Rightward prism adaptation ameliorates neglect symptoms while leftward prism adaptation (LPA) induces neglect-like biases in healthy individuals. Similarly, inhibitory repetitive transcranial magnetic stimulation (rTMS) on the right posterior parietal cortex (PPC) induces neglect-like behavior, whereas on the left PPC it ameliorates neglect symptoms and normalizes hyperexcitability of left hemisphere parietal-motor (PPC-M1) connectivity. Based on this analogy we hypothesized that LPA increases PPC-M1 excitability in the left hemisphere and decreases it in the right one. In an attempt to shed some light on the mechanisms underlying LPA’s effects on cognition, we investigated this hypothesis in healthy individuals measuring PPC-M1 excitability with dual-site paired-pulse TMS (ppTMS). We found a left hemisphere increase and a right hemisphere decrease in the amplitude of motor evoked potentials elicited by paired as well as single pulses on M1. While this could indicate that LPA biases interhemispheric connectivity, it contradicts previous evidence that M1-only MEPs are unchanged after LPA. A control experiment showed that input-output curves were not affected by LPA
Unilateral spatial neglect frequently occurs after right hemisphere damage and is an invalidating multicomponent syndrome in which perception of the contralesional side of space is compromised [
Inhibitory rTMS applied to the RH of healthy individuals can induce neglect-like biases, such as a rightward shift in line bisection judgments [
One possible explanation for the mechanism of action of LPA is that it parallels that of a right hemisphere stroke: LPA decreases excitability in the right PPC and alters left PPC excitability through changes in interhemispheric inhibition [
To investigate parietal-motor changes in both the left and right hemispheres we used a dual-site paired-pulse protocol used elsewhere [
Twenty-eight healthy volunteers (14 males, mean age = 25.14 years, SEM = 0.72) participated in Experiment 1. Ten healthy volunteers (8 females, mean age of 28.4 years, SEM = 2.31) participated in Experiment 2. All participants had normal or corrected-to-normal vision, were right-handed according to the Edinburgh Handedness Inventory [
All participants were adapted to prisms that shifted the visual field 15 degrees to the left (LPA) and PPC-M1 connections were measured before and after LPA in either the left (
As a measure of the efficacy of LPA in inducing rightward visuospatial bias in bisection we used a Landmark task [
This between-subjects design experiment consisted of a perceptual line bisection task (Landmark), paired-pulse TMS (ppTMS), and open-loop pointing measures both before and after LPA. The order of administration of the Landmark task and ppTMS was counterbalanced across participants, while the timing of the open-loop measurements was kept constant. Throughout the experiment participants were comfortably seated on an armchair with their head positioned on a neck-rest during the ppTMS and on a chinrest during the LPA, Landmark task, and open-loop pointing tasks. In order to reduce the possibility of deadaptation, after LPA, participants were instructed to keep their eyes closed and avoid moving their right hand during the short intervals between tasks.
Stimuli were white lines (350 mm × ~2 mm) displayed on a black screen positioned 35 cm from the participant’s eyes. Lines were transected at the true center and 2, 4, 6, 8, and 10 mm toward either the left or right side of the true center. Each of the 11 different prebisected lines was presented six times in a random order, yielding a total of 66 trials, which took approximately three minutes to complete. Each prebisected line was displayed for a maximum of five seconds or until a response was made and was then replaced by a black-and-white patterned mask, which stayed on the screen for one second before the next prebisected line was displayed. Presentation software (Neurobehavioral Systems, Inc., USA) was used to generate the stimuli, record responses, and control the timing of stimulus presentation throughout the task. For each participant the percentage of “right” responses was plotted as a function of the position of the transector. These data were then fitted with a sigmoid function and the value on
Participants rested their arm on a pillow placed on their lap or on the armrest of the chair and leaned their head on a neck-rest. After identifying the hotspot, the scalp location where stimulation evoked the largest MEP from the contralateral FDI muscle, we determined the resting motor threshold (RMT), defined as the lowest stimulation intensity that evoked at least five out of ten MEPs of at least 50
The interstimulus intervals (ISIs) between the CS and TS were 2, 4, 6, and 8 ms [
As in Experiment 1, throughout the experiment, participants were comfortably seated in an armchair with their head positioned on a neck-rest during the TMS and on a chinrest during the LPA and open-loop pointing tasks. The experimental protocol, this time a within-subjects design, consisted of measuring input/output (I/O) curves for each hemisphere and open-loop pointing measures both before and after LPA. The order of I/O curve measurements was counterbalanced across participants (i.e., right or left hemisphere first). The open-loop pointing task and LPA procedures were identical to Experiment 1.
Participants rested their arm on a pillow placed on their lap or on the armrest of the chair and leaned their head on a neck-rest. As in Experiment 1, after identifying the FDI hotspot, we determined the resting motor threshold (RMT). Both hotspot and RMT were measured for each hemisphere with the order counterbalanced across participants (i.e., either right or left hemisphere first).
I/O curves were measured by stimulating the left or right M1 (hotspot) twelve times at each of six different intensities, ranging from 90% to 140% of RMT in 10% steps. TMS pulses were delivered in a pseudorandom order, with intertrial intervals between 6 and 10 seconds. Each I/O curve was constructed using data from seventy-two stimulation pulses and the time to acquire the two curves was approximately 25 minutes. Individual participant I/O curves were calculated by averaging the peak-to-peak MEP amplitudes from all trials for a given intensity.
Average landing position for the open-loop pointing measure at baseline, post 1 (immediate after LPA), post 2 (20 minutes after LPA), and post 3 (45 minutes after PA) (
As expected, the average point of subjective equality (PSE) of our preselected pseudoneglect participants was to the left of zero before PA (mean −2.9 mm, SEM 0.3 mm). To determine whether participants shifted their midline judgments after LPA we performed a mixed-design ANOVA on the point of subjective equality with Session (pre; post 1; post 2) as a within-subject variable and Hemisphere stimulated (LH; RH) as a between-subject variable. This analysis revealed no main effects or interactions (Session [
Average point of subjective equality (PSE) measured at baseline, post 1, and post 2 for each stimulated hemisphere group. PSE (
In conclusion, although our behavioral data revealed a tendency for subjects to shift their perceptual line bisection judgment rightward, this largely expected trend did not reach significance. Moreover, the group stimulated on the right hemisphere showed this trend only at post 2.
Raw MEP amplitudes shown separately for the five different stimulation conditions (M1-only and 4 PPC-M1 ISIs) at baseline, post 1, and post 2 for each stimulated hemisphere. MEP amplitude (
To assess the functional connectivity between PPC and M1 before and after LPA, that is, to measure PA-induced changes in functional connectivity, we performed a mixed-design ANOVA with Hemisphere stimulated (left; right) as a between-subject variable and Session (Pre; Post 1; Post 2) and ISI (M1-only; 2; 4; 6; 8) as within-subject variables. This analysis revealed a significant interaction between Session and Hemisphere [
Since there was no main effect of ISI or any interaction between ISI and any other variables this result suggests that MEP amplitudes changed across all stimulation conditions (M1-only stimulation and PPC-M1 stimulations). Since all 5 levels contain a contribution from the M1-only stimulation, we performed a separate analysis on MEP amplitudes from the M1-only condition to investigate its contribution to the interaction observed above. The mixed-design ANOVA with Hemisphere stimulated (left; right) as a between-subject variable and Session (pre; post 1; post 2) as a within-subject variable revealed no main effects of Session or Hemisphere (both
The significant change in the absolute excitability of the motor cortex, increase in LH and decrease in RH M1-only MEP amplitudes, suggests that the general increase we observed in raw MEP amplitudes across all five tested conditions (M1-only and 2, 4, 6, and 8 ms CS-TS ISIs) might have been triggered by M1 excitability changes. While this could indicate that LPA creates an imbalance between the cerebral hemispheres (similar to that created by a lesion of the right hemisphere), this finding is inconsistent with a recent report showing that M1-only MEPs are unchanged following PA [
Increases or decreases in the slope of motor cortex input/output curves provide a good measure of changes in corticospinal excitability (CSE). Thus, to further investigate possible LPA-induced changes in CSE we conducted a second experiment in which we recorded M1 input-output curves from both hemispheres of 10 subjects (none of whom were tested in Experiment 1) before and after adaptation to the same 15-degree leftward deviating prisms used in Experiment 1.
Input/output curve plots of average MEP amplitude across all subjects in Experiment 2 for both hemispheres (
To obtain the slope (and
Finally, for each participant we used the I/O curves to estimate (to the nearest 10% of RMT) the percentage of stimulator output required to produce a MEP of approximately 1 mV before and after LPA. We then performed a two-way repeated measures ANOVA (variables Hemisphere stimulated and Session) on these percentages. This revealed no significant main effects or interactions (all
The aim of this study was to investigate the physiological counterpart of the neglect-like behavior induced by LPA in healthy individuals. Our hypothesis was that neglect-like behavior arises from modulation of the strength of PPC-M1 interactions. Specifically, we hypothesized that LPA would decrease the activation of the right PPC, measured as a decrease in right hemisphere PPC-M1 connectivity, and that via the release of interhemispheric inhibition this would in turn increase activity in the left PPC, measured as an increase in left hemisphere PPC-M1 connectivity. Our findings provide only partial support for this hypothesis, as, at the behavioral level, the expected rightward shift was almost absent following LPA, particularly after parietal-motor ppTMS of the RH. At the neurophysiological level an asymmetrical modulation of MEPs in the predicted direction was observed (increase in the left hemisphere and decrease in the right one), but this was accompanied by a similar change when M1 was stimulated alone, thus casting some doubts on the specificity of this effect. Given that our control experiment ruled out the possibility that absolute CSE was modulated solely by LPA, we discuss the behavioral and neurophysiological results in the context of previous work, and we conclude by suggesting that the combination of LPA and parietal-motor ppTMS stimulation is responsible for differentially altering the excitability of the motor cortex in each hemisphere.
On the behavioral side, the open-loop pointing measure revealed that following LPA participants had a significant rightward visuomotor after effect. Surprisingly, despite the well-documented rightward shift in line bisection judgments after LPA (see, e.g., [
On the neurophysiological side, MEP amplitudes were significantly modulated after LPA, increasing in the LH and decreasing in the RH. Importantly, this modulation was observed for both M1-only and PPC-M1 trials. Our analysis of CSE (M1-only trials) before and after LPA revealed that CSE increased in the LH and decreased in the RH. Since MEP amplitudes recorded on paired-pulse stimulation trials are inevitably affected by CSE, we cannot rule out the possibility that the PPC-M1 connectivity modulations we observed are mainly due to changes in CSE. In this respect, the left hemisphere increase and right hemisphere decrease in CSE could be interpreted as evidence in support of Pisella and colleagues’ [
An alternative interpretation is that the paired PPC-M1 stimulation could have induced corticocortical associative plasticity, as it has recently been shown [
While there is currently no evidence to suggest that paired stimulation of the type used here (i.e., no strict timing between parietal and motor stimuli or between subsequent stimuli pairs) can produce neuromodulatory effects, recent studies have demonstrated that MEP amplitudes are time variant [
The ppTMS technique has been widely used in recent years as a method to assess parietal-motor functional connectivity. Since our primary goal was to investigate the effects of prismatic adaptation, we chose to thoroughly examine the influence of prismatic adaptation on CSE rather than to look at the possible effect of multiple ppTMS measurements. While there is no published data directly investigating changes in CSE induced by multiple ppTMS measures, there is some evidence that repeated ppTMS measures do not alter CSE. For example, some studies include multiple measures because of their experimental design, and when MEP amplitudes for M1-only stimuli are reported (and the intensity of the stimulator was not adjusted to keep the test stimulus at an average amplitude of 1 millivolt [
To conclude, we designed a dual-site paired-pulse experiment with the intention of using parietal-motor interactions as a proxy for changes in parietal cortex excitability following adaptation to LPA. The differential change in CSE in the left and right hemispheres we observed, plus the absence of the well-documented right shift in line bisection judgments, leads us to suggest that LPA interacted with our parietal and motor cortex stimulation. We conclude, therefore, that under the physiological conditions produced by prism adaptation paired parietal-motor stimulation can act as a neuromodulator.
The authors declare that they have no competing interests.
Laure Pisella, Alessandro Farnè, and Karen T. Reilly contributed equally to this work.
This work was performed at the Neuro-Immersion Platform and supported by the Labex/Idex ANR-11-LABX-0042, IHU CeSaMe ANR-10-IBHU-0003, and by grants from the Fondation pour la Recherche Médicale (FRM) and the James S. McDonnell Foundation; Elisa Martín-Arévalo was supported by funding from FRM (SPF20140129218), Michael Vesia was supported by Natural Sciences and Engineering Research Council of Canada (NSERC), and Alessandro Farnè and Selene Schintu were supported by a James S. McDonnell Scholar Award.