Effects of forced movements on learning: Findings from a choice reaction time task in rats

Abstract

To investigate how motor sensation facilitates learning, we used a sensory-motor association task to determine whether the sensation induced by forced movements contributes to performance improvements in rats. The rats were trained to respond to a tactile stimulus (an air puff) by releasing a lever pressed by the stimulated (compatible condition) or nonstimulated (incompatible condition) forepaw. When error rates fell below 15%, the compatibility condition was changed (reversal learning). An error trial was followed by a lever activation trial in which a lever on the correct or the incorrect response side was automatically elevated at a preset time of 120, 220, 320, or 420 ms after tactile stimulation. This lever activation induced forepaw movement similar to that in a voluntary lever release response, and also induced body movement that occasionally caused elevation of the other forepaw. The effects of lever activation may have produced a sensation similar to that of voluntary lever release by the forepaw on the nonactivated lever. We found that the performance improvement rate was increased by the lever activation procedure on the incorrect response side (i.e., with the nonactivated lever on the correct response side). Furthermore, the performance improvement rate changed depending on the timing of lever activation: Facilitative effects were largest with lever activation on the incorrect response side at 320 ms after tactile stimulation, whereas hindering effects were largest for lever activation on the correct response side at 220 ms after tactile stimulation. These findings suggest that forced movements, which provide tactile and proprioceptive stimulation, affect sensory-motor associative learning in a time-dependent manner.

Keywords: Actuator; Behavior; Proprioceptive sensation; Robotics; Voluntary movement.

Fig. 1

Apparatus. (A) The front panel of the operant box. (B) The rat’s posture in front of the panel during task execution

Fig. 2

Lever activation device. (A) Photograph of the device with two solenoid actuators placed behind the front panel. (B) Side view of the device and a rat’s posture during the foreperiod, in which both the levers are depressed by the rat. The lever activation mechanism consists of a solenoid actuator, a footstall, a microswitch, and an optical sensor. The footstall moves the lever arm behind the front panel downward by activating the solenoid actuator. The microswitch returns the lever back to the “up” (OFF) position. The lever movements are detected by the optical sensor. (C) During a planned trial, the solenoids and footstalls are not activated, and the rat voluntarily releases one of the levers. Note the gap between the footstall and the lever. (D) During an additional trial, the solenoid actuator moves the footstall downward, and thereby elevates the lever and the rat’s arm

Fig. 3

Flow chart. The task consisted of planned and additional trials. An additional trial occurred after every planned trial in which the rat made an error

Fig. 4

Serial reversal learning. Changes are shown in the number of planned trials (A) and error rates (ERs) (B) over the four training periods. The compatibility condition was reversed between the training periods (three times in total). The dashed lines are boundaries between the training periods. The dotted line in panel A indicates the number of 100 planned trials required for reaching the learning criterion. The dotted line in panel B indicates the ER of 15%, which should be crossed as a rat reaches the learning criterion

Fig. 5

Training days required for task learning. The number of training days required for reaching the learning criterion is indicated. The data were collected from all rats in the third and fourth training periods

Fig. 6

Distribution of daily error rates (ERs) and median reaction times (RTs). The distribution of the data five days after reversal in the third and fourth training periods indicate that the relationship between ERs and median RTs differed from day to day. The open star and the double open star indicate the means of the ERs and RTs on Days 0 and 1, respectively

Fig. 7

Correlations between the number of training days required for reaching the learning criterion and the ERs on Days 4 and 5 after reversal. The data were collected from all rats in the third and fourth training periods. The graph indicates a relationship between ER and the number of training days required for reaching the learning criterion. The solid line is a fitted logarithmic curve. The approximate formula is y = ln(x) – 1.14 (y, ER; ln, natural-logarithm function; x, number of days for reaching the criterion). The coefficient of determination (R ²) was .501. The horizontal axis is logarithmic

Fig. 8

Effects of lever activation condition on ERs and median RTs. Data collected on Days 4 and 5 after reversal were analyzed. The mean values of the ERs and median RTs for the different lever activation groups are shown in panels A and B, respectively. Their distribution in the x–y plane (x, ER; y, median RT) is shown in panel C. Post-hoc multiple comparisons were conducted by using Hotelling’s T ² test with control of the false discovery rate using the Benjamini–Hochberg correction. The p values for multiple comparisons between groups in the rows and columns are indicated in panel D. Incorrect-side lever activation groups tended to have lower ERs and shorter median RTs than the other lever activation groups. These tendencies were more apparent for groups in which lever activations were applied to rats that made more responses—that is, the C220 and I320 groups. Error bars denote SEMs, and asterisks indicate significance: ^* p < .050, ^** p < .010, ^*** p < .001

Fig. 9

Histograms of lever releases. (A) Lever releases of the non-responded-side forepaw after voluntary correct responses on Day 0. (B) Lever releases of the forepaw opposite the lever activation side in additional trials with lever activations (i.e., in lever activation trials) on Day 1. Rats ordinarily released a lever as a voluntary response. However, in lever activation trials, not only the lever-activation-side forepaw, but also the other forepaw was often elevated. This effect is demonstrated by a peak approximately 25 ms after the lever activation in panel B—that is, rats released the nonactivated lever approximately 25 ms after lever activation. The bin width is 10 ms

Fig. 10

Hypothetical schematic diagrams. In the trials shown in diagrams A–C, the correct response side is assumed to be on the right. (A) Voluntary correct response of the right forepaw. (B) Lever activation on the correct response side (right). (C) Lever activation on the incorrect response side (left). The open and closed circles on forepaws indicate the reduction and increase in lever-press force, respectively. Spinal α motor neurons innervating the brachial biceps and triceps are presumed to be active in voluntary movements and to be activated by the stretch reflex caused by lever activation. The reduction in lever-press force of the forepaw opposite the lever activation side (left in B, and right in C) is considered to be caused by the rat’s body being swung back by lever activation. Together, the three diagrams indicate that lever activation on the incorrect response side (C), but not on the correct response side (B), causes changes in forelimb neuromuscular activities and cutaneous afferent activities similar to those associated with voluntary correct responses (A)