Crusio, Wim E. (1995) Natural Selection on Hippocampal Circuitry Underlying Exploratory Behaviour in Mice: Quantitative-Genetic Analysis. In: E. Alleva, A. Fasolo, H.-P. Lipp, L. Nadel and L. Ricceri (eds.), Behavioural Brain Research in Naturalistic and Seminaturalistic Settings. NATO Advanced Study Institutes Series D, Behavioural and Social Sciences, Kluwer Academic Press, Dordrecht, THE NETHERLANDS, p. 323-342, 1995.

Natural Selection on Hippocampal Circuitry Underlying Exploratory Behaviour in Mice: Quantitative-Genetic Analysis


Génétique, Neurogénétique et Comportement,
UFR Biomédicale,
Université René Descartes (Paris V)
45 rue des Saints-Pères,
75270 Paris Cedex 06,

1. Introduction

Behaviour is an animal's way of interacting with its environment and it is therefore a prime target for natural selection. As behaviour is the output of an animal's nervous system, this indirectly leads to selection pressures on neuronal structures. In consequence, each species' behaviour and nervous system have co-evolved in the context of its natural habitat and can be properly comprehended only when their interrelationships are regarded against that background [7].

This notion implies that to arrive at a profound understanding of neurobehavioral traits, one will have to consider problems of causation. Van Abeelen [59] distinguished between the phenogenetic and the phylogenetic aspects of causation. Both concern the genetic correlates of neurobehavioral traits, the first in a gene-physiological, the latter in an evolutionary sense. Stated otherwise, neurobehavioral geneticists attempt to uncover the physiological pathways underlying the expression of a trait and to provide an answer to the question of what exactly is the adaptive value of this trait for the organism.

As I have argued before [12, 13], quantitative-genetic methods may be employed with profit to address problems related to both aspects of causation. As an illustration of this research strategy, I present here the results of some experiments concerning mouse exploratory behaviour and hippocampal neuroanatomy.

2. The Phylogenetic Aspect of Causation

2.1. Natural Selection and Genetic Architecture

Selection pressures mould the species' genetic make-up which consequently will show traces of this past selection. Therefore, information about the genetic architecture of neurobehavioral traits might permit us to make inferences about the evolutionary history of these traits [5]. In its broadest sense, knowledge about that genetic architecture implies an understanding of the effects of genes governing a particular phenotype in a given population at a given time and includes information concerning the presence and size of certain genetic effects, the number of genetic units involved, etc. Generally, however, information about the presence and nature of dominance suffices [12].

With very few exceptions, natural, non-pathological variation in neurobehavioral phenotypes is polygenically regulated. If dominance is present, we may envisage two different situations: either (uni)directional or ambidirectional dominance. In the first case, dominance acts in the same direction for all genes involved (e.g., for high expression of the trait), whereas in the latter case it acts in one direction for some genes and in the opposite one for others. In its most extreme form, ambidirectional dominance may lead to situations where an F1 hybrid is completely intermediate between its parents, despite the presence of strong dominance effects.

Mather [39] distinguished between three kinds of selection: stabilising, directional, and disruptive. Stabilising selection favours intermediate expression of the phenotype, directional selection favours either high or low expression, whereas with disruptive selection more than one phenotypic optimum exists. The latter type of selection will lead to di- or polymorphisms, which may be in stable equilibrium or may even lead to breeding isolation and incipient speciation [53]. The commonest example of the former is the existence of two sexes, whereas a possible example of the latter are the explosive adaptive radiation and speciation found among fish species belonging to the family Cichlidae in the great East-African lakes [30].

Selection acts in favour of those genotypes that not only produce the phenotype selected for, but are also capable of producing progeny that differs little from this phenotype. In the long run this results in a population whose mean practically coincides with the optimum. Thus, stabilising and directional selection have predictably different consequences for the genetic architecture of a trait. Under directional selection genes of which a dominant allele is correlated with phenotypic expression opposite to the favoured direction will quickly become fixed for the recessive allele. The same applies to genes for which dominance is absent, i.e., where the heterozygote is intermediate between the two homozygotes. In contrast, selection against recessive alleles is much slower. The result will be that the first type of genes are not contributing to the genetic variation within the population anymore, whereas those genes where the allele correlated with the favoured phenotypical expression is dominant remain genetically polymorphic for a much longer time, conserving genetic variance. It may easily be seen that directional selection leads to situations where dominance is directional, in the same direction as the selection. Stabilising selection leads to situations where dominance is either absent or ambidirectional. Furthermore, directional selection generally results in lower levels of genetic variation than ambidirectional selection does.

The genetic architecture of a trait may be uncovered by using quantitative genetic methods (for a brief technical presentation of the most pertinent ones, I refer the reader to [12]).

2.2. Exploratory Behaviour

2.2.1. The Measurement of Behaviour: Naturalistic Approach

If we want to understand behaviour in the context of an animal's natural habitat, then we have to attempt to study behaviour either in the field under natural circumstances or in the laboratory under semi-natural conditions. To quantify the observed behaviour, we may profit from techniques developed by the field of ethology.

"Ethology is the objectivistic, biological study of behaviour and its first purpose is to obtain a total picture of the behaviour of organisms by means of direct observation of species-specific behaviour patterns" [59]. The application of such ethological methods in behaviour-genetic analyses was pioneered more than 30 years ago by van Abeelen [54], who stated that:

"For a further development of behaviour genetics it seems desirable that behaviour should be studied in all its multiformity; in this respect behaviour genetics may greatly profit from ethological attainments and procedures, among which the drawing up of ethograms, i.e. behaviour inventories, comes first" (p. 80).

With the help of an ethogram, seemingly continuous behaviour is described as a sequence of successive, mutually exclusive, and distinct motor-posture patterns that represent species-specific units of behaviour which may be quantified subsequently by measuring their frequency and/or duration. Complex behavioural responses are thus regarded as organised appearances of the behavioural units.

Defining Exploratory Behaviour.

Mice are attracted by novel stimuli and they spend long periods in exploration when exposed to a novel environment, even when satiated in every aspect. Although seemingly simple, some confusion exists on the precise definition of exploratory behaviour.

Most authors merely equate exploratory behaviour with "activity", "open-field behaviour", or even treat it as the opposite of "emotionality". This is a moot point. Some authors feel that even the more sharply defined locomotor activity in a runway or an open-field contains a non-exploratory component and they distinguish between "general activity" and exploratory activity (e.g., 29, 35, 50, 62]). Further, even to measure a "simple" phenotype like locomotor activity, many different devices are used such as photocell cages, running-wheels, open-fields, etc. More importantly, the different measures of activity almost always turn out to be very difficult to compare [2, 9, 45, 61].

Another point of consideration is that animals, and rodents in particular, often show a very rich behavioural repertoire in an open-field (see [54] for an extensive ethogram of the mouse; see also below). Although almost all behaviours are "activities", not all of them can be classified as exploratory. Several of these activities are measured using the above mentioned devices, one of the most frequently used being the open-field.

The concept of exploration is closely associated with that of novelty [2], which may involve some quality never previously experienced or familiar items arranged in an unfamiliar way. O'Keefe and Nadel [41] defined novelty within the framework of their cognitive map theory as follows:

"an item or place is novel if it does not have a representation in the locale system" and exploration as "a direct response of the animal to the detection of a mismatch by the locale system" (p. 241).

The locale system is their term for the cognitive mapping system, presumably located in the hippocampus, that contains mental representations of stimuli previous perceived. In other words, the hippocampal system supposedly signals a lack of information about the current environment.

Not surprisingly, therefore, one of the processes associated with exploratory activity is what is called latent learning or exploratory learning [38, 41, 44]. Latent learning occurs without overt reinforcement. If a satiated animal is allowed to explore a novel environment (for instance, a maze) and subsequently made hungry or thirsty, then the animal will quickly learn to go to the proper place to find food or water, more quickly than an animal lacking such previous experience [3]. Thus, animals acquire information about their surroundings by means of exploratory movements [35, 40]. The biological significance of exploration emerges clearly: entering and exploring new places promotes dispersion and improves the chances of finding life necessities (food, shelter, escape routes, etc.).

In conclusion then, we have formulated previously the following definition of exploration: "exploration is evoked by novel stimuli and consists of behavioural acts and postures that permit the collection of information about new objects and unfamiliar parts of the environment" [22].


The experiments described below address the question of what exactly is the adaptive value of various mouse exploratory behaviours carried out in novel surroundings. As discussed above, one result of exploration is the collection of new, or the updating of previously acquired, information [42, 45]. Obviously, if an animal enters a completely novel environment, it is of prime interest to collect as much information as possible in a short time. On the other hand, novel stimuli have, apart from informational, also stress- or anxiety-inducing [4, 10, 35], most probably because high exploration levels will render the animal more vulnerable to predation. Taken together, we hypothesise an evolutionary history of stabilising selection for exploration.

2.2.2. Methods

Animals and Behavioural Observation.

Male litter mates were housed 2-5, very occasionally singly, in plastic breeding cages with a metal cover and a bedding of wood shavings. Food and water were always available. Single male mice were observed at an age of 14 ± 3 weeks. The behavioural tests took place between 09.00 and 19.00 hours. To evoke exploratory behaviour, animals were placed individually in a novel environment: an illuminated observation cage measuring 109 x 49 x 49 cm. Against its back wall, a prismatic metal object was attached. Its floor was divided into 21 rectangles by painted lines and was wiped, but not rinsed, after sessions. The animals were observed directly and continuously for 20 min plus the time spent grooming and freezing. For full experimental details, the reader is referred to the references cited.


The ethogram used in the experiments described here was based on the one developed by van Abeelen [54], to which the reader is referred for more details. Briefly, behavioural components were defined as follows: Locomotor activity: the number of line crossings, disregarding the tail. Rearing: standing upright on the hind legs, while the forepaws are not touching any surface. Leaning: leaning against the wall; standing on its hind legs, the mouse places one or two forepaws against the wall. Leaning is often but not always combined with sniffing at the wall. Object-leaning: one or two forepaws are placed against the object; this posture is not always combined with sniffing at the object. Sniffing: the nose is held close to a particular spot while movements of the nasal skin take place. It should be noted here that the large differences in strain means obtained in our successive experiments are mainly due to different definitions used for this variable. In our initial experiments, sniffing was also scored while the animal was moving. Later, sniffing was recorded only when the animal was standing still. Object-sniffing: the nose is held close to the object or is actually touching it, showing the characteristic sniffing movements. Gnawing: occasionally animals gnawed at edges of the floor and walls. This was recorded if it was audible. Defecation: recorded by counting the boluses deposited. Grooming: these activities included face-cleaning, fur-licking, and scratching. Both bout frequency and duration were recorded. Freezing: the animal is, apart from breathing, completely motionless. This behaviour occurred only rarely and is not included in the analyses.


Summarised below are the results of one classical Mendelian cross and two diallel crosses. Highly inbred strains were used as parents for these crosses. A strain is considered inbred after at least 20 generations of rigorous sister x brother mating [37]. This procedure, at least in theory, produces homozygosity at 98 percent of all loci that were heterozygous at the start of inbreeding. All strains used in the present experiments had undergone between 80 and 200 generations of sib-mating [51] and should therefore closely approach complete homozygosity.

The classical cross consisted of the two parental inbred strains, C57BL/6J and DBA/2J, their reciprocal F1s, an F2 generation and the backcrosses between an F1 and both parentals. In total, it comprised 411 male mice [22].

A diallel cross consists of a number of inbred strains that are crossed in all possible combinations, including reciprocals. The first diallel cross was a 4x4 one, for which the inbred strains C57BL/6J, DBA/2J, C3Hf/St, and CPB-K were used as parentals, comprising 300 animals in all [22]. For the second diallel study the inbred strains BA, C57BL/6J, C57BR/cdJ, BALB/cJ, and DBA/2J, were used as parentals, giving a 5x5 diallel cross totalling 150 males [19].

Quantitative-Genetic Analyses.

At this point, a brief excursion into the field of quantitative genetics is necessary. For the sake of simplicity, considerations of possible interactions between genes (epistatic interactions) will be omitted from the present treatment. Pertinent references and more technical details may be found in [12].

The genetic contribution to a phenotype can be divided into two main sources: additive-genetic effects and dominance deviations. In the case of one single gene, with alleles A and a, we may denote the phenotypical values of the three possible genotypes as follows:

AA = m + da
Aa = m + ha
aa = m - da

The parameter da is used to represent half the difference between the homozygotes, ha designates the deviation of the heterozygote from the midparental point m. Note that in quantitative genetics capital letters are used to indicate increaser alleles, which are not necessarily also the dominant ones. Hence, da is positive by definition, whereas ha may attain all possible values. If we now consider an inbred strain (see above) in a situation where many genes affect the phenotype, we may denote its average phenotype by

m + S(d+) + S(d-) (2)

(shortened to m + [d] for ease of representation), where S(d+) indicates the summed effects of those genes that are represented by their increaser alleles and S(d-) indicates the same for decreaser alleles. Parameter m is a constant, reflecting the average environmental effects both strains have in common as well as genetic effects at loci where the strains are fixed for the same alleles. Similarly, the phenotypic value of an F1 hybrid may be written as

m + S(h+) + S(h­) (3)

(shortened to m + [h]). It must be noted that [h] is the sum of the dominance deviations of many genes. If these effects are balanced in opposite directions, [h] can be low or zero, even with dominance present. The same applies to [d], of course.

Because variations in a phenotype can be thought of as the summed effects of variations in genotype and environment, plus the interaction and covariation between these two factors, we may express the phenotypic variance of a population as:

P = G + E + G*E + 2cov(g,e) (4)

In the controlled situation of animal experiments in the laboratory (but not in the field) the covariance between genotype (g) and environment (e) can be minimised. Further, by choosing an appropriate measurement scale [11], the effects of genotype-environment interaction (G*E) may often be removed, leaving

P = G + E (5)

The genetic component of the variance (G) can, of course, be divided into components due to additive-genetic variation (D) and dominance deviations (H).

We may demonstrate the partitioning of genetic variance into its additive-genetic and dominance components by the example of an F2 between two inbred strains. When only one gene with two alleles influences the phenotype, the expected genetic composition of the F2 population will be AA, 25%; Aa, 50%; aa, 25%. From the foregoing, this leads to a phenotypic mean of

¼ da + ½ ha - ¼ da = ½ ha (6)

(m is set at 0 by a simple shift of the measurement scale). The sum of squares of deviations from the mean then equals

½ da2 + ¼ ha2 (7)

In the absence of epistasis and linkage, the contribution of a number of genes (k) to the F2 variance becomes

½ + ¼ (8)

shortened to

½ D + ¼ H (9)

for ease of representation. The total variance of an F2 is thus

½ D + ¼ H + E (10)

By using groups with different genetic compositions it will be possible to obtain estimates for the three parameters D, H, and E.

When investigating populations other than an F2 between two inbred strains (such as a diallel cross), allele frequencies need not be identical. Using u to indicate the frequency of the increaser allele and v as the frequency of the decreaser allele (with u = 1 - v) we may amend the definitions of D and H as follows:

D = ½ (11)


H = ¼ (12)

(Formally, this definition of H should be called H1, to distinguish it from H2, the other one of the two diallel forms of H; see [12]).

It should be noted that because D and H represent summations of the squared effects of single genes, they can only be zero if additive-genetic effects or dominance, respectively, are absent. This in obvious contrast to [d] and [h].

The cross-breeding designs employed in our studies, the classical Mendelian cross and the diallel cross, render different types of information on the genetic architecture of a trait. The classical cross permits very detailed genetic analyses and the detection of even very small genetic effects, but on a very restricted sample of two inbred strains, only. Furthermore, it is very hard when using this design to distinguish between directional and ambidirectional dominance, a significant parameter [h] only indicating that dominance is present and, at least, not completely balanced. The analysis of a diallel cross renders information on a larger genetic sample and is therefore much more generalizable, but this carries a price in that the information obtained is less detailed. A great advantage, however, is the possibility to distinguish between ambi- and unidirectional dominance.

It is difficult to compare results obtained from different crosses. Obviously, different results may be obtained depending on the genetic make-up of the parental strains used, especially if the cross-breeding design employed has a low generalisability (e.g., the classical cross). However, such results may be combined in order to provide a more complete and generalised picture of the genetic architecture of a trait. For instance, if one cross-breeding experiment indicates dominance in the direction of, say, high expression of the trait, but another cross indicates dominance in the opposite direction, then this constitutes prima-facie evidence for ambidirectional dominance. In fact, the presence of directional dominance may only be inferred if all available evidence indicates dominance is acting in the same direction.

2.2.3. Results and Discussion

Since, in the present treatment, we are solely interested in past natural selection pressures on mouse exploratory behaviour, only results regarding the absence or presence and direction of dominance will be shown here. More detailed results, including parameter estimates, heritabilities, etc., have been given elsewhere [19, 22].

Table 1 presents the findings obtained for the different elements of the ethogram in the classical cross and the two diallel crosses. We may now deduce the genetic architecture for each of the behavioural elements as follows. Locomotor activity: Both diallel crosses indicate ambidirectional dominance, which is not completely balanced in the C57BL/6 x DBA/2 cross. Rearing: The 5x5 diallel cross reveals the presence of ambidirectional dominance. The results of the 4x4 diallel cross indicate dominance for high scores, but are rendered questionable, because of the presence of epistatic interactions. Even if we accept this directional dominance, the fact that dominance is in the direction of low scores in the case of the classical cross, whose parental strains are included in the diallel sample, strongly suggests that dominance is, in fact, ambidirectional. Van Abeelen [57, 58] isolated one of the genetic units, for which strains C57BL/6 and DBA/2 differ, in his inbred selection lines SRH and SRL. A classical cross analysis of these two lines revealed that [h] is positive for this isolated genetic [22]. Yet we find a negative estimate of [h] when more genes are sampled in the cross between C57BL/6 and DBA/2. This is only compatible with an ambidirectional nature of the dominance. Leaning, Object-leaning and Gnawing: Both diallel crosses show ambidirectional dominance for these three variables. In the classical cross, there is a slight tendency for dominance towards higher scores. Sniffing: Ambidirectional dominance appears to be present, but it is rather weak, because no dominance at all was present in the 5x5 diallel cross. Object sniffing: Dominance towards high scores is found in the 4x4 diallel cross, but no dominance at all is present in the 5x5 diallel. In fact, in the latter cross no genetic variation at all is detected for object-sniffing. Defecation: The results obtained for this variable are rather confusing. No dominance is found in the classical cross, ambidirectional dominance in the 4x4 diallel cross (but suspect because of the presence of epistatic interactions), and dominance towards higher scores in the 5x5 diallel cross. Grooming frequency and duration: The genetic architectures of these two variables appear to entail weak ambidirectional dominance.

TABLE 1. Presence and direction of dominance obtained for behavioural responses to novelty in mice
Cross: Classical cross 4x4 diallel cross 5x5 diallel cross
Variable [22] [22] [19]
Locomotion - ambi ambi
Rearing - +? ambi
Leaning + ambi ambi
Object leaning + ambi ambi
Sniffing + ambi 0
Object-sniffing - + 0
Gnawing + ambi ambi
Defecation 0 ambi? +
Grooming freq. - ambi 0
Grooming dur. 0 ambi 0
+, dominance for high expression; -, dominance for low expression; ambi, ambidirectional dominance; 0, no dominance; ?, conclusion doubtful due to the presence of epistatic interactions.

For almost all variables the above results suggest the presence of, sometimes only weak, ambidirectional dominance. Object sniffing is an exception: no genetic variation at all was found in the 5x5 diallel cross. Such a situation can only be produced by a very high selection pressure, implying that every allele that causes a deviation away from the optimum is strongly selected against. The results of the 4x4 study suggest that this selection might have been towards high scores. Still, in the classical cross dominance was in the direction of low scores, presenting the possibility that very strong stabilising selection has depleted almost all genetic variation in the population. Taken together, the evidence from the three cross-breeding analyses strongly supports the hypothesis of an evolutionary past of stabilising selection for exploratory behaviour. As put forward above, the adaptive value of intermediate levels of exploratory activity probably derives from efficient gathering of useful information about the environment, on the one hand, and avoidance of predation, on the other.

2.3. Hippocampal Neuroanatomy

2.3.1. Hypothesis.

Natural selection does not act directly on a neural phenotype but modifies neural characters indirectly by way of selection pressures effectuated on behavioural characters. Because the hippocampus is intimately involved in the regulation of a number of different behaviours [8], selection must have played an important role in the evolution of the hippocampal neural network. Many questions remain, however, on the precise extent of this structure's implication in the regulation of behaviour [26]. For this reason it is rather difficult to formulate a strong, a priori hypothesis about the possible nature of the past selection.

The results of a study of exploratory behaviour in mice that had been made anosmic have led us to the conclusion that the physiological system involved in the processing of olfactory information has undergone selection in such a way that the development of a well-balanced regulatory mechanism has been promoted [23]. Several lines of evidence (see [60]) indicate that the most probable location of this mechanism is the hippocampal formation which, by way of the entorhinal cortex, receives a large input from the olfactory bulbs [41]. Furthermore, pharmacogenetic interventions imply that a well-balanced hippocampal neural network is necessary for the efficient execution of exploratory behaviour [59]. Taken together, these clues suggest that the sizes of different hippocampal terminal fields may have undergone stabilising selection in the evolutionary past.

2.3.2. Methods


Rearing conditions were essentially similar to those described above. Mice from two different cross-breeding experiments were analysed: 150 males from the 5x5 diallel cross between the inbred mouse strains BA, C57BL/6J, C57BR/cdJ, BALB/cJ, and DBA/2J (see ch. 2.2.2.) and 69 male and 72 female mice from a classical Mendelian cross between strains DBA/2J and C3H/HeJ consisting of the two parents, their F1 and F2, and the two backcrosses between the F1 and the parents. Animals from the diallel cross were perfused and processed at an age of 16±3 weeks, those from the classical cross at an age of 13±1 weeks. Full experimental details have been presented elsewhere [15, 32].

Histology and Morphometry.

After perfusion, brains were processed for Timm's staining [24], visualising the terminal fields of the hippocampal projections in the form of coloured bands and patches. The hippocampal fields of CA3 and hilus (together referred to as regio inferior hereafter) were measured in 5 defined horizontal 40 µm sections taken randomly from either the right or the left hippocampus at the midseptotemporal level, drawn with the aid of a projection microscope (magnification 160x) and measured on a graphics tablet connected to a desk computer (for more details see [15, 49]). Areas of the strata oriens, radiatum, pyramidale, and lacunosum-moleculare, the mossy fibre (MF) terminal fields (hilus, suprapyramidal -SPMF-, intra- and infrapyramidal -IIPMF), and of the whole regio inferior were summed for all sections and the relative sizes of the various areas were expressed as percentages of the regio inferior.

TABLE 2. Presence and direction of dominance obtained for hippocampal variables in mice
Cross: Classical cross 5x5 diallel cross
Variable [32] [15]
Stratum oriens + 0
Stratum pyramidale + 0
Stratum radiatum 0 0
Stratum lacunosum-moleculare - +
Hilus (CA4) - 0
SPMF + +
Regio inferior (CA3+CA4) 0 0
+, dominance for high expression; -, dominance for low expression; 0, no dominance.

2.3.3. Results and Discussion

As in the case of exploratory behaviour, the present discussion will be limited to the genetic findings concerning dominance (Table 2). More detailed analyses may be found in [15] and [32].

Strata oriens and pyramidale, Hilus: In both crosses significant genetic variation is detected for these variables, but dominance emerges in the classical cross only. We conclude that the genetic architecture is characterised by only very low levels of dominance, suggesting an evolutionary history of stabilising selection. Stratum radiatum: Here, no dominance is found at all, although additive-genetic effects are clearly present. This points unequivocally to an evolutionary past of stabilising selection. Stratum lacunosum-moleculare: The diallel cross indicates dominance for a large stratum lacunosum-moleculare, whereas in the classical cross dominance in the other direction crops up. This indicates that the genetic underpinnings of the size of the stratum lacunosum-moleculare comprise ambidirectional dominance. Past stabilising selection is, again, inferred. SPMF: This is the only variable for which both crosses suggest the presence of directional dominance for high values. Thus, directional selection for proportionally large areas of this synaptic field has existed in the evolutionary past. Animals possessing larger SPMF projections, at least in the midseptotemporal hippocampus, appear to have had a selectional advantage over animals possessing a smaller one. IIPMF: In the classical cross dominance for larger terminal fields is obtained. Yet, in the diallel cross dominance is completely absent. Genetic variation is present at a very high level: approximately 53% of the total variation can be explained by genotypic variations. Taken together, an evolutionary history of stabilising selection may be inferred. Regio inferior: As a measure of the total area of the hippocampus, we have used the sum of the areas of CA3 and hilus. Although some weak genetic variation was detected in both the classical and the diallel crosses, dominance is completely absent. This genetic architecture appears to be compatible only with an evolutionary past of very strong stabilising selection.

In conclusion, then, past stabilising selection is implied for the relative areas of almost all hippocampal variables analysed. The present findings are in agreement with the notion that a well-balanced hippocampal neural network is selectionally advantageous.

3. The Phenogenetic Aspect of Causation

3.1. The Correlational Approach

3.1.2. Brain Lesions and the Locality Assumption

Recently, Farah [28] reviewed the problems connected with the use of the so-called locality assumption, that more or less equalises the function of a lesioned structure with the defects exhibited by the damaged brain and which is almost always invoked to interpret the results of lesion studies. In an elegant way, Farah [28] provided evidence that this reasoning may lead to false conclusions. An additional disadvantage of lesion studies is that, as illustrated by the genetic analyses above, large interindividual differences in brain structure exist. This heritable variation of the brain is an aspect that many neuroscientists tend to ignore, most likely at their own peril. For example, widely divergent behavioural effects of septal [25, 27] or limbic-system lesions [1] have been reported in mice, depending on which particular inbred strain was being used.

It appears that, in the field of neurobehavioral genetics, an alternative approach that does not suffer from these drawbacks exists: using genetic methods exploiting naturally occurring individual differences as a tool for understanding brain function. No brain is like another and every individual behaves differently. The assumption that there is a link between the variability of the brain and individual talents and propensities seems quite plausible. This approach differs from the usual one in neuropsychology in two important aspects. First, no subjects are studied that, by accident or by design, have damaged brains. Rather, all subjects fall within the range of normal, non-pathological variation (provided animals carrying deleterious neurological mutations are excluded). Second, instead of comparing a damaged group with normal controls, we study a whole range of subjects and try to correlate variation at the behavioural level with that at the neuronal level. In the present framework the correlational approach has a further important advantage in that it is evidently easier to apply in (semi)natural situations than an interventionist approach.

This non-invasive strategy is reminiscent of the phrenological approach propagated by Franz Josef Gall (1758-1828); Lipp has coined the name "microphrenology" for it [36]. It appears that, as long as variation in one neuronal structure is independent of that in another, there will be no need for a locality assumption to interpret results of experiments carried out along these lines. Especially when used in combination with methods permitting the estimation of genetic correlations [12, 13], this strategy yields a very powerful approach.

3.1.2. Genetic Correlations

A weakness inherent in correlational studies is that a phenotypical correlation between characters does not necessarily reflect a functional relationship. On the other hand, if two independent processes, one causing a positive relationship, the other causing a negative relationship, act simultaneously upon two characters, the effects may cancel each other so that no detectable correlation can emerge. These problems can to a large extent be avoided by looking at the genetic correlations, that is, at correlations between the genetic effects that influence certain characters. These are the products of either genes with pleiotropic effects or of linkage disequilibrium. By using inbred strains that are only distantly related, the probability that a linkage disequilibrium occurs may be minimised so that a possible genetic correlation will most probably be caused by pleiotropy, that is, there exists a (set of) gene(s) influencing both characters simultaneously. Thus, for these characters, at least part of the physiological pathways leading from genotype to phenotype must be shared and a causal, perhaps also functional, relationship must exist. It is this special property that renders the genetic-correlational approach so uniquely valuable.

3.2. Hippocampal Mossy Fibres and Exploratory Behaviour

3.2.1. Hippocampal Regulation of Exploration

Although many different theories exist addressing the question of the proper function of the hippocampus, most agree, more or less, that this structure is intimately involved with the processing of information about the environment [47]. This notion is supported by evidence from lesion studies [31], pharmacogenetic findings [56, 60], and electrophysiological data, such as the observation that dentate synapses become potentiated during exploratory learning in rats [40].

The information collected during exploration is mainly of a spatial nature [41, 43]. According to O'Keefe and Nadel's "cognitive map" theory [41], the information acquired permits the animals to construct an internal representation of the spatial properties of their environment in their hippocampus. Thus, if a novel environment is entered, the hippocampus, acting as a comparator, detects this novelty and initiates exploratory behaviour, thereby enabling the animal to collect more information about that environment. The animal will gradually become familiarised with it and exploration will wane (habituation). It is known that hippocampal lesions produce hyperactivity, without habituation becoming evident [41]. One may thus expect to find correlations between hippocampal structural variation and novelty-induced exploration. The magnitude and sign of these correlations could provide important clues to hippocampal functioning.

3.2.2. Hypothesis

The experiment reviewed here attempts to provide answers to the following questions. 1. Do covariations between exploratory behaviour and hippocampal neuroanatomy exist? 2. In that case, what are their genetic underpinnings? 3. What are the implications of the foregoing for our understanding of hippocampal involvement in the regulation of exploratory behaviour?

Given the foregoing, we may hypothesise that sizeable anatomico-behavioural correlations do, in fact, exist. Furthermore, as selective forces have acted on the hippocampus indirectly by affecting behaviour, it is to be expected that such correlations will have an important genetic component. Finally, negative correlations are to be expected between the size of the IIPMF and exploration. The reasoning behind this is as follows. In a number of experiments concerning spatial learning tasks in radial mazes, it has been shown that larger IIPMF facilitate acquisition in such tasks [16, 18, 48]. Furthermore, mice possessing the larger IIPMF projections show larger behavioural changes when observed in the open-field for a second time [14]. Taken together, this strongly suggests that larger IIPMF facilitate the processing and/or storage of spatial information. Consequently, we may expect that animals with larger IIPMF will habituate faster to a novel environment. and, overall, show lower levels of exploratory activity.

3.2.3. Methods

Animals, Behavioural Observation, and Histology.

For this experiment the 150 animals of the 5x5 diallel cross described in ch. 2 were used. Experimental details have been presented above and in [20].

Quantitative-Genetic Analyses

In the univariate analysis, we may partition the phenotypic variation into its components E, D, and H, the environmental, additive-genetic, and dominance contributions. In the bivariate analysis, the covariation between two traits x and y is partitioned into its equivalent components Exy, Dxy, and Hxy. We may define the latter two parameters as:

Dxy = ½ (13)


Hxy = ¼ (14)

where dix and diy are the additive-genetic effects of either one of the alleles of the ith gene on characters x and y, respectively, and hix and hiy are the respective dominance deviations due to the ith gene. Evidently, only genes that have effects on both of the characters x and y contribute to the genetic covariance terms, whereas all genes that affect either x or y contribute to the respective genetic variance terms. Combining these components of the covariance with the components of the variance obtained in the univariate analyses we may estimate genetic correlations as

rD = Dxy/ (15)


rH = Hxy/ (16)

As is the case with normal correlations, genetic correlations are bound by -1 and 1. If a genetic correlation equals unity, then all genes affecting character x also affect character y proportionally. Only in case no gene at all affects both characters simultaneously or in some balanced cases (for instance, if some genes affect both characters in the same direction whereas others do so in opposite directions), will a genetic correlation become zero.

Testing the significance of genetic correlations is problematic. In the present case, theoretical standard errors of Exy and Dxy were calculated according to a method described previously [13]. The significance of environmental and genetic correlations was then evaluated by testing the deviations of Exy and Dxy from zero. The power of this test remains unknown for the moment. Fortunately, when the environmental and genetic correlations are used as input for further, multivariate, analyses, the possible significance or lack thereof of an individual correlation is no longer very important.

3.2.4. Results and Discussion

Genetic correlations.

Since we are mainly interested in the relationships between variation in the IIPMF and exploration, only the relevant parts of the correlation matrices are presented in Table 3; correlations involving other hippocampal areas and correlations among behavioural components are omitted. Furthermore, no dominance was found for the IIPMF (ch. 2.2.4). Consequently, dominance correlations could not be estimated.

To aid in the interpretation of the large correlation matrices thus obtained, we performed factor analyses, using the largest correlation coefficient of a certain variable as an estimate of its prior communality. Only those factors that showed eigenvalues > 1 were retained and an orthoblique Harris-Kaiser rotation was applied [46]. The results of the factor analysis on the phenotypical correlation matrix may be found in Table 4, whereas those of the factor analyses of the environmental and additive-genetic correlation matrix are presented in Tables 5 and 6, respectively.

TABLE 3. Phenotypical, environmental, and additive-genetic correlations between the size of the intra- and infrapyramidal mossy fibre terminal fields and behavioural components displayed by mice in an open-field
Type of the correlation
Variable Phenotypical Environmental Additive-genetic
Locomotion -0.073 0.029* -0.086*
Leaning -0.072 -0.001 -0.286*
Rearing 0.138 -0.083* 0.479*
Sniffing -0.107 0.099* -0.574*
Object-sniffing -0.040 -0.119* ------
Object-leaning -0.145 -0.147* 0.000
Grooming freq. 0.078 0.051* 0.128*
Grooming dur. -0.014 0.053* -0.302*
Gnawing -0.146 -0.184* -0.326*
Defecation -0.139 0.088* -0.704*
------, not calculated; * P < 0.05.

As predicted, many substantial genetic correlations were found between the IIPMF and the behavioural responses to novelty displayed in an open-field. The value of the genetic approach becomes apparent when comparing phenotypical, genetic and environmental correlations (Table 3). In a number of instances, environmental and genetic correlations showed opposite signs. This cancellation of effects resulted in non-significant phenotypical correlations (i.e., correlations between the 150 individual values) and in low loadings of the IIPMF on the three factors found in the factor analysis of the phenotypical correlations. Yet, the factor analysis of the matrix of additive-genetic correlations revealed a close relationship between the IIPMF and two of the three behavioural factors.

TABLE 4. Factor analysis of the matrix of phenotypical correlations between behaviour in the open-field and the size of the intra- and infrapyramidal mossy fibre terminal fields in mice
Variable Factor I Factor II Factor III
Gnawing 0.58 0.31
Defecation 0.48
Sniffing 0.43
IIPMF -0.32
Rearing -0.55 0.33
Leaning 0.74
Locomotion 0.64
Object-leaning 0.63
Grooming dur. 0.92
Grooming freq. 0.91
Only loadings with a value > |0.30| are shown. Variables for which no additive-genetic variance was present were excluded from the analysis.

Disregarding some minor deviations, the main differences between the three different factor analyses lay in the different loadings obtained for the IIPMF. For this variable, loadings in the phenotypical and environmental factor analyses were low or very low, in contrast with the two loadings (one intermediate, one high) obtained in the additive-genetic factor analysis. That environmental correlations between behaviour and the IIPMF (and, indeed, all other hippocampal variables, see [20]) are very low is understandable in view of the nature of environmental effects. Part of the "environmental" variance for a particular variable must be caused by chance fluctuations and measurement error. This error varies randomly over variables and does not contribute to the environmental covariation. The rest of the environmental variance is brought about by microenvironmental influences on the phenotype studied. In the cases of behavioural variables, these microenvironmental effects result from such factors as ambient temperature and slight sounds or movements made by the observer. Naturally, such effects do not affect hippocampal anatomy; that can only be affected by microenvironmental variations during its development. In conclusion, low environmental hippocampus-behaviour correlations are to be expected.

TABLE 5. Factor analysis of the matrix of environmental correlations between behaviour in the open-field and the size of the intra- and infrapyramidal mossy fibre terminal fields in mice
Variable Factor I Factor II Factor III
Gnawing 0.66
Defecation 0.33
Rearing -0.58
Grooming freq. 0.91
Grooming dur. 0.90
Leaning 0.74
Locomotion 0.65
Object-leaning 0.57
IIPMF -0.12
Sniffing -0.30
Only loadings with a value > |0.30| are shown (except for IIPMF). Variables for which no additive-genetic variance was present were excluded from the analysis.

Because low environmental correlations, even if they are of the same sign as the genetic ones, tend to decrease the magnitude of the phenotypical correlations, the additive-genetic correlations will most clearly reflect the relationship between behavioural and neuroanatomical variation. Three factors were found in the analysis of the additive-genetic correlation matrix. The second factor is dominated by grooming (both frequency and duration) and can be interpreted as self-maintaining behaviour, but the first and third factors are more important for the present discussion.

TABLE 6. Factor analysis of the matrix of additive-genetic correlations between behaviour in the open-field and the size of the intra- and infrapyramidal mossy fibre terminal fields in mice
Variable Factor I Factor II Factor III
Leaning 1.11 0.38
Object-leaning 0.95
Locomotion 0.78
Gnawing 0.75 -0.58 0.39
Grooming freq. 1.03
Grooming dur. 1.01
Sniffing 0.78 0.51
Defecation -0.34 0.96
Rearing 0.60 -0.66
IIPMF -0.40 -0.86
Only loadings with a value > |0.30| are shown. Variables for which no additive-genetic variance was present were excluded from the analysis.

The first factor shows positive loadings for behavioural variables that can supply information to the mouse about its environment and appears to represent exploration. As hypothesised above, the IIPMF show a negative loading on this factor. Finally, the third factor has a positive loading on defecation, a behaviour that is usually seen as indicating stress or fear [63]. The IIPMF show a negative loading on this factor, too. It appears that animals with larger IIPMF projections show lower levels of exploration and are not very fearful. This may mean that, within a 20 min observation period, animals with large IIPMF projections collect information in such an efficient way that their levels of exploration and fear are lower than in other mice; the open-field has rapidly become less novel to them. This result is in agreement with our hypothesis that larger IIPMF projections facilitate efficient information processing.

Additional experiments.

One of the strongest additive-genetic correlations found between the IIPMF and behaviour was that with rearing. Because of the low environmental correlation, which in addition had a sign opposite to that of the additive-genetic correlation, only a low and non-significant phenotypical correlation was obtained. Yet, the sizeable, positive additive-genetic correlation (0.479) implies that there exist pleiotropic genes that influence these two phenotypes in the same direction. Selective pressures on rearing should thus provoke neuroanatomical changes in the hippocampus. This hypothesis was tested by examining the inbred selection lines SRH (selection for rearing: high) and SRL (selection for rearing: low), that have been developed by van Abeelen [55, 56]. As expected, SRH mice possessed IIPMF terminal fields that were larger than those of SRL mice [17].

Subsequently, the serendipitous appearance of a mutation in the C57BL/6J inbred strain permitted a further test of the relationship between the IIPMF and rearing. The C57BL/6J//Nmg subline displayed a marked drop in the frequency with which this behaviour is displayed in an open-field when compared with the original C57BL/6J subline. Again, as expected on the basis of the positive genetic correlation between rearing and the IIPMF, the mutated subline was shown to have smaller IIPMF [21]. Furthermore, it showed a poorer learning performance in a spatial radial-maze task [33]. A cross-breeding analysis indicated that the behavioural and neuroanatomical differences between these two sublines are most probably due to only one single genetic unit [34], providing the opportunity to further dissect this relationship between neuroanatomy and behaviour by using neurochemical and molecular-genetic methods.

The results of these two experiments convincingly confirm the validity of the results of the diallel cross. Unfortunately, the latter experimental design requires a large investment in resources and effort in order to breed and test animals from many different groups. The main alternative for the diallel cross as a tool for the genetic dissection of neural and behavioural phenotypes is the estimation of genetic correlations, using a battery of inbred strains. In this approach, we "magnify" individual differences by studying animals from different inbred strains and looking for correlations between the means obtained for different variables (see [16] and references therein for some illustrative examples). Of course, the latter method does not allow the estimation of genetic correlations due to dominance or a test of the assumptions underlying the analysis. On the other hand, because no cross-breeding is necessary, we may use it as a "quick, but dirty" pilot analysis (see [12] for a more elaborate comparison of the different alternative quantitative-genetic methods available). Even in species for which no inbred strains are available, quantitative-genetic methods for estimating genetic correlations have been applied successfully [6].

4. Conclusion

Genetic methods are increasingly being applied to elucidate brain-behaviour relationships. Usually, however, these approaches utilise identified, single genes through the production of transgenic or knock-out mice [52]. In the present chapter, I have attempted to show that quantitative-genetic methods, utilising polygenic variation caused by unidentified genes, may also be used fruitfully to provide answers to questions concerning both the phenogenetic and the phylogenetic aspects of causation. In this way, it appeared possible to show that both exploratory behaviour displayed in an open-field and neuroanatomical variation of the hippocampus have been subjected to past stabilising selection (ch. 2). Furthermore, a close involvement of the hippocampus in the regulation or modulation of behavioural factors associated with exploration and stress or fear could be shown to exist (ch. 3). It should perhaps be emphasised here that this relationship could not be demonstrated at the phenotypical level, but was revealed only after analysing the genetic correlations.

In conclusion, then, quantitative-genetic experiments carried out in natural or seminatural situations, rendering information on the genetic architecture of a trait and about the multivariate genetic structure of complexes of traits are a valuable additional tool in the neuroscientist's arsenal and will greatly enhance our understanding of the genetic and neural foundations of behaviour.

5. Acknowledgements

Parts of the work described here have been supported by a NATO Science Fellowship, awarded by the Netherlands Organization of Pure Research (ZWO), an Alexander-von-Humboldt stipend, CNRS (URA 1294), DRED, and the Fondation pour la Recherche Médicale. I would like to thank Drs. Hans van Abeelen (Nijmegen, The Netherlands) and Herbert Schwegler (Freiburg, F.R. Germany), with whom I had the privilege and pleasure to collaborate on the experiments described in this chapter, for their collegiality, help, and friendship. Drs. Hans van Abeelen, Lynn Nadel (Tucson, Arizona), and Pierre Roubertoux (Paris, France) critically read the manuscript. Our students, Dr. Bernd Heimrich, Mrs. Gaby Genthner-Grimm, Dr. Ingrid Brust (Heidelberg, F.R. Germany), and Mrs. Laure Jamot (Paris, France), as well as Anton Buis, Ron Engels, and Tony Coenen (Nijmegen, The Netherlands) provided valuable technical assistance. Finally, I am grateful to Profs. Friedrich Vogel and Werner Buselmaier (Heidelberg, F.R. Germany) for their hospitality and for providing experimental facilities during my stay at the Institute of Human Genetics and Anthropology.

6. References

1. Ammassari-Teule, M., Fagioli, S., and Rossi-Arnaud, C. (1992) Genotype-dependent involvement of limbic areas in spatial learning and postlesion recovery, Physiol. Behav. 52, 505-510.

2. Barnett, S.A. and Cowan, P.E. (1976) Activity, exploration, curiosity and fear: An ethological study, Interdisc. Sci. Rev. 1, 43-62.

3. Blodgett, H.C. (1929) The effect of the introduction of reward upon the maze performance of rats, Univ. Calif. Publ. Psychol. 4, 113-134.

4. Brennan, M.J., Allen, D., Aleman, D., Azmitia, E.C., and Quartermain, D. (1984) Age differences in within-session habituation of exploratory behavior: Effects of stimulus complexity, Behav. Neural Biol. 42, 61-72.

5. Broadhurst, P.L. and Jinks, J.L. (1974) What genetical architecture can tell us about the natural selection of behavioural traits, in J.H.F. van Abeelen (ed.), The Genetics of Behaviour, North-Holland, Amsterdam, pp. 43-63.

6. Brodie, E.D. III (1989) Genetic correlations between morphology and antipredator behavior in natural populations of the garter snake Thamnophis ordinoides, Nature 342, 542-543.

7. Camhi, J.M. (1993) Neural mechanisms of behavior, Curr. Opinion Neurobiol. 3, 1011-1019.

8. Chozick, B.S. (1983) The behavioral effects of lesions of the hippocampus - A review, Int. J. Neurosci. 22, 63-80.

9. Corey, D.T. (1978) The determinants of exploration and neophobia, Neurosci. Biobehav. Rev. 2, 235-253.

10. Crawley, J.N. (1985) Exploratory behavior models of anxiety in mice, Neurosci. Biobehav. Rev. 9, 37-44.

11. Crusio, W.E. (1990) HOMAL: A computer program for selecting adequate data transformations, J. Hered. 81, 173.

12. Crusio, W.E. (1992) Quantitative Genetics, in D. Goldowitz, D. Wahlsten, and R. Wimer (eds.), Techniques for the Genetic Analysis of Brain and Behavior: Focus on the Mouse, Techniques in the Behavioral and Neural Sciences, Volume 8, Elsevier, Amsterdam, pp. 231-250.

13. Crusio, W.E. (1993) Bi- and multivariate analyses of diallel crosses: A tool for the genetic dissection of neurobehavioral phenotypes, Behav. Genet. 23, 59-67.

14. Crusio, W.E. and Schwegler, H. (1987) Hippocampal mossy fiber distribution covaries with open-field habituation in the mouse, Behav. Brain Res. 26, 153-158.

15. Crusio, W.E., Genthner-Grimm, G., and Schwegler, H. (1986) A quantitative-genetic analysis of hippocampal variation in the mouse, J. Neurogenet. 3, 203-214.

16. Crusio, W.E., Schwegler, H. and Brust, I. (1993) Covariations between hippocampal mossy fibres and working and reference memory in spatial and non-spatial radial maze tasks in mice, Eur. J. Neurosci. 5, 1413-1420.

17. Crusio, W.E., Schwegler, H., Brust, I., and van Abeelen, J.H.F. (1989) Genetic selection for novelty-induced rearing behavior in mice produces changes in hippocampal mossy fiber distributions, J. Neurogenet. 5, 87-93.

18. Crusio, W.E., Schwegler, H., and Lipp, H.-P. (1987) Radial-maze performance and structural variation of the hippocampus in mice: A correlation with mossy fibre distribution, Brain Res. 425, 182-185.

19. Crusio, W.E., Schwegler, H., and van Abeelen, J.H.F. (1989) Behavioral responses to novelty and structural variation of the hippocampus in mice. I. Quantitative-genetic analysis of behavior in the open-field, Behav. Brain Res. 32, 75-80.

20. Crusio, W.E., Schwegler, H., and van Abeelen, J.H.F. (1989) Behavioral responses to novelty and structural variation of the hippocampus in mice. II. Multivariate genetic analysis, Behav. Brain Res., 32, 81-88.

21. Crusio, W.E., Schwegler, H., and van Abeelen, J.H.F. (1991) Behavioural and neuroanatomical divergence between two sublines of C57BL/6J inbred mice, Behav. Brain Res. 42, 93-97.

22. Crusio, W.E. and van Abeelen, J.H.F. (1986) The genetic architecture of behavioural responses to novelty in mice, Heredity 56, 55-63.

23. Crusio, W.E. and van Abeelen, J.H.F. (1987) Zinc-induced peripheral anosmia and behavioral responses to novelty in mice: A quantitative-genetic analysis, Behav. Neural Biol. 48, 63-82.

24. Danscher, G., and Zimmer, J. (1978) An improved Timm sulphide silver method for light and electron microscopic localization of heavy metals in biological tissues, Histochemistry 55, 27-40.

25. Donovick, P.J., Burright, R.G., Fanelli, R.J., and Engellenner, W.J. (1981) Septal lesions and avoidance behavior: Genetic, neurochemical and behavioral considerations, Physiol. Behav. 26, 495-507.

26. Eichenbaum, H., Otto, T., and Cohen, N.J. (1994) Two functional components of the hippocampal memory system, Behav. Brain Sci. 17, 449-517.

27. Fanelli, R.J., Burright, R.J., and Donovick, P.J. (1983) A multivariate approach to the analysis of genetic and septal lesion effects on maze performance in mice, Behav. Neurosci. 97, 354-369.

28. Farah, M. J. (1994) Neuropsychological inference with an interactive brain: A critique of the "locality" assumption, Behav. Brain Sci. 17, 43-104.

29. Foshee, D.P., Vierck, C.J., Meier, G.W., and Federspiel, C. (1965) Simultaneous measure of general activity and exploratory behavior, Perc. Motor Skills 20, 445-451.

30. Fryer, G. and Iles, T.D. (1972) The Cichlid Fishes of the Great Lakes of Africa: Their Biology and Evolution, Oliver and Boyd, Edinburgh.

31. Gray, J.A. and McNaughton, N. (1983) Comparison between the behavioral effects of septal and hippocampal lesions: A review, Neurosci. Biobehav. Rev. 7, 119-188.

32. Heimrich, B., Schwegler, H., and Crusio, W.E. (1985) Hippocampal variation between the inbred mouse strains C3H/HeJ and DBA/2: A quantitative-genetic analysis, J. Neurogenet. 2, 389-401.

33. Jamot, L., Bertholet, J.-Y., and Crusio, W.E. (1994) Hereditary neuroanatomical divergence between two substrains of C57BL/6J inbred mice entails differential radial-maze learning, Brain Res. 644, 352-356.

34. Jamot, L., Bertholet, J.-Y., and Crusio, W.E. (1994) Genetic analysis of hippocampal mossy fibers and radial-maze learning in two substrains of C57BL/6J inbred mice, Behav. Genet. 24, 518.

35. Kelley, A.E., Cador, M., and Stinus, L. (1989) Exploration and its measurement. A psychopharmacological perspective, in A.B. Boulton, G.B. Baker, and A.J. Greenshaw (eds.), Neuromethods, Volume 13: Psychopharmacology, Humana Press, Clifton, pp. 95-144.

36. Lipp, H.P., Schwegler, H., Crusio, W.E., Wolfer, D., Leisinger-Trigona, M.­C., Heimrich, B., and Driscoll, P. (1989) Using genetically-defined rodent strains for the identification of hippocampal traits relevant for two­way avoidance learning: A non-invasive approach, Experientia 45, 845-859.

37. Lyon, M.F. (1981) Rules for nomenclature of inbred strains, in M.C. Green (ed.), Genetic Variants and Strains of the Laboratory Mouse, Gustav Fischer, Stuttgart and New York, pp. 368-372.

38. Manning, A. (1979) An Introduction to Animal Behaviour, 3rd ed., Edward Arnold, London.

39. Mather, K. (1973) The Genetical Stucture of Populations, Chapman and Hall, London.

40. Moser, E.I., Moser, M.-B., and Andersen, P. (1994) Potentiation of dentate synapses initiated by exploratory learning in rats: Dissociation from brain temperature, motor activity, and arousal, Learn. Memory 1, 55-73.

41. O'Keefe, J. and Nadel, L. (1978). The Hippocampus as a Cognitive Map, Clarendon Press, Oxford.

42. Poucet, B. (1993) Spatial cognitive maps in animals: New hypotheses on their structure and neural mechanisms, Psychol. Rev. 100, 163-182.

43. Poucet, B., Chapuis, N., Durup, M., and Thinus-Blanc, C. (1986) A study of exploratory behavior as an index of spatial knowledge in hamsters, Anim. Learn. Behav. 14, 93-100.

44. Renner, M.J. (1988) Learning during exploration: The role of behavioral topography during exploration in determining subsequent adaptive behavior, Int. J. Comp. Psychol. 2, 43-56.

45. Renner, M.J. (1990) Neglected aspects of exploratory and investigatory behavior, Psychobiol. 18, 16-22.

46. SAS Institute Inc. (1987) SAS/STAT Guide for Personal Computers, Version 6 Edition, SAS Institute Inc, Cary, NC.

47. Schmajuk, N.A. (1984) Psychological theories of hippocampal function, Physiol. Psychol. 12, 166-183.

48. Schwegler, H. and Crusio, W.E. (1994) Covariations between radial-maze learning and structural variations of septum and hippocampus in rodents, Behav. Brain Res. (in press).

49. Schwegler, H. and Lipp, H.-P. (1983) Hereditary covariations of neuronal circuitry and behavior: correlations between the proportions of hippocampal synaptic fields in the regio inferior and two-way avoidance in mice and rats, Behav. Brain Res. 7, 1-38.

50. Simmel, E.C. and Bagwell, M. (1983) Genetics of exploratory behavior and activity, in J.L. Fuller and E.C. Simmel (eds.), Behavior Genetics: Principles and Applications, Lawrence Erlbaum Assoc., London, pp. 89-115.

51. Staats, J. (1980) Standardized nomenclature for inbred strains of mice: Sixth listing, Cancer Res. 36, 4333-4377.

52. Takahashi, J.S., Pinto, L.H., and Hotz Vitaterna, M. (1994) Forward and reverse genetic approaches to behavior in the mouse, Science 264, 1724-1733.

53. Thoday, J.M. (1972) Disruptive selection, Proc. R. Soc. Lond. B 182, 109-143.

54. van Abeelen, J.H.F. (1963) Mouse mutants studied by means of ethological methods. I. Ethogram, Genetica 34, 79-94.

55. van Abeelen, J.H.F. (1970) Genetics of rearing behavior in mice, Behav. Genet. 1, 71-76.

56. van Abeelen, J.H.F. (1974) Genotype and the cholinergic control of exploratory behaviour in mice, in J.H.F. van Abeelen (ed.), The Genetics of Behaviour, North-Holland, Amsterdam, pp. 347-374.

57. van Abeelen, J.H.F. (1975) Genetic analysis of behavioural responses to novelty in mice, Nature 254, 239-241.

58. van Abeelen, J.H.F. (1977) Rearing responses and locomotor activity in mice: Single-locus control, Behav. Biol. 19, 401-404.

59. van Abeelen, J.H.F. (1979) Ethology and the genetic foundations of animal behavior, in J.R. Royce and L.P. Mos (eds.), Theoretical Advances in Behavior Genetics, Sijthoff and Noordhoff, Alphen aan den Rijn, pp. 101-112.

60. van Abeelen, J.H.F. (1989) Genetic control of hippocampal cholinergic and dynorphinergic mechanisms regulating novelty-induced exploratory behavior in house mice, Experientia 45, 839-845.

61. Walsh, R.N. and Cummins, R.A. (1976). The open-field test: A critical review, Psychol. Bull. 83, 482-504.

62. Weischer, M.L. (1976) Eine einfache Versuchsanordnung zur quantitativen Beurteilung von Motilität und Neugierverhalten bei Mäusen, Psychopharmacology 50, 275-279.

63. Whimbey, A.E. and Denenberg, V.H. (1967) Two independent behavioral dimensions in open-field performance, J. Comp. Physiol. Psychol. 63, 500-504.