Effects of social environment on memory, hippocampal structure and neurogenesis.

Humans and most animal species live in very diverse environments and environmental conditions appear to play a critical role in developing and maintaining cognitive abilities, such as memory and their underlying neural mechanisms. Because humans and many other animal species are highly social, both social and physical environment could be important for maintaining mental health. Memory is a critical component of mental health and thus it is important to understand the relationship between environmental conditions and memory with its neural mechanisms. The hippocampus is a brain area that has been linked to memory processing in both mammals and birds and it is well known for its high levels of plasticity.  In particular, intensive neurogenesis and neuron replacements have been observed specifically in the hippocampus in both birds and mammals (Barnea & Nottebohm 1994; Gould et al. 1999, 2000; Banta Lavenex et al. 2001; Gage 2002; Kempermann 2002; Nottebohm 2002; Gould & Gross 2002; Prickaert et al. 2004). It has been hypothesized that adult neurogenesis plays an important role in memory function and neurogenesis regulation and its relationship with memory have been under intensive investigation (e.g. Kempermann 2002; Gage 2002; Nottebohm 2002).  

         Despite the fact that humans and most animal species live in highly diverse environments, most animal studies of memory and the brain are based on individuals maintained in impoverished laboratory conditions (review in van Praag et al. 2000). Such impoverished conditions seem to have a negative effect on memory and the hippocampus. Kempermann et al. (1997), for example, demonstrated that young mice placed in enriched environments had a larger hippocampus with more neurons and higher neurogenesis rates than those living in standard “poor” laboratory conditions. At the same time, Kempermann et al. (1998) failed to find an effect of environmental enrichment on hippocampal volume and neuron numbers in adult mice. Most other mammalian studies did not detect an effect of environmental enrichment specifically on hippocampal volume and neuron numbers in adult animals while detecting a significant effect of enrichment on neurogenesis rates (see van Praag et al. 2000 for review). Environmental enrichment does not always involve memory-based tasks, thus it is critical to separate the effects of memory-based experiences on the hippocampus from all other experiences occurring in the enriched environment. For example, it has been shown that increased physical activity (e.g. running) experienced by animals in artificially enriched environments is enough to increase hippocampal cell proliferation and neurogenesis in adult mice while learning experiences per se did not produce any additional effects (van Praag et al. 1999a,b). Gould et al. (1999), on the other hand, reported that hippocampus-dependent learning experience specifically affects adult neurogenesis. Clearly, the issue still remains unresolved because of difficulty separating different behaviors in complex environment and of showing that neural consequences of the enriched environment are specifically related to learning rather than to increased activity levels using the rodent model (van Praag et al. 2000).  

Most of the studies investigating how environmental enrichment affects memory and the brain manipulated mostly physical environment (see van Praag et al. 2000 for review). In natural conditions, however, both physical and social environments are usually much richer that those in most of laboratory studies. Humans and many animal species are highly social and thus social interactions are an inseparable part of their natural environment, yet social aspect of environmental enrichment has received less attention. As most of the laboratory animal studies are conducted in fairly unnatural social conditions, it is absolutely essential to understand whether and how social environment might affect behavior and the brain.

Most research concerning the effects of social environment on memory and the hippocampus has been focused on social dominance (e.g. Fuchs et al. 2001; Kozorovitskiy & Gould 2004). Many animals live in hierarchically structured groups in which dominants have unlimited access to all resources while subordinates might suffer social stress of subordination. Evidence from laboratory studies suggests that social stress resulting from dominance hierarchies in groups may indeed affect memory and the brain in both mammals and birds (Gould et al. 1997; Lucassen et al. 2001; Kozorovitskiy & Gould 2004; Pravosudov et al. 2003; Pravosudov & Omanska 2005). In laboratory rats, for example, compared to subordinate individuals dominant males had more new neurons in the dentate gyrus of the hippocampus irrespective of whether these animals were maintained in artificially enriched environment or in impoverished environment (Kozorovitskiy & Gould 2004). Interestingly, there were no differences between dominants and subordinates in cell proliferation rates suggesting that social dominance specifically influenced neuron survival rates (Kozorovitskiy & Gould 2004). There were also no differences between solitary animals and subordinates in the number of new neurons suggesting that only dominants benefited from being in a social group. An experiment with food-caching mountain chickadees demonstrated that in laboratory conditions dominant chickadees had better spatial memory and higher cell proliferation rates in the ventricular zone than their subordinate conspecifics (Pravosudov et al. 2003; Pravosudov & Omanska 2005). The later study did not specifically look at neuron survival rates and thus it remains possible that in chickadees social dominance affected both cell proliferation and neuron survival rates. Unlike the Kozorovitskiy & Gould (2004) study, preliminary data on mountain chickadees suggest that solitary birds had fewer hippocampal neurons and lower cell proliferation rates in the ventricular zone compared to both dominants and subordinates in social groups (see Preliminary data). This finding suggests that unnatural solitary environment might be more detrimental for the hippocampus than naturally occurring social subordination within social groups.

It has often been assumed that glucocorticoid hormones mediate the effect of social dominance on memory and the brain. Subordinates have been suggested to maintain elevated levels of glucocorticoids as a result of their low social status (e.g. Capitanio et al. 1998, Mendoza 2000; Fuchs et al. 2001) and elevated glucocorticoid levels might directly and negatively impact memory and the hippocampus (e.g. Sapolsky 1992; McEwen 2000a,b; Shors 2001; Shors et al. 2000). A few recent studies, however, demonstrated that the effects of social dominance on memory and the hippocampus are not necessarily related to changes in glucocorticoid hormones (Pravosudov et al. 2003; Kozorovitskiy & Gould 2004; Pravosudov & Omanska 2005). In both rats and mountain chickadees the baseline glucocorticoid levels were statistically indistinguishable between dominants and subordinates yet subordinates showed inferior memory performance (Pravosudov et al. 2003) and had lower cell proliferation (Pravosudov & Omanska 2005) and lower neuron survival rates in the hippocampus (Kozorovitskiy & Gould 2004). Thus more research is needed in order to understand the effect of social dominance and social environment on glucocorticoid hormones, memory and the hippocampus.

Most of the laboratory experiments investigating dominance hierarchy have not truly simulated natural environment. These studies used either only males or females to initiate strong dominance hierarchies (Fuchs et al. 2001; Lucassen et al. 2001; Pravosudov et al. 2003; Kozorovitskiy & Gould 2004). In natural conditions, on the other hand, most adult animals including humans usually form male-female pairs in addition to being in larger groups. In prairie voles, for example, social isolation results in reduced neurogenesis in adult females whereas exposure to males results in elevated neurogenesis levels (Smith et al. 2001; Fowler et al. 2002). In mound-building mice Mus spicilegus, females that preferred their sexual partner had increased levels of neurogenesis in their olfactory bulb (Baudoin et al. 2005). Males and females maintained in the same sex social groups might also have different responses to the unisex social conditions. In rats, experimentally induced stress decreased neurogenesis rates in solitary males but not in males maintained in unisex groups (Westenbroek et al. 2004). In female rats, on the other hand, experimentally induced stress increased neurogenesis rates in solitary individuals but not in animals maintained in social unisex groups (Westenbroek et al. 2004). Westenbroek et al. (2004) concluded that social environment affects neurogenesis rates differently in males and females, but there were no comparisons of animals maintained in male-female groups which is a more natural social environment. Preliminary data on mountain chickadees showed that irrespectively of their dominance status, both males and females maintained in unisex pairs had significantly more hippocampal neurons and higher cell proliferation rates in the ventricular zone compared to solitary males and females (See preliminary results). In chickadees, subordinate individuals showed impaired memory performance and lower cell proliferation rates in the ventricular zone compared to their dominant group mates (Pravosudov et al. 2003; Pravosudov & Omanska 2005), yet solitary chickadees had even fewer neurons and lower cell proliferation rates than subordinate but social birds (see preliminary results). Thus, being a subordinate in a social group might provide greater benefits for memory and the brain than being solitary. On the other hand, Kozorovitskiy & Gould (2004) found that neuron survival rates did not significantly differ between subordinate male rats and solitary rats whereas dominant male rates had significantly higher neuron survival rates than both subordinate and solitary animals. Lu et al. (2003) showed that maintaining young males in social unisex groups resulted in improved memory performance and increased neurogenesis rates irrespective of their dominance status (which was not measured in that study), but these results concerned only developing and not adult animals and it also remains unclear whether females follow the same pattern and whether animals growing in groups containing both males and females show different patterns of behavior and neurogenesis compared to individuals maintained in unisex groups. Maintaining the same sex groups might still provide an unnatural and impoverished social environment because natural groups in many animal species (including humans) almost always include both males and females. Many animal species also live in groups of larger than two individuals but it is unknown whether social environment consisting of groups larger than two individuals has any consequences for memory and the hippocampus. In adult zebra finches, for example, individuals maintained in larger groups had higher neuron survival rates in the forebrain than those maintained in pairs (Lipkind et al. 2002) but it is unknown whether differences in neuron survival rates also occurred in the hippocampus. 

Thus all available data suggest that social environment might have a strong effect on memory, hippocampal structure and neurogenesis rates but published studies using social dominance models in unisex groups do not provide naturalistic enough social complexity to fully understand the relationship between social environment, memory and the hippocampus. The following questions remain poorly investigated and need to be addressed directly: (1) do solitary animals have inferior memory, fewer hippocampal neurons and lower neurogenesis rates (resulting from either lower cell proliferation or lower neuron survival rates) compared to animals maintained in either the same sex or male-female pairs irrespective of their dominance rank? (2) do animals maintained in unisex pairs have inferior memory and fewer hippocampal neurons and lower neurogenesis rates compared to animals maintained in male-female pairs?, and (3) do animals maintained in groups larger than 2 individuals with both males and females have better memory, more hippocampal neurons and higher neurogenesis rates than either solitary animals or animals maintained in pairs (either unisex or male-female)? Answering these questions is critical for our understanding of how naturally occurring social environment might affect memory and its neural mechanisms and it could help better understand the neural consequences of social disorders in humans.

Food-caching birds present a great model to study the effects of social environment on memory and the hippocampus because of several reasons: (1) food-caching chickadees live in complex and stable social groups which contain several male-female pairs; (2) food-caching birds are extremely convenient for studies of spatial memory in controlled laboratory settings because these birds regularly cache food and recover previously made food caches using spatial memory; thus unlike in rodents, spatial memory in food-caching birds can be tested using natural paradigm of food caching and retrieval. Food-caching mountain chickadees have been successfully used in studies of spatial memory, stress hormones and the hippocampus (Pravosudov & Clayton 2001, 2002; Pravosudov et al. 2001, 2002, 2003; Pravosudov 2003; Pravosudov & Omanska 2005a,b). In the proposed study, I will investigate all possible combinations of social environment (solitary, male-male pairs, female-female pairs, male-female pairs and groups containing two male-female pairs) and their effect on spatial memory and the hippocampus. To investigate whether social environment also has an effect on glucocorticoid hormones, which could have a strong effect on memory and the brain, I will also measure baseline and stress-induced corticosterone levels. 

Social structure of mountain chickadees with several male-female pairs living in a group resembles that of majority of other animal species and so the results obtained in this study could be extrapolated to many other species, including humans. Chickadees form long-lasting pair bonds and such pairs form groups during non-breeding season. Such pairs are territorial during breeding season but in the winter the entire group defends their territory from other groups. There are certainly some more unusual social structures in both birds and mammals in which animals live solitary most of their life or form colonies with specialized functions for individuals. Results of the proposed experiment might be difficult to extrapolate to generally unsocial species, for example. However, the results of the proposed experiment could be extrapolated to most social species such as humans and thus mountain chickadees present a great model to gain understanding into general issues of social influences on memory, mental health and the brain. Certainly slight differences in social structure between species might result in slightly different effects of social environment on memory and the brain but the main issue here is to first establish whether social environment is important in regulating memory and brain functions.