CONICET | Buscador de Institutos y Recursos Humanos

Running speed has often been thought to modulate the frequency of the theta band LFP oscillations in the hippocampus and this relation has been proposed as the basis of a path integration mechanism. We tested the hypothesis of a correlation between speed and theta frequency under experimental conditions that allowed us to control the movement of rats in a way that dissociated speed and acceleration contributions. Rats were trained to run inside a bottomless cart that was driven along a 4 meter linear track by an external motor. The cart provided back and front physical barriers to the rats, which in the absence of floor had to actively engage in locomotion at the correct speed in order to keep up. The track was divided into 2 or 3 sections. A constant speed was assigned to each, with sudden transitions between them.The analysis of constant running periods showed no significant effect of speed on theta frequency. The analysis of the transition periods showed that theta frequency increases during positive acceleration but remains at baseline for negative or zero acceleration. The firing of individual neurons is consistent with a faster pace in theta modulation during positive acceleration periods, both in hippocampus and entorhinal cortex, indicating that the LFP effect is locally generated in both areas.Open field recordings confirmed a relationship between theta frequency and positive (but not negative) acceleration, both in LFP and modulation of neural firing. When plotting average theta frequency against speed, however, we found a positive slope, as in previous studies. Interestingly, this slope falls to zero when only instantaneous speed variations are considered by applying a high-pass filter of 0.5 Hz to the speed signal. Thus, the reported correlation captures behavioral changes that happen in typical time windows slower than 2 seconds (peaking at 100 seconds), too slow for path integration purposes. We speculate that it rather reflects a correlation between the general level of arousal of the animal and slow changes in the theta frequency baseline. Acceleration has an opposite effect, improving its correlation with theta frequency when only the fast ´instantaneous´ component of the behavior is considered.Finally, we examined an alternative hypothesis. Could running speed be coded in the firing rate of individual neurons? We found two new functional types of neuron. Speed cells are putative interneurons with a firing rate proportional to running speed. Velocity cells in the medial entorhinal cortex are putative pyramidal cells with a conjunctive representation of speed and head direction. These cells could be the basis of an entorhinal path integration mechanism.