Although much attention has been focused on the hypothalamus as the site of detection for physiological (e.g., hormonal, metabolic) signals relating to energy balance, a number of findings suggest that the hippocampus may be involved with utilization of the information that is provided by those signals. For example, rats with selective neurotoxic lesions of the hippocampus exhibit increased food intake, body weight gain, metabolic activity, and appetitive behavior compared to rats with an intact hippocampus and to rats with damage confined to the medial prefrontal cortex (Davidson et al., 2009
). These findings complement recent results from (a) neuroanatomical studies (e.g.,Cenquizca & Swanson, 2006
) which identified direct projections in rats from hippocampal cell fields to brain regions (e.g., the lateral hypothalamus) that are known to be involved in feeding; (b) studies using functional magnetic resonance imaging (fMRI) with humans and rats that revealed changes in hippocampal activation in response to food stimulation (Thanos et al., 2008
) and to feeding and gastric manipulations (e.g., DelParigi et al., 2004
; Wang et al., 2006
) that were designed to give rise to interoceptive signals of satiety and; (c) studies which showed that humans suffering from severe amnesia associated with hippocampal damage also exhibited a reduced ability to suppress food intake when given repeated opportunities to consume meals (e.g., Hebben, Corkin, Eichenbaum, & Shedlack, 1985
; Rozin, Dow, Moscovitch, & Rajaram, 1998
The goal of the present study was to examine the effects of damage to the hippocampus on the ability of rats to utilize their interoceptive energy state signals to control their appetitive behavior. In this study, rats were trained to solve a deprivation intensity discrimination problem (Davidson, 1987
) in which cues arising from 0-h food deprivation (produced by ad libitum access to food for approximately 24 h prior to a training session) and 24-h food deprivation (no access to food for approximately 24 h prior to the training session) served as discriminative signals for the delivery of sucrose pellets. When asymptotic discrimination performance was achieved, the effects of selective removal of the complete hippocampus on retention of the previously learned discrimination were assessed. If rats rely on the hippocampus to utilize the information provided by their interoceptive energy state signals to anticipate the delivery of sucrose, then retention of deprivation intensity discrimination performance should be impaired for rats with hippocampal lesions compared to controls. The nature or pattern of the impairment exhibited by lesioned rats (e.g., increased or decreased appetitive responding under 0-h compared to 24-h food deprivation) will also be informative about the functional role of the hippocampus in utilizing energy state information.
Previous research found that rats with selective ibotenate lesions were impaired in learning to use interoceptive cues arising from different levels of food deprivation as discriminative cues for a brief shock (Davidson & Jarrard, 1993
). These same rats were unimpaired in learning to use brief auditory conditioned stimuli (CSs) as signals for shock. Thus, the results of this “food deprivation intensity discrimination” training indicated that hippocampal damage interfered with either the detection or the utilization of energy state signals. This basic outcome was also obtained in another study with a shock unconditioned stimulus (US), when rats with lesions confined to the ventral and dorsal hippocampus, respectively, were compared to controls (Hock & Bunsey, 1998
One thing that is unclear based on these findings, is whether or not the hippocampus is needed for rats to use their energy state cues to anticipate appetitive, as well as aversive outcomes. Recent research (Thibaudeau, Dore, & Goulet, 2009) indicates that hippocampal damage impairs Pavlovian trace conditioning with punctate exteroceptive CSs when learning is based on an aversive, but not when it is based on an appetitive US. Thus, it would be important to know whether the effects of hippocampal damage on discrimination performance with interoceptive food deprivation intensity cues, is similarly dependent on the type of US (appetitive or aversive) that is used to reinforce learning. If the hippocampus is needed for rats to use interoceptive stimuli that correspond to “hunger” and “satiety” as signals for appetitive USs, this would support the idea that interfering with hippocampal functioning could also interfere with the regulation of energy intake and body weight as reported by Davidson et al., (2009)
The study by Davidson & Jarrard (1993)
assessed the effects of hippocampal damage that occurred prior to the beginning of food deprivation intensity discrimination training. Energy state signals corresponding to “hunger” and “satiety” are presumably present early in life. Therefore, animals (including humans) would have ample opportunity to learn about these cues before problems with intake and body weight regulation emerge. Thus, if interference with hippocampal functioning contributes to impaired energy and body weight regulation, it is more likely that this effect would be based on interfering with the retention or utilization of previous learning about energy state signals, rather than the acquisition of new learning about those cues. The present experiment also set out to assess the effects of ibotenate lesions of the complete hippocampus on the ability of rats to use prior learning that established interoceptive cues corresponding to “hunger’ and “satiety” as discriminative signals for the availability of an appetitive US.
Following the general procedures and experimental design used previously to establish appetitive food deprivation intensity discrimination learning in non-lesioned animals (Davidson, Kanoski, Tracy et al., 2005
; Kanoski, Walls, & Davidson, 2007
), we trained two groups of rats to use cues arising from 24 h (i.e., 24 h without food) and 0 h food deprivation (i.e., 0 h without food following a 24 h period with food freely available) as discriminative stimuli. For rats in Group 0+, sucrose pellets were delivered at the end of each 4 min session that took place under 0-h food deprivation and no pellets were delivered at the end of sessions that took place when the rats were food deprived for 24 h. Rats in Group 24+ received the reversed deprivation level-sucrose pellet contingency. After asymptotic discrimination performance was achieved, half the rats in each group received hippocampal lesions and half were assigned to control conditions. Following recovery from surgery, deprivation intensity discrimination performance was tested under the same conditions that were used in original training.
Previous studies using this basic design confirmed stimulus control by interoceptive food deprivation intensity cues by showing that neurohormonal manipulations known to promote food intake (e.g., systemic ghrelin administration) generalize to cues produced by a high level of food deprivation (Davidson, Kanoski, Tracy et al., 2005
) whereas treatment with hormones known to suppress intake (e.g., CCK-8) generalize to cues produced by a low level of food deprivation (Kanoski, Walls et al., 2007
In the present study, if hippocampal lesions impair the ability of rats to utilize their deprivation cues as discriminative stimuli, then the difference in responding between hippocampal lesioned rats in Groups 0+ and 24+ should be smaller compared to controls, whether testing occurs under 0-h or under 24-h food deprivation. If removing the hippocampus impairs the ability of deprivation cues to excite appetitive responding, then during testing rats with hippocampal lesions in both Groups 0+ and 24+ should exhibit reduced responding under their rewarded food deprivation level relative to controls. On the other hand, if hippocampal lesions cause rats in these two groups to exhibit increased tendencies to respond under their nonrewarded, but not their rewarded, food deprivation level, this would suggest that the hippocampus is involved with the inhibition of responding to nonreinforced cues. This latter outcome would be consistent with our view that the hippocampus is involved with inhibiting memories of the reinforcing postingestive consequences of eating (e.g., Davidson et al., 2009
; Davidson, Kanoski, Schier, Clegg, & Benoit, 2007
) and other appetitive USs (e.g., Chan, Morell, Jarrard, & Davidson, 2001
; Davidson, Jarrard, & Jarrard, 2004