Head direction cells recorded in the anterior thalamic nuclei of freely moving rats

Previous studies have identified neurons in the postsubiculum which discharge as a function of the animal's head direction in the horizontal plane, independent of its behavior and location in the environment. Anatomical studies have shown that the postsubiculum contains reciprocal connections with the anterior thalamic nuclei (ATN). In order to determine how the head direction (HD) cell signal is processed in the brain, single-unit recordings were monitored in the ATN of freely moving rats in order to characterize their behavioral and spatial correlates. Animals were trained to retrieve food pellets thrown randomly into a cylindrical apparatus containing a single orientation cue. Single unit recordings in the ATN showed that approximately 60% of the recorded cells discharged in relation to the animal's head direction in the horizontal plane. Observation of the animal and quantitative analyses showed that HD cell firing was not dependent on the animal's behavior, trunk position, linear speed, angular head velocity, or location in the environment. Most of these cells were localized to the anterior dorsal thalamic nucleus. Each HD cell contained only one head direction at which the cell discharged maximally and the firing rate decreased linearly away from this preferred direction. The preferred firing directions from all cells recorded were distributed over a 360 degrees range. Quantitative analysis showed that these cells contained similar discharge parameters (peak firing rate, directional firing range) to values reported previously for postsubicular HD cells (Taube et al., 1990a). Experiments involving rotation of the orientation cue showed that the preferred firing direction could be controlled by a salient visual cue. In contrast to postsubicular HD cells, passive rotation of a restrained animal showed that most ATN HD cells ceased discharging when the animal's head was oriented in the preferred direction. These findings demonstrate the presence of HD cells in the ATN and indicate the potential importance of this area for spatial navigation. The origin of the head direction signal is discussed and it is concluded that because of the presence of reciprocal connections between the postsubiculum and the ATN, further studies are required in order to determine the direction in which this head-directional information is flowing. Finally, ATN HD cells differ from postsubicular HD cells by appearing to require volitional motoric input.