Learning and memory require that neural activity be flexibly organized. Neurons must sometimes fire together in ensembles so that important signals are not lost in the complex activity of the brain, but these ensembles must also be capable of forming, breaking apart, or reforming at a moment’s notice. Such transient synchrony of neural activity may be achieved by means of oscillatory firing. Oscillatory firing at various frequencies is commonly found in electrophysiological recordings from the brain and may underlie the dynamic organization of neural activity. We investigated oscillatory activity in a brain region critical to learning and memory, the hippocampus. The power of oscillatory activity in the hippocampus at several frequency bands including gamma (30-80 hz) has been linked to successful encoding and recall in hippocampus-dependent memory tasks. Prior work has also shown that inputs from the ventral tegmental area are necessary for normal hippocampal function and for spatial learning and memory. These inputs occur in two forms: tonic baseline firing and intermittent phasic bursts. We tested genetic knockout mice that lack phasic but not tonic inputs from the ventral tegmental area while animals foraged for food on a spatial navigation task. We simultaneously recorded local field potentials in the hippocampus, which represent the net inputs to a region of the brain and allow us to assess the power of oscillatory activity. In the preliminary data, knockout mice demonstrated more rapid learning in response to successive manipulations of the reward location compared to controls. Knockouts also showed higher gamma-frequency but not theta-frequency (4-10 hz) power. These results suggest that phasic inputs from the ventral tegmental area selectively inhibit gamma-frequency oscillations in the hippocampus. This role may be behaviorally relevant, reflecting the possibility that these oscillations play a role in organizing learning and memory.

Ventral tegmental area (VTA) dopaminergic cell firing encodes reward information and has been implicated in reinforcement learning, spatial memory and decision-making. However the mechanisms which modulate VTA dopaminergic burst firing are not well known. A division of the mesonpontine tegmentum, the laterodorsal tegmental nucleus (LDTg) provides inputs into the VTA. Inactivation of the LDTg greatly reduces VTA dopaminergic burst firing, implicating the LDTg’s role in permitting VTA activity. In the current study we sought to determine the nature of the information coded by LDTg neurons in order to better understand the type of information that regulates dopamine cell activity. We assessed this by neurophysiological single unit recordings of LDTg cell activity in freely behaving rats. Rats were trained on an 8-arm radial maze, which required spatial and working memory in order to, without errors, forage for large and small rewards located at the end of arms. Experimental manipulations consisted of omitting expected rewards, switching the location of rewards of different magnitudes and imposing darkness during the maze trials. We recorded a total of 83 different cells, which were histologically confirmed to be located in the LDTg. We found a large number of reward-related cells (9.6% became excited when rats encountered reward, 8.4% showed excitation in anticipation of reward, 4.8% showed excitatory, then inhibitory and then excitatory responses around the time of reward) and movement related cells (76% were correlated with the rat’s velocity). The large proportion of velocity-correlated cells supports the LDTg’s role in motor behavior. Context and reward manipulations did not consistently affect the firing rate of LDTg cellular activity. However, the current study demonstrates that with 25% of LDTg cells exhibiting reward-related responses, the LDTg has a role in reward encoding by a permissive affect on VTA dopaminergic burst firing.

Decision-making is a complex behavior where one must weigh cost and benefit to maximize the likelihood of the best possible outcome. A common example of this is a decision between a small yet certain reward (e.g. $1 or 1 piece of candy), and a larger uncertain reward (e.g. $10 or 10 pieces of candy 50% of the time). It has been suggested that context and dopamine levels play a critical role in the willingness of decision makers to take risks. We tested this hypothesis by assessing behavioral and neural differences between groups of rats trained on an equivalent risk-based paradigm conducted in different contexts. Rats started training in an operant chamber or a decision maze, and then switched contexts (i.e. tasks). These tasks involved the rat making a decision between two levers or doors, one of which was associated with a certain small reward (1 sugar pellet), and the other which is associated with a large (4 sugar pellets) reward that appeared with decreasing probability (100%, 50%, 25%, 12.5%). To test the influence that dopamine has on risk taking behavior, we used a dopamine antagonist (flupenthixol) to inhibit dopamine in both the maze and operant chamber contexts. Our results suggest that when analyzed as a group, rats are less willing to take risks in the maze context. We hypothesize this is because risk taking on the maze is a more complex behavior, perhaps because of a longer delay between a choice and reward delivery. Our pharmacological findings of high numbers of trials in which levers were not pressed in the chamber, or trials not run on the maze, suggest that blocking dopamine interferes with the ability of rats to make risky decisions. Together, these results support the idea that risk-based decision-making is influenced by many factors, including context and dopamine.

The hippocampus is well known for its role in spatial memory. A better understanding of how the hippocampus functions will provide us with insight into how memories are made and stored. This knowledge may ultimately lead to improved preventative and treatment strategies under conditions in which the hippocampus is altered, such as after traumatic brain injury or in cases of age-related memory impairment. Within the hippocampus there are cells commonly known as “place cells”. These cells are most likely to fire when a navigating organism is in a certain space in an environment, known as that cell’s “place field”. The activity of place cells may be modulated by dopamine input, which can be released tonically or phasically. For example, when overall dopamine input from the ventral tegmental area is disrupted, the place fields of cells within the CA1 subregion of the hippocampus are disrupted and spatial memory deficits occur. To test the idea that phasic dopamine, in particular, is necessary for normal place cell activity and spatial memory performance, we tested spatial working memory while recording electrophysiological activity of neurons in the hippocampus in normal mice and genetically engineered ‘knockout’ mice whose phasic dopamine signaling was selectively disrupted. Preliminary electrophysiological results show differences between the groups, including a nearly significant increase in the spatial selectivity that corresponds to a significant decrease in the size of place fields from the knockout mice when compared to controls. Behaviorally, knockout mice adapt more quickly than control mice following a change in the location of a reward, indicating that phasic dopamine input is not required to perform this task. Rather, removal of the phasic firing pattern may have unmasked a potential facilitating role of tonic dopamine release in flexible goal directed behaviors.