE 70

^'Vl^^'/s•^VV^\'rt^^V,',•;,nV/.•;.'>, 61

PAGE 71

Figure 19. Comparison of the records of a control S and a penicillin-injected S_. Top trace: left (injected) hippocampus. Bottom trace: right hippocampus. First record: control S_, Day 0 (S #49). No spiking. Second record: unilateral penicillin S, Day 0 (S #41). Bilateral spiking. Third record: control S, Day 37. Fourth record: unilateral penicillin S^, Day 42. Pulses on time trace occur every second. Calibration mark at beginning of time trace represents 0.5 mV.

PAGE 72

63

PAGE 73

64 ately after the penicillin injection appeared to have any correlation with performance of the animal on the behavioral task. Among the S^s of Group 5, which had penicillin implanted in the left hippocampus and the right hippocampus ablated, two of the six Ss never learned the avoidance task; the remaining four Ss learned the task in 3-5 days. Yet all Ss shov/ed spiking from one or both hippocampi; and the S_ which showed seizures learned faster than any other Ss. Only one S in the group did not show regular spiking, but instead showed infrequent ipsilateral and contralateral spikes; this S was one of the two v^/hich never learned the task (Figures 20 and 21) . However, when records taken tv/o weeks after the injection are examined, the two Ss which showed the greatest deficits also appeared to have the greatest spike height relative to background activity (Figures 22 and 23) . Since lesions removed the penicillin focus, behavioral correlations with other groups cannot be made.

PAGE 74

Figures 20 and 21. EEG records one hour post-injection for all Group 5A £s (ipsilateral penicillin; contralateral lesion). Top trace: left (injected) hippocampus. Bottom trace: right hippocampus. First record: S #2, neverreached criterion. Second record: £ #13, never readied criterion. Third record: #14, 4 days to criterion. Fourth record: #32, 5 days to criterion. Fifth record: S #35, 4 days to criterion. Sixth record: S #37, 3 days to criterion. No apparent correlation between electrical activity and behavior. Pulses on time trace occur every second. Calibration mark at beginning of time trace represents 0.5 mV.

PAGE 75

iffi ' ^H'^'jJ'Jtw.WJUtyH t.^. ' ^ty^yy i \ 66

PAGE 76

4 I, . , _ / 67

PAGE 77

Figures 22 and 23. EEG records 14 days post-injection for all Group 5A Ss (ipsilateral penicillin; contralateral lesion). Nenbutal anesthesia. Top trace: left (injected) hippocampus. Bottom trace: right hippocampus. First record: #2, never reached criterion. Second record: S #13, never reached criterion. Third record: S #14, 4 days to criterion. Fourth record: S #32, 5 days to criterion. Fifth record: S_ #35, 4 days to criterion. Sixth record: S #37, 3 days to criterion.

PAGE 78

69

PAGE 80

CHAPTER V: BEHAVIORAL RESULTS Table 1 presents the mean number of trials which the nine main groups of subjects took to reach the behavioral criterion of 23 out of 25 consecutive avoidance responses. Appendix B gives the number of avoidance responses made by the individual Ss for each block of 25 trials. Table 1 also shows the standard deviation for each group, and variances were also computed. Because the within-group variances differed so greatly among groups, parametric statistical tests were considered inappropriate to analyze the data. The greatest variance was found in Groups 1, 5, and 6. Groups 5 and 6, which had penicillin implanted unilaterally and a subsequent ipsilateral or contralateral lesion, were the two groups in v;hich it was expected that behavioral deficits might be found. However, although the mean scores were highest for these two groups, and particularly for Group 5, indicating that they took more trials to reach criterion, the overlap in scores was so great that the groups do not differ significantly from control groups using the nonparametric iMann-Whitney U Test. The source of this variance will be discussed later. Despite the lack of significant differences between groups, certain trends are evident in Table 1. Group 2 71

PAGE 81

Eh 00 in "'3' CO CM 3 O M O o in 00 CO in 00 00 o CO H (0 C H O U -H EH O M -P 0) a -p S U

PAGE 82

73 scores were very sirailar to Group 1 scores, indicating that the operative procedure and the permanent bilateral electrode implants alone had no effect on performance of the task. The scores of animals in Group 4, which received unilateral hippocampal lesions alone, were very similar to control group scores. Group 1, which received bilateral lesions alone, learned the task more quickly than control groups. Only two subjects v/ere run in each of these groups because the results have been confirmed in several other studies (e.g., Olton and Isaacson, 1958; Schmaltz, 1968; Olton, 1969) . Animals in Group 3, v/hich received unilateral penicillin injections, were also unimpaired on the task. This result is in contrast to the findings of Schmaltz (1968); however, it does agree v/ith the human data on temporal lobe epileptics prior to surgical intervention. Animals in Group 5, which received unilateral hippocampal penicillin and a subsequent contralateral lesion, were deficient on the task, confirming the findings of schmaltz (1968). In contrast, Ss in Group 8, which also received unilateral hippocampal penicillin and a subsequent contralateral lesion, but also received a lesion of the injected hippocampus, showed no deficits on the task. Animals in Group 6, v/hich received unilateral hippocampal penicillin and a subsequent lesion of this injected side to isolate a possible mirror focus, performed more

PAGE 83

74 poorly than the control groups, although the difference did not reach significance. It seeras likely that, at least for some S_s, the mirror focus had a behavioral effect. Animals in Group 9, which also received unilateral hippocampal penicillin and a subsequent lesion of this injected side but also received a lesion of the mirror focus side, showed no deficits on the task. The large variance found in Group 5 was unexpected: in previous studies, this treatment resulted in more consistently poor performance (Schraaltz, 1968; Isaacson, Schmaltz, and Hamilton, in preparation) . Four additional groups were run in an effort to determine some variable which would increase performance deficits over those found in this study. The treatment changes investigated were: decreased recovery period; decreased time elapsing between implantation and contralateral lesion; and changes in the type of drug used. Table 2 presents the mean number of trials to criterion for each subgroup of Group 5. An analysis of variance of the means yielded a significant difference between subgroups (F = 4.31, df 4/23, p < .01) . Clearly, reducing the amount of time elapsing between drug implantation and behavioral testing did not increase performance deficits, whereas in Group 5A, 26-28 days elapsed; in Group 5B, 13 days elapsed; in Group 5C, 14 days elapsed; and in Groups 5D and 5E only 5 days elapsed. Variation in length of time elapsing between injection and contralateral lesion also appears to have no significant effect on trials to criterion. In Groups 5C, 5D, and 52

PAGE 84

75 Table 2 Avoidance Response Data for Group 5 Standard Deviations Given in Parentheses Group 5A 5B 5C 5D 5E Mean Trials 198 91 120 91 375 to Criterion (162.3) (39.19) (125.5) (41.63) (0.000)

PAGE 85

76 the lesion was made either immediately before or immediately after the drug injection; in Group 5B the lesion was made 7 days later; and in Group 5A the lesion V7as made 14 days later. Penicillin G-K was used to determine whether the form of penicillin used might affect results, possibly because of active ingredients other than the penicillin itself. However, penicillin G-K appeared to have effects similar to those of S-R penicillin. In Groups 5A, 5B, 5C, and 5E S-R penicillin was injected. In Group 5D , penicillin G-K, mixed with cobalt powder, was used. The same drug, S-R penicillin, was used in the studies of Schmaltz (1968); Isaacson, Schmaltz, and Hamilton (in preparation), and this study. However, in the present study the results v/ere less consistent for Group 5A, and the drug appeared to have no effect at all in Groups 5B and 5C. Although spiking for the two-hour recording period immediately after injection was of the same frequency and amplitude for all groups it is quite possible that the effects of the drug used in this study were more quickly dissipated v/hen used for Group 5A; and that when the drug was injected 2-3 months later in Groups 5B and 5C, its potency had entirely dissipated. This hypothesis is supported by the results from Group 5E, which was treated exactly like Group 5D, except that a fresh shipment of S-R penicillin was used. The Ss of this group showed large and consistent deficits. If this is the case, then an interesting interpretation of the data can be made. The decreased potency of the drug

PAGE 86

77 quite probably causes it to activate fewer cells within the area of diffusion, so that locus of drug implant becomes critical. The locus of penicillin injection for all Ss in which the site of penicillin injection was intact at the time of histology was therefore carefully analyzed by three different observers. In all cases, close agreement was found between observers: in a few cases, two observers desig nated the immediately neighboring field. Photographs of each of these loci have been presented in Figures 1-6; Appendix A gives reconstruction on coronal sections. Table 3 presents the histological analysis. Injection of penicillin into CA3 or CA4 produced an inability to learn the avoidance task within 375 trials. Injection into CAl or CA2 did not produce this deficit, nor did dentate injections. Two components are necessary for successful performance of the avoidance task. The first component is learning the escape response: once the shock has begun, it can be terminated by crossing to the other side. For days 1 and 2, mean escape latencies were computed for escape trials for Ss listed in Table 3. No differences were found between groups: the impaired Ss were able to learn and retain the escape response as well as other animals. The second component is, of course, learning to make the avoidance response (crossing to the other side as a response to the CS) . It is this response that the Ss were unable to learn.

PAGE 87

78 Table 3 Locus of Penicillin Injection Subject Group Locus of Injection Criterion 41 3 X ^ y 44 LAI 70 35 5A 76 tin CAl 152 65 CAl 63 67 5C PA 1 7 "7 / / Moan — Q 19 3 CA2 95 32 5C CA2 85 Mean =90 2 5A CA3-4 375 34 5A CA3-4 375 66 5C CA3 375 Mean=375 39 3 dentate 137 40 3 dentate 66 38 5A dentate 60 59 5B dentate 50 61 5B dentate 101 63 5B dentate 66 64 5B dentate 87 69 5C dentate 55 70 5C dentate 64 Mean= 76

PAGE 88

CHAPTER VI: DISCUSSION AND SUM-IARY The basic purpose of this study v;as to investigate the hypothesis that hippocampal mirror foci as v/ell as original foci have behaviorally debilitating effects and that these effects can be alleviated by removing the epileptogenic focus. When results v;ere analyzed without regard to locus of implant within the hippocampus, no significant deficits were found in any of the groups receiving penicillin. However, when the determination of locus of implant vjas made in those groups in which the focus v/as not ablated, it was found that an implant in fields CA3 and CA4 , but not in CAl, CA2 , or the dentate gyrus, produced severe avoidance deficits. These results indicate a functional dif f erentation of the hippocampus. The finding that in all rats, regardless of locus, mean trials to criterion for Group 5 (unilateral penicillin; contralateral lesion) and Group 6 (unilateral penicillin; ipsilateral lesion) were larger than for the control groups, whereas Group 8 (unilateral penicillin; contralateral lesion; ipsilateral lesion) and Group 9 (unilateral penicillin; ipsilateral lesion; contralateral lesion) learned at the same rate as controls, suggests that the hypothesis merits further investigation, with careful attention to locus of implant. In view of the important implications of the hypothesis, such research seems warranted. 79

PAGE 89

80 In order to assess the significance of the functional differentiation of the hippocampus found in this study, it is necessary to reviev; the nature of the cytoarchitectonic fields of the hippocampus and to determine to vrhat extent and in what manner they may represent functionally distinct entities . An important cell type of the hippocampus is the pyramidal cell. Axons of the pyramidal cells form the hippocampal efferent and commissural systems and their collaterals and various association pathways. Lorente de No identified four different cytoarchitectonic fields of the hippocampus on the basis of the morphology and arrangement of these pyramidal cells. These layers are called CAl, CA2, CA3, and CA4 . CAl is composed of small pyramidal cells arranged in two layers. The apical dendrites of these pyramids typically have many fine branches given off at right angles into the stratum radiatum. A review of the layers of the hippocampus is given in Table 4. The pyramidal cells of CA2 are considerably larger than those of CAl and are not arranged in two layers, nor do they have branches in the stratum radiatum. The pyramids of CA3 are the largest in the hippocampus. Their apical dendrites usually bifurcate before crossing the stratum radiatum, but they have no side branches. The proximal part of the apical dendrite shaft has thick thorns, which are in contact with mossy fibre endings. The axons give off Schaffer collaterals v/hich leave the stratum

PAGE 90

. 81 TABLE 4 Layers v;ithin the hippocampus, beginning with ventricular surface alveus axons, afferent and efferent; rostrally continuous with fornix, laterally with fimbria stratum oriens stratum pyramidale stratum lucidum stratum radiatum stratum lacunosum stratum moleculare Dentate gyrus : stratum moleculare stratum granulosura pyramidal axons heading for alveus basal dendrites of pyramidal cells afferents to hippocampus from fimbria and entorhinal area commissural fibre inputs pyramidal cell bodies basked cell terminations mossy fibres (axons of granule cells of dentate gyrus) end on thorny projections on initial part of apical dendrites in the stratum lucidum of CAS and CA4 unbranched large apical dendrites, closely packed in radial rows unbranched apical dendrites large Schaffer collaterals from axons of CA3 and CA4 pyramids form a dense cuff of synapses on the pyramidal apical dendrites in CAl and CA2 fine terminal branches of apical dendrites input from perforant path from entorhinal cortex peripheral dendrites of granule cells; continuous with stratum moleculare of hippocampus tightly packed granule cells axons from mossy fibres to end on thick thorns of proximal apical dendrite trunks of pyramids in s. lucidum stratum pol\^orphe in hilum of dentate gyrus; not clearly separable from CA 4 of hippocampus

PAGE 91

82 lacunosum of CAl. CA4 cells, which lie in the hilum of the dentate gyrus, do not have a typical pyramidal shape. They do, however, send axons to the alveus, and have both thorns in contact v/ith mossy fibres and Schaffer axon collaterals. The spatial arrangement of the cytoarchitectonic fields of the hippocampus is shown in Figure 24. Figure 24 -A shows the position of the hippocampus within the whole brain: the fornix extends anterior and ventral. Figure 24 -B shov/s a section cut through the middle portion of the hippocampus. Two interlocking U-shaped structures can be seen: the hippocampus proper, and the dentate gyrus. A close-up of this section is drawn in Figure 24-C. If another section showing these structures cut on the same plane were made in the dorsal hippocampus, it would then be a coronal rather than a horizontal section, and CA2 would be located ventrally rather than medially. With this background, the evidence concerning the functional significance of the cytoarchitectonic fields can be reviev/ed. The work of Blackstad (1956, 1958, 1970), and Raisman (1965, 1966), based on earlier anatomical analyses of Cajal, Lorente de No (1933, 1934) , Nauta (1956) , and numerous other investigators, suggests that both afferent and efferent projections of the hippocampus in the rat show considerable localization with respect to field of origin and termination. The analysis below draws on the work of all these investigators. Two major direct sources of afferent fibres to the

PAGE 92

83 B. Schematic diagram of side view of the hippocampus in the rat, with section through the middle portion. Fig. 24. Diagram of the hippocampus in the rat, redrav/n and relabelled from Douglas" (1967) ,pp. 417-418.

PAGE 93

84 hippocampus exist. The first source is the entorhinal area. The lateral entorhinal area sends fibres into the hippocampus via the perforant tempo roammonic tract to the stratum lacunosum-moleculare of the hippocampus, contacting distal apical dendrites of pyramidal cells in CAl, and to a lesser extent in CA2 . A further group of fibres reaches the stratum moleculare of granule cells. The medial entorhinal area sends fibres through the alvear tract to the basal dendrites of stratum oriens in CAl. The second source of afferent fibres to the hippocampus is the septum. Fibres originating in the medial septal nucleus pass through the fimbria to terminate in the basal dendrites of stratum oriens and the apical dendrites of stratum radiatum of fields CA3 and CA4 of the hippocampus (not CAl) , and to a lesser extent in the dentate gyrus. Relatively few axodendritic synapses are formed on the main dendritic trunk; most are on the small fine side branches. Other proposed afferents influence the hippocampus only indirectly: the cingulum projects to the presubiculum and the entorhinal area; the pyriform cortex projects to the lateral entorhinal area; and the dorsal fornix projects primarily to the presubiculum, although it may relay a few afferents from the hypothalamus and rostral midbrain to CAl. McLardy (1963) presents evidence for one additional projection: he observed in both guinea pig and monkey brains a small fascicle of myelinated fibres projecting between the amygdaloid nucleus and the ventral alveus of the hippocampus.

PAGE 94

85 The commisural system of the hippocampus has been studied by Blackstad (1956) . At the rostral pole of the hippocampus, fibres of the fimbria separate into two distinct groups: the extrinsic hippocampal efferents continuing forv/ard to form the anterior pillars of the fornix; and the commissural fibres. The latter cross the midline as the ventral hippocampal commissure, and turn back in the fimbria of the opposite side to terminate in the stratum oriens and radiatum in all fields of the hippocampus. However, the percentage of these fibres terminating in CAl and CA2 is much greater than in CA3 and CA4. Some fibres also project through the ventral commissure to the stratum moleculare of the dentate gyrus. Also, a sm.all proportion of fibres pass through the dorsal hippocampal commissure, projecting to the stratum moleculare of anterior CAl and to the subiculum; the precise origin of these fibres is unknown . It is probable (and verified for CAl, Blackstad, 1956) that each cytoarchitectonic field is reciprocally connected through the ventral hippocampal commissure with its contralateral counterpart. Sharp transitions in fibre distribution occur exactly at the architectonic limits. Whenever heterotopic rather than homotopic projections are found, the fibres pass only to precise loci on the contralateral side. Thus the idea that commissural fibres from one point spread diffusely over the contralateral hippocampus, with at most a greater percentage of fibres projecting to the homologous

PAGE 95

86 point, appears to be as inaccurate in the hippocampus as it. is in the cortex. The function of such precise organization in unknown. However, the precision of the organization suggests that a specific function is involved. Association fibres in the hippocampus also have precise loci. The largest group of association fibres consists of very short axonal relays within the complex fibre plexus of the various hippocampal strata. A second group of collaterals consists of Schaffer's axon collaterals of fields CA3 and CA4, which turn back through the stratum ratiatum to the stratum lacunosum-moleculare of CA3, where they pass to the same layer of CA2 and CAl. CA2 and CAl, however, do not have reciprocal interhippocampal connections with CAS and CA4 ; their axon collaterals pass to the stratum lacunosxam-moleculare of their ov/n field, and also to the subiculum. The last group of collaterals consists of those involving the dentate gyrus. Posterior CAl, but not anterior CAl, CAS, or CA4, projects to the stratum moleculare of the dentate gyrus. Reciprocally, axons of granule cells of the dentate gyrus gather into supra-and infra-pyramidal bundles and pass to the stratum lucidum of CA3 and CA4 , where they make contact V7ith the thorns at the bases of the apical dendrites of the pyramids. Blackstad (1970) found that each level of the dentate gyrus along a septo-temporal axis projects to an equally restricted transverse level of CA3 and CA4 of the hippocampus. Blackstad suggested that this precise level-tolevel localization implied that other associated systems

PAGE 96

87 (such as the perforant path, Schaffer collaterals, and basket cell system) might be organized v/ith comparable topographic precision. Concerning the functional significance of this level-to-level localization, "all that can be said at present is that the preciseness of the observed localization clearly suggests some precise function in need of a high degree of structural order " (Blackstad, 1970, p. 448). In summary, the regional differences in term.ination of afferents are diagrammed in Figure 25, adapted from Raisman (1965, p. 990). The functional significance of this patterning of cellular connections can at present only be speculated upon. The efferent projections from the hippocampus are gathered together in the alveus and distributed through tv;o pathways, the fimbria and the dorsal fornix. These efferents are well described by Nauta (1955) . (The dentate gyrus appears to have no extrahippocampal projections.) The anterior part of CAl sends its efferents through the dorsal and postcomanis sural fornix to terminate in the anterior thalamic nuclei and the medial and lateral maramillary nuclei, with the exception of the ventral lamina of the posterior part of the medial mammillary nucleus. The posterior part of CAl has fibres distributed both through the dorsal fornix and fimbria. These have a precommissural termination in the septofimbria] , medial septal, and diagonal band nuclei, the ventromedial quadrant of the lateral septal nucleus and the nucleus accumbens of the

PAGE 97

t 88 CO < o < o o m +» c 0) u 0) «H (M Rt o c o -H -!-> fd (D in -P en -H m C Id in CN •H fa

PAGE 98

89 same side. They have a postcoramissural termination in the anterior thalamic nuclei and in the medial and lateral mammillary nuclei, res trie bed to the mid-dorso-ventral lamina of the posterior part of the medial mammillary nucleus. CA2 distributes its fibres through the fimbria and the pes tcommis sural fornix to the anterior thalamic nuclei and to the medial and lateral mammillary nuclei, being confined to the ventral lamina of the posterior part of the medial mammillary nucleus. It is not certain whether this field projects into the precommissural fornix. CAS and CA4 distribute their fibres through the fimbria and precommissural fornix to terminate in the septof imbrial, medial septal and accximbens nuclei, and bilaterally in both the dorsolateral quadrant of the lateral septal nuclei and in the diagonal band nuclei. To summarize and simplify, anterior CAl projects to anterior thalamic nuclei and mammillary bodies; CA3 and CA4 project to septal region nuclei; and posterior CAl projects to all of these structures. CA2 fibre projections do not fall into one clear category. Taking together all of the findings on afferent and efferent connections of the hippocampus, the evidence is persuasive that the cytoarchitectonic fields are functionally distinct, and "each area will be shov/n to have a different pattern of activity, determined by its afferent connections and reflected in its efferent projection " (Raisman, 1966, p. 104). The hippocampus appears to divide functionally into tv;o parts with reference to afferent and efferent

PAGE 99

90 systems. The first part, adjacent to the subiculum, includes CAl and corresponds to Cajal's regie superior . The second part, adjoining the dentate area, includes fields CA3 and CA4 , and corresponds to Cajal's regio inferior . ICA2 connections seem to be intermediate betv;een the two divisions and exact boundaries are difficult to determine. It is interesting, however, that in many epileptic patients with hippocampal sclerosis, the pyramidal cells in CA2 are the only remaining intact cells (McLardy, 196 9) . McLardy suggests that these cells, as a direct result of their isolation, may be responsible for a positive feedback seizure-progatating system acting through the amygdaloid complex. ] The first part, field CAl, projects to the anterior thalamic nuclei, both directly and indirectly, through the mammillary nuclei and the mammillothalamic tract. This projection is relayed on to the cingulate gyrus, and hence through the cingulum, to the entorhinal area. From here fibres arise v/hich are distributed principally to the cells of CAl. The alvear tract terminates solely in CAl of the hippocampus, and the perforant tract terminates mainly in CAl and the dentate gyrus, although with some overlap to CA2 and CA3. This system corresponds roughly to the "limbic circuit" of classical anatomy. The second part, fields CAS and CA4 , projects directly to the medial septal nucleus and the nucleus of the diagonal band and also indirect].y to these nuclei via the lateral

PAGE 100

91 septal nucleus through an interseptal relay to the diagonal band. The medial septal and diagonal band nuclei in turn project back to CA3 and CA4. The dentate gyrus acts as a link between these two parts. The efferents from the dentate gyrus, the inossy fibres, project only to CAS and CA4 . So the first division acts upon the second through a relay in the dentate gyrus , while a reciprocal pathway is provided by the Schaffer collaterals v/hich allow CA3 and CA4 to act upon CAl (see Figure 26) . The significance of these two systems is not knov/n at present. Raisman (1966) suggests that such circuits, if excitatory, could provide the mechanism for reverberation with prolonged activity; or if inhibitory, could explain a damping down of an isolated burst of activity. The hippocampus is capable of both of these types of activity. The lov; seizure threshold of the hippocampus may be a direct result of these systems. The other striking electrical characteristic of the hippocampus, the 4-7 cps theta rhythm, can also be explained by this system. Von Euler and Green (1960) describe the inactivation process: a brief initial burst of excitation, followed by inhibition. They suggest that this pattern, which occurs synchronously in many neurons, is the mechanism responsible for theta rhythm. They present evidence to support the hypothesis that the pathway from septum to dorsal fornix to granule cells of the dentate gyrus, and thence to CAS and CA4 pyramids, is responsible for this excitatory and inhibitory process.

PAGE 102

93 Adey, Dunlop and Hendrix (1960) found that the theta rhytlim occurred in the dorsal hippocampus and entorhinal area during goal-directed behavior. In early training, they found that dorsal hippocampal records in CA3 and CA4 led the entorhinal record, suggesting mediation via the septohippocampal pathvray. In later training, the entorhinal area led the dorsal hippocampus, suggesting mediation via the temporoammonic tract from entorhinal cortex. Adey hypothesized that the theta rhythm itself arose in the ephaptic relations of the dendritic trees of the hippocampal pyramids, and that the integration involved in theta rhythm was produced by a phase comparator mechanism between the tv;o systems. Several investigators have studied the distribution of neural transmitters within the hippocampus. Cholinergic innervation is differentially distributed in CAl and in CA3~ CA4. Although in all fields the greatest concentration of transmitter appears in the stratum lacunosum-moleculare and the stratum oriens, in CAl the concentration in stratum lacunosura-m.oleculare is greater than in CA3 and CA4 , v/hile the concentration in stratum pyramidale and stratum oriens is smaller (Mathisen and Blackstad, 1964; Shute and Lev/is, 1961; Lewis and Shute, 1967) . A similar change in relative strengths is found in distribution of monoaminergic terminals. In general, highest concentrations of transmitters are found in the stratum radiatum, but these concentrations are much greater in CA3 and CA4 than in CAl and CA2 (Fuxe, 1965; Blackstad, Fuxe, and HSkfelt, 1967).

PAGE 103

94 Very fevinvestigators have used treatments sufficiently localized to allow correlation of behavior v/ith cytoarchi tectonic fields. Indeed, as Jackson (1969) points out, only about 15% of the behaviorally-oriented hippocampal articles have used lesions even as small as one third of the hippocampus. And these few articles analyze hippocampal function in terms of anterodorsal and posteroventral divisions (Grant and Jarrard, 1967; Siegel and Flynn, 196 8; Elul, 1964a, 1964b).Grant, Jarrard, and Barenfeld (1967) did find that lesions in the posterior hippocampus, mainly involving CA3 and CA4 cells, impaired one-way avoidance learning, while no such impairment occurred with lesions that destroyed CAl and CA2 cells in the anterodorsal hippocampus. Olds (1969) suggests that the input to the hippocampus through the perforant pathway to the dentate granule cells carries sensory information. The mossy fibres projecting from dentate granule cells to CA3 and CA4 carry sensory patterns in a coded form. The cells of CA3 and CA4 are memory elements, which store sensory patterns by modified synaptic connections in their dendrites. Simultaneously, the CA3 and CA4 cells monitor and store concomitant motor patterns through modified connections along their Schaffer collaterals. Patterns of CAl activity would later carry the output codes of appropriate motor behavior. At present, the evidence for such a schema is chiefly anatomical (McLardy and Kilmer, 1970) .

PAGE 104

95 Schmaltz (196 8) and Olton (196 9) have provided the only other relevant behavioral data. Both investigators injected penicillin into the hippocampus and found behavioral deficits on a tv70-way active avoidance task. Unfortunately, they do not present precise anatomical loci of their implantations v;ithin the hippocampus. However, Olton did find that when penicillin was injected into the medial edge of the dentate gyrus, no performance deficits occurred. He also tried to investigate extra-hippocampal pathways controlling avoidance behavior, particularly the septum and entorhinal cortex. As discussed above, the septum is primarily connected v/ith CA3 and CA4 , while the entorhinal cortex is primarily connected with CAl. Unfortunately, his results did not suggest any clear differences betv;een septum and entorhinal cortex: penicillin injected into either structure, in contrast to hippocampal injections, did not alter normal performance. When penicillin was injected into these structures and the hippocampus was simultaneously ablated, again no clear differences appeared. A more precise analysis in terms of cytoarchitectonic fields, such as was made in this study, is necessary to indicate the pathv/ays involved in avoidance behavior.

PAGE 105

APPENDICES

PAGE 106

APPENDIX A Reconstructions of lesions, drug implant, and electrod tracts. Where both electrode tract and penicillin implant are on the same side, "P" designates penicillin implant.

PAGE 107

GROUP 3 98

PAGE 108

99

PAGE 109

100 GROUP 6

PAGE 110

101

PAGE 111

102 GROUP 9

PAGE 112

APPENDIX B Performance Records of Individual Animals by Groups

PAGE 113

104 o +> to 1-1 -H u Eh C O •H in U tN +) n -H CN 5-) u in in in cn in o 00 in H CO in iH 1-1 O (1) (d Q o (0 (0 (U (Q c o ta (0 o •H o CO CM rH H 00 00 iH in 03 (0 O 00 CM CN CO CO in rH CN o CM o CO iH CN 0, o CN in in vo VD CN iH o iH iH

PAGE 114

105 Q x: o (0 w (0 m o &, 0) Pi o c fd 'd •H o > -p m o •H o c +J o •H m U H -H M 5-1 Eh U CO +J Id VI 4) O •0 CM 0. 3 O CN 00 CO 00 00 03 ) u 1 (N CN CN CN CN CN (N in
PAGE 115

106 >i a Xi u fd W m 0) w o &. 03 0) p; u -H o > g G) CN o <: 0 -p p 0 u u c CN -H Q) 1 X5 Q c iH <: <: 0 m •H (0 -p (0 H vo m w CN CN CN CN (0 a Q 0 0 -p 0 -H in to in c^^ o f-l CN ro • iH iH n o in CN 3 O a> o i-i 'a' O wl s cn CTi (0

PAGE 116

107 >i (« a o M cn Q) m d O a, w Q) 03 O m •H o > c o -H 0) Q> ^^ iH ft e (0 o o Dt H M (U 4J KS H •H G 3 O Days after operation t H iH rH Trials to Criterion 23/25 CO '3' CO in rH iH n CM iH 1^ iH iH CO in CN
PAGE 117

108 (0 Q Xi o M w M o w 0) « O C -d •H O > o •H CO 0) 1-^ (0 a. u o •H K rH u Q) -P fO rH u -p o u -H rH rH •H U H c (1) r-i as Pi O o ft •H K rH >H 0) -P Iti rH •H fl D ID a. 3 o >H o Locus of Implant 1 1 1 i CA3-4 CA2 I 1 u dentate dentate dentate CAl dentate subliculum dentate dentate No. Stages CN CN CN CN CN CN CN CN CN CN Days after Penicillin^ n CN CN CN rCN in CN n rH C^ rH n rH ro rH CO H Trials to Criterion 23/25 in CO O rH in m vo ro vo Mean =19 8 o in CN in rH TOT VD VD CO 14 15 in H CN o o m rH o rH CN rH o o IT o o o n rH VD H 00 rH O r» in VD o CN CN in o VD H CTl rH CN o rH CN CN rH CN ro rH CN CN CN CN o n
PAGE 118

109 0) a ri +> a o o in o o Locus of Implant <^ O CAl CA3 ! CAl dentate dentate No. Stages •H rH rH rH Days after Penicillin f-H iH iH iH rH I — 1 Trials to Criterion 23/25 in C30 n in rn r~ in in VD in H H in 3 r-\ o o CN o CN 00 rH rH O iH o CM rH rH rH O rH Subgroup tn 5C 5C 5C 5C 5C CSl vol in VD VD VD r~VD o o CM (D rH rH rH rH rH H rH rH rH in in in in in in in in H CNJ in VD o VD CM in H in rH CO CN rH H H rH CM CM r-l CN CO CM VD ri-l O CM CM o o rH CTl CM CM rH CM rH CM rH CO o rH o in CM rH rH CM rH H CO 03 VD CM in 00 in VD rH H D Q Q Q Q o Q Q Q in in in in in in in in in 00 CTl o rH CM 00 r00 00 CO 00 00 . rH CTl (0

PAGE 119

110 >i as a Si o
PAGE 120

Ill >1 as Q O w Ui

CN CO CJ\ CN rH ro ro CN rH C5 0)

PAGE 121

112 (It Q ji o (0 (0 <0 to d o cu to 07 0$ <0 o c (0 'd •H o o •H to 0) 1^ rH n] I V o a< &. iH Id nJ rH •rl n o o Days after Penicillin rH rH Trials to Criterion 23/25 CO n rin in rH rH n rHI fNJ r-i i-i rH O rH CTl 00 rvo in CN CM CO rH rH in rH in rH col o rH CN II 0) 2

PAGE 122

113 (d P Xi o W CQ 0) to o a, w < •H I-) iH •H O -H G Q) k] iH o, g (ti o o o O Qa p., 04 -H •H m rH nj nJ 5h Q) -P -p to (0 H rH -H •H W C Oi D H 00 Qi o Days after Penicillin eg rsi CN rH CM rH CN Trials to Criterion 23/25 CN H CTi CO in CO CN CN rH n U5 rH rH CN rv) H rH CO CM CO CM rH og rH O 00 H wl 1 r-l rH CO rH in S

PAGE 123

114 o o w CO 0) to o Pi 0) (D 0 O c rd 'd H O > o •H (0 <0 i4 H t u O r-i X rH («
PAGE 124

LIST OF REFERENCES Adey, W.R., Dunlop, C.,& Hendrix, C. Hippocampal slow waves: distribution and phase relationships in the course of approach learning. Archives of Neurology (Chicago) , 1960, 3_' 74-90. Blacks tad, T.W. Commissural connections of the hippocampal region in the rat, with special reference to their mode of termination. Journal of Comparative Neurology , 1956 , 10_5, 417-538. Blackstad, T.W. , Brink, K. , Hem, J., & Jeune , B. Distribution of hippocampal mossy fibres in the rat. An experimental study v/ith silver impregnation methods. Journal of Comparative Neurology , 1970, 138 , 433449. Blackstad, T.W. On the termination of some afferents to the hiprjocamous and fascia dentata. Acta Anatomica , 1958, 35^, 202-214 . Blackstad, T.W. , Fuxe, K. , & Hflkfelt, T. Noradrenaline nerve terminals in the hippocampal region of the rat and the guinea pig. Zeitschrift fiir Zellfor schunq , 1967, 7_8, 463-473. Deutsch, J. A. & Hamburg, M.C., Anti-ChE-induced amnesia and its temporal aspects. Science , 1966, 151 , 221223, Deutsch, J.A., & Leibov/itz, S.F. Amnesia or reversal of forgetting by anti-ChE, depending simply on time of injection. Science , 1966, 153 , 1017-1018. Deutsch, J.A.,& Rocklin, K.W. Amnesia induced by scopolamine and temporal variations. Nature , 1967, 216 , 89-90. Douglas, R.J. The hippocampus and behavior. Psychological Bulletin , 1967, 67, 416-442. Douglas, R.J.,£, Pribram, K.H. Learning and limbic lesions. Neuropsychologia , 1966, £, 197-220. 115

PAGE 125

116 Elul, R. Regional differences in the hippocampus of the cat. I. Specific discharge patterns of the dorsal and ventral hippocampus and their role in generalized seizures. Electroencephalography and Clinical Neurophysiology , 1964 , 16_, 470-488.* Elul, R. Regional differences in the hippocampus of the cat. II. Projections of the dorsal and ventral hippocampus . Electroencephalography and Clinical Neurophysiology , 1964 , 16^, 489-502. von Euler, C, & Green J.D. Activity in single hippocampal pyramids. Acta Physiologica Scandinavia , 1960, 48 , 95-109. von Euler, C, & Green J.D. Excitation, inhibition, and rhytl-unical activity in hippocampal pyramidal cells in rabbit. Acta Physiologica Scandinavia , 1960, _48, 110-125. Flexner, L.B., Flexner, J.B., & Roberts, R.B. Memory in mice analyzed with antibiotics. Science, 1967, 155 , 1377-1383. Flexner, L.B., Flexner, J.B., & Stellar, E. Memory and cerebral protein synthesis in mice as affected by graded amounts of puromycin. Experimental Neurol ogy, 1965, 13, 264-272. Fuxe, K. The distribution of monoamine terminals in the CNS. Acta Physiologica Scandinavia , 1965, 64, Suppl. 24 7". Grant, L.D., & Jarrard, L.E. Functional dissociation of hippocampal cell fields. Unpublished Paper, 1967. Green, J.D. The hippocampus. Physiologic al Reviews, 1964, 44, 561-608. ~~ Herrick, C.J. The function of the olfactory parts of the cerebral cortex. Proceed in gs of the National Academy of Science USA , 1933, 19"7~7^^m Hostetter, G. Hippocampal lesions in rats weakened the retrograde amnesic effect of ECS. Journ al of Comp arative and Physiologic al Psychology, 195'8, 66, 349-353." " — Isaacson, R.L., Douglas, R.J., & Moore, R.Y. The effect of radical hippocampal ablation on acauisition of avoidance response. Journal of Comparative and Physiological Psyc holocry, 1961, 54 ,' 625-628.

PAGE 126

117 Isaacson, R.L., Schmaltz, L.W. , and Hamilton, G. The effect of irritative lesions of the hippocampus on active avoidance behavior. In preparation. Jackson, W.J. A comment on "The hippocampus and behavior." Psychological Bulletin , 1968, 6_9, 20-22. Jarrard, L.E., Grant, L.D., & Barenfeld, M. Paper presented at MPA, Chicago, May, 196 7. Kimble, D.P. Hippocampus and internal inhibition. Psychological Bulletin , 1968, 7_0, 285-295. Kltlver, H. Neurobiology of normal and abnormal perception. Psychooathology of perception . New York: Grune and Stratton, 1965. Lewis, P.R., & Shute, CCD. The cholinergic limbic system: projections to hippocampal formation, medial cortex, nuclei of the ascending cholinergic reticular formation , and the subfornical organ and supraoptic crest. Brain, 1967, 90.' 521-540. Lorente De No,R. Studies on the structure of the cerebral cortex. I. The area entorhinalis . Journal of Psychology and Neurology , 1933, 45, 381-438. Lorente De No, R. Studies on the structure of the cerebral cortex. II. Continuation of the study of the Ammonic system. Journal of Psychology and Neurology , 1934 , 46_, 113-177. Mathisen, J.S., & Blackstad, T. Cholinesterase in the hippocampal region. Acta Anatomica , 1964 , 56_, 216-253. McLardy, T. Some cell and fiber peculiarities of uncal hiopocarapus. Progress in Brain Research (Elsevier) , 1963, 3 , 71-88': ' ' ~ McLardy, T. Ammonshorn pathology and epileptic dyscontrol. Nature , 1969, 221, 877-878. McLardy, T. , & Kilmer, W.L. Hippocampal circuitry. American Psychologist , 1970 , 2_5, 563-566. Nakajima, S. Interference with relearning in the rat after hippocampal injection of Actinomysin D. Journal of Com.parative and Physiological Psychology;, 19"69, 67, 457-461. Nauta, W.J.H. An experimental study of the fornix system in the rat. Journal of Comparative N eurology, 1956, 104, 247-271. "

PAGE 127

118 Olds, J. The central nervous system and the reinforcement of behavior. American Psychologist , 196 9, 2_4, 114.132. Olton, D.S. Penicillin and the hippocampus. Doctoral Dissertation, The University of Michigan, 1969. Olton, D.S., & Isaacson, R. L. Hippocampal lesions and active avoidance. Physiology and Behavior , 1968, 3, 719-724. Pellegrino, L.J., & Cushman, A.J. A stereotaxic atlas of the rat brain . New York: Appleton-Century-Crof ts , 1969. Penfield, W. , & Milner, B. Memory deficit produced by bilateral lesions in the hippocampal zone. American Medical Association Archives of Neurology : . . and Psychiatry , 1958, 475-479. Raisman, G. , Cowan, W.M. , & Powell, T.P.S. The extrinsic afferent, commissural and association fibres of the hippocampus. Brain , 1965 , 8_8, 963-996. Raisman, G. , Covjan, W.M. , & Pov;ell, T.P.S. An experimental analysis of the efferent projection of the hippocampus. Brain , 1966 , 8_9, 83-108. Schmaltz, L.W. The hippocampus and recent memory loss. Doctoral Dissertation, The University of Michigan, 1968. Schmidt, R.P., & Wilder, B.J. Epilepsy. Philadelphia: F.A. Davis Co., 1968. . Scoville, W.B., & Milner, B. Loss of recent memory after bilateral hippocampal lesions. Journal of Neurology, Neurosurgery, and Psvchiatry , r9 5 7 , 20_, 11-21. Seraf etinides , E.A., & Falconer, M.A. Some observations on memory impairment after temporal lobectomy for epilepsy. Journal of Neurology, Neurosurgery , .-. and Psychiatry , 1962 , 25_, 251-255. Shute, CCD., & Lewis, P.R. The 'use of cholinesterase techniques combined with operative procedures to follow nervous pathways. Bibliotheca Anatomica , 1961, 2, 34-49.

PAGE 128

119 Siegel, A., & Flynn, J. P. Differential effects of electrical stimulation and lesions of the hippocampus and adjacent regions upon attack behavior in cats. Brain Research , 1968 , 1_, 252-267. Stepien, L., & Sierpinski, S. The effect of focal lesions of the brain upon auditory and visual recent memory in man. Journal of Neurology, Neurosurgery, and Psy chiatry , 1960, 2^, 334-340. Terzian, H. Observations on the clinical symptomatology of bilateral partial or total removal of the temporal lobes in man. In M. Baldwin and P. Bailey (Eds.), Temporal-lobe Epilepsy . Springfield, 111.: Thomas, 1958, p. 510-529.

PAGE 129

BIOGRAPHICAL SKETCH Gillian Hamilton was born on February 23, 1946, in New York City. She was graduated from the Woodstock County School, Woodstock, Vermont in June, 1963. In June, 1967 she received the degree of Bachelor of Arts from Bryn Mawr College under the direction of Dr. M. E. Bitterman. During the summer of 1967 she worked as a research trainee for Dr. Larry Stein at Wyeth Laboratories, Radnor, Pennsylvania. In August, 1968 she received the Master of Arts degree from the University of Michigan under the direction of Dr. Robert L. Isaacson. At the University of Florida, she worked as a research assistant to Dr. Isaacson v/hile studying toward the Doctor of Philosophy degree in Psychology. She was married to Dan Hamilton on December 24, 1968, and acquired Sloan Hamilton on December 29, 1969. 120

PAGE 130

This dissertation was prepared under the direction of the chairman of the candidate's supervisory committee and has been approved by all members of that c<»iBRittee. It was submitted to the Dean of the College of Arts and Sciences and to the Graduate Council, and was approved as partial fulfill' ment of the requirements for the degree of Doctor of Philosophy. August, 1970. v/^ \ Dean, College of Arts ('and Sciences Dean, Graduate School Supervisory Committee t cEalrEan Robert L. Isaacson, Ph.D. AlCljjirles Weiss, M. D. 1 Satz, Ph.D. Robert King, Ph.D. ' Prederick A. R±hg, Ph.D