Many functional imaging studies have demonstrated age-related alterations in cerebral blood flow during the resting state. However, few studies have addressed possible differences in functional response to cerebral activation. We assessed the response of visual cortex to photic stimulation in 9 normal elderly subjects and 17 normal younger subjects with blood oxygenation level dependent functional magnetic resonance imaging. We found that the amplitude of response in elderly subjects was significantly decreased compared to younger subjects (2.5 +/- 1.0% versus 4.0 +/- 1.6%, p = 0.01), suggesting a reduction in functional activation or an age-related alteration in the coupling of blood oxygenation to focal activation.

No abstract

Thirsty rodents will persistently lick a stream of dry air pumped through a standard drinking tube. This air-licking is attenuated by experimental manipulations which reduce the evaporative cooling of the tongue and mouth produced by the airstream. This suggests that such cooling is itself an effective reward for thirstry rodents. We tested this hypothesis by presenting thirsty rodents with a piece of cold, dry metal. Different species spent from 9 to 40 percent of their session time licking the cold metal. When deprived of water hamsters reared from birth without access to drinking water licked cold metal in preference to metal maintained at room or body temperature. This preference was approximately equal to that of littermates reared normally. We conclude that tongue cooling is a primary reward for thirsty rodents.

Satiated rats receiving continuous electrical stimulation in hypothalamic "feeding areas" learned to choose the T-maze arm containing food. 2 experiments were then performed to determine in which portions of the maze stimulation-induced hunger is necessary and sufficient for maintaining food-seeking behavior. Ss chose randomly if stimulation was terminated as they entered the goal box. However, if stimulation was not turned on until they entered the goal box, they consistently made food-directed choices. These results are interpreted as showing that, in order for food-seeking behavior to occur in a T maze, hunger is necessary and sufficient only in the goal boxes, and food-seeking behavior can be independent of hunger at the choice point.In order to get rats to engage repeatedly and persistently in food-seeking behavior it is generally found necessary to make them hungry, either by depriving them of food or by administering continuous electrical stimulation to specific parts of the lateral hypothalamus.If one assumes that lateral hypothalamic stimulation induces eating and food-seeking behavior in the same way as food deprivation, it should be possible to design experiments to test two prevalent and contrasting hypotheses concerning the role of hunger drive in the performance of instrumental responses for food. According to one view, hunger selectively facilitates the performance of responses previously successful in obtaining food, and such facilitation is necessary for performance (e.g., Hull, 1952, p. 7). According to the other view, the memory of having eaten in a particular goal box on previous trials provides incentive moti-1 The research reported is based upon a dissertation submitted in partial fulfillment of the requirements for the PhD degree at the Massachusetts Institute of Technology. The author would like to thank S. L. Chorover for his guidance throughout all phases of this research, William Anderson for his assistance with the histology, and A. M. DeLuea for help in preparing the figures.

Clearly the inactive phase cannot be explained by a general depression of the "ascending reticular system" because ECoG desynchronization (occurring either spontaneously or following stimulation of the reticular formation) was observed in cortex 1 cm or more from the site of depression. This inference is consistent with the finding of Bures et al. (18) that in the unanesthetized rat most reticular neurons only increased in rate during SD. (By contrast, the majority of medial thalamic neurons they studied showed only an inactive phase.) Neither synaptic fatigue (for example, transmitter depletion) nor delayed inhibition, caused by the high level of corticofugal discharge, adequately account for the inactive phase, because it was readily observed during cortical cooling without a preceding phase of hyperactivity. The mean rates of discharge attained during SD excitation exceeded those occurring either spontaneously or in response to stimulation of the reticular formation. However, some innervated neurons may be driven at comparable rates for many seconds by peripheral stimulation but return quickly to the control rate of discharge following such stimulation (19).Although reduction of corticothalamic discharge is associated with reduced activity of VP neurons, our experiments do not directly differentiate between reduced direct excitatory drive and reduced inhibition of tonically active inhibitory interneurons (disinhibition). Either mechanism would lead to reduced facilitation of VP neurons. However, because increased inhibition of tonic inhibition probably could not alone account for the very high level of discharge during the excitation phase we conclude that the dominant effect exerted by corticothalamic neurons is synaptic excitation. Therefore, the relatively small decrease in excitability that was observed during the early part of the inactive phase probably resulted from reduced synaptic facilitation. Our finding that depression of sigmoid gyrus and adjacent cortex leads to reduced VP neuron activity is consistent with anatomical evidence that this region of cortex projects to VP (17). The data imply that the response of VP neurons to reticular formation stimulation re-Clearly the inactive phase cannot be explained by a general depression of the "ascending reticular system" because ECoG desynchronization (occurring either spontaneously or following stimulation of the reticular formation) was observed in cortex 1 cm or more from the site of depression. This inference is consistent with the finding of Bures et al. (18) that in the unanesthetized rat most reticular neurons only increased in rate during SD. (By contrast, the majority of medial thalamic neurons they studied showed only an inactive phase.) Neither synaptic fatigue (for example, transmitter depletion) nor delayed inhibition, caused by the high level of corticofugal discharge, adequately account for the inactive phase, because it was readily observed during cortical cooling without a preceding phase of hyperactivity. The mean rates of discharge atta...