Tsuneyuki Kobayashi

Contribution of hippocampal place cell activity to learning and formation of goal-directed navigation in rats

Although extensive behavioral studies have demonstrated that hippocampal lesions impair navigation toward specific places, the role of hippocampal neuronal activity in the development of efficient navigation during place learning remains unknown. The aim of the present study was to investigate how hippocampal neuronal activity changes as rats learn to navigate efficiently to acquire rewards in an open field. Rats were pre-trained in a random reward task where intracranial self-stimulation rewards were provided at random locations. Then, the rats were trained in a novel place task where they were rewarded at two specific locations as they repeatedly shuttled between them. Hippocampal neuronal activity was recorded during the course of learning of the place task. The rats learned reward sites within several sessions, and gradually developed efficient navigation strategies throughout the learning sessions. Some hippocampal neurons gradually changed spatial firing as the learning proceeded, and discharged robustly near the reward sites when efficient navigation was established. Over the learning sessions, the neuronal activity was highly correlated to formation of efficient shuttling trajectories between the reward sites. At the end of the experiment, spatial firing patterns of the hippocampal neurons were re-examined in the random reward task. The specific spatial firing patterns of the neurons were preserved if the rats navigated, as if they expected to find rewards at the previously valid locations. However, those specific spatial firing patterns were not observed in rats pursuing random trajectories. These results suggest that hippocampal neurons have a crucial role in formation of an efficient navigation.

Altered accumbens neural response to prediction of reward associated with place in dopamine D2 receptor knockout mice

Midbrain dopaminergic activity seems to be important in forming the prediction of future events such as rewards. The nucleus accumbens (NAc) plays an important role in the integration of reward with motor function, and it receives dense dopamine innervation and extensive limbic and cortical afferents. Here, we examined the specific role of the dopamine D2 receptor (D2R) in mediating associative learning, locomotor activity, and regulating NAc neural responses by using D2R-knockout (KO) mice and their wild-type littermates. D2R-KO mice displayed reduced locomotor activity and slower acquisition of a place-learning task. D2R-KO eliminated the prereward inhibitory response of neurons in the NAc. In contrast, an increased number of neurons in D2R-KO mice displayed place-related activity. These results provide evidence that D2R in the NAc participates in coding for a specific type of neural response to incentive contingencies and partly in spatial learning.

Effects of facial expression on shared attention mechanisms.

We investigated the effects of facial expression on shared attention mechanisms. A female or male facial stimulus with one of 3 facial expressions (happiness, neutral, or anger) was presented at the center of a display. This facial stimulus gazed toward a subject, or toward the left or right side of the display. After the facial stimulus was offset, a target appeared on the left or right side of the display and the reaction time to the target was measured. In the statistical analysis by ANOVA, there was a significant main effect of congruity between the target position and the gaze direction in both the female and male facial cues, indicating that gaze direction significantly affected reaction time. When the female facial cues were presented, the reaction times for the congruent target position to the gaze direction were significantly shorter in the happy than other facial expressions. However, there were no significant differences in reaction time when the facial stimuli were presented in an inverted orientation. The results demonstrated that facial expression significantly affected shared attention mechanisms.

HIPPOCAMPAL PLACE CELL ACTIVITY DURING CHASING OF A MOVING OBJECT ASSOCIATED WITH REWARD IN RATS

Hippocampal place cells encode location of animals in the environment. However, it remains unknown whether the hippocampal place cells encode a continuously moving object in the environment. To investigate this topic, we analyzed the place cell activity of freely moving rats when a toy car was introduced into an arena. First, in a freely moving task without the car, the rats freely navigated inside the arena to earn an intracranial stimulation (ICS) reward for each 150 cm traveled. Second, they were divided into two groups and tested using two different tasks. In the car-dependent navigation (CDN) task, the car was placed inside the arena, and the rat received ICS if it chased and came within 20 cm of the car. In the car-independent navigation (CIN) task, the rat acquired ICS rewards if it traveled 150 cm regardless of its relation to the car. Place fields remapped more frequently in the CDN than the CIN tasks. In both the CDN and CIN tasks, the place cell activity inside the place fields displayed moderate tuning to the movement parameters of the rats and car, and the distance between the car and rats. However, tuning of the place cells to movement variables of the car was more selective in the CDN than the CIN tasks, while information regarding movement variables of the car represented by the place cell activity was larger in the CDN than the CIN task. These results indicated that place cell activity within the place fields represents not only an animals own location but also the movement variables of another moving object if that object is associated with rewards. The present results provide new evidence that place cell activity conveys relevant information in a task even if this information is derived from other moving objects.

Brain learning control representation in nucleus accumbens

Brain information processing is supported by the dual architecture of the cortical and limbic systems for the knowledge-based and emotional information, respectively. We hypothesize this dual architecture of the brain contributes to brain learning control. In order to examine the role of emotion in forming memory that is, automatic algorithm acquisition, solidification and retrieval, single unit recording was executed to nucleus accumbens of the rat. The rat was trained in a circular open field to develop its learning ability for food and water reward. After this reward acquisition task was trained, electrical activities were recorded in nucleus accumbens neurons of the in vivo brain while the rat continued the originally-trained reward acquisition task or executed some other combinations of food and water reward task. In nucleus accumbens, some neurons were found to respond to anticipation of reward. Some other neurons changed their activities while the rat continued to perform its training. These results suggest that activities of nucleus accumbens are learning-controlled by the reward value evaluated possibly by both amygdala and ventral tegmental area.

Central action of endogenous sugar acid (2-buten-4-olide): comparison with local anesthesia in hypothalamus.

Rats were trained to discriminate cue tone stimuli (CTS) predicting reward (CTS+) [juice or intracranial self-stimulation (ICSS)], or aversion (CTS-) (mild electric shock or tail pinch). Unit activity in the lateal hypothalamus (LHA) and lateral preoptic-anterior hypothalamic area (lPOA-AHA) of the rat was recorded during CTS learning. The effects of local anesthesia of the amygdala (AM), ventral tegmental area (VTA) or LHA by procaine hydrochloride, and the effects of intraperitoneal or intravenous 2-buten-4-olide (2-B4O) on LHA neural activity and licking behavior were compared. LHA neurons differentiated between rewarding and aversive stimuli, and acquired corresponding discrimination of CTS+ and CTS-. In the lPOA-AHA, neurons responded similarly to CTS+, rewarding stimuli, CTS- and aversive stimuli. Procainization of the AM suppressed LHA neural responses to CTS1+ predicting juice, and stopped licking for juice. Procainization of the VTA suppressed LHA neural responses to CTS2+ predicting ICSS, and stopped licking for ICSS. LHA procainization suppressed both licking for juice and ICSS. Both intraperitoneal and intravenous 2-B4O stopped licking for juice and ICSS, but did not influence LHA responses to CTS1+ or CTS2+. The results suggest that dynamic interaction of AM-LHA-VTA are important for CTS+ learning, and 2-B4O acts directly on LHA neurons while maintaining afferent sensory inputs to the LHA.

Study of brain activation related to memory organization by near-infrared spectroscopy

The visual acuity of rats is poorer than their auditory acuity. This study examined whether visual or auditory memory is superior in rats. Either a light stimulus (L) or a tone stimulus (T) was presented for 3 s as a discriminative stimulus (SD) in an operant chamber. Then, there was a delay of 0.25, 3, 6, or 9 s. After the delay, the rat had to press one of two levers depending on the SD: when L was presented, pressing the right lever gave a food pellet reward; when T was presented, pressing the left lever gave a similar reward. In order to perform this task correctly, the rats had to remember the SD during the delay interval. Therefore, performance of this task reflects their short-term memory (STM). When L was presented as the SD, the percent of correct responses declined as the delay interval lengthened. When T was presented as the SD, the decline in the percent of correct responses was slower than in the trials in which L was presented. These results imply that in rats STM is better for auditory information than for visual information.

Dopamine D2 receptor-knockout changed accumbens neural response to prediction of reward associated with place in mice

Abstract The amygdala (AM) and hippocampal formation (HF) play important roles in stimulus-affect association and episodic memory, respectively. The prefrontal cortex (PFC) encodes the prediction of reward or aversion. The nucleus accumbens (NAc) receives converging inputs from these areas, and also mesolimbic dopamine (DA) input originated from the ventral tegmental area (VTA) to predict reward or aversion; this is an essential determinant for learning approach behaviors. However, the involvement of dopamine receptor subtypes in assessing reward information and in learning still remains to be specified. In the present study, we examined the specific role of dopamine D2 receptor (D2R) subtype in mediating associative learning, in regulating NAc neural responses and locomotor activity, using new genetic tool: D2R-knockout mice. The specific contribution of the D2R to reward process and associative learning at neural and behavioral levels is discussed in this article.

Hippocampal and Parahippocampal Neuronal Responses to Spatial and Non-Spatial Factors in Rats and Monkeys

Rat hippocampal (HF) neurons were recorded while the rat ran on a treadmill affixed to a motion stage that was translocated along a figure 8-shaped track. Comparison of HF spatial firing patterns across different experimental conditions indicated that place neuron activity encodes intra-maze multiple information including location of animals, locomotion, the reinforcement episodes, and vestibular sensation or optic flow. Place neurons were also recorded from the monkey HF during virtual navigation. Most place-differential responses disappeared or changed their spatial tuning (i.e., remapping) when the arrangements of the distal cues were altered/moved in the virtual spaces. The results suggest that the HF encodes multifold information within the maze, which are gated by the extra-maze distal cues.

Role of the hippocampal formation in sequence memory

It is suggested that episodic memory is defined as a context-dependent sequence memory, and that the hippocampal formation (HF) is essential in such memory. To investigate HF involvement in a context-dependent sequence memory, multiple single unit activities were recorded from the monkey HF during performance of real translocation and virtual navigation tasks. The results indicated that place-related neuronal activity in the HF was task-or context-dependent, and cross-correlation data suggest that the context-dependent information may be encoded by interaction among pyramidal neurons based on asymmetrical connections. Rat CA1 HF neurons were also recorded during a conditional sequence memory task. Consistent with the computational studies, 2 types of the HF neurons were found; some neurons responded to single item regardless of sequences in which the item was presented, while other neurons displayed sustained firing during serial presentation of several items. In humans, event-related potentials (ERPs) were recorded during a sound-sequence memory task. The results suggest that the ERPs around 300-700 msec latency were specifically involved in sound sequence information processing. Furthermore, equivalent dipoles for the ERPs were localized in the medial temporal lobe including the HF and parahippocampal gyrus. These results suggest that the HF is crucial in context-dependent sequence information processing, which may be the neural basis of episodic memory.