Inactivation of hippocampus or caudate nucleus with lidocaine differentially affects expression of place and response learning

NEUROBIOLOGY OF LEARNING AND MEMORY 65, 65 – 72 (1996)
Article No. 0007
Inactivation of Hippocampus or Caudate Nucleus with Lidocaine Differentially Affects Expression of Place and Response Learning MARK G. PACKARD* AND JAMES L. MCGAUGH†,1 *Department of Psychology, University of New Orleans 70148; and †Center for the Neurobiology of Learning and Memory and Department of Psychobiology, University of California, Irvine 92717 response learning, respectively (2), in a visually cued ex- Involvement of the hippocampus and caudate nucleus in place and response learning was examined by functionally learning is acquired faster than caudate-dependent re- inactivating these brain regions bilaterally with infusions sponse learning, and (3) when animals shift to caudate- of lidocaine. Rats were trained to approach a consistently dependent response learning with extended training, the baited arm in a cross-maze from the same start box (four hippocampal-based place representation remains intact.
trials/day/14 total days). On Days 8 and 16 a single probe trial was given, in which rats were placed in the start boxopposite that used in training and allowed to approach amaze arm. Three minutes prior to the probe trial, rats received bilateral injections of either saline or a 2% lido-caine solution (in order to produce neural inactivation) Throughout the history of research on animal into either the dorsal hippocampus or dorsolateral caudate learning there have been conflicting views concern- nucleus. On the probe trials, rats which entered the baited ing the fundamental issue of what animals learn.
maze arm (i.e., approached the place where food was lo- Cognitive theorists such as Tolman (1932; 1948) pro- cated during training) were designated place learners, and posed that animals acquire knowledge of ‘‘what- rats which entered the unbaited maze arm (i.e., made the leads-to-what’’ that result in expectations of the con- same turning response as during training) were desig- sequences of their behavior. In contrast, following nated response learners. Saline-treated rats displayed the pioneering work of Thorndike (1898; 1933) and place learning on the Day 8 probe trial and response learn- Pavlov (1927), other theorists proposed that animal ing on the Day 16 probe trial, indicating that with ex- learning consists of the formation of stimulus – re- tended training there is a shift in learning mechanisms sponse (S-R) habits (Hull, 1943; Guthrie, 1935) or, controlling behavior. Rats given lidocaine injections intothe hippocampus showed no preference for place or re- simply, the learning of motor responses.
sponse learning on the Day 8 probe trial, but displayed In experiments addressing this issue, rats were response learning on the Day 16 probe trial, indicating a trained in mazes and subsequently tested in a vari- blockade of place learning following inactivation of the ety of ways in an attempt to discover whether the hippocampus. Rats given lidocaine injections into the cau- learning was based on acquisition of knowledge or date nucleus displayed place learning on both the Day 8 learning of responses (e.g., Tolman, Ritchie, & Kal- and the Day 16 probe trials, indicating a blockade of re- ish, 1946, 1947; Blodgett & McCutchan, 1947, 1948; sponse learning following inactivation of the caudate nu- Blodgett, McCutchan, & Matthews, 1949; Thomp- cleus. The findings indicate: (1) the hippocampus and cau- son & Thompson, 1949; Hill & Thune, 1952; Schar- date nucleus selectively mediate expression of place and lock, 1955). For example, after animals were trainedto enter one arm of a T-maze where they were re- 1 This research was supported by NRSA Grant 1 F32NS08973- warded with food in a goal box, had they learned a 01 (M.G.P.) and USPHS Grant MH12526 from NIDA and NIMH turning response or, alternatively, had they learned (J.L.M.). The authors thank Norman White, Norman Wienberger, where food was located? In order to examine this and Richard Thompson for helpful comments on an earlier version question, Tolman and colleagues introduced the of the manuscript. Address correspondence and reprint requests cross-maze paradigm (Tolman et al., 1946, 1947).
to Mark G. Packard, Department of Psychology, University ofNew Orleans, New Orleans, LA 70148.
The cross-maze is essentially a T-maze built such Copyright ᭧ 1996 by Academic Press, Inc.
All rights of reproduction in any form reserved.
that the choice point connecting the two goal boxes Findings of double dissociations of the effects of (e.g., east – west), can be approached from start boxes brain lesions and intracerebral drug treatments on located on either side of the maze (e.g., north – memory strongly support the hypothesis that the south). During an initial training period, animals hippocampal system and caudate nucleus mediate are placed in the same start box and trained on sev- different forms of memory. Lesions of the hippocam- eral trials to obtain a food pellet from the same goal pal system (Packard, Hirsh, & White, 1989; Pack- box. Following this training period, a critical probe ard & McGaugh, 1992; McDonald & White, 1993; trial is given in which animals are placed in the Kesner, Bolland, & Dakis, 1993), and posttraining start box opposite the one used during training and injections of drugs into the hippocampus (Packard & allowed to approach either goal box. According to White, 1991; Packard, Cahill, & McGaugh, 1994), cognitive theory, animals should approach the goal alter cognitive learning but do not affect response box which was baited during training, since they learning. Conversely, lesions of the caudate nucleus acquired information concerning the spatial location (Packard et al., 1989; Packard & McGaugh, 1992; of the food reward. Animals displaying this behavior McDonald & White, 1993; Kesner et al., 1993), and on the probe trial are designated ‘‘place learners.’’ posttraining injections of drugs into the caudate In contrast, S-R theory predicts that animals should (Packard & White, 1991; Packard et al., 1994), approach the goal box which was unbaited during alter response learning but do not affect cognitive training, since they have learned a response ten- dency (i.e., a specific body turn at the choice point).
The present study was designed to determine Animals displaying this behavior are designated ‘‘re- whether the evidence suggesting differential roles of sponse learners.’’ Although the cross-maze paradigm the hippocampus and caudate nucleus in memory provides a procedure to contrast cognitive and S-R might provide a resolution of the place versus re- learning theories, the findings of many experiments sponse learning question. Rats implanted bilaterally using this task were inconclusive. Under some condi- with cannulae in either the hippocampus or caudate tions animals indicated knowledge of the location of nucleus were first trained (four trials per day) using rewards (i.e., place learning), and under other condi- one starting alley of a cross-maze (e.g., south) and tions they made specific body turn responses on the consistently rewarded in the goal box of one alley probe trial (for review see Restle, 1957).
(e.g., west). On Days 8 and 16 the rats were given a Findings indicating that both place and response single test trial using the other starting alley (e.g., learning occurred raises the possibility that these two north) to determine whether they made the same forms of learning might be mediated by distinct neu- turning response made in training or made a differ- ral mechanisms. This hypothesis is consistent with ent turning response and went to the place where extensive evidence suggesting that different forms of food was located during the training. Prior to the learning and memory are mediated by different neu- test trials lidocaine or saline was infused bilaterally ral systems in humans (e.g., Milner, 1962; Corkin, into either the caudate nucleus or hippocampus. Li- 1965; Weiskrantz & Warrington, 1979; Cohen & docaine produces a temporary functional blockade of Squire, 1980; Schacter, 1992), monkeys (e.g., Gaffan, neural activity following local administration in the 1974; Zola-Morgan, Squire, & Mishkin, 1982; Ma- brain and has been used in studies examining brain hut & Moss, 1984), and rats (e.g., Hirsh, 1974; function and memory (e.g., Perez-Ruiz & Prado-Al- O’Keefe & Nadel, 1978; Olton, Becker, & Handel- cala, 1989; Salinas, Packard, & McGaugh, 1993). If mann, 1979; Sutherland & Rudy, 1989). Within the cognitive and response learning occur in parallel and context of the debate between cognitive and S-R are mediated by different neural systems, selective learning theorists, it has been suggested that the hip- temporary inactivation of one of these neural sys- pocampal system may selectively mediate ‘‘cognitive’’ tems prior to testing should influence the type of memory (Hirsh, 1974; Mishkin & Petri, 1984), and learning, i.e., place or response learning expressed that the caudate nucleus may selectively mediate ‘‘S- R habit’’ formation (Mishkin & Petri, 1984; Packardand White, 1987; Packard, Hirsh, & White, 1989).
Similarly, Kesner and colleagues hypothesized thatthe hippocampus selectively mediates allocentric spa- tial behavior within a ‘‘data-based’’ memory system,while the caudate nucleus selectively mediates ego- The subjects were 50 male Sprague – Dawley rats centric learning within an ‘‘expectancy-based’’ mem- (275 – 300 g). Animals were individually housed in a ory system (Kesner & DiMattia, 1987).
temperature-controlled environment on a 12-h light/ INACTIVATION OF HIPPOCAMPUS OR CAUDATE NUCLEUS dark cycle with the lights on from 7 AM to 7 PM. All intracerebral administration (Parent & McGaugh, animals were given ad lib. access to water.
1993). The injections were delivered over a period of37 s using a syringe pump (Sage Instruments), and the injection needles (extending 1 mm from the endof the guide cannulae) were left in place an addi- The apparatus was a wooden cross maze painted tional 60 s to allow for diffusion of the solution away flat gray. The maze consisted of four arms (north, from the needle tip. Lidocaine injections were ad- south, east, and west) of an eight-arm radial maze ministered approximately 2 – 3 min prior to the (other maze arms were removed). The arms of the probe trial tests on Days 8 and 16 of training. Saline cross maze measured 60 1 9 cm. The center platform injections were administered using procedures iden- of the maze connecting the four arms measured 40 tical to those used for lidocaine injections.
cm in diameter. A clear Plexiglas cross-shaped alley-way structure placed on the center platform of the modified radial maze connected the four arms of thecross-maze. The alleyways measured 20 1 9 1 15 At the completion of behavioral training, animals cm. A recessed food well was present at the end of were deeply anesthetized with a 1.0-ml injection of the west arm of the maze. The maze was located in sodium pentobarbital and perfused with saline fol- a testing room that contained many extramaze cues lowed by 10% formal-saline. The brains were re- including wall posters, a lamp, table, animal cage moved and subsequently sectioned at 20-mm sections through the cannula tract region and stained withCresyl violet. Cannula placements were examined for verification of needle tip location using the atlasof Paxinos and Watson (1986).
Animals were anesthetized with sodium pentobar- Cannulae placements for both dorsolateral cau- bital (50 mg/kg) and implanted with bilateral guide date nucleus and dorsal hippocampus are shown in cannula in the dorsal hippocampus (10 mm length) Fig. 1 (left, right, respectively). Caudate nucleus or dorsolateral caudate nucleus (15 mm length) us- placements were located in the dorsolateral caudate, ing standard stereotaxic techniques. The cannulae ranging from 0.20 mm to 00.30 mm from bregma (23 gauge) were anchored to the skull with jewelers (Fig. 1, left). Hippocampal placements were located screws and dental acrylic. Coordinates for the dorsal in the dorsal hippocampus, ranging from 02.8 mm hippocampal placements were AP Å 03.1 mm, ML to 03.3 mm from bregma (Fig. 1, right).
Å {2.5 mm, and DV Å 02.0 mm from bregma. Coor-dinates for the dorsolateral caudate nucleus place- ments were AP Å 0.26 mm, ML Å {4.2 mm, DVÅ 04.0 mm. After surgery, stylets (30 gauge) were Prior to training, all rats were reduced to 85% of inserted and left in place to ensure cannulae patency ad lib. body weights over 7 days and maintained at until injections were made. Behavioral testing began this weight throughout the experiment. On 2 consec- utive days, rats were placed into the cross-maze inthe start box (south arm) and allowed to explore the maze for 5 min. No food was present in the maze on A 2% lidocaine hydrochloride solution (Western either of these 2 habituation days. Access to the Medical Supply, Inc.) was used to produce reversible north arm of the cross-maze was blocked during ha- inactivation of brain sites. Injections (0.5 ml) were bituation sessions and the subsequent food rewarded administered intracerebrally using 30-gauge injec- training trials with a clear Plexiglas shield. Follow- tion needles inserted into the guide cannulae. The ing habituation on both days, the animals were al- needles were connected by polyethylene tubing to lowed to consume ten 45-mg Noyes food pellets in 10-ml Hamilton microsyringes (Hamilton Co., Reno, their home cage. Food trials began on Day 3. On NV). The volume of 2% lidocaine solution used was each food trial rats were placed into the start box chosen on the basis of previous evidence indicating and allowed to traverse the maze and consume a that this volume produces functional inactivation of single Noyes food pellet located in the food cup at the caudate nucleus (Perez-Ruiz & Prado-Alcala, the end of the goal arm of the cross maze (west arm).
1989) and amygdala (Salinas et al., 1993) sufficient On the initial food trial only, a trail of four pellets to cause memory impairment. Other findings indi- leading to the food cup was placed along the length cate that volumes of 2% lidocaine solution as low as of the goal arm. Each rat received four food rewarded 0.25 ml can produce memory impairment following trials per day. Entries into the unbaited arm of the Dorsolateral caudate nucleus (left) and dorsal hippocampus (right) cannulae placements showing the anterior/posterior extent of needle tip locations at 0.5-mm sections. Caudate nucleus placements ranged from 0.20 to 0.30 mm from bregma. Hippocampalplacements ranged from 02.8 to 03.0 mm from bregma. (Plates adapted from atlas of Paxionos and Watson, 1986).
cross-maze (east) were scored as incorrect responses On Day 8 of the food-rewarded trials, rats were during the training trials, and entries into the baited assigned to experimental groups (n Å 12–14 per arm of the cross-maze (west) were scored as correct group) in a rank-order method to assure that levels responses. A correction procedure was used such of learned performance (i.e., number of correct re- that rats making an incorrect response were allowed sponses over the first 7 days of training) of the groups to trace back to the baited maze arm and consume were comparable prior to treatment. On Day 8 a sin- the food pellet. If a rat failed to consume the food gle probe trial was given. Three minutes prior to the pellet within 2 min, the trial was terminated and probe trial, half of the animals with caudate nucleus the rat was manually placed in the goal box and implants received a saline injection and the other half allowed to consume the pellet. After consuming the received a lidocaine injection. Similarly, half of the pellet on a given trial, the rat was placed in a holding animals with hippocampal implants received a saline cage located directly behind the start arm for a 30- injection prior to the probe trial, and half received an injection of lidocaine. On the probe trial, animals INACTIVATION OF HIPPOCAMPUS OR CAUDATE NUCLEUS the Day 8 probe trial, animals given saline injectionsinto either the hippocampus or caudate nucleus werepredominantly place learners (hippocampus – salinex2 Å 7.14, p õ .05; caudate nucleus – saline x2 Å5.34, p õ .05). Rats given intracaudate injections oflidocaine were also predominantly place learners (x2Å 5.34, p õ .05), while those receiving intrahippo-campal injections of lidocaine did not show a signifi-cant trend toward either place or response learning(x2 Å 0, n.s.). These findings indicate that functionalinactivation of the hippocampus, but not caudate nu-cleus, prevented the expression of place learning.
On the Day-16 probe trial, animals given saline injections into either the hippocampus or caudatenucleus were predominantly response learners (hip-pocampus – saline x2 Å 4.58, p õ .05; caudate nu-cleus – saline x2 Å 5.34, p õ .05). Rats receiving in-trahippocampal injections of lidocaine were also pre-dominantly response learners (x2 Å 5.34, p õ .05), Number of rats in each treatment group that exhib- while those receiving intracaudate injections of lido- ited place or response learning on both the Day 8 and Day 16 test caine were predominantly place learners (x2 Å 8.32, p õ .05). These findings indicate that with extendedtraining, saline-treated rats switched from dis-playing place learning to response learning. Func- were placed into the start box opposite that used dur- tional inactivation of the caudate nucleus, but not ing training (i.e., north arm) and were allowed to hippocampus, prevented the expression of response make a single entry into either the baited or the un- learning and, in fact, preserved place learning.
baited maze arm. The entrance to the south maze arm(i.e., the arm containing the start box used during training) was blocked by a clear Plexiglas shield onthe probe trials. Animals entering the baited arm on The results indicate that place and response learn- the probe trial were designated ‘‘place’’ learners (i.e., ing occur concurrently and are mediated by different animals going to the place where food was located neural systems involving the hippocampus and cau- during training), and animals entering the unbaited date nucleus, respectively. The findings are consis- arm on the probe trial were designated ‘‘response’’ tent with previous studies using pretraining irre- learners (i.e., animals making the same turning re- versible lesions of the caudate nucleus (Thompson, Guilford, & Hicks, 1980), and fimbria – fornix (Decas- On Day 9 of training, food-rewarded training (four tro, 1974), which reported that such lesions result in trials per day) was reinstated using procedures iden- a predominant tendency toward place and response tical to those of Training Days 1 – 7. On Day 16, a learning, respectively. Our findings indicate that in second probe trial was given using procedures identi- addition to acquisition, the functional integrity of cal to the probe trial given on Day 8. Animals re- the hippocampus and caudate nucleus are necessary ceived the same intracerebral saline or lidocaine for the expression of these two forms of learned be- treatments given on the Day 8 probe trial and were again allowed to make a single entry into either the The finding that rats initially acquire place infor- mation and that the expression of such learning isblocked by inactivation of the hippocampus is consis- tent with extensive evidence implicating the hippo-campus in spatial learning (e.g., O’Keefe & Nadel, Figure 2 shows the results of the test trials on Days 8 and 16. x2 analyses (p õ .05 for all compari- The hippocampal injection sites in the present sons) were computed in order to determine whether study were located along the dorsal edge of the hip- groups showed a significant tendency to display pocampus, and areas of parietal cortex located dorso- place or response learning on the probe trials. On lateral to the lidocaine injection sites may also be involved in acquisition of spatial information. For rats, a response learning tendency had apparently example, large lesions of parietal cortex selectively not been acquired by this system prior to the eighth impair acquisition of a spatial cheese-board task day of training. Thus, one difference between the (Kesner, Farnsworth, & DiMattia, 1989) and pro- operating characteristics of the hippocampal-based duce deficits in acquisition and retention of spatial and caudate-based memory systems is that the for- behavior in the Morris water maze (DiMattia & mer system appears to be involved in the rapid ac- Kesner, 1988). However, the posterior extent of the quisition of new information, while the latter large parietal cortex lesions reported to impair spa- system acquires information in a slower, incremen- tial behavior (Kesner, Farnsworth, & DiMattia, 1989) is approximately 0.5 – 0.8 mm anterior and With extended training in the cross-maze, animals 1.0 – 1.2 mm lateral to our effective injection sites in acquired a response learning tendency (i.e., on the dorsal hippocampus. Furthermore, small lesions of Day 16 probe trial control rats made the same turn- parietal cortex which may more closely mimic the ing response as during training). The finding that behavioral effects of any limited spread of lidocaine the animals shifted from place learning to response from hippocampus in the present study do not im- learning with increased training is consistent with pair acquisition of spatial behavior in the Morris that of previous research (e.g., Ritchie, Aeschli- water maze or the radial maze (Kolb, Sutherland, & man, & Pierce, 1950; Hicks, 1964). Importantly, se- Whishaw, 1983). Thus, although the possibility that lective functional inactivation of the caudate nucleus the spread of lidocaine to adjacent parietal cortex prevented the expression of the acquired response influenced the expression of place learning cannot and revealed preserved place learning. Thus, the pre- be completely ruled out, the effect of lidocaine injec- dominance of response learning expressed after ex- tions on place learning observed in the present study tensive training is not due to elimination of the place was likely due to neural inactivation of the hippo- representation. These findings provide strong evi- dence that the two memory systems are functioning The finding that the expression of response learn- ing is blocked by functional inactivation of the cau- In 1957, Restle proposed a resolution of the place date nucleus is consistent with the view that the versus response learning debate based on findings caudate nucleus selectively mediates learning in- suggesting that either type of learning could be pre- volving the use of ‘‘egocentric’’ cues to guide naviga- dicted based on the availability of various ex- tion (Potegal, 1972; Kesner & DiMattia, 1987). For tramaze/intramaze cues present (Restle, 1957).
example, lesions of the caudate nucleus impair ac- Thus, in open mazes located in visually heteroge- quisition of left – right maze discrimination tasks neous environments place learning is dominant (Cook and Kesner, 1988; Kesner et al., 1993). As (Blodgett & McCutchan, 1947; Blodgett, McCut- the region of the caudate targeted by the lidocaine chan, & Matthews, 1949; Tolman et al., 1946, 1947), injections (dorsolateral) receives cortical input from while in closed mazes or visually homogeneous envi- the somatosensory cortex (Heimer, Alheid, & Zar- ronments response learning is dominant (Blodgett & bosky, 1985) such input may be critical for establish- McCutchan, 1948; Thompson & Tompson, 1949; ing an association between the maze cues and the Hill & Thune, 1952; Scharlock, 1955). According to egocentric cues mediating the specific body turn re- Restle’s multiple-cue theory, a single learning mech- quired for response learning. However, it is im- anism was responsible for both place and response portant to note that egocentric learning may be but learning, and this mechanism was ‘‘tuned’’ to pro- one example of a class of stimulus-response learning duce either type of learning by the nature of the functions mediated by the caudate (Packard et al., extramaze/intramaze environment. However, al- 1989; Packard and McGaugh, 1992; Viaud & White, though evidence indicates that the nature of the test- 1989), each of which is hypothesized (White, 1989) ing environment contributes to the relative expres- to be organized on the basis of the sensory cortical sion of place and response learning, Restle’s resolu- input the caudate receives (Heimer et al., 1985).
tion of the debate is, at least, incomplete. The nature Rats given saline injections into the hippocampus of the environment appears to interact with separate or caudate nucleus displayed a strong place learning learning systems in determining the expression of tendency on the Day 8 probe trial. However, animals place or response learning. Consistent with this sug- with the hippocampus inactivated with lidocaine gestion, acquisition of a caudate nucleus-mediated prior to the Day 8 probe trial did not display either win – stay radial maze task by intact rats is facili- place or response learning tendencies. Although the tated when extramaze cues are reduced by sur- caudate nucleus was presumably functional in these rounding the maze with curtains (Packard & White, INACTIVATION OF HIPPOCAMPUS OR CAUDATE NUCLEUS Hirsh, R. (1974). The hippocampus and contextual retrevial of 1987). Future studies are necessary to determine information from memory: A theory. Behavioral Biology, 12,
what other factors in addition to the nature of the testing environment contribute to the use of a given Hull, C. L. (1943). Principles of Behavior. Appleton-Century- Finally, these results indicate that the place ver- Kesner, R. P., Farnsworth, G., and DiMattia, B. D. (1989). Double sus response controversy was based on a false prem- dissociation of egocentric and allocentric space following me- ise: The debate was primarily based on the assump- dial prefrontal and parietal cortex lesions in the rat. Behav- tion that there is only one kind of learning and that ioral Neuroscience, 103, 956 – 961.
the appropriate test would reveal the nature of that Kesner, R. P., Bolland, B. L., & Dakis, M. (1993). Memory for spatial locations, motor responses, and objects: Triple dissoci- learning (but see Tolman, 1949). Evidence that the ation among the hippocampus, caudate nucleus, and extra- brain uses different neural systems for acquiring striate visual cortex. Experimental Brain Research, 93, 462 –
and expressing different kinds of learning would ap- pear to resolve this long-lasting controversy.
Kesner, R. P., & DiMattia, B. V. (1987). Neurobiology of an attri- bute model of memory. Progress in Psychobiology and Physio-
logical Psychology, 12, 207 – 277.
Kolb, B., Sutherland, R. J., and Whishaw, I. Q. (1983). A compari- Blodgett, H. C., & McCutchan, K. (1947). Place versus response son of the contributions of the frontal and parietal association learning in the simple T-maze. Journal of Experimental Psy- cortex to spatial localization in rats. Behavioral Neuroscience, chology, 37, 412 – 422.
97, 13 – 27.
Blodgett, H. C., and McCutchan, K. (1948). The relative strength Mahut, H., & Moss, M. (1984). Consolidation of memory: The of place and response learning in the T-maze. Journal of hippocampus revisited. In N. Butters and L. R. Squire (Eds.), Comparative and Physiological Psychology, 41, 17 – 24.
Neuropsychology of Memory (pp. 297 – 315). New York: Guil- Blodgett, H. C., McCutchan, K., & Matthews, R. (1949). Spatial learning in the T-maze: The influence of direction, turn, and McDonald, R. J., & White, N. M. (1993). A triple dissociation of food location. Journal of Experimental Psychology, 39, 800 –
memory systems: Hippocampus, amygdala, and dorsal stria- tum. Behavioral Neuroscience, 107, 3 – 22.
Cohen, N. J., & Squire, L. R. (1980). Preserved learning and reten- Milner, B. (1962). Les troubles de la memoire accompagnant des tion of pattern analyzing skill in amnesics: Dissociation of lesions hippocampiques bilaterales. In P. Passant (Ed.), Psys- knowing how and knowing that. Science, 210, 207 – 210.
iologie de l’Hippocampae (pp. 257 – 272). Paris: Centre de la Cook, D., & Kesner, R. P. (1988). Caudate nucleus and memory for egocentric localization. Behavioral and Neural Biology, Mishkin, M., & Petri, H. L. (1984). Memories and habits: Some 49, 332 – 343.
implications for the analysis of learning and retention. In Corkin, S. (1965). Tactually-guided maze learning in man: Effect L. R. Squire and N. Butters (Eds.), Neuropsychology of Mem- of unilateral cortical excision and bilateral hippocampal le- ory (pp. 287 – 296). New York: Guilford.
sions. Neuropsychologia, 3, 339 – 351.
Mishkin, M., Malamut, B., & Bachevalier, J. (1984). Memories DeCastro, J. M. (1974). A selective spatial discrimination deficit and habits: Two neural systems. In G. Lynch, J. L. McGaugh, after fornicotomy in the rat. Behavioral Biology, 12, 373 –
and N. M. Weinberger (Eds.), Neurobiology of Learning and Memory (pp. 65 – 77). New York: Guilford.
DiMattia, B. D., and Kesner, R. P. (1988). Spatial cognitive maps: O’Keefe, J., & Nadel, L. (1978). The Hippocampus as a Cognitive Differential roles of parietal cortex and hippocampal forma- Map. New York: Oxford University Press.
tion. Behavioral Neuroscience, 102, 471 – 480.
Olton, D. S., Becker, J. T., & Handelmann, G. E. (1979). Hippo- Gaffan, D. (1974). Recognition impaired and association intact campus, space, and memory. Behavioral and Brain Sciences, in the memory of monkeys after transsection of the fornix.
2, 313 – 365.
Journal of Comparative and Physiological Psychology, 86,
Packard, M. G., & McGaugh, J. L. (1992). Double dissociation of fornix and caudate nucleus lesions on acquisition of two water Galanter, E. H., & Shaw, W. A. (1954). ‘‘Cue’’ vs. ‘‘reactive inhibi- maze tasks: Further evidence for multiple memory systems.
tion’’ in place and response learning. Journal of Comparative Behavioral Neuroscience, 106, 439 – 446.
and Physiological Psychology, 47, 395 – 398.
Packard, M. G., & White, N. M. (1987). Differential roles of the Guthrie, E. R. (1935). The Psychology of Learning. Harper and hippocampus and caudate nucleus in memory: Selective me- diation of ‘‘cognitive’’ and ‘‘associative’’ learning. Society for Heimer, L., Alheid, G. F., & Zaborsky, L. (1985). Basial Ganglia.
Neuroscience Abstracts, 13, 1005.
In G. Paxinos, (Ed.) The Rat Nervous System (pp. 37 – 87).
Packard, M. G., & White, N. M. (1990). Lesions of the caudate nucleus selectively impair acquisition of ‘‘reference memory’’ Hicks, L. H. (1964). Effects of overtraining on acquisition and in the radial maze. Behavioral and Neural Biology, 53, 39 –
reversal of place and response learning. Psychological Re- ports, 15, 459 – 462.
Packard, M. G., & White, N. M. (1991). Dissociation of hippocam- Hill, C. W., & Thune, L. E. (1952). Place and response learning pus and caudate nucleus memory systems by posttraining in the white rat under simplified and mutually isolated condi- intracerebral injection of dopamine agonists. Behavioral Neu- tions. Journal of Experimental Psychology, 43, 289 – 297.
roscience, 105, 295 – 306.
Packard, M. G., Cahill, L., & McGaugh, J. L. (1994). Amygdala theory: The role of the hippocampal formation in learning, modulation of hippocampal-dependent and caudate nucleus- memory, and amnesia. Psychobiology, 17, 129 – 144.
dependent memory processes. Proceedings of the National Thompson, M. E., & Thompson, J. P. (1949). Reactive inhibition Academy of Sciences USA 91, 8477 – 8481.
as a factor in maze learning: II. The role of reactive inhibition Packard, M. G., Hirsh, R., & White, N. M. (1989). Differential in studies of place learning versus response learning. Journal effects of fornix and caudate nucleus lesions on two radial of Experimental Psychology, 39, 883 – 891.
maze tasks: Evidence for multiple memory systems. Journal Thompson, W. G., Guilford, M. O., & Hicks, L. H. (1980). Effects of Neuroscience, 9, 1465 – 1472.
of caudate and cortical lesions on place and response learning
in rats. Physiological Psychology, 8, 473 – 479.
Parent, M. B., & McGaugh, J. L. (1993). Time-dependent retro- grade impairment produced by lidocaine injections into the Thorndike, E. L. (1898). Animal Intelligence: An experimental basolateral complex of the amygdala. Society for Neuroscience study of the associative processes in animals. Psychological Review, 8, 28 – 31.
Thorndike, E. L. (1933). A proof of the law of effect. Science, 77,
Pavlov, I. P. (1927). Conditioned Reflexes. Clarendon Press, Lon- Tolman, E. C. (1932). Purposive Behavior in Animals and Men. Paxinos, G., & Watson, C. (1986). The Rat Brain in Stereotaxic Coordinates. New York: Academic Press.
Tolman, E. C. (1948). Cognitive maps in rats and men. Psychologi- Perez-Ruiz, C., & Prado-Alcala, R. A. (1989). Retrograde amnesia cal Review, 56, 144 – 155.
induced by lidocaine injection into the striatum: protective Tolman, E. C. (1949). There is more than one kind of learning.
effect of the negative reinforcer. Brain Research Bulletin, 22,
Psychological Review, 55, 189 – 208.
Tolman, E. C., & Gleitman, H. (1949). Studies in spatial learning: Potegal, M. (1972). The caudate nucleus egocentric localization VII. Place and response learning under different degrees of system. Acta Neurobiologica Experimentia, 32, 379 – 494.
motivation. Journal of Experimental Psychology, 39, 633 –
Restle, F. (1957). Discrimination of cues in mazes: A resolution of the place vs. response controversy. Psychological Review, Tolman, E. C., Ritchie, B. F., & Kalish, D. (1946). Studies in 64, 217 – 228.
spatial learning: II. Place learning versus response learning.
Ritchie, B. F., Aeschliman, B., & Pierce, P. (1950). Studies in Journal of Experimental Psychology, 36, 221 – 229.
spatial learning: VIII. Place performance and the acquisition Tolman, E. C., Ritchie, B. F., & Kalish, D. (1947). Studies in of place dispositions. Journal of Comparative and Physiologi- spatial learning: V. Response versus place learning by the cal Psychology, 43, 73 – 85.
noncorrection method. Journal of Experimental Psychology, Salinas, J. A., Packard, M. G., & McGaugh, J. L. (1993). Amygdala 37, 285 – 292.
modulates memory for changes in reward magnitude: Revers- Viaud, M. D., & White, N. M. (1989). Dissociation of visual and ible post-training inactivation with lidocaine attenuates the olfactory conditioning in the neostriatum of rats. Behavioral response to a reduction in reward. Behavioral Brain Re- Brain Research, 32, 31 – 42.
search, 59, 153 – 159.
Weiskrantz, L., and Warrington, E. K. (1979). Conditioning in Schacter, D. L. (1992). Priming and multiple memory systems: amnesic patients. Neuropsychologia, 17, 187 – 194.
Perceptual mechanisms of implicit memory. Journal of Cog- White, N. M. (1989). A functional hypothesis concerning the stria- nitive Neuroscience, 4, 244 – 256.
tal matrix and patches: Mediation of S-R memory and re- Scharlock, D. P. (1955). The role of extramaze cues in place and ward. Life Sciences, 45, 1943 – 1947.
response learning. Journal of Experimental Psychology, 50,
Zola-Morgan, S., Squire, L., & Mishkin, M. (1982). The neuroanat- omy of amnesia: Amygdala – hippocampus versus temporal Sutherland, R. J., & Rudy, J. W. (1989). Configural association stem. Science, 218, 1337 – 1339.

Source: http://gureckislab.org/courses/fall13/learnmem/papers/Packard1996.pdf