Action does not enhance but attenuates predicted touch
- Department of Neuroscience, Karolinska Institute, Sweden
- Donders Institute for Brain, Cognition and Behaviour, Radboud University, Netherlands
Figures
Figure 1 with 1 supplement
Experimental methods, hypotheses, and results of Experiment 1 (n=30).
(A) In the contact condition (magenta), participants tapped with their right index finger ‘R’ on a force sensor placed above the probe that delivered a test force of 2N to their left index finger ‘L’, followed by a comparison force randomly varying between 1 and 3N. In the no-contact condition (green), participants approached a distance sensor with their right index finger, which triggered the test force on their left index finger, thus receiving no touch on their active finger. In the baseline condition (blue), participants relaxed both hands and passively received the forces on their left index finger. (B) Hypotheses based on the attenuation model. Touch in the contact condition (magenta) should be perceived as weaker than in the baseline (blue), but touch in the no-contact condition (green) should be perceived similarly to that in the baseline (blue). (C) Hypotheses based on the enhancement model. Touch in the no-contact condition (green) should be perceived as stronger than in the baseline (i.e. enhanced) (blue), but touch in the contact condition (magenta) should be perceived as weaker than the baseline (blue). Note that attenuation effects in the contact condition are predicted both by the attenuation and the enhancement model with the difference that the attenuation model attributes these effects to action prediction, while the enhancement model attributes these effects to the simultaneous touch on the active hand. (D–H) Data are color-coded per condition. (D) Box plots show the median and interquartile ranges for the point of subjective equality (PSE) values per condition, black circles and error bars show the mean PSE ± SEM, and the raincloud plots show the individual PSE values and their distributions. No enhancement effects were observed in the no-contact condition. (E) Group psychometric functions for each condition. The leftward shift of the curve in the contact condition indicates attenuated PSE values compared to the other two conditions. (F–H). Line plots for individual participant PSE values illustrate significantly lower PSEs for the contact versus baseline (F) and no-contact (G) conditions, but no significant differences in the PSE values between no-contact and baseline (H). (***p<.001, n.s. non-significant).
Figure 1—figure supplement 1
Fitted logistic models based on the participants’ responses under each condition of Experiment 1.
Figure 2 with 1 supplement
Experimental methods, hypotheses, and results of Experiment 2 (n=30).
(A) The contact (magenta) and no-contact (green) trials were identical to those of Experiment 1, with the only difference that in 50% of the trials, the participants had to inhibit their movement (NOGO trials - yellow), and the test force was delivered automatically. The baseline condition (blue) was identical to that of Experiment 1. (B) Hypotheses based on the attenuation model. Touch in the contact condition (magenta) should be perceived as weaker than in the baseline (blue), but touch in the no-contact condition (green) should be perceived similarly to that in the baseline (blue). Critically, touch may also be perceived as weaker than baseline in the NOGO trials, resulting in a ‘false enhancement’ of the no-contact trials. (C) Hypotheses based on the enhancement model. Touch in the no-contact condition (green) should be perceived as stronger than in the baseline (i.e. enhanced) (blue), but touch in the contact condition (magenta) should be perceived as weaker than the baseline (blue). (D–H) Data are color-coded per condition. (D–E) Box plots show the median and interquartile ranges for the point of subjective equality (PSE) values in the baseline and no-contact blocks (D) and in the baseline and contact blocks (E). Black circles and error bars show the mean PSE ± SEM, and the raincloud plots show the individual PSE values and their distributions. (F–H) Line plots for individual participant PSE values illustrating no significant differences in the PSE values between no-contact and baseline (F), significantly higher PSEs for the no-contact versus NOGO trials in the same block (G) and significantly lower PSEs in the contact versus baseline trials (H). (**p<0.01, ***p<0.001, n.s. non-significant).
Figure 2—figure supplement 1
Fitted logistic models based on the participants’ responses under each condition of Experiment 2.
Figure 3 with 3 supplements
Experimental methods, hypotheses, and results of Experiment 3 (n=30).
(A) The contact (magenta) and no-contact (green) conditions were identical to those of Experiment 1, with the only difference being their relative proportion (contact trials 80%, no-contact trials 20%). In the no-contact trials, the force sensor was automatically retracted, unbeknownst to the participant, revealing the distance sensor placed below. The baseline condition (blue) was identical to that of Experiments 1 and 2. (B) Hypotheses based on the attenuation model. If attenuation is due to action prediction, then the perceived magnitude of touch should be reduced in both the contact (magenta) and no-contact conditions (green) compared to the baseline (blue). (C) Hypotheses based on the enhancement model. If attenuation effects are driven by simultaneous touch on the active hand, then the perceived magnitude of touch should be reduced only in the contact condition (magenta) compared to the baseline (blue) and not in the no-contact condition (green) which should be similar to the baseline (blue). (D–H) Data are color-coded per condition. (D) Box plots show the median and interquartile ranges for the point of subjective equality (PSE) values per condition, black circles and error bars show the mean PSE± SEM, and the raincloud plots show the individual PSE values and their distributions. (E) Group psychometric functions for each condition. The leftward shift of the curves in the contact and no-contact conditions indicates attenuated PSE values compared to the baseline. (F) Line plots for individual participant PSE values illustrate significantly lower PSEs for the contact versus baseline condition, (G) significantly lower PSEs in the contact versus no-contact condition and (H) significantly lower PSEs in the no-contact versus baseline. (***p<0.001).
Figure 3—figure supplement 1
Fitted logistic models based on the participants’ responses under each condition of Experiment 3.
Figure 3—figure supplement 2
Distance between the fingers (contact condition minus no-contact condition) for each experiment at the time of the test force for each experiment (n=30).
Box plots show the median and interquartile ranges, raincloud plots show individual participant values and their distribution. Values below zero indicate that the fingers were closer together for the contact condition compared to the no-contact condition when the test force was delivered. The fingers were slightly closer in the contact condition compared to the no-contact conditions of Experiment 1 and 2 but were closer in the no-contact condition compared to the contact condition of Experiment 3 since the sensor was removed unbeknownst to the participants.
Figure 3—figure supplement 3
Forces exerted on the left index finger (Test force) and on the force sensor by the right index finger (Peak active force) (n=30).
(A–C) Average test forces (mean ± SEM) exerted on the participants’ left index finger for the contact (magenta traces), no-contact (green traces), and baseline (blue traces) across all experimental trials. The average active force exerted on the force sensor by the participants’ right index finger (gray traces) is overlayed for the contact conditions only. Error bands represent ± SEM. Note that the error bars for the test forces are small since these forces were delivered by the electric motor. (D–F) Average active forces exerted on the force sensor in the contact condition across all experimental trials (mean ± SEM). As it can be seen, participants applied forces of similar magnitude on the force sensor with their right index finger, thereby establishing an approximately fixed gain between the applied force and the received test force across all experimental trials.
Additional files
Download links
A two-part list of links to download the article, or parts of the article, in various formats.
Downloads (link to download the article as PDF)
Open citations (links to open the citations from this article in various online reference manager services)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
Action does not enhance but attenuates predicted touch
eLife 12:e90912.
https://doi.org/10.7554/eLife.90912
