My research has been focused on the effects of both physical and social environments on adrenocortical function, spatial memory and the hippocampus.
Most food caching birds live in highly seasonal environments; food is abundant during summer and early autumn but during the winter it is often limited and unpredictable. Food caching patterns mostly follow a seasonal cycle; birds cache most when food is plentiful outside of breeding season – late summer and early fall and, sometimes, in spring when there is abundance of conifer seeds (Haftorn 1956; Pravosudov 1985; Brodin 1994). During these periods, food-caching parids might cache thousands of food items (Haftorn 1956; Pravosudov 1985; Brodin 1994). These food caches appear to be crucial for winter survival and thus ability to retrieve them should be of paramount importance. Parids continue to cache throughout winter although they cache fairly small amount of food during that time (Pravosudov 1985). It is not clear whether winter caching involves storing newly found food or re-caching of previously hidden food stores (Pravosudov & Grubb 1997), but winter short-term caching might be important for birds’ survival on a daily basis (Pravosudov & Lucas 2001).
In seasonal environments, conditions vary during the seasons and demands for spatial memory do not appear to be equal throughout the year. If a bird does not remember where it stored food during summer and early autumn when natural food is abundant, it is unlikely to die of starvation. During the winter, however, failure to recover food caches is more likely to result in mortality and thus demands for spatial memory should be higher during the winter. Naturally available food is usually abundant in summer and autumn but becomes much less abundant during the winter. Winter is also most energetically demanding season because of short days and low air temperature, and successful retrieval of previously made food caches becomes critically important specifically during the winter (Pravosudov & Lucas 2001).
When food supply was made limited and unpredictable in a laboratory, mountain chickadees (Poecile gambeli) demonstrated more efficient cache recovery and showed better spatial memory performance compared to the individuals maintained on ad libitum food (Pravosudov & Clayton 2001). Thus, it appears that chickadees might have better spatial memory resulting in more efficient cache retrieval during the winter when available food is rather scarce.
Hippocampal volume and the total number of neurons, however, were statistically indistinguishable between mountain chickadees maintained on limited and unpredictable food supply and those on ad libitum food supply (Pravosudov et al. 2002) which suggests that changes in spatial memory triggered by changes in food supply were not supported by structural changes in the hippocampus. Interestingly, baseline corticosterone levels were significantly elevated in birds maintained on limited food (Pravosudov et al. 2001). Relatively long-term (two months) limited and unpredictable food supply resulted in moderately but significantly elevated baseline corticosterone and in enhanced spatial memory. Thus it is possible that elevated corticosterone was responsible for spatial memory improvements observed in mountain chickadees in the absence of any changes in hippocampal volume or neuron numbers.
Baseline corticosterone levels may vary seasonally in food-caching parids and highest corticosterone levels have been observed during the winter (Silverin 1998). Several proximate factors such as photoperiod, abundance of food, air temperature or a combination of these factors could be responsible for the observed variation in corticosterone levels. Experimental studies on mountain chickadees ruled out photoperiod (Pravosudov et al. 2002) because there were no significant differences in baseline cortiosterone levels between the birds maintained on a short, winter-like day and the individuals switched to a long day. Food supply, on the other hand, turned out to be a significant factor (Pravosudov et al. 2001). Thus enhanced spatial memory performance in birds maintained on limited and unpredictable food supply correlated with elevated corticosterone levels suggesting that elevated corticosterone could be responsible for spatial memory enhancements (Pravosudov & Clayton 2001; Pravosudov et al. 2001).
Alaska black-capped chickadees live in an environment which is predictably more energetically challenging that that of Colorado chickadees every winter (Pravosudov & Clayton 2002). In such conditions more food caching and better spatial memory for food caches can be crucial for survival. In identical laboratory conditions, Alaska chickadees cached more food and demonstrated better spatial memory than Colorado birds suggesting that such differences are not simply caused by immediate environment (Pravosudov & Clayton 2002). However, there were no significant differences in either baseline or stress-induced corticosterone levels between Alaska and Colorado chickadees which indicates that differences in memory found between these two populations were not related to changes in corticosterone levels (Pravosudov et al. 2004). The fact that Alaska and Colorado chickadees did not differ in their corticosterone levels in identical laboratory conditions also suggests that different environmental conditions did not cause permanently elevated corticosterone levels in Alaska birds due to their predictably more energetically challenging environment. Instead, it appears that baseline corticosterone levels might only respond to immediate environmental conditions in both populations.
Compared to birds in Colorado, corticosterone levels in Alaska environment may be elevated in black-capped chickadees during the winter as a result of differences in immediate conditions between Alaska and Colorado during the winter. Such elevation (if it exists) may aid spatial memory benefits achieved through enlarged hippocampus (Pravosudov & Clayton 2002). Whereas the volume of the hippocampus and its number of neurons appear to be permanently larger in Alaska chickadees providing them with permanent advantages when it comes to spatial memory performance, temporary corticosterone elevations during most difficult times may provide additional memory benefits (Pravosudov 2004).
Most parids live in hierarchically structured social groups in which dominants enjoy priority of access to all available resources (Ekman 1989). It has been suggested that subordinates experience less predictable conditions because of their low social status; even when they find food they can be supplanted by dominants at any time (Ekman & Lilliendahl 1993). Because social classes differ in predictability of their environment they might involve different tactics of managing their available energy resources. In particular, it has been suggested that birds in less predictable environment should accumulate and maintain more energy reserves as a buffer against higher probability of starvation in case adequate amount of food cannot be obtained at a critical time (Ekman & Lilliendahl 1993; Brodin et al. 2001). Indeed, there is some experimental evidence that subordinate birds maintain higher levels of fat reserves than dominants (Ekman & Lilliendahl 1993 – willow tits), but other studies found either a reverse pattern (Koivula et al. 1995 – willow tits; Verhulst & Hogstad 1996 – willow tits; Pravosudov & Lucas 2000 – Carolina chickadees, P. carolinensis) or no differences between dominants and subordinates (Pravosudov & Lucas 2000 – Carolina chickadees).
Most of the studies investigating the effect of dominance hierarchy on food caching behavior used behavioral and ecological approaches while ignoring possible mechanisms of food caching regulation. For example, unpredictable food triggers more intense food caching (Hurly 1992 - marsh tits; Pravosudov & Grubb 1997 – tufted titmice, Baeolophus bicolor) and at the same time it also triggers corticosterone elevation (Pravosudov et al. 2001 – mountain chickadees). Implant study with mountain chickadees (Pravosudov 2004) showed that elevated corticosterone facilitates more eating, more intense food caching and enhanced spatial memory. Thus, corticosterone might be a likely mechanistic link between unpredictable foraging conditions and energy management tactics. If socially subordinate individuals experience unpredictable environment their corticosterone levels might be elevated, which would support the idea that subordinates should cache more food and even have better memory for food caches. Available data on the relationship between dominance status and corticosterone levels from observational unmanipulated studies, however, do not show a consistent pattern (see Pravosudov et al. 2004 for review). In some studies corticosterone levels were higher in subordinate individuals whereas in other studies there were no differences in baseline corticosterone levels between dominants and subordinates.
To test the hypothesis that dominance hierarchy has an effect on food caching and spatial memory, we compared experimentally created pairs of dominants and subordinates in mountain chickadees. Previous laboratory studies investigating the effect of social dominance on food caching failed to find significant differences between dominants and subordinates (Pravosudov & Lucas 2000; Lundberg & Brodin 2003). However, in all of these studies, dominant and subordinates were tested together in a fairly small room in which subordinates could not escape the dominant’s view. Thus it is likely that immediate presence of dominants might have influenced the behavior of subordinates in these studies. In natural conditions, subordinates can easily move out of dominant’s sight, and in fact, most birds avoid caching when in a presence of a group mate (Woodrey 1991; Lahti et al. 1998). We attempted to simulate more natural environment by testing dominants and subordinates separately in an adjacent room. In this experimental design, subordinate were not affected by the presence of the dominants during the short duration of food caching trials. Results of that study were directly opposite to all predictions; subordinates cached significantly less food than dominants and they also performed significantly worse in a cache recovery task. Baseline corticosterone levels measured after almost two months following formation of dominance relationships were statistically indistinguishable between dominants and subordinates and maximum corticosterone levels achieved during a standardized stress response were actually higher in dominants. These results suggest that whereas corticosterone might be involved in mediating food caching behavior and spatial memory, changes in caching rates and memory can occur without changes in baseline corticosterone levels. Inferior spatial memory found in subordinate chickadees was not related to any changes in hippocampal volume or neuron numbers either (Pravosudov & Omanska 2005a).