A model of the hippocampus combining self- 
organization and associative memory function. 
Michael E. Hasselmo, Eric Schnell 
Joshua Berke and Edi Barkai 
Dept. of Psychology, Harvard University 
33 Kirkland St., Cambridge, MA 02138 
hasselmokatla.harvard.edu 
Abstract 
A model of the hippocampus is presented which forms rapid self-orga- 
nized representations of input arriving via the perforant path, performs 
recall of previous associations in region CA3, and performs comparison 
of this recall with afferent input in region CA1. This comparison drives 
feedback regulation of cholinergic modulation to set appropriate 
dynamics for learning of new representations in region CA3 and CA1. 
The network responds to novel patterns with increased cholinergic mod- 
ulation, allowing storage of new self-organized representations, but 
responds to familiar patterns with a decrease in acetylcholine, allowing 
recall based on previous representations. This requires selectivity of the 
cholinergic suppression of synaptic transmission in stratum radiatum of 
regions CA3 and CA1, which has been demonstrated experimentally. 
1 INTRODUCTION 
A number of models of hippocampal function have been developed (Burgess et al., 1994; 
Myers and Gluck, 1994; Touretzky et al., 1994), but remarkably few simulations have 
addressed hippocampal function within the constraints provided by physiological and ana- 
tomical data. Theories of the function of specific subregions of the hippocampal forma- 
tion often do not address physiological mechanisms for changing dynamics between 
learning of novel stimuli and recall of familiar stimuli. For example, the afferent input to 
the hippocampus has been proposed to form orthogonal representations of entorhinal 
activity (Mart, 1971; McNaughton and Morris, 1987; Eichenbaum and Buckingham, 
1990), but simulations have not addressed the problem of when these representations 
78 Michael E. Hasselmo, Eric Schne!l, Joshlea Berke, Edi Barkai 
should remain stable, and when they should be altered. In addition, models of autoasso- 
ciative memory function in region CA3 (Mart, 1971; McNaughton and Morris, 1987; 
Levy, 1989; Eichenbaum and Buckingham, 1990) and heteroassociative memory function 
at the Schaffer collaterals projecting from region CA3 to CA1 (Levy, 1989; McNaughton, 
1991) require very different activation dynamics during learning versus recall. 
Acetylcholine may set appropriate dynamics for storing new information in the cortex 
(Hasselmo et al., 1992, 1993; Hasselmo, 1993, 1994; Hasselmo and Bower, 1993). Ace- 
tylcholine has been shown to selectively suppress synaptic transmission at intrinsic but 
not afferent fiber synapses (Hasselmo and Bower, 1992), to suppress the neuronal adapta- 
tion of cortical pyramidal cells (Hasselmo et al., 1994; Barkai and Hasselmo, 1994), and 
to enhance long-term potentiation of synaptic potentials (Hasselmo, 1994b). Models 
show that suppression of synaptic transmission during learning prevents recall of previ- 
ously stored information from interfering with the storage of new information (Hasselmo 
et al., 1992, 1993; Hasselmo, 1993, 1994a), while cholinergic enhancement of synaptic 
modification enhances the rate of learning (Hasselmo, 1994b). 
Feedback regulation of cholinergic modulation may set the appropriate level of cholin- 
ergic modulation dependent upon the novelty or familiarity of a particular input pattern. 
We have explored possible mechanisms for the feedback regulation of cholinergic modu- 
lation in simulations of region CA1 (Hasselmo and Schnell, 1994) and region CA3. Here 
we show that self-regulated learning and recall of self-organized representations can be 
obtained in a network simulation of the hippocampal formation. This model utilizes selec- 
tive cholinergic suppression of synaptic transmission in stratum radiatum of region CA3, 
which has been demonstrated in brain slice preparations of the hippocampus. 
2 METHODS 
2.1. SIMPLIFIED REPRESENTATION OF HIPPOCAMPAL NEURONS. 
In place of the sigmoid input-output functions used in many models, this model uses a 
simple representation in which the output of a neuron is not explicitly constrained, but the 
total network activity is regulated by feedback from inhibitory interneurons and adapta- 
tion due to intracellular calcium concentration. Separate variables represent pyramidal 
cell membrane potential a, intracellular calcium concentration c, and the membrane poten- 
tial of inhibitory intemeurons h: 
Aa i = A i - rla i - gc +  Wijg(a j - 0o) - Hitg(ht - Oh) 
J 
AC i = Tg(ai- Oc) - c 
Ah: = Wtjg(a j - 0o)- 11h k - '.Hklg(h I - 0o) 
j l 
where A = afferent input, rl = passive decay of membrane potential, ix = strength of cal- 
A Model of Hippocampus 79 
cium-dependent potassium current (proportional to intracellular calcium), Wij = excitatory 
recurrent synapses (longitudinal association path terminating in stratum radiatum), gO is a 
threshold linear function proportional to the amount by which membrane potential 
exceeds an output threshold 0o or threshold for calcium current 0c, t = strength of voltage- 
dependent calcium current, '1 = diffusion constant of calcium, Wki = excitatory synapses 
inhibitory intemeurons, Hik = inhibitory synapses from intemeurons to pyramidal cells, 
Hkl= inhibitory synapses between intemeurons. This representation gives neurons adapta- 
tion characteristics similar to those observed with intracellular recording (Barkai and Has- 
selmo, 1994), including a prominent afterhyperpolarization potential (see Figure 1). 
A 
B C 
Figure 1. Comparison of pyramidal cell model with experimental data. 
In Figure 1, A shows the membrane potential of a modeled pyramidal cell in response to 
simulated current injection. Output of this model is a continuous variable proportional to 
how much membrane potential exceeds threshold. This is analogous to the reciprocal of 
interspike interval in real neuronal recordings. Note that the model displays adaptation 
during current injection and afterhyperpolarization afterwards, due to the calcium-depen- 
dent potassium current. B shows the intracellularly recorded membrane potential in a pir- 
iform cortex pyramidal cell, demonstrating adaptation of fu'ing frequency due to 
activation of calcium-dependent potassium current. The firing rate falls off in a manner 
similar to the smooth decrease in firing rate in the simplified representation. C shows an 
intracellular recording illustrating long-term afterhyperpolarization caused by calcium 
influx induced by spiking of the neuron during current injection. 
2.2. NETWORK CONNECTIVITY 
A schematic representation of the network simulation of the hippocampal formation is 
shown in Figure 2. The anatomy of the hippocampal formation is summarized on the left 
in A, and the function of these different subregions in the model is shown on the right in 
B. Each of the subregions in the model contained a population of excitatory neurons with 
a single inhibitory intemeuron mediating feedback inhibition and keeping excitatory 
activity bounded. Thus, the local activation dynamics in each region follow the equations 
presented above. The connectivity of the network is further summarized in Figure 3 in the 
Results section. A learning rule of the Hebbian type was utilized at all synaptic connec- 
tions, with the exception of the mossy fibers from the dentate gyms to region CA3, and the 
connections to and from the medial septurn. Self-organization of perforant path synapses 
was obtained through decay of synapses with only pre or post-synaptic activity, and 
growth of synapses with combined activity. Associative memory function at synapses 
80 Michael E. Hassebno, Eric Schnell, Joshua Berke, Edi Barkai 
arising from region CA3 was obtained through synaptic modification during cholinergic 
suppression of synaptic transmission. 
A I Entrhinal crtexl -. B 
 Self-organized// [ 'ira 
[ I Region CAI-- Region CA3,  represenmtinl-- 
ptum I .. 
Fdback gulafion of Regulation of 
cholinergic mulation leing dynamics 
Figure 2. Schematic representation of hippocampal circuitry 
and the corresponding function of connections in the model. 
2.3. CHOLINERGIC MODULATION 
The total output from region CA1 determined the level of cholinergic modulation within 
both region CA3 and CA1, with increased output causing decreased modulation. This is 
consistent with experimental evidence suggesting that activity in region CA1 and region 
CA3 can inhibit activity in the medial septurn, and thereby downregulate cholinergic mod- 
ulation. This effect was obtained in the model by excitatory connections from region CA1 
to an inhibitory intemeuron in the medial septurn, which suppressed the activity of a cho- 
linergic neuron providing modulation to the full network. When levels of cholinergic 
modulation were high, there was strong suppression of synaptic transmission at the excite- 
tory recurrent synapses in CA3 and the Schaffer collaterals projecting from region CA3 to 
CA1. This prevented the spread of activity due to previous learning from interfering with 
self-organization. When levels of cholinergic modulation were decreased, the strength of 
synaptic transmission was increased, allowing associative recall to dominate. Cholinergic 
modulation also increased the rate of synaptic modification and alepolarized neurons. 
2.4. TESTS OF SELF-REGULATED LEARNING AND RECALL 
Simulations of the full hippocampal network evaluated the response to the sequential pre- 
sentation of a series of highly overlapping activity patterns in the entorhinal cortex. Recall 
was tested with interspersed presentation of degraded versions of previously presented 
activity patterns. For effective recall, the pattern of activity in entorhinal cortex layer IV 
evoked by degraded patterns matched the pattern evoked by the full learned version of 
these patterns. The function of the full network is illustrated in Figure 3. In simulations 
A Model of Hippocampus 81 
focused on region CA3, activity patterns were induced sequentially in region CA3, repre- 
senting afferent input from the entorhinal cortex. Different levels of external activation of 
the cholinergic neuron resulted in different levels of learning of new overlapping patterns. 
These results are illustrated in Figure 4. 
2.5. BRAIN SLICE EXPERIMENTS 
The effects in the simulations of region CA3 depended upon the cholinergic suppression 
of synaptic transmission in stratum radiatum of this region The cholinergic suppression of 
glutamatergic synaptic transmission in region CA3 was tested in brain slice preparations 
by analysis of the influence of the cholinergic agonist carbachol on the size of field poten- 
rials elicited by stimulation of stratum radiatum. These experiments used techniques sim- 
ilar to previously published work in region CA1 (Hasselmo and Schnell, 1994). 
3 RESULTS 
In the full hippocampal simulation, input of an unfamiliar pattern to entorhinal cortex 
layer II resulted in high levels of acetylcholine. This allowed rapid self-organization of 
the perforant path input to the dentate gyms and region CA1. Cholinergic suppression of 
synaptic transmission in region CA1 prevented recall from interfering with self-organiza- 
tion. Instead, recurrent collaterals in region CA3 stored an autoassociative representation 
of the input from the dentate gyms to region CA3, and connections from CA3 to CA1 
stored associations between the pattem of activity in CA3 and the associated self-orga- 
nized representation in region CA1. 
. identity  
E/  r auto- I NN,self'rg 'Natr'x  > 
  Self-org  identity o. hetero- .R hetero- 
matrix ' assoc  assoc 
' "' ' " ,,, 
 311I I! !"' t I I I, I II l 
4, ,,, . .., " '"""'" 
4 II I ' !' I ' I 'i ,If' 
ld,I I I ! I i tl I, I I iI I II, I I   t 
2ll II' I I[ I I I Ii II t It l 
1 II 1 1 3C 1 3C 1 11 
Neuron # 
Figure 3. Activity in each subregion of the full network simulation of the hippocampal 
formation during presentation of a sequence of activity patterns in entorhinal cortex. 
82 Michael E. Hasselmo, Eric Schnell, Joshua Berke, Edi Barkai 
In Figure 3, width of the lines represents the activity of each neuron at a particular time 
step. As seen here, the network forms a self-organized representation of each new pattern 
consisting of active neurons in the dentate gyms and region CA1. At the same time, an 
association is formed between the self-organized representation in region CA1 and the 
same afferent input pattern presented to entorhinal cortex layer IV. Four overlapping pat- 
terns (14) are presented sequentially, each of which results in learning of a separate self- 
organized representation in the dentate gyms and region CA1, with an association formed 
between this representation and the full input pattern in entorhinal cortex. 
The recall characteristics of the network are apparent when degraded versions of the affer- 
ent input patterns are presented in the sequence (ld-4d). This degraded afferent input 
weakly activates the same representations previously formed in the dentate gyms. Recur- 
rent excitation in region CA3 enhances this activity, giving robust recall of the full version 
of this pattern. This activity then reaches CA1, where it causes strong activation if it 
matches the pattern of afferent input from the entorhinal cortex. Strong activation in 
region CA1 decreases cholinergic modulation, preventing formation of a new representa- 
tion and allowing recall to dominate. Strong activation of the representation stored in 
region CA1 then activates the full representation of the pattern in entorhinal cortex layer 
IV. Thus, the network can accurately recall each of many highly overlapping patterns. 
The effect of cholinergic modulation on the level of learning or recall can be seen more 
clearly in a simulation of auto-associative memory function in region CA3 as shown in 
Figure 4. Each box shows the response of the network to sequential presentation of full 
and degraded versions of two highly overlapping input patterns. The width of the black 
traces represents the activity of each of 10 CA3 pyramidal cells during each simulation 
step. In the top row, level of cholinergic modulation (ACh) is plotted. In A, external acti- 
vation of the cholinergic neuron is absent, so there is no cholinergic supp .ression of synap- 
tic transmission. In this case, the first pattern is learned and recalled properly, but 
subsequent presentation of a second overlapping pattern results only in recall of the previ- 
ously learned pattern. In B, with greater cholinergic suppression, recall is suppressed suf- 
ficiently to allow learning of a combination of the two input patterns. Finally, in C, strong 
cholinergic suppression prevents recall, allowing learning of the new overlapping pattern 
to dominate over the previously stored pattern. 
Stood A B C 
patterns ACh input = 0.0 ACh input = 0.15 ACh input = 0.3 
I Ill I1 
AC 
.. ,il 411 ',1111 'ill 
,,',,,,,,1 dlllll ...llffil ,IfilIll 
qlillll ,,#filll .111ffi# 
'"'ffillffi..,,,'IH..}I ll... ,,I INII[ 
I ........ "1 'll 
I ,,,.,,.,,. .ll .IHI "#1111 '".1! .18i8 "ll ,Ill[ 
I .... ,. '""1 '111B 
L ,,-,u,,ql .,11111ffi ,111111111 ,fillIll ',,.IHil..a] ,,ill .11 ll 
Figure 4. Increased cholinergic suppression of synaptic transmission in region CA3 
causes greater learning of new aspects of afferent input patterns. 
A Model of Hippocampus 83 
Extracellular recording in brain slice preparations of hippocampal region CA3 have dem- 
onstrated that perfusion of the cholinergic agonist carbachol strongly suppresses synaptic 
potentials recorded in stratum radiatum, as shown in Figure 5. In contrast, suppression of 
synaptic transmission at the afferent fiber synapses arising from entorhinal cortex is much 
weaker. At a concentration of 20tM, carbachol suppressed synaptic potentials in stratum 
radiatum on average by 54.4% (n=5). Synaptic potentials elicited in stratum lacunosum 
were more weakly suppressed, with an average suppression of 28%. 
Figure 5. 
Carbachol Wash 
Control (20M) 
Cholinergic suppression of synaptic transmission in stratum radiatum of CA3. 
4 DISCUSSION 
In this model of the hippocampus, self-organization at perforant path synapses forms com- 
pressed representations of specific patterns of cortical activity associated with events in 
the environment. Feedback regulation of cholinergic modulation sets appropriate dynam- 
ics for learning in response to novel stimuli, allowing predominance of self-organization, 
and appropriate dynamics for recall in response to familiar stimuli, allowing predomi- 
nance of associative memory function. This combination of self-organization and associa- 
tive memory function may also occur in neocortical structures. The selective cholinergic 
suppression of feedback and intrinsic synapses has been proposed to allow self-organiza- 
tion of feedforward synapses while feedback synapses mediate storage of associations 
between higher level representations and activity in primary cortical areas (Hasselmo, 
1994b). This previous proposal could provide a physiological justification for a similar 
mechanism utilized in recent models (Dayan et al., 1995). Detailed modeling of cholin- 
ergic effects in the hippocampus provides a theoretical framework for linking the consid- 
erable behavioral evidence for a role of acetylcholine in memory function (Hagan and 
Morris, 1989) to the neurophysiological evidence for the effects of acetylcholine within 
cortical structures (Hasselmo and Bower, 1992; 1993; Hasselmo, 1994a, 1994b). 
Acknowledgements 
This work supported by a pilot grant from the Massachusetts Alzheimer's Disease 
Research Center and by an NIMH FIRST award MH52732-01. 
References 
Barkai E, Hasselmo ME (1994) Modulation of the input/output function of rat pitiform 
cortex pyramidal cells. J. Neurophysiol. 72: 644-658. 
84 Michael E. Hasselmo, Eric Schnell, Joshua Berke, Edi Barkai 
Barkai E, Bergman RE, Horwitz G, Hasselmo ME (1994) Modulation of associative mem- 
ory function in a biophysical simulation of rat piriform cortex. J. Neurophysiol. 72:659- 
677. 
Burgess N, Recce M, O'Keefe J (1994) A model of hippocampal function. Neural Net- 
works 7: 1065-1081. 
Dayan P, Hinton GE, Neal RM and Zemel RS (1995) The Helmholtz machine. Neural 
computation in press. 
Eichenbaum, H. and Buckingham, J. (1990) Studies on hippocampal processing: experi- 
ment, theory and model. In: Learning and computational neuroscience: foundations of 
adaptive networks, M. Gabriel and J. Moore, eds., Cambridge, MA: MIT Press. 
Hagan, JJ and Morris, RGM (1989) The cholinergic hypothesis of memory: A review of 
animal experiments. In Psychopharmacology of the Aging Nervous System, L.L. Iversen, 
S.D. Iversen and S.H. Snyder, eds. New York: Plenum Press, p. 237-324. 
Hasselmo, M.E. (1993) Acetylcholine and learning in a cortical associative memory. Neu- 
ral Comp. 5: 22-34. 
Hasselmo ME (1994a) Runaway synapfic modification in models of cortex: Implications 
for Alzheimer's disease. Neural Networks 7: 1340. 
Hasselmo ME (1994b) Neuromodulation and cortical function. Behav. Brain Res. in press 
Hasselmo ME, Anderson, BP and Bower, JM (1992) Cholinergic modulation of cortical 
associative memory function. J. Neurophysiol. 67(5): 1230-1246. 
Hasselmo ME, Bower JM (1992) Cholinergic suppression specific to intrinsic not afferent 
fiber synapses in rat pitiform (olfactory) cortex. J. Neurophysiol. 67(5): 1222-1229. 
Hasselmo ME, Bower JM (1993) Acetylcholine and memory. Trends Neurosci 16:218- 
222. 
Hasselmo ME, Barkai E, Horwitz G, Bergman RE (1993) Modulation of neuronal adapta- 
tion and cortical associative memory function. In: Computation and Neural Systems II 
(Eeckman F, Bower JM, ed). Norwell, MA: Kluwer Academic Publishers. 
Hasselmo ME, Schnell E (1994) Laminar selectivity of the cholinergic suppression of syn- 
aptic transmission in rat hippocampal region CAI: Computational modeling and brain 
slice physiology. J. Neurosci. 14: 3898-3914. 
Levy WB (1989) A computational approach to hippocampal function. In: Computational 
models of learning in simple neural systems (Hawkins RD, Bower GH, ed), pp. 243-305. 
Orlando, FL: Academic Press. 
Myers CE and Gluck M (1994) Context, conditioning and hippocampal rerepresentation 
in animal learning. Behav. Neurosci. 108: 835-847. 
Man' D (1971) Simple memory: A theory for archicortex. Phil. Trans. Roy. Soc. B 
B262:23-81 
McNaughton BL (1991) Associative pattern completion in hippocampal circuits: New 
evidence and new questions. Brain Res. Rev. 16:193-220. 
McNaughton BL, Morris RGM (1987) Hippocampal synaptic enhancement and informa- 
tion storage within a distributed memory system. Trends Neurosci. 10:408415. 
Touretzky DS, Wan HS and Redish AD (1994) Neural representation of space in rats and 
robots. In Zurada JM and Marks RJ (eds) Computational Intelligence: Imitating life. 
IEEE Press. 
