Motor imagery involves predicting the sensory consequences of the imagined movement

Abstract

Research on motor imagery has identified many similarities between imagined and executed actions at the behavioral, physiological and neural levels, thus supporting their "functional equivalence". In contrast, little is known about their possible "computational equivalence"-specifically, whether the brain's internal forward models predict the sensory consequences of imagined movements as they do for overt movements. Here, we address this question by assessing whether imagined self-generated touch produces an attenuation of real tactile sensations. Previous studies have shown that self-touch feels less intense compared with touch of external origin because the forward models predict the tactile feedback based on a copy of the motor command. Our results demonstrate that imagined self-touch is attenuated just as real self-touch is and that the imagery-induced attenuation follows the same spatiotemporal principles as does the attenuation elicited by overt movements. We conclude that motor imagery recruits the forward models to predict the sensory consequences of imagined movements.

Fig. 1

Conditions of Experiment 1 and Results. a–c Participants received a reference force on their relaxed left index finger by a probe attached to a lever controlled by a DC motor. During the application of this reference force (3 s), the participants were instructed to (i) keep their right hand and right index finger relaxed on top of a support (base; a); (ii) press a sensor with the right index finger (press; b); or (iii) imagine pressing the sensor with their right index finger (imagine; c). When pressing or imagining pressing, they were instructed to use as much force as they felt was required to match the reference force that they simultaneously felt on the left index finger. Immediately afterwards, they were asked to reproduce the reference force by using a slider that controlled the force output on their left index finger. d Forces generated by the participants (matched forces) as a function of the reference force. Points represent the matched forces of the participants by condition, averaged across the repetitions of each reference force level. Errors represent the SEM ( ± SE). The dotted line indicates the theoretically perfect performance. Colored lines represent the fitted regression lines per condition. For illustration purposes, the position of the markers has been horizontally adjusted to avoid overlapping points (for detailed statistical results see main text). e Matched forces per condition, averaged across the reference force levels. Error bars represent the SEM ( ± SE). Participants produced significantly weaker forces (p < 0.05) when matching the previously felt reference force on the left index finger during the press or imagine conditions compared to the baseline (base). f Somatosensory attenuation displayed per condition; here, the matched forces are subtracted from the reference forces (mean ± SE). g Root-mean-square EMG activity of the right FDI muscle during the application of the reference forces per condition, averaged across all trials. Errors represent the SEM ( ± SE). No significant difference was detected between the base and imagine conditions, demonstrating that the FDI muscle remained relaxed during the motor imagery

Fig. 2

Conditions of Experiment 2 and Results. a–c The conditions of Experiment 2 were identical to those of Experiment 1, with the only difference being that the participants’ right hands, the sensor, and the slider were all placed 25 cm to the right of their left index fingers. As it has been shown that introducing such a distance between the hands eliminates somatosensory attenuation, we tested whether the same holds true for motor imagery. d Forces generated by participants (matched forces) as a function of the reference force (mean ± SE). There was no significant effect of condition on the participants’ performances, i.e., we found no evidence for somatosensory attenuation (see main text for detailed results). e Mean matched forces displayed per condition (mean ± SE). There were no significant differences among the three conditions. f Somatosensory attenuation expressed as the difference between the reference forces and the matched forces per condition (mean ± SE). g Root-mean-square EMG activity of the right FDI muscle averaged across all applications of the reference force per condition (mean ± SE). Importantly, no significant difference was detected between the base_far and imagine_far conditions, showing that the participants were able to relax their hand in the motor imagery condition

Fig. 3

Comparing attenuation and task performance across the imagery conditions of the two experiments. a Baseline-corrected attenuation for the imagery conditions used in the two experiments (mean ± SE). There was significantly stronger attenuation (p < 0.05) during imagery of pressing the sensor when the sensor was placed on top of the left index finger compared with when it was placed at a distance of 25 cm from left index. b, c Boxplots for the ratings of difficulty and vividness of the motor imagery. After each experiment, we asked the participants to rate, on a 7-point Likert scale, how difficult they experienced the task and how vivid their mental imagery was. The horizontal black bars represent the medians. The boxes represent the interquartile ranges. The whiskers extend to the lowest and highest value within the 1.5 × interquartile range. There were no significant differences in imagery difficulty or imagery vividness between the groups, thus supporting similar performance

Fig. 4

Computational equivalence between imagined and executed movements. a Somatosensory attenuation associated with pressing one index finger against the other. Given a copy of the motor command, the forward models predict the next state of the body and the associated sensory consequences of this state. When pressing one finger against the other, these sensory consequences include the tactile feedback from the self-touch. The predicted touch is used to attenuate the actual tactile feedback (comparator). This model is based on an earlier proposal. b Somatosensory attenuation during imagining pressing one index finger against the other. For the covert action, the forward models predict tactile feedback based on the efference copy that is generated as part of the internal simulation of the action. When touch is applied externally on the finger in a way that matches the predicted feedback from the imagined movement, the somatosensory sensation is attenuated, just as occurs during real self-touch, because the forward models have already predicted it