Citation
Acquisition and reversal of a two manipulanda differentiation in sham, neocortically, and hippocampally lesioned rats

Material Information

Title:
Acquisition and reversal of a two manipulanda differentiation in sham, neocortically, and hippocampally lesioned rats
Creator:
Milan, Michael Arnold, 1938- ( Dissertant )
Pennypacker, H. S. ( Thesis advisor )
Isaacson, Robert L. ( Reviewer )
Levy, Michael ( Reviewer )
Webb, Wilse B. ( Reviewer )
King, Frederick A. ( Reviewer )
Place of Publication:
Gainesville, Fla.
Publisher:
University of Florida
Publication Date:
Copyright Date:
1970
Language:
English
Physical Description:
vi, 67 leaves : illus. ; 28 cm.

Subjects

Subjects / Keywords:
Discrimination learning ( jstor )
Experimentation ( jstor )
Fornix ( jstor )
Hippocampus ( jstor )
Learning ( jstor )
Lesions ( jstor )
Mental stimulation ( jstor )
Neocortex ( jstor )
Paradigms ( jstor )
Rats ( jstor )
Brain -- Localization of functions ( lcsh )
Dissertations, Academic -- Psychology -- UF
Learning, Psychology of ( lcsh )
Psychology thesis Ph. D
Rats ( lcsh )
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Abstract:
Sham, neocortically, and hippo car.pally lesioned rats were examined in the acquisition of a two manipulanda differentiation under conditions which insured that either the absolute number of reinforced or non-reinforced responses each S_ emitted on each manipulandum during each experimental session were equated. Acquisition performance was not differentially affected by the three lesions, nor by equated reinforced or non-reinforced responding. Reversal performance did not differ for the neocortically and hippocanpally lesioned Ss, but both appeared to be facilitated when compared to sham control Ss. As in acquisition, equated reinforced or non-reinforced responding did not differentially affect performance of the three lesion groups.
Thesis:
Thesis - University of Florida.
Bibliography:
Bibliography: leaves 60-66.
Additional Physical Form:
Also Available on World Wide Web
General Note:
Manuscript copy.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright [name of dissertation author]. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
022274708 ( AlephBibNum )
13586206 ( OCLC )
ACZ2241 ( NOTIS )

Downloads

This item has the following downloads:


Full Text









ACQUISITION AND REVERSAL OF A TWO MANIPULANDA

DIFFERENTIATION IN SHAM, NEOCORTICALLY,

AND HIPPOCAMPALLY LESIONED RATS










By
MICHAEL ARNOLD MILAN


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




















































'JrJr. ERS T r' F:* L F RIDA


3 1262 08552 4113












AC-f ',U`..1_ -. .- :TS


I gratefully t 1 -.,. le: the support and direction provides by

Dr. H. S. Pecrnn.:-c. both in the E; '.;.tiOn of the present project

and, ~,o e iportantly, thiou3' _it ny granite career.

I also express y appreciation to Dr. Robcot L. Ilc?.cson,

Dr. Frederick A. King, Dr. C. Michlel I' ;, and Dr. Wilse B. Webb

for their assistance in th2 devaloccant of this dissertation.

I thank Hrs. Pauletta 3--.::ra ari Mrs. Gloria S:mith for thcir

histological assistance, andi Mrs. Ira Snith who ably typ-d the

present riuscript.










TABLE OF CO :.:.T3


Page
ACKNO'TL z-:E-- f .. .... . .. .... . . .. i

LIST OF TTLS . . . . . . . . . . . . iv

LIST OF FIh.- . . . ........ . . v

ABSTRACT. ................ . .. vi

INTRODUCTICO:. . . . . . o ..* . . 1

METEOD. . . . . .. . . . .. 14

RESULTS .. .. .. .. .. ..* .. ........ . 24

DISCUSSION.. ... . ........... .. 47

PRz-c C3. . .. .. . . .. .. ......... 60

BIOGrHICAL S::: *; ........ . ..... . . .a 67
























ill












LIST OF TA-LL3


Table P-e3

1. TRIALS TO CRT TTRIO IN ACiJSITT(r A'D PCR TrST, FOR
SEAM, -.' 2CTICALT., AiTD HlPPOCX:0 PALLY I:: IO:D10
Ss ASSIC .TO T E R21,n1O;C, 3 1.- :0SZ3 I .'D
cc" -i ( . . .. . . . .. . . 27

2. TRIALS TO CRITZ IO IN ACQUISITION A. D R-4? P5TAL FOR
SEA:, NEOCOYTICA.LY, AD EIDPPOCA'TALLY LESIO3ZD
Ss ASSIGCI TO TH~ ::.;-.. :i.:OCaD 01533 EqUATED
COL.TIO . . . . . . . . . . 28

3. SP'1 ..... P.'.... 70,,,R COR2r2ATION 0'C0 ICi .."?1 (r,) KOR
SHAM, ". TIiLC, y-iD HIP2POPC-'-, LZ3ION L... 3 . 30

4.. SP'... 2,,- R : O.- 2 COijL..L i0: C'* ICl l-i- (rs) %R
:S'+A2, :..3....IC.., ND rIPOC7r2?AL LiSIO: G OUPS
).;^A --:..-._) R:+I2'ORC3D OR N01-.R-0. -ORCED
.J .'i DU I:.T TRilIG . . . . . . 31

5. ANALYSIS OF VARIA TCE ON TRIALS TO CRITICC, . . . 37

6. ANALYSIS 073 VA''IC 0:C 7 " -.7-' TA-S OPTION OF
TRIALS TO CRI:0- . . . . . . . . 38

7. STUDL:,?ZID PR..E STATISTIC A P,.'. IO' i TI.S. . . 46








LIST OF FIGcURIS

Figure Page

1. Traci"-'s of representative cross sections through the
hippoc::;.1 lesion. . . . . . . . . 25
2. Tracings of representative cross sections through the
neocortical lesion . . . . . . . 26

3. N'-:ber of sc:sicr; for Ss in c:m' lion g:c-,p to attain
criterion in acquisition: ReiL:C.orced responses
equated . . . . . . . . . . . 32

4. 2nter of sessions for Ss in e.-ch lesion group to attain
criterion in reversal: R:ir.Corced responses
equated . . . . . . .. . . . . 33

5. ,N-etr of s:.-:lo.:: for Ss in each legion group to attain
criterio:n in acquisition: ...-rclnforce re:~ ones
equated . . . . . . . . . . . 34
6. Nu-:i'r of sessions for Ss in each lesion groups to attain
criterion in reversal: .:'.-reinforce. responses
equated . . . . . . . . . .. 35
7. C',3.-tive percentage of Ss in each lesion condition
attaining criterion in successive 5 session
blocks. . .. . . . .. . . . . 39

8. Cunalative percentage of Ss in each re.'ss'-reinforcesrnt
condition attaining criterion in uc-.csive 5 session
blocks. . . . . . . . . . . . 40

9. Ciurlative psrcezntage of Ss attainirn criterion in
acquisition and rLv-erzal in successive 5 session
blocks. .. . . . . . . . . .. . 41
10. Cumulative percentage of Ss in e-ic lesion condition attain-
ing the acquisition criterion in successive 5 session
blocks. . . . .. . . .. . . . 43
11. Cumulative percentage of Ss in each lesion condition
attaining the reversal criterion in successive 5
session blocks. . . . . . . . . . 44
12. Mean trials to criterion in a:quoisition and reversal for
Ss in each lesion group . .. . . . . 45









Abst'rct of Dissertatioa Presented to the Gtr>uate C-.' -i in
Partial ualfinl-ent cf th3 Rcquirm:ent. for the Dir:ec of
Doctor of Philosophy at the University of Florida


AC:.iI.-ITION AND RE7V'.. OF A l..1 MANIPU DA Di ATIlON IN
SH:. :, 2 :W)CORTICJLLY, A!iD HIPPOC'AALLY LZSIOJISD PRATS

By

Michael Arnold Milan

June, 1970

Ch'aI ,-,n: H. S. P eny '-
Major D treatment: Psychology

Sham, ncocortically, and hbippoca.pally lesione.C rats voer ez-

a-tI:~ in the a::-. 1itito:I of a tvo 1nil-.)a diffc tietition under

c;,:-:litions zhicli I,".1.- that either the absolute 'nop-, 7.r of reinfo J .'.

or non-reinforcod. responses each S emitted on each .ip. '.; duri

each expcrJ.-.:r-t:.l session wore equated. Acquisition performLance was

not diffe.:: .t.1 affecte.l by th three le,.io-L, nor by equated

reinforcA-.l or non-reinforced r:.;c:-in>. Reversal performance did not

differ for the n0ozortically and hisp.-:_-.:-' lly lezio:.:d Ss, but both

app:-c' to be facilitated i,'-- compared to sl.'uJ control Ss. As in

acquisition, equated reinfoicel or non-reinforced responding did not

differentially affect performance of the three lesion groups.












The hip_ ,- _-:pus, nestled in the ir-erl folds of the temporal

lobe, has been subjected to more intense experimental investigation

within the past deri.e than in all previous years combined (after

Douglas, 1967). The early neuroanatonical investigations of Papez

(1937) which first linked the paleocortical structures of the rhinen-
cephalon with emotional behavior were z:..:rted by the contemporary

experir.i-cnt.l investigations, perfor7,- by Kluver and Bucy (1939), of

the effects of tc;- :.:1 lobestec-.j on beha-ior. MacLean (1954, 1955,

1957, 1958) fcx Jised this orientation, postulating a dichotomy bet'.-ee
the phylog:-.n:ticcally older p leocortex and the more recent neocortex.

The former was p s:er;ted as involved in variety of emotional and

visceral f; .cions, while the latter was t.c... t to be-coicarnd with

more cognitive functions. However, as the hippocaarpus was subjected

to more intense experiimental investigation a more cop-lex picture of

hippoc-:?--l function emerged. Before suinarizing the results of more

recent investigations of the role of the hippoc.- pus in behavior and,

concurrently, reviewing the physiological theories of hippocanpal

function which habe emerged from those data, a general description of the

hippocampus and its' interconnections with other brain structures will

be provided.

The hippoc--Apal formation, composed of the hippocampus (Ammon's

horn), the hippocampal gyrus, the fascia dentata and the fornix,lies along

the medial and ventral border of the temporal lobe where it is wrapped


IN:rOLUCTIO'N










around the posterior surface of the th!a':1us. T-- hippocerspus,

re-ininrcent of the common sea horse .o_ c ous. hitoooca: ous from which

its name is derived, is the -ajor structural component of the hippoca-.:pal

formation. Gross description of the hippoc~apus was provided by lorente

de No (1934, cite! in 1o1ilack, 1967) xho divided it into four s-egJnts:

CA1, located pr:-: 1 l to the subicuilun, Ck2, CA3, and CA4, lying in the

fold of the gra'~-le cell layer of the fascia dentata. The most generally

accepted cytoarchitectonic description of the c.-plex internal structure

of the hipn.c-p--us was provided by Cajal (1955). Starting front the

ventral surface atove CA2, thenprocedi-D vertically, seven Zajor

layers are evident: the ventricular epen,,_ -_, alves, stratau orie-ns,

stratu rial strataui radiun, stratum lacunosu.-, and stratum

molecular. A detailed exposition of the int. al norpholos/ of the

hippocampus and hippc.- .-.1l formation is provided by Meissner (1966).

Two major afferent p-th :-ys serve the hippocampus: the alvear

path through the fornix system and the perforant path through the subicu-

luem. The fibers of the fornix arise primarily in the septal area and

the intral-nir.t.r nuclei of the thalanius (Gr';:n & Adey, 1956). The

system is more involved than this, however, for inputs also reach the

hippocampus from the asc-endin-r reticular activating system of the mid-

brain and th2la-us (Sermn & Arduini, 1954) and from the hypothal. lnus as

well (Feldman, 1962). The perforant path reaches the hippoc:::pus by way

of the entorhinal cortex and sublculu'-. The temporoamonic tracts pass

from the entorhinal cortex through the subiculum to the hippocampus

proper. The entorhinal cortex, in turn, receives its afferents from

considerable areas of the neocortex (Green, 1964). In addition, there is








evidence for direct fibers from the cingulum attaining the hippoci-pus

via the perforant rath'ay (Adey, 1961).

Fibers passing into the fimbria constitute the main efferent

pathway of the hippoca..-pus. These fibers cross to the contralateral

hippoc-:.i'Js via the hiv..caepal commissure, or enter the fornix and

project variously to the septum, hypothalanus, anterior thalamus, and

rostral portions of the brain stem (Green & Adey, 1956). The hippo-

campus also gives rise to efferent fibers to the entorhinal area via

the temporoanonic pathway. In addition, Gloor (1960) presents evidence

for primary hippoc ::2'"-'.gdaloid fibers as well as secondary &.:10.lo-

hippoc':--- connections. An int-7-ive review of the literature pertain-

ing to the neurco%-to:.ic.l investigation of the hippocampal afferent and

efferent systems may be found in Gr.;n (1964) and Str::f (1965).

In an attempt to assess the contribution of the hippocampus to

the physiolLgic21 substrata of behavior, researchers have examined the

effect of hippoc- pectomy upon a wide variety of behaviors. One of the

most studied classes of behavior falls under the general notation of

avoidance conditioning. Interest was spared in this paradigm by the

relatively early study of Kimura (1958) who found that rats with bilateral

posterior hippoceapal lesions were deficient when compared to neo-

cortically lesioned and sham operated subjects (Ss) in their ability to

vithold a well-practiced, food motivated approach response following the

introduction of punishment (electric shock) of the consumatory response.

The essential characteristics of this experimental paradigm are proto-

typic of what is generally referred to as passive avoidance conditioning.

Such deficits in passive avoidance have since been replicated under









varyih- conditions in nMuwrous studies (e.g. ICscon :;:ic~.:1- l,

1962; Ki:ble, 1963; cite]L:-l & !:il r, 19?3). Those styles which

have failed to replicate thcso findir s have reported restricted. lesions

involving only the dora1 portion of the hippoca-.s (3oit':.o & I cs. or,

1966; Kvcin, Sotcklicv & KaYv, 1964), or have employed a response of

low probability (Ki:.ble, Kirkby & Stein, 1966; 'i ocuir & jills, 1969).

It is possible to ccistract various cxplc.tions of the i cderlyinr;

nature of the. o- .hippoc- pcto yi deficit; o'ne such s.. ;estioa

is that hippos., l lesions in so.e wvy vitiate the aversive effect of

the p nishiig st.ili s. Such a position woui 1. lead to the prediction

that hippoccI- actoeiz cd S vould 12 deficicit in a vide rP of sheI

moti e1 tb:; :'.-ir This cdoes not arp to b the case, ho- er, for

.,hip .3cto:id Ss are not nc. C ;: rily roI rd .' :1ei coi .rei to

control Ss in their ability to a"ter a variety of active avoid.ace

tasks.

In the typical one-:ay active avoidance par-ad.ii Ss is required

to zIOve front c-.-. conpart-irent of a shuttlebox to a second in the presence

of a warning signal to avoid an avcrsive stimulus is delivered and S mist

then perform the response in order to escape from it. Follovitlr completion

of the trial S is returned to the original compart.,nt of the shuttle-

box and the procedure repeated. Although Niki (1962) reported that

destruction of the hippocatpus had no effect on this variation of

avoidance conditioning, more recent investigatiL:z have indict--t that

there is a lesion-i:2.ced deficit which, h: ::'.cr, appears to be of a

lesser relative ?-ritude than that found in the passive avoidance

paradigm (':::"-' & .." son, 1966; Olton & Isaacson, 1967).









If the above avoidance paradigm is modified so that S is not

returned to the original compartment of the shuttlebox following each

trial but inste-'f must return to it as the response in the following

avoidance trial t."e task is reter.-.A t-:o-.':y active avoid.:e. .:hen

compared to neocortically lesioned and sham opi'rat>.d Ss, hippocanpectom-

ized Ss appear to be facilitated in the acquisition of this iespons

(Isaacson, Dou.-.s & Moore, 1961). It has been sj;:3ted that facili-

tation is due to the p.-ese.ice of a passive avoidance component involved

in the t'o-: 2 active avoidance paradigm which interferes with acquisi-

tion in control Ss (S is required to return to the compartment which has

most recently betn associated with aversive stimulation). Hipoc:-.,.:-

tomized Ss which have been demonstrated to be relatively ip-rvious to

the effect of the contingencies necessary for the instate-ent of the

passive avoidance response are not so hampered r.d consequently acquire

the two-way active avoidance response more readily (T-:la-s, 1967). The

slight deficit seen in th- hippocampectomized Ss' acquisition of the one-

way active avoidance task can also be related to the deficit hippo-

campectomized Ss manifest in passive avoidance. Here, however, the

hippocampectomized Ss' tendency not to avoid the compartment associated

with aversive stimulation retards acquisition relative to control Ss

(Olton & Isaacson, 1967).

A second formulation of the underlying nature of the hippocampal

contribution to behavior which, like the aversive stimulus position,

relates to the passive avoidance deficit suggests that hippocampal

lesions enhance the reinforcing properties of appetitive stimuli or,









altei.-:tively, holds that the hipo ac:.>l lesion in so:e fashion eleDates

drive level relative to non-hippocSsupectodized Ss under equal levels of

deprivation. J:-. .rd (193) has briefly reviewed the lite-ture which

supports this position and points out that in a:lition to the intimate

conr-ctions of the hippoc- ,; ,ith structures important for physiological

ho.eostasis, behavioral evide..:a indicates that hi;.*-."e. -ctonized Ss

are more active in both novel and non-novel situations, ini-ease their

response rate for food and water, atid show sloc. extinction of a food-

motivated running response. Altho-ugh hip:oc=ap-ctoMzcd Ss have not

been^ found to eat more food then control Ss, they have been found to

drink more after It has b a-r ed thl-t the increased drive

hypothesis zs hasz ,.:r-7 been abtE. oned o (: .jlas, 197). However, the

ar---.J-.:its marshalled against this position have stressed the findings

of the avoidance conditio..i,' ;-:-_l:.. and reasoned that because

hippoc:.:':tr.-e. Ss do not appear to be more sensitive to the drive-

inducing properties of aversive stimuli than do intact Ss, it is

inappropriate to posit that the reinforcing. or drive reducing prcp-erties

of appetitive stinuli might differe-.tially affect h .poc.-.-:'pectomized al.v

norLal Ss. Such a critique is cogent orly if the theorist holds that

a unitary or one-process theory of hippocampal function will explain the

whole spectrun of lesion-induced behavioral ar.:, ..ies. Whether it is

possible to formulate a one-process theory of hippoca2pl function has

yet to be demonstrated.

The deficit seen in the passive avoidance performance of hippo-

campectomized Ss has also been viewed as a manifestation of a general

tendency towards response p:r..coveration or, alterr:-tively, an inability

to inhibit responses. A considerable body of evidence is available in

support of such a position. Correll (1957) foun-d that cats subjected







to bilateral hi,:. -'-.1 sti:.ola.tion during the acquisition and ex-
tinction of a food-moti,.atel straIsht alleyway running response sh:e-l

no differcace in rate or acquisition :*.:- compared to control Ss but

did require a greater number of trials in extinction. This firing has

been reliably rc-lic.%ted in rats following hih-.n:J1 destruction

(Jarrard & Isaacson, 1965, Rapholson, Is:.:c3:n & Douglas, 1966). A

closer e~:-irnation of the phenomenon indicates that the interval

between extinction trials is an important variable to be consid:1rl,

for while the incre-;:i1 r:-sistance to extinction is deLonstrable when

trials are sp-ced, the h1p;..'capal lesion deficit dliw:ppears in the

massed presentation of the extf..ction trials (J:- ?,rd & Is:aacson, 1965;

Jarrard, Isaacson & Uic'" l'.cn, 1964). Both Peretz (1965) and c .L.i..l

and Pribran (1966) have reported that hipn. :."pectomized Ss show shorter

response latencies and a greater number of responses to extinction than

do control Ss. Increased resistance to extinction has also been dem:on-

strated in the t'o-'.a active avoidance paradigm (Isaacson, Douglas &

Moore, 1961).

However, Schaltz and Isa_-con (1967) h-7e presented slightly
divergent findings concerning the perfcrLance of hippoc:npectoOized Ss

in extinction. They ran hippocampally lesioned and control Ss to complete

extinction in as many 30-minute free operant sessions as were required

for the attainment of their stringent criterion. No difference was found

between the experimental and control Ss in the total number of sessions

required for extinction. In addition, the hippocTmpectonized Ss shoved

shorter response latencies in only the first extinction session; no

differences bztw'-?n groups were four.1 for any of the subsequent sessions.









Kal.ni (1967) has reported that hippocvazp3cto:;ized Ss sho'r faster

extinction of a freezing reaction t..:':-, as ;'. .:tive of a classically

conditic .:1 emotional response.

The general inability of hippoc .; lly lesioned Ss to inhibit

responses has becn widely demonstration in a number of other situations.

Ellen anrd ilson (1963) found hipnocaipeactonized rats impaired in their

ability to inhibit one type of bar press o-. alopt a second follc'.:i:. a

change in the response requirements for reinforcen nt. Both .iki

(1965) and S,anson and. Isa2cson (1967) have de. -..st \,i a hippoca-pal

lesion-iniduce. deficl-.:,- ; in yielding to stir-lus control following

the initiation of SD-S6 training. Ho' .x:r, the latter authors also

dcnonstratel that hip-Scspctoiz d could readily acquire the

discrimination provided' they were not subjected to a lorg pst history

of continuous reinforc-::nt for r... 'I..n.j prior to the initiation of

discrimination training. Clark and. Is" -.;:... (1965) found that hippo-

campecto.ized Ss were less efficient than control Ss on DUL s:-;illes

of reinfo:cPe'--_nt. A follow-up study by Schialtz and Isaacson (1966)

presented findings analogous to those of C1: a-d Is-,:son (1965),

indicating th-t hippocampally lesioned Ss could perform well on DRL

schedules if not first subjected to prolonged crf training.

The apparently critical role of past learnn.r. in the demonstra-

tion of hippoc.:...pl lesion-induced deficits in discririnartlon and DRL

performance su listedd to some that the hippocampus was not involved in

the inhibition of b.?hr;ior in general, but was more specifically necessary

for the inhibition of well practiced rcsp'nses. The demonstrations by

Kimble, Kirkby and Stein (1966) and W'inocur a!.' Mills (1969) that






hippoc- :-ctomized Ss showed ito deficit in their ability to inhibit an

unlearned e.r-..i response fro- a snail, elevated perch when the response

was punished lead to their for:l! statement of that position. However,

Isaacson, Olton, Bauer and Swart (1966) and Teitelbau: and Milner (1963)

have presented cc-.tradictory data, inlicatirg that hippoca pectonized

Ss are deficient in withholding a naturally occurring response involving

a step-C-& !L from a platform to an electrified grid. The former authors,

who shook the platfoi to increase the probability of response oc..c..nce,

s1i.-ested that tV- escape response employed by Kile, Kirkby arnd Stein

(1966) was too weak or inprobable in nature to adequately reveal a

hiIp:,,capal lesion-indnuccd deficit.

Inhibitory deficits of bippSSoc.:pcto::izcd Ss hae also been

widely exa2ined wit.;;: the context of exploration and s::.:itaieous

alternation paz-rL :s. Roberts, Deuber a:d Broduick (1962) c.o-:' cd

exploration rates of hirzocanpecto)ized and control Ss in T- a-.- Y-Mazes

and fo.._1 no differences in initial rates, but a more rapid decrease in

exploration rate in control than in lesioned Ss. An additional analysis

revealed that Ss with small hippoca!ipal lesions showed a moderately, but

significant, slo,..er exploration rate decrease than controls, and that Ss

with "iLsive hlppoc...;p:l destruction showed no rate decrease whats.osvcr.

Leaton (1965) studied opportunity for exploration as a reinforcer of a

T-maze turning response and found evird:,e.o for acquisition in normal and

sham operated Ss while hippocanpectomized Ss were unable to overcome

perscver-ative teidencies and cons-.u; tly show.ied no acquisition effect.

Forced training was instituted in the second phase of the experiment and

.o.'-ui. of running speed .;cre ta cn. I'.e hippocai:poctomized Ss showed

slower habituation to the reinforcer, i :..: by a slower decline in









runiig3 speed over trials than control Ss. Kirkby, Stein, Kiftle and

Kiible (1967) examined perseveration of a T-7aze response as a function

of gojl-box confinement. With short confinement periods (50 seconds)

hippocaspsl lesioned Ss showed perseverative L- vior vhil- control

Ss spontaneously alte.:-'2:. their responses on successive trials. With

longer c:.fiae:::,.t periods (10 e 50 rcin-.es) both hippoca pectonized

and control Ss den.onstrateJd s:; t'.: :.s alteiration.. A supple. -:ntry

analysis rv"aled hippocatpectoaize3 Ss'perseverate responses Der se

rather than responses to s52ified locations.

Studies of the effect of hippoc- -.l-' lesions upon caze learning

have yieldeol rather consistent res-ults. In ::-"ral, the hippoc.2pap

lesion-induced deficit is sl .:t, if present at all, in very simple

mazes, but as maze complexity increases the lesion-induced deficit in

acquisition becomes incr-:s:l:7 :--r:.' rEanifest. These fi. r.ls have

be-en attributed to the hippocantpectomized Ss' inability to inhibit the

reentry of previously explored blinds and the g-ter fre&-:r.cy of

blinds in p:o-,r:sively more complex cazes (Kaada, Pasnussen & Kviea, 1961;

Kimble, 1963; Kimble & Kimble, 1965). Hosteller and Ih-.:is (1967) have

demonstrated that the hippocampal deficits in maze learning cannot be

attributed to enh:ncel thigmotaxis. The hippoc-:pal lesion-induced

changes in spontaneous alternation and maze performance suggested to

Kimble and his co-workers (Kimble, Kir'rby & Stein, 1966; Kirkby, Stein, Kimble

& Kimble, 1967) that hipp.-ca.-jpectomized Ss suffer from a reduced rate

of information acquisition. This position is inco.:plete, however, for it

fails to account for the unic-ired acquisition rates hippocaapectomized

Ss demonstrate in alternative learning p.rad5.,s.








Although h-ix:, jectoaized Ss appear deficient in their ability

to withhold rs.ons ,s in successive or go-no go discrimination problenis

(Kimble, 1963), numerous cz-1..is have demonstrated that theydo not

differ from control Ss on a vide ~-,-iety of simultaneous discriL.-i:rAtion
problems (Alleyu, 15'..:, 1941; Bro.n, Kaufnan & :-rco, 1969; Grastyan &

Karmos, 1962; Hir:nr,, 1966; Kimble, 1963; i'Nble & Zack, 1967; Swann,

1934, 1935; Teitell.-.., 1964; Webster & Voneida, 1964). *;h-n, hippo-

campectonized Ss are requiral to reverse sic a discrimination, a

pronounced deficit in shifting responding froi that which was previously

reinforced to that which is ne.ly reinforced is regularly observed

(Bro.:.:, ?.-.: _. & Marco, 19;2; Kiqbie & ..-'le, 1965; Rabe, 1963; Stutz

& Roc;:lin, 1968; S'anson & Isaacson, 1967; Teitelb h-a, 1964; T*'!-.-o.pon &

Langer, 1963).

In an attempt to explain the c.'-o. s observed in positively
reinforced b:h:nvior following hippoc-'.;ctoay in terns of the loss of a

single process contributing to such behavior in the intact organism,

Douglas and Pribram (1966) develope. a sophisticated neurophysiological

theory of "problem solving." Although the authors were initially con-

cerned with the hippacczpus, they found it necessary to include in their

theory a second limbic system structure, the anygdala, in order to account

for the behavior of which hippocampectomized Ss are capable. Each of

these structures is postulated to be intimately involved in two distinct

processes underlying problem solving or discrimination learning: the

hippocampus-centered "error-evaluate" process and the complementary

aeyg-di-l.-centerZs "reinforce-register" process. The terns are indicative

of the function of each: the reinforce-register process is depicted as









increasing the future probability of a response which has been folloJeal

by reinforcement; the error-ovaluate process is postulated as decreasing

the future probability of a r: ;nse which has not 1K-' followed by

reinforce&.ent. During discrimination learning in the intact organism

both these processes or systems are cooperative as behavior is brought

under stimulus c:,ntrol.

The propose. neuronal system undcrlyirg the error-evaluate

process involves hippoca.:pally mediated inhibition in a P -:.!.-like

mechanism within afferent systems which serves to "gate out" non-

reinforced stimuli. In the absence of the hippocampus non-reinforcL-:nt

cannot alter bh ior ard discrimination learning cust be accomplished

by the remaining reinforce-register system The effect of reinforce-

ment termed "iipellence" is i:-::- -'ta over reinfo._:! training,

constant in size, and related to the roagnitude of reinforcement and the

effort required for its production. At the primary neuronal level,

impellence is depicted as involving normally occurring collateral

inhibitory processes in afferent systems. The work of De.:'son, Nobel

and Pribra,. (1966) and Spinelli and Pribram (1966) is ta-en as direct

evidence for the exist-nce of these proposed systems.

To sumirarize the Douglas-Pribram theory: It has been suggested

that the hippoc.-.pus is a key structure in an error evaluating system

which mediates the effect of non-reinforced responses during learning.

Organisms with hippocampal disruption are rendered relatively in-

sensitive to the effects of non-reinforcer.czt a.d are therefore required

to learn appetitively motivated tasks via the remaining reinforceuint







sensitive amygdaloid system. Although t::o theory is a posteriori

in construction, Douglas and Pribram (1966) do present soie data

confirming predictions made from the theory.

The present experiment focuses upon the hippocampus and its

proposed involv;::wit in situations involving non-reinforced respon ir .

An experimental paradig3 in which manipulation of reinforccen:, and

non-reinrfoi'c-.t contingencies gr.cvctes differential predictions

concerning the behavior of hippoccapectomized rats has been developed

from the theory in question. In both the acquisition and reversal

phases of a position discrimination, equation of the absolute number

of reinforced res:--r.ses to each of two to-be-discri:-~..tci r:.ipu-r,.'-

ccbinrr2 with differentiation beaten the two in terms of the absolute

number of ncn-rei:.:'orced responses would be predicted from the theo'zy

to retard both acquisition and reversal in hiL:, ..-p-ctoaized Ss when

co-parcl to neocortically-lesioned and sham o;,rated controls. Ho-.:cver,

when the absolute number of Lor.-reinforcdl re;jonic: to the nr-mnule:Ja

are equated and the number of reinforcJ.l responses differ, any hippo-

cazpal lesion-ind-.-ed deficit would be predicted to be of a significantly

lesser magnitude if present at all. Positive results would constitute

support of the Douglas-Pribram theory (Douglas, personal communication,

1968).















Subjects

The Ss *'_--- 60 male Lo -' rats .pp-roxi;ately 125 to 175

days old at the start of training.


Awparatus

A total of four expert' ::tal chambers were employed. One was

constructed in the laboY "-.tory uhile the remaining three were coLsercially

obtained. The.c.atber :tactoed in thie laboratorY *a:s a converted ice

chest with a she.t.-ta1 partition dUi-.ldi it into two cc_:.rtients.

One coipartrenat co2ntained a pellet dis3nse~ r and relat-ed reinforce:_ent

delivery equ'. .-:'t; the second co-par- .At, with the inclusion of a

hardware cloth floor, ': -"r:i 28.5 :.'. by 28 :-... by 23 :-1. high and

served as the ex priLental space. A 7?lTh GerbranLds Co:-.:'!y rat lever

was situated along the verticle center linr. of ore wall, 2.25 ::.. above

the hardwarecloth floor. Reinforce> ::it was delivered to a food cup

situated 5 ms. above the Lanipr.l=.n.u. An e:I.-: st fan provided ventil-

ation, and a 20 VDC bulb located in the center of the ceiling provided

illumination during experimental sessions. The coanercially obtained

chambers were all Lehigh Valley Electronics Model 1316 small cubicles.

A metal food cup was located along the verticle center line of one

vall and rested on the grid floor. Two Lehigh Valley Electronics Model

1352 rat levers were mounted on the same wall, one on each side of the

food cup. The center point of each ranipul--.' .: was 3 ic. above the







floor and 5 m from the neari7t side wall. Illumination was provided by
a 20 VDC bulb located 2 mn. above the center of the plexiglass ceiling.

All c --nipulanda were c--JLil:-ut so t'-.t a weight of approximately 20
grams would activate the r.;ponse circuitry. All e:;2ricental operations

and contirgencies were controlled by a 't..:- tic elcctro-nechanical

programming equipment. Reinforcement consisted of 45 ag. Noyes rat

pellets. A plexiglass cover, measuring 3 :.=. by 7 mn. by 14.5 22. high,

was available to cover either -:.npul-:r.tu in the two manipulanda

chambers, thereby forcir.:- Ss to r3,rnri on the uncovered .I.. i-- T.--iA
when the condition: of trin rj so required.




The 60 Ss were assigned in equal :rn.'rs to the 6 cells pre-

scribed by the first two factors of a 3 x 2 x 2 experimental design

involving re-:zted measures as the third factor. Animals subjected to

hippoc~r:-.1, neocortical, or sham lesions (factor A) were assigned to

conditions of differentiation training which insured that for each daily

session either the number of reinforced responses or the number of non-

reinforced responses (factor B) emitted on each of two mnaipulanda were

equal, and were then tested in both the acquisition and reversal

(factor C) of a two manipulanda differentiation.

Procedure

Upon receipt from the supplier all Ss were placed on ad lib

food and water. Following recovery from the rigors of shipment a mean

base weight derived from five con:ccutive days weighing was established









for each S, and Ss were r :.' i to 8,' of these valr':. an maintained

at that level for the duration of pretraining.

The goal of the pretra in.L phase of the experiment was to

establish in each S a bar ;r;:3 response free of a~,y procedurally-

induced left or right position preference. To accol-..1ih this pre-

training was corKucted in the single anipulandi.u: chaToier. Subjects

were first magazine_ trai-ned and then z'. to pt'&z the manipulannd.a

by the delivery of food rl-...orcemsnt. Special care was t:h'-:- to

insure that no S received a disprcoiriLer-.te amount of training under

crf and low FR reinforcement schedules. Trh reinforcement ratio was

gr-.l.lly escalaPted and. preu.:-.-.v1 was terainatod pon each S's

de:oi.stration of stable respondin under the requireants of an FR 10

reinforcement schedule. Subjects were then returned to ad lib food

maintenance.

Following ric.overy of lost weight Ss as i :7:; to the appropriate

cells of the factorial desi-, were subjected to bilatteal hippoc:..I.:.l

removal, bilat.r.l removal of the neocortex overlyinr. the hippocampus,

or bilateral shank operations in which the dura overlying the neocortex

removed in the neocortical lesions was exposed. Follo'rin. recovery fr'o

surgery a mean base weight derived fro: five consecutive days weighing

was again established for each S, and Ss were reduceLe to 85; of these

values and maintained at that level for the duration of the experiment.

Subjects were then returned to the single manipulandun chamber

and retrained to respond under the conditions of the FR 10 reinforce:.:nt

schedule. With few exceptions reestablishment of control of responding

by the FR 10 sche.le was accomplished during one session of approximately









45 minute: duration. In no instance did this retrALT.ir, require more

than three daily sessions. Following completion of retraining in the

single ',.iril.-P..,, churTar Ss '.",- adva:.-c, to the two manipulanda

chambers in which the expert -:,tal operations were conducted.

During preli .ry trC in- ij in the two manipulanda chambers the

right L,:n.-_i2,-A-.2 .n was first c). i:- with the plexiglass cover provided

for forced traid.1j: Responding on the left r'anipl-.:.': was first

maintained by a crf schedule of reinforce-ient, a,-. then by intermittent

reinforcement. The reinforce -..t ratio was escLl'.'c 1 one step follow-

ing every tenth reinforcement u.'tll 10 reinforce-ents onan 7R 10 schedule

were delivered. Subjects were then removed froi the chibs-r, the plexi-

glass cover roved to the left manipuilandc-, and the preliminary training

regimen r:,:..ted. Those Ss which failed to earn 10 FR 10 reinforcements

on either -::,:.p-aiudum within a 5-minute period during which that

schedule was in effect repeated the preliminary training region the

following day. With few exceptions pretri-.1in._ required no more than

one session approximating one bour in duration; in no case were more than

4 daily sessions required. On the day following the co:.pletion of pre-

liminary training the expericental procedures were initiated.

In discrimination acquisition responses on one ranipulandur

were reinforced onan FR 5 schedule and responses on the second were

reinforced on an FR 9 schedule. Of the 10 Ss in each of the two hippo-

campal and sham lesion groups, 6 vere assigned to one chamber and 4 to

a second. Within each group the relationship between manipulandum and

reinforcemcalt schedule was. counter l. .anced. The two groups subjected

to neocortical destruction were assigned to the third chamber and the

relationship between manipulandum and reinforcement schedule also counter-

balanced. The first portion of each daily e:,:perimental session consisted










of a 5-minute free choice period drin which both manipulania were

ex,:" :e for re.- :-r.-.r. andr reinforced on the appropriate sc'.: es. At

the con: .uj-ion of the test period, which sered to monitor the formations

of the discrimination, the ntber of responses emitted ar. the -I'"sbr

of reinforceaeats L rns d on eacb Laniplai wer-a re .:"'ei. The-

forced training p',.tion to the epriental session .as then initiated.

The purpose of fo:c'2 ti 1. '- differed for each of the two

hippocapal, neocor-tical, and sha'2 lesion ,.ops. One each of the

hippocli.:pal, neocortical, and sham lesion groups Vias rtn under the

ccr..l'tion presc:bin the e:uation, for each S, of the absolute r.:::-

of reirforccl reo r.:s e tced or, e .: rr::2le : during ceCh

daily session. The second hippoca-.pally neocoruically, 2 sa

lesic .?. groups were run unicr the conditioning prescribir the'

equation, for _:'_ S, of the absolute 1: .br of no..-rinr.forced

responses emitted on each ianipi: n> .- d.'ing each daily session. It

should be noted t-.at as a result of the utilization of an !- 5 and an

FR 9 schedule of reinforcement, Ss :.':h emitted an equal number of

reinforced rcsponc.. on each r.:nijl-1,. also emitted twice as :rny

non-reinforced responses on the FR 9 ca.'..la..l.'u as on the FR 5

manipuland-m. Conversely, Ss which emitted an equal number of non-reinforced

responses on the two ,nip..lardi. also emitted twice as a=ny reinforced

responses on the FR 5 m-nipulanrtv as on the FR 9 nLnipula.'nr.

Subjects assigned to the reinforced responses equated procedure

fulfilled a dual relulre: nt during each complete experimental session.

These requirements were: (a) each S earn an equal nurbcr of reinforcements








on the 7. 5 and P2 9 m i:.-l.., a-d (b) a total of at least 50 rein-

forcements be earned on each of the two manip!l?,I-. If S did not

earn the minimum 50 reinforcements or either : :.i- i'..... during the

5-minute free choice period, the forced training portion of the

session involved r n fiJ.g. on both Lip -. alala. At th en d of the

free c-hoice period the required number of make-up reinforcements to be

earned on each r..: ni 'P-'-' was deteri'. 1, one anipul .:.. was

co-ve ed, and S was allowed to respond on the :::.. until the required

number of -':e.-up reinforcc::ats for that r-'.1 ulandun had been

deli,;-1v. The cover vwas then moved to the seco-d manipulanidu and

S was allowed to respod on the first until th2 require :ents for that

B:.';i..-.-' : had been falfille1. For ex::ple, if S ea:ned 40 F2 5

reinro'cennts and 25 Fa 9 rei-.rorce.ets du-ins the free cL..icc

period, S_ would be required to earnn a additional 10 PR 5 reinforce-

ments and 25 FR 9 reinforcements during the forced training period.

As a result, S would have earned an equal nr'.Lber of reinfo:rc':r:ts (50)

on each m aipula:.i during the course of the experi -r:tal session.

The order in which the manipula-.da were covered alternated across

daily sessions.

If S earned 50 or rLre reinforcements on either, or both

manipulanda during the free choice period, the forced training period

involved rzspo,,iirg only one r.niii'ulznda_. At the ter.iination of

the free choice period the difference between the number of reinforce-

ments earned on the two manipulanda was determined, and the manipulniduL

on which S_ had ea:..:. the greater rio'-:r of reinforcements was covered.

The S was then allowed to respond on the second raripLil-ndum until the









reureq ed n ... c' c.:..e-'.. re ifoc k sts deivered. For e: .ple,

if S_ earicd 65 FR 5 reinforcc.:nts ar.1 20 FR 9 reinforce;n2ts during

the free choice p-rioi, the :-" 5 mcanipulnda would bs c;. ered during

the forced trailiin priod and.' S would be alle.cd to rcspondl or- the l1

9 ranip, 1 -tl 45 reipforc,:i:.x ts htr ta been Ieli-vcr:.. This fu l-

filled thVe reqijirc:- nt that S e:it an equvi nmu:ber of reinforced

responses, in this instance 65, on each r.ipulan'2 during cech expcri-

mental session.

Subjects signedc to the L: inf'c:c: responscs equated

proc,-; ce fulfillei a different dual reful : ..t curing each experi-

mental session. 'Th7s r ce ir:Sc nts were: (a) each S cr'.n t:ice as

maniy reinforce-:2ts on th: FR 5 .ii.!-..l c s oc the 2R 9 2nipl -.

a-.l (b) a r.ini: i of 60 reirForc c:onts be carnl! on the 1 5 cTniz2 "-

and, consequently, a mini m'- of 30 reinforcements be earned on the

FR 9 manipulandun. -- free choice period, and subseqcnt forced tii.,-

ing proceeded in a manner analogous to that d.: ,clir t? above for the Ss

assigned to the reinforced responses equated relgien. In the percent

condition, if S failed to earn the minilau 60 0r.1 30 reinforcements on

both the F? 5 and :-' 9 n .ipulaada, respectively, durir.: the free choice

period, the force?, training perica would insure that these minima were

earnel. For exa-ple, if S earned 40 FR 5 reinforce..ents and 25 FR 9

reinforce:eants during the free choice period, he would be required to

earn an additional 20 FR 5 reinforc:-.:nts and 5 FR 9 reinforc::.A.its

during the forcEi training period. The S would have therefore er1rned

the c.ninru, 60 and 30 rein'orc.e-nts on the FR 5 and F: 9 ra'i;']2:?a-

durii:- the c.,urse cf the exoerinrntal session.







If S earned L:O:.2 than 60 ri,;..force:onts on the I-; 5 ranipuland~:1
and/or more than 30 reinforcements on the FR 9 man pulandum during the

free choice period, the forced, training period i'.A:ved only one r. '.,-

landun ..-d served to insure ti t the ratio of reinforcements earned of

the FR 5 r-Fnipulandti;i to those ea:nc on the FR 9 :,1-ifll.-..:' vas 2 to 1.

For e:,- i,1e, if S earned 70 FR 5 reinforcements and 40 PR 9 reinforce-
mt:-ts during the free choice period, S was recruire3d to ear an additional

10 FR 5 reinfo;c :e-nts during the forced training period. As a result,

S earned a total of 80 -i1 5 reinforceLsnts and 40 l. 9 reinforcements
dui-in the c'-..:. of the experimental session, and fulfilled the require-

ment that the ratio of :: 5 to Ji 9 reinforce:;3nts be 2 to 1. If S earnCed

90 :", 5 reinforcements and 10 FR 9 reinforcc:'ents during the free choice

period, S[ as require to earn an additior?.! 35 FR 9 reinfcAc.: .ts

during the forced tr,1,tuit,.' period. ThI S ther.7,ice earned a total of 90

FR 5 reinfo-,--:.,ts and 45 FR 9 reinforcE::ents, and the ratio of FR 5 to

FR 9 reinfoc.: -:-rts was again the required 2 to 1.

Discrimination training was terminated when S attained a
criterion of at least 90, rDc-.?i:ng on the -R 5 r-. i~ul r..- in 9 of 10

consecutive free choice periods. Upon completion of this require -lnt

S_ was subjected to one half esin as r-any daily sessions as were required

for attainment of the criterion and then n,'ed to the discrimination

reversal phase of the experiment. If it became statistically irpossible

for S[ to satisfy the criterion within 40 days of training Sq was con-

sidered to have failed to satisfy the requirements for discrimination

and was then moved to the discrimination reves.:-l phase of the study.










Discriimntiont re-iersal training ;'as instituted for cach S

on the day following termination of the acquisition portion of the

experir.ent In rer-?21 training the roA tion chip I.'. '*e reinforce-

menit sch1'nule a 3anip-~!..'. was rIseerse for each S. Training

in reversal proceeded in tie s" e fashion for the trio groups as

described in the aequisition p -se above. Reversal training was

teKrinated "- each S cat either the criterion of acquisition or

failure ~ est:.b11".: for the acquisition ph.mse of the stidy.


Sur er:y

All Ss -- re op-re:-'1 under :0 /1i,: Ilc:itPl astheei

su.-,.ete ted with .30 cc. of atropine injected int cton: All

opra.t.. p V- prforl: while S was held in a -ti -re stereotaxic

instris-,-nt. A dissecting sccpe was e-rployed to assist in the vi:i,1

0.I, _i of neocorcical ansd hi-.-:,c: .al removal. In all operations the

skull was exposed by Erans of a midline incision, bilateral trephine

holes were placed lateral to the r:i.dl.;' and posterior to bregna. The

holes were e:-L :*_ with rougures to expose the neocortex o;erlying

the dorsal and lateral portions of the hipp;:c..'.:.us. In the sham operated

Ss surg-.ry was terminated at this point. For those Ss sustaining neo-

cortical rc'-oval the dural was cut and the neocortex overlying the hippo-

campus was aspiratel off, with care te':ea not to damage the hi.ppoc:r-:.'s.

For those Ss subjected to hippoc-:L:pctony the operation proceeded until

the th-la-.us was exposed. In addition, hippocan-pal removal was e-;peni.3.

anteriorally as far as the hippc--.pal couissvure as vell as c:.:te:.ld

around the posterio-lateral surface of the thlanVus. Care was taken not







to : a Ie the t'ausis. Followi-j co!pl etion of surgery Ss were

ret .' to their hose cages and maintained on a water amd tetracycline

solution for there to five days.


Histo 1 7

Follol.0 the ter.. i-tic of the e.. .- iEent all Ss were sacri-

ficea with a lethl.1 dose of :eatbutal aresthesia -- intracardially

perfused with saline follow ezd by a 10C for1.7lin solution. All brains

wer.- then i,_., froi the brain c-.ities and those of the Ss in the

nec-.rtical anirl hippoc--. 1.. -1., were infiltrated with, a C.Id embedded

in, celloiin &ai1 sectioned at 15 u. .:c; tenth section was retained,

moutcd on a sl'i and stailnd with thio.in. In addition, for 5 Ss

in each legion --oup t'.:: section follo:'irg the thionin section was

stained with wile r". then mounted.











R SUJI .S


Tr'i.' s of represen-tativ cross sections through the hip, po-.

campal and neocortical le:i :- are presLented in Fi .i:-: 1 a7nd 2,

r-. :: tivaly. Eizpoaapal destracticn reSalarly involved at least 75;,

of that struactu.e and in all instance resulted in the complete

separation of th3 dorsal a.! lateral aspects of the hippoca' al foro-

ation. Spciic da::a to the tha1 .-as was Dninial aindi, when evizi-it,

was typically unilat- .1 it, nature. .. neocortical lesions did not

en...:- a vol 7e of tissue cora-.ale to t -' rc:oe,1 inr the hippo-

c.;:- lesi.cs bc.it d.id apr'- y ...t the nrocortical ,e.' .- tion incurred

by the hippoc2.-;ctcies. .:pc<-?al d: .:g resulting fro the neo-

cortlc.! lesions ,-,as minir-:1 if pr .-:.t, usually unilateral in

nat.It-. CGrozS e.x-zirnatioa of the intact brains of those Ss subjected

to sham, oporat:".:- r.-:--ei no discernible neocortical da:.age.

The sc-ii:s requiral to attain criterion in acquisition and

reversal by these shaa, neoe. :,tically, and hippocamipally lesioned Ss

trained under the reinforced responses equated re-:;- are presented in

Table 1. The trials to criterion for Ss assigned to the non-reinforced

responses equated cor.lition are presented in Table 2. Irspectinn of these

tables suggests there is an inverse relatic.:.-c'ip between perfoi_:-nce in

acquisition and rz7cr:al. It appears that Ss who readily attain criterion

in acquisition are retarded in reversal and, to a lezc:r de:'ec, Ss

who are retardc. in acquisition rpp^:r to perform well in rc.'er.-. To

test th: possil.lity of sucL an inverse relatic,.. .'ip, all Ss were ranked



























j j-

-/ i .


Fig. 1.-Tracings of representative cross sections through the
hiproc~ ':i lesion (After Pellegrino and. C;-:-- -., 1967).
















1 1 .


'N



0~


7
7


J,'


(I j~


Fig. 2.-Tr,-ciPs of representative cross sections through the
neocortical lesion (After Pellejrino and C-usan, 197).








TABLE 1

TRIALS TO CIT'i' '1.0I IN ACQUISITION AID PDSTESA'U FOR SHAM, lNI'OCORTICALLY,
ANID HIPPOCA71AILLY J :1 O 0 Sz !SSI.,.- -. TO THE REItO1.C. )
R --C' -; UJAlTi!) CODITiOI


Acquisition


Sham Co-.trol


Neocortical Lesion










Hippocri rl Lesion











TRIALS TO CzPRIT 100iO -i ACYUISITIC.: AD R -L 5OR SI.I, KC CORIICoLY,
AID 1. ". TY IO ) S ASSIG~ TO T:` -
P-' i. :A2C) P O:3S 1.JATE) CO D>TIOl


Acquisition


Rverzsal


ShamI C-. -.t


Hippocr #p1 Lesion.


Neocortic-al lesicn








from low to high on the nr.,' :.- of sessions to attain criterion in

acquisition, and front high to low on the I.Ln-? of sessions required

to attain criterion in reversal. A Spearrcan rank order correlation

coefficient (rs) was then co.puted (Siegel, 1956), and found to be

significant (rs = .47, t = 4.056?, df = 59, p <.001). Spearman rank

ornf2cx correlation coefficients were also comiputed in the sa.;e r,:.-lrer

for the shza, neocortical, and hippoca: pal lesion groups (see Table 3),

and for the three lesion :..'p; when further divided on the basis of the

reinforul% ve;-', non-reinforced r-..p-r... e.-u-ted dimension (see

Table 4). The former analysis indicates the inverse relatic,.:hip between

poifor:_:nce in acquisition and reversal. is present in only the shY.j

control Ss; the latter analysis r--:.1. that uhile the shan control Ss

unl;.r the rn:.,-reilnorced re: ,::: equated condition do sho! this

relationsh'-,p, their c':. terparts under the reinforced reson.%s equated

condition do not. In addition, the finer grain a.-al.ysis indicates that

the hipp:.-:- ctc.:Iz .1 Ss under;" the reinforced responses equated con-

dition also manifest this relatic.'ship, albeit to a lesser degree.

The sessions required to attain criterion in acquisition and

reversal for sham, neocortically, and hippocampally lesioned Ss under

the reinforced responses equated requirement are presented graphically

in Figures 3 ar.i 4, respectively. Analoges data for Ss under the

non-reinforced re:.ponses equated condition are presented in Figures 5 and

6. An analysis of variance assessing the effects of the three lesion

conditions, the two re-.rponse-reinforccct contingencies, and acquisition

and reversal upon performance as ii..c::i by the trials to criterion

measure was perfor.-c-1 (Winer, 1962). Since the trials to criterion









TU3LE 3

SP?.'7..: R I 0.i CO TiATIOI CO.L iICI.iS (r.) : SIAI,
NEPKCC..iCCL, 1D :IlPPO'CTA L vCI-0 G 3 .


Shmn Control


r = .51

p<.05
rs = .10


Neocortical Lesic:i


p>.05


Eippocac-pal l] sion


rs = .36

p >.05


*See text force i---i:i" pr.o. u2s.








TABLE 4

SPi2J.UA R%':: ORDIl CO.- .L. 'O.i CO-, ,-ICi .. (rn) FOR S LAM, NEOCO.iIC.'L,
AND HIPPOC C'PAL I-:I C:.'":"; U..-. .,'-' : ORGE2),
OR 1 "- : ".. O ".D R .BDINC- DU-N I .D:. ',


Reinforced IRc. ,,..,-s
Equated


Oi-Rji fzorceIl Responses
E uated.


Sham Control



Neocortical Lesion



Hip :,.:- 1 Lesion


*See text for ranging procedure.


p> .05


rs = .06

p> .05

r = .68
S
p<.05


rs = .76

p <.01

r = .29

p> .05

rs = .35


p>.05


rs = .12




















































WV)
* .tvf.. ~ *. -


CI L-
C14 -


N0OIl31IIJD 01 SNOISS3S


-. *.


0
- |










V)
-




0
CLL
O
* -. 0 C



z





0





0
U -3 - mI


-r4

-4

0


C->


















to

4-,













o
0








or-4
..-4
^-4


C=11
I.u-
cO- UCO


& I0
CT, CLsi











1-1





0











Le
- - 0.




- -
< <- -- C- -
^u 0 u .. ... ) '

0 U .-- o





0I0
00'" LC










C-1o
IU -


















V)
a .. ,::a ,
-f-













*C- .-
z -I
NOI11:11D 01 *~0IS9& 1";
-~~~~~ ~ ~ ~ ~~~ ~~~~~~~~ ~ 9.' .. .. -** "**::- ::**..: .*.-*: J ^ 4-
r ~ ~ ~ ~ ~~~~ ~ ~ ~ ~ ~ *L. -..-I- ,: ; ^ ':
^ "^ -- ^ *^ '' ^ ^ -S g

^~~~~ ~ C) s^"i
0 <
N01^11^301 SOISSS T0

a.,


























z_ i







--- 1
0 0 _0

c- <
0 O u
u 0O
0 c
Z IC









C) 10 CD
cj3 ro, C-4 c14


U)


r
P






I.

F

F:


) : C 7


NOI-D3IPD 01 SNOISS-S


CO

U

VO


z
0

I--






-0

L-


r4










C-









4- 0
o















0



o<
t4I



o
0
r-i 0


C-:)




























[iF1F~


, ----

< <-
I LIC-

n 0 U
u 0





0 r





I3 I^ C CCI


F

r


uI cn





. 0



C- -





C <


L 0


C D 0 9
I' ;


*rl

*rd
0





*-4
4-1




t-,






a







4r-
'-3
0



















eS
0














IL)


Iro


NOIri...I.rD 01 SNOISS:S










me2.sura pr ouced- slightly s'-i.-d d. t:a, thse vJlucs 1 %:I-, subjected to a

square-root tr" n-.ora.tion a a second analysis of variance perfor:cd

upon the resultant data. The results of thcse, t:wo analyses are presented.

in su':y fashion, in Tables 5 a.Ld 6, r ctivly. A comparison of the

two tables roce.l little inconsist ..y in the r .'.?ts of th> tVo



An e C.:'..tion of Fi:-. 7, which depicts the c:u.ulative per-

centafe of Ss in each lesion group attainIr: criterion in successive

five session blocks, su-;::ts that the neocorti-.1. c.-.3 hippocarpally

lesior. SS are sli'gttly facilitated ,ith r.e_ :-t to the s'- control

Ss in discVi'_inatico prfo:: Lco. The an lyse' of variance indicate

that this difference is not lare e.c'-; to 1' statistically reliable.

However, the rEsnIlts of the a.alyses do reve l a sitaificant int tion

between this factor and the acquisition and reversal phases of tr:'rir.

which -aust be e...:-.-.-:; b:ore it can be concluded that the various lesion

cor.itions have no effect on discrimination learniri.

A co>rison of thb effects of equating either reinforced or non-

reinfo;'c-l r:or.:zes during discrimination training is depicted in

Figure 8. Neither le-.l of this factor, nor this factor's interaction

with the lesion dimension, w,.. indicated by the analyses of variance as

differentially affecting performance in the discrimination task.

Inspection of Figure 9, which reprLZEnts perform-ance in the

acquisition c.i reversal phases of the study, su. -c:ts that Ss attained

criterion more rapidly in acquisition training than in reversal training,

and this is verified as a si.,'Lficant diffc :'-:.e by the analyses of

variance.







Ti 5

ANALYSIS OF VONi.:CE 01 1:1,.JLS TO CRIi -,0ON



So1 CI-'-: MS df F

Between Ss 59
Lesions (A) 117.11 2 2.23
Response-Reinf,:.: .:t Ccnti: : c; (3) 151.89 1 2.90

A x B 2.79 2 0.05

Ss Within Groups (Error) 52.46 54

Within Ss 60
Acquisition.o-..- -:- T (C) 795.69 1 7.93:'
A x C 417.62 2 4.16*
B x C 1.86 1 0.02
A x B x C 83.21 2 0.83

C x Ss Within Groups ( 0-1rz) 100.35 54

*p<.05
** <.01









TATL7 6

ANALYSIS O0 VMARIIANC3 0 S 0 A5 T TRU S OE 0TION O TIAITS TO CRIT-RTC



Source YS df F

Between Ss 59

Lesions (A) 1.23 2 1.7

Respaonse-Rlifor c:ent Con~tinrscy (T) 1.33 1 1.91

A x B 0.02 2 0.03

Ss With Groups ( ror) 0.72 54

Within SZ 60

Acquisitio n.-.;ra., 1 (C) 10.96 1 6.61-

A x C 5.42 2 3.27-

B x C 2.01 1 1.21

A x B x C 0.04 2 0.02

C x Ss within G--.,: (Zrror) 1.66 54

*p<.05
























F' 0 )

o o 0

^ 0 t

0) 0' U
19 U l r

< ?: *4




1...y~. ~-'- 9 I I C
-. V; C~t. tx-


C2 C~
c;:~ I~)
9'-


Cz C) C- e C. CC D C,3 C^ C
co r- &to IA ;* c13 C" -


N01iJ 1i11 ON iI '1 I V.uV S 31rtsi f

:10 a0VIN3_2iEXd 3ALViflWO


C2


54-









tl
*,-4









d.)



o








L3












cs
1
















*r-
^.-- iii .,

1 'C ;























43

0V



u J L'! L rI


o
\ t + (o 0 ",-,







,outo





Lt C- I
\L' L"
\ \ 1 -d \









+) tr)

) o

4-' V


I.t-

__ I rf0)


cm (- ti C> c' C' C' C) c.) c) C
I | *3
I I 5










I ~ '~'
0 Ar-.I

ri) i
V I
-~


C~)







K


4 -


C-1 C) C
cZ co cC3


C~J ~~Z) ~


IlOii~ flD siv SA--jra!s
:1o0 1ci P ^AIj.V FiVIAD


C-4









L -
C)


4-,
0 0





en
OT
f-r4
1 o
0


















J a









As indicatc.- .;-vl.icrsly, a sijni.c-it inter'ction bat:cc'. the.

lesions f,.tor j-.3 the A.-. .ition an: roeitsal phases of training vas

revealed by the c-. yses of variance. The co::poneuts of the inter-

action are depicted, in Fi-..- 10, hich pre:sets the prforaicc of

the three lesion groups in rquisition trnin, in Fgure 11,

which pre -..its their perfir -"c; in reversal tra-in'jnj. 1: ean triris

to criterion for Ss in each of the thrk~ lesicn groups is presented

for acquisition and r -:..1 in FiJure 12. F:-, r.-tion of those

fi-urcs su.s:ct- th-t the tL.',. lesion grou did not differ in the

acquisition pf:se of tr. .in, lut th-.t in thi revc-.1 pltase the

neocortic .y r ad hip po-t .ly Fic :.i.. S, -ils not clffuerin"e 2--

th-. "--2ves, did attain th criteric rc.. r.-:dil aid in geter narbrs

tl.:'" the sh.e c.:.:.L,.! .- -: co:S. AisC.s b tv:~ a the cell

means involve in this interaction vero I .:oric utilizin- the

Stulentizcd range statistic (,Uinr, 19(63) The results of the compari-

sons are pr. .cnted, in s u::,'ry fashion, in Table 7. The results s. -ct

the above observations and reveal, in addition, S'.,.t shan cont..l Ss

attained criterion significantly faster in acquisition th,-. in re:o.r:",

but that such a differ::!. is not prcs:,it in the neocortically and

hippoc: L11.y leio,.: Ss. Neither the r- :'L.,; first order inter-

action (lesions by response-reinforcement c.:tingency) nor the single

second order interaction (lesions by r.:.:'n.-reinforce-crt contin;.- y

by acquisition-reversal) attained significance.















































































t.> c co ca ca c~ ci c0 c. ca
C Q1 C C. C r- CL) LC3 C CN d (




ro tID> aAIJ)/irCMnJ


C-)
c





















ot
41











0





O o









U
W) m

+-0

V-4
o c






o r-#




4-)l









-W-4
)




.0












Ir--






0

(C)






































































C3 Cr C C) ct C)
C- c 4- u? c" U-


O

0












O
, I



0.
cI.
0~.
c*A.


L)I

U

0



V)


ti
::- L'J


L'J

LI


4-i3



CO
0









.-4
Cd

C-)
4-i














-21
Cj










.C


5 t-4
0
--I
C!i























( )


t-
*'-





U


wB-


C*-, cr> C73 l=
c, 11 C'j -


NO Il'l 9NSliV .V .!DEffnS

0 B ")"J. i'D 3Ai)y'"i"0 D














LII


7!


[i ~ -- --* -





L3 ccI
.,+,-j C+ I," "
*! CSI +"
N+Ol I iyH d. O.i SNO'S I NV'+++ + .+++ l',':+W+\:+,


N
,-1
* P
dfl pi
'-4 0

0
4)
r-j











STUJ -'TIZD ... S iSTIC A P03> O TI3S'


. tV..., .. . .


Shra vs. oecortic;.l Lesion

Sham vs. Hip c....plJ Lesion

Neocortical vs. Hippoc pal Ls.ion


Reverse .

Shi vs. Neoco rticc,. Lesion

Shra vs. ," ^c 1 Lesion



ShzR Cotrol

Acquisition vs. ?. -.1i


Neocortic:r. Lesicn

Acquisition vs. Revcrsal1

1* ;0


Acquisition vs. Reversal


3.09

0.86

0.83


20.29

13.'.

0.23


19.77r-


0.01


2.30










DISCUSSION


This study exauincd the p-irforiance of sham, ncocortically, and

hipp_:.c ll.y lesionea rats in t';.- acquisition an.1 reversal of a two

.manilj.:-. differ:-.tia.tion as affect,:... by certain L -,.,itic:. of

response-rrirnforc .ient contingencies. 7:: one-half the shaa, neo-

cortically, art. hip:.:.-. 1 :ly lesic:--. Ss, these manipulations insured.

that each S emitted n a jal l.;.-: of reinforced responses on the-

two .:ij.:i:.i duri--,: a complete expert. ..:.1 session. As a result

of this proc-:.:. e,ich S also c; itt.i t '- l: as .i, y non-reinforce

responses on one :-.:.li.u!,a:.: (th3 >: 9 r" ,.:lan) as on the second

(the FPR 5 r :u:.iladulm). For the re:-.inirj s, the experiknntal

m_;nipul.'.ti; insured that each S emitted an equal L.L:.:'r of non-

reinforced responses on. the t:wo r.:.:21.-L:. during a ceo.:lete experi-

ment..l session. As a r.-.it of this proc;l.,r, each S emitted twice as

many rei forcedc ri- .p.css on the FR 5 : 1-.:"..-'.: as on the FR 9

manipu~l:.."

The Doula.s-Prif:-i> theory of hippocampal fun.ction (Dc- ils &

Pribr-.n, 1966; D.ugl.s, 1967) is a vigorous attempt to integrate the

wide variety of beh-.vioral chng:s following hippoc:.-'..1 disruption,

and leads to clear-cut predictions of the bE-.vioral effects of the

manipulations performed in this stuiy. Specifically, the theory

predicts retarded acquisition and reversal in hippoc.:in-,l Ss when

compared to shai. and r-:cortical control Ss under the tre:.tment condition

which s pcifies e.'-ti.on of the ni-.i'ber of reinforced responses emitted

47









on th3 tw'o 01- 1. le n or no ro -,tic % in 2l -o. Ss

under th3 condition which s. -,. to equate the n'r2ar of no,-r1inforccd

responses emitted on the discrininanda, and, indirectly, no difference

bt'-:ee shan ard neocor'tical Ss uHiLrx either of those t;o. treat..nt

conditions. These predictions are contradicted by the re-suts of this

exnpsr-L .nt.

Tn:' trials to criterion Ld.t2:, do not reC7Tl ony differencEs

betweca the three lesion groups in acquisition perfor-. :e. This

finding is in g-ne. I1 accord with the r. .-ts of recent studies of the

role of the hippocampus in discriMin.tejon IoPai n. However, such

studies h7ve not e- ,.i:i ts :i 'fcc' c diffe.ti3 de1 cities of

reinforced and non-r iocel res3'.in to tVr clic e: tite disic i -

inanr.a upon discri.iJat.tioCi lsoini in hipca cto.i2ed Ss. The

r:.it fi''.., alt.' i-i inconsistcit with predicti.. deri..

the Doumlas-?Prith:: :'\el, i-licate that such differences have no sig-

nificant effect on acquisition perform: u'ce in either shaen, neocort'. :17,

and hippoca:pally lesioned Ss.

Thi performance of the three lesion groups in reversal also is

inconsistent with predictions derived froa the Douglas-Pribrai for,

as in acquisition, the two response-reinforcement contin,:ncies do not

differentially affect in either sham, neocortically, or hipt: :.:.W11Y

lesioned Ss. In addition, the reversal data appear to be inconsistent

with the bull- of the data on the perfor-.-ce of hippocampectomized ,.1

control Ss in discrimination reversal; namely the hitc.-.c.-lly lesione"

Ss are typically reported as retarded in discrimination reversal

when c..-. .lred to neocortically lesi(;..2 and shc: i control Ss who -: -.lly







do not differ eacich other. E :i results of tho present study indicate

that it is nec-:-ortically ae.: hip, .- .117 lesioned Ss who do not

differ fi.:. ech other, anl both : :-- facilitated when compared to

shm: operated Ss on the trials required to attain criterion in rv'::.-1.

A comparison cf the se-c.i-: to criterion data with alternative psr-

forance ness, su-ch as trials to successive criteria and the ab-

solute per ceot of :-' 5 reszi:: .-r for successive days, revealed that

all three c.. -: dpictc2' acquisition and r -:rsal performance in a

similar fash ica.

A p .:1. Mle epleanation of this da:-;:-ity is offered by the

relationship aciiticn aid re.--1 :.-Corc as revealed

by the Spear. r.. ': order correlation coef icicnt, i:- s. c opIrat t

Ss, w'-o r ,.i a signific:;tly .'-ter niI or of trials to attain

criterion in i T~r 2L. when c *-;. to the rc.cortically and hippo-

campally lesicr~d Ss as ir. ic?.ted by the St.'i-::tized r._:-s statistic,

are also the 3 who rianifest a s..iljicant inverse relatlo:..hip bet-ion

trials to cri'.. ,ion in acn,:-:ition and reversal. In addition, a greater

number of Ss in the sh:u.:i control pioui:, attained the criterion in 11 or

less sessions (15 of 20 Ss) than in either the neocortical (11 of 20

Ss) or hippoc: ;-.1 (12 of 2,' Ss) lesion 'c-r S. When the three lesion

groups are p-' L-tio:,'! in t-:..z of the response-reinforcement con-

tingencies, similar phcno-:ra are observed. These findings sug est

that scene Ss possess an initial position preference which, when in

accord with thec requirements of the initial differentiation, facilitates

acquisition ard retards reer:.'-. Moreover, it appears that a grcater

proportion of Ss in the shabi operated group fall into this category









tL-'. in ei'. "- of the t.:o cther cle ion goup. It : be e s;e:i, then,

that the poorer p or.nce of th sh:: control Ss in reversal is an arti-

fact resultirj fro.: a failv.: to co.- ...tely control for initial position

preference, anal that if this h-a b,-n done the differ.-ices bet'.ee the

three lesion groups in rcv.ersal ;:ul: be eli'irnatcl.

The expr 'i ental procolures e.ployc*. in this stv-U differ in a

nmb..ber of details fir:.: those i ' hippoo al lesio-,inTucac. dFeficits

in discri-1. 17 C:.. reversal ?are con obserc: It is possible that the

lack of a lesieo-i;iucc: deficit in the prcs-et st"' le r attribt'.ble

to one or .ore of these cdiffer.c.ces. One such uod.ific otion involves the

utiliza.ticn of an 1C rtri 3 (rain oscc,'in7 fiollvi ttr i:.rt o, Of ritesrio!

in acquisitic. I nvestiga tions of the ef c, _L i u on revr l. in

the T- zcz ivdic- to that Ss subjected to, on the aere "A at C s';t 1.3

(Macintosh, 1962) and 3 (Publos, 1956) ti aes as -- ovevtrainin trials

as Vwer rcquizrei to reach criterion in the acruisiti:. of a discri' nation

perform~ better in reversal than, Ss without oei ,... (the v.:. ti.

reversal effect). An ezanination of Reid's (1953) data indicates that when

overtraining involve: less trials than were required to attain criteria., the

overtraininS revel-.1 effect is not seen. It is difficult to c':;:.- re those

procedures and the present one, for different responses and p.oc':dr.s were

employed. However, sir,.- this study employed only one-half as .mny over-

training sessions as were required to attain the acquisition criterion it is

unlikely that the overtraining Icv:.. 1 effect was operative. L:-_pite this,

an examination of the o;crtrair:ing reversal ph:-.::-on does add to an under-

standirj of the results of this study.

Macintosh (1965) r,'Les that it is prL. w-o L.j justifiable to

regard reversal learning as consistir: of two parts: (a) extinction







of a tendency to select the f.oir r SD, ad (b) r:--.ition of a ten-acy

to select the Ls'- SD. Stage (a), extinction, is usually reCard.ed as

contln.'l. so long as S scores below' chance level, and stage (b),

acquisition, is typically depicted as co uencing as soon as S begins

to perform .,:.- chance level. It should be noted that implicit in

the formnilaticx- described by Nacintosh is the generally accepted

assumptiontha.t learning is a continuous, rather than discontJh.1:-us,

process. ."n-: ther this is indeed the case has not yet been completely

resolved. T-: :. .t discussion is c-:,-]rned primarily with the

effect of att:...tional f:ct'L upon discrimination reversal rather than

with the ude.l .in- nature of the learning p:'ocess.

It is oft$n repo: that' ortrainij of a run.:&:.y response

rejlts in reaa.c, resistance to extinction (>.-:- r, 1963). It is

logical to -:. .:, ther.-:ore, that this p-:.: -.. is what underlies

the oci. Li. ni r.::'.1 o effect; namely, overtraining facilitate
extinction of ::-.,-.:s to th3 for-.-r SD in the formulation descriL:i

by Macintosh. This is not the case, h..cver, for overtrain :- in the

discrimination 1aradigm regularly increases resistance to e::tinction

of respoi,.es to the fo:.er SD (Macintosh, 1962). As Macintosh (1965)

points out, overtraining facilitates reversal of a simultaneous dis-

crimination not ,2c,.u:e of, but in spite of its effect on extinction.

This finding would appear to negate an extension of the frustration

(Lawrence & Festinger, 1962) and generalization decrement (Kimble, 1961)

explanations of extinction to the overt'i-.1nij-g reversal effect and,

by implication, to reversal training in general.









IIacintosh (1965) reports that c: isCeLit evJ cea indicates

ov;c tli-1 facilitates reversal by shortening, runs of i:.-; ct

responses durirG the middle of the reversal. This su -ts that ovcr-

training 1i c.s Ss' tendencies to respond to irrelevant cues c.. '-

reversal. There are to possible explanations for this: Either over-

trainin3 effectively c -,cs Ss to "adapt out" cues along irrelevant

dirensiois; or overtraining allows a.ple opportu-ity for Ss to learn

to attend to the releva-nt cue dimension. IcI: tosh presents evidence

which indicates that it is the latter alternative whichh uaierlie the

overtrai.n :.. reversal effect, -- I points out a distinction bstvecn

research utilizir, visual asnd spatial cues. Stve.iss whicl have in-

volved a si ulta-,, us Iviiel di C xincinuation (bicht .es;, sjttern, etc.)

invariiably produce the o rtraininr revc-sal effect, .. 1e those

studies y:Kch c;-l.oy a spatial discr1:.i..tion (left turn versus irght

turn in ?.and Y-:azes, etc.) f~ -.F ently do not. A major reason for

this, Macintosh contends, is that the rat (the cou.only used e:pr'i-

mental organism) is primarily spatially oriented and, as a c-:.cLuence,

spatial cues :-.-:, a high priority ver, without ovrtrlirnfr,;. Since the

rat is already attending mainly to spatial or position cues, over-

training would not be expected to havc much effect on performance in

reverzl. Conversely, the lower the relevant cue dl?-.c.o' is on the

Ss attendingg hierarchy," the more valuable overtraining .:.'ld be ex-

pected to be in fji-ly establishing the relevant cues in a position of

dominance. The L-.nitude of the ovrtraining rever.::! effect should be

inversely related to the probability that S vill attend to the relevant

cue at the begin;i.- of discriirI.tion training, ern to the nr:..l.r of

irrelevant cues involved in the discri::r,:-.tiotn.







rI..- p:z,..- di uZion is carkedly similar to the r., -

Prlbi-ar c...-.: tualization -:: the role of the c .: al. in discrimination

luAnlr. ; n e'ly, the registration of the effects of reinforceennt or,

altr,.lively ti:- directi. of attention to the ,'..'. of the ts.Y.

(relevant cues) associated ith reinforce;:eit. HFoever, the above

formulation of the functic' of attentic: -.' factors in discrimination

learning does .0a t inco: :. '. a pI;. cc: lanalco to t,':..t attributed to

the hip' ::.: ; the gatinf ..L of sti-..ai associated with non-reinforce-

rent. It will be recalled that -.) alternative to the atteiitional

r.':icl ir. Icate. N.t the c,- 1.Lr .:1 ij rcversal effect caIm be attributti

to the op:ort~-yity for Ss to effectively 'a apt out" irrelevant cus

(Sp!;.cc., dc3cc;lo2 in Mac -',sh, 1965). Perhaps an explalatioi of the

ov-tr3i','l-; ::-.'-;1 effect involves both those processes. The

Douglas-Pritra-.. model su:. t,-s that this is so. In addition, the

results of this study Lay be as iiconsistsent with the Doublas-

Pribra r.. :1. as was first -.:i.:ted. The differentiation required in

the present sr .7 involves s.-ti-.l cues, a di:-.:ion thought to be high

on the rat's "..tsntic al hierarchy." If this is the case, the role of

the hi.poc : that of g.-nin out irrelearnt stimuli, would be

nir.al in the intact S, and Ss without the hippocampus should not be

greatly impairel. Perl.->o: a hi;.poc.-':.1 lesion-ir'duc'cd deficit would

be evident in the prc-s:-t r ..r-.]ig if a task involving a non-spatial

differentiLatio,_, or, more probably, a differentiation between a large

number of equ-...y saliojit cues was employe... Support for this possi-

bility is pr,'-,-: .-d by Pril:- _-_ (1969), Uho reports that bi.c-:..ctonized

monkeys show r.tar-L-tic) in discrimination 1 rning, provided there are









a .; nrm-.b.r of non-r' e.:a:3 I to:ZtiT:2s in the sit ti.on

(pS. 137).

A second procedural innovation e: loycd in the px- -.t stury

ir,.'lv,, s the rcspoose selectca foe- st.dy. Those studied, reported

prev-iouly, which h1 do: str ated the hippocnrnal lec.n re;vorr:1

deficit h-e typically e:ployc& an inst.rient-J. respo4s rcqu.irinC

sonc foriL of gross loccl otion on the part of S. In con'. -t, this

study utiliz7ct an 0oeiit response, a be, press, vliich, v.!I.: the

typical inst.r t-...I:a resp:.nes rqiies a ai.ii-i.: of locoCotion, ta':es

a short ti:sm to execute, requires relatively little effort, and leaves

S in th. s ... p>.ce rsy to resp:id a2ain. Althoal h it is go eral.ly



isU u'icr,-l-: t i appear to ta a.:loO cs ta;s in th tilo 1per -

metal approa-ches Cdo not differ in rany critical aspect, a thoro_ -A

cc_; :.isoin of these t'.o pr- -curs hs not yet be"n attempted. Hoveier,

a recent study by Means, "Ker, and Is ..,:: (1969) i: ic .tes th"at the

effect of hippocapal d.is,.'_.ion upon go-no go perform :.':e mray be

resp.:.--.-specific. Althoieui it is typically reported that hI, c, ca:pecto:.:y

interferes with this behavior when cxaZined in an instic..:;tal par...' .:A],

such as an allei. '.y (.o "., Kaufcan & KMrco, 1969), Means et al. rcep:rt

that hippoca:.,pal ablations facilitate performance in this task when a

bar press rc. :-,.I is utilized. Fin'lTir-3 such as these qu-.ti .,', the

trans-situational Lr:ture of the pattern of lbt';-ioral disruption ob-

serv...l following hippoce,-pal destruction and, consequently, an1' formu-

lation which attempts to :co.:';. for these effects with global concepts







su :h as 0'1-:.T ration," _.'.ir _," or "inhibition" without further

re D.--o it oir Qalificatio.

The third r'ajor di fe1rence tet,~ccn this and contemporary in-

.'.tiatic. s Of the role c_- the hippocarupis in discri. *- Ion learning

inyoli c-' the sche'lecs of reinflc::. nt associated with the two to-bCc-

discrimiiiated responses. -.:- reszc:. i -r. .d u iously has typically

p:-1 .ir .CI cont. .i c. rei' .. nt for "co -c'" r:'-.:.30es and withheld

rcl:.2.orceent for "incorrZct" r ..:::: ... In the pr...-, st .';, con-

C,..:cnt o;i :-:a: !s were utilized: Both respc:::- iere reinforced, one

on Ln FR 5 scLm'.le and t'. other on an l: 9 -s 'le, r':'. the fora'-

atiori of ths c'-cri- 'i.ticr- Was 'zbsed on a relat-.:, r.-' :: t n

al:-SDlute, dif fotial in rein for1e: 2t dJ.:ity. Tf;: e 7 j-ri it1

, -,vsis of concurrent ratio sch:,.iles inicates that with unequal FR

reqir-';.-ts, espoiing t: .' to be r-aintained only by the schedule

wizh the scal :FR requiT .. t; with equal F. requirements, responding

carn be naintai_-cJd by either one, and shifting fro c:,.: schedule to

the second oc.-.lonally c::i"v. (C-t'.1-., 1966, E-.2:rnstein, 1958). In

a study which i" only sup finallyly co: _.r.ble to the o.:- re cted

he7o, Douglas C.-. Prilr.. (1966) c-.::.in:.. the effects of probabilistic

reinforcc-r:ct -z:.r, the foi ::,tion of a disci ,.inatc.t pr-:.- press in

moaLeys. As 12 the present study the d'.crinination rested upon a

relative diff.-rer.tial in r:'iLfo'cc.1c:it dcisity; one response was

reinforced 70;' of the tine and the scc-;3 reinforced 30;3 of the time.

T'r ir results, in contrast to those of the present study, irUicat-ed

th:-t hip'co:::.-toized Ss "re retarded with respect to control Ss









in their ability to acquire a discrj.t.'. oi .ic- suc c itions.

fortui-r'tely, no data Ce snted oi the pcfor.- :.ce of thsc, Ss in

di inination revere. .. iL reasons for these eaprrently con"'. "ctoy

finding: are unno:7 ut thse0 st' dies ,. I that J. J Ciciet

attention: has bzan directed toia.' s C.T declaration of the eficcts of

sc''.lcS of reinfocrc-:ct on discrll.iration fori:.ltion ad.! reveirsl

in hippocp cctoriizd Ss.

Another difference b ,&en this a-I other studies of hippoc.::~ 1

functici involves the spc.ir.s of test trials. Most r.-c .rch in this

area !e.s employed a discrete trle proc: r & td ta-v related te ir Cue

of massed train trils d each daily e: irrcl s ion. In

the present study the e .ai. .nt of tc:t trials, the five '-nute free-

choice periods, wore ." ':ly spaced for they r. I. at the bseginir2

of each daily ex-peri ental se.ssion. No direct evidence is c :-,lable

conic-..i:- the effect of this factor on discriL.in:ation learning and

reverl.- in hippocampectoaized Ss. HO-. ., there is evidence that the

interval bete,:en trials does inf]i',: ".:- the bihavicral effects of hippo-

camipal disruption. As reported previously, Kirkby, Stein, Kimble and

Kimble (1967) have demonstrated that the lac- of T.-maze spontaneous

alternation co:'ionly reported in hippocan.pal Ss can be reestablished by

lengthening the intertrial interval froa 50 .si.. -os to 10 minutes.

Although their explanation of this ph'no-::.ion, a postulated lesion-

induced reduced infor-mtion ac uisition rate, has been generally
abandr,:.-i on the praise that :c.ch preparations do not show deficits in

a number of alter'.-tive learning tasks, no adcc.ate .pl:. -tion has l. .n







fo :.ated. Little or no *..tional research has i,: ..directed to':2r-3

an Tnderstanrig of this finding, and. until this phenomenon is investi-

:.'.:0 in i c :.., detail such an explanation of the results of the

pr..eent e.1,-:- :,t cjinot be fully evaluated.

A fifth majorr departure of this e ,:' :.. relative to previous

rec:: .rch is tnso utili :tion of a forced training technique. As a

function of fulfil.:I the requirements of the rcsponse-.reinforceent

contingentcies, this procclre ir.3.'edi that e..c' S was fully exposed to

the co:,.'.itions of reinforceent throughout both acquisition tan

re...-sal training. In addition it can be ass~nd thL.t this innovation

mo s prolc.bly raintained the strength of the FIZ 9 response at a

re. -.tively his,or level th..n discdieinnation studies which hwve not

e:.l-ed forceL tr-'. ong s.d reinfo7rcGent c. the "Incorrect"

reE--n-. Isa.:c.on, Olton, azuer and Svart (1966) present evi: :,.:

which indicate-s that t'rf;e hippocc,:i.pally lesioned S's inability to

with.l! a ret -.e in the passive; avoidance ta.': is directly related

to the strcnpt' of that response. It is also possible that the ease

with vhich hi( .:c:.:pctcmized Ss can inhibit one ic,,se and initiate

an alternative is dependent upon the relative strength, or probability,

of th.se two responses.

These findi~-3, which dc-onstrate that hlppocr-:p:l Ss are

capable of inhMCiting an e.t-,ablihed response and initiatir'3 another,

st&nd in ra--.'. contrast to the talkl of the data on the performance of

such Ss vihen f.- :- with similar tasks. It is not surprising that task

vari..-.bles, so--- of which have been discussed c,.cv, have the potential

to pr-ofoundly i-flue-ce the behavioral effect of physiological Lniru-

lations. What is surprising is that no concertcd effort has been made









to explain s-uc fir.ling: 't.thi:i the c7 f. ts cf pra "c.t fo 'L:latioC:s

of hippoca .pl fi icticn, or to ie-vise thL:se forulations so that thly

may incorporate t hse .. Its. M11 too often fiUlins as have

been disc,,l. hEsrc are neglected or dis:issec Ce aberrent. Pci-.aps a

detailed e:a iination of the n Tner in which exs? '.2-.l r 'nipulations

can chasge or counteract the effects of physiolcical c.rniil-ations '.Jill

provide increcad insight into th role of neuroTphysioloic.l1 systr -'s

in the intact or.u nis:.

The unexpected facilitation of discr:' 1- i,-: reversal per-

formance resulting froi tlhe necorticl, da'"-u: sustained by the

control Ss is most likely a.'.:.1t?.. le to m.control!ed position

prefe...es, v, ciscussi previously. O'thr se ri ent Stioi on t

effects of hip: c .l ablation has t:, icilly nvolv/d aalogous nec-

cortically lesior.u. control Ss, and has re~..I- rly reported that such

Ss do not differ fr1o their uno-.rate c-:. terparts. There are ex-

ceptic.:,:. to this ho::..cr, for Mearns, ot (1069) ha;e found that

destruction of the noocortex overiyl g the hippoc.. _,. leads to a

retardation of performance in the go no-go t:.:; and Olt':. c: Isaacson

(1967) have reported that d-.: -:e of this area, as well as this area

plus the hippocampus, lengthens response late!cies in avoid:-rnc a-:d

escape tasks.

In the present experiment neocortical e.dttiu;tion ir:,'lOed

considerable portio',s of the rat r..:',2ortex cc-L.;-.tr.ble to .170:" i's a c:'

7, which is involved in so.:..-,thesis, particularly the integration of

information on weight ar.n th- state of muscles vnd joints; areas 17








and 18, the vi -_1 projection and association -..-;-,, respectively;

area 25, the entor1:,-_ cortex; a:d area 37, which receives sz.; ;esthetic

&ud optic as-.:'-t.o-n fibers andq in ana, is thouiht to be involved in

the rcco n:ition c- body i 9', indivi-' ty continuity of

per:----lity a1nd of the self in relation to the E .vironment (Kreig,

1957). Since a considerable portion of the hi .:.-. pal research has
involved so.~a (dOe of these arCes it is possible that co-o.only ob-

served hippo( ..:.! lesion deficits ePL0 in actuality a function of an

intr action of the hippoc"npus and the ne.ccn. 't.ex ,lhich overlies it.

Within this context, Ir-.,: .- (1967) has obsrvd that electrolytic

lesions i ::.tricted to the hi-poc- c friaently do _..t produce the

deficits seen in ablttion studies involve 2 neocorticAl destructirc .

It is also possible that thV facilit'nted pertorLance shoWn by the

hippoc,. .:ctoeizcd Ss in the present sti. is fully accounted for by

the effects of .::o0tical dcstLrction. Questio.!vi such as these point

to the relative prinitiveness of our U..i': t.:Ti,., of the role of the

hippoc-:iPu--; in behavior, and to the importance of further research in

this area.















Aoy, W. R. St....es cf hippoc:: el.ctricl activity di.i:. appr:.;ch
,learni_. In J. 1'. DJl af .: ( :1.), ."' i r K :': e s' in
3 Carl:10.-. Ozfoi :e1, 1961, p./--*- -

Allen, ',. ffect or ablatin[ the f:' :tal lo s, hippoci:pi ai
occipit rieto -2 (exc tin p ior;: a s) lob
on polite ,ti -e olfactory co.inti is reflexes.

Allen, FW. ect of ;.bti:- the pirofor>::. :'aloid are:s and
hippoc pi on positive "."l nc'ative olf'octor-y diff rcnt,tiation.
Alerricc Jo l of. M s1!iov 191.1 132, 81-92.

Boit_'o, J. J., & I: .co R. L. Effects of f 7.iation in shoc:-.
int.:.st. 01 thv >-'.0 of do--al-o 0c,. ctcn-zed rz.ts
in t-o -.3t gi'; c .o : -c sitcuatiO:.9. 2. eric T JC t,:.] of?
Phjyc c 196 8, 3, -

;.::,, T. S., ; P. G., '.?co, L. A. Tihe hiWoru s and
respond e p:s... ..tic.. ...,i in le cat .- Ess. ch, 1969,
12, 86-98.

Cajal, S. R. Y. Strie- on thO ce: ,r:-'l tort?. Vol. 11. 1-lIbic
stIII:t ( ira. by 1 1. ra) ... C hicao: Year 1or:,
19.5.

Cataia,; C. A. Co:.c:-.'..t o:nt. In 'J. K. Honig, (21.), O rant
b c c:, .:o; .re.of -_ C. i.. nolic.tion.. New YAo'k:
Apple : Centl"-. C;.%. 17966, pp. 213-Z2O.

Clark, C. V. H., 8 Isan.cson, R. L. Effect of bilateral hippoca :pal
ablation on P.-. perforznance. Journal of Cc:urative and
Physioloica'l Psychololr, 19697 9, ~137--'70.

Correll, R. E. i:- effect of bilateral hippoc:. ..-1 stimulation on the
acquisition ard exti..tion of an in t.....:tal rezponsc. Journal
of C -- ive n Pysi l. P:. : 1957, 0., 627-17-6

Dewson, J. H., III, Nobel, K. \I. & Pribram, K. H. Corticoft-.
influcncee of cochlear nucleus of the cat: Sorie effects of
ablation of ir.:.."r-te:.;o-al cortex. I .in Research, 1966,
2, 151-159.








Dorjas, R. J. T; hip -;)` u.ous and behavior. P: eocical hu1le1it,
1967, 67, 4 .':'. : ..

Dou.gls, R. J., & Prib:.:- K. H. Leacrel' and linbic lesions.
Neuronejsch1olo:a, 1966, 4, 197-C."!.

Ellen, P., & llson, A. S. Pcr:;- .- .tion in the rat following hippo.
campal lesions. -imJ^t2 ll' m-:mole 1963, 8, 310-31?.

Feldman, S. .. *ophysiol" ical :..... s ;. 1. ; -:l1ing afferent
hypot'. :1- o-hippocys pal con:uction. Exerimenttal Neurolo,7y,
1962, 5, 269-29i.

Gloor, P. Iyrgala. In J. Field (Ed.) F--:.': oJ. o -',. Vol. 2.
Neuro o.olc'-. '.ashingtoIn, D. C.: r1ic... ,-ological
So-iety, 1Y60,' t 1395-1420.

Grastyan, E. & Far-os, G. The influence of hippoc:-.ipal lesions on
simple delon, in st* ...tal con.ditionc.a ref] :; In
P. Poso'nmnt (,d.), P.ysio..i.'. e l'hic..:;, 1962,
pp. 225-239.

Green, J. D. 1:.? hippoc" pus. P,?S IOe PJP1Ci, 1964, ', 561-6C3.

Grecn, J. D. & ..:'y, ;. R. Electro:'-sioJ.r cal studies of hippoc...pal
connections ar.d excitaJbility. FT: tronce:r' 'ic-:-! .
C ].i n :, .- 1 !.... : .. ..' .-,, ." "

GrIeU, J. D., & Arduini, A. Hip-, ..I electrical activity in arousal.
Jolryi.l of r.:.i ei 19'.," 17, 533-557.

Herrnstein, R. J. Soze factors influencing :'r- '-:or in a tuo-r.'-.pr...
situation. Ti:r: -rtions of the Ue;o York Ar-0.'t y of Sciences,
1958, 21, 35-45.
Hirano, T. Effects of hippi:- .i:.i electrical sti;.:Ytion on memory
consoliOc2.tion. ,vc]'.-i-}, 1966, 2, 63-75.

Hostetter, C-,/le, & Thoz-s,q G. J. E-;:..l--.tion of enhanced thi:..otaxis as
a condition of impaired mraze learning by rats with hippocarmpal
lesions. Jci-.:1 of Co --.tj- -:.1 P.;-: ic'c-ic-' Fzvc\elo-,
1967, L6, s105-1T.

Isaacson, R. L., Douglas, R. J. & Moore, K. Y. The effect of radical
hippocarmpal ablation on acquisition of avoidance response.
Joori-l of Compa-;:A.ive aT Physiolo-ical PsvcholoSy, 1961, i4,
-625-6K -

Iso. :'coi, R. L., Olton, D. S., F-.iT-r, B., & Swart, P. The effect of
the nmiber of tu '. g trials on the d-lIcit in passive
avoidance behavior in the hippocaspectonized rat. Paper read
at Midwestern Psychological Association, Chicago, 1966.









I^.cson, R. L., : Vfic'el c, T. 0. Hicpoc:- I aMlation a-d r. iv
avoid .lcc. Scl 19/2, 13, 1100-1106.

Jcr. :;., I. E o havi"r c hiupo : ?, lesionri rats in hoe ca __d
novel si..tic, s. ol "'. ,io^ 1968, .?0.

Ja~ rd, L. E, & I .c.on, R. L. Hi-.c 1 Jl ttoa it. i.ts: ffe cts
of inter-L La into.._ i11: ;, 1i6, 2L .G -110.

J ard, L. E., Is'mcso:, R. I., & *icl ,.<, C. 0. ....: of hippO
ca[pal ablati on ari intertrial intero.;- on rui. acquisition
and extinction. Jo_: t.. of C' .i t t h d ? 1-o0 i: .

Ka B. ., R n, E. II., &K T. 0. K ects o- hif "
lesions on Cnze 1 Icran: and roteteic in r.. S. F-'Iult2.
N ~rl -, 1961, 2, 333-355.
Keyplan, J. T. Inhlbitori r .tic s in rats vith bi)toel hi oc :- .
lesions Prc r ocf t'> 7?J C:.tno c: th'


Ki ble, D. P. Th effects of bio t: hi. :. .1. l a in rT ts.
Journ .1 of Co.--mw.Ue L i1-. ^-'i; Pc:1: 'o 7, 1963, f6,

Kinle, D. P., KiMble, R. J. Hippc::-ctc7 and res onse pecrsever-
atlon in tho rat. Jo, :I -' o C' -.tivea 2 0 i'i'OlccAl
Pjy, :k-i, 1965, 2 7-., "

Kir-ble, D. P., & Z-:':, S. 1Ofactc discriir.. i ': in rats with
hippoc.-l1l lesions. Psic'on:::1 Scie::cc, 1967, 8, 211-212.
Kir '.le, D. P., Kirki:y, R. J., & Stein, D. G. ?:.:.:. e '. ::vcration
interpretation of passive avoidance deficits in hippnc: .:ctonized
rats. Jc." : of Co'a-ative and. ;F .ol'ical. P.ichlo:1 1966,
61, 141-143.

Kirble, G. A. Hil ard ad '' li. conditionrijn d' le.nrf M:.. New Yor':
Appletoi -Cent j-C.:-fts, I-ro, 96l .

KiLura, D. Effects of selective hip.-. : pal dz Se on avoidance be-
havior in the rat. C' -2 anu Jo' :I of Ps0c'o1o r, 1958, 12,
213-218.

Kir kby, R. J., Stein, D. G., Kinble, Recva J., & Kirble, D. P. Effects
of hi..;-:1c r:- lesions E-i C.l -,tion of scnsL-y ir.;:.:t on
spontaneous alternation. Jour?.-l of CAmo'ative .- Physoio0_ica1
.L.... 1967, 64, 392-3 5.







Kluver, H., & :-.-, P. C. F,:l. :- .P- a I;::-" of I.:: t,,.., of the
ter~ oral lobes in __o_::es. :rc?'ive c: ,- ,^ ...' T -
chiatry 1939, 42, 979-1000.


Krieg, W. J. S.
Pa, -.


Iii~< ~ri~ 's in Qi-~chr@ >~.
PIinLi; -~ --


Kviei, 0., Sctclie", J., :9Kn.1, B. R. Differential effects of
'.:.: ..1 lesions on raze an'l passive a7oid.SnTce learning in
rats,. _~_.;.^ t i..irl.? 1964' 59-72.

D1 y *Fce.st, Di n. & esti erLD terzent an d re inforc: S ant .


Leaton, R. N. Exploratory "bhaavior in ra'Lc with hi -al
Jour c: C '. I .- - :.1 P:
3Z53~33C.


lesions.
1965, 19,


Macintosh, N. J. Selecctive atte-tion in ;ii:21 discrininction le
Psycho!oical <:-! ., 1965, _:t, 12-150.

.acintoshs N. J. The effects of o .' .' on a r'eversl an. a
ro::.-; : .: .l shift. S Jo r oft C - :'_'. : ' ., .
PF s .:C 19 6 2 -6 .. '
-.1.. "_..-_-. ''9 / ... J J O /


acl.;-'nP, P. D. The lirbic
Stu'.Ites in animals
Jo.r:.- of :: _. .


5: -:-. and its hi-. :c .?.1 formation:
ar. their possible t sicztion to :nan.
- .***., 195,, 1., 29-Vt.


:-'.c:-., P. D. Th3: linbic ,ote-: ('visceral brain') c-., emotional
b :i-:ior. Archives of 5.olp..-il Ps,,chi,.tj, 1955, 7L,
130-134.

.acLe-r,, P. D. Chemical ar-i electrical stinulation of the hippocanpus
in una,':.tr in -1 aninrals. IAerica.n Medical Association Archives
of U!1. ::-'. a :'-'ch7-it, 195?7, z, 1i3-i_-.

MacLe_-n, P. D. Contraltir-2 functions of limbic and neocortical systems
of the brain and their relev:n., :e to psychoDhysioloical aspects
of ::c.icine. A.erc- Journral of IMedicine, 1958, 25, 611-626.

McNew, J., & TR:cpon, R. eRole of the limbic system in active and
passive avoidance conditioning in the rat. Journal of Comrarative
and PI"-io_,--oc: F :-cho-ior, 1966 61, 173-i-8.

Means, L. W., -I-lkEr, D. W. & Isaacson, R. L. Facilitated single
alternation go-no go perforrc.nce follovivni hrip- c.-:p2ctcoy in
rat. F:.per present?. at the meeting of the P-ch:.;:ic Society,
St. I'..:, Nove-,ber, 1969.









Melscnr, U. EipCir: 1 f :Ccti in lai A. Jc, v.-j) P&-C '9'',r

Niki, H. The effects of hip-oc-.' ablation on th: be. lvol' in the
rat. J^ cPE: 1J .1hc% c, 1962, 139-153.
i1, H. The effects of hipc 21 aM.,tlon on the i IItory
control of cp:rznt b -'i in theo rat. J. ...?.'clo. 10l
R. .. 19.. 5, 7, 126-137.

Olton, D. S. & Isesoco:a R. L. Hi0moc.al lesions nA activee
avoidance. r Aio.f 0;/ i- 1967 ?, 7197.

Papez, J. W. A pro I ZCI uec inis i of e otion. Archiivc of Nurol- r
a. PSchi ', 19 -7, ^-, 7?25-7 3.
Pelle-rino, L. J. & Cush-an, A. J. A stcer ic atlas of the rat

Peretz E. tion of a fuCd-rinfor : ons in h. cc.pctcz'i
cbts. Je.: l of C tiwe Uc :l5. c1 p> c -,,
1965, ..-1 A --
Pri... :, K. H. The r';n:.tc sy..ri.:-. Di:'.)c"3 in ccAing? In
E. R. Joihn (-:.), r. c- cf: 2 e.: York: Acc..nic
Press, 1969, pp.1 Z/.157

Pubols, B. H., Jr. T"': facilitation of vs!o.l a. s-:tial disci-ination
re : :.-1 by oerleariing, Jour ':. of_ C ar.tie rr P-: -lo-
l10 1r ?:.:' -'. 1956, 2-a -.

Rabe, Aus'?. Discrimination reversal deficit f-..' hippoc!'pal stiu-
lation in the rat. JourIal c^Ps-'cclcic 1 Stiooe1, 1963, 14,
139-150.
Rarp-:.L-on, A. C., Isaracson, R. L., & Do--las, R. J. Thr' effect of
limbic damage on the retention F3ni performance of a rui..:y
response. l: ?'L.o'o-'-, 1966, 4, 253-264.
Reid, L. S. The development of noncontinuity blh:iior tl:".:;::h continuity
learng. Journcl of yocriTenta Pscholor, 1953, 6, 107-112.
Roberts, W. W., Denbcr, W. N., & Brodwick, M. Alta-.-.tion end explor-
ation in rats with hippocczaal lesions. Jc-:.':-. of ConriTrative
ar.l Pi': I r- 1 .I c :-c 1962, ) c 65- J- .
Sch:':Iltz, L. W., & Isaacson, R. L. Fffect of bilateral hip.oc.-: :1'
destruction on the acquisition .-.i e.:.ti-ction of an operant
rc.p.:&:'e. Phycio 7 Py lBehl n.oL, 1967, 2, 291-298.





65

Sc.?.ltz, L., Isson, L. The effects '.2 prelinairy traininS
c'.ltios vrder I;L pr3foir,'nce in the hippoca .p cto' izcd
rat. IiEsjoloyv A z sravior, 1966, 1, 175-182.

Si: :1, S. C ....etric statistics for the b' a.vioral science-. Ne
York: .,c.. .;Ul' 1956.
Sp :11i, D. ., & Pribr'!n, K. H. C -.: se in visual recovery functions
prc;: *.ai by tenor al lobe sti:.ation in ....':eys. Electro-
enceDT -:-.v f' as C! I c u -i.5ozlo.-, 1966,9 20,

St" -, C. H. Dr'u actio on the electrical activity of the hippso-
capu~. InteMhtc U 1 ?7ie5 e o^2 .. !-0 ., 1965, 8, 77-138.
Stutz, R. M., K:15., K W. Cirjl-te. fornix lesions: 1fects
on two. types of reversal i:'.r .. Jouimra,l of CoTorative :.
',- .- "... : ... --:, 1968, 5-.-' -. ............

Sw: ,n H, G. G. --, fv: -'.tic ." th2 brain in olfaction. :: effects of
Ir.':- crtica1 lesions on olfactory discr.n:ir.tion. !, 7.ic; n
Jc -l.: 'P-sic1^ ? 19359 I1-, 357-26.
Sw.'-, E. function of the brain in olfacticon. II. The results
of destruction of olfactory 7d other ge.s Ue ctt res upon
the diiscrizination 2' odors. Journc, of C,.aative .eurol..,
1934, 175-201.
S7 tsCn, A. M, & I.::.-:--, R. L. Hippo- .1 ablation a-: pLrfoir :-,c:
durj-. 'thii"aral ": reinforcement. J,.o ':*-1 of Co.:Lrative and
P,'.': "' c ' P- ".: -' '., 1967, 64, in press.

Tei l11:-- H.. A comparison of the effects of orbitcro'atal and hippo-
ca.,ipal. lesions upon discrimination learning and reversal in
the cat. 7.:-rmental Isuroloy, 1964, %, 452-462.
Teit.clbaun, H.. & Milner, P. Activity ch-:n:.-. following partial hippo-
Ca-.:.2' lesions in rats. Jour..-1 of Co:narative and Physiolo:ical
Psychol 1963, 9 284-2.-- .
Thc:.c.r.., R., & Langer, S. K. Deficits in position re,:rs:l learning
follc'. lesions cf the linbic system. Journal of Comoarative
a:,-.d PL--. 1- ir 'l Pr-cFol-.-Co 1963, 96_, 9J7-Y95.
Wagnesr, A. R. ertrainir..g and frustration. Psychological Renorts,
1963, J, 717-718.

Wet-:'cr, D. B.v & Voneida, T. J. Learning deficits following hippo-
camipal lesions in split-brain cats. 1.-.r-nt:1 .-cr--_,
1964, L0, 170-182.







66

Winer, B. J. St:i.stpc. rc is 1 } r::c8:2l die n. jo. Yorl::

Wino:.1, G., Iz Mills, J. A. Hippo ca -pu? a' septun in response
inhibition. JcR u.Il of C5:n: 3.ve "-i T. :loic'! P.c'ioo. ",
1969, t?, 352-3,5.














Michael Arnold Mirlan '.s lorn on January 25, 1938, at New York,

New Yorl:< He attended -.Ll"c school in e":', Yo': state and gradetcd

from F.- yetteville-!anlius High School, Fayette671e, Nce' York, in

June, 1956. In May, 1960, he received the dc-rs of _1-.'..lor of Arts

froi Si.:. :e University. Froi 1960 until 1962, he sez.,il in the

Adjutant General Corps of the United States Amy. Follo'.ing his release

fr:i the Aany khe uas cnploycd ,s a psyc'.:-lojic.l research assistant

with the Veteran's A1:1nistration EospitJl in Syracuse, xe' Yor'k. In

Septe:.ct.i', 1963, he enroll~i in the gr -to s:' ool of the Univer~'.i

of Floridfa an r :eceivcd the Master of Arts d:rca in Decehber, 1965c He

served a- a r:.:.. .'ch assistant for Dr. H. S. Pe;:. -: .e and as an
interim inst-auctor With the I-:. rtnnt of P: L'c-.'l.- while r;triculating

for the Doctor of Philosophy degree in i:, chology.


BIOGRAPHICAL S (.-,C:'










This dissereNtion .:as prepared undel the direction of the chair-

man of the candidate's sup rvisory cc... .:ttee oand has been approved by

all cx'Abe'rs of that committee. It vas subLitted to the Dcan of the

Collo-s of Artt aind Sciences anid to the Gi..' to C: -'c l, a,. :?s

approved as pa:.tial fu fill3eent of the requirements for the dcgric of

Doctor of Philosophy.











LiEU .," C:.-."i. O- Sc: ool






; /t / 0











/ /
-r6^ ^^ ^ JS^^^'i/

















































7714B









As vindicated previously, a sign~if~c-nt interaction bet::oon thei
lesions factor and the acquisition sand reversal phrases of training was
revealed by the analyses of" varioace. The co-?ponents of the inter-

action are depictedi in Figu~re 10, whnichl presents the per~fornarace of

the three losion groups in acq~uisitiion training,, and in Figure 11,
which presents their perfor::mnrce in reversal trainin,-. The mean trials
to criterion for Ss in eachl of the throc lesion groups is presented
for acquisitions? and reverce.1 in Figure '12. Examination of these

ficares sucgests t'-t the three lesion groupFs did not differ in the

acquisition ph-se of training, hlt that in the reversal ph~se the3
neocortically and hippocragally lesioned Ss, while not differin3 a ong3
the-:selves, did attain the criterion? core r;pidly Reld in greater Lu:terss
than the shanl control Ss. A tosterioril comparisons bat::een the cell

mceanis involved in this interactions wer~e performed utilizing the
Studentized rangeo stiatisic (:inecr, 1962). The results. of the compari^

sons are presented, in surstary fashion, in Table 7. The results support
the above observations a!d. reveal, in addition, that shan control Ss

attained criterion significantly faster in acquisition than in reversal,
but that such a difference is not present in the neocortically and

hippocampally lesioned Ss. Neither the r~e-aining first order inter-
action (lesions by response-reinforcement contingency) nor the single
second order interaction (lesions by response-reinforcenent contingency

by acquisit ion-reverjsl) attained significance.
























nI~B
Z,
O C. O
C'
L' LI
%'L~
C
'
.:

O c r
;: "r /i






r
,-_--
JI;:
I
1----I- 'I -- r
---~a-
-rclm


O

o

o

Cl
ce
o1


1


E)LDO


N01~3dli13 Od. SNOISS3S










TABLE 7

STUDENIZEI RAN1GP STATISTIC A PO"7TI TESTS





Shan vs. Necocrtical Lesion. 3.01

Sham vs. Hippoccanpl Lesionl 0.86

seocortical vs. H~ippocompa;l Lesion 0.83




Shan3 VS. Neocortica~l Lesion 20.29"

Sham~ vs. HipIpoco pa~l Lesionl 13.92:"

Neocortical v;. Hippoc-:?ral Lesion 0.23




Acquisition vs. Rev`erssl 19.77-~


Neocortical Lesion

Acquisition vs. Reversal 0.01




Acquisition vs. Reversal 2.30


i'p<.01







to bilateral hippocampal stimulation during the acquisition and ex-
tinction of a food-motivated straight alleyway running response showed
no difference in rate or acquisition when compar~ied to control Ss but

did require a greatelr numbr of trials in extinction. This finding has
been reliably replicated in rats following hippocanpal destruction

(Jarrard &c Isaacson, 1965, Rapholson, Isaacson &c Douglas, 1966). A
closer eratination of the phenomenon indica~tes that the interval

between extinction trials is an important variable to be considered,
for while the increased resistance to extinction is deoonstrable when

trials are spaced, the hippocampal lesion deficit disappears in the
massed presentation of the extinction trials (Jarrard & Isaa~cson, 1905;
Jarrard, Islacson &c Wickelgren, 1964). Both Peretz (1965) and Douglas
and Pribrea (1966) have reported that hippocam~pctomized Ss shnow shorter.
response latencies and a greater number of responses to extinction than
do control Ss. Increased resistance to extinction ha~s also been de:;on-

strated in the two-wiay active avoidance paradigm (Isaacson, Douglas &r
Moore, 1961).

However, Schmaltz and Isaacson (1967) have presented slightly
divergent findings concerning the performance of hippocampectomized Ss
in extinction. They ran hippocamplly lesioned and control Ss to complete
extinction in as many 30-minute free operant sessions as were required
for the attainment of their stringent criterion. No difference was found
between the experimental and control Ss in the total number of sessions

required for extinction. In addition, the hippocampectomized Ss showJed
shorter response latencies in only the first extinction session; no
differences between groups were found for any of the subsequent sessions.














:3Z









Mr
Q o i io
ag U~









INTRODUCTION


The hippocampus, nestled in the inner folds of the temporal

lobe, has been subjected to more intense experimental investigation
within the past decade than in all previous years combined (after

Douglas, 1967). The early neuroanatomical investigations of Papez
(1937) which first linked the paleocortical structures of the rhinen-

cephalon with e-ctional behavior were supported by the contemporary

experimental investigations, performed by Kluver and Bacy (1933), of
the effects of temporal lo'osctorny on behav-ior. MacLean (1954, 1955,

19571 1958) for;alized this orientation, postulating a dichotomy between
the phylogenetically older paleocortex andl the more recent neocortex.
The former was presented as involved in. a~variety of emotional and

visceral functions, whiile the latter was thought to be concerned with

more cognitive functions. However, as the hippcampus wras subjected

to more intense experimental investigation a more complex picture of

hippocampal function emerged. Before summarizing the results of more
recent investigations of the role of the hippocampus in behavior and,

concurrently, reviewing the physiological theories of hippocanpal
function which have emerged from those data, a general description of the

hippocampus and its' interconnections with other brain structures will
be provided.

The hippocamipal formation, composed of the hippocampus (Ammon's

born), the hippocampal gyrus, the fascia dentata and. the fornix,11es along
the medial and ventral border of the temporal lobe where it is wrapped










Abstract of Disser~tation Pretentecd to the Graduatee Counacil in
Partial Fulfillmient of the Requirirents for the De,-ree of
Doctor of Philosophyi at the University of Florida


ALCQUISITIONJ AND R2EiESALr OF A TWO RAI~PUIANDA DIFF~IER:TITIONj IN
SHAM, NEOICORTICA.LY, AND HIPPOCAMPALL LESjIONED P,1TS


Michael Arnold Hiilan

June, 1970


Ch~airman: E., S. Pennypa~ckr
MlaJor Dealrtment: Psychology

Sham, ne~ocortically, and hiI~ppocpaly lesioned. rats were exr-

amined in the acquisition of a two an~ipuilana different~iation imrder

conditions which insured that either the absolute number of reinforced

or non-reinforced responses each S elitted on ech renipulandu2l during-

each expTerimental session were equasted. Acquisition performance was

not differentially affected by the thnree lesions, nor by equated

reinforced or non-reinforced responding. Reversal performance did not

differ for the neocortically and hippocampally lesioned Ss, but both

appeared to be facilitated when comparedl to shamn control Ss. As in

acquisition, equated reinforced or noa-reinforcedi responding: did not

differentially affect perf~ormancea of the three lesion groups.









varying conditionsr in n-eaous studies (e6. Isaacson? & Wickelgren,

1962; Kimbnle, 1963; Teiteltrum & H!ilner, 1953). Those studies which

have failed to replicate these finding-s have reportedly restricted lesions

involving only the dorsal portionl of th~e hippocenpuis (Eoitiano & Isaacson,

1966; K~vein, Setekcliev & K~anda, 190'4), or have employred a response of

low probability (Kinble, K~i~rby & Stein, 1966; Wiivocur &c Mills, 1969).

It is possible to construlct various cxplcnations of the un-derlyin,

nature of thle observed hippoca r3pectolyi:nducll, deficit; one such sugg~esticn

is thact hippocrcparl lesions in sole way: vitiate th~e aversive effect of"

the punishingp stirulus. Such a position would9 lead to the prediction-

that hip~potampectomisca Ss would be def~icient inl a wide range: of shrock

motiv-atel behav-iors. This does not appearn to be the case, however, for

hippocac~pcto-ized Ss are not necesterily r 1 'e;;le co;;~e to

control Ss ir. their ability to r-ster a variety of active avoidance

task~s.

In the typical onle-:fay active avoidance paradign Ss is required

to m~ove from one8 coSparttent of a ShuttletoX to a second in the presence

of a warnin,- signal to avoid an aversipe stimulus is delivered and 5 must

then perform the response in order to escape fro:: it. FollowringP completion

of the trial S is returned to the original com~partment of the shuttle-

box and the procedure repeated. Although Niiki (1962) reported that

destruction of the hippocampus had no effect on this variation of

avoidance conditioning, more recent invecstigations have indicated that

there is aI lesion-ind~uced deficit which, however, appears to be of a

lesser relative mag-nitude than that found in the passive avoidance

paradign (Ilc~iirew Thnonpson, 1966; 01~ton & Isaacson, 1967).







Such as 'perseveration,"! "6gating," or "inhibition" without further

refrinement or qua~lificatin.k

The third r:-jor difference betwooen this and contemporary in-

vestigaztions of^ the- role of the hippoca~pus. in discrinination learning

involves the schedule of reinforceloent associated with the tw-o to-ba-

discriminrated responses. The research reported previously hazs typically

provided cont-nuous reinfo:: :ae t for "cor~rect" responses and writhheld

reinforcement f~or "incorrect"' responses. In thez present study con-

curfrent ope~rans were utilized: Both responses were reinforced, one

on ian FR 5 schedule and th~e other on an PR 9 schedule, and the forn-

ation of the discrimninatio~n was based on a relative, rather than

absolute, diffecrential in! 1einf'oi~rceaent desity. The experilental

anarlys-is of cocuc~rrent rattio schediules indlicates t'-at withi unequl FR

requirlements, respoding tends to be rlintainedt only by the schle~dle

with the smaller FR requnir;ent; with equal FR requirements, responding

can be maintained by either one, and shlifting fromi one schedule to

the second occ-rsionally occurs (Catania, 1966, Herrnstein, 1958). In

a study which is only superficially corwarable to the one reported

here, DougEla~s and Pribram (1966) examined the effects of probabilistic

reinforcement upon the for;3ation of a discriminated panel press in

mor3keys. As inl the present study the discriminationl rested upon a

relative differential in reinforcwe.2e density; one response was

reinfocced 70$ of the time and the second reinforced 30i0 of the time.

Their results, in contract to those of the present study, indicated

thact hippocampectomized Ss are ret~ared with respect to control Ss





31

TABLE 4

SPEARRAN RLUKf ORDER CORRSITION6 COFFICIN;TS (r ) FOR SKS', N-OCORTICAL~,
ANID EIPPO)C?:PAL ~IDSIO GROUPS UNDER EQULTED REINFJORCED
OR ~OI:-REINFORD.CED ICSPONDING DURI ;G TXINP(ITC~



REinf~orced Response ~ on-Reinfo~r-ced Responses
Equted. Equated

Sham Control rs = .42 rs =.7
p> .05 p <.01

Neocortical Lesion rs =.01 rs = .29
p>.05 p>.o5

H~ippoc 1r~ Lesion rs = .68 rs = *35
p<.05 p>.05

ESee text for rank-in~g procedure.










Meissner, N. Hlippoc1C'Sl functions. in learning. Joa-rn:1 of ?'
Rese -rch, 1966, 4, 235-3 A'1

N~iki, H. The effects of hipp~cenpal abtleaion on the teleiror in the,


Nii .The effects of hip ~xc?7ysl abletion onl the i~h~ibitory-
control of coran~iet behavior in the rat. 2 .
Rescarch, 19 5, 2, 126-137.

01ton, D. S. & Isaacson, R. L. Hiippoempa:l lesions aond active
avoidance. Physio0_~~nlo-o;i7ir ed ,ir 1967, 3, 719-724.

Papez, J. U!. A proposed Dech~nisn of elction. Arcihnes ol F ur N}-
F.- I. .~ P._ 1932, 23, 725-243.

PellegDrino, L. J. &; Cusban, .'.. ?. .4_ : '`'.* -'1 '-


Peretz, E. 'Extinction of a. food-reinforce" response: in hipp~oc"-.pectc~~izt
cats. 7-- I -1


Pribrea, K. E. The alnestic syndrl-res: Di":r3na in ecding? In
E. R. John (Edi.), .' r NewYork: Aca aic
Press, 1969, pp. 11,-1_.

Pubols, B. H., Jr. Tihe facilitation of visual ad satial discrimination
reversal by overlearning. ..- i :'; : i7- --i T',~ 1
logical Psycholy,~; 1956, L,

Rate, Aue-s,. Discrimiination reversal deficit from hippor 31s stin;u-
1 Sosn n thle rat.Jcr : :e -: 93, ,

Raphelson, A.. C., Isacson, R. L., & Douglas, R.. J. The effect of
limbic damage on the retention and perfor)ance of a runwa~y
response. Neuroggyfoogi~a, 1966, 4, 253-264.

Reid, L. S. The develop-ient of noncontinuity behaviior through continuity
learning. ...:-1. F-r i.35. 93 ,171

Roberts, W. Vl., Denber, 8. N., & BrodJickc, M. Alternaztion and explor-
ation in rats with hippocompll lesioins. ;->1o ~-ptv


Schmaltz, L. U., &; Iszracson, R. L. Effect of bilateral hippocaomal
destruction on the acqulisition and extinction of an operaent
response. Physi l' Wavor 1967, 2,~~ 291-298.








Dougelas,1 R. J The 1 "p;.~ us and b~ehavor. Psgchological Balletin,


Douglas, P.. J., & Prihra-1, K. H. Learning and liabic lesions.
.~:r-:: -1- 'l ''~1666, 4, 1.97-220.

Ellen, P., &E Wilson, A. S. Persove-ration in the rat foll.owing hippo-
ca'upal lesions. .1 'lii-1: i:.1 1963, 8, 310-317.

Feldman, S. Neurophysolo~ical tech~anisms andl modifyJing afferent
bypothalamo-hippocaamp.1 condiuction. Exoermnta Nurl
1962, j_, 269-291.

Gloor, P. inyg~dala. In J. Field (Ed.) :' -!: ::(. i :i .}.2



Grastyan, E. & Famaos, G. The inf~luence of hippoclcpal lesions on
sim;ple and delaying instrulcatal conditioned re~flexes. In
P. Passosnnut (Ec.) IT:L. to ..~_ill~ ~ 1962,
pp. 225-239.
Green, J. D. The hippocarpuiis. i 0 1 1964, fi4 561-608.

Green, J. D. &: Aey, ". R. Electropy!siolo,-ical studies of hipp~ocomp;.l
colnnections and excitability. ..., : 1


Green, J. D., & ArduinJi, A. Eippocargal electrical activity in arorusal.
Journal of Neurovhysioloff, 1951) 1? 533-557.

Herrnstein, R. J. Soale factors influenlcingo behavior in a twoP-response
situation. Transactions of thc N~ewr Yorki Academyv of Sciences,
1958, 21, 35-4 i7P1 ----

Hirano, T. Effects of hippocanpal electrical stimulationl on memlory
consolidation. Pacgg, 19661 2. 63-75.

Hostetter, Gayle, & Thomas, G. J. Evaluation of enhanced thigl~otaxis as
a condition of impaired maze learning by rats with hippocampal
lesions. Jounasl of Compnrtv 1n Phsooia 7cool,


Isaacson, R. L., IDouglas, R. J. & Moore, K. Y. The effect of radical
bippocapal ablation on acquisition of avoidance response.
Jounlo Coarative and Physiological Pylog, 1961, j2i,


Isaa3cson, R. L., 01tojn, D. S., Bsuer, B., &r Swart, P. The effect of
the nuat~er of training trials on the deficit in Iassive
avoidance behavior in the hippocampectomized rat. Paper read
at Midwecstern PsycholoDical Association, ChicaCgo, 1966.







hippoc-rmpectomized Ss showed. no deficit in their ability to inhibit an1
unlearn-d escape response from a small, elevated perch when the response

was punished lead to their fonrml statement of that position. H~oucver,
Isaacson, 01ton, Bauer and Swart (1966) and Teitelbaum and Mlilner (1963)

have presented contradictory data, indicating thait hippocanpectomized

Ss are deficient in withholding~ a natuorally occurring- response involving

a step-down from3 a platform to an electrified grid. The foraor authors,

who shooke the platform to increase the probability of response occurrence,

suggested that the escape response elrployed by K~imble, Kilrkby and Stein

(1966) was too weak or improbable in nature to adequaately reveal a

hippocampal lesion-ind!uced deficit.

Inhibitory deficits of hippoclr:Tctomized 55 have also been

widely examined within the context of exploration and spontaneous

alternation paradigns. Roberts, Dentler a';d Brodwick (1962) compared

exploration rates of hippocacpectonized and control Ss in T- and Y~-azes
and found no differences in initial rates, but a more rapid decrease in

exploration rate in control than in lesione-d Ss. An additional analysis
revealed that Ss with small hippocanpal lesions showed a moderately, but

significant, slower exploration rate decrease than controls, and that Ss
with massive hippocaupal destruction showed no rate decrease whatsoever.

Leaton (1965) studied opportunity for exploration as a reinforcer of a

T-maze turning response and found evidence for acquisition in normal and

shamn operated Ss while hippocaspectomized Ss were unable to overcome

perseverative tendencies and consequently showed no acquisition effect.

Forced training was instituted in the second phase of the experiment and

measures of" running speed were takecn. The hippocomp-eomized Ss showed

slower habituation to the reinforcer, indexed by a slower decline in










-77
000
zon


J ch
<1 e. I'
24 00
0oo
i~u E en
z ,
b--~n U)(l c



It~~ 1 8 ~










DISCUSSION


This study examined the performance of sham, neocortically, anld

hippocaupally losioned rats in the a~cquisition and reversal of a two

manipuland a ifferentiastion as affected by certain magnipulations of

response-rzinforcem`ent contingencies. For one-half the shes, neo-

cortically, and hippc 11yll lesion~ed Ss, these manipulations insured!

that each S e-itted an equal number of reinforced responses on the

twro manipulanda1 durinQ a complete experimental session. As a result

of this procedure, each S_ also emitted twice as -many non-reinforced

responses on one ma2nipulantumd (the FR 9 manipu~lcaduo as on the secondl

(the FR 5 manipulanudum). For the remaining Ss, the experimental

manipulations insured that each S, saitted an equal number of non?-

reinforced responses on the tw;o man~ipulana during a complete exp~er~i-

mental session. As a result of this procedure, each S emitted twice as

many reinforced responses on the FR 5 nanipulandun as on the FR 9

manipulandum.

The Douglas-Pritram theory of hippocampal function (Dougalas &

Pribraml, 1966; Doug~las, 1967) is a vigorous attcnpt to integrate the

wide variety of behavsioral changes following hippocamal disruption,

and leads to clear-cat predictions of the behavioral effects of the

mninpulations performed in this study. Specifically, the theory

predicts retarded acquisition and reversal in hippocampal Ss wihen

compared to sh-3 and neocortical control Ss under the treatment condition

which specifies eqluation of the number of reinforced responses emitted







sensitive anyiglaloid system. Although thie theory is a posteriori
in construction, Dougla-s and Pribram (1966) do present some data
confirmi~n predictions made from the theory.

The present experiment focuses upon the hippocam;pus an~d its
proposed involverent in situations involving nonl-reinf~orced responding~.
An experimental pa~radig2 in wihich manipulation of reinforcement and

non-reinforcement contingencies generates differential predictions
concerning the behavior of hippocampectosized rats has been developed
from the theory in question. In both the acquisition and reversal

phases of a position discrimination, equation of" the absolute number
of reinforced responses to each of two to-be-discriminated manipulandla
combined with differentiation between the two in terns of the absolute

number of non-reinforced responses would be predicted from7 the theory
to retard both acquisition eand reversal in hippocarpectomized Ss when

compared to neocortically-lesioned andZ shank operated controls. However,
when the absolute number of non-reinforced responses to the Da~nipulanlda

are equated and the number of reinforced responses differ, any hippo-

camp~al lesion-induced deficit would be predicted to be of a significantly
lesser magnitude if present at all. Positive results would constitute

support of the Douglas-Pribram theory (Douglas, personal communication,
1968).








from low to high on the matber of sessions to attain criterion in

acquisition, and from high to lold on, the number of sessions required

to attain criterion in reversal. A Spearcan rank: order correlation

coefficient (rs) was then computed (Siegel, 1956), and found to be

significant (rE; = .47, t = 4.0567, at' = 59t p <.001). sp~eaman rnkl
order correlation coefficients were also co7putEd in the same nsaner

for the shcz, neocortical, anid hippocanpal lesion groups (see Table 3),

and for the three lesion groups when further divided on thle basis of the

reinforced versus non-reinforced responses equated dinension (see

Table 4). The form-er analysis indicates the inverse relationship between

performance in acquisition a~n? re-ve-sal is present in only the shce

control Ss,; the latter analysis reveals that c-hil~e the shat control 5_s

under the non-reinfiorced responses equated condition do show this

relationship, their counterparts under the reinforced responses equated

condition do not. In addition, the finer grain analysis indicates that

the hippoctapectolized 55 murder the reinforced responses equated con-

dition also mnrifest this relationship, albeit to a lesser degree.

The sessions required to attain criterion in acquisition and

reversal for sham, neocortically, and hippocaapclly lesioned S~s under

the reinforced responses equatedl requirement are presented graphically

in Figu~res 3 and 4, respectively. Analogous data for Ss under the

non-reinforced responses equated condition are presented in Figulres 5 and

6. An analysis of variance assessing the effects of the three lesion

conditions, the two response-reinforcement contingencies, and acquisition

and reversal upon performance as indexed by the trials to criterion

measure was performed (Winer, 1962). Since the trials to criterion













IL~r~a":CES


A~day, W. R. Stt'die o" 3'i-^r- -1 .-Oct-rictl activiity du~ringp approch
learnirs. In J. i. 11.! .. e (E4.), J- : i
lerrpirp Oxiford: Blackuellr, 1961, pp. ;.-

Allen, Wi. F. Effect of ablatin; the frontal lotas, hippocc.:api, and
occipite-- rieto-tc:porcl (exccyting py'rif~orm ares) lobes
on pocsitivec oil negative 01fcactory cjalitione-: reflexes.
Anori e 7 1 1 ,1940, 128, 7 54-771.

Allen, Y. F. Effect of ablating the: pirofoMrrn~:ayalo~id? aeaBsand
hippocr-l 01 piositive a1 negative 01factory differentiation.,

Boitano, J. J., & ICICSreso R. L.. Effects ol variat~ion in shock-
inten~sity on the3 Fi:I110? of dorsal-hIpo:co otenized'` ra~ts
in two pa1ssive COrI .eF Situations mei nJ$n1o
i 1966, 80, 7T'80.
Brown, T. S., Kauf-~:i12, P. G., &- D:rco, L. A. The hippetepuis and.
respornce perseveratic.! in. the cat. 3-r in Eases~rch, 1969,
12, 86-.98.

Caljal, S. Y. :TL;'i 2 .- .: -C: -1 .---t- l. 11 Linbic


Catanic, C. A. Concurrent operant. In W. K. Honig, (Ei.), OAcrafnt
; ?r : -11 -- ... N w o


Clark, C. V. 9., 8: Isaacson, Ri. L. Effect of bilat~era~l hippocanpal
ablation on DRLZ performances. J_':rn:.1 cf r.----rt:ti- Ir;


Correll, R. E. Thle effect of bilateral hippocampal stimulation on the
acquisition ean extinction of an instrumental response. Journal
of Connrat~?;ive ad Pysilgcal Psyholog 1957, 2, 62 i29.~

1)ewson, J. Hi., III, Nobel, K. Y. & Pribranl, K. H. Corticofugal
influence of cochnlear nucleus of the cat: Some effects of
ablation of insular-tepor~al cortex. Brain Research, 1966,








Although hippoceapectomized Ss appear deficient in their ability

to withhold responses in successive or go-no go discrimination problems

(Kimble, 1963), nmr~orus studies have demonstrated that they do not
differ from control Ss on a iiide variety ofT simultaneous discrimination

pr-oble-,s (Alleni, 19410, 1941; Brojn, Kaufnan &; MaIrco, 1969; Grastyan &:
Karaos, 1962; Hirano, 1966; Kimble, 1963; Kimble & Zackc, 1967; Swann,

193rr, 1935; Teitenaum~, 1964; webster &- Yoneida, 1964). Wlhen hippo-
caspectomized Ss are required to reverse such a discrimination, a.

pronounced deficit in shif~ting responding from that which was previously
reinforced to that which is nearly reinforced is regularly observed

(Brown, Kaufcan &c Marco, 1969; Kintle & Kintle, 1965; ?abet 1963; Stutz
& Rock~lin, 1968; Suanson & Isaacson, 1967; Teitelbau, 1961j; Thocmpson &:

Langer, 1963).
In an attempt to explain the changes observed in positively

reinforced behavior following hippcom~pectony in terns of the loss of a

single process contributing to such behavior in the intact orgainisl,

Douglas and Pribram (1966) developed a sophisticated neurophyrsiological
theory of "problem solving." Although~ the authors were initially con-

cerned with the hippocampus, they found it necessary to include in their

theory a second limbic system structure, the amygdala, in order to account

for the behavior of which hippocampactomized Ss are capable. Each of
these structures is postulated to be intimately involved in two distinct

processes underlying problem solvinga or discrimination learning: the

hippocampus-centered "error-evaluate" process and the complementary
anygdala-centered "reinforce-register" process. The terms are indicative
of the function of each: the reinforce-register process is depicted as

















;i ii
p C;:
i-i



i "-,
i




u
d


FiE. 2.P-Trcings of representative cross sections through the
neocortical lesion (After Pelle,"rino and C71 n71, 1907).











BIOGRAPHICAL SKETCHI


Miichael Arnold Mlilan wavs bor on January 25, 1938, at Newl York,

NewI Yo~r. He attendied public school in Hewr York stzte and graduated

from Pay~etteville-':an~lius High School, Payetteville, Niew York, in

June, 1956. In May-., 1960, he received the degree of Eachelor of Arts,

fron Syracuse Universi~ty. From 1960 until 1962, he served in the

Adjutant General Corp~s of the United States Army. Following his release

from the Arn-iy he waes elployed a~s a psycho.logical research assistant

with the Veteran's Adinirstration Easpitc1 in. Syracuse, Kel; York. In

September, 1963, he enrolled in toe grandnate school of the University

of FloriCda andc received the Malster of Arts degree: in December, 1965. Be

served as a research assistant for Dr. HI. S. Pennypackier and as an

interim instructor with th~e Departrant of Psyrcholoy while ratriculating

for the Doctor of Philosophyl degree in psycholog-y.











-' J




:L- CI

2 > a

U U4



'" :
LO


t~
D
o




Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EJ047HQRP_N63ZME INGEST_TIME 2017-07-14T22:52:51Z PACKAGE UF00097730_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES



PAGE 1

ACQUISITION AND REVERSAL OF A TWO MANIPULANDA DIFFERENTIATION IN SHAM, NEOCORTICALLY, AND HIPPOCAMPALLY LESIONED RATS By MICHAEL ARNOLD MILAN 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

PAGE 2

3 1262 08552 4113

PAGE 3

AOTOtfLEDGMENTS I gratefully acknowledge the support and direction provided by Dr. H. S. Pennypacker both in the execution of the present project and, more importantly, throughout cy graduate career. I also express ny appreciation to Dr. Robert L. Isaacson, Dr. Frederick A, King, Dr. C. Michael Levy, and Dr. Wilse B. Webb for their assistance in the development of this dissertation. I thank Mrs. Pauletta Sanders and Mrs. Gloria Smith for their histological assistance, and Mrs. Irna Saith who ably typed the present manuscript. ii

PAGE 4

TABLE OF CONTENTS Page ACKNOWLEDGEMENTS ii LIST OF TABLES i 7 LIST OF FIGURES v ABSTRACT vi INTRODUCTION 1 METHOD 14 RESULTS 2k DISCUSSION ' k7 REFERENCES &) BIOGRAPHICAL SKETCH 6? ill

PAGE 5

LIST 0? TABLE Table Pcg3 1. TRIALS TO CRITERION IN ACQUISITION AND REVERSAL FOR SEAM, NEQCORTICALLY, AND HIPPOCAMPALLY L23I0NED Ss ASSIGNED TO THE REINFORCED RESPONSES EQUATED CONDITION 2? 2. TRIALS TO CRITERION IN ACQUISITION AND REVERSAL FOR SHAM, NEOCORTICALLY, AND HIPPOCAMPALLY LB3I0NED Ss ASSIGNED TO THE NONRE IN FORCED RESPONSES EQUATED CONDITION 28 3. SPEARMAN RANK ORDER CORRELATION COEFFICIENTS (r ) FOR SHAM, NEOCORTICAL, AND HIPFOCAMPAL LESION GROUPS . . 30 4. SPEARMAN RANK ORDER CORRELATION COEFFICIENTS (r s ) FOR SEAM, NEOCORTICAL, AND HIPFOCAMPAL LESION GROUPS UNDER EQUATED REINFORCED OR NON-REINFORCED RESPONDING DURING TRAINING 31 5. ANALYSIS OF VARIANCE ON TRIALS TO CRITERION 3? 6. ANALYSIS OF VARIANCE ON SQUARE-ROOT TRANSFORATION OF TRIALS TO CRITERION 38 7. STUDENTIZED RANGE STATISTIC A POSTERIORI TESTS k6 iv

PAGE 6

LIST OF FIGURES Figure Page 1. Tracings of representative cross sections through the nippocanpal lesion 25 2. Tracings of representative cross sections through the neocortical lesion . . . 26 3. Number of sessions for Ss in each lesion group to attain criterion in acquisition: Reinforced responses equated „ 32 4. Nusiber of sessions for Ss in each lesion group to attain criterion in reversal: Reinforced responses equated « 5. Hunter of sessions for Ss in each lesion grout) to attain criterion in acquisition: Non-reinforced responses equated # ^ 6. Number of sessions for Ss in each lesion group to attain criterion in reversal: Non-reinforced responses equated ^ 7. Cumulative percentage of Ss in each lesion condition attaining criterion in successive 5 session blocks. 29 8. Cunulative percentage of Ss in each response-reinforcenent condition attaining criterion in successive 5 session blocks ^ 9. Cunulative percentage of Ss attaining criterion in aquisition and reversal in successive 5 session blocks ^ 10. Cunulative percentage of Ss in each lesion condition attaining the acquisition criterion in successive 5 session blocks ^ 11. Cumulative percentage of Ss in each lesion condition attaining the reversal criterion in successive 5 session blocks ^ 12. Mean trials to criterion in acquisition and reversal for Ss in each lesion group 45 v

PAGE 7

Abstract of Dissertation Presented to the Graduate Council in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy at the University of Florida ACQUISITION AND REVERSAL OF A TdO MANIPULANDA DIFFERENTIATION IN SHAM, NEOCORTICALLY, AND HIPPOCAMPALLY LESIONED RATS Michael Arnold Milan June, 1970 Chairman: H. S. Pennypacker Major Department: Psychology Sham, neocortically, and hippo car.pally lesioned rats were examined in the acquisition of a two manipulanda differentiation under conditions which insured that either the absolute number of reinforced or non-reinforced responses each S_ emitted on each manipulandum during each experimental session were equated. Acquisition performance was not differentially affected by the three lesions, nor by equated reinforced or non-reinforced responding. Reversal performance did not differ for the neocortically and hippocanpally lesioned Ss, but both appeared to be facilitated when compared to sham control Ss. As in acquisition, equated reinforced or non-reinforced responding did not differentially affect performance of the three lesion groups. vi

PAGE 8

INTRODUCTION The hippocampus, nestled in the inner folds of the temporal lobe, has been subjected to more intense experimental investigation within the past decade than in all previous years combined (after Douglas, 1967}. The early neuroanatomical investigations of Papez (193?) which first linked the paleocortical structures of the rhinencephalon with emotional behavior were supported by the contemporary experimental investigations, performed by Kluver and Bucy (1939), of the effects of temporal lobectomy on behavior. MacLean (1954, 1955, 195?, 1958) formalized this orientation, postulating a dichotomy between the phylogenetically older paleocortex and the more recent neocortex. The former was presented as involved in a„ variety of emotional and visceral functions, while the latter was thought to be-concerned with more cognitive functions. However, as the hippocampus was subjected to more intense experimental investigation a more complex picture of hippocampal function emerged. Before summarizing the results of more recent investigations of the role of the hippocampus in behavior and, concurrently, reviewing the physiological theories of hippocampal function which have emerged from those data, a general description of the hippocampus and its' interconnections with other brain structures will be provided. The hippocampal formation, composed of the hippocampus (Amnion's horn), the hippocampal gyrus, the fascia dentata and the fornix, lies alon< the medial and ventral border of the temporal lobe where it is wrapped 1

PAGE 9

around the posterior surface of the thalamus. The hippocampus, reminiscent of the common sea horse Hippocampus hi ppocam pus from which its name is derived, is the major structural component of the hippo campal formation. Gross description of the hippocampus was provided by Lorente de Ko (193-'» cited in Douglas, I96?) who divided it into four segments: CA1, located proximal to the subiculum, CA2, CA3, and CA4, lying in the fold of the granule cell layer of the fascia dentata. The most generally accepted cytoarchitectonic description of the complex internal structure of the hippocampus was provided by Cajal (1955) • Starting from the ventral surface above CA2, then proceeding vertically, seven major layers are evident: the ventricular ependyma, alveus, stratum oriens, stratum pyramidale, stratum radium, stratum lacunosum, and stratum moleculare, A detailed exposition of the internal morphology of the hippocampus and hippocampal formation is provided by Meissner (1966). Two major afferent pathways serve the hippocampus: the alvear path through the fornix systea and the perforant path through the subiculum. The fibers of the fornix arise primarily in the septal area and the intralaminar nuclei of the thalamus (Green & Adey, 195^) • The system is more involved than this, however, for inputs also reach the hippocampus from the ascending reticular activating system of the midbrain and thalamus (Green & Arduini, 195*0 and from the hypothalamus as well (Feldman, I962) . The perforant path reaches the hippocampus by way of the entorhinal cortex and subiculum. The temporoamonic tracts pass from the entorhinal cortex through the subiculum to the hippocampus proper. The entorhinal cortex, in turn, receives its afferents from considerable areas of the neocortex (Green, 19&0. In addition, there is

PAGE 10

evidence for direct f iters from the cingulum attaining the hippocampus via the perforant pathway (Adey, 1961). Fibers passing into the fimbria constitute the main efferent pathway of the hippocampus. These fibers cross to the contralateral hippocampus via the hippocampal commissure, or enter the fornix and project variously to the septum, hypothalamus, anterior thalamus, and rostral portions of the brain stem (Green & Adey, 19 56). The hippocampus also gives rise to efferent fibers to the entorhinal area via the temporoamonic pathway. In addition, Gloor (i960) presents evidence for primary hippocampo-amygdaloid fibers as well as secondary amygdalohippocampal connections. An intensive review of the literature pertaining to the neuroanatomical investigation of the hippocampal afferent and efferent systems may be found in Green (1964) and Stumpf (1965). In an attempt to assess the contribution of the hippocampus to the physiological substrata of behavior, researchers have examined the effect of hippocampectomy upon a wide variety of behaviors. One of the most studied classes of behavior falls under the general notation of avoidance conditioning. Interest was sparked in this paradigm by the relatively early study of Kimura (1953) who found that rats with bilateral posterior hippocampal lesions were deficient when compared to neocortically lesioned and sham operated subjects (Ss) in their ability to vithold a well-practiced, food motivated approach response following the introduction of punishment (electric shock) of the consignatory response. The essential characteristics of this experimental paradigm are prototypic of what is generally referred to as passive avoidance conditioning. Such deficits in passive avoidance have since been replicated under

PAGE 11

varying conditions in numerous studies (e.g. Isaacson & tfickelgren, 1962; Kimble, 1?63; Teitelbaum & Milner, I963). Those studies which have failed to replicate these findings have reported restricted lesions involving only the dorsal portion of the hippocampus (Boitano & Isaacson, 1966; Kvein, Setekliev & Kaada, 196*4), or have employed a response of low probability (Kimble, Kirkby & Stein, 1966; tfinocur & Mills, 1969). It is possible to construct various explanations of the underlying nature of the observed hippocampectoray-induced deficit; one such suggestion is that hippocampal lesions in some way vitiate the aversive effect of the punishing stimulus. Such a position would lead to the prediction that hippocampectomized Ss would be deficient in a wide range of shock motivated behaviors. This does not appear to be the case, however, for hippocampectomized S_s are not necessarily retarded when compared to control Ss in their ability to master a variety of active avoidance tasks. In the typical one-way active avoidance paradigm Ss is required to move from one compartment of a shuttlebox to a second in the presence of a warning signal to avoid an aversive stimulus is delivered and S_ must then perform the response in order to escape from it. Following completion of the trial S_ is returned to the original compartment of the shuttlebox and the procedure repeated. Although Niki (1962) reported that destruction of the hippocampus had no effect on this variation of avoidance conditioning, more recent investigations have indicated that there is a lesion-induced deficit which, however, appears to be of a lesser relative magnitude than that found in the passive avoidance paradigm (McNew & Thompson, 1966; Olton & Isaacson, 196?).

PAGE 12

If the ahove avoidance paradigm is modified so that S_ is not returned to the original coapartaent of the shuttlehox following each trial but instead must return to it as the response in the following avoidance trial the task is reteraed two-way active avoidance. When compared to neocortically lesioned and sham operated Ss, hippocampectoaized Ss appear to he facilitated in the acquisition of this response (Isaacson, Douglas & Moore, 19&L). It has heen suggested that facilitation is due to the presence of a passive avoidance component involved in the two-way active avoidance paradigm which interferes with acquisition in control Ss (S_ is required to return to the compartment which has most recently teen associated with aversive stimulation). Hippocampectoaized Ss which have been demonstrated to he relatively impervious to the effect of the contingencies necessary for the instatement of the passive avoidance response are not so hampered and consequently acquire the two-way active avoidance response more readily (Douglas, I967). The slight deficit seen in the hippocampectomized S_s* acquisition of the oneway active avoidance task can also he related to the deficit hippocampectoaized Ss manifest in passive avoidance. Here, however, the hippocaapectomized Ss* tendency not to avoid the compartment associated with aversive stimulation retards acquisition relative to control Ss (Olton & Isaacson, 196?). A second formulation of the underlying nature of the hippocampal contribution to behavior which, like the aversive stimulus position, relates to the passive avoidance deficit suggests that hippocampal lesions enhance the reinforcing properties of appetitive stimuli or,

PAGE 13

alternatively, holds that the hippocaarpal lesion in some fashion elevates drive level relative to non-hippocampectomized Ss under equal levels of deprivation. Jarrard (1968) has briefly reviewed the literature which supports this position and points out that in addition to the intimate connections of the hippocampus with structures important for physiological homeostasis, behavioral evidence indicates that hippocampectomized S_s are more active in both novel and non-novel situations, increase their response rate for food and water, and show slower extinction of a foodmotivated running response. Although hippocaiapectomized Ss have not been found to eat more food than control S_s, they have been found to drink more water. It has been argued that the increased drive hypothesis has generally been abandoned (Douglas, 1967). However, the arguments marshalled against this position have stressed the findings of the avoidance conditioning paradigms and reasoned that because hippocampectooized Ss do not appear to be more sensitive to the driveinducing properties of aversive stimuli than do intact Ss, it is inappropriate to posit that the reinforcing or drive reducing properties of appetitive stimuli might differentially affect hippo caapectomi zed and normal S_s. Such a critique is cogent only if the theorist holds that a unitary or one-process theory of hippocampal function will explain the whole spectrum of lesion-induced behavioral anomalies. Whether it is possible to formulate a one-process theory of hippocampal function has yet to he demonstrated. The deficit seen in the passive avoidance performance of hippocaspectoaized Ss has also been viewed as a manifestation of a general tendency towards response perseveration or, alternatively, an inability to inhibit responses. A considerable body of evidence is available in eitr>TM"iT>t ri-P en/Oi o nn^l -H nn fi-v-fr-pl 1 MOO^ fminrl tViO t rate: eniTi Iprt.pfl

PAGE 14

to bilateral hippocanpal stimulation during the acquisition and extinction of a food-motivated straight alleyway running response showed no difference in rate or acquisition when compared to control Ss but did require a greater number of trials in extinction. This finding has been reliably replicated in rats following hippocanpal destruction (Jarrard & Isaacson, 1965, Raphelson, Isaacson & Douglas, 1966). A closer examination of the phenomenon indicates that the interval between extinction trials is an important variable to be considered, for while the increased resistance to extinction is demonstrable when trials are spaced, the hippocanpal lesion deficit disappears in the massed presentation of the extinction trials (Jarrard & Isaacson, 1965; Jarrard, Isaacson & Wickelgren, 19&0. Both Peretz (1965) and Douglas and Pribram (196*6") have reported that hippocanpectoaized Ss show shorter response latencies and a greater number of responses to extinction than do control S_s. Increased resistance to extinction has also been demonstrated in the two-way active avoidance paradigm (Isaacson, Douglas & Moore, 1961). However, Schmaltz and Isaacson (196?) have presented slightly divergent findings concerning the performance of hippo campectoini zed Ss in extinction. They ran hippocampally lesioned and control Ss to complete extinction in as many 30-minute free operant sessions as were required for the attainment of their stringent criterion. No difference was found between the experimental and control Ss in the total number of sessions required for extinction. In addition, the hippocampectomized Ss showed shorter response latencies in only the first extinction session; no differences between groups were found for any of the subsequent sessions.

PAGE 15

Kaplan ( 19 6? ) has reported that hippocaiapectomized Ss show faster extinction of a freezing reaction taken as indicative of a classically conditioned emotional response. The general inability of hippocampally lesioned S_s to inhibit responses has "been widely demonstrated in a number of other situations. Ellen and Wilson (19^3) found hippocanpectomized rats impaired in their ability to inhibit one type of bar press and adopt a second following a change in the response requirements for reinforcement. Both Niki (1965) and Swanson and Isa3.cson (1967) have demonstrated a hippocampal lesion-induced deficiency in yielding to stimulus control following the initiation of S D «S training. However, the latter authors also demonstrated that hippocanpectomized Ss could readily acquire the discrimination provided they were not subjected to a long past history of continuous reinforcement for responding prior to the initiation of discrimination training. Clark and Isaacson (1965) found that hippocanpectomized Ss were less efficient than control Ss on DHL schedules of reinforcement. A follow-up study by Schmaltz and Isaacson (1966) presented findings analogous to those of Clark and Isaacson (1965)5 indicating that hippocampally lesioned Ss could perform well on DRL schedules if not first subjected to prolonged erf training. The apparently critical role of past learning in the demonstration of hippocampal lesion-induced deficits in discrimination and DRL performance suggested to some that the hippocampus was not involved in the inhibition of behavior in general, but was more specifically necessary for the inhibition of well practiced responses. The demonstrations by Kimble, Kirkby and Stein (1966) and Winocur and Mills (1969) that

PAGE 16

hippocampectomized Ss showed no deficit in their ability to inhibit an unlearned escape response from a small, elevated perch when the response was punished lead to their formal statement of that position. However, Isaacson, Olton, Bauer and Swart (1966) and Teitelbaum and Milner (1963) have presented contradictory data, indicating that hippocampectomized Ss are deficient in withholding a naturally occurring response involving a step-down from a platform to an electrified grid. The former authors, who shook the platform to increase the probability of response occurrence, suggested that the escape response employed by Kimble, Kirkby and Stein (1966) was too weak or improbable in nature to adequately reveal a hippocampal lesion-induced deficit. Inhibitory deficits of hippocampectomized Ss have also been widely examined within the context of exploration and spontaneous alternation paradigms. Roberts, Dember and Brodwick (1962) compared exploration rates of hippocampectomized and control Ss in Tand Y~mazes and found no differences in initial rates, but a more rapid decrease in exploration rate in control than in lesioned Ss. An additional analysis revealed that Ss with small hippocampal lesions showed a moderately, but significant, slower exploration rate decrease than controls, and that S_s with massive hippocampal destruction showed no rate decrease whatsoever. Leaton (1965) studied opportunity for exploration as a reinforcer of a T-maze turning response and found evidence for acquisition in normal and sham operated Ss while hippocampectomized Ss were unable to overcome perseverative tendencies and consequently showed no acquisition effect. Forced training was instituted in the second phase of the experiment and measures of running speed were taken. The hippocampectomized Ss showed slower habituation to the reinforcer, indexed by a slower decline in

PAGE 17

10 running speed over trials than control Ss. Kirkby, Stein, Kimble and Kimble (19o?) examined perseveration of a T-maze response as a function of goal-box confinement. With short confinement periods (50 seconds) hippocampal lesioned Ss showed perseverative behavior while control Ss spontaneously alternated their responses on successive trials. With longer confinement periods (10 and 50 minutes) both hippocampectomized and control Ss demonstrated spontaneous alternation. A supplementary analysis revealed hippocampectomized Ss'perseverate responses per se rather than responses to specified locations. Studies of the effect of hippocampal lesions upon maze learning have yielded rather consistent results. In general, the hippocampal lesion-induced deficit is slight, if present at all, in very simple mazes, but as maze complexity increases the lesion-induced deficit in acquisition becomes increasingly more manifest. These findings have been attributed to the hippocampectomized S_s' inability to inhibit the reentry of previously explored blinds and the greater frequency of blinds in progressively more complex mazes (Kaada, Rasmussen & KviSn, l?6l; Kimble, 19^3; Kimble & Kimble, 1965). Hosteller and Thomas (19&7) have demonstrated that the hippocampal deficits in maze learning cannot be attributed to enhanced thigmotaxis. The hippocampal lesion-induced changes in spontaneous alternation and maze performance suggested to Kimble and his co-workers (Kimble, Kirkby & Stein, 1966; Kirkby, Stein, Kimble & Kimble, 196?) that hippocampectomized .Ss suffer from a reduced rate of information acquisition. This position is incomplete, however, for it fails to account for the unimpaired acquisition rates hippocampectomized Ss demonstrate in alternative learning paradigms.

PAGE 18

11 Although hippocanpectomized Ss appear deficient in their ability to withhold responses in successive or go-no go discrimination problems (Kimble, 1963), numerous studies have demonstrated that they do not differ from control Ss on a vide variety of simultaneous discrimination problems (Allen, 1940, 1941; Brovn, Kaufman & Marco, 1969; Grastyan & Karmos, 1962; Hirano, I966; Kimble, 1963; Kimble & Zack, 196?; Swann, 1934, 1935; Teitelbaum, 1964; Webster & Yoneida, 1964). When hippocampectomized Ss are required to reverse such a discrimination, a pronounced deficit in shifting responding from that which was previously reinforced to that which is newly reinforced is regularly observed (Brown, Kaufman & Marco, 1969; Kimble 8c Kinble, 1965; Habe, 1963; Stutz & Rocklin, 1968; Swanson & Isaacson, 1967; Teitelbaum, 1964; Thompson & Langer, I963). In an attempt to explain the changes observed in positively reinforced behavior following hippo camp ectomy in terms of the loss of a single process contributing to such behavior in the intact organism, Douglas and Pribram (1966) developed a sophisticated neurophysiological theory of "problem solving." Although the authors were initially concerned with the hippocampus, they found it necessary to include in their theory a second limbic system structure, the aniygdala, in order to account for the behavior of which hippocampectomized Ss are capable. Each of these structures is postulated to be intimately involved in two distinct processes underlying problem solving or discrimination learning: the hippocampus-centered "error-evaluate" process and the complementary amygdala-centered "reinforce-register" process. The terms are indicative of the function of each: the reinforce-register process is depicted as

PAGE 19

12 increasing the future probability of a response which has been followed by reinforcement; the error-evaluate process is postulated as decreasing the future probability of a response which has not been followed by reinforcement. During discrimination learning in the intact organism both these processes or systems are cooperative as behavior is brought under stimulus control. The proposed neuronal system underlying the error-evaluate process involves hippocampally mediated inhibition in a Renshaw-like mechanism within afferent systems which serves to "gate out" nonreinforced stimuli. In the absence of the hippocampus non-reinforcement cannot alter behavior and discrimination learning must be accomplished by the remaining reinforce-register system. The effect of reinforcement termed "impellence" is incremental over reinforced training, constant in size, and related to the magnitude of reinforcement and the effort required for its production. At the primary neuronal level, impellence is depicted as involving normally occurring collateral inhibitory processes in afferent systems. The work of Dews on, Hobel and Pribram (1966) and Spinelli and Pribram (1966) is taken as direct evidence for the existence of these proposed systems. To summarize the Douglas-Pribram theory: It has been suggested that the hippocampus is a key structure in an error evaluating system which mediates the effect of non-reinforced responses during learning. Organisms with hippocarapal disruption are rendered relatively insensitive to the effects of non-reinforcement and are therefore required to learn appetitively motivated tasks via the remaining reinforcement

PAGE 20

13 sensitive amygdaloid system. Although the theory is a posteriori in construction, Douglas and Pribram (1966) do present some data confirming predictions made from the theory. The present experiment focuses upon the hippocampus and its proposed involvement in situations involving non-reinforced responding. An experimental paradigm in which manipulation of reinforcement and non-reinforcement contingencies generates differential predictions concerning the behavior of hippocampectomized rats has been developed from the theory in question. In both the acquisition and reversal phases of a position discrimination, equation of the absolute number of reinforced responses to each of two to-be-discriminated manipulanda combined with differentiation between the two in terms of the absolute number of non-reinforced responses would be predicted from the theory to retard both acquisition and reversal in hippocampectomized Ss when compared to neocortically-lesioned and sham operated controls. However, when the absolute number of non-reinforced responses to the manipulanda are equated and the number of reinforced responses differ, any hippo campal lesion-induced deficit would be predicted to be of a significantly lesser magnitude if present at all. Positive results would constitute support of the Douglas-Pribram theory (Douglas, personal communication, 1968).

PAGE 21

METHOD Subjects The Ss were 60 male Long-Evaus rats approximately 125 to 1?5 days old. at the start of training. Apparatus A total of four experimental chambers were employed. One was constructed in the laboratory while the remaining three were commercially obtained. The chamber constructed in the laboratory was a converted ice chest with a sheetmetal partition dividing it into two compartments. One compartment contained a pellet dispenser and related reinforcement delivery equipment; the second compartment, with the inclusion of a hardware cloth floor, measured 28.5 mm. by 28 mm. by 23 mm. high and served as the experimental space. A Ralph Gerbrands Company rat lever was situated along the verticle center line of one wall, 2.25 mm. above the hardware cloth floor. Reinforcement was delivered to a food cup situated 5 Jnau above the manipulandum. An exhaust fan provided ventilation, and a 20 VDC bulb located in the center of the ceiling provided illumination during experimental sessions. The commercially obtained chambers were all Lehigh Valley Electronics Model 1316 small cubicles. A metal food cup was located along the verticle center line of one vail and rested on the grid floor. Two Lehigh Valley Electronics Model 1352 rat levers were mounted on the same wall, one on each side of the food cup. The center point of each manipulandum was 3 nw» above the 14

PAGE 22

15 floor and 5 mm. from the nearest side wall. Illumination was provided by a 20 VDC bulb located 2 mm. above the center of the plexiglass ceiling. All manipulanda were calibrated so that a weight of approximately 20 grams would activate the response circuitry. All experimental operations and contingencies. were controlled by automatic electro-mechanical programming equipment. Reinforcement consisted of k$ mg. Noyes rat pellets. A plexiglass cover, measuring 3 mm. by 7 mm. by 14.5 mm. high, was available to cover either manipulandum in the two manipulanda chambers, thereby forcing Ss to respond on the uncovered manipulandum when the conditions of training so required. Experimental Design The 60 Ss were assigned in equal numbers to the 6 cells prescribed by the first two factors of a 3 x 2 x 2 experimental design involving repeated measures as the third factor. Animals subjected to hippocampal, neocortical, or sham lesions (factor A) were assigned to conditions of differentiation training which insured that for each daily session either the number of reinforced responses or the number of nonreinforced responses (factor B) emitted on each of two manipulanda were equal, and were then tested in both the acquisition and reversal (factor C) of a two manipulanda differentiation. Procedure Upon receipt from the supplier all Ss were placed on ad lib food and water. Following recovery from the rigors of shipment a mean base weight derived from five consecutive days weighing was established

PAGE 23

16 for each S_, and Ss were reduced to %$% of these values and maintained at that level for the duration of pretraining. The goal of the pretraining phase of the experiment was to establish in each _S a tar press response free of any procedurallyinduced left or right position preference. To accomplish this pretraining was conducted in the single manipulandum chamber. Subjects were first magazine trained and then shaped to press the manipulandum "by the delivery of food reinforcement. Special care was taken to insure that no S_ received a disproportionate amount of training under erf and low FR reinforcement schedules. The reinforcement ratio was gradually escalated and pretraining was terminated upon each S_'s demonstration of stable responding' under the requirements of an FR 10 reinforcement schedule. Subjects were then returned to ad lib food maintenance. Following recovery of lost weight Ss assigned to the appropriate cells of the factorial design were subjected to bilateral hippocampal removal, bilateral removal of the neocortex overlying the hippocampus, or bilateral sham operations in which the dura overlying the neocortex removed in the neocortical lesions was exposed. Following recovery from surgery a mean base weight derived from five consecutive days weighing was again established for each S_, and Ss were reduced to %$% of these values and maintained at that level for the duration of the experiment. Subjects were then returned to the single manipulandum chamber and retrained to respond under the conditions of the FR 10 reinforcement schedule. With few exceptions reestablishment of control of responding by the FR 10 schedule was accomplished during one session of approximately

PAGE 24

1? kS minutes? duration. In no instance did this retraining require more than three daily sessions. Following completion of retraining in the single manipulandum chamber Ss were advanced to the two manipulanda chancers in which the experimental operations were conducted. During preliminary training in the two manipulanda chambers the right manipulandum was first covered with the plexiglass cover provided for forced training. Responding on the left manipulandum was first maintained by a erf schedule of reinforcement, and then by intermittent reinforcement. The reinforcement ratio was escalated one step following every tenth reinforcement until 10 reinforcements onanFR 10 schedule were delivered. Subjects were then removed from the chamber, the plexiglass cover moved to the left manipulandum, and the preliminary training regimen repeated. Those S_s which failed to earn 10 ?R 10 reinforcements on either manipulandum within a 5-minute period during which that schedulewas in effect repeated the preliminary training regimen the following day. With few exceptions pretraining required no more than one session approximating one hour in duration; in no case were more than 4 daily sessions required. On the day following the completion of preliminary training the experimental procedures were initiated. In discrimination acquisition responses on one manipulandum were reinforced onan FR 5 schedule and responses on the second were reinforced on an FR 9 schedule. Of the 10 Ss in each of the two hippocampal and sham lesion groups, 6 were assigned to one chamber and 4 to a second. Within each group the relationship between manipulandum and reinforcement schedule was counterbalanced. The two groups subjected to neocortical destruction were assigned to the third chamber and the relationship between manipulandum and reinforcement schedule also counter-

PAGE 25

18 of a 5-nimite free choice period during which both manipulanda were exposed for responding and reinforced on the appropriate schedules. At the conclusion of the test period, which served to monitor the formation of the discrimination, the number of responses emitted and the number of reinforcements earned on each manipulandum were recorded. The forced training portion to the experimental session was then initiated. The purpose of forced training differed for each of the two hippocampal , neocortical, arid sham lesion groups. One each of the hippocampal, neocortical, and sham lesion groups was run under the condition prescribing the equation, for each S, of the absolute number of reinforced responses emitted on each manipulandum during each daily session. The second hippocampally, neocortically, and ; lesioned groups were run under the conditioning prescribing the* equation, for each _S, of the absolute number of non-reinforced responses emitted on each manipulandum during each daily session. It should be noted that as a result of the utilization of an FR 5 and an FR 9 schedule of reinforcement, Ss which emitted an equal number of reinforced responses on each manipulandum also emitted twice as many non-reinforced responses on the FR 9 manipulandum as on the FR 5 manipulandum. Conversely, Ss which emitted an equal number of non-reinforced responses on the two manipulanda also emitted twice as many reinforced responses on the FR 5 manipulandum as on the FR 9 manipulandum. Subjects assigned to the reinforced responses equated procedure fulfilled a dual requirement during each complete experimental session. These requirements were: (a) each S. earn an equal number of reinforcements

PAGE 26

19 on the FR 5 and PR 9 manipulanda, and (b) a total of at least 50 reinforcements be earned on each of the two manipulanda. If S. did not earn the minimum 50 reinforcements on either manipulandum during the 5-minute free choice period, the forced training portion of the session involved responding on "both manipulanda. At the end of the free choice period the required number of make-up reinforcements to be earned on each manipulandum was determined, one manipulandum was covered, and S_ was allowed to respond on the second until the required number of make-up reinf orcemsrts for that manipulandum had been delivered. The cover was then moved to the second manipulandum and _S was allowed to respond on the first until the requirements for that manipulandum had been fulfilled. For example, if £ earned 40 FR 5 reinforcements and 25 FR 9 reinforcements during the free choice period, S_ would be required to earn an additional 10 FR 5 reinforcements and 25 FR 9 reinforcements during the forced training period. As a result, £ would have earned an equal number of reinforcements (50) on each manipulandum during the course of the experimental session. The order in which the manipulanda were covered alternated across daily sessions. If S_ earned 50 or more reinforcements on either, or both manipulanda during the free choice period, the forced training period involved responding only one manipulandum. At the termination of the free choice period the difference between the number of reinforcements earned on the two manipulanda was determined, and the manipulandum on which S. had earned the greater number of reinforcements was covered. The S_ was then allowed to respond on the second manipulandum until the

PAGE 27

20 required ake-up reinforcements vas delivered. For example, if S_ earned 65 PR 5 reinforcements and 20 FR 9 reinforcements during the free choice period, the FR 5 manipulandum would he covered during the forced training period and S. would he allowed to respond on the FR 9 manipulandum until ^5 reinforcements had heen delivered. This fulfilled the requirement that S emit an equal numher of reinforced responses, in this instance 65, on each manipulandum during each experimental session. Subjects assigned to the non-reinforced responses equated procedure fulfilled a different dual requirement during each experimental session. These requirements were: (a) each S earn twice as many reinforcements on the FR 5 manipulandum as on the FR 9 manipulandum, and (h) a minimum of 60 reinforcements he earned on the FR 5 manipulandum and, consequently, a minimum of 30 reinforcements he earned on the FR 9 manipulandum. The free choice period and suhsequent forced training proceeded in a manner analogous to that descrihed ahove for the S_s assigned to the reinforced responses equated regimen. In the present condition, if S_ failed to earn the minimum 60 and 30 reinforcements on hoth the FR 5 and FR 9 manipulanda, respectively, during the free choice period, the forced training period would insure that these minima were earned. For example, if S_ earned kO FR 5 reinforcements and 25 FR 9 reinforcements during the free choice period, he would "be required to earn an additional 20 FR 5 reinforcements and 5 FR 9 reinforcements during the forced training period. The S_ would have therefore earned the minimum 60 and 30 reinforcements on the FR 5 and FR 9 manipulanda during the course of the experimental session.

PAGE 28

21 If £ earned more than 60 reinforcements on the FR 5 manipulandua and/or more than 30 reinforcements on the FR 9 manipulandum during the free choice period, the forced training period involved only one manipulandum and served to insure that the ratio of reinforcements earned of the FR 5 manipulanduin to those earned on the FR 9 manipulandum was 2 to 1. For example, if S earned ?0 FR 5 reinforcements and 40 FR 9 reinforcements during the free choice period, S. was required to earn an additional 10 FR 5 reinforcements during the forced training period. As a result, S earned a total of 80 FR 5 reinforcements and 40 FR 9 reinforcements during the course of the experimental session, and fulfilled the requirement that the ratio of FR 5 to FR 9 reinforcements he 2 to 1. If S earned 90 •R 5 reinforcements and 10 FR 9 reinforcements during the free choice period, £ uas required to earn an additional 35 FR 9 reinforcements during the forced training period. The S. therefore earned a total of 90 FR 5 reinforcements and k$ FR 9 reinforcements, and the ratio of FR 5 to FR 9 reinforcements was again the required 2 to 1. Discrimination training was terminated when S_ attained a criterion of at least J0% responding on the FR 5 manipulandum in 9 of 10 consecutive free choice periods. Upon completion of this requirement S. vas subjected to one half again as many daily sessions as were required for attainment of the criterion and then moved to the discrimination reversal phase of the experiment. If it became statistically impossible for S. to satisfy the criterion within hO days of training S, was considered to have failed to satisfy the requirements for discrimination and was then moved to the discrimination reversal phase of the study.

PAGE 29

1Z Discrimination reversal training was instituted for each S on the day following termination of the acquisition portion of the experiment* In reversal training the relationship "between reinforcement schedule arid manipulandum was reversed for each S_. Training in reversal proceeded in the same fashion for the two groups as described in the acquisition phase above. Reversal training was terminated when each S net either the criterion of acquisition or failure established for the acquisition phase of the study. J. All Ss were operated under kO nig/kg Nembutal anesthesia supplesented with .30 cc. of atropine injected interperitoneally. All operations were performed while S_ was held in a Baltimore stereotaxic instrument. A dissecting scope was employed to assist in the visual guiding of neocortical and hippocampal removal. In all operations the skull was exposed by means of a midline incision, bilateral trephine holes were placed lateral to the midline and posterior to bregma. The holes were enlarged with rougares to expose the neocortex overlying the dorsal and lateral portions of the hippocampus. In the sham operated Ss surgery was terminated at this point. For those Ss sustaining neocortical removal the dural was cut and the neocortex overlying the hippocampus was aspirated off, with care taken not to damage the hippocaapus. For those Ss subjected to hippocampectoay the operation proceeded until the thalamus was exposed. In addition, hippocampal removal was expended anter-iorally as far as the hippocampal commissure as well as extended around the posterio-lateral surface of the thalamus. Care was taken not

PAGE 30

23 to damage the thalamus. Following completion of surgery Ss were returned to their home cages and maintained on a water and tetracycline solution for three to five days. Histology Following the termination of the experiment all Ss were sacrificed with a lethal dose of Nembutal anesthesia and intracardially perfused with saline followed "by a 10fo fornalin solution. All trains were then reared from the "brain cavities and those of the Ss in the neocortical and hippocanipal groups were infiltrated with, and eahedded in, celloidin an! sectioned at 15 u. Ever/ tenth section was retained, :. . ted on a slide and stained with thionin. In addition, for 5 Ss in ezch lesion group the section following the thionin section was stained with vile and then mounted.

PAGE 31

RESULTS Tracings of representative cross sections through the hippocanpal and neocortical lesions are presented in Figures 1 and 2, respectively. Hippocampal destruction regularly involved at least 7& of that structure and in all instances resulted in the complete separation of the dorsal and lateral aspects of the hippocampal formation. Specific damage to the thalamus was minimal and, when evident, was typically unilateral in nature. The neocortical lesions did not encompass a volume of tissue comparable to that removed in the hippocampal lesions, but did approximate the neocortical destruction incurred by the hipp . i resulting from the neocortical lesions was minimal end, if present, usually unilateral in .na.tu.re. Gross examination of the intact brains of those Ss subjected to sham operations revealed no discernible neocortical dan : ?e , The sessions required to attain criterion in acquisition and reversal by those sham, neoeortically, and hippocampally lesioned Ss trained under the reinforced responses equated regimen are presented in Table 1. The trials to criterion for Ss assigned to the non-reinforced responses equated condition are presented in Table 2. Inspection of these tables suggests there is an inverse relationship between performance in acquisition and reversal. It appears that Ss who readily attain criterion in acquisition are retarded in reversal and, to a lesser degree, Ss who are retarded in acquisition appear to perform well in reversal. To test the possibility of such an inverse relationship, all Ss were ranked 24

PAGE 32

25 u »«^ i«iS~t^'^^^'! n 22!.«Jgar «™*«» ino and Cusiunan, 1967).

PAGE 33

Z6 Fig. 2.— Tracings of representative cross sections through the neocortical lesion (After Pellegrino and Cushman, 1967).

PAGE 34

27 TABLE 1 TRIALS TO CRITERION IN ACQUISITION AND REVERSAL FOR SHAM, NEOCORTICALLY, AND HIPPOCAMPALLY LESIONED Ss ASSIGNED TO THE REINFORCED RESPONSES EQUATED CONDITION Shaia Control Neocortical Lesion Hippocaapal Lesion Acquisition

PAGE 35

28 TABLE 2 TRIALS TO CRITERION IN ACQUISITION AND REVERSAL FOR SHAM, NEOCORTICALLY, AND HIPPOCAMPALLY LE3I0NED Ss ASSIGNED TO THE NONREINFORCED ES EQUATED CONDITION Acauisition Reversal Sham Control 21 10 15 13 9 10 9 10 18 11 12 13 16 24 40 22 32 16 22 Neocortical Lesion I'll 40 13 12 10 10 10 31 9 1? 19 15 13 12 15 20 16 14 13 Hippocanpal Lesion. 14 25 14 14 1?. l? 40 17 14 13

PAGE 36

29 from lov; to high on the number of sessions to attain criterion in acquisition, and from high to low on the number of sessions required to attain criterion in reversal. A Spearman rank order correlation coefficient (r s ) was then computed (Siegel, 1956), and found to be significant (r g = A?, t 4.0567, df = 59 » p <.001). Spearman rank order correlation coefficients were also computed in the same manner for the sham, neocortical, and hippocampal lesion groups (see Table 3), and for the three lesion groups when further divided on the basis of the reinforced versus non-reinforced responses equated dimension (see Table k). The former analysis indicates the inverse relationship between performance in acquisition and reversal is present in only the sha 1 control Ss; the latter analysis reveals that while the sham control Ss under the non-reinforced responses equated condition do show this relationship, their counterparts under the reinforced responses equated condition do not. In addition, the finer grain analysis indicates that the hippocampectoaized Ss under the reinforced responses equated condition also manifest this relationship, albeit to a lesser degree. The sessions required to attain criterion in acquisition and reversal for sham, neocortically, and hippocaapally lesioned Ss under the reinforced responses equated requirement are presented graphically in Figures 3 and 4, respectively. Analogous data for Ss under the non-reinforced responses equated condition are presented in Figures 5 an<* 6, An analysis of variance assessing the effects of the three lesion conditions, the two response-reinforcement contingencies, and acquisition and reversal upon performance as indexed by the trials to criterion measure was performed (Winer, 1962). Since the trials to criterion

PAGE 37

30 TABLE 3 SPEARMAN RANK ORDER CORRELATION COEFFICIENTS (r ) FOR SHAM, NEOCORTICALo AND HIPPO CAMPAL LESION GROUPS* Shan Control Neocortical Lesion Eippocaipal lesion V

PAGE 38

31 TABLE k SPEARMAN RANK ORDER CORRELATION COEFFICIENTS (rj FOR SHAM, NEOCORTICAL, AND HIPPOCAMPAL LESION GROUPS UNDER EQUATED REINFORCED OR NON-REINFORCED RESPONDING DURING TRAINING*

PAGE 39

32

PAGE 40

33 r . . • NOItf31ISD Ol SNOISS3S CO

PAGE 41

3* . CNl •— NOIU3II^D Ol SNOISS3S to

PAGE 42

3>

PAGE 43

% measure produced slightly si ! ta, these values were subjected to a square-root transformation and a second analysis of variance performed upon the resultant data. The results of these two analyses are presented in summary fashion, in Tables 5 and 6, respectively. A comparison of the two tables reveals little inconsistency in the results of the two analyses. An examination of Figure 7, which depicts the cumulative percentage of Ss in each lesion group attaining criterion in successive five session blocks, suggests that the neocortically and hippocampally lesioned _Ss are slightly facilitated with respect to the sham control Ss in discrimination performance. The analyses of variance indicate that this difference is not lcr to be statistically reliable. However, the results of the analyses do reveal a significant interaction between this factor and the acquisition and reversal phases of training which must be examined before it can be concluded that the various lesion conditions have no effect on discrimination learning. A comparison of the effects of equating either reinforced or nonreinforced responses during discrimination training is depicted in Figure 8. Neither level of this factor, nor this factor's interaction with the lesion dimension, were indicated by the analyses of variance as differentially affecting performance in the discrimination task. Inspection of Figure 9, which represents performance in the acquisition and reversal phases of the study, suggests that Ss attained criterion more rapidly in acquisition training than in reversal training, and this is verified as a significant difference by the analyses of variance.

PAGE 44

37 TABLE 5 ANALYSIS OF VARIANCE ON TRIALS TO CRITERION Source

PAGE 45

38 TABLE 6 MALTS IS 0? VARHNCE ON SQUARE-ROOT TRMSFORHATION 0? TRIALS TO CRITERION Source

PAGE 46

3/ © 7 O o

PAGE 47

*0 in

PAGE 48

to w u

PAGE 49

42 As indicated previously , a significant interaction between the lesions factor and the acquisition and reversal phases of training was revealed by the analyses of variance. The components of the interaction are depicted in Figure 10, which presents the performance of the three lesion groups in acquisition training, and in Figure 11, which presents their performance in reversal training. The mean trials to criterion for Ss in each of the three lesion groups is presented for acquisition and reversal in Figure 12. Examination of these figures suggests that the three lesion groups did not differ in the acquisition phase of training, but that in the reversal phase the neocortically and hippocampally lesioned Ss, v;hile not differing among themselves, did attain the criterion more rapidly and in greater numbers than the sham control Ss. A poster iori comparisons between the cell means involved in this interaction were performed utilizing the Studentized range statistic ( T ./iner, 1962). The results of the comparisons are presented, in summary fashion, in Table ?. The results support the above observations and reveal, in addition, that sham control Ss attained criterion significantly faster in acquisition than in reversal, but that such a difference is not present in the neocortically and hippocampally lesioned Ss. Neither the remaining first order interaction (lesions by response-reinforcement contingency) nor the single second order interaction (lesions by response-reinforcement contingency by acquisition-reversal) attained significance.

PAGE 50

43 C-r-a' •r ; ."».---i.';%.T^-/, , . .1 vy

PAGE 51

44 ^O co u O Z g CO CO LU CO UJ > El c:| tnt) o

PAGE 52

*5 z o

PAGE 53

46 TABLE 7 STUDENTIZED RANGE STATISTIC A POSTERIORI TESTS Acquisition Sham vs. Neocortical Lesion 3.09 Sham vs. Hippocampal Lesion 0.86 Neocortical vs. Hippocampal Lesion 0.83 Reversal Sham vs. Neocortical Lesion 20.29** Sham vs. Hippocampal Lesion 13.9^** Neocortical vs. Hippocampal Lesion 0.23 Sham Control Acquisition vs. Reversal 19.77** Neocortical Lesion Acquisition vs. Reversal 0.01 Hippocampal Lesion Acquisition vs. Reversal 2.30 **p<.01

PAGE 54

DISCUSSION This study examined the performance of sham, neocortically, and hippocampally lesioned rats in the acquisition and reversal of a two manipulanda differentiation as affected by certain manipulations of response-reinforcement contingencies. For one-half the sham, neocortically, and hippocampally lesioned Ss, these manipulations insured that each £ emitted an equal number of reinforced responses on the tv/o manipulanda during a complete experimental session. As a result of this procedure, each S_ also emitted tv:ice as cany non-reinforced responses on one manipulandum (the FR 9 manipulandum) as on the second (the FR 5 manipulandum). For the remaining Ss, the experimental manipulations insured that each S_ emitted an equal number of nonreinforced responses on the two manipulanda during a complete experimental session. As a result of this procedure, each S emitted twice as many reinforced responses on the FR 5 manipulandum as on the FR 9 manipulandum. The Douglas-Pribram theory of hippocaoipal function (Douglas & Pribram, 1966% Douglas, 196?) is a vigorous attempt to integrate the wide variety of "behavioral changes following hippocampal disruption, and leads to clear-cut predictions of the behavioral effects of the manipulations performed in this study. Specifically, the theory predicts retarded acquisition and reversal in hippocampal Ss when compared to sham and neocortical control Ss under the treatment condition which specifies equation of the number of reinforced responses emitted 47

PAGE 55

on th3 two manipulanda, less or no retardation in hippocanpal Ss under the condition which served to equate the number of non-reinforced responses emitted on the discriminanda, and, indirectly, no difference between sham and neccortical S_s under either of these two treatment conditions. These predictions are contradicted "by the results of this experiment. The trials to criterion data do not reveal any differences between the three lesion groups in acquisition performance. This finding is in general accord with the results of recent studies of the role of the hippocampus in discrimination learning. However, such studies have not examined the effect of differential densities of reinforced and non-reinforced responding to the alternative discrininanda upon discrimination learning in hippocampectoaized Ss. The present findings, although inconsistent with predictions derived from the Douglas-Pribram model, indicate that such differences have no significant effect on acquisition performance in either sham, neocortically, and hippo campally lesioned Ss. The performance of the three lesion groups in reversal also is inconsistent with predictions derived from the Douglas-Pribram for, as in acquisition, the two response-reinforcement contingencies do not differentially affect in either sham, neocortically, or hippocampally lesioned S_s. In addition, the reversal data appear to be inconsistent with the bulk of the data on the performance of hippocampectomized and control Ss in discrimination reversal; namely the hippocampally lesioned Ss are typically reported as retarded in discrimination reversal when compared to neocortically lesioned and sham control Ss who usually

PAGE 56

4? do not differ from each other. The results of the present study indicate that it is neceortically and hippocaipally lesioned Ss who do not differ from eada. other, and both appear facilitated when compared to sham operated Ss on the trials required to attain criterion in reversal. A comparison cf the sessions to criterion data with alternative performance measures, such as trials to successive criteria and the absolute per cent of FR 5 responding for successive days, revealed that all three measures depicted acquisition and reversal performance in a similar fashj A possible explanation of this disparity is offered by the relationship between acquisition and reversal performance as revealed by the Spearmaa rank order correlation coefficient. The sham operated Ss, who require a significantly' greater number of trials to attain criterion in reversal when compared to the neocortically and hippocaapally lesiossed Ss as indicated by the Studentized range statistic, are also the Ss who manifest a significant inverse relationship between trials to criterion in acquisition and reversal. In addition, a greater number of Ss in the sham control group attained the criterion in 11 or less sessions (15 of 20 Ss) than in either the neocortical (11 of 20 S_s) or hippocaixal (12 of 20 Ss) lesion groups. When the three lesion groups are partitioned in terms of the response-reinforcement contingencies, similar phenomena are observed. These findings suggest that some Ss possess an initial position preference which, when in accord with ths requirements of the initial differentiation, facilitates acquisition and retards reversal. Moreover, it appears that a greater proportion of Ss in the sham operated group fall into this category

PAGE 57

50 than in either of the two other lesion groups. It may he assumed, then, that the poorer performance of the sham control S_s in reversal is an artifact resulting from a failure to completely control for initial position preference, and that if this had heen done the differences between the three lesion groups in reversal would he eliminated. The experimental procedures employed in this study differ in a number of details from those in which' hippocampal lesion-induced deficits in discrimination reversal are commonly observed. It is possible that the lack of a lesion-induced deficit in the present study may be attributable to one or more of these differences. One such modification involves the utilization of an overtraining procedure following attainment of criterion in acquisition. Investigations of the effect overtraining upon reversal in the T-maze indicate that Ss subjected to, on the average, at least 1.3 (Macintosh, 1962) and 3 (Publcs, 1956) times as many overtraining trials as were required to reach criterion in the acquisition of a discrimination perform better in reversal than Ss without overtraining (the overtraining reversal effect). An examination of Reid's (1953) data indicates that when overtraining involves less trials than were required to attain criterion the overtraining reversal effect is not seen. It is difficult to compare those procedures and the present one, for different responses and procedures were employed. However, since this study employed only one-half as many overtraining sessions as were required to attain the acquisition criterion it is unlikely that the overtraining reversal effect was operative. Despite this, an examination of the overtraining reversal phenomenon does add to an understanding of the results of this study. Macintosh (19^5) notes that it is presumably justifiable to regard reversal learning as consisting of two parts: (a) extinction

PAGE 58

51 of a tendency to select the former S D , and (b) acquisition of a tendency to select the new S D . Stage (a), extinction, is usually regarded as continuing so long as S_ scores below chance level, and stage (b), acquisition, is typically depicted as commencing as soon as _S begins to perform above chance level. It should be noted that implicit in the formulation described by Macintosh is the generally accepted assumption that learning is a continuous, rather than discontinuous, process. Whether this is indeed the case has not yet been completely resolved. The present discussion is concerned primarily with the effect of attentional factors upon discrimination reversal rather than with the underlying nature of the learning process. It is often reported that overtraining of a runway response results in reduced resistance to extinction (Wagner, I963). It is logical to assume, therefore, that this phenomenon is what underlies the overtraining reversal effect; namely, overtraining facilitates extinction of responses to the former S D in the formulation described by Macintosh. This is not the case, however, for overtraining in the discrimination paradigm regularly increases resistance to extinction of responses to the former S D (Macintosh, 1962). As Macintosh (19^5) points out, overtraining facilitates reversal of a simultaneous discrimination not because of, but in spite of its effect on extinction. This finding would appear to negate an extension of the frustration (Lawrence & Festinger, 1962) and generalization decrement (Kimble, 1961) explanations of extinction to the overtraining reversal effect and, by implication, to reversal training in general.

PAGE 59

52 Macintosh (19 65) reports that existent evidence indicates overtraining facilitates reversal by shortening runs of incorrect responses during the middle of the reversal. This suggests that overtraining reduces SS 1 tendencies to respond to irrelevant cues during reversal. There are two possible explanations for this: Either overtraining effectively enables Ss to "adapt out" cues along irrelevant dimensions; or overtraining allows ample opportunity f or Ss to learn to attend to the relevant cue dimension. Macintosh presents evidence which indicates that it is the latter alternative which underlies the overtraining reversal effect, and points out a distinction between research utilizing visual and spatial cues. Studies which have involved a simultaneous visual discrimination (brightness, pattern, etc.) invariably produce the overtraining reversal effect, while those studies which employ a spatial discrimination (left turn versus right turn in 5and Y-mazes, etc.) frequently do not. A major reason for this, Macintosh contends, is that the rat (the commonly used experimental organism) is primarily spatially oriented and, as a consequence, spatial cues have a high priority even without overtraining. Since the rat is already attending mainly to spatial or position cues, overtraining would not be expected to have much effect on performance in reversal. Conversely, the lower the relevant cue dimension is on the Ss "attending hierarchy," the more valuable overtraining would be expected to be in firmly establishing the relevant cues in a position of dominance. The magnitude of the overtraining reversal effect should be inversely related to the probability that S will attend to the relevant cue at the beginning of discrimination training, and to the number of irrelevant cues involved in the discrimination.

PAGE 60

53 The preceding discussion is markedly similar to the DouglasPribram conceptualization of the role of the amygdala in discrimination learning; namely, the registration of the effects of reinforcement or, alternatively, the direction of attention to the aspects of the task (relevant cues) associated with reinforcement. Hovever, the above formulation of the function of attentional factors in discrimination learning does, not incorporate a process analogous to that attributed to the hippocampttss the gating out of stimuli associated with non-reinforcemerit. It will be recalled that an alternative to the attentional model indicates that the overtraining reversal effect can be attributed to the opportunity for Ss to effectively "adapt out" irrelevant cues (Spence, described in 1 h, 1965)0 Perhaps an explanation of the overtraining reversal effect involves both these processes. The Douglas-Pribras model suggests that this is so. In addition, the results of this study may not be as inconsistent with "the DoublasPribram model as was first indicated. The differentiation required in the present study involves spatial cues, a dimension thought to be high on the rat's "attentional hierarchy." If this is the case, the role of the hippocampus, that of gating out irrelevant stimuli, would be minimal in the intact _S, and Ss without the hippocampus should not be greatly impaired* Perhaps a hippocampal lesion-induced deficit would be evident in the present paradigm if a task involving a non-spatial differentiation, or, more probably, a differentiation between a large number of equally salient c^es was employed. Support for this possibility is provided by Pribras (I969), who reports that hlppocaapectonized monkeys show retardation in discrimination learning, provided there are

PAGE 61

a large number of non-rewarded alternatives in the situation (PS137). A second procedural innovation employed in the present study involves the response selected forstudy. Those studies, reported previously, which have demonstrated the hippocampal lesion reversal deficit have typically employed an instrumental response requiring some form of gross locomotion on the part of S. In contrast, this study utilized an operant response, a bar press, which, unlike the typical instrumental response, requires a minimum of locomotion, takes a short time to execute, requires relatively little effort, and leaves S in the seme place ready to respond again. Although it is generally assumed that the behavioral principles and neurophysiological mechanisms underlying what appear to be analogous tasks in the two experimental approaches do not differ in any critical aspect, a thorough comparison of these two procedures has not yet been attempted. However, a recent study by Means, VJalker, and Isaacson (1969) indicates that the effect of hippocampal disruption upon go-no go performance may be response-specific. Although it is typically reported that hippocampectomy interferes with this behavior when examined in an instrumental parad^ such as an alleyway (Brown, Kaufman & Marco, 1969), Means et al . report that hippocampal ablations facilitate performance in this task when a bar press response is utilized. Findings such as these question the trans-situational nature of the pattern of behavioral disruption observed following hippocampal destruction and, consequently, any formulation which attempts to account for these effects with global concepts

PAGE 62

55 such as "perseveration," "gating, 11 or "inhibition" without further refinement or qualificatic , The third major difference between this and contemporary investigations of the role of the hippocampus in discrimination learning involves the schedules of reinforcement associated with the two to-bediscriminated responses. The research reported previously has typically pi Lded c nous reinforcement for "correct" responses and withheld reinforcement for "incorrect" responses. In the present study concurrent operants were utiliz-ed: Both responses were reinforced, one on an PR 5 schedule and the other on an PR 9 schedule, and the formation of the discrimination was based on a relative, rather than absolute, differential in reinforcement density. The experimental analysis of concurrent ratio schedules indicates that with unequal PR requirements, responding tends to be maintained only by the schedule with the smaller FR requirement; with equal PR requirements, responding cam be maintained by eitherone, and shifting from one schedule to the second occasionally occurs (Catania, 19&>» Herrnstein, 1958). In a study which is only superficially comparable to the one reported here, Douglas and Pribram (1966) examined the effects of probabilistic reinforcement upon the formation of a discriminated panel press in monkeys. As in the present study the discrimination rested upon a relative differential in reinforcement density; one response was reinforced ?0% of the time and the second reinforced 30$ of the time. Their results 8 in contrast to those of the present study, indicated that hippocaapectomized Ss are retarded with respect to control Ss

PAGE 63

56 in their ability to acquire a discrimination under such conditions. Unfortunately, no data are presented on the performance of these Ss in discrimination reversal. The reasons for these apparently contradictory findings are unknown, but these studies reveal that insufficient attention has been directed towards an elaboration of the effects of schedules of reinforcement on discrimination formtion and reversal in hippocampectomized Ss. Another difference between this and other studies of hippocanpal function involves the spacing of test trials. Most research in this area has employed a discrete trial procedure and the related technique of massed training trials during each daily experimental session. In the present study the equivalent of test trials, the five minute freechoice periods, vera widely spaced for they occurred at the beginning of each daily experimental session. No direct evidence is available concerning the effect of this factor on discrimination learning and reversal in hippocampectomized Ss. However, there is evidence that the interval between trials does influence the behavioral effects of hippocaapal disruption. As reported previously, Kirkby, Stein, Kimble and Kimble (196?) have demonstrated that the lack of T«maze spontaneous alternation commonly reported in hippocampal Ss can be reestablished by lengthening the intertrial interval from 50 seconds to 10 minutes. Although their explanation of this phenomenon, a postulated lesioninduced reduced information acquisition rate, has been generally abandoned on the premise that such preparations do not show deficits in a number of alternative learning tasks, no adequate explanation has been

PAGE 64

57 f lated. Little or no additional research has been directed toy an understanding of this finding, and until this phenomenon is investig 3 in greater detail such an explanation of the results of the present experiment cannot he fully evaluated. A fifth major departure of this experiment relative to previous I rch is tfte utilization of a forced training technique. As a function of fulfilling the requirements of the response-reinforcement contingencies, this procedure insured that each S_ was fully exposed to the conditions of reinforcement throughout "both acquisition and reversal training. In addition, it can he assumed that this innovation most prohahly maintained the strength of the IB. 9 response at a relatively higher level than discrimination studies which have not employed forced training on 9 and reinforcement o£ the "incorrect" response. Isaacson, Olton, Bauer and Swart (1966) present evidence indicates that the hippocampally lesioned S_ e s inability to withhold a response in the passive avoidance task is directly related to the strength of that response. It is also possible that the ease with which hippocampe eternized Ss can inhibit one response and initiate an alternative is dependent upon the relative strength, or probability, of those two responses. These findings, which demonstrate that hippocampal Ss are capable of inhibiting an established response and initiating another, stand in marked contrast to the bulk of the data on the performance of suc^h Ss when faced with similar tasks. It is not surprising that task variables, some of which have been discussed above, have the potential to profoundly influence the behavioral effect of physiological manipulations. What is surprising is that no concerted effort bas been made

PAGE 65

5* to explain such findings within ths contexts of present formulations of hippocampal function, or to revise these formulations so that they may incorporate these results. All too often findings such as have been discussed here are neglected or dismissed as aherrent. Perhaps a detailed examination of the manner in which experimental manipulations can change or counteract the effects of physiological manipulations will provide increased insight into the role of neurophysiological systems in ths intact organism. The unexpected facilitation of discrimination reversal perforcance resulting from the neocortical damage sustained "by the control Ss is most likely attributable to uncontrolled position preferences, as vac discussed previously. Other experimentation on the effects of hippocampal ablation has typically involved analogous neocortically lesioned control Ss, and has regularly reported that such Ss do not differ from their unoperated counterparts. There are exceptions to this however, for Means, ot al . (19^9) have found that destruction of the neocortex overlying the hippocampus leads to a retardation of performance in the go no-go task; and Olton and Isaacson (I967) have reported that damage of this area, as well as this area plus the hippocampus, lengthens response latencies in avoidance and escape tasks. In the present experiment neocortical destruction involved considerable portions of the rat neocortex comparable to Broadman's area 7, which is involved in somesthesis, particularly the integration of information on weight and the state of muscles and joints; areas 1?

PAGE 66

59 and 18, the visual projection and association areas, respectively; area 25, the entorhinal cortex; and area 3?, which receives soaesthetic and optic association fibers and, in man, is thought to "be involved in the recognition of "body image, individuality end continuity of personality, and of the self in relation to the environment (Kreig, 1957) • Since a considerable portion of the hippocampal research has involved some damage of these areas it is possible that commonly observed hippocampal lesion deficits are in actuality a function of an interaction of the hippocampus and the neocortex which overlies it. Within this context, Douglas (1?6?) has observed that electrolytic lesions restricted to the hippocampus frequently do not produce the deficits seen in ablation studies involving neocortical destruction. It is also possible that the facilitated performance shown by the hippocampectosized Ss in the present study is fully accounted for by the effects of neocortical destruction. Questions such as these point to the relative prinitiveness of our understanding of the role of the hippocampus in behavior, and to the importance of further research in this area.

PAGE 67

} Adey, W. R. Studies of Mppocampal electrical activity during approach learning. In J. I'. D2l?.fresnaye (Ed.), Brain mechanisms in learning . Oxford; Blackwell, 1961, pp. Sn-S&. Allen, V. E. Effect of ablating the frontal lobes, hippocampi, and occipito-parieto-temporal (excepting pyrifors areas) lobes on positive and negative olfactory conditioned reflexes. American Journal of Physiology , 1940, 128, ?54-7?l. Allen, \L F. Effect of ablating the piroform-amygdaloid areas and hippocampi on positive and negative olfactory differentiation, Ageri can Journal o^.P^y,siol£gv « 1941, 132., 81-92. Boitano, J. J., & Isaacson, R. L. Effects of variation in shockintensity on the behj ..'::• of dorsal-hippocampectomized rats in two passive avoidance situations. American Journal of Hiy^siology, 1966, 80, 73-80. Brown, T. S., Kaufman, P. G. , & Marco, L. A. The hippocampus and response perseveration in the cat. Brain Research, 1969, 12, 86-98. ' ' Cajal, S. R. I. Studies on the cerebral cortex. Vol. 11. Limbic s true tares (Trans, by i. H. Kraftf. Chicago: rear Book, 1955. Catania, C. A. Concurrent operant. In W. K. Honig, (Ed.), Operant behavior; Areas of research and appJULcatipn. New York: App] ituryCrofts, 1966, pp. 213-270. Clark, C. V. H. , & Isaacson, R. L. Effect of bilateral hippocampal ablation on DHL performance. Journal of Comparative and Physiological Psychology . 196£T]>2» 137-140. Correll, R. E. The effect of bilateral hippocampal stimulation on the acquisition and extinction of an instrumental response. Journal of Comparative and Physiological Psychology . 1957, j>0, 62^629". Dewson, J. E.* Ill, Nobel, K. \U & Pribram, K. H. Corticofugal influence of cochlear nucleus of the cat: Some effects of ablation of insular-temporal cortex. Brain Research, 1966, 2, 151-159. . ~ Co

PAGE 68

61 Douglas, R. J. The hippocampus and behavior. Ps ychologica l Bulletin, 196?, 67, 416-44 Douglas, R. J., & Pribram, K. H. Learning and limbic lesions. ££HS22S£SJi2l£aL§:» i?^, iL> 197-220. Ellen, P., & Wilson, A. S. Perseveration in the rat following hippocampal lesions. Experimental Neurology , 1963, 8, 310-31?. Feldman, S. Heurophysiological mechanisms and modifying afferent hypothalamo-hippoeampal conduction. Experimental Neurology, 1962, £, 269-291. ' Gloor, P. Amygdala. In J. Field (Ed.) Handbook of physiology. Vol. 2. Neurophy si ology . Washington, D. C: American Physiological Society, I960, pp. 1395-1420. Grastyan, E. & Karmos, G. The influence of hippocampal lesions on simple and delaying instrumental conditioned reflexes. In P. Passouant (Eel.), Physiologia de l'hlppocampe, 19 62, pp. 225-239. Green, J. D. The hippocampus. Physiological Review, 1964, 44, 561-603. Green, J. D. & Adey, W. R. Electrophysiological studies of hippocampal connections and excitability. Electroencephalography and Clinical Neurophysiology,. 1956, 8, 245-262. Green, J. D., & Arduini, A. Hippocampal electrical activity in arousal. Journal of Neurophysiology . 1954, 37, 533-557Herrnstein, R. J. Some factors influencing behavior in a tuo-response situation. Transactions of the New York Academy of Sciences, 1958, 21, 35~$57~ — — — Hirano, T. Effects of hippocampal electrical stimulation on memory consolidation. Psychologia, 1966,
PAGE 69

6*2 I son, R. L.j & Wickelgren, W. 0. Rippocarapal ablation and pas avoidance. Science, 1962, 1/33, 1104-1106. Jarrard, 1. Ee Behavior of hippocampal lesioned rats in hone cage and novel situations. Physiology and Behavior , 1968, j}, 65-70. Jarrard, L. E. & Isaacson, R. I. Hippocampal ablation in rats: Effects of intertrial interval. Nature, 1965, 20?, 109-110. J d, L. E. , Isaacson, R. I., & Wickelgren, tf. 0. Effects of hippocampal ablation and intertrial interval on runway acquisition and extinction. Journal of Cc t arative and Physiological Psychology, 1964, j£, 442A'-'; .. Kaada, B. R., Rasinussen, E. •. , & Kyi;::, 0. Effects of hippocampal lesions on maze learning and retention in rats. Experimental ffeurology. 1961, %, 333-3551, J. II. Inhibitor/ reactions in rats with bilateral hippocampal lesions. Pro c \ rf the 75th Annual Convention of the America n PsycSTo^./ Association, 196*7, 2, 73^ph, 3, D. P. The effects of bilateral hi 1 lesions in rats. Journal of Comparative and Physiological Psychology , 19o3» i£, 273-283. Kinble, D. P., a Kimble, R. J. Hippocaspectomy and response perseveration in the rat. Jc 1 ' Comparative and Physiological Psychology, 1965, 6oT Kinble, D. P., & Zac>, S. Olfactory discrimination in rats with hippocampal lesions. Psychonomics Science , 1967, 8, 211-212. Kiahle, D. P., Kirkhy, R. J., & Stein, D. G. Response perseveration interpretation of passive avoidance deficits in hippocampectonized rats. Journal of Comparative and Physiological Psychology, 1966, 61, 141-143. Kinble, G. A. Hilgard and Marquis 1 conditionin g and learning . New York: Apple ton-Century-Crofts, Inc., £90*17 Kisaara, D. Effects of selective hippocampal damage on avoidance behavior in the rat. Canadia n Journal of Psychology , 1958, 12, 213-218. Kirkby, R. J., Stein, D. G. , Kinble, Reeva J., & Kimble, D. P. Effects of hippocampal lesions and duration of sensory input on spontaneous alternation. Journal of Co mparative and Physiological Psychology, 1967, 64, 342~34~5.

PAGE 70

(3 Kluver, H., & Bucy, P. C« Prelimi] . lysis of functions of the temporal lob:?. 1 aonkeys. Archives of Neurology and Psychiatry, 1939, £2, 979-1000. ™"~ Krieg, W. J. S, Brain mechanisms indiachrome . Bloomington, Illinois: PantagrapE Prinfing C0.7 1957. Kviem, 0., Setekliev, J., & Kaada, B. R. Differential effects of hippocampal lesions on maze and passive avoidance learning in rats » -Experimental Neurology . 1964, £, 59-7Z* Lawrence, D. H. & Festlnger, L. Deterrents and reinforcement. The psychology of insufficient reward. Stanford: Stanford University Press, WSz". leaton, R. 21. Exploratory "behavior in rats with hippocampal lesions. Journal of Comparative and Physiological Psychology, 1965, 59, 325-330. Macintosh, N. J. Selective attention in animal discrimination learning. Psychological Bulletin , 1965, 6£, 124-150. Macintosh, N. J. The effects of overtraining on a reversal and a nonreversal shift. Journal of Cjur^rative and Physiological Psychology, 1962, jg, 555-559 . Maclean, P. D. The limbic system and its hippocampal formation: Studies in animals and their possible application to man. Journal of Neurosurgery, 1954, 11, 29-44. Maclean, P. D. The limbic system ('visceral brain') and emotional behavior. Archives of Neurological Psychiatry, 1955, 7^» 130-13*. Maclean, P. D. Chemical and electrical stimulation of the hippocampus in unrestrained aninals. American Medical Association Archives of Neurology and Psychiatry ^ 1957, £8, 113-142. Maclean, P. D. Contrasting functions of limbic and neocortical systems of the "brain and their relevance to psychophysiological aspects of medicine. American Journal of Medicine, 1958, 2£, 611-626. McNew, J., & Thompson, R. Sole of the limbic system in active and passive avoidance conditioning in the rat. Journal of Comparative and Physiological Psychology, 1966, 61, 173-TBb. Means, I. W., Walker, D. W. & Isaacson, R. L. Facilitated single alternation go-no go performance following hippocampectomy in rat. Paper presented at the meeting of the Psychonomic Society, St. Louis, November, 1969.

PAGE 71

64 Meissner, 17. Hippocampal functions in learning. Journal of Psychiatric Research, 1966, 4, 235-304. ~~ ~~*~ ~~" Niki, H. The effects of hippocampal ablation on the behavior in the rat. Japanese Psychological Research . 1962, 4, 139-153. Niki, H. The effects of hippocaapal ablation on the inhibitory control of operant behavior in the rat. Japanese Psychological Research , 1965, 7, 126-137. ™ " ~ £lz ~~ Olton, D. S. & Isaacson, R. L. Hippocampal lesions and active avoidance. Physiology and Behavior, 1967, %» 719-.724. Papez, J. tf. a proposed mechanism of emotion. Archives of Neurology and Psychiatry. 193?, ,33, 725-743. ~~~ — — Pellegrino, L. J. & Cushman, A. J. A stereotaxic atlas of the rat brain. New York: Appleton-Century-Crof ts , 19677""""" Peretz, E. Extinction of a food-reinforced response in hippocampectomized cats. Journal of Comparative and Physiolog ical Psycho" 19^5, 607T82-I85. — J — — Pribram, K. H. The amnestic syndromes: Disturbances in coding? In E. R. John (Ed.), The patholoj iry. New York: Academic Press, 1969, pp. 127-15?. Pubols, B. H., Jr. The facilitation of visual and spatial discrimination reversal by overlearning. Journal of Comparative and Physiological Psychology . 1956, jV-T, ..." ~~" — '— — Rabe, Ausma. Discrimination reversal deficit from hippocampal stimulation in the rat. Journal of Psy chological Studies, 1963, 14, 139-150. — — l_&_ — _. 7 ^, __, Raphelson, A. C, Isaacson, R. L. s & Douglas, R. J. The effect of limbic damage on the retention and performance of a runway response. Nej^usy_cholO£3^a, I966, 4, 253-264. Reid, L. S. The development of noncontinuity behavior through continuity learning. Journal of ^^j^^toljsj/cholo^i, 1953, 46, 107-112. Roberts, T ,f. W., Denber, W. N., & Brodvick, M. Alternation and exploration in rats with hippocampal lesions. Jou rnal of Comparative £5L^ZSl£l£^|caLl§Z£M2fSLt 19&, Hi 695-700. Schmaltz, L. w., & Isaacson, R. L. Effect of bilateral hippocampal destruction on the acquisition and extinction of an operant response. Physiology and Behavior . 196?, 2, 291-298.

PAGE 72

65 £ tltz, I., & Isaacson, R. L. The effects of preliminary training conditions under DHL performance in the hippocampectomized rat * Miysiology and Behavior . 1966, _!, 175-182/ 1, S. Fcrrr.r-."8tric stati stics for the behavioral sciences . New York: kcGraw-Hill, 1956™ — — — Splnelli, D. I., & Pribram, K. H. Changes in visual recovery functions produced by temporal lobe stimulation in monkeys. Electroencephalography and Clinical Neurophysiology. 1966,~20* 'kkJ$* ~ ———*-» — St pf, c. H. Drug action on the electrical activity of the hipoocampus* International Revie w of Neurology, 1965, 8, 77-138. St . 5, R. H., & Rocklin, K. W. Cingulate and fornix lesions: Effects on twotypes of reversal learning. Journal of_ Comparative and Physiological Psychology, 1968, 6£, 520-523. S s H. G. Sis function of the brain in olfaction. The effects of large cortical lesions on olfactory discrimination. American Journal of Physiology , 1935, lll f 357-262. Swaim, E. G. •be function of the brain in olfaction. II. The results of destruction of olfactory and other nervous structures upon the discrimination of odors. Journal of Comparative Neurology, 1934* i£, 175-201. — — — Sv son., A. H. s & Isaacson, R. L. Hippocampal ablation and performance during srithdrawal of reinforcement. Journal _of Comparative and Physiological Psychology . 1967, 64, in press. Teitelbaum, H«A comparison of the effects of orbitofrontal and hippocampal lesions upon discrimination learning and reversal in the cat. feperinental neurology, 1964, £, 452-462. Teltelbaum, H. s & Milner, P. Activity changes following partial hippocampal lesions in rats. Journal of Compa rative and Ph ysiological Psychology. 1963, j?6, 284-289. on, R., & langer, S. K. Deficits in position reversal learning folloulrg lesions of the limbic system. Journal of Comparative and Ph ysiological Psycholog y, 1963, j>6, 987-995. Vagasr, A. R. Overtraining and frustration. Psychological Reoorts, 1963, 12, 717-718. -— " ' ~"~" Webster, P. B.» & Voneida, T. J. Learning deficits following hippocampal lesions in snlit-brain cats. Experimental Neurology, 1964, 10, 170-182. ~~*~~ ~ ~~"

PAGE 73

t' Winer, B. J. St : . ' principles in experimental desig n. New York: McGraw-Hill, lf6zT~ VTinocur, G. , & Mills, J. A. Hippocampus and septum in response inhibition. Journal of Comparative and Physiological Psych ] 1^9tiZ.» 352"35A '

PAGE 74

BIOGRAPHICAL SKETCH Michael Arnold Milan was torn on January 25, 1938, at New York, New York. He attended public school in New York state and graduated from Fayetteville-Manlius High School, Fayetteville, New York, in June, 1956. In May, 19 60, he received the degree of Bachelor of Arts from Syracuse University. From i960 until 19^2, he served in the Adjutant General Corps of the United States Army. Following his release from the Army he was employed as a psychological research assistant with the Veteran's Administration Hospital in Syracuse, New York. In September, 19^3, he enrolled in the graduate school of the University of Florida and received the Master of Arts degree in December, 1965* He served as a research assistant for Dr. H. S. Pennypacker and as an interim instructor with the Department of Psychology while matriculating for the Doctor of Philosophy degree in psychology. 6?

PAGE 75

This disseration was prepared under the direction of the chairroan of the candidate's supervisory committee and has been approved "by all members of that committee. It was submitted to the Dean of the College of Arts and Sciences and to the Graduate Council, and was approved as partial fulfillment of the requirements for the degree of Doctor of Philosophy. June, 19?0 Dean, College ofi Arts sm\ Sciences Dean, Graduate School

PAGE 76

7714B