Fig 1.
(A) Schematic diagram of the operant lever press task in the first stage of conditioning (side view). Rats were trained to press one of the levers ipsilateral to the illuminating LED to receive food. The right (R) and left (L) levers were placed in the conditioning box and attached to micro-switches at their bases. (B) Schematic diagram of the robot controlling task in the second and the third stages of conditioning (top view). Restrained rats were trained to press the correct lever ipsilateral to the robot several times until the robot came to the area within their reach. The robot had a dish with food and an illuminating LED.
Fig 2.
Electrode position and the hippocampal local field potential (LFP).
(A) Typical electrode position in the dorsal hippocampus. The arrows indicate the tips of the twisted electrode, which were marked electrically after the completion of behavioral experiment. (B) Typical hippocampal LFP data around the lever press. The LFP data in four trials are shown on a longer and shorter time scale in upper and lower traces, respectively. The times before the first lever press are indicated as negative values. Horizontal lines under each LFP trace indicate the lever presses. The rat took a food from the robot after the final lever off, indicated by arrowheads in the upper traces. (C) Typical power spectral density of the hippocampal LFP during a session, plotted in arbitrary units.
Fig 3.
Increase in the relative and absolute theta power before the first lever press in the correct and incorrect trials.
(A) Dynamic power spectra in the correct (a, c) and incorrect trials (b, d). A time window of 1 s and a stepping width of 0.2 s were used. The frequency resolution was approximately 0.36 Hz. The data were averaged over the trial type in each rat and then over all the rats. The relative power divided by the total power of 1–30 Hz (a, b) and the absolute power subtracted and normalized by that 10 s before the first lever press (c, d) were calculated for each frequency. The times before the first lever press are indicated as negative values. The pseudocolor scales in the right indicate the relative power and the normalized absolute power. (B) Increase in the relative (a) and normalized absolute (b) power of theta frequency band. The power data of each frequency in A were integrated over the theta frequency band of 6–9 Hz. The black and dotted lines indicate the data averaged over the correct and incorrect trials in each rat, respectively, and then over the rats. The shaded areas associated with the lines are standard error of the mean.
Fig 4.
Increase in the relative and absolute theta power before the first lever press in the right- and left-forelimb trials.
(A) Dynamic power spectra before the first lever press with the right (a, c) or the left (b, d) forelimb in correct trials. The parameters used for the analysis were the same as in Fig 3. The relative power (a, b) and normalized absolute power (c, d) of each frequency were calculated as in Fig 3. (B) Increase in the relative (a) and normalized absolute (b) power of theta frequency band (6–9 Hz). The black and dotted lines indicate the data averaged over the right- and left-forelimb trials, respectively, and then over the rats. The shaded areas associated with the lines are standard error of the mean.
Fig 5.
Decrease in the relative and absolute theta power after the termination of multiple lever pressing in the right- and left-forelimb trials.
(A) Dynamic power spectra after the final lever off, with the right (a, c) or left (b, d) forelimb in correct trials. The parameters used for the analysis were the same as in Fig 3. The relative power divided by the total power of 1–30 Hz (a, b) and the absolute power subtracted and normalized by that at the time of lever off (c, d), were calculated for each frequency. The times before the final lever off are indicated as negative values. (B) Decrease in the relative (a) and normalized absolute (b) power of the theta frequency band (6–9 Hz). The black and dotted lines indicate the data averaged over the right- and left-forelimb trials in each rat, respectively, and then over the rats. The shaded areas associated with the lines are the standard error of the mean.
Fig 6.
Differences between the correct and incorrect trials during the first and the second lever presses.
(A) Duration of the time between the first lever-on and first lever-off (1st press), between the first lever-off and second lever-on (1st release), and between the second lever-on and second lever-off (2nd press). The data were averaged over the trial type in each rat and then over all the rats. (B) Absolute power spectral density normalized by the total power of the first press (see the Methods for details) in the correct (a) and incorrect trials (b). The spectrum was averaged over the trial type in each rat, and then over all the rats. (C) Normalized power of theta frequency band (6–9 Hz) in the correct (a) and incorrect trials (b). The error-bars indicate the standard error of the mean.
Fig 7.
Change in the theta frequency before the first lever press and after the last lever release.
(A) Increase in the dominant frequency of theta oscillation before the first lever press in the correct and incorrect trials. The filled and open circles indicate the data averaged over the correct and incorrect trials in each rat, respectively, and then over all the rats. (B) Increase in the dominant frequency of theta oscillation before the first lever press (a) and decrease after the final lever off (b) in the right and left forelimb trials. The filled and open squares indicate the data averaged over the right and left trials in each rat, respectively, and then over all the rats. The error-bars indicate the standard error of the mean.